From 6a806a214af42ac951e2d85e64d1bf4463482e16 Mon Sep 17 00:00:00 2001 From: Sifan Naeem Date: Thu, 18 Jun 2015 13:50:54 +0100 Subject: spi: img-spfi: fix support for speeds up to 1/4th input clock Setting the Same Edge bit indicates to the spfi block to receive and transmit data on the same edge of the spfi clock, which in turn doubles the operating frequency of spfi. The maximum supported frequency is limited to 1/4th of the spfi input clock, but without this bit set the maximum would be 1/8th of the input clock. The current driver calculates the divisor with maximum speed at 1/4th of the input clock, this would fail if the requested frequency is higher than 1/8 of the input clock. Any requests for 1/8th of the input clock would still pass. Fixes: 8543d0e72d43 ("spi: img-spfi: Limit bit clock to 1/4th of input clock") Signed-off-by: Sifan Naeem Signed-off-by: Mark Brown Cc: diff --git a/drivers/spi/spi-img-spfi.c b/drivers/spi/spi-img-spfi.c index 788e2b1..acce90a 100644 --- a/drivers/spi/spi-img-spfi.c +++ b/drivers/spi/spi-img-spfi.c @@ -40,6 +40,7 @@ #define SPFI_CONTROL_SOFT_RESET BIT(11) #define SPFI_CONTROL_SEND_DMA BIT(10) #define SPFI_CONTROL_GET_DMA BIT(9) +#define SPFI_CONTROL_SE BIT(8) #define SPFI_CONTROL_TMODE_SHIFT 5 #define SPFI_CONTROL_TMODE_MASK 0x7 #define SPFI_CONTROL_TMODE_SINGLE 0 @@ -491,6 +492,7 @@ static void img_spfi_config(struct spi_master *master, struct spi_device *spi, else if (xfer->tx_nbits == SPI_NBITS_QUAD && xfer->rx_nbits == SPI_NBITS_QUAD) val |= SPFI_CONTROL_TMODE_QUAD << SPFI_CONTROL_TMODE_SHIFT; + val |= SPFI_CONTROL_SE; spfi_writel(spfi, val, SPFI_CONTROL); } -- cgit v0.10.2 From 32c848e33ace75fce388cceff76223d12b46eaa3 Mon Sep 17 00:00:00 2001 From: Krzysztof Kozlowski Date: Wed, 24 Jun 2015 19:48:43 +0900 Subject: regulator: s2mps11: Fix GPIO suspend enable shift wrapping bug Status of enabling suspend mode for regulator was stored in bitmap-like long integer. However since adding support for S2MPU02 the number of regulators exceeded 32 so on devices with more than 32 regulators (S2MPU02 and S2MPS13) overflow happens when shifting the bit. This could lead to enabling suspend mode for completely different regulator than intended or to switching different regulator to other mode (e.g. from always enabled to controlled by PWRHOLD pin). Both cases could result in larger energy usage and issues when suspending to RAM. Fixes: 00e2573d2c10 ("regulator: s2mps11: Add support S2MPU02 regulator device") Reported-by: Dan Carpenter Signed-off-by: Krzysztof Kozlowski Signed-off-by: Mark Brown Cc: diff --git a/drivers/regulator/s2mps11.c b/drivers/regulator/s2mps11.c index 326ffb5..72fc3c3 100644 --- a/drivers/regulator/s2mps11.c +++ b/drivers/regulator/s2mps11.c @@ -34,6 +34,8 @@ #include #include +/* The highest number of possible regulators for supported devices. */ +#define S2MPS_REGULATOR_MAX S2MPS13_REGULATOR_MAX struct s2mps11_info { unsigned int rdev_num; int ramp_delay2; @@ -49,7 +51,7 @@ struct s2mps11_info { * One bit for each S2MPS13/S2MPS14/S2MPU02 regulator whether * the suspend mode was enabled. */ - unsigned long long s2mps14_suspend_state:50; + DECLARE_BITMAP(suspend_state, S2MPS_REGULATOR_MAX); /* Array of size rdev_num with GPIO-s for external sleep control */ int *ext_control_gpio; @@ -500,7 +502,7 @@ static int s2mps14_regulator_enable(struct regulator_dev *rdev) switch (s2mps11->dev_type) { case S2MPS13X: case S2MPS14X: - if (s2mps11->s2mps14_suspend_state & (1 << rdev_get_id(rdev))) + if (test_bit(rdev_get_id(rdev), s2mps11->suspend_state)) val = S2MPS14_ENABLE_SUSPEND; else if (gpio_is_valid(s2mps11->ext_control_gpio[rdev_get_id(rdev)])) val = S2MPS14_ENABLE_EXT_CONTROL; @@ -508,7 +510,7 @@ static int s2mps14_regulator_enable(struct regulator_dev *rdev) val = rdev->desc->enable_mask; break; case S2MPU02: - if (s2mps11->s2mps14_suspend_state & (1 << rdev_get_id(rdev))) + if (test_bit(rdev_get_id(rdev), s2mps11->suspend_state)) val = S2MPU02_ENABLE_SUSPEND; else val = rdev->desc->enable_mask; @@ -562,7 +564,7 @@ static int s2mps14_regulator_set_suspend_disable(struct regulator_dev *rdev) if (ret < 0) return ret; - s2mps11->s2mps14_suspend_state |= (1 << rdev_get_id(rdev)); + set_bit(rdev_get_id(rdev), s2mps11->suspend_state); /* * Don't enable suspend mode if regulator is already disabled because * this would effectively for a short time turn on the regulator after @@ -960,18 +962,22 @@ static int s2mps11_pmic_probe(struct platform_device *pdev) case S2MPS11X: s2mps11->rdev_num = ARRAY_SIZE(s2mps11_regulators); regulators = s2mps11_regulators; + BUILD_BUG_ON(S2MPS_REGULATOR_MAX < s2mps11->rdev_num); break; case S2MPS13X: s2mps11->rdev_num = ARRAY_SIZE(s2mps13_regulators); regulators = s2mps13_regulators; + BUILD_BUG_ON(S2MPS_REGULATOR_MAX < s2mps11->rdev_num); break; case S2MPS14X: s2mps11->rdev_num = ARRAY_SIZE(s2mps14_regulators); regulators = s2mps14_regulators; + BUILD_BUG_ON(S2MPS_REGULATOR_MAX < s2mps11->rdev_num); break; case S2MPU02: s2mps11->rdev_num = ARRAY_SIZE(s2mpu02_regulators); regulators = s2mpu02_regulators; + BUILD_BUG_ON(S2MPS_REGULATOR_MAX < s2mps11->rdev_num); break; default: dev_err(&pdev->dev, "Invalid device type: %u\n", -- cgit v0.10.2 From 923a8c1d8069104726bde55c37cec66324ccc328 Mon Sep 17 00:00:00 2001 From: Emmanuel Grumbach Date: Sun, 31 May 2015 21:44:22 +0300 Subject: iwlwifi: mvm: fix antenna selection when BT is active When BT is active, we want to avoid the shared antenna for management frame to make sure we don't disturb BT. There was a bug in that code because it chose the antenna BIT(ANT_A) where ANT_A is already a bitmap (0x1). This means that the antenna chosen in the end was ANT_B. While this is not optimal on devices with 2 antennas (it'd disturb BT), it is critical on single antenna devices like 3160 which couldn't connect at all when BT was active. This fixes: https://bugzilla.kernel.org/show_bug.cgi?id=97181 CC: [3.17+] Fixes: 34c8b24ff284 ("iwlwifi: mvm: BT Coex - avoid the shared antenna for management frames") Signed-off-by: Emmanuel Grumbach diff --git a/drivers/net/wireless/iwlwifi/mvm/tx.c b/drivers/net/wireless/iwlwifi/mvm/tx.c index 7ba7a118..8911686 100644 --- a/drivers/net/wireless/iwlwifi/mvm/tx.c +++ b/drivers/net/wireless/iwlwifi/mvm/tx.c @@ -252,7 +252,7 @@ void iwl_mvm_set_tx_cmd_rate(struct iwl_mvm *mvm, struct iwl_tx_cmd *tx_cmd, if (info->band == IEEE80211_BAND_2GHZ && !iwl_mvm_bt_coex_is_shared_ant_avail(mvm)) - rate_flags = BIT(mvm->cfg->non_shared_ant) << RATE_MCS_ANT_POS; + rate_flags = mvm->cfg->non_shared_ant << RATE_MCS_ANT_POS; else rate_flags = BIT(mvm->mgmt_last_antenna_idx) << RATE_MCS_ANT_POS; -- cgit v0.10.2 From 11828dbce6f769ed631919aa378b645b1af6bda6 Mon Sep 17 00:00:00 2001 From: Matti Gottlieb Date: Mon, 1 Jun 2015 15:15:11 +0300 Subject: iwlwifi: mvm: Avoid accessing Null pointer when setting igtk Sometimes when setting an igtk key the station maybe NULL. In the of case igtk the function will skip to the end, and try to print sta->addr, if sta is Null - we will access a Null pointer. Avoid accessing a Null pointer when setting a igtk key & the sta == NULL, and print a default MAC address instead. Signed-off-by: Matti Gottlieb Signed-off-by: Emmanuel Grumbach diff --git a/drivers/net/wireless/iwlwifi/mvm/sta.c b/drivers/net/wireless/iwlwifi/mvm/sta.c index d68dc69..26f076e 100644 --- a/drivers/net/wireless/iwlwifi/mvm/sta.c +++ b/drivers/net/wireless/iwlwifi/mvm/sta.c @@ -1401,6 +1401,7 @@ int iwl_mvm_set_sta_key(struct iwl_mvm *mvm, bool mcast = !(keyconf->flags & IEEE80211_KEY_FLAG_PAIRWISE); u8 sta_id; int ret; + static const u8 __maybe_unused zero_addr[ETH_ALEN] = {0}; lockdep_assert_held(&mvm->mutex); @@ -1467,7 +1468,7 @@ int iwl_mvm_set_sta_key(struct iwl_mvm *mvm, end: IWL_DEBUG_WEP(mvm, "key: cipher=%x len=%d idx=%d sta=%pM ret=%d\n", keyconf->cipher, keyconf->keylen, keyconf->keyidx, - sta->addr, ret); + sta ? sta->addr : zero_addr, ret); return ret; } -- cgit v0.10.2 From 0fd72ff92d6bea32bba612744abbe6a0abd25e43 Mon Sep 17 00:00:00 2001 From: Dan Carpenter Date: Wed, 24 Jun 2015 17:27:43 +0300 Subject: HID: wacom: NULL dereferences on error in probe() We can't pass a NULL to input_unregister_device(). Fixes: 2a6cdbdd4cc0 ('HID: wacom: Introduce new 'touch_input' device') Signed-off-by: Dan Carpenter Reviewed-by: Jason Gerecke Signed-off-by: Jiri Kosina diff --git a/drivers/hid/wacom_sys.c b/drivers/hid/wacom_sys.c index 4c0ffca..44958d7 100644 --- a/drivers/hid/wacom_sys.c +++ b/drivers/hid/wacom_sys.c @@ -1271,11 +1271,13 @@ fail_leds: pad_input_dev = NULL; wacom_wac->pad_registered = false; fail_register_pad_input: - input_unregister_device(touch_input_dev); + if (touch_input_dev) + input_unregister_device(touch_input_dev); wacom_wac->touch_input = NULL; wacom_wac->touch_registered = false; fail_register_touch_input: - input_unregister_device(pen_input_dev); + if (pen_input_dev) + input_unregister_device(pen_input_dev); wacom_wac->pen_input = NULL; wacom_wac->pen_registered = false; fail_register_pen_input: -- cgit v0.10.2 From c5b2b809cee8db018ac68566fe2114c175d79b5b Mon Sep 17 00:00:00 2001 From: Reyad Attiyat Date: Sun, 28 Jun 2015 19:22:37 -0500 Subject: HID: microsoft: Add quirk for MS Surface Type/Touch cover The newer firmware on MS Surface 2 tablets causes the type and touch cover keyboards to timeout when waiting for reports. The quirk HID_QUIRK_NO_INIT_REPORTS allows them to function normally. Signed-off-by: Reyad Attiyat Signed-off-by: Jiri Kosina diff --git a/drivers/hid/usbhid/hid-quirks.c b/drivers/hid/usbhid/hid-quirks.c index 53e7de7..20f9a65 100644 --- a/drivers/hid/usbhid/hid-quirks.c +++ b/drivers/hid/usbhid/hid-quirks.c @@ -87,6 +87,9 @@ static const struct hid_blacklist { { USB_VENDOR_ID_LOGITECH, USB_DEVICE_ID_LOGITECH_MOUSE_C05A, HID_QUIRK_ALWAYS_POLL }, { USB_VENDOR_ID_LOGITECH, USB_DEVICE_ID_LOGITECH_MOUSE_C06A, HID_QUIRK_ALWAYS_POLL }, { USB_VENDOR_ID_MGE, USB_DEVICE_ID_MGE_UPS, HID_QUIRK_NOGET }, + { USB_VENDOR_ID_MICROSOFT, USB_DEVICE_ID_MS_SURFACE_PRO_2, HID_QUIRK_NO_INIT_REPORTS }, + { USB_VENDOR_ID_MICROSOFT, USB_DEVICE_ID_MS_TYPE_COVER_2, HID_QUIRK_NO_INIT_REPORTS }, + { USB_VENDOR_ID_MICROSOFT, USB_DEVICE_ID_MS_TOUCH_COVER_2, HID_QUIRK_NO_INIT_REPORTS }, { USB_VENDOR_ID_MICROSOFT, USB_DEVICE_ID_MS_TYPE_COVER_3, HID_QUIRK_NO_INIT_REPORTS }, { USB_VENDOR_ID_MICROSOFT, USB_DEVICE_ID_MS_TYPE_COVER_3_JP, HID_QUIRK_NO_INIT_REPORTS }, { USB_VENDOR_ID_MICROSOFT, USB_DEVICE_ID_MS_POWER_COVER, HID_QUIRK_NO_INIT_REPORTS }, -- cgit v0.10.2 From 8b0a9d42301e45d501d751074a6f767fded680b1 Mon Sep 17 00:00:00 2001 From: "Michael S. Tsirkin" Date: Mon, 29 Jun 2015 11:16:25 +0200 Subject: virtio_net: document VIRTIO_NET_CTRL_GUEST_OFFLOADS Document VIRTIO_NET_CTRL_GUEST_OFFLOADS and the relevant feature bits. Will allow ethtool control of the offloads down the road. Reported-by: Yan Vugenfirer Signed-off-by: Michael S. Tsirkin diff --git a/include/uapi/linux/virtio_net.h b/include/uapi/linux/virtio_net.h index 7bbee79..ec32293 100644 --- a/include/uapi/linux/virtio_net.h +++ b/include/uapi/linux/virtio_net.h @@ -34,6 +34,7 @@ /* The feature bitmap for virtio net */ #define VIRTIO_NET_F_CSUM 0 /* Host handles pkts w/ partial csum */ #define VIRTIO_NET_F_GUEST_CSUM 1 /* Guest handles pkts w/ partial csum */ +#define VIRTIO_NET_F_CTRL_GUEST_OFFLOADS 2 /* Dynamic offload configuration. */ #define VIRTIO_NET_F_MAC 5 /* Host has given MAC address. */ #define VIRTIO_NET_F_GUEST_TSO4 7 /* Guest can handle TSOv4 in. */ #define VIRTIO_NET_F_GUEST_TSO6 8 /* Guest can handle TSOv6 in. */ @@ -226,4 +227,19 @@ struct virtio_net_ctrl_mq { #define VIRTIO_NET_CTRL_MQ_VQ_PAIRS_MIN 1 #define VIRTIO_NET_CTRL_MQ_VQ_PAIRS_MAX 0x8000 +/* + * Control network offloads + * + * Reconfigures the network offloads that Guest can handle. + * + * Available with the VIRTIO_NET_F_CTRL_GUEST_OFFLOADS feature bit. + * + * Command data format matches the feature bit mask exactly. + * + * See VIRTIO_NET_F_GUEST_* for the list of offloads + * that can be enabled/disabled. + */ +#define VIRTIO_NET_CTRL_GUEST_OFFLOADS 5 +#define VIRTIO_NET_CTRL_GUEST_OFFLOADS_SET 0 + #endif /* _LINUX_VIRTIO_NET_H */ -- cgit v0.10.2 From bcfeacab45e6d419c6bafc0e57ea4b1125e23231 Mon Sep 17 00:00:00 2001 From: Igor Mammedov Date: Tue, 16 Jun 2015 18:33:35 +0200 Subject: vhost: use binary search instead of linear in find_region() For default region layouts performance stays the same as linear search i.e. it takes around 210ns average for translate_desc() that inlines find_region(). But it scales better with larger amount of regions, 235ns BS vs 300ns LS with 55 memory regions and it will be about the same values when allowed number of slots is increased to 509 like it has been done in kvm. Signed-off-by: Igor Mammedov Signed-off-by: Michael S. Tsirkin diff --git a/drivers/vhost/vhost.c b/drivers/vhost/vhost.c index 9e8e004..71bb468 100644 --- a/drivers/vhost/vhost.c +++ b/drivers/vhost/vhost.c @@ -25,6 +25,7 @@ #include #include #include +#include #include "vhost.h" @@ -663,6 +664,16 @@ int vhost_vq_access_ok(struct vhost_virtqueue *vq) } EXPORT_SYMBOL_GPL(vhost_vq_access_ok); +static int vhost_memory_reg_sort_cmp(const void *p1, const void *p2) +{ + const struct vhost_memory_region *r1 = p1, *r2 = p2; + if (r1->guest_phys_addr < r2->guest_phys_addr) + return 1; + if (r1->guest_phys_addr > r2->guest_phys_addr) + return -1; + return 0; +} + static long vhost_set_memory(struct vhost_dev *d, struct vhost_memory __user *m) { struct vhost_memory mem, *newmem, *oldmem; @@ -682,9 +693,11 @@ static long vhost_set_memory(struct vhost_dev *d, struct vhost_memory __user *m) memcpy(newmem, &mem, size); if (copy_from_user(newmem->regions, m->regions, mem.nregions * sizeof *m->regions)) { - kfree(newmem); + kvfree(newmem); return -EFAULT; } + sort(newmem->regions, newmem->nregions, sizeof(*newmem->regions), + vhost_memory_reg_sort_cmp, NULL); if (!memory_access_ok(d, newmem, 0)) { kfree(newmem); @@ -992,17 +1005,22 @@ EXPORT_SYMBOL_GPL(vhost_dev_ioctl); static const struct vhost_memory_region *find_region(struct vhost_memory *mem, __u64 addr, __u32 len) { - struct vhost_memory_region *reg; - int i; + const struct vhost_memory_region *reg; + int start = 0, end = mem->nregions; - /* linear search is not brilliant, but we really have on the order of 6 - * regions in practice */ - for (i = 0; i < mem->nregions; ++i) { - reg = mem->regions + i; - if (reg->guest_phys_addr <= addr && - reg->guest_phys_addr + reg->memory_size - 1 >= addr) - return reg; + while (start < end) { + int slot = start + (end - start) / 2; + reg = mem->regions + slot; + if (addr >= reg->guest_phys_addr) + end = slot; + else + start = slot + 1; } + + reg = mem->regions + start; + if (addr >= reg->guest_phys_addr && + reg->guest_phys_addr + reg->memory_size > addr) + return reg; return NULL; } -- cgit v0.10.2 From 93aa6c7bbc4124c8361c26a8b2c5c40afb185619 Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Fri, 3 Jul 2015 09:28:09 -0400 Subject: SUNRPC: Don't reencode message if transmission failed with ENOBUFS If we're running out of buffer memory when transmitting data, then we want to just delay for a moment, and then continue transmitting the remainder of the message. Signed-off-by: Trond Myklebust diff --git a/net/sunrpc/clnt.c b/net/sunrpc/clnt.c index cbc6af9..23608eb 100644 --- a/net/sunrpc/clnt.c +++ b/net/sunrpc/clnt.c @@ -1902,6 +1902,7 @@ call_transmit_status(struct rpc_task *task) switch (task->tk_status) { case -EAGAIN: + case -ENOBUFS: break; default: dprint_status(task); @@ -1928,7 +1929,6 @@ call_transmit_status(struct rpc_task *task) case -ECONNABORTED: case -EADDRINUSE: case -ENOTCONN: - case -ENOBUFS: case -EPIPE: rpc_task_force_reencode(task); } @@ -2057,12 +2057,13 @@ call_status(struct rpc_task *task) case -ECONNABORTED: rpc_force_rebind(clnt); case -EADDRINUSE: - case -ENOBUFS: rpc_delay(task, 3*HZ); case -EPIPE: case -ENOTCONN: task->tk_action = call_bind; break; + case -ENOBUFS: + rpc_delay(task, HZ>>2); case -EAGAIN: task->tk_action = call_transmit; break; -- cgit v0.10.2 From b5872f0c67edf3714dd46f04d73c3644f3addaf9 Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Fri, 3 Jul 2015 09:32:23 -0400 Subject: SUNRPC: Don't confuse ENOBUFS with a write_space issue ENOBUFS means that memory allocations are failing due to an actual low memory situation. It should not be confused with being out of socket buffer space. Handle the problem by just punting to the delay in call_status. Reported-by: Neil Brown Signed-off-by: Trond Myklebust diff --git a/net/sunrpc/xprtsock.c b/net/sunrpc/xprtsock.c index ee359fc..44c1927 100644 --- a/net/sunrpc/xprtsock.c +++ b/net/sunrpc/xprtsock.c @@ -539,6 +539,7 @@ static int xs_local_send_request(struct rpc_task *task) switch (status) { case -ENOBUFS: + break; case -EAGAIN: status = xs_nospace(task); break; @@ -692,7 +693,6 @@ static int xs_tcp_send_request(struct rpc_task *task) status = -ENOTCONN; /* Should we call xs_close() here? */ break; - case -ENOBUFS: case -EAGAIN: status = xs_nospace(task); break; @@ -703,6 +703,7 @@ static int xs_tcp_send_request(struct rpc_task *task) case -ECONNREFUSED: case -ENOTCONN: case -EADDRINUSE: + case -ENOBUFS: case -EPIPE: clear_bit(SOCK_ASYNC_NOSPACE, &transport->sock->flags); } -- cgit v0.10.2 From 8b895ce652dc73bcf42a0c24acfc0708a06ea7c3 Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Sun, 5 Jul 2015 15:12:06 -0400 Subject: NFSv4.1: Handle SEQ4_STATUS_EXPIRED_SOME_STATE_REVOKED status bit correctly. If the server tells us that only some state has been revoked, then we need to run the full TEST_STATEID dog and pony show in order to discover which locks and delegations are still OK. Currently we blow away all state, which means that we lose all locks! Signed-off-by: Trond Myklebust diff --git a/fs/nfs/nfs4state.c b/fs/nfs/nfs4state.c index 605840d..f769161 100644 --- a/fs/nfs/nfs4state.c +++ b/fs/nfs/nfs4state.c @@ -2191,12 +2191,20 @@ static void nfs41_handle_server_reboot(struct nfs_client *clp) } } -static void nfs41_handle_state_revoked(struct nfs_client *clp) +static void nfs41_handle_all_state_revoked(struct nfs_client *clp) { nfs4_reset_all_state(clp); dprintk("%s: state revoked on server %s\n", __func__, clp->cl_hostname); } +static void nfs41_handle_some_state_revoked(struct nfs_client *clp) +{ + nfs4_state_mark_reclaim_helper(clp, nfs4_state_mark_reclaim_nograce); + nfs4_schedule_state_manager(clp); + + dprintk("%s: state revoked on server %s\n", __func__, clp->cl_hostname); +} + static void nfs41_handle_recallable_state_revoked(struct nfs_client *clp) { /* This will need to handle layouts too */ @@ -2231,10 +2239,11 @@ void nfs41_handle_sequence_flag_errors(struct nfs_client *clp, u32 flags) if (flags & SEQ4_STATUS_RESTART_RECLAIM_NEEDED) nfs41_handle_server_reboot(clp); - if (flags & (SEQ4_STATUS_EXPIRED_ALL_STATE_REVOKED | - SEQ4_STATUS_EXPIRED_SOME_STATE_REVOKED | + if (flags & (SEQ4_STATUS_EXPIRED_ALL_STATE_REVOKED)) + nfs41_handle_all_state_revoked(clp); + if (flags & (SEQ4_STATUS_EXPIRED_SOME_STATE_REVOKED | SEQ4_STATUS_ADMIN_STATE_REVOKED)) - nfs41_handle_state_revoked(clp); + nfs41_handle_some_state_revoked(clp); if (flags & SEQ4_STATUS_LEASE_MOVED) nfs4_schedule_lease_moved_recovery(clp); if (flags & SEQ4_STATUS_RECALLABLE_STATE_REVOKED) -- cgit v0.10.2 From 4099287feb5833c24b5a91e1fe55207cf7559350 Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Sun, 5 Jul 2015 15:20:53 -0400 Subject: NFSv4.1: Handle SEQ4_STATUS_RECALLABLE_STATE_REVOKED status bit correctly Try to handle this for now by invalidating all outstanding layouts for this server and then testing all the open+lock+delegation stateids. At some later stage, we may want to optimise by separating out the testing of delegation stateids only, and adding testing of layout stateids. Signed-off-by: Trond Myklebust diff --git a/fs/nfs/nfs4state.c b/fs/nfs/nfs4state.c index f769161..4694529 100644 --- a/fs/nfs/nfs4state.c +++ b/fs/nfs/nfs4state.c @@ -2207,8 +2207,10 @@ static void nfs41_handle_some_state_revoked(struct nfs_client *clp) static void nfs41_handle_recallable_state_revoked(struct nfs_client *clp) { - /* This will need to handle layouts too */ - nfs_expire_all_delegations(clp); + /* FIXME: For now, we destroy all layouts. */ + pnfs_destroy_all_layouts(clp); + /* FIXME: For now, we test all delegations+open state+locks. */ + nfs41_handle_some_state_revoked(clp); dprintk("%s: Recallable state revoked on server %s!\n", __func__, clp->cl_hostname); } -- cgit v0.10.2 From b13529059cf782f9b4e4ac0ca9d524bd922163da Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Sun, 5 Jul 2015 15:26:58 -0400 Subject: NFSv4.1: Handle SEQ4_STATUS_BACKCHANNEL_FAULT correctly RFC5661 states: The server has encountered an unrecoverable fault with the backchannel (e.g., it has lost track of the sequence ID for a slot in the backchannel). The client MUST stop sending more requests on the session's fore channel, wait for all outstanding requests to complete on the fore and back channel, and then destroy the session. Ensure we do so... Signed-off-by: Trond Myklebust diff --git a/fs/nfs/nfs4state.c b/fs/nfs/nfs4state.c index 4694529..f2e2ad8 100644 --- a/fs/nfs/nfs4state.c +++ b/fs/nfs/nfs4state.c @@ -2217,9 +2217,9 @@ static void nfs41_handle_recallable_state_revoked(struct nfs_client *clp) static void nfs41_handle_backchannel_fault(struct nfs_client *clp) { - nfs_expire_all_delegations(clp); - if (test_and_set_bit(NFS4CLNT_SESSION_RESET, &clp->cl_state) == 0) - nfs4_schedule_state_manager(clp); + set_bit(NFS4CLNT_SESSION_RESET, &clp->cl_state); + nfs4_schedule_state_manager(clp); + dprintk("%s: server %s declared a backchannel fault\n", __func__, clp->cl_hostname); } -- cgit v0.10.2 From b15c7cdde4991be5058f442c6d08d404d56f662c Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Sun, 5 Jul 2015 15:01:36 -0400 Subject: NFSv4.1: nfs41_sequence_done should handle sequence flag errors Instead of just kicking off lease recovery, we should look into the sequence flag errors and handle them. Signed-off-by: Trond Myklebust diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c index 6f228b5..1607b41 100644 --- a/fs/nfs/nfs4proc.c +++ b/fs/nfs/nfs4proc.c @@ -616,8 +616,7 @@ int nfs41_sequence_done(struct rpc_task *task, struct nfs4_sequence_res *res) clp = session->clp; do_renew_lease(clp, res->sr_timestamp); /* Check sequence flags */ - if (res->sr_status_flags != 0) - nfs4_schedule_lease_recovery(clp); + nfs41_handle_sequence_flag_errors(clp, res->sr_status_flags); nfs41_update_target_slotid(slot->table, slot, res); break; case 1: -- cgit v0.10.2 From be824167e33a8b747423c90f72479deb03255d54 Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Sun, 5 Jul 2015 14:50:46 -0400 Subject: NFSv4: Leases are renewed in sequence_done when we have sessions Ensure that the calls to renew_lease() in open_done() etc. only apply to session-less versions of NFSv4.x (i.e. NFSv4.0). Signed-off-by: Trond Myklebust diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c index 1607b41..9f24238 100644 --- a/fs/nfs/nfs4proc.c +++ b/fs/nfs/nfs4proc.c @@ -467,7 +467,10 @@ static void do_renew_lease(struct nfs_client *clp, unsigned long timestamp) static void renew_lease(const struct nfs_server *server, unsigned long timestamp) { - do_renew_lease(server->nfs_client, timestamp); + struct nfs_client *clp = server->nfs_client; + + if (!nfs4_has_session(clp)) + do_renew_lease(clp, timestamp); } struct nfs4_call_sync_data { @@ -7572,13 +7575,8 @@ static int nfs4_proc_sequence(struct nfs_client *clp, struct rpc_cred *cred) goto out; } ret = rpc_wait_for_completion_task(task); - if (!ret) { - struct nfs4_sequence_res *res = task->tk_msg.rpc_resp; - - if (task->tk_status == 0) - nfs41_handle_sequence_flag_errors(clp, res->sr_status_flags); + if (!ret) ret = task->tk_status; - } rpc_put_task(task); out: dprintk("<-- %s status=%d\n", __func__, ret); -- cgit v0.10.2 From 159eeea4cbc142bd5fd396a84cc63345455586a6 Mon Sep 17 00:00:00 2001 From: Robert Jarzmik Date: Thu, 25 Jun 2015 19:14:18 +0200 Subject: MAINTAINERS: mtd: docg3: add docg3 maintainer Add myself as maintainer of the NAND based MSystems DiskOnChip G3 driver. Signed-off-by: Robert Jarzmik Signed-off-by: Brian Norris diff --git a/MAINTAINERS b/MAINTAINERS index 8133cef..9567329 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -6824,6 +6824,12 @@ T: git git://linuxtv.org/anttip/media_tree.git S: Maintained F: drivers/media/usb/msi2500/ +MSYSTEMS DISKONCHIP G3 MTD DRIVER +M: Robert Jarzmik +L: linux-mtd@lists.infradead.org +S: Maintained +F: drivers/mtd/devices/docg3* + MT9M032 APTINA SENSOR DRIVER M: Laurent Pinchart L: linux-media@vger.kernel.org -- cgit v0.10.2 From 9b353cc8f199e4bc78023e8300306b5a16b48e75 Mon Sep 17 00:00:00 2001 From: Sagi Grimberg Date: Mon, 29 Jun 2015 13:07:06 +0300 Subject: target/pr: Fix possible uninitialized variable usage Triggered a compilation warning. Fixes: 2650d71e2 target: move transport ID handling to the core Signed-off-by: Sagi Grimberg Signed-off-by: Nicholas Bellinger diff --git a/drivers/target/target_core_pr.c b/drivers/target/target_core_pr.c index 0fdbe43..5ab7100 100644 --- a/drivers/target/target_core_pr.c +++ b/drivers/target/target_core_pr.c @@ -1474,7 +1474,7 @@ core_scsi3_decode_spec_i_port( LIST_HEAD(tid_dest_list); struct pr_transport_id_holder *tidh_new, *tidh, *tidh_tmp; unsigned char *buf, *ptr, proto_ident; - const unsigned char *i_str; + const unsigned char *i_str = NULL; char *iport_ptr = NULL, i_buf[PR_REG_ISID_ID_LEN]; sense_reason_t ret; u32 tpdl, tid_len = 0; -- cgit v0.10.2 From 27e6772b0d492375621f496a114617bf9c17c8d2 Mon Sep 17 00:00:00 2001 From: Sagi Grimberg Date: Mon, 29 Jun 2015 18:32:03 +0300 Subject: target/spc: Set SPT correctly in Extended INQUIRY Data VPD page LIO supports protection types 1,3 so setting a hard-coded SPT=3 is fine for now. Signed-off-by: Sagi Grimberg Signed-off-by: Nicholas Bellinger diff --git a/drivers/target/target_core_spc.c b/drivers/target/target_core_spc.c index b074443..5df4d9b 100644 --- a/drivers/target/target_core_spc.c +++ b/drivers/target/target_core_spc.c @@ -458,6 +458,9 @@ spc_emulate_evpd_86(struct se_cmd *cmd, unsigned char *buf) buf[4] = 0x4; } + /* logical unit supports type 1 and type 3 protection */ + buf[4] |= (0x3 << 3); + /* Set HEADSUP, ORDSUP, SIMPSUP */ buf[5] = 0x07; -- cgit v0.10.2 From 3aa3c67b266fd6114c019bbac35de79aca02d520 Mon Sep 17 00:00:00 2001 From: Nicholas Bellinger Date: Mon, 6 Jul 2015 15:26:37 -0700 Subject: target: Add extra TYPE_DISK + protection checks for INQUIRY SPT Cc: Sagi Grimberg Signed-off-by: Nicholas Bellinger diff --git a/drivers/target/target_core_spc.c b/drivers/target/target_core_spc.c index 5df4d9b..b5ba1ec 100644 --- a/drivers/target/target_core_spc.c +++ b/drivers/target/target_core_spc.c @@ -454,12 +454,16 @@ spc_emulate_evpd_86(struct se_cmd *cmd, unsigned char *buf) cmd->se_sess->sess_prot_type == TARGET_DIF_TYPE1_PROT) buf[4] = 0x5; else if (dev->dev_attrib.pi_prot_type == TARGET_DIF_TYPE3_PROT || - cmd->se_sess->sess_prot_type == TARGET_DIF_TYPE3_PROT) + cmd->se_sess->sess_prot_type == TARGET_DIF_TYPE3_PROT) buf[4] = 0x4; } /* logical unit supports type 1 and type 3 protection */ - buf[4] |= (0x3 << 3); + if ((dev->transport->get_device_type(dev) == TYPE_DISK) && + (sess->sup_prot_ops & (TARGET_PROT_DIN_PASS | TARGET_PROT_DOUT_PASS)) && + (dev->dev_attrib.pi_prot_type || cmd->se_sess->sess_prot_type)) { + buf[4] |= (0x3 << 3); + } /* Set HEADSUP, ORDSUP, SIMPSUP */ buf[5] = 0x07; -- cgit v0.10.2 From 5dacbfc934cab0778852b18dd5d67ecebfcf9a2e Mon Sep 17 00:00:00 2001 From: Sagi Grimberg Date: Sat, 4 Jul 2015 14:05:24 +0300 Subject: target/rd: Set ramdisk as non rotational device Since a RAM backend device is not really a rotational device, we set it as is_nonrot=1 which will be forwarded in VPD page 0xb1 (block device characteristics) response. Signed-off-by: Sagi Grimberg Signed-off-by: Nicholas Bellinger diff --git a/drivers/target/target_core_rd.c b/drivers/target/target_core_rd.c index 4703f40..384cf88 100644 --- a/drivers/target/target_core_rd.c +++ b/drivers/target/target_core_rd.c @@ -333,6 +333,7 @@ static int rd_configure_device(struct se_device *dev) dev->dev_attrib.hw_block_size = RD_BLOCKSIZE; dev->dev_attrib.hw_max_sectors = UINT_MAX; dev->dev_attrib.hw_queue_depth = RD_MAX_DEVICE_QUEUE_DEPTH; + dev->dev_attrib.is_nonrot = 1; rd_dev->rd_dev_id = rd_host->rd_host_dev_id_count++; -- cgit v0.10.2 From ae128293d97404f491dc76f1843c7adacfec3441 Mon Sep 17 00:00:00 2001 From: Krzysztof Kozlowski Date: Mon, 15 Jun 2015 17:25:16 +0900 Subject: dmaengine: pl330: Fix overflow when reporting residue in memcpy During memcpy operations the residue was always set to an u32 overflowed value. In pl330_tx_status() function number of currently transferred bytes was subtracted from internal "bytes_requested" field. However this "bytes_requested" was not initialized at start to length of memcpy buffer so transferred bytes were subtracted from 0 causing overflow. Signed-off-by: Krzysztof Kozlowski Cc: Fixes: aee4d1fac887 ("dmaengine: pl330: improve pl330_tx_status() function") Signed-off-by: Vinod Koul diff --git a/drivers/dma/pl330.c b/drivers/dma/pl330.c index f513f77..c535cd0 100644 --- a/drivers/dma/pl330.c +++ b/drivers/dma/pl330.c @@ -2623,6 +2623,7 @@ pl330_prep_dma_memcpy(struct dma_chan *chan, dma_addr_t dst, desc->rqcfg.brst_len = 1; desc->rqcfg.brst_len = get_burst_len(desc, len); + desc->bytes_requested = len; desc->txd.flags = flags; -- cgit v0.10.2 From 5dd90e5b91e0f5c925b12b132c7cd27538870256 Mon Sep 17 00:00:00 2001 From: Krzysztof Kozlowski Date: Mon, 15 Jun 2015 23:00:09 +0900 Subject: dmaengine: pl330: Really fix choppy sound because of wrong residue calculation When pl330 driver was used during sound playback, after some time or after a number of plays the sound became choppy or totally noisy. For example on Odroid XU3 board the first four executions of aplay with small WAVE worked fine, but fifth was unrecognizable with errors: $ aplay /usr/share/sounds/alsa/Front_Right.wava underrun!!! (at least 0.095 ms long) Issue was caused by wrong residue reported by pl330 driver to pcm_dmaengine for its cyclic dma transfers. The pl330_tx_status(), residue reporting function, used a "last" flag in a descriptor to indicate that there is no more data to send. The pl330_tx_submit() iterated over descriptors trying to remove this flag from them and then mark last descriptor as "last". However when iterating it actually removed the flag not from descriptors but always from last of it (and then reset it). Thus effectively once some descriptor was marked as last, then it stayed like this forever causing residue to be reported too low. Signed-off-by: Krzysztof Kozlowski Fixes: aee4d1fac887 ("dmaengine: pl330: improve pl330_tx_status() function") Cc: Reported-by: gabriel@unseen.is Suggested-by: Marek Szyprowski Tested-by: Lars-Peter Clausen Signed-off-by: Vinod Koul diff --git a/drivers/dma/pl330.c b/drivers/dma/pl330.c index c535cd0..ecab4ea0 100644 --- a/drivers/dma/pl330.c +++ b/drivers/dma/pl330.c @@ -2328,7 +2328,7 @@ static dma_cookie_t pl330_tx_submit(struct dma_async_tx_descriptor *tx) desc->txd.callback = last->txd.callback; desc->txd.callback_param = last->txd.callback_param; } - last->last = false; + desc->last = false; dma_cookie_assign(&desc->txd); -- cgit v0.10.2 From 908a5544cd29ed60114ed60bded6dbe8cdd56326 Mon Sep 17 00:00:00 2001 From: Stephen Rothwell Date: Tue, 30 Jun 2015 10:59:04 +1000 Subject: virtio scsi: fix unused variable warning drivers/scsi/virtio_scsi.c: In function 'virtscsi_probe': drivers/scsi/virtio_scsi.c:952:11: warning: unused variable 'host_prot' [-Wunused-variable] int err, host_prot; ^ Signed-off-by: Stephen Rothwell Reviewed-by: Michael S. Tsirkin Signed-off-by: Michael S. Tsirkin diff --git a/drivers/scsi/virtio_scsi.c b/drivers/scsi/virtio_scsi.c index f164f24..55441c7 100644 --- a/drivers/scsi/virtio_scsi.c +++ b/drivers/scsi/virtio_scsi.c @@ -944,7 +944,7 @@ static int virtscsi_probe(struct virtio_device *vdev) { struct Scsi_Host *shost; struct virtio_scsi *vscsi; - int err, host_prot; + int err; u32 sg_elems, num_targets; u32 cmd_per_lun; u32 num_queues; @@ -1003,6 +1003,8 @@ static int virtscsi_probe(struct virtio_device *vdev) shost->nr_hw_queues = num_queues; if (virtio_has_feature(vdev, VIRTIO_SCSI_F_T10_PI)) { + int host_prot; + host_prot = SHOST_DIF_TYPE1_PROTECTION | SHOST_DIF_TYPE2_PROTECTION | SHOST_DIF_TYPE3_PROTECTION | SHOST_DIX_TYPE1_PROTECTION | SHOST_DIX_TYPE2_PROTECTION | SHOST_DIX_TYPE3_PROTECTION; -- cgit v0.10.2 From d768f32aec8c0ebb8499ffca89cfed8f5f1a4432 Mon Sep 17 00:00:00 2001 From: Thomas Huth Date: Thu, 2 Jul 2015 09:21:22 +0200 Subject: virtio: Fix typecast of pointer in vring_init() The virtio_ring.h header is used in userspace programs (ie. QEMU), too. Here we can not assume that sizeof(pointer) is the same as sizeof(long), e.g. when compiling for Windows, so the typecast in vring_init() should be done with (uintptr_t) instead of (unsigned long). Signed-off-by: Thomas Huth Signed-off-by: Michael S. Tsirkin diff --git a/include/uapi/linux/virtio_ring.h b/include/uapi/linux/virtio_ring.h index 915980a..c072959 100644 --- a/include/uapi/linux/virtio_ring.h +++ b/include/uapi/linux/virtio_ring.h @@ -31,6 +31,9 @@ * SUCH DAMAGE. * * Copyright Rusty Russell IBM Corporation 2007. */ +#ifndef __KERNEL__ +#include +#endif #include #include @@ -143,7 +146,7 @@ static inline void vring_init(struct vring *vr, unsigned int num, void *p, vr->num = num; vr->desc = p; vr->avail = p + num*sizeof(struct vring_desc); - vr->used = (void *)(((unsigned long)&vr->avail->ring[num] + sizeof(__virtio16) + vr->used = (void *)(((uintptr_t)&vr->avail->ring[num] + sizeof(__virtio16) + align-1) & ~(align - 1)); } -- cgit v0.10.2 From 3121bb023e2db4f00ed6678898c09e35ed4b5301 Mon Sep 17 00:00:00 2001 From: "Michael S. Tsirkin" Date: Thu, 2 Jul 2015 10:56:49 +0200 Subject: virtio: define virtio_pci_cfg_cap in header. We already have VIRTIO_PCI_CAP_PCI_CFG, let's define the structure that goes with it. Signed-off-by: Michael S. Tsirkin diff --git a/include/uapi/linux/virtio_pci.h b/include/uapi/linux/virtio_pci.h index 7530146..90007a1 100644 --- a/include/uapi/linux/virtio_pci.h +++ b/include/uapi/linux/virtio_pci.h @@ -157,6 +157,12 @@ struct virtio_pci_common_cfg { __le32 queue_used_hi; /* read-write */ }; +/* Fields in VIRTIO_PCI_CAP_PCI_CFG: */ +struct virtio_pci_cfg_cap { + struct virtio_pci_cap cap; + __u8 pci_cfg_data[4]; /* Data for BAR access. */ +}; + /* Macro versions of offsets for the Old Timers! */ #define VIRTIO_PCI_CAP_VNDR 0 #define VIRTIO_PCI_CAP_NEXT 1 -- cgit v0.10.2 From f2dbda3b4fc2833f0a6240ac28b0e3a17f09893b Mon Sep 17 00:00:00 2001 From: Cornelia Huck Date: Wed, 1 Jul 2015 17:15:37 +0200 Subject: MAINTAINERS: separate section for s390 virtio drivers The s390-specific virtio drivers have probably more to do with virtio than with kvm today; let's move them out into a separate section to reflect this and to be able to add relevant mailing lists. CC: Christian Borntraeger Signed-off-by: Cornelia Huck Signed-off-by: Michael S. Tsirkin diff --git a/MAINTAINERS b/MAINTAINERS index af802b3..280a568 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -5690,7 +5690,6 @@ S: Supported F: Documentation/s390/kvm.txt F: arch/s390/include/asm/kvm* F: arch/s390/kvm/ -F: drivers/s390/kvm/ KERNEL VIRTUAL MACHINE (KVM) FOR ARM M: Christoffer Dall @@ -10571,6 +10570,15 @@ F: drivers/block/virtio_blk.c F: include/linux/virtio_*.h F: include/uapi/linux/virtio_*.h +VIRTIO DRIVERS FOR S390 +M: Christian Borntraeger +M: Cornelia Huck +L: linux-s390@vger.kernel.org +L: virtualization@lists.linux-foundation.org +L: kvm@vger.kernel.org +S: Supported +F: drivers/s390/kvm/ + VIRTIO HOST (VHOST) M: "Michael S. Tsirkin" L: kvm@vger.kernel.org -- cgit v0.10.2 From 1b568d934eec1c5c99565c41f6c8ca66e9743e96 Mon Sep 17 00:00:00 2001 From: Cornelia Huck Date: Tue, 7 Jul 2015 11:41:01 +0200 Subject: virtio/s390: rename drivers/s390/kvm -> drivers/s390/virtio This more accurately reflects what these drivers actually do. Suggested-by: Paolo Bonzini Acked-by: Christian Borntraeger Signed-off-by: Cornelia Huck Signed-off-by: Michael S. Tsirkin diff --git a/MAINTAINERS b/MAINTAINERS index 280a568..fbef7d0 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -10577,7 +10577,7 @@ L: linux-s390@vger.kernel.org L: virtualization@lists.linux-foundation.org L: kvm@vger.kernel.org S: Supported -F: drivers/s390/kvm/ +F: drivers/s390/virtio/ VIRTIO HOST (VHOST) M: "Michael S. Tsirkin" diff --git a/drivers/s390/Makefile b/drivers/s390/Makefile index 95bccfd..e5225ad 100644 --- a/drivers/s390/Makefile +++ b/drivers/s390/Makefile @@ -2,7 +2,7 @@ # Makefile for the S/390 specific device drivers # -obj-y += cio/ block/ char/ crypto/ net/ scsi/ kvm/ +obj-y += cio/ block/ char/ crypto/ net/ scsi/ virtio/ drivers-y += drivers/s390/built-in.o diff --git a/drivers/s390/kvm/Makefile b/drivers/s390/kvm/Makefile deleted file mode 100644 index 241891a..0000000 --- a/drivers/s390/kvm/Makefile +++ /dev/null @@ -1,9 +0,0 @@ -# Makefile for kvm guest drivers on s390 -# -# Copyright IBM Corp. 2008 -# -# This program is free software; you can redistribute it and/or modify -# it under the terms of the GNU General Public License (version 2 only) -# as published by the Free Software Foundation. - -obj-$(CONFIG_S390_GUEST) += kvm_virtio.o virtio_ccw.o diff --git a/drivers/s390/kvm/kvm_virtio.c b/drivers/s390/kvm/kvm_virtio.c deleted file mode 100644 index dd65c8b..0000000 --- a/drivers/s390/kvm/kvm_virtio.c +++ /dev/null @@ -1,510 +0,0 @@ -/* - * virtio for kvm on s390 - * - * Copyright IBM Corp. 2008 - * - * This program is free software; you can redistribute it and/or modify - * it under the terms of the GNU General Public License (version 2 only) - * as published by the Free Software Foundation. - * - * Author(s): Christian Borntraeger - */ - -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include - -#define VIRTIO_SUBCODE_64 0x0D00 - -/* - * The pointer to our (page) of device descriptions. - */ -static void *kvm_devices; -static struct work_struct hotplug_work; - -struct kvm_device { - struct virtio_device vdev; - struct kvm_device_desc *desc; -}; - -#define to_kvmdev(vd) container_of(vd, struct kvm_device, vdev) - -/* - * memory layout: - * - kvm_device_descriptor - * struct kvm_device_desc - * - configuration - * struct kvm_vqconfig - * - feature bits - * - config space - */ -static struct kvm_vqconfig *kvm_vq_config(const struct kvm_device_desc *desc) -{ - return (struct kvm_vqconfig *)(desc + 1); -} - -static u8 *kvm_vq_features(const struct kvm_device_desc *desc) -{ - return (u8 *)(kvm_vq_config(desc) + desc->num_vq); -} - -static u8 *kvm_vq_configspace(const struct kvm_device_desc *desc) -{ - return kvm_vq_features(desc) + desc->feature_len * 2; -} - -/* - * The total size of the config page used by this device (incl. desc) - */ -static unsigned desc_size(const struct kvm_device_desc *desc) -{ - return sizeof(*desc) - + desc->num_vq * sizeof(struct kvm_vqconfig) - + desc->feature_len * 2 - + desc->config_len; -} - -/* This gets the device's feature bits. */ -static u64 kvm_get_features(struct virtio_device *vdev) -{ - unsigned int i; - u32 features = 0; - struct kvm_device_desc *desc = to_kvmdev(vdev)->desc; - u8 *in_features = kvm_vq_features(desc); - - for (i = 0; i < min(desc->feature_len * 8, 32); i++) - if (in_features[i / 8] & (1 << (i % 8))) - features |= (1 << i); - return features; -} - -static int kvm_finalize_features(struct virtio_device *vdev) -{ - unsigned int i, bits; - struct kvm_device_desc *desc = to_kvmdev(vdev)->desc; - /* Second half of bitmap is features we accept. */ - u8 *out_features = kvm_vq_features(desc) + desc->feature_len; - - /* Give virtio_ring a chance to accept features. */ - vring_transport_features(vdev); - - /* Make sure we don't have any features > 32 bits! */ - BUG_ON((u32)vdev->features != vdev->features); - - memset(out_features, 0, desc->feature_len); - bits = min_t(unsigned, desc->feature_len, sizeof(vdev->features)) * 8; - for (i = 0; i < bits; i++) { - if (__virtio_test_bit(vdev, i)) - out_features[i / 8] |= (1 << (i % 8)); - } - - return 0; -} - -/* - * Reading and writing elements in config space - */ -static void kvm_get(struct virtio_device *vdev, unsigned int offset, - void *buf, unsigned len) -{ - struct kvm_device_desc *desc = to_kvmdev(vdev)->desc; - - BUG_ON(offset + len > desc->config_len); - memcpy(buf, kvm_vq_configspace(desc) + offset, len); -} - -static void kvm_set(struct virtio_device *vdev, unsigned int offset, - const void *buf, unsigned len) -{ - struct kvm_device_desc *desc = to_kvmdev(vdev)->desc; - - BUG_ON(offset + len > desc->config_len); - memcpy(kvm_vq_configspace(desc) + offset, buf, len); -} - -/* - * The operations to get and set the status word just access - * the status field of the device descriptor. set_status will also - * make a hypercall to the host, to tell about status changes - */ -static u8 kvm_get_status(struct virtio_device *vdev) -{ - return to_kvmdev(vdev)->desc->status; -} - -static void kvm_set_status(struct virtio_device *vdev, u8 status) -{ - BUG_ON(!status); - to_kvmdev(vdev)->desc->status = status; - kvm_hypercall1(KVM_S390_VIRTIO_SET_STATUS, - (unsigned long) to_kvmdev(vdev)->desc); -} - -/* - * To reset the device, we use the KVM_VIRTIO_RESET hypercall, using the - * descriptor address. The Host will zero the status and all the - * features. - */ -static void kvm_reset(struct virtio_device *vdev) -{ - kvm_hypercall1(KVM_S390_VIRTIO_RESET, - (unsigned long) to_kvmdev(vdev)->desc); -} - -/* - * When the virtio_ring code wants to notify the Host, it calls us here and we - * make a hypercall. We hand the address of the virtqueue so the Host - * knows which virtqueue we're talking about. - */ -static bool kvm_notify(struct virtqueue *vq) -{ - long rc; - struct kvm_vqconfig *config = vq->priv; - - rc = kvm_hypercall1(KVM_S390_VIRTIO_NOTIFY, config->address); - if (rc < 0) - return false; - return true; -} - -/* - * This routine finds the first virtqueue described in the configuration of - * this device and sets it up. - */ -static struct virtqueue *kvm_find_vq(struct virtio_device *vdev, - unsigned index, - void (*callback)(struct virtqueue *vq), - const char *name) -{ - struct kvm_device *kdev = to_kvmdev(vdev); - struct kvm_vqconfig *config; - struct virtqueue *vq; - int err; - - if (index >= kdev->desc->num_vq) - return ERR_PTR(-ENOENT); - - if (!name) - return NULL; - - config = kvm_vq_config(kdev->desc)+index; - - err = vmem_add_mapping(config->address, - vring_size(config->num, - KVM_S390_VIRTIO_RING_ALIGN)); - if (err) - goto out; - - vq = vring_new_virtqueue(index, config->num, KVM_S390_VIRTIO_RING_ALIGN, - vdev, true, (void *) config->address, - kvm_notify, callback, name); - if (!vq) { - err = -ENOMEM; - goto unmap; - } - - /* - * register a callback token - * The host will sent this via the external interrupt parameter - */ - config->token = (u64) vq; - - vq->priv = config; - return vq; -unmap: - vmem_remove_mapping(config->address, - vring_size(config->num, - KVM_S390_VIRTIO_RING_ALIGN)); -out: - return ERR_PTR(err); -} - -static void kvm_del_vq(struct virtqueue *vq) -{ - struct kvm_vqconfig *config = vq->priv; - - vring_del_virtqueue(vq); - vmem_remove_mapping(config->address, - vring_size(config->num, - KVM_S390_VIRTIO_RING_ALIGN)); -} - -static void kvm_del_vqs(struct virtio_device *vdev) -{ - struct virtqueue *vq, *n; - - list_for_each_entry_safe(vq, n, &vdev->vqs, list) - kvm_del_vq(vq); -} - -static int kvm_find_vqs(struct virtio_device *vdev, unsigned nvqs, - struct virtqueue *vqs[], - vq_callback_t *callbacks[], - const char *names[]) -{ - struct kvm_device *kdev = to_kvmdev(vdev); - int i; - - /* We must have this many virtqueues. */ - if (nvqs > kdev->desc->num_vq) - return -ENOENT; - - for (i = 0; i < nvqs; ++i) { - vqs[i] = kvm_find_vq(vdev, i, callbacks[i], names[i]); - if (IS_ERR(vqs[i])) - goto error; - } - return 0; - -error: - kvm_del_vqs(vdev); - return PTR_ERR(vqs[i]); -} - -static const char *kvm_bus_name(struct virtio_device *vdev) -{ - return ""; -} - -/* - * The config ops structure as defined by virtio config - */ -static const struct virtio_config_ops kvm_vq_configspace_ops = { - .get_features = kvm_get_features, - .finalize_features = kvm_finalize_features, - .get = kvm_get, - .set = kvm_set, - .get_status = kvm_get_status, - .set_status = kvm_set_status, - .reset = kvm_reset, - .find_vqs = kvm_find_vqs, - .del_vqs = kvm_del_vqs, - .bus_name = kvm_bus_name, -}; - -/* - * The root device for the kvm virtio devices. - * This makes them appear as /sys/devices/kvm_s390/0,1,2 not /sys/devices/0,1,2. - */ -static struct device *kvm_root; - -/* - * adds a new device and register it with virtio - * appropriate drivers are loaded by the device model - */ -static void add_kvm_device(struct kvm_device_desc *d, unsigned int offset) -{ - struct kvm_device *kdev; - - kdev = kzalloc(sizeof(*kdev), GFP_KERNEL); - if (!kdev) { - printk(KERN_EMERG "Cannot allocate kvm dev %u type %u\n", - offset, d->type); - return; - } - - kdev->vdev.dev.parent = kvm_root; - kdev->vdev.id.device = d->type; - kdev->vdev.config = &kvm_vq_configspace_ops; - kdev->desc = d; - - if (register_virtio_device(&kdev->vdev) != 0) { - printk(KERN_ERR "Failed to register kvm device %u type %u\n", - offset, d->type); - kfree(kdev); - } -} - -/* - * scan_devices() simply iterates through the device page. - * The type 0 is reserved to mean "end of devices". - */ -static void scan_devices(void) -{ - unsigned int i; - struct kvm_device_desc *d; - - for (i = 0; i < PAGE_SIZE; i += desc_size(d)) { - d = kvm_devices + i; - - if (d->type == 0) - break; - - add_kvm_device(d, i); - } -} - -/* - * match for a kvm device with a specific desc pointer - */ -static int match_desc(struct device *dev, void *data) -{ - struct virtio_device *vdev = dev_to_virtio(dev); - struct kvm_device *kdev = to_kvmdev(vdev); - - return kdev->desc == data; -} - -/* - * hotplug_device tries to find changes in the device page. - */ -static void hotplug_devices(struct work_struct *dummy) -{ - unsigned int i; - struct kvm_device_desc *d; - struct device *dev; - - for (i = 0; i < PAGE_SIZE; i += desc_size(d)) { - d = kvm_devices + i; - - /* end of list */ - if (d->type == 0) - break; - - /* device already exists */ - dev = device_find_child(kvm_root, d, match_desc); - if (dev) { - /* XXX check for hotplug remove */ - put_device(dev); - continue; - } - - /* new device */ - printk(KERN_INFO "Adding new virtio device %p\n", d); - add_kvm_device(d, i); - } -} - -/* - * we emulate the request_irq behaviour on top of s390 extints - */ -static void kvm_extint_handler(struct ext_code ext_code, - unsigned int param32, unsigned long param64) -{ - struct virtqueue *vq; - u32 param; - - if ((ext_code.subcode & 0xff00) != VIRTIO_SUBCODE_64) - return; - inc_irq_stat(IRQEXT_VRT); - - /* The LSB might be overloaded, we have to mask it */ - vq = (struct virtqueue *)(param64 & ~1UL); - - /* We use ext_params to decide what this interrupt means */ - param = param32 & VIRTIO_PARAM_MASK; - - switch (param) { - case VIRTIO_PARAM_CONFIG_CHANGED: - virtio_config_changed(vq->vdev); - break; - case VIRTIO_PARAM_DEV_ADD: - schedule_work(&hotplug_work); - break; - case VIRTIO_PARAM_VRING_INTERRUPT: - default: - vring_interrupt(0, vq); - break; - } -} - -/* - * For s390-virtio, we expect a page above main storage containing - * the virtio configuration. Try to actually load from this area - * in order to figure out if the host provides this page. - */ -static int __init test_devices_support(unsigned long addr) -{ - int ret = -EIO; - - asm volatile( - "0: lura 0,%1\n" - "1: xgr %0,%0\n" - "2:\n" - EX_TABLE(0b,2b) - EX_TABLE(1b,2b) - : "+d" (ret) - : "a" (addr) - : "0", "cc"); - return ret; -} -/* - * Init function for virtio - * devices are in a single page above top of "normal" + standby mem - */ -static int __init kvm_devices_init(void) -{ - int rc; - unsigned long total_memory_size = sclp_get_rzm() * sclp_get_rnmax(); - - if (!MACHINE_IS_KVM) - return -ENODEV; - - if (test_devices_support(total_memory_size) < 0) - return -ENODEV; - - rc = vmem_add_mapping(total_memory_size, PAGE_SIZE); - if (rc) - return rc; - - kvm_devices = (void *) total_memory_size; - - kvm_root = root_device_register("kvm_s390"); - if (IS_ERR(kvm_root)) { - rc = PTR_ERR(kvm_root); - printk(KERN_ERR "Could not register kvm_s390 root device"); - vmem_remove_mapping(total_memory_size, PAGE_SIZE); - return rc; - } - - INIT_WORK(&hotplug_work, hotplug_devices); - - irq_subclass_register(IRQ_SUBCLASS_SERVICE_SIGNAL); - register_external_irq(EXT_IRQ_CP_SERVICE, kvm_extint_handler); - - scan_devices(); - return 0; -} - -/* code for early console output with virtio_console */ -static __init int early_put_chars(u32 vtermno, const char *buf, int count) -{ - char scratch[17]; - unsigned int len = count; - - if (len > sizeof(scratch) - 1) - len = sizeof(scratch) - 1; - scratch[len] = '\0'; - memcpy(scratch, buf, len); - kvm_hypercall1(KVM_S390_VIRTIO_NOTIFY, __pa(scratch)); - return len; -} - -static int __init s390_virtio_console_init(void) -{ - if (sclp_has_vt220() || sclp_has_linemode()) - return -ENODEV; - return virtio_cons_early_init(early_put_chars); -} -console_initcall(s390_virtio_console_init); - - -/* - * We do this after core stuff, but before the drivers. - */ -postcore_initcall(kvm_devices_init); diff --git a/drivers/s390/kvm/virtio_ccw.c b/drivers/s390/kvm/virtio_ccw.c deleted file mode 100644 index 6f1fa17..0000000 --- a/drivers/s390/kvm/virtio_ccw.c +++ /dev/null @@ -1,1380 +0,0 @@ -/* - * ccw based virtio transport - * - * Copyright IBM Corp. 2012, 2014 - * - * This program is free software; you can redistribute it and/or modify - * it under the terms of the GNU General Public License (version 2 only) - * as published by the Free Software Foundation. - * - * Author(s): Cornelia Huck - */ - -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include - -/* - * virtio related functions - */ - -struct vq_config_block { - __u16 index; - __u16 num; -} __packed; - -#define VIRTIO_CCW_CONFIG_SIZE 0x100 -/* same as PCI config space size, should be enough for all drivers */ - -struct virtio_ccw_device { - struct virtio_device vdev; - __u8 *status; - __u8 config[VIRTIO_CCW_CONFIG_SIZE]; - struct ccw_device *cdev; - __u32 curr_io; - int err; - unsigned int revision; /* Transport revision */ - wait_queue_head_t wait_q; - spinlock_t lock; - struct list_head virtqueues; - unsigned long indicators; - unsigned long indicators2; - struct vq_config_block *config_block; - bool is_thinint; - bool going_away; - bool device_lost; - void *airq_info; -}; - -struct vq_info_block_legacy { - __u64 queue; - __u32 align; - __u16 index; - __u16 num; -} __packed; - -struct vq_info_block { - __u64 desc; - __u32 res0; - __u16 index; - __u16 num; - __u64 avail; - __u64 used; -} __packed; - -struct virtio_feature_desc { - __u32 features; - __u8 index; -} __packed; - -struct virtio_thinint_area { - unsigned long summary_indicator; - unsigned long indicator; - u64 bit_nr; - u8 isc; -} __packed; - -struct virtio_rev_info { - __u16 revision; - __u16 length; - __u8 data[]; -}; - -/* the highest virtio-ccw revision we support */ -#define VIRTIO_CCW_REV_MAX 1 - -struct virtio_ccw_vq_info { - struct virtqueue *vq; - int num; - void *queue; - union { - struct vq_info_block s; - struct vq_info_block_legacy l; - } *info_block; - int bit_nr; - struct list_head node; - long cookie; -}; - -#define VIRTIO_AIRQ_ISC IO_SCH_ISC /* inherit from subchannel */ - -#define VIRTIO_IV_BITS (L1_CACHE_BYTES * 8) -#define MAX_AIRQ_AREAS 20 - -static int virtio_ccw_use_airq = 1; - -struct airq_info { - rwlock_t lock; - u8 summary_indicator; - struct airq_struct airq; - struct airq_iv *aiv; -}; -static struct airq_info *airq_areas[MAX_AIRQ_AREAS]; - -#define CCW_CMD_SET_VQ 0x13 -#define CCW_CMD_VDEV_RESET 0x33 -#define CCW_CMD_SET_IND 0x43 -#define CCW_CMD_SET_CONF_IND 0x53 -#define CCW_CMD_READ_FEAT 0x12 -#define CCW_CMD_WRITE_FEAT 0x11 -#define CCW_CMD_READ_CONF 0x22 -#define CCW_CMD_WRITE_CONF 0x21 -#define CCW_CMD_WRITE_STATUS 0x31 -#define CCW_CMD_READ_VQ_CONF 0x32 -#define CCW_CMD_SET_IND_ADAPTER 0x73 -#define CCW_CMD_SET_VIRTIO_REV 0x83 - -#define VIRTIO_CCW_DOING_SET_VQ 0x00010000 -#define VIRTIO_CCW_DOING_RESET 0x00040000 -#define VIRTIO_CCW_DOING_READ_FEAT 0x00080000 -#define VIRTIO_CCW_DOING_WRITE_FEAT 0x00100000 -#define VIRTIO_CCW_DOING_READ_CONFIG 0x00200000 -#define VIRTIO_CCW_DOING_WRITE_CONFIG 0x00400000 -#define VIRTIO_CCW_DOING_WRITE_STATUS 0x00800000 -#define VIRTIO_CCW_DOING_SET_IND 0x01000000 -#define VIRTIO_CCW_DOING_READ_VQ_CONF 0x02000000 -#define VIRTIO_CCW_DOING_SET_CONF_IND 0x04000000 -#define VIRTIO_CCW_DOING_SET_IND_ADAPTER 0x08000000 -#define VIRTIO_CCW_DOING_SET_VIRTIO_REV 0x10000000 -#define VIRTIO_CCW_INTPARM_MASK 0xffff0000 - -static struct virtio_ccw_device *to_vc_device(struct virtio_device *vdev) -{ - return container_of(vdev, struct virtio_ccw_device, vdev); -} - -static void drop_airq_indicator(struct virtqueue *vq, struct airq_info *info) -{ - unsigned long i, flags; - - write_lock_irqsave(&info->lock, flags); - for (i = 0; i < airq_iv_end(info->aiv); i++) { - if (vq == (void *)airq_iv_get_ptr(info->aiv, i)) { - airq_iv_free_bit(info->aiv, i); - airq_iv_set_ptr(info->aiv, i, 0); - break; - } - } - write_unlock_irqrestore(&info->lock, flags); -} - -static void virtio_airq_handler(struct airq_struct *airq) -{ - struct airq_info *info = container_of(airq, struct airq_info, airq); - unsigned long ai; - - inc_irq_stat(IRQIO_VAI); - read_lock(&info->lock); - /* Walk through indicators field, summary indicator active. */ - for (ai = 0;;) { - ai = airq_iv_scan(info->aiv, ai, airq_iv_end(info->aiv)); - if (ai == -1UL) - break; - vring_interrupt(0, (void *)airq_iv_get_ptr(info->aiv, ai)); - } - info->summary_indicator = 0; - smp_wmb(); - /* Walk through indicators field, summary indicator not active. */ - for (ai = 0;;) { - ai = airq_iv_scan(info->aiv, ai, airq_iv_end(info->aiv)); - if (ai == -1UL) - break; - vring_interrupt(0, (void *)airq_iv_get_ptr(info->aiv, ai)); - } - read_unlock(&info->lock); -} - -static struct airq_info *new_airq_info(void) -{ - struct airq_info *info; - int rc; - - info = kzalloc(sizeof(*info), GFP_KERNEL); - if (!info) - return NULL; - rwlock_init(&info->lock); - info->aiv = airq_iv_create(VIRTIO_IV_BITS, AIRQ_IV_ALLOC | AIRQ_IV_PTR); - if (!info->aiv) { - kfree(info); - return NULL; - } - info->airq.handler = virtio_airq_handler; - info->airq.lsi_ptr = &info->summary_indicator; - info->airq.lsi_mask = 0xff; - info->airq.isc = VIRTIO_AIRQ_ISC; - rc = register_adapter_interrupt(&info->airq); - if (rc) { - airq_iv_release(info->aiv); - kfree(info); - return NULL; - } - return info; -} - -static void destroy_airq_info(struct airq_info *info) -{ - if (!info) - return; - - unregister_adapter_interrupt(&info->airq); - airq_iv_release(info->aiv); - kfree(info); -} - -static unsigned long get_airq_indicator(struct virtqueue *vqs[], int nvqs, - u64 *first, void **airq_info) -{ - int i, j; - struct airq_info *info; - unsigned long indicator_addr = 0; - unsigned long bit, flags; - - for (i = 0; i < MAX_AIRQ_AREAS && !indicator_addr; i++) { - if (!airq_areas[i]) - airq_areas[i] = new_airq_info(); - info = airq_areas[i]; - if (!info) - return 0; - write_lock_irqsave(&info->lock, flags); - bit = airq_iv_alloc(info->aiv, nvqs); - if (bit == -1UL) { - /* Not enough vacancies. */ - write_unlock_irqrestore(&info->lock, flags); - continue; - } - *first = bit; - *airq_info = info; - indicator_addr = (unsigned long)info->aiv->vector; - for (j = 0; j < nvqs; j++) { - airq_iv_set_ptr(info->aiv, bit + j, - (unsigned long)vqs[j]); - } - write_unlock_irqrestore(&info->lock, flags); - } - return indicator_addr; -} - -static void virtio_ccw_drop_indicators(struct virtio_ccw_device *vcdev) -{ - struct virtio_ccw_vq_info *info; - - list_for_each_entry(info, &vcdev->virtqueues, node) - drop_airq_indicator(info->vq, vcdev->airq_info); -} - -static int doing_io(struct virtio_ccw_device *vcdev, __u32 flag) -{ - unsigned long flags; - __u32 ret; - - spin_lock_irqsave(get_ccwdev_lock(vcdev->cdev), flags); - if (vcdev->err) - ret = 0; - else - ret = vcdev->curr_io & flag; - spin_unlock_irqrestore(get_ccwdev_lock(vcdev->cdev), flags); - return ret; -} - -static int ccw_io_helper(struct virtio_ccw_device *vcdev, - struct ccw1 *ccw, __u32 intparm) -{ - int ret; - unsigned long flags; - int flag = intparm & VIRTIO_CCW_INTPARM_MASK; - - do { - spin_lock_irqsave(get_ccwdev_lock(vcdev->cdev), flags); - ret = ccw_device_start(vcdev->cdev, ccw, intparm, 0, 0); - if (!ret) { - if (!vcdev->curr_io) - vcdev->err = 0; - vcdev->curr_io |= flag; - } - spin_unlock_irqrestore(get_ccwdev_lock(vcdev->cdev), flags); - cpu_relax(); - } while (ret == -EBUSY); - wait_event(vcdev->wait_q, doing_io(vcdev, flag) == 0); - return ret ? ret : vcdev->err; -} - -static void virtio_ccw_drop_indicator(struct virtio_ccw_device *vcdev, - struct ccw1 *ccw) -{ - int ret; - unsigned long *indicatorp = NULL; - struct virtio_thinint_area *thinint_area = NULL; - struct airq_info *airq_info = vcdev->airq_info; - - if (vcdev->is_thinint) { - thinint_area = kzalloc(sizeof(*thinint_area), - GFP_DMA | GFP_KERNEL); - if (!thinint_area) - return; - thinint_area->summary_indicator = - (unsigned long) &airq_info->summary_indicator; - thinint_area->isc = VIRTIO_AIRQ_ISC; - ccw->cmd_code = CCW_CMD_SET_IND_ADAPTER; - ccw->count = sizeof(*thinint_area); - ccw->cda = (__u32)(unsigned long) thinint_area; - } else { - indicatorp = kmalloc(sizeof(&vcdev->indicators), - GFP_DMA | GFP_KERNEL); - if (!indicatorp) - return; - *indicatorp = 0; - ccw->cmd_code = CCW_CMD_SET_IND; - ccw->count = sizeof(vcdev->indicators); - ccw->cda = (__u32)(unsigned long) indicatorp; - } - /* Deregister indicators from host. */ - vcdev->indicators = 0; - ccw->flags = 0; - ret = ccw_io_helper(vcdev, ccw, - vcdev->is_thinint ? - VIRTIO_CCW_DOING_SET_IND_ADAPTER : - VIRTIO_CCW_DOING_SET_IND); - if (ret && (ret != -ENODEV)) - dev_info(&vcdev->cdev->dev, - "Failed to deregister indicators (%d)\n", ret); - else if (vcdev->is_thinint) - virtio_ccw_drop_indicators(vcdev); - kfree(indicatorp); - kfree(thinint_area); -} - -static inline long do_kvm_notify(struct subchannel_id schid, - unsigned long queue_index, - long cookie) -{ - register unsigned long __nr asm("1") = KVM_S390_VIRTIO_CCW_NOTIFY; - register struct subchannel_id __schid asm("2") = schid; - register unsigned long __index asm("3") = queue_index; - register long __rc asm("2"); - register long __cookie asm("4") = cookie; - - asm volatile ("diag 2,4,0x500\n" - : "=d" (__rc) : "d" (__nr), "d" (__schid), "d" (__index), - "d"(__cookie) - : "memory", "cc"); - return __rc; -} - -static bool virtio_ccw_kvm_notify(struct virtqueue *vq) -{ - struct virtio_ccw_vq_info *info = vq->priv; - struct virtio_ccw_device *vcdev; - struct subchannel_id schid; - - vcdev = to_vc_device(info->vq->vdev); - ccw_device_get_schid(vcdev->cdev, &schid); - info->cookie = do_kvm_notify(schid, vq->index, info->cookie); - if (info->cookie < 0) - return false; - return true; -} - -static int virtio_ccw_read_vq_conf(struct virtio_ccw_device *vcdev, - struct ccw1 *ccw, int index) -{ - vcdev->config_block->index = index; - ccw->cmd_code = CCW_CMD_READ_VQ_CONF; - ccw->flags = 0; - ccw->count = sizeof(struct vq_config_block); - ccw->cda = (__u32)(unsigned long)(vcdev->config_block); - ccw_io_helper(vcdev, ccw, VIRTIO_CCW_DOING_READ_VQ_CONF); - return vcdev->config_block->num; -} - -static void virtio_ccw_del_vq(struct virtqueue *vq, struct ccw1 *ccw) -{ - struct virtio_ccw_device *vcdev = to_vc_device(vq->vdev); - struct virtio_ccw_vq_info *info = vq->priv; - unsigned long flags; - unsigned long size; - int ret; - unsigned int index = vq->index; - - /* Remove from our list. */ - spin_lock_irqsave(&vcdev->lock, flags); - list_del(&info->node); - spin_unlock_irqrestore(&vcdev->lock, flags); - - /* Release from host. */ - if (vcdev->revision == 0) { - info->info_block->l.queue = 0; - info->info_block->l.align = 0; - info->info_block->l.index = index; - info->info_block->l.num = 0; - ccw->count = sizeof(info->info_block->l); - } else { - info->info_block->s.desc = 0; - info->info_block->s.index = index; - info->info_block->s.num = 0; - info->info_block->s.avail = 0; - info->info_block->s.used = 0; - ccw->count = sizeof(info->info_block->s); - } - ccw->cmd_code = CCW_CMD_SET_VQ; - ccw->flags = 0; - ccw->cda = (__u32)(unsigned long)(info->info_block); - ret = ccw_io_helper(vcdev, ccw, - VIRTIO_CCW_DOING_SET_VQ | index); - /* - * -ENODEV isn't considered an error: The device is gone anyway. - * This may happen on device detach. - */ - if (ret && (ret != -ENODEV)) - dev_warn(&vq->vdev->dev, "Error %d while deleting queue %d", - ret, index); - - vring_del_virtqueue(vq); - size = PAGE_ALIGN(vring_size(info->num, KVM_VIRTIO_CCW_RING_ALIGN)); - free_pages_exact(info->queue, size); - kfree(info->info_block); - kfree(info); -} - -static void virtio_ccw_del_vqs(struct virtio_device *vdev) -{ - struct virtqueue *vq, *n; - struct ccw1 *ccw; - struct virtio_ccw_device *vcdev = to_vc_device(vdev); - - ccw = kzalloc(sizeof(*ccw), GFP_DMA | GFP_KERNEL); - if (!ccw) - return; - - virtio_ccw_drop_indicator(vcdev, ccw); - - list_for_each_entry_safe(vq, n, &vdev->vqs, list) - virtio_ccw_del_vq(vq, ccw); - - kfree(ccw); -} - -static struct virtqueue *virtio_ccw_setup_vq(struct virtio_device *vdev, - int i, vq_callback_t *callback, - const char *name, - struct ccw1 *ccw) -{ - struct virtio_ccw_device *vcdev = to_vc_device(vdev); - int err; - struct virtqueue *vq = NULL; - struct virtio_ccw_vq_info *info; - unsigned long size = 0; /* silence the compiler */ - unsigned long flags; - - /* Allocate queue. */ - info = kzalloc(sizeof(struct virtio_ccw_vq_info), GFP_KERNEL); - if (!info) { - dev_warn(&vcdev->cdev->dev, "no info\n"); - err = -ENOMEM; - goto out_err; - } - info->info_block = kzalloc(sizeof(*info->info_block), - GFP_DMA | GFP_KERNEL); - if (!info->info_block) { - dev_warn(&vcdev->cdev->dev, "no info block\n"); - err = -ENOMEM; - goto out_err; - } - info->num = virtio_ccw_read_vq_conf(vcdev, ccw, i); - size = PAGE_ALIGN(vring_size(info->num, KVM_VIRTIO_CCW_RING_ALIGN)); - info->queue = alloc_pages_exact(size, GFP_KERNEL | __GFP_ZERO); - if (info->queue == NULL) { - dev_warn(&vcdev->cdev->dev, "no queue\n"); - err = -ENOMEM; - goto out_err; - } - - vq = vring_new_virtqueue(i, info->num, KVM_VIRTIO_CCW_RING_ALIGN, vdev, - true, info->queue, virtio_ccw_kvm_notify, - callback, name); - if (!vq) { - /* For now, we fail if we can't get the requested size. */ - dev_warn(&vcdev->cdev->dev, "no vq\n"); - err = -ENOMEM; - goto out_err; - } - - /* Register it with the host. */ - if (vcdev->revision == 0) { - info->info_block->l.queue = (__u64)info->queue; - info->info_block->l.align = KVM_VIRTIO_CCW_RING_ALIGN; - info->info_block->l.index = i; - info->info_block->l.num = info->num; - ccw->count = sizeof(info->info_block->l); - } else { - info->info_block->s.desc = (__u64)info->queue; - info->info_block->s.index = i; - info->info_block->s.num = info->num; - info->info_block->s.avail = (__u64)virtqueue_get_avail(vq); - info->info_block->s.used = (__u64)virtqueue_get_used(vq); - ccw->count = sizeof(info->info_block->s); - } - ccw->cmd_code = CCW_CMD_SET_VQ; - ccw->flags = 0; - ccw->cda = (__u32)(unsigned long)(info->info_block); - err = ccw_io_helper(vcdev, ccw, VIRTIO_CCW_DOING_SET_VQ | i); - if (err) { - dev_warn(&vcdev->cdev->dev, "SET_VQ failed\n"); - goto out_err; - } - - info->vq = vq; - vq->priv = info; - - /* Save it to our list. */ - spin_lock_irqsave(&vcdev->lock, flags); - list_add(&info->node, &vcdev->virtqueues); - spin_unlock_irqrestore(&vcdev->lock, flags); - - return vq; - -out_err: - if (vq) - vring_del_virtqueue(vq); - if (info) { - if (info->queue) - free_pages_exact(info->queue, size); - kfree(info->info_block); - } - kfree(info); - return ERR_PTR(err); -} - -static int virtio_ccw_register_adapter_ind(struct virtio_ccw_device *vcdev, - struct virtqueue *vqs[], int nvqs, - struct ccw1 *ccw) -{ - int ret; - struct virtio_thinint_area *thinint_area = NULL; - struct airq_info *info; - - thinint_area = kzalloc(sizeof(*thinint_area), GFP_DMA | GFP_KERNEL); - if (!thinint_area) { - ret = -ENOMEM; - goto out; - } - /* Try to get an indicator. */ - thinint_area->indicator = get_airq_indicator(vqs, nvqs, - &thinint_area->bit_nr, - &vcdev->airq_info); - if (!thinint_area->indicator) { - ret = -ENOSPC; - goto out; - } - info = vcdev->airq_info; - thinint_area->summary_indicator = - (unsigned long) &info->summary_indicator; - thinint_area->isc = VIRTIO_AIRQ_ISC; - ccw->cmd_code = CCW_CMD_SET_IND_ADAPTER; - ccw->flags = CCW_FLAG_SLI; - ccw->count = sizeof(*thinint_area); - ccw->cda = (__u32)(unsigned long)thinint_area; - ret = ccw_io_helper(vcdev, ccw, VIRTIO_CCW_DOING_SET_IND_ADAPTER); - if (ret) { - if (ret == -EOPNOTSUPP) { - /* - * The host does not support adapter interrupts - * for virtio-ccw, stop trying. - */ - virtio_ccw_use_airq = 0; - pr_info("Adapter interrupts unsupported on host\n"); - } else - dev_warn(&vcdev->cdev->dev, - "enabling adapter interrupts = %d\n", ret); - virtio_ccw_drop_indicators(vcdev); - } -out: - kfree(thinint_area); - return ret; -} - -static int virtio_ccw_find_vqs(struct virtio_device *vdev, unsigned nvqs, - struct virtqueue *vqs[], - vq_callback_t *callbacks[], - const char *names[]) -{ - struct virtio_ccw_device *vcdev = to_vc_device(vdev); - unsigned long *indicatorp = NULL; - int ret, i; - struct ccw1 *ccw; - - ccw = kzalloc(sizeof(*ccw), GFP_DMA | GFP_KERNEL); - if (!ccw) - return -ENOMEM; - - for (i = 0; i < nvqs; ++i) { - vqs[i] = virtio_ccw_setup_vq(vdev, i, callbacks[i], names[i], - ccw); - if (IS_ERR(vqs[i])) { - ret = PTR_ERR(vqs[i]); - vqs[i] = NULL; - goto out; - } - } - ret = -ENOMEM; - /* We need a data area under 2G to communicate. */ - indicatorp = kmalloc(sizeof(&vcdev->indicators), GFP_DMA | GFP_KERNEL); - if (!indicatorp) - goto out; - *indicatorp = (unsigned long) &vcdev->indicators; - if (vcdev->is_thinint) { - ret = virtio_ccw_register_adapter_ind(vcdev, vqs, nvqs, ccw); - if (ret) - /* no error, just fall back to legacy interrupts */ - vcdev->is_thinint = 0; - } - if (!vcdev->is_thinint) { - /* Register queue indicators with host. */ - vcdev->indicators = 0; - ccw->cmd_code = CCW_CMD_SET_IND; - ccw->flags = 0; - ccw->count = sizeof(vcdev->indicators); - ccw->cda = (__u32)(unsigned long) indicatorp; - ret = ccw_io_helper(vcdev, ccw, VIRTIO_CCW_DOING_SET_IND); - if (ret) - goto out; - } - /* Register indicators2 with host for config changes */ - *indicatorp = (unsigned long) &vcdev->indicators2; - vcdev->indicators2 = 0; - ccw->cmd_code = CCW_CMD_SET_CONF_IND; - ccw->flags = 0; - ccw->count = sizeof(vcdev->indicators2); - ccw->cda = (__u32)(unsigned long) indicatorp; - ret = ccw_io_helper(vcdev, ccw, VIRTIO_CCW_DOING_SET_CONF_IND); - if (ret) - goto out; - - kfree(indicatorp); - kfree(ccw); - return 0; -out: - kfree(indicatorp); - kfree(ccw); - virtio_ccw_del_vqs(vdev); - return ret; -} - -static void virtio_ccw_reset(struct virtio_device *vdev) -{ - struct virtio_ccw_device *vcdev = to_vc_device(vdev); - struct ccw1 *ccw; - - ccw = kzalloc(sizeof(*ccw), GFP_DMA | GFP_KERNEL); - if (!ccw) - return; - - /* Zero status bits. */ - *vcdev->status = 0; - - /* Send a reset ccw on device. */ - ccw->cmd_code = CCW_CMD_VDEV_RESET; - ccw->flags = 0; - ccw->count = 0; - ccw->cda = 0; - ccw_io_helper(vcdev, ccw, VIRTIO_CCW_DOING_RESET); - kfree(ccw); -} - -static u64 virtio_ccw_get_features(struct virtio_device *vdev) -{ - struct virtio_ccw_device *vcdev = to_vc_device(vdev); - struct virtio_feature_desc *features; - int ret; - u64 rc; - struct ccw1 *ccw; - - ccw = kzalloc(sizeof(*ccw), GFP_DMA | GFP_KERNEL); - if (!ccw) - return 0; - - features = kzalloc(sizeof(*features), GFP_DMA | GFP_KERNEL); - if (!features) { - rc = 0; - goto out_free; - } - /* Read the feature bits from the host. */ - features->index = 0; - ccw->cmd_code = CCW_CMD_READ_FEAT; - ccw->flags = 0; - ccw->count = sizeof(*features); - ccw->cda = (__u32)(unsigned long)features; - ret = ccw_io_helper(vcdev, ccw, VIRTIO_CCW_DOING_READ_FEAT); - if (ret) { - rc = 0; - goto out_free; - } - - rc = le32_to_cpu(features->features); - - if (vcdev->revision == 0) - goto out_free; - - /* Read second half of the feature bits from the host. */ - features->index = 1; - ccw->cmd_code = CCW_CMD_READ_FEAT; - ccw->flags = 0; - ccw->count = sizeof(*features); - ccw->cda = (__u32)(unsigned long)features; - ret = ccw_io_helper(vcdev, ccw, VIRTIO_CCW_DOING_READ_FEAT); - if (ret == 0) - rc |= (u64)le32_to_cpu(features->features) << 32; - -out_free: - kfree(features); - kfree(ccw); - return rc; -} - -static int virtio_ccw_finalize_features(struct virtio_device *vdev) -{ - struct virtio_ccw_device *vcdev = to_vc_device(vdev); - struct virtio_feature_desc *features; - struct ccw1 *ccw; - int ret; - - if (vcdev->revision >= 1 && - !__virtio_test_bit(vdev, VIRTIO_F_VERSION_1)) { - dev_err(&vdev->dev, "virtio: device uses revision 1 " - "but does not have VIRTIO_F_VERSION_1\n"); - return -EINVAL; - } - - ccw = kzalloc(sizeof(*ccw), GFP_DMA | GFP_KERNEL); - if (!ccw) - return -ENOMEM; - - features = kzalloc(sizeof(*features), GFP_DMA | GFP_KERNEL); - if (!features) { - ret = -ENOMEM; - goto out_free; - } - /* Give virtio_ring a chance to accept features. */ - vring_transport_features(vdev); - - features->index = 0; - features->features = cpu_to_le32((u32)vdev->features); - /* Write the first half of the feature bits to the host. */ - ccw->cmd_code = CCW_CMD_WRITE_FEAT; - ccw->flags = 0; - ccw->count = sizeof(*features); - ccw->cda = (__u32)(unsigned long)features; - ret = ccw_io_helper(vcdev, ccw, VIRTIO_CCW_DOING_WRITE_FEAT); - if (ret) - goto out_free; - - if (vcdev->revision == 0) - goto out_free; - - features->index = 1; - features->features = cpu_to_le32(vdev->features >> 32); - /* Write the second half of the feature bits to the host. */ - ccw->cmd_code = CCW_CMD_WRITE_FEAT; - ccw->flags = 0; - ccw->count = sizeof(*features); - ccw->cda = (__u32)(unsigned long)features; - ret = ccw_io_helper(vcdev, ccw, VIRTIO_CCW_DOING_WRITE_FEAT); - -out_free: - kfree(features); - kfree(ccw); - - return ret; -} - -static void virtio_ccw_get_config(struct virtio_device *vdev, - unsigned int offset, void *buf, unsigned len) -{ - struct virtio_ccw_device *vcdev = to_vc_device(vdev); - int ret; - struct ccw1 *ccw; - void *config_area; - - ccw = kzalloc(sizeof(*ccw), GFP_DMA | GFP_KERNEL); - if (!ccw) - return; - - config_area = kzalloc(VIRTIO_CCW_CONFIG_SIZE, GFP_DMA | GFP_KERNEL); - if (!config_area) - goto out_free; - - /* Read the config area from the host. */ - ccw->cmd_code = CCW_CMD_READ_CONF; - ccw->flags = 0; - ccw->count = offset + len; - ccw->cda = (__u32)(unsigned long)config_area; - ret = ccw_io_helper(vcdev, ccw, VIRTIO_CCW_DOING_READ_CONFIG); - if (ret) - goto out_free; - - memcpy(vcdev->config, config_area, sizeof(vcdev->config)); - memcpy(buf, &vcdev->config[offset], len); - -out_free: - kfree(config_area); - kfree(ccw); -} - -static void virtio_ccw_set_config(struct virtio_device *vdev, - unsigned int offset, const void *buf, - unsigned len) -{ - struct virtio_ccw_device *vcdev = to_vc_device(vdev); - struct ccw1 *ccw; - void *config_area; - - ccw = kzalloc(sizeof(*ccw), GFP_DMA | GFP_KERNEL); - if (!ccw) - return; - - config_area = kzalloc(VIRTIO_CCW_CONFIG_SIZE, GFP_DMA | GFP_KERNEL); - if (!config_area) - goto out_free; - - memcpy(&vcdev->config[offset], buf, len); - /* Write the config area to the host. */ - memcpy(config_area, vcdev->config, sizeof(vcdev->config)); - ccw->cmd_code = CCW_CMD_WRITE_CONF; - ccw->flags = 0; - ccw->count = offset + len; - ccw->cda = (__u32)(unsigned long)config_area; - ccw_io_helper(vcdev, ccw, VIRTIO_CCW_DOING_WRITE_CONFIG); - -out_free: - kfree(config_area); - kfree(ccw); -} - -static u8 virtio_ccw_get_status(struct virtio_device *vdev) -{ - struct virtio_ccw_device *vcdev = to_vc_device(vdev); - - return *vcdev->status; -} - -static void virtio_ccw_set_status(struct virtio_device *vdev, u8 status) -{ - struct virtio_ccw_device *vcdev = to_vc_device(vdev); - u8 old_status = *vcdev->status; - struct ccw1 *ccw; - int ret; - - ccw = kzalloc(sizeof(*ccw), GFP_DMA | GFP_KERNEL); - if (!ccw) - return; - - /* Write the status to the host. */ - *vcdev->status = status; - ccw->cmd_code = CCW_CMD_WRITE_STATUS; - ccw->flags = 0; - ccw->count = sizeof(status); - ccw->cda = (__u32)(unsigned long)vcdev->status; - ret = ccw_io_helper(vcdev, ccw, VIRTIO_CCW_DOING_WRITE_STATUS); - /* Write failed? We assume status is unchanged. */ - if (ret) - *vcdev->status = old_status; - kfree(ccw); -} - -static struct virtio_config_ops virtio_ccw_config_ops = { - .get_features = virtio_ccw_get_features, - .finalize_features = virtio_ccw_finalize_features, - .get = virtio_ccw_get_config, - .set = virtio_ccw_set_config, - .get_status = virtio_ccw_get_status, - .set_status = virtio_ccw_set_status, - .reset = virtio_ccw_reset, - .find_vqs = virtio_ccw_find_vqs, - .del_vqs = virtio_ccw_del_vqs, -}; - - -/* - * ccw bus driver related functions - */ - -static void virtio_ccw_release_dev(struct device *_d) -{ - struct virtio_device *dev = container_of(_d, struct virtio_device, - dev); - struct virtio_ccw_device *vcdev = to_vc_device(dev); - - kfree(vcdev->status); - kfree(vcdev->config_block); - kfree(vcdev); -} - -static int irb_is_error(struct irb *irb) -{ - if (scsw_cstat(&irb->scsw) != 0) - return 1; - if (scsw_dstat(&irb->scsw) & ~(DEV_STAT_CHN_END | DEV_STAT_DEV_END)) - return 1; - if (scsw_cc(&irb->scsw) != 0) - return 1; - return 0; -} - -static struct virtqueue *virtio_ccw_vq_by_ind(struct virtio_ccw_device *vcdev, - int index) -{ - struct virtio_ccw_vq_info *info; - unsigned long flags; - struct virtqueue *vq; - - vq = NULL; - spin_lock_irqsave(&vcdev->lock, flags); - list_for_each_entry(info, &vcdev->virtqueues, node) { - if (info->vq->index == index) { - vq = info->vq; - break; - } - } - spin_unlock_irqrestore(&vcdev->lock, flags); - return vq; -} - -static void virtio_ccw_int_handler(struct ccw_device *cdev, - unsigned long intparm, - struct irb *irb) -{ - __u32 activity = intparm & VIRTIO_CCW_INTPARM_MASK; - struct virtio_ccw_device *vcdev = dev_get_drvdata(&cdev->dev); - int i; - struct virtqueue *vq; - - if (!vcdev) - return; - /* Check if it's a notification from the host. */ - if ((intparm == 0) && - (scsw_stctl(&irb->scsw) == - (SCSW_STCTL_ALERT_STATUS | SCSW_STCTL_STATUS_PEND))) { - /* OK */ - } - if (irb_is_error(irb)) { - /* Command reject? */ - if ((scsw_dstat(&irb->scsw) & DEV_STAT_UNIT_CHECK) && - (irb->ecw[0] & SNS0_CMD_REJECT)) - vcdev->err = -EOPNOTSUPP; - else - /* Map everything else to -EIO. */ - vcdev->err = -EIO; - } - if (vcdev->curr_io & activity) { - switch (activity) { - case VIRTIO_CCW_DOING_READ_FEAT: - case VIRTIO_CCW_DOING_WRITE_FEAT: - case VIRTIO_CCW_DOING_READ_CONFIG: - case VIRTIO_CCW_DOING_WRITE_CONFIG: - case VIRTIO_CCW_DOING_WRITE_STATUS: - case VIRTIO_CCW_DOING_SET_VQ: - case VIRTIO_CCW_DOING_SET_IND: - case VIRTIO_CCW_DOING_SET_CONF_IND: - case VIRTIO_CCW_DOING_RESET: - case VIRTIO_CCW_DOING_READ_VQ_CONF: - case VIRTIO_CCW_DOING_SET_IND_ADAPTER: - case VIRTIO_CCW_DOING_SET_VIRTIO_REV: - vcdev->curr_io &= ~activity; - wake_up(&vcdev->wait_q); - break; - default: - /* don't know what to do... */ - dev_warn(&cdev->dev, "Suspicious activity '%08x'\n", - activity); - WARN_ON(1); - break; - } - } - for_each_set_bit(i, &vcdev->indicators, - sizeof(vcdev->indicators) * BITS_PER_BYTE) { - /* The bit clear must happen before the vring kick. */ - clear_bit(i, &vcdev->indicators); - barrier(); - vq = virtio_ccw_vq_by_ind(vcdev, i); - vring_interrupt(0, vq); - } - if (test_bit(0, &vcdev->indicators2)) { - virtio_config_changed(&vcdev->vdev); - clear_bit(0, &vcdev->indicators2); - } -} - -/* - * We usually want to autoonline all devices, but give the admin - * a way to exempt devices from this. - */ -#define __DEV_WORDS ((__MAX_SUBCHANNEL + (8*sizeof(long) - 1)) / \ - (8*sizeof(long))) -static unsigned long devs_no_auto[__MAX_SSID + 1][__DEV_WORDS]; - -static char *no_auto = ""; - -module_param(no_auto, charp, 0444); -MODULE_PARM_DESC(no_auto, "list of ccw bus id ranges not to be auto-onlined"); - -static int virtio_ccw_check_autoonline(struct ccw_device *cdev) -{ - struct ccw_dev_id id; - - ccw_device_get_id(cdev, &id); - if (test_bit(id.devno, devs_no_auto[id.ssid])) - return 0; - return 1; -} - -static void virtio_ccw_auto_online(void *data, async_cookie_t cookie) -{ - struct ccw_device *cdev = data; - int ret; - - ret = ccw_device_set_online(cdev); - if (ret) - dev_warn(&cdev->dev, "Failed to set online: %d\n", ret); -} - -static int virtio_ccw_probe(struct ccw_device *cdev) -{ - cdev->handler = virtio_ccw_int_handler; - - if (virtio_ccw_check_autoonline(cdev)) - async_schedule(virtio_ccw_auto_online, cdev); - return 0; -} - -static struct virtio_ccw_device *virtio_grab_drvdata(struct ccw_device *cdev) -{ - unsigned long flags; - struct virtio_ccw_device *vcdev; - - spin_lock_irqsave(get_ccwdev_lock(cdev), flags); - vcdev = dev_get_drvdata(&cdev->dev); - if (!vcdev || vcdev->going_away) { - spin_unlock_irqrestore(get_ccwdev_lock(cdev), flags); - return NULL; - } - vcdev->going_away = true; - spin_unlock_irqrestore(get_ccwdev_lock(cdev), flags); - return vcdev; -} - -static void virtio_ccw_remove(struct ccw_device *cdev) -{ - unsigned long flags; - struct virtio_ccw_device *vcdev = virtio_grab_drvdata(cdev); - - if (vcdev && cdev->online) { - if (vcdev->device_lost) - virtio_break_device(&vcdev->vdev); - unregister_virtio_device(&vcdev->vdev); - spin_lock_irqsave(get_ccwdev_lock(cdev), flags); - dev_set_drvdata(&cdev->dev, NULL); - spin_unlock_irqrestore(get_ccwdev_lock(cdev), flags); - } - cdev->handler = NULL; -} - -static int virtio_ccw_offline(struct ccw_device *cdev) -{ - unsigned long flags; - struct virtio_ccw_device *vcdev = virtio_grab_drvdata(cdev); - - if (!vcdev) - return 0; - if (vcdev->device_lost) - virtio_break_device(&vcdev->vdev); - unregister_virtio_device(&vcdev->vdev); - spin_lock_irqsave(get_ccwdev_lock(cdev), flags); - dev_set_drvdata(&cdev->dev, NULL); - spin_unlock_irqrestore(get_ccwdev_lock(cdev), flags); - return 0; -} - -static int virtio_ccw_set_transport_rev(struct virtio_ccw_device *vcdev) -{ - struct virtio_rev_info *rev; - struct ccw1 *ccw; - int ret; - - ccw = kzalloc(sizeof(*ccw), GFP_DMA | GFP_KERNEL); - if (!ccw) - return -ENOMEM; - rev = kzalloc(sizeof(*rev), GFP_DMA | GFP_KERNEL); - if (!rev) { - kfree(ccw); - return -ENOMEM; - } - - /* Set transport revision */ - ccw->cmd_code = CCW_CMD_SET_VIRTIO_REV; - ccw->flags = 0; - ccw->count = sizeof(*rev); - ccw->cda = (__u32)(unsigned long)rev; - - vcdev->revision = VIRTIO_CCW_REV_MAX; - do { - rev->revision = vcdev->revision; - /* none of our supported revisions carry payload */ - rev->length = 0; - ret = ccw_io_helper(vcdev, ccw, - VIRTIO_CCW_DOING_SET_VIRTIO_REV); - if (ret == -EOPNOTSUPP) { - if (vcdev->revision == 0) - /* - * The host device does not support setting - * the revision: let's operate it in legacy - * mode. - */ - ret = 0; - else - vcdev->revision--; - } - } while (ret == -EOPNOTSUPP); - - kfree(ccw); - kfree(rev); - return ret; -} - -static int virtio_ccw_online(struct ccw_device *cdev) -{ - int ret; - struct virtio_ccw_device *vcdev; - unsigned long flags; - - vcdev = kzalloc(sizeof(*vcdev), GFP_KERNEL); - if (!vcdev) { - dev_warn(&cdev->dev, "Could not get memory for virtio\n"); - ret = -ENOMEM; - goto out_free; - } - vcdev->config_block = kzalloc(sizeof(*vcdev->config_block), - GFP_DMA | GFP_KERNEL); - if (!vcdev->config_block) { - ret = -ENOMEM; - goto out_free; - } - vcdev->status = kzalloc(sizeof(*vcdev->status), GFP_DMA | GFP_KERNEL); - if (!vcdev->status) { - ret = -ENOMEM; - goto out_free; - } - - vcdev->is_thinint = virtio_ccw_use_airq; /* at least try */ - - vcdev->vdev.dev.parent = &cdev->dev; - vcdev->vdev.dev.release = virtio_ccw_release_dev; - vcdev->vdev.config = &virtio_ccw_config_ops; - vcdev->cdev = cdev; - init_waitqueue_head(&vcdev->wait_q); - INIT_LIST_HEAD(&vcdev->virtqueues); - spin_lock_init(&vcdev->lock); - - spin_lock_irqsave(get_ccwdev_lock(cdev), flags); - dev_set_drvdata(&cdev->dev, vcdev); - spin_unlock_irqrestore(get_ccwdev_lock(cdev), flags); - vcdev->vdev.id.vendor = cdev->id.cu_type; - vcdev->vdev.id.device = cdev->id.cu_model; - - ret = virtio_ccw_set_transport_rev(vcdev); - if (ret) - goto out_free; - - ret = register_virtio_device(&vcdev->vdev); - if (ret) { - dev_warn(&cdev->dev, "Failed to register virtio device: %d\n", - ret); - goto out_put; - } - return 0; -out_put: - spin_lock_irqsave(get_ccwdev_lock(cdev), flags); - dev_set_drvdata(&cdev->dev, NULL); - spin_unlock_irqrestore(get_ccwdev_lock(cdev), flags); - put_device(&vcdev->vdev.dev); - return ret; -out_free: - if (vcdev) { - kfree(vcdev->status); - kfree(vcdev->config_block); - } - kfree(vcdev); - return ret; -} - -static int virtio_ccw_cio_notify(struct ccw_device *cdev, int event) -{ - int rc; - struct virtio_ccw_device *vcdev = dev_get_drvdata(&cdev->dev); - - /* - * Make sure vcdev is set - * i.e. set_offline/remove callback not already running - */ - if (!vcdev) - return NOTIFY_DONE; - - switch (event) { - case CIO_GONE: - vcdev->device_lost = true; - rc = NOTIFY_DONE; - break; - default: - rc = NOTIFY_DONE; - break; - } - return rc; -} - -static struct ccw_device_id virtio_ids[] = { - { CCW_DEVICE(0x3832, 0) }, - {}, -}; -MODULE_DEVICE_TABLE(ccw, virtio_ids); - -static struct ccw_driver virtio_ccw_driver = { - .driver = { - .owner = THIS_MODULE, - .name = "virtio_ccw", - }, - .ids = virtio_ids, - .probe = virtio_ccw_probe, - .remove = virtio_ccw_remove, - .set_offline = virtio_ccw_offline, - .set_online = virtio_ccw_online, - .notify = virtio_ccw_cio_notify, - .int_class = IRQIO_VIR, -}; - -static int __init pure_hex(char **cp, unsigned int *val, int min_digit, - int max_digit, int max_val) -{ - int diff; - - diff = 0; - *val = 0; - - while (diff <= max_digit) { - int value = hex_to_bin(**cp); - - if (value < 0) - break; - *val = *val * 16 + value; - (*cp)++; - diff++; - } - - if ((diff < min_digit) || (diff > max_digit) || (*val > max_val)) - return 1; - - return 0; -} - -static int __init parse_busid(char *str, unsigned int *cssid, - unsigned int *ssid, unsigned int *devno) -{ - char *str_work; - int rc, ret; - - rc = 1; - - if (*str == '\0') - goto out; - - str_work = str; - ret = pure_hex(&str_work, cssid, 1, 2, __MAX_CSSID); - if (ret || (str_work[0] != '.')) - goto out; - str_work++; - ret = pure_hex(&str_work, ssid, 1, 1, __MAX_SSID); - if (ret || (str_work[0] != '.')) - goto out; - str_work++; - ret = pure_hex(&str_work, devno, 4, 4, __MAX_SUBCHANNEL); - if (ret || (str_work[0] != '\0')) - goto out; - - rc = 0; -out: - return rc; -} - -static void __init no_auto_parse(void) -{ - unsigned int from_cssid, to_cssid, from_ssid, to_ssid, from, to; - char *parm, *str; - int rc; - - str = no_auto; - while ((parm = strsep(&str, ","))) { - rc = parse_busid(strsep(&parm, "-"), &from_cssid, - &from_ssid, &from); - if (rc) - continue; - if (parm != NULL) { - rc = parse_busid(parm, &to_cssid, - &to_ssid, &to); - if ((from_ssid > to_ssid) || - ((from_ssid == to_ssid) && (from > to))) - rc = -EINVAL; - } else { - to_cssid = from_cssid; - to_ssid = from_ssid; - to = from; - } - if (rc) - continue; - while ((from_ssid < to_ssid) || - ((from_ssid == to_ssid) && (from <= to))) { - set_bit(from, devs_no_auto[from_ssid]); - from++; - if (from > __MAX_SUBCHANNEL) { - from_ssid++; - from = 0; - } - } - } -} - -static int __init virtio_ccw_init(void) -{ - /* parse no_auto string before we do anything further */ - no_auto_parse(); - return ccw_driver_register(&virtio_ccw_driver); -} -module_init(virtio_ccw_init); - -static void __exit virtio_ccw_exit(void) -{ - int i; - - ccw_driver_unregister(&virtio_ccw_driver); - for (i = 0; i < MAX_AIRQ_AREAS; i++) - destroy_airq_info(airq_areas[i]); -} -module_exit(virtio_ccw_exit); diff --git a/drivers/s390/virtio/Makefile b/drivers/s390/virtio/Makefile new file mode 100644 index 0000000..241891a --- /dev/null +++ b/drivers/s390/virtio/Makefile @@ -0,0 +1,9 @@ +# Makefile for kvm guest drivers on s390 +# +# Copyright IBM Corp. 2008 +# +# This program is free software; you can redistribute it and/or modify +# it under the terms of the GNU General Public License (version 2 only) +# as published by the Free Software Foundation. + +obj-$(CONFIG_S390_GUEST) += kvm_virtio.o virtio_ccw.o diff --git a/drivers/s390/virtio/kvm_virtio.c b/drivers/s390/virtio/kvm_virtio.c new file mode 100644 index 0000000..dd65c8b --- /dev/null +++ b/drivers/s390/virtio/kvm_virtio.c @@ -0,0 +1,510 @@ +/* + * virtio for kvm on s390 + * + * Copyright IBM Corp. 2008 + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License (version 2 only) + * as published by the Free Software Foundation. + * + * Author(s): Christian Borntraeger + */ + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#define VIRTIO_SUBCODE_64 0x0D00 + +/* + * The pointer to our (page) of device descriptions. + */ +static void *kvm_devices; +static struct work_struct hotplug_work; + +struct kvm_device { + struct virtio_device vdev; + struct kvm_device_desc *desc; +}; + +#define to_kvmdev(vd) container_of(vd, struct kvm_device, vdev) + +/* + * memory layout: + * - kvm_device_descriptor + * struct kvm_device_desc + * - configuration + * struct kvm_vqconfig + * - feature bits + * - config space + */ +static struct kvm_vqconfig *kvm_vq_config(const struct kvm_device_desc *desc) +{ + return (struct kvm_vqconfig *)(desc + 1); +} + +static u8 *kvm_vq_features(const struct kvm_device_desc *desc) +{ + return (u8 *)(kvm_vq_config(desc) + desc->num_vq); +} + +static u8 *kvm_vq_configspace(const struct kvm_device_desc *desc) +{ + return kvm_vq_features(desc) + desc->feature_len * 2; +} + +/* + * The total size of the config page used by this device (incl. desc) + */ +static unsigned desc_size(const struct kvm_device_desc *desc) +{ + return sizeof(*desc) + + desc->num_vq * sizeof(struct kvm_vqconfig) + + desc->feature_len * 2 + + desc->config_len; +} + +/* This gets the device's feature bits. */ +static u64 kvm_get_features(struct virtio_device *vdev) +{ + unsigned int i; + u32 features = 0; + struct kvm_device_desc *desc = to_kvmdev(vdev)->desc; + u8 *in_features = kvm_vq_features(desc); + + for (i = 0; i < min(desc->feature_len * 8, 32); i++) + if (in_features[i / 8] & (1 << (i % 8))) + features |= (1 << i); + return features; +} + +static int kvm_finalize_features(struct virtio_device *vdev) +{ + unsigned int i, bits; + struct kvm_device_desc *desc = to_kvmdev(vdev)->desc; + /* Second half of bitmap is features we accept. */ + u8 *out_features = kvm_vq_features(desc) + desc->feature_len; + + /* Give virtio_ring a chance to accept features. */ + vring_transport_features(vdev); + + /* Make sure we don't have any features > 32 bits! */ + BUG_ON((u32)vdev->features != vdev->features); + + memset(out_features, 0, desc->feature_len); + bits = min_t(unsigned, desc->feature_len, sizeof(vdev->features)) * 8; + for (i = 0; i < bits; i++) { + if (__virtio_test_bit(vdev, i)) + out_features[i / 8] |= (1 << (i % 8)); + } + + return 0; +} + +/* + * Reading and writing elements in config space + */ +static void kvm_get(struct virtio_device *vdev, unsigned int offset, + void *buf, unsigned len) +{ + struct kvm_device_desc *desc = to_kvmdev(vdev)->desc; + + BUG_ON(offset + len > desc->config_len); + memcpy(buf, kvm_vq_configspace(desc) + offset, len); +} + +static void kvm_set(struct virtio_device *vdev, unsigned int offset, + const void *buf, unsigned len) +{ + struct kvm_device_desc *desc = to_kvmdev(vdev)->desc; + + BUG_ON(offset + len > desc->config_len); + memcpy(kvm_vq_configspace(desc) + offset, buf, len); +} + +/* + * The operations to get and set the status word just access + * the status field of the device descriptor. set_status will also + * make a hypercall to the host, to tell about status changes + */ +static u8 kvm_get_status(struct virtio_device *vdev) +{ + return to_kvmdev(vdev)->desc->status; +} + +static void kvm_set_status(struct virtio_device *vdev, u8 status) +{ + BUG_ON(!status); + to_kvmdev(vdev)->desc->status = status; + kvm_hypercall1(KVM_S390_VIRTIO_SET_STATUS, + (unsigned long) to_kvmdev(vdev)->desc); +} + +/* + * To reset the device, we use the KVM_VIRTIO_RESET hypercall, using the + * descriptor address. The Host will zero the status and all the + * features. + */ +static void kvm_reset(struct virtio_device *vdev) +{ + kvm_hypercall1(KVM_S390_VIRTIO_RESET, + (unsigned long) to_kvmdev(vdev)->desc); +} + +/* + * When the virtio_ring code wants to notify the Host, it calls us here and we + * make a hypercall. We hand the address of the virtqueue so the Host + * knows which virtqueue we're talking about. + */ +static bool kvm_notify(struct virtqueue *vq) +{ + long rc; + struct kvm_vqconfig *config = vq->priv; + + rc = kvm_hypercall1(KVM_S390_VIRTIO_NOTIFY, config->address); + if (rc < 0) + return false; + return true; +} + +/* + * This routine finds the first virtqueue described in the configuration of + * this device and sets it up. + */ +static struct virtqueue *kvm_find_vq(struct virtio_device *vdev, + unsigned index, + void (*callback)(struct virtqueue *vq), + const char *name) +{ + struct kvm_device *kdev = to_kvmdev(vdev); + struct kvm_vqconfig *config; + struct virtqueue *vq; + int err; + + if (index >= kdev->desc->num_vq) + return ERR_PTR(-ENOENT); + + if (!name) + return NULL; + + config = kvm_vq_config(kdev->desc)+index; + + err = vmem_add_mapping(config->address, + vring_size(config->num, + KVM_S390_VIRTIO_RING_ALIGN)); + if (err) + goto out; + + vq = vring_new_virtqueue(index, config->num, KVM_S390_VIRTIO_RING_ALIGN, + vdev, true, (void *) config->address, + kvm_notify, callback, name); + if (!vq) { + err = -ENOMEM; + goto unmap; + } + + /* + * register a callback token + * The host will sent this via the external interrupt parameter + */ + config->token = (u64) vq; + + vq->priv = config; + return vq; +unmap: + vmem_remove_mapping(config->address, + vring_size(config->num, + KVM_S390_VIRTIO_RING_ALIGN)); +out: + return ERR_PTR(err); +} + +static void kvm_del_vq(struct virtqueue *vq) +{ + struct kvm_vqconfig *config = vq->priv; + + vring_del_virtqueue(vq); + vmem_remove_mapping(config->address, + vring_size(config->num, + KVM_S390_VIRTIO_RING_ALIGN)); +} + +static void kvm_del_vqs(struct virtio_device *vdev) +{ + struct virtqueue *vq, *n; + + list_for_each_entry_safe(vq, n, &vdev->vqs, list) + kvm_del_vq(vq); +} + +static int kvm_find_vqs(struct virtio_device *vdev, unsigned nvqs, + struct virtqueue *vqs[], + vq_callback_t *callbacks[], + const char *names[]) +{ + struct kvm_device *kdev = to_kvmdev(vdev); + int i; + + /* We must have this many virtqueues. */ + if (nvqs > kdev->desc->num_vq) + return -ENOENT; + + for (i = 0; i < nvqs; ++i) { + vqs[i] = kvm_find_vq(vdev, i, callbacks[i], names[i]); + if (IS_ERR(vqs[i])) + goto error; + } + return 0; + +error: + kvm_del_vqs(vdev); + return PTR_ERR(vqs[i]); +} + +static const char *kvm_bus_name(struct virtio_device *vdev) +{ + return ""; +} + +/* + * The config ops structure as defined by virtio config + */ +static const struct virtio_config_ops kvm_vq_configspace_ops = { + .get_features = kvm_get_features, + .finalize_features = kvm_finalize_features, + .get = kvm_get, + .set = kvm_set, + .get_status = kvm_get_status, + .set_status = kvm_set_status, + .reset = kvm_reset, + .find_vqs = kvm_find_vqs, + .del_vqs = kvm_del_vqs, + .bus_name = kvm_bus_name, +}; + +/* + * The root device for the kvm virtio devices. + * This makes them appear as /sys/devices/kvm_s390/0,1,2 not /sys/devices/0,1,2. + */ +static struct device *kvm_root; + +/* + * adds a new device and register it with virtio + * appropriate drivers are loaded by the device model + */ +static void add_kvm_device(struct kvm_device_desc *d, unsigned int offset) +{ + struct kvm_device *kdev; + + kdev = kzalloc(sizeof(*kdev), GFP_KERNEL); + if (!kdev) { + printk(KERN_EMERG "Cannot allocate kvm dev %u type %u\n", + offset, d->type); + return; + } + + kdev->vdev.dev.parent = kvm_root; + kdev->vdev.id.device = d->type; + kdev->vdev.config = &kvm_vq_configspace_ops; + kdev->desc = d; + + if (register_virtio_device(&kdev->vdev) != 0) { + printk(KERN_ERR "Failed to register kvm device %u type %u\n", + offset, d->type); + kfree(kdev); + } +} + +/* + * scan_devices() simply iterates through the device page. + * The type 0 is reserved to mean "end of devices". + */ +static void scan_devices(void) +{ + unsigned int i; + struct kvm_device_desc *d; + + for (i = 0; i < PAGE_SIZE; i += desc_size(d)) { + d = kvm_devices + i; + + if (d->type == 0) + break; + + add_kvm_device(d, i); + } +} + +/* + * match for a kvm device with a specific desc pointer + */ +static int match_desc(struct device *dev, void *data) +{ + struct virtio_device *vdev = dev_to_virtio(dev); + struct kvm_device *kdev = to_kvmdev(vdev); + + return kdev->desc == data; +} + +/* + * hotplug_device tries to find changes in the device page. + */ +static void hotplug_devices(struct work_struct *dummy) +{ + unsigned int i; + struct kvm_device_desc *d; + struct device *dev; + + for (i = 0; i < PAGE_SIZE; i += desc_size(d)) { + d = kvm_devices + i; + + /* end of list */ + if (d->type == 0) + break; + + /* device already exists */ + dev = device_find_child(kvm_root, d, match_desc); + if (dev) { + /* XXX check for hotplug remove */ + put_device(dev); + continue; + } + + /* new device */ + printk(KERN_INFO "Adding new virtio device %p\n", d); + add_kvm_device(d, i); + } +} + +/* + * we emulate the request_irq behaviour on top of s390 extints + */ +static void kvm_extint_handler(struct ext_code ext_code, + unsigned int param32, unsigned long param64) +{ + struct virtqueue *vq; + u32 param; + + if ((ext_code.subcode & 0xff00) != VIRTIO_SUBCODE_64) + return; + inc_irq_stat(IRQEXT_VRT); + + /* The LSB might be overloaded, we have to mask it */ + vq = (struct virtqueue *)(param64 & ~1UL); + + /* We use ext_params to decide what this interrupt means */ + param = param32 & VIRTIO_PARAM_MASK; + + switch (param) { + case VIRTIO_PARAM_CONFIG_CHANGED: + virtio_config_changed(vq->vdev); + break; + case VIRTIO_PARAM_DEV_ADD: + schedule_work(&hotplug_work); + break; + case VIRTIO_PARAM_VRING_INTERRUPT: + default: + vring_interrupt(0, vq); + break; + } +} + +/* + * For s390-virtio, we expect a page above main storage containing + * the virtio configuration. Try to actually load from this area + * in order to figure out if the host provides this page. + */ +static int __init test_devices_support(unsigned long addr) +{ + int ret = -EIO; + + asm volatile( + "0: lura 0,%1\n" + "1: xgr %0,%0\n" + "2:\n" + EX_TABLE(0b,2b) + EX_TABLE(1b,2b) + : "+d" (ret) + : "a" (addr) + : "0", "cc"); + return ret; +} +/* + * Init function for virtio + * devices are in a single page above top of "normal" + standby mem + */ +static int __init kvm_devices_init(void) +{ + int rc; + unsigned long total_memory_size = sclp_get_rzm() * sclp_get_rnmax(); + + if (!MACHINE_IS_KVM) + return -ENODEV; + + if (test_devices_support(total_memory_size) < 0) + return -ENODEV; + + rc = vmem_add_mapping(total_memory_size, PAGE_SIZE); + if (rc) + return rc; + + kvm_devices = (void *) total_memory_size; + + kvm_root = root_device_register("kvm_s390"); + if (IS_ERR(kvm_root)) { + rc = PTR_ERR(kvm_root); + printk(KERN_ERR "Could not register kvm_s390 root device"); + vmem_remove_mapping(total_memory_size, PAGE_SIZE); + return rc; + } + + INIT_WORK(&hotplug_work, hotplug_devices); + + irq_subclass_register(IRQ_SUBCLASS_SERVICE_SIGNAL); + register_external_irq(EXT_IRQ_CP_SERVICE, kvm_extint_handler); + + scan_devices(); + return 0; +} + +/* code for early console output with virtio_console */ +static __init int early_put_chars(u32 vtermno, const char *buf, int count) +{ + char scratch[17]; + unsigned int len = count; + + if (len > sizeof(scratch) - 1) + len = sizeof(scratch) - 1; + scratch[len] = '\0'; + memcpy(scratch, buf, len); + kvm_hypercall1(KVM_S390_VIRTIO_NOTIFY, __pa(scratch)); + return len; +} + +static int __init s390_virtio_console_init(void) +{ + if (sclp_has_vt220() || sclp_has_linemode()) + return -ENODEV; + return virtio_cons_early_init(early_put_chars); +} +console_initcall(s390_virtio_console_init); + + +/* + * We do this after core stuff, but before the drivers. + */ +postcore_initcall(kvm_devices_init); diff --git a/drivers/s390/virtio/virtio_ccw.c b/drivers/s390/virtio/virtio_ccw.c new file mode 100644 index 0000000..6f1fa17 --- /dev/null +++ b/drivers/s390/virtio/virtio_ccw.c @@ -0,0 +1,1380 @@ +/* + * ccw based virtio transport + * + * Copyright IBM Corp. 2012, 2014 + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License (version 2 only) + * as published by the Free Software Foundation. + * + * Author(s): Cornelia Huck + */ + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +/* + * virtio related functions + */ + +struct vq_config_block { + __u16 index; + __u16 num; +} __packed; + +#define VIRTIO_CCW_CONFIG_SIZE 0x100 +/* same as PCI config space size, should be enough for all drivers */ + +struct virtio_ccw_device { + struct virtio_device vdev; + __u8 *status; + __u8 config[VIRTIO_CCW_CONFIG_SIZE]; + struct ccw_device *cdev; + __u32 curr_io; + int err; + unsigned int revision; /* Transport revision */ + wait_queue_head_t wait_q; + spinlock_t lock; + struct list_head virtqueues; + unsigned long indicators; + unsigned long indicators2; + struct vq_config_block *config_block; + bool is_thinint; + bool going_away; + bool device_lost; + void *airq_info; +}; + +struct vq_info_block_legacy { + __u64 queue; + __u32 align; + __u16 index; + __u16 num; +} __packed; + +struct vq_info_block { + __u64 desc; + __u32 res0; + __u16 index; + __u16 num; + __u64 avail; + __u64 used; +} __packed; + +struct virtio_feature_desc { + __u32 features; + __u8 index; +} __packed; + +struct virtio_thinint_area { + unsigned long summary_indicator; + unsigned long indicator; + u64 bit_nr; + u8 isc; +} __packed; + +struct virtio_rev_info { + __u16 revision; + __u16 length; + __u8 data[]; +}; + +/* the highest virtio-ccw revision we support */ +#define VIRTIO_CCW_REV_MAX 1 + +struct virtio_ccw_vq_info { + struct virtqueue *vq; + int num; + void *queue; + union { + struct vq_info_block s; + struct vq_info_block_legacy l; + } *info_block; + int bit_nr; + struct list_head node; + long cookie; +}; + +#define VIRTIO_AIRQ_ISC IO_SCH_ISC /* inherit from subchannel */ + +#define VIRTIO_IV_BITS (L1_CACHE_BYTES * 8) +#define MAX_AIRQ_AREAS 20 + +static int virtio_ccw_use_airq = 1; + +struct airq_info { + rwlock_t lock; + u8 summary_indicator; + struct airq_struct airq; + struct airq_iv *aiv; +}; +static struct airq_info *airq_areas[MAX_AIRQ_AREAS]; + +#define CCW_CMD_SET_VQ 0x13 +#define CCW_CMD_VDEV_RESET 0x33 +#define CCW_CMD_SET_IND 0x43 +#define CCW_CMD_SET_CONF_IND 0x53 +#define CCW_CMD_READ_FEAT 0x12 +#define CCW_CMD_WRITE_FEAT 0x11 +#define CCW_CMD_READ_CONF 0x22 +#define CCW_CMD_WRITE_CONF 0x21 +#define CCW_CMD_WRITE_STATUS 0x31 +#define CCW_CMD_READ_VQ_CONF 0x32 +#define CCW_CMD_SET_IND_ADAPTER 0x73 +#define CCW_CMD_SET_VIRTIO_REV 0x83 + +#define VIRTIO_CCW_DOING_SET_VQ 0x00010000 +#define VIRTIO_CCW_DOING_RESET 0x00040000 +#define VIRTIO_CCW_DOING_READ_FEAT 0x00080000 +#define VIRTIO_CCW_DOING_WRITE_FEAT 0x00100000 +#define VIRTIO_CCW_DOING_READ_CONFIG 0x00200000 +#define VIRTIO_CCW_DOING_WRITE_CONFIG 0x00400000 +#define VIRTIO_CCW_DOING_WRITE_STATUS 0x00800000 +#define VIRTIO_CCW_DOING_SET_IND 0x01000000 +#define VIRTIO_CCW_DOING_READ_VQ_CONF 0x02000000 +#define VIRTIO_CCW_DOING_SET_CONF_IND 0x04000000 +#define VIRTIO_CCW_DOING_SET_IND_ADAPTER 0x08000000 +#define VIRTIO_CCW_DOING_SET_VIRTIO_REV 0x10000000 +#define VIRTIO_CCW_INTPARM_MASK 0xffff0000 + +static struct virtio_ccw_device *to_vc_device(struct virtio_device *vdev) +{ + return container_of(vdev, struct virtio_ccw_device, vdev); +} + +static void drop_airq_indicator(struct virtqueue *vq, struct airq_info *info) +{ + unsigned long i, flags; + + write_lock_irqsave(&info->lock, flags); + for (i = 0; i < airq_iv_end(info->aiv); i++) { + if (vq == (void *)airq_iv_get_ptr(info->aiv, i)) { + airq_iv_free_bit(info->aiv, i); + airq_iv_set_ptr(info->aiv, i, 0); + break; + } + } + write_unlock_irqrestore(&info->lock, flags); +} + +static void virtio_airq_handler(struct airq_struct *airq) +{ + struct airq_info *info = container_of(airq, struct airq_info, airq); + unsigned long ai; + + inc_irq_stat(IRQIO_VAI); + read_lock(&info->lock); + /* Walk through indicators field, summary indicator active. */ + for (ai = 0;;) { + ai = airq_iv_scan(info->aiv, ai, airq_iv_end(info->aiv)); + if (ai == -1UL) + break; + vring_interrupt(0, (void *)airq_iv_get_ptr(info->aiv, ai)); + } + info->summary_indicator = 0; + smp_wmb(); + /* Walk through indicators field, summary indicator not active. */ + for (ai = 0;;) { + ai = airq_iv_scan(info->aiv, ai, airq_iv_end(info->aiv)); + if (ai == -1UL) + break; + vring_interrupt(0, (void *)airq_iv_get_ptr(info->aiv, ai)); + } + read_unlock(&info->lock); +} + +static struct airq_info *new_airq_info(void) +{ + struct airq_info *info; + int rc; + + info = kzalloc(sizeof(*info), GFP_KERNEL); + if (!info) + return NULL; + rwlock_init(&info->lock); + info->aiv = airq_iv_create(VIRTIO_IV_BITS, AIRQ_IV_ALLOC | AIRQ_IV_PTR); + if (!info->aiv) { + kfree(info); + return NULL; + } + info->airq.handler = virtio_airq_handler; + info->airq.lsi_ptr = &info->summary_indicator; + info->airq.lsi_mask = 0xff; + info->airq.isc = VIRTIO_AIRQ_ISC; + rc = register_adapter_interrupt(&info->airq); + if (rc) { + airq_iv_release(info->aiv); + kfree(info); + return NULL; + } + return info; +} + +static void destroy_airq_info(struct airq_info *info) +{ + if (!info) + return; + + unregister_adapter_interrupt(&info->airq); + airq_iv_release(info->aiv); + kfree(info); +} + +static unsigned long get_airq_indicator(struct virtqueue *vqs[], int nvqs, + u64 *first, void **airq_info) +{ + int i, j; + struct airq_info *info; + unsigned long indicator_addr = 0; + unsigned long bit, flags; + + for (i = 0; i < MAX_AIRQ_AREAS && !indicator_addr; i++) { + if (!airq_areas[i]) + airq_areas[i] = new_airq_info(); + info = airq_areas[i]; + if (!info) + return 0; + write_lock_irqsave(&info->lock, flags); + bit = airq_iv_alloc(info->aiv, nvqs); + if (bit == -1UL) { + /* Not enough vacancies. */ + write_unlock_irqrestore(&info->lock, flags); + continue; + } + *first = bit; + *airq_info = info; + indicator_addr = (unsigned long)info->aiv->vector; + for (j = 0; j < nvqs; j++) { + airq_iv_set_ptr(info->aiv, bit + j, + (unsigned long)vqs[j]); + } + write_unlock_irqrestore(&info->lock, flags); + } + return indicator_addr; +} + +static void virtio_ccw_drop_indicators(struct virtio_ccw_device *vcdev) +{ + struct virtio_ccw_vq_info *info; + + list_for_each_entry(info, &vcdev->virtqueues, node) + drop_airq_indicator(info->vq, vcdev->airq_info); +} + +static int doing_io(struct virtio_ccw_device *vcdev, __u32 flag) +{ + unsigned long flags; + __u32 ret; + + spin_lock_irqsave(get_ccwdev_lock(vcdev->cdev), flags); + if (vcdev->err) + ret = 0; + else + ret = vcdev->curr_io & flag; + spin_unlock_irqrestore(get_ccwdev_lock(vcdev->cdev), flags); + return ret; +} + +static int ccw_io_helper(struct virtio_ccw_device *vcdev, + struct ccw1 *ccw, __u32 intparm) +{ + int ret; + unsigned long flags; + int flag = intparm & VIRTIO_CCW_INTPARM_MASK; + + do { + spin_lock_irqsave(get_ccwdev_lock(vcdev->cdev), flags); + ret = ccw_device_start(vcdev->cdev, ccw, intparm, 0, 0); + if (!ret) { + if (!vcdev->curr_io) + vcdev->err = 0; + vcdev->curr_io |= flag; + } + spin_unlock_irqrestore(get_ccwdev_lock(vcdev->cdev), flags); + cpu_relax(); + } while (ret == -EBUSY); + wait_event(vcdev->wait_q, doing_io(vcdev, flag) == 0); + return ret ? ret : vcdev->err; +} + +static void virtio_ccw_drop_indicator(struct virtio_ccw_device *vcdev, + struct ccw1 *ccw) +{ + int ret; + unsigned long *indicatorp = NULL; + struct virtio_thinint_area *thinint_area = NULL; + struct airq_info *airq_info = vcdev->airq_info; + + if (vcdev->is_thinint) { + thinint_area = kzalloc(sizeof(*thinint_area), + GFP_DMA | GFP_KERNEL); + if (!thinint_area) + return; + thinint_area->summary_indicator = + (unsigned long) &airq_info->summary_indicator; + thinint_area->isc = VIRTIO_AIRQ_ISC; + ccw->cmd_code = CCW_CMD_SET_IND_ADAPTER; + ccw->count = sizeof(*thinint_area); + ccw->cda = (__u32)(unsigned long) thinint_area; + } else { + indicatorp = kmalloc(sizeof(&vcdev->indicators), + GFP_DMA | GFP_KERNEL); + if (!indicatorp) + return; + *indicatorp = 0; + ccw->cmd_code = CCW_CMD_SET_IND; + ccw->count = sizeof(vcdev->indicators); + ccw->cda = (__u32)(unsigned long) indicatorp; + } + /* Deregister indicators from host. */ + vcdev->indicators = 0; + ccw->flags = 0; + ret = ccw_io_helper(vcdev, ccw, + vcdev->is_thinint ? + VIRTIO_CCW_DOING_SET_IND_ADAPTER : + VIRTIO_CCW_DOING_SET_IND); + if (ret && (ret != -ENODEV)) + dev_info(&vcdev->cdev->dev, + "Failed to deregister indicators (%d)\n", ret); + else if (vcdev->is_thinint) + virtio_ccw_drop_indicators(vcdev); + kfree(indicatorp); + kfree(thinint_area); +} + +static inline long do_kvm_notify(struct subchannel_id schid, + unsigned long queue_index, + long cookie) +{ + register unsigned long __nr asm("1") = KVM_S390_VIRTIO_CCW_NOTIFY; + register struct subchannel_id __schid asm("2") = schid; + register unsigned long __index asm("3") = queue_index; + register long __rc asm("2"); + register long __cookie asm("4") = cookie; + + asm volatile ("diag 2,4,0x500\n" + : "=d" (__rc) : "d" (__nr), "d" (__schid), "d" (__index), + "d"(__cookie) + : "memory", "cc"); + return __rc; +} + +static bool virtio_ccw_kvm_notify(struct virtqueue *vq) +{ + struct virtio_ccw_vq_info *info = vq->priv; + struct virtio_ccw_device *vcdev; + struct subchannel_id schid; + + vcdev = to_vc_device(info->vq->vdev); + ccw_device_get_schid(vcdev->cdev, &schid); + info->cookie = do_kvm_notify(schid, vq->index, info->cookie); + if (info->cookie < 0) + return false; + return true; +} + +static int virtio_ccw_read_vq_conf(struct virtio_ccw_device *vcdev, + struct ccw1 *ccw, int index) +{ + vcdev->config_block->index = index; + ccw->cmd_code = CCW_CMD_READ_VQ_CONF; + ccw->flags = 0; + ccw->count = sizeof(struct vq_config_block); + ccw->cda = (__u32)(unsigned long)(vcdev->config_block); + ccw_io_helper(vcdev, ccw, VIRTIO_CCW_DOING_READ_VQ_CONF); + return vcdev->config_block->num; +} + +static void virtio_ccw_del_vq(struct virtqueue *vq, struct ccw1 *ccw) +{ + struct virtio_ccw_device *vcdev = to_vc_device(vq->vdev); + struct virtio_ccw_vq_info *info = vq->priv; + unsigned long flags; + unsigned long size; + int ret; + unsigned int index = vq->index; + + /* Remove from our list. */ + spin_lock_irqsave(&vcdev->lock, flags); + list_del(&info->node); + spin_unlock_irqrestore(&vcdev->lock, flags); + + /* Release from host. */ + if (vcdev->revision == 0) { + info->info_block->l.queue = 0; + info->info_block->l.align = 0; + info->info_block->l.index = index; + info->info_block->l.num = 0; + ccw->count = sizeof(info->info_block->l); + } else { + info->info_block->s.desc = 0; + info->info_block->s.index = index; + info->info_block->s.num = 0; + info->info_block->s.avail = 0; + info->info_block->s.used = 0; + ccw->count = sizeof(info->info_block->s); + } + ccw->cmd_code = CCW_CMD_SET_VQ; + ccw->flags = 0; + ccw->cda = (__u32)(unsigned long)(info->info_block); + ret = ccw_io_helper(vcdev, ccw, + VIRTIO_CCW_DOING_SET_VQ | index); + /* + * -ENODEV isn't considered an error: The device is gone anyway. + * This may happen on device detach. + */ + if (ret && (ret != -ENODEV)) + dev_warn(&vq->vdev->dev, "Error %d while deleting queue %d", + ret, index); + + vring_del_virtqueue(vq); + size = PAGE_ALIGN(vring_size(info->num, KVM_VIRTIO_CCW_RING_ALIGN)); + free_pages_exact(info->queue, size); + kfree(info->info_block); + kfree(info); +} + +static void virtio_ccw_del_vqs(struct virtio_device *vdev) +{ + struct virtqueue *vq, *n; + struct ccw1 *ccw; + struct virtio_ccw_device *vcdev = to_vc_device(vdev); + + ccw = kzalloc(sizeof(*ccw), GFP_DMA | GFP_KERNEL); + if (!ccw) + return; + + virtio_ccw_drop_indicator(vcdev, ccw); + + list_for_each_entry_safe(vq, n, &vdev->vqs, list) + virtio_ccw_del_vq(vq, ccw); + + kfree(ccw); +} + +static struct virtqueue *virtio_ccw_setup_vq(struct virtio_device *vdev, + int i, vq_callback_t *callback, + const char *name, + struct ccw1 *ccw) +{ + struct virtio_ccw_device *vcdev = to_vc_device(vdev); + int err; + struct virtqueue *vq = NULL; + struct virtio_ccw_vq_info *info; + unsigned long size = 0; /* silence the compiler */ + unsigned long flags; + + /* Allocate queue. */ + info = kzalloc(sizeof(struct virtio_ccw_vq_info), GFP_KERNEL); + if (!info) { + dev_warn(&vcdev->cdev->dev, "no info\n"); + err = -ENOMEM; + goto out_err; + } + info->info_block = kzalloc(sizeof(*info->info_block), + GFP_DMA | GFP_KERNEL); + if (!info->info_block) { + dev_warn(&vcdev->cdev->dev, "no info block\n"); + err = -ENOMEM; + goto out_err; + } + info->num = virtio_ccw_read_vq_conf(vcdev, ccw, i); + size = PAGE_ALIGN(vring_size(info->num, KVM_VIRTIO_CCW_RING_ALIGN)); + info->queue = alloc_pages_exact(size, GFP_KERNEL | __GFP_ZERO); + if (info->queue == NULL) { + dev_warn(&vcdev->cdev->dev, "no queue\n"); + err = -ENOMEM; + goto out_err; + } + + vq = vring_new_virtqueue(i, info->num, KVM_VIRTIO_CCW_RING_ALIGN, vdev, + true, info->queue, virtio_ccw_kvm_notify, + callback, name); + if (!vq) { + /* For now, we fail if we can't get the requested size. */ + dev_warn(&vcdev->cdev->dev, "no vq\n"); + err = -ENOMEM; + goto out_err; + } + + /* Register it with the host. */ + if (vcdev->revision == 0) { + info->info_block->l.queue = (__u64)info->queue; + info->info_block->l.align = KVM_VIRTIO_CCW_RING_ALIGN; + info->info_block->l.index = i; + info->info_block->l.num = info->num; + ccw->count = sizeof(info->info_block->l); + } else { + info->info_block->s.desc = (__u64)info->queue; + info->info_block->s.index = i; + info->info_block->s.num = info->num; + info->info_block->s.avail = (__u64)virtqueue_get_avail(vq); + info->info_block->s.used = (__u64)virtqueue_get_used(vq); + ccw->count = sizeof(info->info_block->s); + } + ccw->cmd_code = CCW_CMD_SET_VQ; + ccw->flags = 0; + ccw->cda = (__u32)(unsigned long)(info->info_block); + err = ccw_io_helper(vcdev, ccw, VIRTIO_CCW_DOING_SET_VQ | i); + if (err) { + dev_warn(&vcdev->cdev->dev, "SET_VQ failed\n"); + goto out_err; + } + + info->vq = vq; + vq->priv = info; + + /* Save it to our list. */ + spin_lock_irqsave(&vcdev->lock, flags); + list_add(&info->node, &vcdev->virtqueues); + spin_unlock_irqrestore(&vcdev->lock, flags); + + return vq; + +out_err: + if (vq) + vring_del_virtqueue(vq); + if (info) { + if (info->queue) + free_pages_exact(info->queue, size); + kfree(info->info_block); + } + kfree(info); + return ERR_PTR(err); +} + +static int virtio_ccw_register_adapter_ind(struct virtio_ccw_device *vcdev, + struct virtqueue *vqs[], int nvqs, + struct ccw1 *ccw) +{ + int ret; + struct virtio_thinint_area *thinint_area = NULL; + struct airq_info *info; + + thinint_area = kzalloc(sizeof(*thinint_area), GFP_DMA | GFP_KERNEL); + if (!thinint_area) { + ret = -ENOMEM; + goto out; + } + /* Try to get an indicator. */ + thinint_area->indicator = get_airq_indicator(vqs, nvqs, + &thinint_area->bit_nr, + &vcdev->airq_info); + if (!thinint_area->indicator) { + ret = -ENOSPC; + goto out; + } + info = vcdev->airq_info; + thinint_area->summary_indicator = + (unsigned long) &info->summary_indicator; + thinint_area->isc = VIRTIO_AIRQ_ISC; + ccw->cmd_code = CCW_CMD_SET_IND_ADAPTER; + ccw->flags = CCW_FLAG_SLI; + ccw->count = sizeof(*thinint_area); + ccw->cda = (__u32)(unsigned long)thinint_area; + ret = ccw_io_helper(vcdev, ccw, VIRTIO_CCW_DOING_SET_IND_ADAPTER); + if (ret) { + if (ret == -EOPNOTSUPP) { + /* + * The host does not support adapter interrupts + * for virtio-ccw, stop trying. + */ + virtio_ccw_use_airq = 0; + pr_info("Adapter interrupts unsupported on host\n"); + } else + dev_warn(&vcdev->cdev->dev, + "enabling adapter interrupts = %d\n", ret); + virtio_ccw_drop_indicators(vcdev); + } +out: + kfree(thinint_area); + return ret; +} + +static int virtio_ccw_find_vqs(struct virtio_device *vdev, unsigned nvqs, + struct virtqueue *vqs[], + vq_callback_t *callbacks[], + const char *names[]) +{ + struct virtio_ccw_device *vcdev = to_vc_device(vdev); + unsigned long *indicatorp = NULL; + int ret, i; + struct ccw1 *ccw; + + ccw = kzalloc(sizeof(*ccw), GFP_DMA | GFP_KERNEL); + if (!ccw) + return -ENOMEM; + + for (i = 0; i < nvqs; ++i) { + vqs[i] = virtio_ccw_setup_vq(vdev, i, callbacks[i], names[i], + ccw); + if (IS_ERR(vqs[i])) { + ret = PTR_ERR(vqs[i]); + vqs[i] = NULL; + goto out; + } + } + ret = -ENOMEM; + /* We need a data area under 2G to communicate. */ + indicatorp = kmalloc(sizeof(&vcdev->indicators), GFP_DMA | GFP_KERNEL); + if (!indicatorp) + goto out; + *indicatorp = (unsigned long) &vcdev->indicators; + if (vcdev->is_thinint) { + ret = virtio_ccw_register_adapter_ind(vcdev, vqs, nvqs, ccw); + if (ret) + /* no error, just fall back to legacy interrupts */ + vcdev->is_thinint = 0; + } + if (!vcdev->is_thinint) { + /* Register queue indicators with host. */ + vcdev->indicators = 0; + ccw->cmd_code = CCW_CMD_SET_IND; + ccw->flags = 0; + ccw->count = sizeof(vcdev->indicators); + ccw->cda = (__u32)(unsigned long) indicatorp; + ret = ccw_io_helper(vcdev, ccw, VIRTIO_CCW_DOING_SET_IND); + if (ret) + goto out; + } + /* Register indicators2 with host for config changes */ + *indicatorp = (unsigned long) &vcdev->indicators2; + vcdev->indicators2 = 0; + ccw->cmd_code = CCW_CMD_SET_CONF_IND; + ccw->flags = 0; + ccw->count = sizeof(vcdev->indicators2); + ccw->cda = (__u32)(unsigned long) indicatorp; + ret = ccw_io_helper(vcdev, ccw, VIRTIO_CCW_DOING_SET_CONF_IND); + if (ret) + goto out; + + kfree(indicatorp); + kfree(ccw); + return 0; +out: + kfree(indicatorp); + kfree(ccw); + virtio_ccw_del_vqs(vdev); + return ret; +} + +static void virtio_ccw_reset(struct virtio_device *vdev) +{ + struct virtio_ccw_device *vcdev = to_vc_device(vdev); + struct ccw1 *ccw; + + ccw = kzalloc(sizeof(*ccw), GFP_DMA | GFP_KERNEL); + if (!ccw) + return; + + /* Zero status bits. */ + *vcdev->status = 0; + + /* Send a reset ccw on device. */ + ccw->cmd_code = CCW_CMD_VDEV_RESET; + ccw->flags = 0; + ccw->count = 0; + ccw->cda = 0; + ccw_io_helper(vcdev, ccw, VIRTIO_CCW_DOING_RESET); + kfree(ccw); +} + +static u64 virtio_ccw_get_features(struct virtio_device *vdev) +{ + struct virtio_ccw_device *vcdev = to_vc_device(vdev); + struct virtio_feature_desc *features; + int ret; + u64 rc; + struct ccw1 *ccw; + + ccw = kzalloc(sizeof(*ccw), GFP_DMA | GFP_KERNEL); + if (!ccw) + return 0; + + features = kzalloc(sizeof(*features), GFP_DMA | GFP_KERNEL); + if (!features) { + rc = 0; + goto out_free; + } + /* Read the feature bits from the host. */ + features->index = 0; + ccw->cmd_code = CCW_CMD_READ_FEAT; + ccw->flags = 0; + ccw->count = sizeof(*features); + ccw->cda = (__u32)(unsigned long)features; + ret = ccw_io_helper(vcdev, ccw, VIRTIO_CCW_DOING_READ_FEAT); + if (ret) { + rc = 0; + goto out_free; + } + + rc = le32_to_cpu(features->features); + + if (vcdev->revision == 0) + goto out_free; + + /* Read second half of the feature bits from the host. */ + features->index = 1; + ccw->cmd_code = CCW_CMD_READ_FEAT; + ccw->flags = 0; + ccw->count = sizeof(*features); + ccw->cda = (__u32)(unsigned long)features; + ret = ccw_io_helper(vcdev, ccw, VIRTIO_CCW_DOING_READ_FEAT); + if (ret == 0) + rc |= (u64)le32_to_cpu(features->features) << 32; + +out_free: + kfree(features); + kfree(ccw); + return rc; +} + +static int virtio_ccw_finalize_features(struct virtio_device *vdev) +{ + struct virtio_ccw_device *vcdev = to_vc_device(vdev); + struct virtio_feature_desc *features; + struct ccw1 *ccw; + int ret; + + if (vcdev->revision >= 1 && + !__virtio_test_bit(vdev, VIRTIO_F_VERSION_1)) { + dev_err(&vdev->dev, "virtio: device uses revision 1 " + "but does not have VIRTIO_F_VERSION_1\n"); + return -EINVAL; + } + + ccw = kzalloc(sizeof(*ccw), GFP_DMA | GFP_KERNEL); + if (!ccw) + return -ENOMEM; + + features = kzalloc(sizeof(*features), GFP_DMA | GFP_KERNEL); + if (!features) { + ret = -ENOMEM; + goto out_free; + } + /* Give virtio_ring a chance to accept features. */ + vring_transport_features(vdev); + + features->index = 0; + features->features = cpu_to_le32((u32)vdev->features); + /* Write the first half of the feature bits to the host. */ + ccw->cmd_code = CCW_CMD_WRITE_FEAT; + ccw->flags = 0; + ccw->count = sizeof(*features); + ccw->cda = (__u32)(unsigned long)features; + ret = ccw_io_helper(vcdev, ccw, VIRTIO_CCW_DOING_WRITE_FEAT); + if (ret) + goto out_free; + + if (vcdev->revision == 0) + goto out_free; + + features->index = 1; + features->features = cpu_to_le32(vdev->features >> 32); + /* Write the second half of the feature bits to the host. */ + ccw->cmd_code = CCW_CMD_WRITE_FEAT; + ccw->flags = 0; + ccw->count = sizeof(*features); + ccw->cda = (__u32)(unsigned long)features; + ret = ccw_io_helper(vcdev, ccw, VIRTIO_CCW_DOING_WRITE_FEAT); + +out_free: + kfree(features); + kfree(ccw); + + return ret; +} + +static void virtio_ccw_get_config(struct virtio_device *vdev, + unsigned int offset, void *buf, unsigned len) +{ + struct virtio_ccw_device *vcdev = to_vc_device(vdev); + int ret; + struct ccw1 *ccw; + void *config_area; + + ccw = kzalloc(sizeof(*ccw), GFP_DMA | GFP_KERNEL); + if (!ccw) + return; + + config_area = kzalloc(VIRTIO_CCW_CONFIG_SIZE, GFP_DMA | GFP_KERNEL); + if (!config_area) + goto out_free; + + /* Read the config area from the host. */ + ccw->cmd_code = CCW_CMD_READ_CONF; + ccw->flags = 0; + ccw->count = offset + len; + ccw->cda = (__u32)(unsigned long)config_area; + ret = ccw_io_helper(vcdev, ccw, VIRTIO_CCW_DOING_READ_CONFIG); + if (ret) + goto out_free; + + memcpy(vcdev->config, config_area, sizeof(vcdev->config)); + memcpy(buf, &vcdev->config[offset], len); + +out_free: + kfree(config_area); + kfree(ccw); +} + +static void virtio_ccw_set_config(struct virtio_device *vdev, + unsigned int offset, const void *buf, + unsigned len) +{ + struct virtio_ccw_device *vcdev = to_vc_device(vdev); + struct ccw1 *ccw; + void *config_area; + + ccw = kzalloc(sizeof(*ccw), GFP_DMA | GFP_KERNEL); + if (!ccw) + return; + + config_area = kzalloc(VIRTIO_CCW_CONFIG_SIZE, GFP_DMA | GFP_KERNEL); + if (!config_area) + goto out_free; + + memcpy(&vcdev->config[offset], buf, len); + /* Write the config area to the host. */ + memcpy(config_area, vcdev->config, sizeof(vcdev->config)); + ccw->cmd_code = CCW_CMD_WRITE_CONF; + ccw->flags = 0; + ccw->count = offset + len; + ccw->cda = (__u32)(unsigned long)config_area; + ccw_io_helper(vcdev, ccw, VIRTIO_CCW_DOING_WRITE_CONFIG); + +out_free: + kfree(config_area); + kfree(ccw); +} + +static u8 virtio_ccw_get_status(struct virtio_device *vdev) +{ + struct virtio_ccw_device *vcdev = to_vc_device(vdev); + + return *vcdev->status; +} + +static void virtio_ccw_set_status(struct virtio_device *vdev, u8 status) +{ + struct virtio_ccw_device *vcdev = to_vc_device(vdev); + u8 old_status = *vcdev->status; + struct ccw1 *ccw; + int ret; + + ccw = kzalloc(sizeof(*ccw), GFP_DMA | GFP_KERNEL); + if (!ccw) + return; + + /* Write the status to the host. */ + *vcdev->status = status; + ccw->cmd_code = CCW_CMD_WRITE_STATUS; + ccw->flags = 0; + ccw->count = sizeof(status); + ccw->cda = (__u32)(unsigned long)vcdev->status; + ret = ccw_io_helper(vcdev, ccw, VIRTIO_CCW_DOING_WRITE_STATUS); + /* Write failed? We assume status is unchanged. */ + if (ret) + *vcdev->status = old_status; + kfree(ccw); +} + +static struct virtio_config_ops virtio_ccw_config_ops = { + .get_features = virtio_ccw_get_features, + .finalize_features = virtio_ccw_finalize_features, + .get = virtio_ccw_get_config, + .set = virtio_ccw_set_config, + .get_status = virtio_ccw_get_status, + .set_status = virtio_ccw_set_status, + .reset = virtio_ccw_reset, + .find_vqs = virtio_ccw_find_vqs, + .del_vqs = virtio_ccw_del_vqs, +}; + + +/* + * ccw bus driver related functions + */ + +static void virtio_ccw_release_dev(struct device *_d) +{ + struct virtio_device *dev = container_of(_d, struct virtio_device, + dev); + struct virtio_ccw_device *vcdev = to_vc_device(dev); + + kfree(vcdev->status); + kfree(vcdev->config_block); + kfree(vcdev); +} + +static int irb_is_error(struct irb *irb) +{ + if (scsw_cstat(&irb->scsw) != 0) + return 1; + if (scsw_dstat(&irb->scsw) & ~(DEV_STAT_CHN_END | DEV_STAT_DEV_END)) + return 1; + if (scsw_cc(&irb->scsw) != 0) + return 1; + return 0; +} + +static struct virtqueue *virtio_ccw_vq_by_ind(struct virtio_ccw_device *vcdev, + int index) +{ + struct virtio_ccw_vq_info *info; + unsigned long flags; + struct virtqueue *vq; + + vq = NULL; + spin_lock_irqsave(&vcdev->lock, flags); + list_for_each_entry(info, &vcdev->virtqueues, node) { + if (info->vq->index == index) { + vq = info->vq; + break; + } + } + spin_unlock_irqrestore(&vcdev->lock, flags); + return vq; +} + +static void virtio_ccw_int_handler(struct ccw_device *cdev, + unsigned long intparm, + struct irb *irb) +{ + __u32 activity = intparm & VIRTIO_CCW_INTPARM_MASK; + struct virtio_ccw_device *vcdev = dev_get_drvdata(&cdev->dev); + int i; + struct virtqueue *vq; + + if (!vcdev) + return; + /* Check if it's a notification from the host. */ + if ((intparm == 0) && + (scsw_stctl(&irb->scsw) == + (SCSW_STCTL_ALERT_STATUS | SCSW_STCTL_STATUS_PEND))) { + /* OK */ + } + if (irb_is_error(irb)) { + /* Command reject? */ + if ((scsw_dstat(&irb->scsw) & DEV_STAT_UNIT_CHECK) && + (irb->ecw[0] & SNS0_CMD_REJECT)) + vcdev->err = -EOPNOTSUPP; + else + /* Map everything else to -EIO. */ + vcdev->err = -EIO; + } + if (vcdev->curr_io & activity) { + switch (activity) { + case VIRTIO_CCW_DOING_READ_FEAT: + case VIRTIO_CCW_DOING_WRITE_FEAT: + case VIRTIO_CCW_DOING_READ_CONFIG: + case VIRTIO_CCW_DOING_WRITE_CONFIG: + case VIRTIO_CCW_DOING_WRITE_STATUS: + case VIRTIO_CCW_DOING_SET_VQ: + case VIRTIO_CCW_DOING_SET_IND: + case VIRTIO_CCW_DOING_SET_CONF_IND: + case VIRTIO_CCW_DOING_RESET: + case VIRTIO_CCW_DOING_READ_VQ_CONF: + case VIRTIO_CCW_DOING_SET_IND_ADAPTER: + case VIRTIO_CCW_DOING_SET_VIRTIO_REV: + vcdev->curr_io &= ~activity; + wake_up(&vcdev->wait_q); + break; + default: + /* don't know what to do... */ + dev_warn(&cdev->dev, "Suspicious activity '%08x'\n", + activity); + WARN_ON(1); + break; + } + } + for_each_set_bit(i, &vcdev->indicators, + sizeof(vcdev->indicators) * BITS_PER_BYTE) { + /* The bit clear must happen before the vring kick. */ + clear_bit(i, &vcdev->indicators); + barrier(); + vq = virtio_ccw_vq_by_ind(vcdev, i); + vring_interrupt(0, vq); + } + if (test_bit(0, &vcdev->indicators2)) { + virtio_config_changed(&vcdev->vdev); + clear_bit(0, &vcdev->indicators2); + } +} + +/* + * We usually want to autoonline all devices, but give the admin + * a way to exempt devices from this. + */ +#define __DEV_WORDS ((__MAX_SUBCHANNEL + (8*sizeof(long) - 1)) / \ + (8*sizeof(long))) +static unsigned long devs_no_auto[__MAX_SSID + 1][__DEV_WORDS]; + +static char *no_auto = ""; + +module_param(no_auto, charp, 0444); +MODULE_PARM_DESC(no_auto, "list of ccw bus id ranges not to be auto-onlined"); + +static int virtio_ccw_check_autoonline(struct ccw_device *cdev) +{ + struct ccw_dev_id id; + + ccw_device_get_id(cdev, &id); + if (test_bit(id.devno, devs_no_auto[id.ssid])) + return 0; + return 1; +} + +static void virtio_ccw_auto_online(void *data, async_cookie_t cookie) +{ + struct ccw_device *cdev = data; + int ret; + + ret = ccw_device_set_online(cdev); + if (ret) + dev_warn(&cdev->dev, "Failed to set online: %d\n", ret); +} + +static int virtio_ccw_probe(struct ccw_device *cdev) +{ + cdev->handler = virtio_ccw_int_handler; + + if (virtio_ccw_check_autoonline(cdev)) + async_schedule(virtio_ccw_auto_online, cdev); + return 0; +} + +static struct virtio_ccw_device *virtio_grab_drvdata(struct ccw_device *cdev) +{ + unsigned long flags; + struct virtio_ccw_device *vcdev; + + spin_lock_irqsave(get_ccwdev_lock(cdev), flags); + vcdev = dev_get_drvdata(&cdev->dev); + if (!vcdev || vcdev->going_away) { + spin_unlock_irqrestore(get_ccwdev_lock(cdev), flags); + return NULL; + } + vcdev->going_away = true; + spin_unlock_irqrestore(get_ccwdev_lock(cdev), flags); + return vcdev; +} + +static void virtio_ccw_remove(struct ccw_device *cdev) +{ + unsigned long flags; + struct virtio_ccw_device *vcdev = virtio_grab_drvdata(cdev); + + if (vcdev && cdev->online) { + if (vcdev->device_lost) + virtio_break_device(&vcdev->vdev); + unregister_virtio_device(&vcdev->vdev); + spin_lock_irqsave(get_ccwdev_lock(cdev), flags); + dev_set_drvdata(&cdev->dev, NULL); + spin_unlock_irqrestore(get_ccwdev_lock(cdev), flags); + } + cdev->handler = NULL; +} + +static int virtio_ccw_offline(struct ccw_device *cdev) +{ + unsigned long flags; + struct virtio_ccw_device *vcdev = virtio_grab_drvdata(cdev); + + if (!vcdev) + return 0; + if (vcdev->device_lost) + virtio_break_device(&vcdev->vdev); + unregister_virtio_device(&vcdev->vdev); + spin_lock_irqsave(get_ccwdev_lock(cdev), flags); + dev_set_drvdata(&cdev->dev, NULL); + spin_unlock_irqrestore(get_ccwdev_lock(cdev), flags); + return 0; +} + +static int virtio_ccw_set_transport_rev(struct virtio_ccw_device *vcdev) +{ + struct virtio_rev_info *rev; + struct ccw1 *ccw; + int ret; + + ccw = kzalloc(sizeof(*ccw), GFP_DMA | GFP_KERNEL); + if (!ccw) + return -ENOMEM; + rev = kzalloc(sizeof(*rev), GFP_DMA | GFP_KERNEL); + if (!rev) { + kfree(ccw); + return -ENOMEM; + } + + /* Set transport revision */ + ccw->cmd_code = CCW_CMD_SET_VIRTIO_REV; + ccw->flags = 0; + ccw->count = sizeof(*rev); + ccw->cda = (__u32)(unsigned long)rev; + + vcdev->revision = VIRTIO_CCW_REV_MAX; + do { + rev->revision = vcdev->revision; + /* none of our supported revisions carry payload */ + rev->length = 0; + ret = ccw_io_helper(vcdev, ccw, + VIRTIO_CCW_DOING_SET_VIRTIO_REV); + if (ret == -EOPNOTSUPP) { + if (vcdev->revision == 0) + /* + * The host device does not support setting + * the revision: let's operate it in legacy + * mode. + */ + ret = 0; + else + vcdev->revision--; + } + } while (ret == -EOPNOTSUPP); + + kfree(ccw); + kfree(rev); + return ret; +} + +static int virtio_ccw_online(struct ccw_device *cdev) +{ + int ret; + struct virtio_ccw_device *vcdev; + unsigned long flags; + + vcdev = kzalloc(sizeof(*vcdev), GFP_KERNEL); + if (!vcdev) { + dev_warn(&cdev->dev, "Could not get memory for virtio\n"); + ret = -ENOMEM; + goto out_free; + } + vcdev->config_block = kzalloc(sizeof(*vcdev->config_block), + GFP_DMA | GFP_KERNEL); + if (!vcdev->config_block) { + ret = -ENOMEM; + goto out_free; + } + vcdev->status = kzalloc(sizeof(*vcdev->status), GFP_DMA | GFP_KERNEL); + if (!vcdev->status) { + ret = -ENOMEM; + goto out_free; + } + + vcdev->is_thinint = virtio_ccw_use_airq; /* at least try */ + + vcdev->vdev.dev.parent = &cdev->dev; + vcdev->vdev.dev.release = virtio_ccw_release_dev; + vcdev->vdev.config = &virtio_ccw_config_ops; + vcdev->cdev = cdev; + init_waitqueue_head(&vcdev->wait_q); + INIT_LIST_HEAD(&vcdev->virtqueues); + spin_lock_init(&vcdev->lock); + + spin_lock_irqsave(get_ccwdev_lock(cdev), flags); + dev_set_drvdata(&cdev->dev, vcdev); + spin_unlock_irqrestore(get_ccwdev_lock(cdev), flags); + vcdev->vdev.id.vendor = cdev->id.cu_type; + vcdev->vdev.id.device = cdev->id.cu_model; + + ret = virtio_ccw_set_transport_rev(vcdev); + if (ret) + goto out_free; + + ret = register_virtio_device(&vcdev->vdev); + if (ret) { + dev_warn(&cdev->dev, "Failed to register virtio device: %d\n", + ret); + goto out_put; + } + return 0; +out_put: + spin_lock_irqsave(get_ccwdev_lock(cdev), flags); + dev_set_drvdata(&cdev->dev, NULL); + spin_unlock_irqrestore(get_ccwdev_lock(cdev), flags); + put_device(&vcdev->vdev.dev); + return ret; +out_free: + if (vcdev) { + kfree(vcdev->status); + kfree(vcdev->config_block); + } + kfree(vcdev); + return ret; +} + +static int virtio_ccw_cio_notify(struct ccw_device *cdev, int event) +{ + int rc; + struct virtio_ccw_device *vcdev = dev_get_drvdata(&cdev->dev); + + /* + * Make sure vcdev is set + * i.e. set_offline/remove callback not already running + */ + if (!vcdev) + return NOTIFY_DONE; + + switch (event) { + case CIO_GONE: + vcdev->device_lost = true; + rc = NOTIFY_DONE; + break; + default: + rc = NOTIFY_DONE; + break; + } + return rc; +} + +static struct ccw_device_id virtio_ids[] = { + { CCW_DEVICE(0x3832, 0) }, + {}, +}; +MODULE_DEVICE_TABLE(ccw, virtio_ids); + +static struct ccw_driver virtio_ccw_driver = { + .driver = { + .owner = THIS_MODULE, + .name = "virtio_ccw", + }, + .ids = virtio_ids, + .probe = virtio_ccw_probe, + .remove = virtio_ccw_remove, + .set_offline = virtio_ccw_offline, + .set_online = virtio_ccw_online, + .notify = virtio_ccw_cio_notify, + .int_class = IRQIO_VIR, +}; + +static int __init pure_hex(char **cp, unsigned int *val, int min_digit, + int max_digit, int max_val) +{ + int diff; + + diff = 0; + *val = 0; + + while (diff <= max_digit) { + int value = hex_to_bin(**cp); + + if (value < 0) + break; + *val = *val * 16 + value; + (*cp)++; + diff++; + } + + if ((diff < min_digit) || (diff > max_digit) || (*val > max_val)) + return 1; + + return 0; +} + +static int __init parse_busid(char *str, unsigned int *cssid, + unsigned int *ssid, unsigned int *devno) +{ + char *str_work; + int rc, ret; + + rc = 1; + + if (*str == '\0') + goto out; + + str_work = str; + ret = pure_hex(&str_work, cssid, 1, 2, __MAX_CSSID); + if (ret || (str_work[0] != '.')) + goto out; + str_work++; + ret = pure_hex(&str_work, ssid, 1, 1, __MAX_SSID); + if (ret || (str_work[0] != '.')) + goto out; + str_work++; + ret = pure_hex(&str_work, devno, 4, 4, __MAX_SUBCHANNEL); + if (ret || (str_work[0] != '\0')) + goto out; + + rc = 0; +out: + return rc; +} + +static void __init no_auto_parse(void) +{ + unsigned int from_cssid, to_cssid, from_ssid, to_ssid, from, to; + char *parm, *str; + int rc; + + str = no_auto; + while ((parm = strsep(&str, ","))) { + rc = parse_busid(strsep(&parm, "-"), &from_cssid, + &from_ssid, &from); + if (rc) + continue; + if (parm != NULL) { + rc = parse_busid(parm, &to_cssid, + &to_ssid, &to); + if ((from_ssid > to_ssid) || + ((from_ssid == to_ssid) && (from > to))) + rc = -EINVAL; + } else { + to_cssid = from_cssid; + to_ssid = from_ssid; + to = from; + } + if (rc) + continue; + while ((from_ssid < to_ssid) || + ((from_ssid == to_ssid) && (from <= to))) { + set_bit(from, devs_no_auto[from_ssid]); + from++; + if (from > __MAX_SUBCHANNEL) { + from_ssid++; + from = 0; + } + } + } +} + +static int __init virtio_ccw_init(void) +{ + /* parse no_auto string before we do anything further */ + no_auto_parse(); + return ccw_driver_register(&virtio_ccw_driver); +} +module_init(virtio_ccw_init); + +static void __exit virtio_ccw_exit(void) +{ + int i; + + ccw_driver_unregister(&virtio_ccw_driver); + for (i = 0; i < MAX_AIRQ_AREAS; i++) + destroy_airq_info(airq_areas[i]); +} +module_exit(virtio_ccw_exit); -- cgit v0.10.2 From 6fec919b61b66e7f92646a7bd6fada9850f5cedc Mon Sep 17 00:00:00 2001 From: Jiri Prchal Date: Thu, 25 Jun 2015 13:44:19 +0200 Subject: spi: spidev: add compatible value for LTC2488 Since spidev is no more allowed to use in DT and is really loudly warned about it I'd like to add this compatible value. (Geert Uytterhoeven wrote: "Add the compatible value for your device to the spidev_dt_ids[] array in drivers/spi/spidev.c.") Signed-off-by: Mark Brown diff --git a/drivers/spi/spidev.c b/drivers/spi/spidev.c index dd616ff..c7de641 100644 --- a/drivers/spi/spidev.c +++ b/drivers/spi/spidev.c @@ -693,6 +693,7 @@ static struct class *spidev_class; #ifdef CONFIG_OF static const struct of_device_id spidev_dt_ids[] = { { .compatible = "rohm,dh2228fv" }, + { .compatible = "lineartechnology,ltc2488" }, {}, }; MODULE_DEVICE_TABLE(of, spidev_dt_ids); -- cgit v0.10.2 From 2e1c75f4d3ecf127032f2511aba76927a8703c1d Mon Sep 17 00:00:00 2001 From: Geert Uytterhoeven Date: Fri, 26 Jun 2015 14:07:12 +0200 Subject: spi: SPI_ZYNQMP_GQSPI should depend on HAS_DMA If NO_DMA=y: ERROR: "dma_unmap_single" [drivers/spi/spi-zynqmp-gqspi.ko] undefined! ERROR: "dma_mapping_error" [drivers/spi/spi-zynqmp-gqspi.ko] undefined! ERROR: "dma_map_single" [drivers/spi/spi-zynqmp-gqspi.ko] undefined! Add a dependency on HAS_DMA to fix this. Signed-off-by: Geert Uytterhoeven Acked-by: Ranjit Waghmode Signed-off-by: Mark Brown diff --git a/drivers/spi/Kconfig b/drivers/spi/Kconfig index 0cae169..b0f30fb 100644 --- a/drivers/spi/Kconfig +++ b/drivers/spi/Kconfig @@ -612,7 +612,7 @@ config SPI_XTENSA_XTFPGA config SPI_ZYNQMP_GQSPI tristate "Xilinx ZynqMP GQSPI controller" - depends on SPI_MASTER + depends on SPI_MASTER && HAS_DMA help Enables Xilinx GQSPI controller driver for Zynq UltraScale+ MPSoC. -- cgit v0.10.2 From 127e10624232c5d605cad840d21af3b842541719 Mon Sep 17 00:00:00 2001 From: Laxman Dewangan Date: Wed, 1 Jul 2015 18:31:42 +0530 Subject: regulator: max8973: Fix up control flag option for bias control The control flag for the bias control is MAX8973_CONTROL_BIAS_ENABLE rather than MAX8973_BIAS_ENABLE which is macro for the bits in register. Fix this typo. Signed-off-by: Laxman Dewangan Signed-off-by: Mark Brown diff --git a/drivers/regulator/max8973-regulator.c b/drivers/regulator/max8973-regulator.c index 6f2bdad..e94ddcf 100644 --- a/drivers/regulator/max8973-regulator.c +++ b/drivers/regulator/max8973-regulator.c @@ -450,7 +450,7 @@ static struct max8973_regulator_platform_data *max8973_parse_dt( pdata->control_flags |= MAX8973_CONTROL_FREQ_SHIFT_9PER_ENABLE; if (of_property_read_bool(np, "maxim,enable-bias-control")) - pdata->control_flags |= MAX8973_BIAS_ENABLE; + pdata->control_flags |= MAX8973_CONTROL_BIAS_ENABLE; return pdata; } -- cgit v0.10.2 From 861a481c5ede75bf31bc118f1f7f6d1f420182b5 Mon Sep 17 00:00:00 2001 From: Dan Carpenter Date: Wed, 24 Jun 2015 17:31:33 +0300 Subject: spi: zynq: missing break statement There is a missing break statement here so selecting both only selects upper. Fixes: dfe11a11d523 ('spi: Add support for Zynq Ultrascale+ MPSoC GQSPI controller') Signed-off-by: Dan Carpenter Acked-by: Michal Simek Signed-off-by: Mark Brown diff --git a/drivers/spi/spi-zynqmp-gqspi.c b/drivers/spi/spi-zynqmp-gqspi.c index 87b20a5..f23f36e 100644 --- a/drivers/spi/spi-zynqmp-gqspi.c +++ b/drivers/spi/spi-zynqmp-gqspi.c @@ -214,6 +214,7 @@ static void zynqmp_gqspi_selectslave(struct zynqmp_qspi *instanceptr, case GQSPI_SELECT_FLASH_CS_BOTH: instanceptr->genfifocs = GQSPI_GENFIFO_CS_LOWER | GQSPI_GENFIFO_CS_UPPER; + break; case GQSPI_SELECT_FLASH_CS_UPPER: instanceptr->genfifocs = GQSPI_GENFIFO_CS_UPPER; break; -- cgit v0.10.2 From 5f867db63473f32cce1b868e281ebd42a41f8fad Mon Sep 17 00:00:00 2001 From: Scott Wood Date: Fri, 26 Jun 2015 19:43:58 -0500 Subject: mtd: nand: Fix NAND_USE_BOUNCE_BUFFER flag conflict Commit 66507c7bc8895f0da6b ("mtd: nand: Add support to use nand_base poi databuf as bounce buffer") added a flag NAND_USE_BOUNCE_BUFFER using the same bit value as the existing NAND_BUSWIDTH_AUTO. Cc: Kamal Dasu Fixes: 66507c7bc8895f0da6b ("mtd: nand: Add support to use nand_base poi databuf as bounce buffer") Signed-off-by: Scott Wood Signed-off-by: Brian Norris diff --git a/include/linux/mtd/nand.h b/include/linux/mtd/nand.h index f25e2bd..272f429 100644 --- a/include/linux/mtd/nand.h +++ b/include/linux/mtd/nand.h @@ -178,17 +178,17 @@ typedef enum { /* Chip may not exist, so silence any errors in scan */ #define NAND_SCAN_SILENT_NODEV 0x00040000 /* - * This option could be defined by controller drivers to protect against - * kmap'ed, vmalloc'ed highmem buffers being passed from upper layers - */ -#define NAND_USE_BOUNCE_BUFFER 0x00080000 -/* * Autodetect nand buswidth with readid/onfi. * This suppose the driver will configure the hardware in 8 bits mode * when calling nand_scan_ident, and update its configuration * before calling nand_scan_tail. */ #define NAND_BUSWIDTH_AUTO 0x00080000 +/* + * This option could be defined by controller drivers to protect against + * kmap'ed, vmalloc'ed highmem buffers being passed from upper layers + */ +#define NAND_USE_BOUNCE_BUFFER 0x00100000 /* Options set by nand scan */ /* Nand scan has allocated controller struct */ -- cgit v0.10.2 From 66337b7c67f6237720ba13f6d41b9d8dbcb59cda Mon Sep 17 00:00:00 2001 From: "Dreyfuss, Haim" Date: Thu, 4 Jun 2015 11:45:33 +0300 Subject: iwlwifi: pcie: Fix bug in NIC's PM registers access While cleanig the access to those hw-dependent registers, instead of using the product family type, wrong condition was added mistakenly and enabled 8000 family devices a forbidden access to HW registers, fix it. Fixes: 95411d0455cc ("iwlwifi: pcie: Control access to the NIC's PM registers via iwl_cfg") Signed-off-by: Dreyfuss, Haim Signed-off-by: Emmanuel Grumbach diff --git a/drivers/net/wireless/iwlwifi/pcie/trans.c b/drivers/net/wireless/iwlwifi/pcie/trans.c index 43ae658..608ba1e 100644 --- a/drivers/net/wireless/iwlwifi/pcie/trans.c +++ b/drivers/net/wireless/iwlwifi/pcie/trans.c @@ -182,7 +182,7 @@ static void iwl_trans_pcie_write_shr(struct iwl_trans *trans, u32 reg, u32 val) static void iwl_pcie_set_pwr(struct iwl_trans *trans, bool vaux) { - if (!trans->cfg->apmg_not_supported) + if (trans->cfg->apmg_not_supported) return; if (vaux && pci_pme_capable(to_pci_dev(trans->dev), PCI_D3cold)) -- cgit v0.10.2 From af3f2f740173f8b5c61dba46f900998c5984ccd9 Mon Sep 17 00:00:00 2001 From: Emmanuel Grumbach Date: Thu, 4 Jun 2015 09:51:11 +0300 Subject: iwlwifi: pcie: don't panic if pcie transport alloc fails iwl_trans_pcie_alloc needs to return a non-zero value if it fails. Otherwise the iwl_drv_start will think that the allocation succeeded. Remove the duplication of err and ret variable and use ret which is the name we usually use in other places of the driver. Fixes: c278754a21e6 ("iwlwifi: mvm: support family 8000 B2/C steps") CC: [4.1] Signed-off-by: Emmanuel Grumbach diff --git a/drivers/net/wireless/iwlwifi/pcie/trans.c b/drivers/net/wireless/iwlwifi/pcie/trans.c index 608ba1e..fe61d0a 100644 --- a/drivers/net/wireless/iwlwifi/pcie/trans.c +++ b/drivers/net/wireless/iwlwifi/pcie/trans.c @@ -2459,7 +2459,7 @@ struct iwl_trans *iwl_trans_pcie_alloc(struct pci_dev *pdev, struct iwl_trans_pcie *trans_pcie; struct iwl_trans *trans; u16 pci_cmd; - int err; + int ret; trans = iwl_trans_alloc(sizeof(struct iwl_trans_pcie), &pdev->dev, cfg, &trans_ops_pcie, 0); @@ -2474,8 +2474,8 @@ struct iwl_trans *iwl_trans_pcie_alloc(struct pci_dev *pdev, spin_lock_init(&trans_pcie->ref_lock); init_waitqueue_head(&trans_pcie->ucode_write_waitq); - err = pci_enable_device(pdev); - if (err) + ret = pci_enable_device(pdev); + if (ret) goto out_no_pci; if (!cfg->base_params->pcie_l1_allowed) { @@ -2491,23 +2491,23 @@ struct iwl_trans *iwl_trans_pcie_alloc(struct pci_dev *pdev, pci_set_master(pdev); - err = pci_set_dma_mask(pdev, DMA_BIT_MASK(36)); - if (!err) - err = pci_set_consistent_dma_mask(pdev, DMA_BIT_MASK(36)); - if (err) { - err = pci_set_dma_mask(pdev, DMA_BIT_MASK(32)); - if (!err) - err = pci_set_consistent_dma_mask(pdev, + ret = pci_set_dma_mask(pdev, DMA_BIT_MASK(36)); + if (!ret) + ret = pci_set_consistent_dma_mask(pdev, DMA_BIT_MASK(36)); + if (ret) { + ret = pci_set_dma_mask(pdev, DMA_BIT_MASK(32)); + if (!ret) + ret = pci_set_consistent_dma_mask(pdev, DMA_BIT_MASK(32)); /* both attempts failed: */ - if (err) { + if (ret) { dev_err(&pdev->dev, "No suitable DMA available\n"); goto out_pci_disable_device; } } - err = pci_request_regions(pdev, DRV_NAME); - if (err) { + ret = pci_request_regions(pdev, DRV_NAME); + if (ret) { dev_err(&pdev->dev, "pci_request_regions failed\n"); goto out_pci_disable_device; } @@ -2515,7 +2515,7 @@ struct iwl_trans *iwl_trans_pcie_alloc(struct pci_dev *pdev, trans_pcie->hw_base = pci_ioremap_bar(pdev, 0); if (!trans_pcie->hw_base) { dev_err(&pdev->dev, "pci_ioremap_bar failed\n"); - err = -ENODEV; + ret = -ENODEV; goto out_pci_release_regions; } @@ -2527,9 +2527,9 @@ struct iwl_trans *iwl_trans_pcie_alloc(struct pci_dev *pdev, trans_pcie->pci_dev = pdev; iwl_disable_interrupts(trans); - err = pci_enable_msi(pdev); - if (err) { - dev_err(&pdev->dev, "pci_enable_msi failed(0X%x)\n", err); + ret = pci_enable_msi(pdev); + if (ret) { + dev_err(&pdev->dev, "pci_enable_msi failed(0X%x)\n", ret); /* enable rfkill interrupt: hw bug w/a */ pci_read_config_word(pdev, PCI_COMMAND, &pci_cmd); if (pci_cmd & PCI_COMMAND_INTX_DISABLE) { @@ -2547,7 +2547,6 @@ struct iwl_trans *iwl_trans_pcie_alloc(struct pci_dev *pdev, */ if (trans->cfg->device_family == IWL_DEVICE_FAMILY_8000) { unsigned long flags; - int ret; trans->hw_rev = (trans->hw_rev & 0xfff0) | (CSR_HW_REV_STEP(trans->hw_rev << 2) << 2); @@ -2591,13 +2590,14 @@ struct iwl_trans *iwl_trans_pcie_alloc(struct pci_dev *pdev, /* Initialize the wait queue for commands */ init_waitqueue_head(&trans_pcie->wait_command_queue); - if (iwl_pcie_alloc_ict(trans)) + ret = iwl_pcie_alloc_ict(trans); + if (ret) goto out_pci_disable_msi; - err = request_threaded_irq(pdev->irq, iwl_pcie_isr, + ret = request_threaded_irq(pdev->irq, iwl_pcie_isr, iwl_pcie_irq_handler, IRQF_SHARED, DRV_NAME, trans); - if (err) { + if (ret) { IWL_ERR(trans, "Error allocating IRQ %d\n", pdev->irq); goto out_free_ict; } @@ -2617,5 +2617,5 @@ out_pci_disable_device: pci_disable_device(pdev); out_no_pci: iwl_trans_free(trans); - return ERR_PTR(err); + return ERR_PTR(ret); } -- cgit v0.10.2 From 1cc1cc92c4c4891abc48a777fb9fbc69077d5673 Mon Sep 17 00:00:00 2001 From: Brent Adam Date: Fri, 19 Jun 2015 10:53:35 -0600 Subject: HID: multitouch: Fix fields from pen report ID being interpreted for multitouch Fields like HID_DG_CONTACTCOUNT are outside of the physical collection, but within the application collection and report ID. Make sure to catch those fields that are not part of the mt_report_id and return 0 so they can be processed with the pen. Otherwise, the wrong HID_DG_CONTACTCOUNT will be applied to cc_index and result in dereferencing a null pointer in mt_touch_report. Signed-off-by: Brent Adam Signed-off-by: Jiri Kosina diff --git a/drivers/hid/hid-multitouch.c b/drivers/hid/hid-multitouch.c index 6a9b05b..7c81125 100644 --- a/drivers/hid/hid-multitouch.c +++ b/drivers/hid/hid-multitouch.c @@ -778,9 +778,16 @@ static int mt_input_mapping(struct hid_device *hdev, struct hid_input *hi, /* * some egalax touchscreens have "application == HID_DG_TOUCHSCREEN" * for the stylus. + * The check for mt_report_id ensures we don't process + * HID_DG_CONTACTCOUNT from the pen report as it is outside the physical + * collection, but within the report ID. */ if (field->physical == HID_DG_STYLUS) return 0; + else if ((field->physical == 0) && + (field->report->id != td->mt_report_id) && + (td->mt_report_id != -1)) + return 0; if (field->application == HID_DG_TOUCHSCREEN || field->application == HID_DG_TOUCHPAD) -- cgit v0.10.2 From 9633920e5ef65f81d30f2acd642d6d1c25cf7bc7 Mon Sep 17 00:00:00 2001 From: Jason Gerecke Date: Wed, 1 Jul 2015 13:01:00 -0700 Subject: HID: wacom: Enable pad device for older Bamboo Touch tablets Commit 862cf55 ("HID: wacom: Introduce a new WACOM_DEVICETYPE_PAD device_type") neglected to set the WACOM_DEVICETYPE_PAD flag for older two-finger Bamboo Touch tablets. Not only does this result in the pad device not appearing when such a tablet is plugged in, but also causes a segfault when 'wacom_bpt_touch' tries to send pad events. This patch adds the flag to resolve these issues. Signed-off-by: Jason Gerecke Signed-off-by: Jiri Kosina diff --git a/drivers/hid/wacom_wac.c b/drivers/hid/wacom_wac.c index 232da89..0d24423 100644 --- a/drivers/hid/wacom_wac.c +++ b/drivers/hid/wacom_wac.c @@ -2213,6 +2213,9 @@ void wacom_setup_device_quirks(struct wacom *wacom) features->x_max = 4096; features->y_max = 4096; } + else if (features->pktlen == WACOM_PKGLEN_BBTOUCH) { + features->device_type |= WACOM_DEVICETYPE_PAD; + } } /* -- cgit v0.10.2 From 6debce6f4e787a8eb4cec94e7afa85fb4f40db27 Mon Sep 17 00:00:00 2001 From: Antonio Borneo Date: Sun, 21 Jun 2015 14:20:25 +0800 Subject: HID: cp2112: fix to force single data-report reply Current implementation of cp2112_raw_event() only accepts one data report at a time. If last received data report is not fully handled yet, a new incoming data report will overwrite it. In such case we don't guaranteed to propagate the correct incoming data. The trivial fix implemented here forces a single report at a time by requesting in cp2112_read() no more than 61 byte of data, which is the payload size of a single data report. Cc: stable@vger.kernel.org Signed-off-by: Antonio Borneo Tested-by: Ellen Wang Signed-off-by: Jiri Kosina diff --git a/drivers/hid/hid-cp2112.c b/drivers/hid/hid-cp2112.c index 3318de6..a2dbbbe 100644 --- a/drivers/hid/hid-cp2112.c +++ b/drivers/hid/hid-cp2112.c @@ -356,6 +356,8 @@ static int cp2112_read(struct cp2112_device *dev, u8 *data, size_t size) struct cp2112_force_read_report report; int ret; + if (size > sizeof(dev->read_data)) + size = sizeof(dev->read_data); report.report = CP2112_DATA_READ_FORCE_SEND; report.length = cpu_to_be16(size); -- cgit v0.10.2 From 69146e7bfc38139a134c79a4ee6607c881891786 Mon Sep 17 00:00:00 2001 From: Zhen Lei Date: Fri, 26 Jun 2015 09:32:58 +0100 Subject: iommu/arm-smmu: Fix the index calculation of strtab The element size of cfg->strtab is just one DWORD, so we should use a multiply operation instead of a shift when calculating the level 1 index. Signed-off-by: Zhen Lei Signed-off-by: Will Deacon diff --git a/drivers/iommu/arm-smmu-v3.c b/drivers/iommu/arm-smmu-v3.c index 8e9ec81..606852f 100644 --- a/drivers/iommu/arm-smmu-v3.c +++ b/drivers/iommu/arm-smmu-v3.c @@ -1064,7 +1064,7 @@ static int arm_smmu_init_l2_strtab(struct arm_smmu_device *smmu, u32 sid) return 0; size = 1 << (STRTAB_SPLIT + ilog2(STRTAB_STE_DWORDS) + 3); - strtab = &cfg->strtab[sid >> STRTAB_SPLIT << STRTAB_L1_DESC_DWORDS]; + strtab = &cfg->strtab[(sid >> STRTAB_SPLIT) * STRTAB_L1_DESC_DWORDS]; desc->span = STRTAB_SPLIT + 1; desc->l2ptr = dma_zalloc_coherent(smmu->dev, size, &desc->l2ptr_dma, -- cgit v0.10.2 From d2e88e7c081efb2c5a9e1adb2a065d373167af4b Mon Sep 17 00:00:00 2001 From: Will Deacon Date: Tue, 30 Jun 2015 10:02:28 +0100 Subject: iommu/arm-smmu: Fix LOG2SIZE setting for 2-level stream tables STRTAB_BASE_CFG.LOG2SIZE should be set to log2(entries), where entries is the *total* number of entries in the stream table, not just the first level. This patch fixes the register setting, which was previously being set to the size of the l1 thanks to a multi-use "size" variable. Reported-by: Zhen Lei Tested-by: Zhen Lei Signed-off-by: Will Deacon diff --git a/drivers/iommu/arm-smmu-v3.c b/drivers/iommu/arm-smmu-v3.c index 606852f..6b1ae4e 100644 --- a/drivers/iommu/arm-smmu-v3.c +++ b/drivers/iommu/arm-smmu-v3.c @@ -2020,21 +2020,23 @@ static int arm_smmu_init_strtab_2lvl(struct arm_smmu_device *smmu) { void *strtab; u64 reg; - u32 size; + u32 size, l1size; int ret; struct arm_smmu_strtab_cfg *cfg = &smmu->strtab_cfg; /* Calculate the L1 size, capped to the SIDSIZE */ size = STRTAB_L1_SZ_SHIFT - (ilog2(STRTAB_L1_DESC_DWORDS) + 3); size = min(size, smmu->sid_bits - STRTAB_SPLIT); - if (size + STRTAB_SPLIT < smmu->sid_bits) + cfg->num_l1_ents = 1 << size; + + size += STRTAB_SPLIT; + if (size < smmu->sid_bits) dev_warn(smmu->dev, "2-level strtab only covers %u/%u bits of SID\n", - size + STRTAB_SPLIT, smmu->sid_bits); + size, smmu->sid_bits); - cfg->num_l1_ents = 1 << size; - size = cfg->num_l1_ents * (STRTAB_L1_DESC_DWORDS << 3); - strtab = dma_zalloc_coherent(smmu->dev, size, &cfg->strtab_dma, + l1size = cfg->num_l1_ents * (STRTAB_L1_DESC_DWORDS << 3); + strtab = dma_zalloc_coherent(smmu->dev, l1size, &cfg->strtab_dma, GFP_KERNEL); if (!strtab) { dev_err(smmu->dev, @@ -2055,8 +2057,7 @@ static int arm_smmu_init_strtab_2lvl(struct arm_smmu_device *smmu) ret = arm_smmu_init_l1_strtab(smmu); if (ret) dma_free_coherent(smmu->dev, - cfg->num_l1_ents * - (STRTAB_L1_DESC_DWORDS << 3), + l1size, strtab, cfg->strtab_dma); return ret; -- cgit v0.10.2 From 5d58c6207c300340151931ad9c2cdea2d1685dc4 Mon Sep 17 00:00:00 2001 From: Zhen Lei Date: Fri, 26 Jun 2015 09:32:59 +0100 Subject: iommu/arm-smmu: Fix the values of ARM64_TCR_{I,O}RGN0_SHIFT The arm64 CPU architecture defines TCR[8:11] as holding the inner and outer memory attributes for TTBR0. This patch fixes the ARM SMMUv3 driver to pack these bits into the context descriptor, rather than picking up the TTBR1 attributes as it currently does. Signed-off-by: Zhen Lei Signed-off-by: Will Deacon diff --git a/drivers/iommu/arm-smmu-v3.c b/drivers/iommu/arm-smmu-v3.c index 6b1ae4e..98e987a 100644 --- a/drivers/iommu/arm-smmu-v3.c +++ b/drivers/iommu/arm-smmu-v3.c @@ -269,10 +269,10 @@ #define ARM64_TCR_TG0_SHIFT 14 #define ARM64_TCR_TG0_MASK 0x3UL #define CTXDESC_CD_0_TCR_IRGN0_SHIFT 8 -#define ARM64_TCR_IRGN0_SHIFT 24 +#define ARM64_TCR_IRGN0_SHIFT 8 #define ARM64_TCR_IRGN0_MASK 0x3UL #define CTXDESC_CD_0_TCR_ORGN0_SHIFT 10 -#define ARM64_TCR_ORGN0_SHIFT 26 +#define ARM64_TCR_ORGN0_SHIFT 10 #define ARM64_TCR_ORGN0_MASK 0x3UL #define CTXDESC_CD_0_TCR_SH0_SHIFT 12 #define ARM64_TCR_SH0_SHIFT 12 -- cgit v0.10.2 From e2f4c2330f08ba73d9a3c919a3d6ca33dce7d2c2 Mon Sep 17 00:00:00 2001 From: Zhen Lei Date: Tue, 7 Jul 2015 04:30:17 +0100 Subject: iommu/arm-smmu: Enlarge STRTAB_L1_SZ_SHIFT to support larger sidsize Because we will choose the minimum value between STRTAB_L1_SZ_SHIFT and IDR1.SIDSIZE, so enlarge STRTAB_L1_SZ_SHIFT will not impact the platforms whose IDR1.SIDSIZE is smaller than old STRTAB_L1_SZ_SHIFT value. Signed-off-by: Zhen Lei Signed-off-by: Will Deacon diff --git a/drivers/iommu/arm-smmu-v3.c b/drivers/iommu/arm-smmu-v3.c index 98e987a..29cba32 100644 --- a/drivers/iommu/arm-smmu-v3.c +++ b/drivers/iommu/arm-smmu-v3.c @@ -199,9 +199,10 @@ * Stream table. * * Linear: Enough to cover 1 << IDR1.SIDSIZE entries - * 2lvl: 8k L1 entries, 256 lazy entries per table (each table covers a PCI bus) + * 2lvl: 128k L1 entries, + * 256 lazy entries per table (each table covers a PCI bus) */ -#define STRTAB_L1_SZ_SHIFT 16 +#define STRTAB_L1_SZ_SHIFT 20 #define STRTAB_SPLIT 8 #define STRTAB_L1_DESC_DWORDS 1 -- cgit v0.10.2 From 5e92946c39ca6abc65e34775a93cc1d1a819c0e3 Mon Sep 17 00:00:00 2001 From: Zhen Lei Date: Tue, 7 Jul 2015 04:30:18 +0100 Subject: iommu/arm-smmu: Skip the execution of CMD_PREFETCH_CONFIG Hisilicon SMMUv3 devices treat CMD_PREFETCH_CONFIG as a illegal command, execute it will trigger GERROR interrupt. Although the gerror code manage to turn the prefetch into a SYNC, and the system can continue to run normally, but it's ugly to print error information. Signed-off-by: Zhen Lei [will: extended binding documentation] Signed-off-by: Will Deacon diff --git a/Documentation/devicetree/bindings/iommu/arm,smmu-v3.txt b/Documentation/devicetree/bindings/iommu/arm,smmu-v3.txt index c03eec1..3443e0f 100644 --- a/Documentation/devicetree/bindings/iommu/arm,smmu-v3.txt +++ b/Documentation/devicetree/bindings/iommu/arm,smmu-v3.txt @@ -35,3 +35,6 @@ the PCIe specification. NOTE: this only applies to the SMMU itself, not masters connected upstream of the SMMU. + +- hisilicon,broken-prefetch-cmd + : Avoid sending CMD_PREFETCH_* commands to the SMMU. diff --git a/drivers/iommu/arm-smmu-v3.c b/drivers/iommu/arm-smmu-v3.c index 29cba32..da902ba 100644 --- a/drivers/iommu/arm-smmu-v3.c +++ b/drivers/iommu/arm-smmu-v3.c @@ -543,6 +543,9 @@ struct arm_smmu_device { #define ARM_SMMU_FEAT_HYP (1 << 12) u32 features; +#define ARM_SMMU_OPT_SKIP_PREFETCH (1 << 0) + u32 options; + struct arm_smmu_cmdq cmdq; struct arm_smmu_evtq evtq; struct arm_smmu_priq priq; @@ -603,11 +606,35 @@ struct arm_smmu_domain { static DEFINE_SPINLOCK(arm_smmu_devices_lock); static LIST_HEAD(arm_smmu_devices); +struct arm_smmu_option_prop { + u32 opt; + const char *prop; +}; + +static struct arm_smmu_option_prop arm_smmu_options[] = { + { ARM_SMMU_OPT_SKIP_PREFETCH, "hisilicon,broken-prefetch-cmd" }, + { 0, NULL}, +}; + static struct arm_smmu_domain *to_smmu_domain(struct iommu_domain *dom) { return container_of(dom, struct arm_smmu_domain, domain); } +static void parse_driver_options(struct arm_smmu_device *smmu) +{ + int i = 0; + + do { + if (of_property_read_bool(smmu->dev->of_node, + arm_smmu_options[i].prop)) { + smmu->options |= arm_smmu_options[i].opt; + dev_notice(smmu->dev, "option %s\n", + arm_smmu_options[i].prop); + } + } while (arm_smmu_options[++i].opt); +} + /* Low-level queue manipulation functions */ static bool queue_full(struct arm_smmu_queue *q) { @@ -1037,7 +1064,8 @@ static void arm_smmu_write_strtab_ent(struct arm_smmu_device *smmu, u32 sid, arm_smmu_sync_ste_for_sid(smmu, sid); /* It's likely that we'll want to use the new STE soon */ - arm_smmu_cmdq_issue_cmd(smmu, &prefetch_cmd); + if (!(smmu->options & ARM_SMMU_OPT_SKIP_PREFETCH)) + arm_smmu_cmdq_issue_cmd(smmu, &prefetch_cmd); } static void arm_smmu_init_bypass_stes(u64 *strtab, unsigned int nent) @@ -2575,6 +2603,8 @@ static int arm_smmu_device_dt_probe(struct platform_device *pdev) if (irq > 0) smmu->gerr_irq = irq; + parse_driver_options(smmu); + /* Probe the h/w */ ret = arm_smmu_device_probe(smmu); if (ret) -- cgit v0.10.2 From 690edcfad0e570a9a2a42d9b98cd04efaf21489d Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Wed, 8 Jul 2015 20:19:23 +0200 Subject: NFSv4.2/flexfiles: Fix a typo in the flexfiles layoutstats code Signed-off-by: Trond Myklebust diff --git a/fs/nfs/flexfilelayout/flexfilelayout.c b/fs/nfs/flexfilelayout/flexfilelayout.c index c12951b..b3289d7 100644 --- a/fs/nfs/flexfilelayout/flexfilelayout.c +++ b/fs/nfs/flexfilelayout/flexfilelayout.c @@ -1852,7 +1852,7 @@ ff_layout_mirror_prepare_stats(struct nfs42_layoutstat_args *args, struct nfs42_layoutstat_devinfo *devinfo; int i; - for (i = 0; i <= FF_LAYOUT_MIRROR_COUNT(pls); i++) { + for (i = 0; i < FF_LAYOUT_MIRROR_COUNT(pls); i++) { if (*dev_count >= dev_limit) break; mirror = FF_LAYOUT_COMP(pls, i); -- cgit v0.10.2 From 70f19f5869c5accfd9f371c099f21c71516591b2 Mon Sep 17 00:00:00 2001 From: "Steven J. Magnani" Date: Tue, 7 Jul 2015 13:06:05 -0500 Subject: udf: Don't corrupt unalloc spacetable when writing it For a UDF filesystem configured with an Unallocated Space Table, a filesystem operation that triggers an update to the table results in on-disk corruption that prevents remounting: udf_read_tagged: tag version 0x0000 != 0x0002 || 0x0003, block 274 For example: 1. Create a filesystem $ mkudffs --media-type=hd --blocksize=512 --lvid=BUGTEST \ --vid=BUGTEST --fsid=BUGTEST --space=unalloctable \ /dev/mmcblk0 2. Mount it # mount /dev/mmcblk0 /mnt 3. Create a file $ echo "No corruption, please" > /mnt/new.file 4. Umount # umount /mnt 5. Attempt remount # mount /dev/mmcblk0 /mnt This appears to be a longstanding bug caused by zero-initialization of the Unallocated Space Entry block buffer and only partial repopulation of required fields before writing to disk. Commit 0adfb339fd64 ("udf: Fix unalloc space handling in udf_update_inode") addressed one such field, but several others are required. Signed-off-by: Steven J. Magnani Signed-off-by: Jan Kara diff --git a/fs/udf/inode.c b/fs/udf/inode.c index 6afac3d..8d0b3ad 100644 --- a/fs/udf/inode.c +++ b/fs/udf/inode.c @@ -1652,17 +1652,9 @@ static int udf_update_inode(struct inode *inode, int do_sync) iinfo->i_ext.i_data, inode->i_sb->s_blocksize - sizeof(struct unallocSpaceEntry)); use->descTag.tagIdent = cpu_to_le16(TAG_IDENT_USE); - use->descTag.tagLocation = - cpu_to_le32(iinfo->i_location.logicalBlockNum); - crclen = sizeof(struct unallocSpaceEntry) + - iinfo->i_lenAlloc - sizeof(struct tag); - use->descTag.descCRCLength = cpu_to_le16(crclen); - use->descTag.descCRC = cpu_to_le16(crc_itu_t(0, (char *)use + - sizeof(struct tag), - crclen)); - use->descTag.tagChecksum = udf_tag_checksum(&use->descTag); + crclen = sizeof(struct unallocSpaceEntry); - goto out; + goto finish; } if (UDF_QUERY_FLAG(inode->i_sb, UDF_FLAG_UID_FORGET)) @@ -1782,6 +1774,8 @@ static int udf_update_inode(struct inode *inode, int do_sync) efe->descTag.tagIdent = cpu_to_le16(TAG_IDENT_EFE); crclen = sizeof(struct extendedFileEntry); } + +finish: if (iinfo->i_strat4096) { fe->icbTag.strategyType = cpu_to_le16(4096); fe->icbTag.strategyParameter = cpu_to_le16(1); @@ -1791,7 +1785,9 @@ static int udf_update_inode(struct inode *inode, int do_sync) fe->icbTag.numEntries = cpu_to_le16(1); } - if (S_ISDIR(inode->i_mode)) + if (iinfo->i_use) + fe->icbTag.fileType = ICBTAG_FILE_TYPE_USE; + else if (S_ISDIR(inode->i_mode)) fe->icbTag.fileType = ICBTAG_FILE_TYPE_DIRECTORY; else if (S_ISREG(inode->i_mode)) fe->icbTag.fileType = ICBTAG_FILE_TYPE_REGULAR; @@ -1828,7 +1824,6 @@ static int udf_update_inode(struct inode *inode, int do_sync) crclen)); fe->descTag.tagChecksum = udf_tag_checksum(&fe->descTag); -out: set_buffer_uptodate(bh); unlock_buffer(bh); -- cgit v0.10.2 From 322dfa6402ec08da1ca182ac0061a1cf2c5c3101 Mon Sep 17 00:00:00 2001 From: Yi Zhang Date: Thu, 9 Jul 2015 18:11:31 +0530 Subject: regulator: 88pm800: fix LDO vsel_mask value As per datasheet, Except LDO2, all other LDO's use bit [3:0] for VOUT select. Current code uses wrong mask value of 0x1f, So this patch fixes it to use 0xf. Signed-off-by: Yi Zhang [vaibhav.hiremath@linaro.org: Updated changelog with more detailed description] Signed-off-by: Vaibhav Hiremath Signed-off-by: Mark Brown diff --git a/drivers/regulator/88pm800.c b/drivers/regulator/88pm800.c index 832932b..7fd4f51 100644 --- a/drivers/regulator/88pm800.c +++ b/drivers/regulator/88pm800.c @@ -130,7 +130,7 @@ struct pm800_regulators { .owner = THIS_MODULE, \ .n_voltages = ARRAY_SIZE(ldo_volt_table), \ .vsel_reg = PM800_##vreg##_VOUT, \ - .vsel_mask = 0x1f, \ + .vsel_mask = 0xf, \ .enable_reg = PM800_##ereg, \ .enable_mask = 1 << (ebit), \ .volt_table = ldo_volt_table, \ -- cgit v0.10.2 From 95dd8653de658143770cb0e55a58d2aab97c79d2 Mon Sep 17 00:00:00 2001 From: Pablo Neira Ayuso Date: Thu, 9 Jul 2015 22:56:00 +0200 Subject: netfilter: ctnetlink: put back references to master ct and expect objects We have to put back the references to the master conntrack and the expectation that we just created, otherwise we'll leak them. Fixes: 0ef71ee1a5b9 ("netfilter: ctnetlink: refactor ctnetlink_create_expect") Reported-by: Tim Wiess Signed-off-by: Pablo Neira Ayuso diff --git a/net/netfilter/nf_conntrack_netlink.c b/net/netfilter/nf_conntrack_netlink.c index d1c2394..6b8b0ab 100644 --- a/net/netfilter/nf_conntrack_netlink.c +++ b/net/netfilter/nf_conntrack_netlink.c @@ -2995,11 +2995,6 @@ ctnetlink_create_expect(struct net *net, u16 zone, } err = nf_ct_expect_related_report(exp, portid, report); - if (err < 0) - goto err_exp; - - return 0; -err_exp: nf_ct_expect_put(exp); err_ct: nf_ct_put(ct); -- cgit v0.10.2 From c5d73716e95d510c0655272ce44f318932309cb1 Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Thu, 9 Jul 2015 17:58:39 +0200 Subject: pNFS: Layoutreturn must invalidate all existing layout segments. Signed-off-by: Trond Myklebust diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c index 9f24238..671498c 100644 --- a/fs/nfs/nfs4proc.c +++ b/fs/nfs/nfs4proc.c @@ -7964,16 +7964,19 @@ static void nfs4_layoutreturn_release(void *calldata) { struct nfs4_layoutreturn *lrp = calldata; struct pnfs_layout_hdr *lo = lrp->args.layout; + LIST_HEAD(freeme); dprintk("--> %s\n", __func__); spin_lock(&lo->plh_inode->i_lock); if (lrp->res.lrs_present) pnfs_set_layout_stateid(lo, &lrp->res.stateid, true); + pnfs_mark_matching_lsegs_invalid(lo, &freeme, &lrp->args.range); pnfs_clear_layoutreturn_waitbit(lo); clear_bit(NFS_LAYOUT_RETURN_BEFORE_CLOSE, &lo->plh_flags); rpc_wake_up(&NFS_SERVER(lo->plh_inode)->roc_rpcwaitq); lo->plh_block_lgets--; spin_unlock(&lo->plh_inode->i_lock); + pnfs_free_lseg_list(&freeme); pnfs_put_layout_hdr(lrp->args.layout); nfs_iput_and_deactive(lrp->inode); kfree(calldata); -- cgit v0.10.2 From df9cecc1a36d70ec09d34e83bad452064754fdc4 Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Thu, 9 Jul 2015 23:38:11 +0200 Subject: pNFS: pnfs_roc_drain should return 'true' when sleeping Also clean up the case where we don't find a return-on-close layout segment. Signed-off-by: Trond Myklebust diff --git a/fs/nfs/pnfs.c b/fs/nfs/pnfs.c index 0ba9a02..8e9f467 100644 --- a/fs/nfs/pnfs.c +++ b/fs/nfs/pnfs.c @@ -1146,14 +1146,14 @@ bool pnfs_roc_drain(struct inode *ino, u32 *barrier, struct rpc_task *task) struct pnfs_layout_segment *lseg; nfs4_stateid stateid; u32 current_seqid; - bool found = false, layoutreturn = false; + bool layoutreturn = false; spin_lock(&ino->i_lock); list_for_each_entry(lseg, &nfsi->layout->plh_segs, pls_list) if (test_bit(NFS_LSEG_ROC, &lseg->pls_flags)) { rpc_sleep_on(&NFS_SERVER(ino)->roc_rpcwaitq, task, NULL); - found = true; - goto out; + spin_unlock(&ino->i_lock); + return true; } lo = nfsi->layout; current_seqid = be32_to_cpu(lo->plh_stateid.seqid); @@ -1162,23 +1162,21 @@ bool pnfs_roc_drain(struct inode *ino, u32 *barrier, struct rpc_task *task) * a barrier, we choose the worst-case barrier. */ *barrier = current_seqid + atomic_read(&lo->plh_outstanding); -out: - if (!found) { - stateid = lo->plh_stateid; - layoutreturn = - test_and_clear_bit(NFS_LAYOUT_RETURN_BEFORE_CLOSE, + stateid = lo->plh_stateid; + layoutreturn = test_and_clear_bit(NFS_LAYOUT_RETURN_BEFORE_CLOSE, &lo->plh_flags); - if (layoutreturn) { - lo->plh_block_lgets++; - pnfs_get_layout_hdr(lo); - } + if (layoutreturn) { + lo->plh_block_lgets++; + pnfs_get_layout_hdr(lo); } + spin_unlock(&ino->i_lock); if (layoutreturn) { rpc_sleep_on(&NFS_SERVER(ino)->roc_rpcwaitq, task, NULL); pnfs_send_layoutreturn(lo, stateid, IOMODE_ANY, false); + return true; } - return found; + return false; } /* -- cgit v0.10.2 From 7f27392cd4cbcadeeaff9eedebcaec8fae9aec8e Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Thu, 9 Jul 2015 18:40:01 +0200 Subject: pNFS: Fix races between return-on-close and layoutreturn. If one or more of the layout segments reports an error during I/O, then we may have to send a layoutreturn to report the error back to the NFS metadata server. This patch ensures that the return-on-close code can detect the outstanding layoutreturn, and not preempt it. Signed-off-by: Trond Myklebust diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c index 671498c..c5c9e0d 100644 --- a/fs/nfs/nfs4proc.c +++ b/fs/nfs/nfs4proc.c @@ -7972,8 +7972,6 @@ static void nfs4_layoutreturn_release(void *calldata) pnfs_set_layout_stateid(lo, &lrp->res.stateid, true); pnfs_mark_matching_lsegs_invalid(lo, &freeme, &lrp->args.range); pnfs_clear_layoutreturn_waitbit(lo); - clear_bit(NFS_LAYOUT_RETURN_BEFORE_CLOSE, &lo->plh_flags); - rpc_wake_up(&NFS_SERVER(lo->plh_inode)->roc_rpcwaitq); lo->plh_block_lgets--; spin_unlock(&lo->plh_inode->i_lock); pnfs_free_lseg_list(&freeme); diff --git a/fs/nfs/pnfs.c b/fs/nfs/pnfs.c index 8e9f467..27e2bca 100644 --- a/fs/nfs/pnfs.c +++ b/fs/nfs/pnfs.c @@ -352,7 +352,7 @@ pnfs_layout_need_return(struct pnfs_layout_hdr *lo, { struct pnfs_layout_segment *s; - if (!test_bit(NFS_LSEG_LAYOUTRETURN, &lseg->pls_flags)) + if (!test_and_clear_bit(NFS_LSEG_LAYOUTRETURN, &lseg->pls_flags)) return false; list_for_each_entry(s, &lo->plh_segs, pls_list) @@ -362,6 +362,18 @@ pnfs_layout_need_return(struct pnfs_layout_hdr *lo, return true; } +static bool +pnfs_prepare_layoutreturn(struct pnfs_layout_hdr *lo) +{ + if (test_and_set_bit(NFS_LAYOUT_RETURN, &lo->plh_flags)) + return false; + lo->plh_return_iomode = 0; + lo->plh_block_lgets++; + pnfs_get_layout_hdr(lo); + clear_bit(NFS_LAYOUT_RETURN_BEFORE_CLOSE, &lo->plh_flags); + return true; +} + static void pnfs_layoutreturn_before_put_lseg(struct pnfs_layout_segment *lseg, struct pnfs_layout_hdr *lo, struct inode *inode) { @@ -372,17 +384,16 @@ static void pnfs_layoutreturn_before_put_lseg(struct pnfs_layout_segment *lseg, if (pnfs_layout_need_return(lo, lseg)) { nfs4_stateid stateid; enum pnfs_iomode iomode; + bool send; stateid = lo->plh_stateid; iomode = lo->plh_return_iomode; - /* decreased in pnfs_send_layoutreturn() */ - lo->plh_block_lgets++; - lo->plh_return_iomode = 0; + send = pnfs_prepare_layoutreturn(lo); spin_unlock(&inode->i_lock); - pnfs_get_layout_hdr(lo); - - /* Send an async layoutreturn so we dont deadlock */ - pnfs_send_layoutreturn(lo, stateid, iomode, false); + if (send) { + /* Send an async layoutreturn so we dont deadlock */ + pnfs_send_layoutreturn(lo, stateid, iomode, false); + } } else spin_unlock(&inode->i_lock); } @@ -924,6 +935,7 @@ void pnfs_clear_layoutreturn_waitbit(struct pnfs_layout_hdr *lo) clear_bit_unlock(NFS_LAYOUT_RETURN, &lo->plh_flags); smp_mb__after_atomic(); wake_up_bit(&lo->plh_flags, NFS_LAYOUT_RETURN); + rpc_wake_up(&NFS_SERVER(lo->plh_inode)->roc_rpcwaitq); } static int @@ -978,6 +990,7 @@ _pnfs_return_layout(struct inode *ino) LIST_HEAD(tmp_list); nfs4_stateid stateid; int status = 0, empty; + bool send; dprintk("NFS: %s for inode %lu\n", __func__, ino->i_ino); @@ -1007,17 +1020,18 @@ _pnfs_return_layout(struct inode *ino) /* Don't send a LAYOUTRETURN if list was initially empty */ if (empty) { spin_unlock(&ino->i_lock); - pnfs_put_layout_hdr(lo); dprintk("NFS: %s no layout segments to return\n", __func__); - goto out; + goto out_put_layout_hdr; } set_bit(NFS_LAYOUT_INVALID_STID, &lo->plh_flags); - lo->plh_block_lgets++; + send = pnfs_prepare_layoutreturn(lo); spin_unlock(&ino->i_lock); pnfs_free_lseg_list(&tmp_list); - - status = pnfs_send_layoutreturn(lo, stateid, IOMODE_ANY, true); + if (send) + status = pnfs_send_layoutreturn(lo, stateid, IOMODE_ANY, true); +out_put_layout_hdr: + pnfs_put_layout_hdr(lo); out: dprintk("<-- %s status: %d\n", __func__, status); return status; @@ -1097,13 +1111,9 @@ bool pnfs_roc(struct inode *ino) out_noroc: if (lo) { stateid = lo->plh_stateid; - layoutreturn = - test_and_clear_bit(NFS_LAYOUT_RETURN_BEFORE_CLOSE, - &lo->plh_flags); - if (layoutreturn) { - lo->plh_block_lgets++; - pnfs_get_layout_hdr(lo); - } + if (test_and_clear_bit(NFS_LAYOUT_RETURN_BEFORE_CLOSE, + &lo->plh_flags)) + layoutreturn = pnfs_prepare_layoutreturn(lo); } spin_unlock(&ino->i_lock); if (layoutreturn) { @@ -1163,16 +1173,14 @@ bool pnfs_roc_drain(struct inode *ino, u32 *barrier, struct rpc_task *task) */ *barrier = current_seqid + atomic_read(&lo->plh_outstanding); stateid = lo->plh_stateid; - layoutreturn = test_and_clear_bit(NFS_LAYOUT_RETURN_BEFORE_CLOSE, - &lo->plh_flags); - if (layoutreturn) { - lo->plh_block_lgets++; - pnfs_get_layout_hdr(lo); - } + if (test_and_clear_bit(NFS_LAYOUT_RETURN_BEFORE_CLOSE, + &lo->plh_flags)) + layoutreturn = pnfs_prepare_layoutreturn(lo); + if (test_bit(NFS_LAYOUT_RETURN, &lo->plh_flags)) + rpc_sleep_on(&NFS_SERVER(ino)->roc_rpcwaitq, task, NULL); spin_unlock(&ino->i_lock); if (layoutreturn) { - rpc_sleep_on(&NFS_SERVER(ino)->roc_rpcwaitq, task, NULL); pnfs_send_layoutreturn(lo, stateid, IOMODE_ANY, false); return true; } @@ -1693,7 +1701,6 @@ void pnfs_error_mark_layout_for_return(struct inode *inode, spin_lock(&inode->i_lock); /* set failure bit so that pnfs path will be retried later */ pnfs_layout_set_fail_bit(lo, iomode); - set_bit(NFS_LAYOUT_RETURN, &lo->plh_flags); if (lo->plh_return_iomode == 0) lo->plh_return_iomode = range.iomode; else if (lo->plh_return_iomode != range.iomode) -- cgit v0.10.2 From bdc59cf233433ddd2cd671db02bd6b52323ce63d Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Thu, 9 Jul 2015 23:54:09 +0200 Subject: pNFS: pnfs_roc_drain() fix a race with open If a process reopens the file before we can send off the CLOSE/DELEGRETURN, then pnfs_roc_drain() may end up waiting for a new set of layout segments that are marked as return-on-close, but haven't yet been returned. Fix this by only waiting for those layout segments that were invalidated in pnfs_roc(). Signed-off-by: Trond Myklebust diff --git a/fs/nfs/pnfs.c b/fs/nfs/pnfs.c index 27e2bca..b02e32e 100644 --- a/fs/nfs/pnfs.c +++ b/fs/nfs/pnfs.c @@ -1159,12 +1159,15 @@ bool pnfs_roc_drain(struct inode *ino, u32 *barrier, struct rpc_task *task) bool layoutreturn = false; spin_lock(&ino->i_lock); - list_for_each_entry(lseg, &nfsi->layout->plh_segs, pls_list) - if (test_bit(NFS_LSEG_ROC, &lseg->pls_flags)) { - rpc_sleep_on(&NFS_SERVER(ino)->roc_rpcwaitq, task, NULL); - spin_unlock(&ino->i_lock); - return true; - } + list_for_each_entry(lseg, &nfsi->layout->plh_segs, pls_list) { + if (!test_bit(NFS_LSEG_ROC, &lseg->pls_flags)) + continue; + if (test_bit(NFS_LSEG_VALID, &lseg->pls_flags)) + continue; + rpc_sleep_on(&NFS_SERVER(ino)->roc_rpcwaitq, task, NULL); + spin_unlock(&ino->i_lock); + return true; + } lo = nfsi->layout; current_seqid = be32_to_cpu(lo->plh_stateid.seqid); -- cgit v0.10.2 From faa4a54f0be15b5d81b574fb5a40db24345d1a6c Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Thu, 9 Jul 2015 18:24:07 +0200 Subject: pNFS: Don't throw out valid layout segments It is OK for layout segments to remain hashed even if no-one holds any references to them, provided that the segments are still valid. Signed-off-by: Trond Myklebust diff --git a/fs/nfs/pnfs.c b/fs/nfs/pnfs.c index b02e32e..18aa3b7 100644 --- a/fs/nfs/pnfs.c +++ b/fs/nfs/pnfs.c @@ -422,6 +422,10 @@ pnfs_put_lseg(struct pnfs_layout_segment *lseg) pnfs_layoutreturn_before_put_lseg(lseg, lo, inode); if (atomic_dec_and_lock(&lseg->pls_refcount, &inode->i_lock)) { + if (test_bit(NFS_LSEG_VALID, &lseg->pls_flags)) { + spin_unlock(&inode->i_lock); + return; + } pnfs_get_layout_hdr(lo); pnfs_layout_remove_lseg(lo, lseg); spin_unlock(&inode->i_lock); @@ -462,6 +466,8 @@ pnfs_put_lseg_locked(struct pnfs_layout_segment *lseg) test_bit(NFS_LSEG_VALID, &lseg->pls_flags)); if (atomic_dec_and_test(&lseg->pls_refcount)) { struct pnfs_layout_hdr *lo = lseg->pls_layout; + if (test_bit(NFS_LSEG_VALID, &lseg->pls_flags)) + return; pnfs_get_layout_hdr(lo); pnfs_layout_remove_lseg(lo, lseg); pnfs_free_lseg_async(lseg); -- cgit v0.10.2 From 7865598ec24ab4162d976d0bd75a9b2a2f58cd64 Mon Sep 17 00:00:00 2001 From: Felix Fietkau Date: Sun, 21 Jun 2015 19:47:46 +0200 Subject: ath9k_hw: fix device ID check for AR956x Because of the missing return, the macVersion value was being overwritten with an invalid register read Signed-off-by: Felix Fietkau Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/ath/ath9k/hw.c b/drivers/net/wireless/ath/ath9k/hw.c index 5e15e8e..a31a680 100644 --- a/drivers/net/wireless/ath/ath9k/hw.c +++ b/drivers/net/wireless/ath/ath9k/hw.c @@ -279,6 +279,7 @@ static void ath9k_hw_read_revisions(struct ath_hw *ah) return; case AR9300_DEVID_QCA956X: ah->hw_version.macVersion = AR_SREV_VERSION_9561; + return; } val = REG_READ(ah, AR_SREV) & AR_SREV_ID; -- cgit v0.10.2 From f9e5554cd8ca1d1212ec922755b397a20f737923 Mon Sep 17 00:00:00 2001 From: Emmanuel Grumbach Date: Thu, 4 Jun 2015 11:09:47 +0300 Subject: iwlwifi: pcie: prepare the device before accessing it For 8000 series, we need to access the device to know what firmware to load. Before we do so, we need to prepare the device otherwise we might not be able to access the hardware. Fixes: c278754a21e6 ("iwlwifi: mvm: support family 8000 B2/C steps") CC: [4.1] Signed-off-by: Emmanuel Grumbach diff --git a/drivers/net/wireless/iwlwifi/pcie/trans.c b/drivers/net/wireless/iwlwifi/pcie/trans.c index fe61d0a..6203c4a 100644 --- a/drivers/net/wireless/iwlwifi/pcie/trans.c +++ b/drivers/net/wireless/iwlwifi/pcie/trans.c @@ -2551,6 +2551,12 @@ struct iwl_trans *iwl_trans_pcie_alloc(struct pci_dev *pdev, trans->hw_rev = (trans->hw_rev & 0xfff0) | (CSR_HW_REV_STEP(trans->hw_rev << 2) << 2); + ret = iwl_pcie_prepare_card_hw(trans); + if (ret) { + IWL_WARN(trans, "Exit HW not ready\n"); + goto out_pci_disable_msi; + } + /* * in-order to recognize C step driver should read chip version * id located at the AUX bus MISC address space. -- cgit v0.10.2 From 00fd233ac491708d1f64b7176be73c84a18a54a2 Mon Sep 17 00:00:00 2001 From: Johannes Berg Date: Fri, 26 Jun 2015 16:04:01 +0200 Subject: iwlwifi: mvm: check time-event vif to avoid bad deletion The time event is initialized relatively late in interface (mvmvif) initialization, so it's possible to fail before that happens. As a consequence, the driver will crash if it ever tries to delete this time event in case initialization was unsuccessful. Avoid this by using the time event's vif pointer to indicate validity. The vif pointer is != NULL whenever the id is != TE_MAX, except for this special error case where the vif pointer will have the correct property (as the whole memory is cleared on allocation) whereas the id is 0, causing a crash in trying to delete the time event from the list. Signed-off-by: Johannes Berg Signed-off-by: Emmanuel Grumbach diff --git a/drivers/net/wireless/iwlwifi/mvm/time-event.c b/drivers/net/wireless/iwlwifi/mvm/time-event.c index d24b6a8..e472729 100644 --- a/drivers/net/wireless/iwlwifi/mvm/time-event.c +++ b/drivers/net/wireless/iwlwifi/mvm/time-event.c @@ -86,7 +86,7 @@ void iwl_mvm_te_clear_data(struct iwl_mvm *mvm, { lockdep_assert_held(&mvm->time_event_lock); - if (te_data->id == TE_MAX) + if (!te_data->vif) return; list_del(&te_data->list); -- cgit v0.10.2 From b8eee757077e2aced778cfad8c1d989e2deec584 Mon Sep 17 00:00:00 2001 From: Oren Givon Date: Mon, 22 Jun 2015 13:44:05 +0300 Subject: iwlwifi: edit the 3165 series and 8000 series PCI IDs Add new 3165 devices support. Add one new 8000 series device support. Remove support for 0x0000, 0xC030 and 0xD030 sub-system IDs in the 8000 series. Signed-off-by: Oren Givon Signed-off-by: Emmanuel Grumbach diff --git a/drivers/net/wireless/iwlwifi/pcie/drv.c b/drivers/net/wireless/iwlwifi/pcie/drv.c index 2ed1e4d..9f65c1c 100644 --- a/drivers/net/wireless/iwlwifi/pcie/drv.c +++ b/drivers/net/wireless/iwlwifi/pcie/drv.c @@ -368,12 +368,14 @@ static const struct pci_device_id iwl_hw_card_ids[] = { /* 3165 Series */ {IWL_PCI_DEVICE(0x3165, 0x4010, iwl3165_2ac_cfg)}, {IWL_PCI_DEVICE(0x3165, 0x4012, iwl3165_2ac_cfg)}, + {IWL_PCI_DEVICE(0x3166, 0x4212, iwl3165_2ac_cfg)}, {IWL_PCI_DEVICE(0x3165, 0x4410, iwl3165_2ac_cfg)}, {IWL_PCI_DEVICE(0x3165, 0x4510, iwl3165_2ac_cfg)}, {IWL_PCI_DEVICE(0x3165, 0x4110, iwl3165_2ac_cfg)}, {IWL_PCI_DEVICE(0x3166, 0x4310, iwl3165_2ac_cfg)}, {IWL_PCI_DEVICE(0x3166, 0x4210, iwl3165_2ac_cfg)}, {IWL_PCI_DEVICE(0x3165, 0x8010, iwl3165_2ac_cfg)}, + {IWL_PCI_DEVICE(0x3165, 0x8110, iwl3165_2ac_cfg)}, /* 7265 Series */ {IWL_PCI_DEVICE(0x095A, 0x5010, iwl7265_2ac_cfg)}, @@ -426,9 +428,8 @@ static const struct pci_device_id iwl_hw_card_ids[] = { {IWL_PCI_DEVICE(0x24F4, 0x1130, iwl8260_2ac_cfg)}, {IWL_PCI_DEVICE(0x24F4, 0x1030, iwl8260_2ac_cfg)}, {IWL_PCI_DEVICE(0x24F3, 0xC010, iwl8260_2ac_cfg)}, + {IWL_PCI_DEVICE(0x24F3, 0xC110, iwl8260_2ac_cfg)}, {IWL_PCI_DEVICE(0x24F3, 0xD010, iwl8260_2ac_cfg)}, - {IWL_PCI_DEVICE(0x24F4, 0xC030, iwl8260_2ac_cfg)}, - {IWL_PCI_DEVICE(0x24F4, 0xD030, iwl8260_2ac_cfg)}, {IWL_PCI_DEVICE(0x24F3, 0xC050, iwl8260_2ac_cfg)}, {IWL_PCI_DEVICE(0x24F3, 0xD050, iwl8260_2ac_cfg)}, {IWL_PCI_DEVICE(0x24F3, 0x8010, iwl8260_2ac_cfg)}, -- cgit v0.10.2 From 8465fe6ac5cc054a7969e78e56aea40cd7808540 Mon Sep 17 00:00:00 2001 From: Avraham Stern Date: Sun, 21 Jun 2015 12:06:13 +0300 Subject: iwlwifi: mvm: Add preemptive flag to scheulded scan Add preemptive flag to scheduled scan command flags. Without this flag, all scan requests after scheduled scan was started will be delayed until scheduled scan stops. As a result, P2P_FIND will be blocked while scheduled scan is active. This flag was omitted during refactoring. Signed-off-by: Avraham Stern Signed-off-by: Emmanuel Grumbach diff --git a/drivers/net/wireless/iwlwifi/mvm/fw-api-scan.h b/drivers/net/wireless/iwlwifi/mvm/fw-api-scan.h index 5e4cbdb..737774a 100644 --- a/drivers/net/wireless/iwlwifi/mvm/fw-api-scan.h +++ b/drivers/net/wireless/iwlwifi/mvm/fw-api-scan.h @@ -660,7 +660,8 @@ struct iwl_scan_config { * iwl_umac_scan_flags *@IWL_UMAC_SCAN_FLAG_PREEMPTIVE: scan process triggered by this scan request * can be preempted by other scan requests with higher priority. - * The low priority scan is aborted. + * The low priority scan will be resumed when the higher proirity scan is + * completed. *@IWL_UMAC_SCAN_FLAG_START_NOTIF: notification will be sent to the driver * when scan starts. */ diff --git a/drivers/net/wireless/iwlwifi/mvm/scan.c b/drivers/net/wireless/iwlwifi/mvm/scan.c index 5de1449..5000bfcd 100644 --- a/drivers/net/wireless/iwlwifi/mvm/scan.c +++ b/drivers/net/wireless/iwlwifi/mvm/scan.c @@ -1109,6 +1109,9 @@ static int iwl_mvm_scan_umac(struct iwl_mvm *mvm, struct ieee80211_vif *vif, cmd->uid = cpu_to_le32(uid); cmd->general_flags = cpu_to_le32(iwl_mvm_scan_umac_flags(mvm, params)); + if (type == IWL_MVM_SCAN_SCHED) + cmd->flags = cpu_to_le32(IWL_UMAC_SCAN_FLAG_PREEMPTIVE); + if (iwl_mvm_scan_use_ebs(mvm, vif, n_iterations)) cmd->channel_flags = IWL_SCAN_CHANNEL_FLAG_EBS | IWL_SCAN_CHANNEL_FLAG_EBS_ACCURATE | -- cgit v0.10.2 From 255ba06533f09875973d6b400f9c5b88065938df Mon Sep 17 00:00:00 2001 From: Emmanuel Grumbach Date: Sat, 11 Jul 2015 22:30:49 +0300 Subject: Revert "iwlwifi: pcie: New RBD allocation model" This reverts commit 5f17570354f91275b0a37a4da33d29a2ab57d32e. This patch introduced a high latency in buffer allocation under extreme load. This latency caused a firmwre crash. The same scenario works fine with this patch reverted. Signed-off-by: Emmanuel Grumbach diff --git a/drivers/net/wireless/iwlwifi/iwl-fh.h b/drivers/net/wireless/iwlwifi/iwl-fh.h index d560648..d45dc02 100644 --- a/drivers/net/wireless/iwlwifi/iwl-fh.h +++ b/drivers/net/wireless/iwlwifi/iwl-fh.h @@ -438,6 +438,12 @@ static inline unsigned int FH_MEM_CBBC_QUEUE(unsigned int chnl) #define RX_QUEUE_MASK 255 #define RX_QUEUE_SIZE_LOG 8 +/* + * RX related structures and functions + */ +#define RX_FREE_BUFFERS 64 +#define RX_LOW_WATERMARK 8 + /** * struct iwl_rb_status - reserve buffer status * host memory mapped FH registers diff --git a/drivers/net/wireless/iwlwifi/pcie/internal.h b/drivers/net/wireless/iwlwifi/pcie/internal.h index 31f72a6..376b84e 100644 --- a/drivers/net/wireless/iwlwifi/pcie/internal.h +++ b/drivers/net/wireless/iwlwifi/pcie/internal.h @@ -44,15 +44,6 @@ #include "iwl-io.h" #include "iwl-op-mode.h" -/* - * RX related structures and functions - */ -#define RX_NUM_QUEUES 1 -#define RX_POST_REQ_ALLOC 2 -#define RX_CLAIM_REQ_ALLOC 8 -#define RX_POOL_SIZE ((RX_CLAIM_REQ_ALLOC - RX_POST_REQ_ALLOC) * RX_NUM_QUEUES) -#define RX_LOW_WATERMARK 8 - struct iwl_host_cmd; /*This file includes the declaration that are internal to the @@ -86,29 +77,29 @@ struct isr_statistics { * struct iwl_rxq - Rx queue * @bd: driver's pointer to buffer of receive buffer descriptors (rbd) * @bd_dma: bus address of buffer of receive buffer descriptors (rbd) + * @pool: + * @queue: * @read: Shared index to newest available Rx buffer * @write: Shared index to oldest written Rx packet * @free_count: Number of pre-allocated buffers in rx_free - * @used_count: Number of RBDs handled to allocator to use for allocation * @write_actual: - * @rx_free: list of RBDs with allocated RB ready for use - * @rx_used: list of RBDs with no RB attached + * @rx_free: list of free SKBs for use + * @rx_used: List of Rx buffers with no SKB * @need_update: flag to indicate we need to update read/write index * @rb_stts: driver's pointer to receive buffer status * @rb_stts_dma: bus address of receive buffer status * @lock: - * @pool: initial pool of iwl_rx_mem_buffer for the queue - * @queue: actual rx queue * * NOTE: rx_free and rx_used are used as a FIFO for iwl_rx_mem_buffers */ struct iwl_rxq { __le32 *bd; dma_addr_t bd_dma; + struct iwl_rx_mem_buffer pool[RX_QUEUE_SIZE + RX_FREE_BUFFERS]; + struct iwl_rx_mem_buffer *queue[RX_QUEUE_SIZE]; u32 read; u32 write; u32 free_count; - u32 used_count; u32 write_actual; struct list_head rx_free; struct list_head rx_used; @@ -116,32 +107,6 @@ struct iwl_rxq { struct iwl_rb_status *rb_stts; dma_addr_t rb_stts_dma; spinlock_t lock; - struct iwl_rx_mem_buffer pool[RX_QUEUE_SIZE]; - struct iwl_rx_mem_buffer *queue[RX_QUEUE_SIZE]; -}; - -/** - * struct iwl_rb_allocator - Rx allocator - * @pool: initial pool of allocator - * @req_pending: number of requests the allcator had not processed yet - * @req_ready: number of requests honored and ready for claiming - * @rbd_allocated: RBDs with pages allocated and ready to be handled to - * the queue. This is a list of &struct iwl_rx_mem_buffer - * @rbd_empty: RBDs with no page attached for allocator use. This is a list - * of &struct iwl_rx_mem_buffer - * @lock: protects the rbd_allocated and rbd_empty lists - * @alloc_wq: work queue for background calls - * @rx_alloc: work struct for background calls - */ -struct iwl_rb_allocator { - struct iwl_rx_mem_buffer pool[RX_POOL_SIZE]; - atomic_t req_pending; - atomic_t req_ready; - struct list_head rbd_allocated; - struct list_head rbd_empty; - spinlock_t lock; - struct workqueue_struct *alloc_wq; - struct work_struct rx_alloc; }; struct iwl_dma_ptr { @@ -285,7 +250,7 @@ iwl_pcie_get_scratchbuf_dma(struct iwl_txq *txq, int idx) /** * struct iwl_trans_pcie - PCIe transport specific data * @rxq: all the RX queue data - * @rba: allocator for RX replenishing + * @rx_replenish: work that will be called when buffers need to be allocated * @drv - pointer to iwl_drv * @trans: pointer to the generic transport area * @scd_base_addr: scheduler sram base address in SRAM @@ -308,7 +273,7 @@ iwl_pcie_get_scratchbuf_dma(struct iwl_txq *txq, int idx) */ struct iwl_trans_pcie { struct iwl_rxq rxq; - struct iwl_rb_allocator rba; + struct work_struct rx_replenish; struct iwl_trans *trans; struct iwl_drv *drv; diff --git a/drivers/net/wireless/iwlwifi/pcie/rx.c b/drivers/net/wireless/iwlwifi/pcie/rx.c index a3fbaa0..adad8d0 100644 --- a/drivers/net/wireless/iwlwifi/pcie/rx.c +++ b/drivers/net/wireless/iwlwifi/pcie/rx.c @@ -1,7 +1,7 @@ /****************************************************************************** * * Copyright(c) 2003 - 2014 Intel Corporation. All rights reserved. - * Copyright(c) 2013 - 2015 Intel Mobile Communications GmbH + * Copyright(c) 2013 - 2014 Intel Mobile Communications GmbH * * Portions of this file are derived from the ipw3945 project, as well * as portions of the ieee80211 subsystem header files. @@ -74,29 +74,16 @@ * resets the Rx queue buffers with new memory. * * The management in the driver is as follows: - * + A list of pre-allocated RBDs is stored in iwl->rxq->rx_free. - * When the interrupt handler is called, the request is processed. - * The page is either stolen - transferred to the upper layer - * or reused - added immediately to the iwl->rxq->rx_free list. - * + When the page is stolen - the driver updates the matching queue's used - * count, detaches the RBD and transfers it to the queue used list. - * When there are two used RBDs - they are transferred to the allocator empty - * list. Work is then scheduled for the allocator to start allocating - * eight buffers. - * When there are another 6 used RBDs - they are transferred to the allocator - * empty list and the driver tries to claim the pre-allocated buffers and - * add them to iwl->rxq->rx_free. If it fails - it continues to claim them - * until ready. - * When there are 8+ buffers in the free list - either from allocation or from - * 8 reused unstolen pages - restock is called to update the FW and indexes. - * + In order to make sure the allocator always has RBDs to use for allocation - * the allocator has initial pool in the size of num_queues*(8-2) - the - * maximum missing RBDs per allocation request (request posted with 2 - * empty RBDs, there is no guarantee when the other 6 RBDs are supplied). - * The queues supplies the recycle of the rest of the RBDs. + * + A list of pre-allocated SKBs is stored in iwl->rxq->rx_free. When + * iwl->rxq->free_count drops to or below RX_LOW_WATERMARK, work is scheduled + * to replenish the iwl->rxq->rx_free. + * + In iwl_pcie_rx_replenish (scheduled) if 'processed' != 'read' then the + * iwl->rxq is replenished and the READ INDEX is updated (updating the + * 'processed' and 'read' driver indexes as well) * + A received packet is processed and handed to the kernel network stack, * detached from the iwl->rxq. The driver 'processed' index is updated. - * + If there are no allocated buffers in iwl->rxq->rx_free, + * + The Host/Firmware iwl->rxq is replenished at irq thread time from the + * rx_free list. If there are no allocated buffers in iwl->rxq->rx_free, * the READ INDEX is not incremented and iwl->status(RX_STALLED) is set. * If there were enough free buffers and RX_STALLED is set it is cleared. * @@ -105,32 +92,18 @@ * * iwl_rxq_alloc() Allocates rx_free * iwl_pcie_rx_replenish() Replenishes rx_free list from rx_used, and calls - * iwl_pcie_rxq_restock. - * Used only during initialization. + * iwl_pcie_rxq_restock * iwl_pcie_rxq_restock() Moves available buffers from rx_free into Rx * queue, updates firmware pointers, and updates - * the WRITE index. - * iwl_pcie_rx_allocator() Background work for allocating pages. + * the WRITE index. If insufficient rx_free buffers + * are available, schedules iwl_pcie_rx_replenish * * -- enable interrupts -- * ISR - iwl_rx() Detach iwl_rx_mem_buffers from pool up to the * READ INDEX, detaching the SKB from the pool. * Moves the packet buffer from queue to rx_used. - * Posts and claims requests to the allocator. * Calls iwl_pcie_rxq_restock to refill any empty * slots. - * - * RBD life-cycle: - * - * Init: - * rxq.pool -> rxq.rx_used -> rxq.rx_free -> rxq.queue - * - * Regular Receive interrupt: - * Page Stolen: - * rxq.queue -> rxq.rx_used -> allocator.rbd_empty -> - * allocator.rbd_allocated -> rxq.rx_free -> rxq.queue - * Page not Stolen: - * rxq.queue -> rxq.rx_free -> rxq.queue * ... * */ @@ -267,6 +240,10 @@ static void iwl_pcie_rxq_restock(struct iwl_trans *trans) rxq->free_count--; } spin_unlock(&rxq->lock); + /* If the pre-allocated buffer pool is dropping low, schedule to + * refill it */ + if (rxq->free_count <= RX_LOW_WATERMARK) + schedule_work(&trans_pcie->rx_replenish); /* If we've added more space for the firmware to place data, tell it. * Increment device's write pointer in multiples of 8. */ @@ -278,44 +255,6 @@ static void iwl_pcie_rxq_restock(struct iwl_trans *trans) } /* - * iwl_pcie_rx_alloc_page - allocates and returns a page. - * - */ -static struct page *iwl_pcie_rx_alloc_page(struct iwl_trans *trans) -{ - struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans); - struct iwl_rxq *rxq = &trans_pcie->rxq; - struct page *page; - gfp_t gfp_mask = GFP_KERNEL; - - if (rxq->free_count > RX_LOW_WATERMARK) - gfp_mask |= __GFP_NOWARN; - - if (trans_pcie->rx_page_order > 0) - gfp_mask |= __GFP_COMP; - - /* Alloc a new receive buffer */ - page = alloc_pages(gfp_mask, trans_pcie->rx_page_order); - if (!page) { - if (net_ratelimit()) - IWL_DEBUG_INFO(trans, "alloc_pages failed, order: %d\n", - trans_pcie->rx_page_order); - /* Issue an error if the hardware has consumed more than half - * of its free buffer list and we don't have enough - * pre-allocated buffers. -` */ - if (rxq->free_count <= RX_LOW_WATERMARK && - iwl_rxq_space(rxq) > (RX_QUEUE_SIZE / 2) && - net_ratelimit()) - IWL_CRIT(trans, - "Failed to alloc_pages with GFP_KERNEL. Only %u free buffers remaining.\n", - rxq->free_count); - return NULL; - } - return page; -} - -/* * iwl_pcie_rxq_alloc_rbs - allocate a page for each used RBD * * A used RBD is an Rx buffer that has been given to the stack. To use it again @@ -324,12 +263,13 @@ static struct page *iwl_pcie_rx_alloc_page(struct iwl_trans *trans) * iwl_pcie_rxq_restock. The latter function will update the HW to use the newly * allocated buffers. */ -static void iwl_pcie_rxq_alloc_rbs(struct iwl_trans *trans) +static void iwl_pcie_rxq_alloc_rbs(struct iwl_trans *trans, gfp_t priority) { struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans); struct iwl_rxq *rxq = &trans_pcie->rxq; struct iwl_rx_mem_buffer *rxb; struct page *page; + gfp_t gfp_mask = priority; while (1) { spin_lock(&rxq->lock); @@ -339,10 +279,32 @@ static void iwl_pcie_rxq_alloc_rbs(struct iwl_trans *trans) } spin_unlock(&rxq->lock); + if (rxq->free_count > RX_LOW_WATERMARK) + gfp_mask |= __GFP_NOWARN; + + if (trans_pcie->rx_page_order > 0) + gfp_mask |= __GFP_COMP; + /* Alloc a new receive buffer */ - page = iwl_pcie_rx_alloc_page(trans); - if (!page) + page = alloc_pages(gfp_mask, trans_pcie->rx_page_order); + if (!page) { + if (net_ratelimit()) + IWL_DEBUG_INFO(trans, "alloc_pages failed, " + "order: %d\n", + trans_pcie->rx_page_order); + + if ((rxq->free_count <= RX_LOW_WATERMARK) && + net_ratelimit()) + IWL_CRIT(trans, "Failed to alloc_pages with %s." + "Only %u free buffers remaining.\n", + priority == GFP_ATOMIC ? + "GFP_ATOMIC" : "GFP_KERNEL", + rxq->free_count); + /* We don't reschedule replenish work here -- we will + * call the restock method and if it still needs + * more buffers it will schedule replenish */ return; + } spin_lock(&rxq->lock); @@ -393,7 +355,7 @@ static void iwl_pcie_rxq_free_rbs(struct iwl_trans *trans) lockdep_assert_held(&rxq->lock); - for (i = 0; i < RX_QUEUE_SIZE; i++) { + for (i = 0; i < RX_FREE_BUFFERS + RX_QUEUE_SIZE; i++) { if (!rxq->pool[i].page) continue; dma_unmap_page(trans->dev, rxq->pool[i].page_dma, @@ -410,144 +372,32 @@ static void iwl_pcie_rxq_free_rbs(struct iwl_trans *trans) * When moving to rx_free an page is allocated for the slot. * * Also restock the Rx queue via iwl_pcie_rxq_restock. - * This is called only during initialization + * This is called as a scheduled work item (except for during initialization) */ -static void iwl_pcie_rx_replenish(struct iwl_trans *trans) +static void iwl_pcie_rx_replenish(struct iwl_trans *trans, gfp_t gfp) { - iwl_pcie_rxq_alloc_rbs(trans); + iwl_pcie_rxq_alloc_rbs(trans, gfp); iwl_pcie_rxq_restock(trans); } -/* - * iwl_pcie_rx_allocator - Allocates pages in the background for RX queues - * - * Allocates for each received request 8 pages - * Called as a scheduled work item. - */ -static void iwl_pcie_rx_allocator(struct iwl_trans *trans) -{ - struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans); - struct iwl_rb_allocator *rba = &trans_pcie->rba; - - while (atomic_read(&rba->req_pending)) { - int i; - struct list_head local_empty; - struct list_head local_allocated; - - INIT_LIST_HEAD(&local_allocated); - spin_lock(&rba->lock); - /* swap out the entire rba->rbd_empty to a local list */ - list_replace_init(&rba->rbd_empty, &local_empty); - spin_unlock(&rba->lock); - - for (i = 0; i < RX_CLAIM_REQ_ALLOC;) { - struct iwl_rx_mem_buffer *rxb; - struct page *page; - - /* List should never be empty - each reused RBD is - * returned to the list, and initial pool covers any - * possible gap between the time the page is allocated - * to the time the RBD is added. - */ - BUG_ON(list_empty(&local_empty)); - /* Get the first rxb from the rbd list */ - rxb = list_first_entry(&local_empty, - struct iwl_rx_mem_buffer, list); - BUG_ON(rxb->page); - - /* Alloc a new receive buffer */ - page = iwl_pcie_rx_alloc_page(trans); - if (!page) - continue; - rxb->page = page; - - /* Get physical address of the RB */ - rxb->page_dma = dma_map_page(trans->dev, page, 0, - PAGE_SIZE << trans_pcie->rx_page_order, - DMA_FROM_DEVICE); - if (dma_mapping_error(trans->dev, rxb->page_dma)) { - rxb->page = NULL; - __free_pages(page, trans_pcie->rx_page_order); - continue; - } - /* dma address must be no more than 36 bits */ - BUG_ON(rxb->page_dma & ~DMA_BIT_MASK(36)); - /* and also 256 byte aligned! */ - BUG_ON(rxb->page_dma & DMA_BIT_MASK(8)); - - /* move the allocated entry to the out list */ - list_move(&rxb->list, &local_allocated); - i++; - } - - spin_lock(&rba->lock); - /* add the allocated rbds to the allocator allocated list */ - list_splice_tail(&local_allocated, &rba->rbd_allocated); - /* add the unused rbds back to the allocator empty list */ - list_splice_tail(&local_empty, &rba->rbd_empty); - spin_unlock(&rba->lock); - - atomic_dec(&rba->req_pending); - atomic_inc(&rba->req_ready); - } -} - -/* - * iwl_pcie_rx_allocator_get - Returns the pre-allocated pages -.* -.* Called by queue when the queue posted allocation request and - * has freed 8 RBDs in order to restock itself. - */ -static int iwl_pcie_rx_allocator_get(struct iwl_trans *trans, - struct iwl_rx_mem_buffer - *out[RX_CLAIM_REQ_ALLOC]) -{ - struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans); - struct iwl_rb_allocator *rba = &trans_pcie->rba; - int i; - - if (atomic_dec_return(&rba->req_ready) < 0) { - atomic_inc(&rba->req_ready); - IWL_DEBUG_RX(trans, - "Allocation request not ready, pending requests = %d\n", - atomic_read(&rba->req_pending)); - return -ENOMEM; - } - - spin_lock(&rba->lock); - for (i = 0; i < RX_CLAIM_REQ_ALLOC; i++) { - /* Get next free Rx buffer, remove it from free list */ - out[i] = list_first_entry(&rba->rbd_allocated, - struct iwl_rx_mem_buffer, list); - list_del(&out[i]->list); - } - spin_unlock(&rba->lock); - - return 0; -} - -static void iwl_pcie_rx_allocator_work(struct work_struct *data) +static void iwl_pcie_rx_replenish_work(struct work_struct *data) { - struct iwl_rb_allocator *rba_p = - container_of(data, struct iwl_rb_allocator, rx_alloc); struct iwl_trans_pcie *trans_pcie = - container_of(rba_p, struct iwl_trans_pcie, rba); + container_of(data, struct iwl_trans_pcie, rx_replenish); - iwl_pcie_rx_allocator(trans_pcie->trans); + iwl_pcie_rx_replenish(trans_pcie->trans, GFP_KERNEL); } static int iwl_pcie_rx_alloc(struct iwl_trans *trans) { struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans); struct iwl_rxq *rxq = &trans_pcie->rxq; - struct iwl_rb_allocator *rba = &trans_pcie->rba; struct device *dev = trans->dev; memset(&trans_pcie->rxq, 0, sizeof(trans_pcie->rxq)); spin_lock_init(&rxq->lock); - spin_lock_init(&rba->lock); if (WARN_ON(rxq->bd || rxq->rb_stts)) return -EINVAL; @@ -637,49 +487,15 @@ static void iwl_pcie_rx_init_rxb_lists(struct iwl_rxq *rxq) INIT_LIST_HEAD(&rxq->rx_free); INIT_LIST_HEAD(&rxq->rx_used); rxq->free_count = 0; - rxq->used_count = 0; - for (i = 0; i < RX_QUEUE_SIZE; i++) + for (i = 0; i < RX_FREE_BUFFERS + RX_QUEUE_SIZE; i++) list_add(&rxq->pool[i].list, &rxq->rx_used); } -static void iwl_pcie_rx_init_rba(struct iwl_rb_allocator *rba) -{ - int i; - - lockdep_assert_held(&rba->lock); - - INIT_LIST_HEAD(&rba->rbd_allocated); - INIT_LIST_HEAD(&rba->rbd_empty); - - for (i = 0; i < RX_POOL_SIZE; i++) - list_add(&rba->pool[i].list, &rba->rbd_empty); -} - -static void iwl_pcie_rx_free_rba(struct iwl_trans *trans) -{ - struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans); - struct iwl_rb_allocator *rba = &trans_pcie->rba; - int i; - - lockdep_assert_held(&rba->lock); - - for (i = 0; i < RX_POOL_SIZE; i++) { - if (!rba->pool[i].page) - continue; - dma_unmap_page(trans->dev, rba->pool[i].page_dma, - PAGE_SIZE << trans_pcie->rx_page_order, - DMA_FROM_DEVICE); - __free_pages(rba->pool[i].page, trans_pcie->rx_page_order); - rba->pool[i].page = NULL; - } -} - int iwl_pcie_rx_init(struct iwl_trans *trans) { struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans); struct iwl_rxq *rxq = &trans_pcie->rxq; - struct iwl_rb_allocator *rba = &trans_pcie->rba; int i, err; if (!rxq->bd) { @@ -687,21 +503,11 @@ int iwl_pcie_rx_init(struct iwl_trans *trans) if (err) return err; } - if (!rba->alloc_wq) - rba->alloc_wq = alloc_workqueue("rb_allocator", - WQ_HIGHPRI | WQ_UNBOUND, 1); - INIT_WORK(&rba->rx_alloc, iwl_pcie_rx_allocator_work); - - spin_lock(&rba->lock); - atomic_set(&rba->req_pending, 0); - atomic_set(&rba->req_ready, 0); - /* free all first - we might be reconfigured for a different size */ - iwl_pcie_rx_free_rba(trans); - iwl_pcie_rx_init_rba(rba); - spin_unlock(&rba->lock); spin_lock(&rxq->lock); + INIT_WORK(&trans_pcie->rx_replenish, iwl_pcie_rx_replenish_work); + /* free all first - we might be reconfigured for a different size */ iwl_pcie_rxq_free_rbs(trans); iwl_pcie_rx_init_rxb_lists(rxq); @@ -716,7 +522,7 @@ int iwl_pcie_rx_init(struct iwl_trans *trans) memset(rxq->rb_stts, 0, sizeof(*rxq->rb_stts)); spin_unlock(&rxq->lock); - iwl_pcie_rx_replenish(trans); + iwl_pcie_rx_replenish(trans, GFP_KERNEL); iwl_pcie_rx_hw_init(trans, rxq); @@ -731,7 +537,6 @@ void iwl_pcie_rx_free(struct iwl_trans *trans) { struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans); struct iwl_rxq *rxq = &trans_pcie->rxq; - struct iwl_rb_allocator *rba = &trans_pcie->rba; /*if rxq->bd is NULL, it means that nothing has been allocated, * exit now */ @@ -740,15 +545,7 @@ void iwl_pcie_rx_free(struct iwl_trans *trans) return; } - cancel_work_sync(&rba->rx_alloc); - if (rba->alloc_wq) { - destroy_workqueue(rba->alloc_wq); - rba->alloc_wq = NULL; - } - - spin_lock(&rba->lock); - iwl_pcie_rx_free_rba(trans); - spin_unlock(&rba->lock); + cancel_work_sync(&trans_pcie->rx_replenish); spin_lock(&rxq->lock); iwl_pcie_rxq_free_rbs(trans); @@ -769,43 +566,6 @@ void iwl_pcie_rx_free(struct iwl_trans *trans) rxq->rb_stts = NULL; } -/* - * iwl_pcie_rx_reuse_rbd - Recycle used RBDs - * - * Called when a RBD can be reused. The RBD is transferred to the allocator. - * When there are 2 empty RBDs - a request for allocation is posted - */ -static void iwl_pcie_rx_reuse_rbd(struct iwl_trans *trans, - struct iwl_rx_mem_buffer *rxb, - struct iwl_rxq *rxq) -{ - struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans); - struct iwl_rb_allocator *rba = &trans_pcie->rba; - - /* Count the used RBDs */ - rxq->used_count++; - - /* Move the RBD to the used list, will be moved to allocator in batches - * before claiming or posting a request*/ - list_add_tail(&rxb->list, &rxq->rx_used); - - /* If we have RX_POST_REQ_ALLOC new released rx buffers - - * issue a request for allocator. Modulo RX_CLAIM_REQ_ALLOC is - * used for the case we failed to claim RX_CLAIM_REQ_ALLOC, - * after but we still need to post another request. - */ - if ((rxq->used_count % RX_CLAIM_REQ_ALLOC) == RX_POST_REQ_ALLOC) { - /* Move the 2 RBDs to the allocator ownership. - Allocator has another 6 from pool for the request completion*/ - spin_lock(&rba->lock); - list_splice_tail_init(&rxq->rx_used, &rba->rbd_empty); - spin_unlock(&rba->lock); - - atomic_inc(&rba->req_pending); - queue_work(rba->alloc_wq, &rba->rx_alloc); - } -} - static void iwl_pcie_rx_handle_rb(struct iwl_trans *trans, struct iwl_rx_mem_buffer *rxb) { @@ -928,13 +688,13 @@ static void iwl_pcie_rx_handle_rb(struct iwl_trans *trans, */ __free_pages(rxb->page, trans_pcie->rx_page_order); rxb->page = NULL; - iwl_pcie_rx_reuse_rbd(trans, rxb, rxq); + list_add_tail(&rxb->list, &rxq->rx_used); } else { list_add_tail(&rxb->list, &rxq->rx_free); rxq->free_count++; } } else - iwl_pcie_rx_reuse_rbd(trans, rxb, rxq); + list_add_tail(&rxb->list, &rxq->rx_used); } /* @@ -944,7 +704,10 @@ static void iwl_pcie_rx_handle(struct iwl_trans *trans) { struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans); struct iwl_rxq *rxq = &trans_pcie->rxq; - u32 r, i, j; + u32 r, i; + u8 fill_rx = 0; + u32 count = 8; + int total_empty; restart: spin_lock(&rxq->lock); @@ -957,6 +720,14 @@ restart: if (i == r) IWL_DEBUG_RX(trans, "HW = SW = %d\n", r); + /* calculate total frames need to be restock after handling RX */ + total_empty = r - rxq->write_actual; + if (total_empty < 0) + total_empty += RX_QUEUE_SIZE; + + if (total_empty > (RX_QUEUE_SIZE / 2)) + fill_rx = 1; + while (i != r) { struct iwl_rx_mem_buffer *rxb; @@ -968,48 +739,29 @@ restart: iwl_pcie_rx_handle_rb(trans, rxb); i = (i + 1) & RX_QUEUE_MASK; - - /* If we have RX_CLAIM_REQ_ALLOC released rx buffers - - * try to claim the pre-allocated buffers from the allocator */ - if (rxq->used_count >= RX_CLAIM_REQ_ALLOC) { - struct iwl_rb_allocator *rba = &trans_pcie->rba; - struct iwl_rx_mem_buffer *out[RX_CLAIM_REQ_ALLOC]; - - /* Add the remaining 6 empty RBDs for allocator use */ - spin_lock(&rba->lock); - list_splice_tail_init(&rxq->rx_used, &rba->rbd_empty); - spin_unlock(&rba->lock); - - /* If not ready - continue, will try to reclaim later. - * No need to reschedule work - allocator exits only on - * success */ - if (!iwl_pcie_rx_allocator_get(trans, out)) { - /* If success - then RX_CLAIM_REQ_ALLOC - * buffers were retrieved and should be added - * to free list */ - rxq->used_count -= RX_CLAIM_REQ_ALLOC; - for (j = 0; j < RX_CLAIM_REQ_ALLOC; j++) { - list_add_tail(&out[j]->list, - &rxq->rx_free); - rxq->free_count++; - } + /* If there are a lot of unused frames, + * restock the Rx queue so ucode wont assert. */ + if (fill_rx) { + count++; + if (count >= 8) { + rxq->read = i; + spin_unlock(&rxq->lock); + iwl_pcie_rx_replenish(trans, GFP_ATOMIC); + count = 0; + goto restart; } } - /* handle restock for two cases: - * - we just pulled buffers from the allocator - * - we have 8+ unstolen pages accumulated */ - if (rxq->free_count >= RX_CLAIM_REQ_ALLOC) { - rxq->read = i; - spin_unlock(&rxq->lock); - iwl_pcie_rxq_restock(trans); - goto restart; - } } /* Backtrack one entry */ rxq->read = i; spin_unlock(&rxq->lock); + if (fill_rx) + iwl_pcie_rx_replenish(trans, GFP_ATOMIC); + else + iwl_pcie_rxq_restock(trans); + if (trans_pcie->napi.poll) napi_gro_flush(&trans_pcie->napi, false); } -- cgit v0.10.2 From be88a1ada9b97bb016196b7f4a1fc2fe2f798529 Mon Sep 17 00:00:00 2001 From: Liad Kaufman Date: Wed, 1 Jul 2015 17:28:34 +0300 Subject: iwlwifi: nvm: remove mac address byte swapping in 8000 family This fixes the byte order copying in the MAO (Mac Override Section) section from the PNVM, as the byte swapping is not required anymore in the 8000 family. Due to the byte swapping, the driver was reporting an incorrect MAC adddress. CC: [4.1] Signed-off-by: Liad Kaufman Signed-off-by: Emmanuel Grumbach diff --git a/drivers/net/wireless/iwlwifi/iwl-nvm-parse.c b/drivers/net/wireless/iwlwifi/iwl-nvm-parse.c index 80fefe7..3b8e85e 100644 --- a/drivers/net/wireless/iwlwifi/iwl-nvm-parse.c +++ b/drivers/net/wireless/iwlwifi/iwl-nvm-parse.c @@ -540,13 +540,11 @@ static void iwl_set_hw_address_family_8000(struct device *dev, hw_addr = (const u8 *)(mac_override + MAC_ADDRESS_OVERRIDE_FAMILY_8000); - /* The byte order is little endian 16 bit, meaning 214365 */ - data->hw_addr[0] = hw_addr[1]; - data->hw_addr[1] = hw_addr[0]; - data->hw_addr[2] = hw_addr[3]; - data->hw_addr[3] = hw_addr[2]; - data->hw_addr[4] = hw_addr[5]; - data->hw_addr[5] = hw_addr[4]; + /* + * Store the MAC address from MAO section. + * No byte swapping is required in MAO section + */ + memcpy(data->hw_addr, hw_addr, ETH_ALEN); /* * Force the use of the OTP MAC address in case of reserved MAC -- cgit v0.10.2 From 484836ec2de24d9a7c6471f022b746d947698725 Mon Sep 17 00:00:00 2001 From: Dmitry Torokhov Date: Thu, 9 Jul 2015 17:15:01 -0700 Subject: netfilter: IDLETIMER: fix lockdep warning Dynamically allocated sysfs attributes should be initialized with sysfs_attr_init() otherwise lockdep will be angry with us: [ 45.468653] BUG: key ffffffc030fad4e0 not in .data! [ 45.468655] ------------[ cut here ]------------ [ 45.468666] WARNING: CPU: 0 PID: 1176 at /mnt/host/source/src/third_party/kernel/v3.18/kernel/locking/lockdep.c:2991 lockdep_init_map+0x12c/0x490() [ 45.468672] DEBUG_LOCKS_WARN_ON(1) [ 45.468672] CPU: 0 PID: 1176 Comm: iptables Tainted: G U W 3.18.0 #43 [ 45.468674] Hardware name: XXX [ 45.468675] Call trace: [ 45.468680] [] dump_backtrace+0x0/0x10c [ 45.468683] [] show_stack+0x10/0x1c [ 45.468688] [] dump_stack+0x74/0x94 [ 45.468692] [] warn_slowpath_common+0x84/0xb0 [ 45.468694] [] warn_slowpath_fmt+0x4c/0x58 [ 45.468697] [] lockdep_init_map+0x128/0x490 [ 45.468701] [] __kernfs_create_file+0x80/0xe4 [ 45.468704] [] sysfs_add_file_mode_ns+0x104/0x170 [ 45.468706] [] sysfs_create_file_ns+0x58/0x64 [ 45.468711] [] idletimer_tg_checkentry+0x14c/0x324 [ 45.468714] [] xt_check_target+0x170/0x198 [ 45.468717] [] check_target+0x58/0x6c [ 45.468720] [] translate_table+0x30c/0x424 [ 45.468723] [] do_ipt_set_ctl+0x144/0x1d0 [ 45.468728] [] nf_setsockopt+0x50/0x60 [ 45.468732] [] ip_setsockopt+0x8c/0xb4 [ 45.468735] [] raw_setsockopt+0x10/0x50 [ 45.468739] [] sock_common_setsockopt+0x14/0x20 [ 45.468742] [] SyS_setsockopt+0x88/0xb8 [ 45.468744] ---[ end trace 41d156354d18c039 ]--- Signed-off-by: Dmitry Torokhov Signed-off-by: Pablo Neira Ayuso diff --git a/net/netfilter/xt_IDLETIMER.c b/net/netfilter/xt_IDLETIMER.c index f407ebc1..29d2c31 100644 --- a/net/netfilter/xt_IDLETIMER.c +++ b/net/netfilter/xt_IDLETIMER.c @@ -126,6 +126,7 @@ static int idletimer_tg_create(struct idletimer_tg_info *info) goto out; } + sysfs_attr_init(&info->timer->attr.attr); info->timer->attr.attr.name = kstrdup(info->label, GFP_KERNEL); if (!info->timer->attr.attr.name) { ret = -ENOMEM; -- cgit v0.10.2 From ea52bf8eda9832ad30e9f059c5ead8d44f882a53 Mon Sep 17 00:00:00 2001 From: Pierre Morel Date: Thu, 9 Jul 2015 14:58:26 +0200 Subject: 9p/trans_virtio: reset virtio device on remove On device shutdown/removal, virtio drivers need to trigger a reset on the device; if this is neglected, the virtio core will complain about non-zero device status. This patch resets the status when the 9p virtio driver is removed from the system by calling vdev->config->reset on the virtio_device to send a reset to the host virtio device. Signed-off-by: Pierre Morel Reviewed-by: Cornelia Huck Reviewed-by: Michael S. Tsirkin Signed-off-by: Michael S. Tsirkin diff --git a/net/9p/trans_virtio.c b/net/9p/trans_virtio.c index 9dd49ca..6e70ddb 100644 --- a/net/9p/trans_virtio.c +++ b/net/9p/trans_virtio.c @@ -704,6 +704,7 @@ static void p9_virtio_remove(struct virtio_device *vdev) mutex_unlock(&virtio_9p_lock); + vdev->config->reset(vdev); vdev->config->del_vqs(vdev); sysfs_remove_file(&(vdev->dev.kobj), &dev_attr_mount_tag.attr); -- cgit v0.10.2 From 4de7255f7d2be5e51664c6ac6011ffd6e5463571 Mon Sep 17 00:00:00 2001 From: Igor Mammedov Date: Wed, 1 Jul 2015 11:07:09 +0200 Subject: vhost: extend memory regions allocation to vmalloc with large number of memory regions we could end up with high order allocations and kmalloc could fail if host is under memory pressure. Considering that memory regions array is used on hot path try harder to allocate using kmalloc and if it fails resort to vmalloc. It's still better than just failing vhost_set_memory() and causing guest crash due to it when a new memory hotplugged to guest. I'll still look at QEMU side solution to reduce amount of memory regions it feeds to vhost to make things even better, but it doesn't hurt for kernel to behave smarter and don't crash older QEMU's which could use large amount of memory regions. Signed-off-by: Igor Mammedov Signed-off-by: Michael S. Tsirkin diff --git a/drivers/vhost/vhost.c b/drivers/vhost/vhost.c index 71bb468..a4ac369 100644 --- a/drivers/vhost/vhost.c +++ b/drivers/vhost/vhost.c @@ -22,6 +22,7 @@ #include #include #include +#include #include #include #include @@ -544,7 +545,7 @@ void vhost_dev_cleanup(struct vhost_dev *dev, bool locked) fput(dev->log_file); dev->log_file = NULL; /* No one will access memory at this point */ - kfree(dev->memory); + kvfree(dev->memory); dev->memory = NULL; WARN_ON(!list_empty(&dev->work_list)); if (dev->worker) { @@ -674,6 +675,18 @@ static int vhost_memory_reg_sort_cmp(const void *p1, const void *p2) return 0; } +static void *vhost_kvzalloc(unsigned long size) +{ + void *n = kzalloc(size, GFP_KERNEL | __GFP_NOWARN | __GFP_REPEAT); + + if (!n) { + n = vzalloc(size); + if (!n) + return ERR_PTR(-ENOMEM); + } + return n; +} + static long vhost_set_memory(struct vhost_dev *d, struct vhost_memory __user *m) { struct vhost_memory mem, *newmem, *oldmem; @@ -686,7 +699,7 @@ static long vhost_set_memory(struct vhost_dev *d, struct vhost_memory __user *m) return -EOPNOTSUPP; if (mem.nregions > VHOST_MEMORY_MAX_NREGIONS) return -E2BIG; - newmem = kmalloc(size + mem.nregions * sizeof *m->regions, GFP_KERNEL); + newmem = vhost_kvzalloc(size + mem.nregions * sizeof(*m->regions)); if (!newmem) return -ENOMEM; @@ -700,7 +713,7 @@ static long vhost_set_memory(struct vhost_dev *d, struct vhost_memory __user *m) vhost_memory_reg_sort_cmp, NULL); if (!memory_access_ok(d, newmem, 0)) { - kfree(newmem); + kvfree(newmem); return -EFAULT; } oldmem = d->memory; @@ -712,7 +725,7 @@ static long vhost_set_memory(struct vhost_dev *d, struct vhost_memory __user *m) d->vqs[i]->memory = newmem; mutex_unlock(&d->vqs[i]->mutex); } - kfree(oldmem); + kvfree(oldmem); return 0; } -- cgit v0.10.2 From c9ce42f72fd0ba180fd35539829e4139dca31494 Mon Sep 17 00:00:00 2001 From: Igor Mammedov Date: Thu, 2 Jul 2015 15:08:11 +0200 Subject: vhost: add max_mem_regions module parameter it became possible to use a bigger amount of memory slots, which is used by memory hotplug for registering hotplugged memory. However QEMU crashes if it's used with more than ~60 pc-dimm devices and vhost-net enabled since host kernel in module vhost-net refuses to accept more than 64 memory regions. Allow to tweak limit via max_mem_regions module paramemter with default value set to 64 slots. Signed-off-by: Igor Mammedov Signed-off-by: Michael S. Tsirkin diff --git a/drivers/vhost/vhost.c b/drivers/vhost/vhost.c index a4ac369..a9fe859 100644 --- a/drivers/vhost/vhost.c +++ b/drivers/vhost/vhost.c @@ -30,8 +30,12 @@ #include "vhost.h" +static ushort max_mem_regions = 64; +module_param(max_mem_regions, ushort, 0444); +MODULE_PARM_DESC(max_mem_regions, + "Maximum number of memory regions in memory map. (default: 64)"); + enum { - VHOST_MEMORY_MAX_NREGIONS = 64, VHOST_MEMORY_F_LOG = 0x1, }; @@ -697,7 +701,7 @@ static long vhost_set_memory(struct vhost_dev *d, struct vhost_memory __user *m) return -EFAULT; if (mem.padding) return -EOPNOTSUPP; - if (mem.nregions > VHOST_MEMORY_MAX_NREGIONS) + if (mem.nregions > max_mem_regions) return -E2BIG; newmem = vhost_kvzalloc(size + mem.nregions * sizeof(*m->regions)); if (!newmem) -- cgit v0.10.2 From 326bf17ea5d4f8f17b54cbf167b8cb504c606ee9 Mon Sep 17 00:00:00 2001 From: Alex Gartrell Date: Fri, 26 Jun 2015 03:18:45 -0700 Subject: ipvs: fix ipv6 route unreach panic Previously there was a trivial panic unshare -n /bin/bash <dev and use that for the purpose of the invocation. Signed-off-by: Alex Gartrell Acked-by: Julian Anastasov Signed-off-by: Simon Horman diff --git a/net/netfilter/ipvs/ip_vs_xmit.c b/net/netfilter/ipvs/ip_vs_xmit.c index bf66a86..b99d806 100644 --- a/net/netfilter/ipvs/ip_vs_xmit.c +++ b/net/netfilter/ipvs/ip_vs_xmit.c @@ -505,6 +505,13 @@ err_put: return -1; err_unreach: + /* The ip6_link_failure function requires the dev field to be set + * in order to get the net (further for the sake of fwmark + * reflection). + */ + if (!skb->dev) + skb->dev = skb_dst(skb)->dev; + dst_link_failure(skb); return -1; } -- cgit v0.10.2 From 4754957f04f5f368792a0eb7dab0ae89fb93dcfd Mon Sep 17 00:00:00 2001 From: Julian Anastasov Date: Sat, 27 Jun 2015 14:39:30 +0300 Subject: ipvs: do not use random local source address for tunnels Michael Vallaly reports about wrong source address used in rare cases for tunneled traffic. Looks like __ip_vs_get_out_rt in 3.10+ is providing uninitialized dest_dst->dst_saddr.ip because ip_vs_dest_dst_alloc uses kmalloc. While we retry after seeing EINVAL from routing for data that does not look like valid local address, it still succeeded when this memory was previously used from other dests and with different local addresses. As result, we can use valid local address that is not suitable for our real server. Fix it by providing 0.0.0.0 every time our cache is refreshed. By this way we will get preferred source address from routing. Reported-by: Michael Vallaly Fixes: 026ace060dfe ("ipvs: optimize dst usage for real server") Signed-off-by: Julian Anastasov Signed-off-by: Simon Horman diff --git a/net/netfilter/ipvs/ip_vs_xmit.c b/net/netfilter/ipvs/ip_vs_xmit.c index b99d806..ec30d68 100644 --- a/net/netfilter/ipvs/ip_vs_xmit.c +++ b/net/netfilter/ipvs/ip_vs_xmit.c @@ -130,7 +130,6 @@ static struct rtable *do_output_route4(struct net *net, __be32 daddr, memset(&fl4, 0, sizeof(fl4)); fl4.daddr = daddr; - fl4.saddr = (rt_mode & IP_VS_RT_MODE_CONNECT) ? *saddr : 0; fl4.flowi4_flags = (rt_mode & IP_VS_RT_MODE_KNOWN_NH) ? FLOWI_FLAG_KNOWN_NH : 0; -- cgit v0.10.2 From 05f00505a89acd21f5d0d20f5797dfbc4cf85243 Mon Sep 17 00:00:00 2001 From: Julian Anastasov Date: Mon, 29 Jun 2015 21:51:40 +0300 Subject: ipvs: fix crash if scheduler is changed I overlooked the svc->sched_data usage from schedulers when the services were converted to RCU in 3.10. Now the rare ipvsadm -E command can change the scheduler but due to the reverse order of ip_vs_bind_scheduler and ip_vs_unbind_scheduler we provide new sched_data to the old scheduler resulting in a crash. To fix it without changing the scheduler methods we have to use synchronize_rcu() only for the editing case. It means all svc->scheduler readers should expect a NULL value. To avoid breakage for the service listing and ipvsadm -R we can use the "none" name to indicate that scheduler is not assigned, a state when we drop new connections. Reported-by: Alexander Vasiliev Fixes: ceec4c381681 ("ipvs: convert services to rcu") Signed-off-by: Julian Anastasov Signed-off-by: Simon Horman diff --git a/net/netfilter/ipvs/ip_vs_core.c b/net/netfilter/ipvs/ip_vs_core.c index 5d2b806..38fbc19 100644 --- a/net/netfilter/ipvs/ip_vs_core.c +++ b/net/netfilter/ipvs/ip_vs_core.c @@ -319,7 +319,13 @@ ip_vs_sched_persist(struct ip_vs_service *svc, * return *ignored=0 i.e. ICMP and NF_DROP */ sched = rcu_dereference(svc->scheduler); - dest = sched->schedule(svc, skb, iph); + if (sched) { + /* read svc->sched_data after svc->scheduler */ + smp_rmb(); + dest = sched->schedule(svc, skb, iph); + } else { + dest = NULL; + } if (!dest) { IP_VS_DBG(1, "p-schedule: no dest found.\n"); kfree(param.pe_data); @@ -467,7 +473,13 @@ ip_vs_schedule(struct ip_vs_service *svc, struct sk_buff *skb, } sched = rcu_dereference(svc->scheduler); - dest = sched->schedule(svc, skb, iph); + if (sched) { + /* read svc->sched_data after svc->scheduler */ + smp_rmb(); + dest = sched->schedule(svc, skb, iph); + } else { + dest = NULL; + } if (dest == NULL) { IP_VS_DBG(1, "Schedule: no dest found.\n"); return NULL; diff --git a/net/netfilter/ipvs/ip_vs_ctl.c b/net/netfilter/ipvs/ip_vs_ctl.c index 285eae3..24c5542 100644 --- a/net/netfilter/ipvs/ip_vs_ctl.c +++ b/net/netfilter/ipvs/ip_vs_ctl.c @@ -842,15 +842,16 @@ __ip_vs_update_dest(struct ip_vs_service *svc, struct ip_vs_dest *dest, __ip_vs_dst_cache_reset(dest); spin_unlock_bh(&dest->dst_lock); - sched = rcu_dereference_protected(svc->scheduler, 1); if (add) { ip_vs_start_estimator(svc->net, &dest->stats); list_add_rcu(&dest->n_list, &svc->destinations); svc->num_dests++; - if (sched->add_dest) + sched = rcu_dereference_protected(svc->scheduler, 1); + if (sched && sched->add_dest) sched->add_dest(svc, dest); } else { - if (sched->upd_dest) + sched = rcu_dereference_protected(svc->scheduler, 1); + if (sched && sched->upd_dest) sched->upd_dest(svc, dest); } } @@ -1084,7 +1085,7 @@ static void __ip_vs_unlink_dest(struct ip_vs_service *svc, struct ip_vs_scheduler *sched; sched = rcu_dereference_protected(svc->scheduler, 1); - if (sched->del_dest) + if (sched && sched->del_dest) sched->del_dest(svc, dest); } } @@ -1175,11 +1176,14 @@ ip_vs_add_service(struct net *net, struct ip_vs_service_user_kern *u, ip_vs_use_count_inc(); /* Lookup the scheduler by 'u->sched_name' */ - sched = ip_vs_scheduler_get(u->sched_name); - if (sched == NULL) { - pr_info("Scheduler module ip_vs_%s not found\n", u->sched_name); - ret = -ENOENT; - goto out_err; + if (strcmp(u->sched_name, "none")) { + sched = ip_vs_scheduler_get(u->sched_name); + if (!sched) { + pr_info("Scheduler module ip_vs_%s not found\n", + u->sched_name); + ret = -ENOENT; + goto out_err; + } } if (u->pe_name && *u->pe_name) { @@ -1240,10 +1244,12 @@ ip_vs_add_service(struct net *net, struct ip_vs_service_user_kern *u, spin_lock_init(&svc->stats.lock); /* Bind the scheduler */ - ret = ip_vs_bind_scheduler(svc, sched); - if (ret) - goto out_err; - sched = NULL; + if (sched) { + ret = ip_vs_bind_scheduler(svc, sched); + if (ret) + goto out_err; + sched = NULL; + } /* Bind the ct retriever */ RCU_INIT_POINTER(svc->pe, pe); @@ -1291,17 +1297,20 @@ ip_vs_add_service(struct net *net, struct ip_vs_service_user_kern *u, static int ip_vs_edit_service(struct ip_vs_service *svc, struct ip_vs_service_user_kern *u) { - struct ip_vs_scheduler *sched, *old_sched; + struct ip_vs_scheduler *sched = NULL, *old_sched; struct ip_vs_pe *pe = NULL, *old_pe = NULL; int ret = 0; /* * Lookup the scheduler, by 'u->sched_name' */ - sched = ip_vs_scheduler_get(u->sched_name); - if (sched == NULL) { - pr_info("Scheduler module ip_vs_%s not found\n", u->sched_name); - return -ENOENT; + if (strcmp(u->sched_name, "none")) { + sched = ip_vs_scheduler_get(u->sched_name); + if (!sched) { + pr_info("Scheduler module ip_vs_%s not found\n", + u->sched_name); + return -ENOENT; + } } old_sched = sched; @@ -1329,14 +1338,20 @@ ip_vs_edit_service(struct ip_vs_service *svc, struct ip_vs_service_user_kern *u) old_sched = rcu_dereference_protected(svc->scheduler, 1); if (sched != old_sched) { + if (old_sched) { + ip_vs_unbind_scheduler(svc, old_sched); + RCU_INIT_POINTER(svc->scheduler, NULL); + /* Wait all svc->sched_data users */ + synchronize_rcu(); + } /* Bind the new scheduler */ - ret = ip_vs_bind_scheduler(svc, sched); - if (ret) { - old_sched = sched; - goto out; + if (sched) { + ret = ip_vs_bind_scheduler(svc, sched); + if (ret) { + ip_vs_scheduler_put(sched); + goto out; + } } - /* Unbind the old scheduler on success */ - ip_vs_unbind_scheduler(svc, old_sched); } /* @@ -1982,6 +1997,7 @@ static int ip_vs_info_seq_show(struct seq_file *seq, void *v) const struct ip_vs_iter *iter = seq->private; const struct ip_vs_dest *dest; struct ip_vs_scheduler *sched = rcu_dereference(svc->scheduler); + char *sched_name = sched ? sched->name : "none"; if (iter->table == ip_vs_svc_table) { #ifdef CONFIG_IP_VS_IPV6 @@ -1990,18 +2006,18 @@ static int ip_vs_info_seq_show(struct seq_file *seq, void *v) ip_vs_proto_name(svc->protocol), &svc->addr.in6, ntohs(svc->port), - sched->name); + sched_name); else #endif seq_printf(seq, "%s %08X:%04X %s %s ", ip_vs_proto_name(svc->protocol), ntohl(svc->addr.ip), ntohs(svc->port), - sched->name, + sched_name, (svc->flags & IP_VS_SVC_F_ONEPACKET)?"ops ":""); } else { seq_printf(seq, "FWM %08X %s %s", - svc->fwmark, sched->name, + svc->fwmark, sched_name, (svc->flags & IP_VS_SVC_F_ONEPACKET)?"ops ":""); } @@ -2427,13 +2443,15 @@ ip_vs_copy_service(struct ip_vs_service_entry *dst, struct ip_vs_service *src) { struct ip_vs_scheduler *sched; struct ip_vs_kstats kstats; + char *sched_name; sched = rcu_dereference_protected(src->scheduler, 1); + sched_name = sched ? sched->name : "none"; dst->protocol = src->protocol; dst->addr = src->addr.ip; dst->port = src->port; dst->fwmark = src->fwmark; - strlcpy(dst->sched_name, sched->name, sizeof(dst->sched_name)); + strlcpy(dst->sched_name, sched_name, sizeof(dst->sched_name)); dst->flags = src->flags; dst->timeout = src->timeout / HZ; dst->netmask = src->netmask; @@ -2892,6 +2910,7 @@ static int ip_vs_genl_fill_service(struct sk_buff *skb, struct ip_vs_flags flags = { .flags = svc->flags, .mask = ~0 }; struct ip_vs_kstats kstats; + char *sched_name; nl_service = nla_nest_start(skb, IPVS_CMD_ATTR_SERVICE); if (!nl_service) @@ -2910,8 +2929,9 @@ static int ip_vs_genl_fill_service(struct sk_buff *skb, } sched = rcu_dereference_protected(svc->scheduler, 1); + sched_name = sched ? sched->name : "none"; pe = rcu_dereference_protected(svc->pe, 1); - if (nla_put_string(skb, IPVS_SVC_ATTR_SCHED_NAME, sched->name) || + if (nla_put_string(skb, IPVS_SVC_ATTR_SCHED_NAME, sched_name) || (pe && nla_put_string(skb, IPVS_SVC_ATTR_PE_NAME, pe->name)) || nla_put(skb, IPVS_SVC_ATTR_FLAGS, sizeof(flags), &flags) || nla_put_u32(skb, IPVS_SVC_ATTR_TIMEOUT, svc->timeout / HZ) || diff --git a/net/netfilter/ipvs/ip_vs_sched.c b/net/netfilter/ipvs/ip_vs_sched.c index 199760c..7e81416 100644 --- a/net/netfilter/ipvs/ip_vs_sched.c +++ b/net/netfilter/ipvs/ip_vs_sched.c @@ -74,7 +74,7 @@ void ip_vs_unbind_scheduler(struct ip_vs_service *svc, if (sched->done_service) sched->done_service(svc); - /* svc->scheduler can not be set to NULL */ + /* svc->scheduler can be set to NULL only by caller */ } @@ -147,21 +147,21 @@ void ip_vs_scheduler_put(struct ip_vs_scheduler *scheduler) void ip_vs_scheduler_err(struct ip_vs_service *svc, const char *msg) { - struct ip_vs_scheduler *sched; + struct ip_vs_scheduler *sched = rcu_dereference(svc->scheduler); + char *sched_name = sched ? sched->name : "none"; - sched = rcu_dereference(svc->scheduler); if (svc->fwmark) { IP_VS_ERR_RL("%s: FWM %u 0x%08X - %s\n", - sched->name, svc->fwmark, svc->fwmark, msg); + sched_name, svc->fwmark, svc->fwmark, msg); #ifdef CONFIG_IP_VS_IPV6 } else if (svc->af == AF_INET6) { IP_VS_ERR_RL("%s: %s [%pI6c]:%d - %s\n", - sched->name, ip_vs_proto_name(svc->protocol), + sched_name, ip_vs_proto_name(svc->protocol), &svc->addr.in6, ntohs(svc->port), msg); #endif } else { IP_VS_ERR_RL("%s: %s %pI4:%d - %s\n", - sched->name, ip_vs_proto_name(svc->protocol), + sched_name, ip_vs_proto_name(svc->protocol), &svc->addr.ip, ntohs(svc->port), msg); } } -- cgit v0.10.2 From 71563f3414e917c62acd8e0fb0edf8ed6af63e4b Mon Sep 17 00:00:00 2001 From: Alex Gartrell Date: Sun, 5 Jul 2015 14:28:26 -0700 Subject: ipvs: skb_orphan in case of forwarding It is possible that we bind against a local socket in early_demux when we are actually going to want to forward it. In this case, the socket serves no purpose and only serves to confuse things (particularly functions which implicitly expect sk_fullsock to be true, like ip_local_out). Additionally, skb_set_owner_w is totally broken for non full-socks. Signed-off-by: Alex Gartrell Fixes: 41063e9dd119 ("ipv4: Early TCP socket demux.") Acked-by: Julian Anastasov Signed-off-by: Simon Horman diff --git a/net/netfilter/ipvs/ip_vs_xmit.c b/net/netfilter/ipvs/ip_vs_xmit.c index ec30d68..34dc142 100644 --- a/net/netfilter/ipvs/ip_vs_xmit.c +++ b/net/netfilter/ipvs/ip_vs_xmit.c @@ -533,6 +533,21 @@ static inline int ip_vs_tunnel_xmit_prepare(struct sk_buff *skb, return ret; } +/* In the event of a remote destination, it's possible that we would have + * matches against an old socket (particularly a TIME-WAIT socket). This + * causes havoc down the line (ip_local_out et. al. expect regular sockets + * and invalid memory accesses will happen) so simply drop the association + * in this case. +*/ +static inline void ip_vs_drop_early_demux_sk(struct sk_buff *skb) +{ + /* If dev is set, the packet came from the LOCAL_IN callback and + * not from a local TCP socket. + */ + if (skb->dev) + skb_orphan(skb); +} + /* return NF_STOLEN (sent) or NF_ACCEPT if local=1 (not sent) */ static inline int ip_vs_nat_send_or_cont(int pf, struct sk_buff *skb, struct ip_vs_conn *cp, int local) @@ -544,12 +559,21 @@ static inline int ip_vs_nat_send_or_cont(int pf, struct sk_buff *skb, ip_vs_notrack(skb); else ip_vs_update_conntrack(skb, cp, 1); + + /* Remove the early_demux association unless it's bound for the + * exact same port and address on this host after translation. + */ + if (!local || cp->vport != cp->dport || + !ip_vs_addr_equal(cp->af, &cp->vaddr, &cp->daddr)) + ip_vs_drop_early_demux_sk(skb); + if (!local) { skb_forward_csum(skb); NF_HOOK(pf, NF_INET_LOCAL_OUT, NULL, skb, NULL, skb_dst(skb)->dev, dst_output_sk); } else ret = NF_ACCEPT; + return ret; } @@ -563,6 +587,7 @@ static inline int ip_vs_send_or_cont(int pf, struct sk_buff *skb, if (likely(!(cp->flags & IP_VS_CONN_F_NFCT))) ip_vs_notrack(skb); if (!local) { + ip_vs_drop_early_demux_sk(skb); skb_forward_csum(skb); NF_HOOK(pf, NF_INET_LOCAL_OUT, NULL, skb, NULL, skb_dst(skb)->dev, dst_output_sk); @@ -851,6 +876,8 @@ ip_vs_prepare_tunneled_skb(struct sk_buff *skb, int skb_af, struct ipv6hdr *old_ipv6h = NULL; #endif + ip_vs_drop_early_demux_sk(skb); + if (skb_headroom(skb) < max_headroom || skb_cloned(skb)) { new_skb = skb_realloc_headroom(skb, max_headroom); if (!new_skb) -- cgit v0.10.2 From 56184858d1fc95c46723436b455cb7261cd8be6f Mon Sep 17 00:00:00 2001 From: Julian Anastasov Date: Wed, 8 Jul 2015 08:31:33 +0300 Subject: ipvs: fix crash with sync protocol v0 and FTP Fix crash in 3.5+ if FTP is used after switching sync_version to 0. Fixes: 749c42b620a9 ("ipvs: reduce sync rate with time thresholds") Signed-off-by: Julian Anastasov Signed-off-by: Simon Horman diff --git a/net/netfilter/ipvs/ip_vs_sync.c b/net/netfilter/ipvs/ip_vs_sync.c index b08ba95..d99ad93 100644 --- a/net/netfilter/ipvs/ip_vs_sync.c +++ b/net/netfilter/ipvs/ip_vs_sync.c @@ -612,7 +612,7 @@ static void ip_vs_sync_conn_v0(struct net *net, struct ip_vs_conn *cp, pkts = atomic_add_return(1, &cp->in_pkts); else pkts = sysctl_sync_threshold(ipvs); - ip_vs_sync_conn(net, cp->control, pkts); + ip_vs_sync_conn(net, cp, pkts); } } -- cgit v0.10.2 From e3895c0334d0ef46e80f22eaf2a52401ff6d5a67 Mon Sep 17 00:00:00 2001 From: Julian Anastasov Date: Thu, 9 Jul 2015 11:15:27 +0300 Subject: ipvs: call skb_sender_cpu_clear Reset XPS's sender_cpu on forwarding. Signed-off-by: Julian Anastasov Fixes: 2bd82484bb4c ("xps: fix xps for stacked devices") Signed-off-by: Simon Horman diff --git a/net/netfilter/ipvs/ip_vs_xmit.c b/net/netfilter/ipvs/ip_vs_xmit.c index 34dc142..258a0b0 100644 --- a/net/netfilter/ipvs/ip_vs_xmit.c +++ b/net/netfilter/ipvs/ip_vs_xmit.c @@ -529,6 +529,8 @@ static inline int ip_vs_tunnel_xmit_prepare(struct sk_buff *skb, if (ret == NF_ACCEPT) { nf_reset(skb); skb_forward_csum(skb); + if (!skb->sk) + skb_sender_cpu_clear(skb); } return ret; } @@ -569,6 +571,8 @@ static inline int ip_vs_nat_send_or_cont(int pf, struct sk_buff *skb, if (!local) { skb_forward_csum(skb); + if (!skb->sk) + skb_sender_cpu_clear(skb); NF_HOOK(pf, NF_INET_LOCAL_OUT, NULL, skb, NULL, skb_dst(skb)->dev, dst_output_sk); } else @@ -589,6 +593,8 @@ static inline int ip_vs_send_or_cont(int pf, struct sk_buff *skb, if (!local) { ip_vs_drop_early_demux_sk(skb); skb_forward_csum(skb); + if (!skb->sk) + skb_sender_cpu_clear(skb); NF_HOOK(pf, NF_INET_LOCAL_OUT, NULL, skb, NULL, skb_dst(skb)->dev, dst_output_sk); } else -- cgit v0.10.2 From 7bee8b08c428b63aa4a3765bb907602e36355378 Mon Sep 17 00:00:00 2001 From: Chris Mason Date: Tue, 14 Jul 2015 16:25:30 -0400 Subject: Bluetooth: btbcm: allow btbcm_read_verbose_config to fail on Apple Commit 1c8ba6d013 moved around the setup code for broadcomm chips, and also added btbcm_read_verbose_config() to read extra information about the hardware. It's returning errors on some macbooks: Bluetooth: hci0: BCM: Read verbose config info failed (-16) Which makes us error out of the setup function. Since this probe isn't critical to operate the chip, this patch just changes things to carry on when it fails. Signed-off-by: Chris Mason Signed-off-by: Marcel Holtmann Cc: stable@vger.kernel.org # v4.1 diff --git a/drivers/bluetooth/btbcm.c b/drivers/bluetooth/btbcm.c index 1e1a432..9ceb8ac 100644 --- a/drivers/bluetooth/btbcm.c +++ b/drivers/bluetooth/btbcm.c @@ -472,12 +472,11 @@ int btbcm_setup_apple(struct hci_dev *hdev) /* Read Verbose Config Version Info */ skb = btbcm_read_verbose_config(hdev); - if (IS_ERR(skb)) - return PTR_ERR(skb); - - BT_INFO("%s: BCM: chip id %u build %4.4u", hdev->name, skb->data[1], - get_unaligned_le16(skb->data + 5)); - kfree_skb(skb); + if (!IS_ERR(skb)) { + BT_INFO("%s: BCM: chip id %u build %4.4u", hdev->name, skb->data[1], + get_unaligned_le16(skb->data + 5)); + kfree_skb(skb); + } set_bit(HCI_QUIRK_STRICT_DUPLICATE_FILTER, &hdev->quirks); -- cgit v0.10.2 From 945b47441d83d2392ac9f984e0267ad521f24268 Mon Sep 17 00:00:00 2001 From: Lior Amsalem Date: Tue, 30 Jun 2015 16:09:49 +0200 Subject: ata: pmp: add quirk for Marvell 4140 SATA PMP This commit adds the necessary quirk to make the Marvell 4140 SATA PMP work properly. This PMP doesn't like SRST on port number 4 (the host port) so this commit marks this port as not supporting SRST. Signed-off-by: Lior Amsalem Reviewed-by: Nadav Haklai Signed-off-by: Thomas Petazzoni Signed-off-by: Tejun Heo Cc: stable@vger.kernel.org diff --git a/drivers/ata/libata-pmp.c b/drivers/ata/libata-pmp.c index 7ccc084..85aa761 100644 --- a/drivers/ata/libata-pmp.c +++ b/drivers/ata/libata-pmp.c @@ -460,6 +460,13 @@ static void sata_pmp_quirks(struct ata_port *ap) ATA_LFLAG_NO_SRST | ATA_LFLAG_ASSUME_ATA; } + } else if (vendor == 0x11ab && devid == 0x4140) { + /* Marvell 4140 quirks */ + ata_for_each_link(link, ap, EDGE) { + /* port 4 is for SEMB device and it doesn't like SRST */ + if (link->pmp == 4) + link->flags |= ATA_LFLAG_DISABLED; + } } } -- cgit v0.10.2 From 08c85d2a599d967ede38a847f5594447b6100642 Mon Sep 17 00:00:00 2001 From: Aleksei Mamlin Date: Wed, 1 Jul 2015 13:48:30 +0300 Subject: libata: add ATA_HORKAGE_BROKEN_FPDMA_AA quirk for HP 250GB SATA disk VB0250EAVER Enabling AA on HP 250GB SATA disk VB0250EAVER causes errors: [ 3.788362] ata3.00: failed to enable AA (error_mask=0x1) [ 3.789243] ata3.00: failed to enable AA (error_mask=0x1) Add the ATA_HORKAGE_BROKEN_FPDMA_AA for this specific harddisk. tj: Collected FPDMA_AA entries and updated comment. Signed-off-by: Aleksei Mamlin Signed-off-by: Tejun Heo Cc: stable@vger.kernel.org diff --git a/drivers/ata/libata-core.c b/drivers/ata/libata-core.c index e83fc3d..58a6db8f 100644 --- a/drivers/ata/libata-core.c +++ b/drivers/ata/libata-core.c @@ -4174,9 +4174,10 @@ static const struct ata_blacklist_entry ata_device_blacklist [] = { { "ST3320[68]13AS", "SD1[5-9]", ATA_HORKAGE_NONCQ | ATA_HORKAGE_FIRMWARE_WARN }, - /* Seagate Momentus SpinPoint M8 seem to have FPMDA_AA issues */ + /* drives which fail FPDMA_AA activation (some may freeze afterwards) */ { "ST1000LM024 HN-M101MBB", "2AR10001", ATA_HORKAGE_BROKEN_FPDMA_AA }, { "ST1000LM024 HN-M101MBB", "2BA30001", ATA_HORKAGE_BROKEN_FPDMA_AA }, + { "VB0250EAVER", "HPG7", ATA_HORKAGE_BROKEN_FPDMA_AA }, /* Blacklist entries taken from Silicon Image 3124/3132 Windows driver .inf file - also several Linux problem reports */ -- cgit v0.10.2 From 50c2e4dd6749725338621fff456b26d3a592259f Mon Sep 17 00:00:00 2001 From: Dan Carpenter Date: Sun, 12 Jul 2015 01:20:55 +0300 Subject: net/xen-netback: off by one in BUG_ON() condition The > should be >=. I also added spaces around the '-' operations so the code is a little more consistent and matches the condition better. Fixes: f53c3fe8dad7 ('xen-netback: Introduce TX grant mapping') Signed-off-by: Dan Carpenter Signed-off-by: David S. Miller diff --git a/drivers/net/xen-netback/netback.c b/drivers/net/xen-netback/netback.c index 880d0d6..7d50711 100644 --- a/drivers/net/xen-netback/netback.c +++ b/drivers/net/xen-netback/netback.c @@ -1566,13 +1566,13 @@ static inline void xenvif_tx_dealloc_action(struct xenvif_queue *queue) smp_rmb(); while (dc != dp) { - BUG_ON(gop - queue->tx_unmap_ops > MAX_PENDING_REQS); + BUG_ON(gop - queue->tx_unmap_ops >= MAX_PENDING_REQS); pending_idx = queue->dealloc_ring[pending_index(dc++)]; - pending_idx_release[gop-queue->tx_unmap_ops] = + pending_idx_release[gop - queue->tx_unmap_ops] = pending_idx; - queue->pages_to_unmap[gop-queue->tx_unmap_ops] = + queue->pages_to_unmap[gop - queue->tx_unmap_ops] = queue->mmap_pages[pending_idx]; gnttab_set_unmap_op(gop, idx_to_kaddr(queue, pending_idx), -- cgit v0.10.2 From 6ae3673debfbf2cfa1972b6d00b974d0e10ad3c6 Mon Sep 17 00:00:00 2001 From: Marc Kleine-Budde Date: Sat, 11 Jul 2015 21:03:07 +0200 Subject: can: at91_can: don't touch skb after netif_receive_skb()/netif_rx() There is no guarantee that the skb is in the same state after calling net_receive_skb() or netif_rx(). It might be freed or reused. Not really harmful as its a read access, except you turn on the proper debugging options which catch a use after free. Signed-off-by: Marc Kleine-Budde diff --git a/drivers/net/can/at91_can.c b/drivers/net/can/at91_can.c index f4e40aa..945c095 100644 --- a/drivers/net/can/at91_can.c +++ b/drivers/net/can/at91_can.c @@ -577,10 +577,10 @@ static void at91_rx_overflow_err(struct net_device *dev) cf->can_id |= CAN_ERR_CRTL; cf->data[1] = CAN_ERR_CRTL_RX_OVERFLOW; - netif_receive_skb(skb); stats->rx_packets++; stats->rx_bytes += cf->can_dlc; + netif_receive_skb(skb); } /** @@ -642,10 +642,10 @@ static void at91_read_msg(struct net_device *dev, unsigned int mb) } at91_read_mb(dev, mb, cf); - netif_receive_skb(skb); stats->rx_packets++; stats->rx_bytes += cf->can_dlc; + netif_receive_skb(skb); can_led_event(dev, CAN_LED_EVENT_RX); } @@ -802,10 +802,10 @@ static int at91_poll_err(struct net_device *dev, int quota, u32 reg_sr) return 0; at91_poll_err_frame(dev, cf, reg_sr); - netif_receive_skb(skb); dev->stats.rx_packets++; dev->stats.rx_bytes += cf->can_dlc; + netif_receive_skb(skb); return 1; } @@ -1067,10 +1067,10 @@ static void at91_irq_err(struct net_device *dev) return; at91_irq_err_state(dev, cf, new_state); - netif_rx(skb); dev->stats.rx_packets++; dev->stats.rx_bytes += cf->can_dlc; + netif_rx(skb); priv->can.state = new_state; } -- cgit v0.10.2 From a18ec1b6c72eb87fe00e8e09c3a3f3cfbec3435b Mon Sep 17 00:00:00 2001 From: Marc Kleine-Budde Date: Fri, 8 May 2015 11:30:29 +0200 Subject: can: flexcan: don't touch skb after netif_receive_skb() There is no guarantee that the skb is in the same state after calling net_receive_skb() or netif_rx(). It might be freed or reused. Not really harmful as its a read access, except you turn on the proper debugging options which catch a use after free. Signed-off-by: Marc Kleine-Budde diff --git a/drivers/net/can/flexcan.c b/drivers/net/can/flexcan.c index 6201c5a..b1e8d72 100644 --- a/drivers/net/can/flexcan.c +++ b/drivers/net/can/flexcan.c @@ -577,10 +577,10 @@ static int flexcan_poll_bus_err(struct net_device *dev, u32 reg_esr) return 0; do_bus_err(dev, cf, reg_esr); - netif_receive_skb(skb); dev->stats.rx_packets++; dev->stats.rx_bytes += cf->can_dlc; + netif_receive_skb(skb); return 1; } @@ -622,10 +622,9 @@ static int flexcan_poll_state(struct net_device *dev, u32 reg_esr) if (unlikely(new_state == CAN_STATE_BUS_OFF)) can_bus_off(dev); - netif_receive_skb(skb); - dev->stats.rx_packets++; dev->stats.rx_bytes += cf->can_dlc; + netif_receive_skb(skb); return 1; } @@ -670,10 +669,10 @@ static int flexcan_read_frame(struct net_device *dev) } flexcan_read_fifo(dev, cf); - netif_receive_skb(skb); stats->rx_packets++; stats->rx_bytes += cf->can_dlc; + netif_receive_skb(skb); can_led_event(dev, CAN_LED_EVENT_RX); -- cgit v0.10.2 From 20926d79244de4e9c698ae429fdc1e7e5832b081 Mon Sep 17 00:00:00 2001 From: Marc Kleine-Budde Date: Sat, 11 Jul 2015 21:16:08 +0200 Subject: can: bfin_can: don't touch skb after netif_rx() There is no guarantee that the skb is in the same state after calling net_receive_skb() or netif_rx(). It might be freed or reused. Not really harmful as its a read access, except you turn on the proper debugging options which catch a use after free. Cc: Aaron Wu Signed-off-by: Marc Kleine-Budde diff --git a/drivers/net/can/bfin_can.c b/drivers/net/can/bfin_can.c index 27ad312..57dadd5 100644 --- a/drivers/net/can/bfin_can.c +++ b/drivers/net/can/bfin_can.c @@ -424,10 +424,9 @@ static void bfin_can_rx(struct net_device *dev, u16 isrc) cf->data[6 - i] = (6 - i) < cf->can_dlc ? (val >> 8) : 0; } - netif_rx(skb); - stats->rx_packets++; stats->rx_bytes += cf->can_dlc; + netif_rx(skb); } static int bfin_can_err(struct net_device *dev, u16 isrc, u16 status) @@ -508,10 +507,9 @@ static int bfin_can_err(struct net_device *dev, u16 isrc, u16 status) priv->can.state = state; - netif_rx(skb); - stats->rx_packets++; stats->rx_bytes += cf->can_dlc; + netif_rx(skb); return 0; } -- cgit v0.10.2 From 83537b6fd6c3752bcb578d2bd46312d1e2c4a73a Mon Sep 17 00:00:00 2001 From: Marc Kleine-Budde Date: Sat, 11 Jul 2015 21:16:08 +0200 Subject: can: grcan: don't touch skb after netif_rx() There is no guarantee that the skb is in the same state after calling net_receive_skb() or netif_rx(). It might be freed or reused. Not really harmful as its a read access, except you turn on the proper debugging options which catch a use after free. Cc: Andreas Larsson Signed-off-by: Marc Kleine-Budde diff --git a/drivers/net/can/grcan.c b/drivers/net/can/grcan.c index e3d7e22..db9538d 100644 --- a/drivers/net/can/grcan.c +++ b/drivers/net/can/grcan.c @@ -1216,11 +1216,12 @@ static int grcan_receive(struct net_device *dev, int budget) cf->data[i] = (u8)(slot[j] >> shift); } } - netif_receive_skb(skb); /* Update statistics and read pointer */ stats->rx_packets++; stats->rx_bytes += cf->can_dlc; + netif_receive_skb(skb); + rd = grcan_ring_add(rd, GRCAN_MSG_SIZE, dma->rx.size); } -- cgit v0.10.2 From a2e78cf7a3f562edca4230b188c8832b6214eccd Mon Sep 17 00:00:00 2001 From: Marc Kleine-Budde Date: Sat, 11 Jul 2015 21:16:08 +0200 Subject: can: slcan: don't touch skb after netif_rx_ni() There is no guarantee that the skb is in the same state after calling net_receive_skb() or netif_rx(). It might be freed or reused. Not really harmful as its a read access, except you turn on the proper debugging options which catch a use after free. Cc: Oliver Hartkopp Signed-off-by: Marc Kleine-Budde diff --git a/drivers/net/can/slcan.c b/drivers/net/can/slcan.c index a23a7af..9a3f15c 100644 --- a/drivers/net/can/slcan.c +++ b/drivers/net/can/slcan.c @@ -218,10 +218,10 @@ static void slc_bump(struct slcan *sl) memcpy(skb_put(skb, sizeof(struct can_frame)), &cf, sizeof(struct can_frame)); - netif_rx_ni(skb); sl->dev->stats.rx_packets++; sl->dev->stats.rx_bytes += cf.can_dlc; + netif_rx_ni(skb); } /* parse tty input stream */ -- cgit v0.10.2 From 05c4456538e9551b8ac47762b21127bf9cf6cc8e Mon Sep 17 00:00:00 2001 From: Marc Kleine-Budde Date: Sat, 11 Jul 2015 21:16:08 +0200 Subject: can: ti_heccn: don't touch skb after netif_rx() There is no guarantee that the skb is in the same state after calling net_receive_skb() or netif_rx(). It might be freed or reused. Not really harmful as its a read access, except you turn on the proper debugging options which catch a use after free. Cc: Anant Gole Signed-off-by: Marc Kleine-Budde diff --git a/drivers/net/can/ti_hecc.c b/drivers/net/can/ti_hecc.c index e95a9e1..cf345cb 100644 --- a/drivers/net/can/ti_hecc.c +++ b/drivers/net/can/ti_hecc.c @@ -747,9 +747,9 @@ static int ti_hecc_error(struct net_device *ndev, int int_status, } } - netif_rx(skb); stats->rx_packets++; stats->rx_bytes += cf->can_dlc; + netif_rx(skb); return 0; } -- cgit v0.10.2 From ef934e89f5e68535c447789b74de42cf389e55de Mon Sep 17 00:00:00 2001 From: Marc Kleine-Budde Date: Sat, 11 Jul 2015 21:16:08 +0200 Subject: can: cc770: don't touch skb after netif_rx() There is no guarantee that the skb is in the same state after calling net_receive_skb() or netif_rx(). It might be freed or reused. Not really harmful as its a read access, except you turn on the proper debugging options which catch a use after free. Cc: Wolfgang Grandegger Signed-off-by: Marc Kleine-Budde diff --git a/drivers/net/can/cc770/cc770.c b/drivers/net/can/cc770/cc770.c index c11d449..70a8cbb 100644 --- a/drivers/net/can/cc770/cc770.c +++ b/drivers/net/can/cc770/cc770.c @@ -504,10 +504,10 @@ static void cc770_rx(struct net_device *dev, unsigned int mo, u8 ctrl1) for (i = 0; i < cf->can_dlc; i++) cf->data[i] = cc770_read_reg(priv, msgobj[mo].data[i]); } - netif_rx(skb); stats->rx_packets++; stats->rx_bytes += cf->can_dlc; + netif_rx(skb); } static int cc770_err(struct net_device *dev, u8 status) @@ -584,10 +584,10 @@ static int cc770_err(struct net_device *dev, u8 status) } } - netif_rx(skb); stats->rx_packets++; stats->rx_bytes += cf->can_dlc; + netif_rx(skb); return 0; } -- cgit v0.10.2 From 889dd06e107d816fe63ff65e5dd3466949878439 Mon Sep 17 00:00:00 2001 From: Marc Kleine-Budde Date: Sat, 11 Jul 2015 21:16:08 +0200 Subject: can: sja1000: don't touch skb after netif_rx() There is no guarantee that the skb is in the same state after calling net_receive_skb() or netif_rx(). It might be freed or reused. Not really harmful as its a read access, except you turn on the proper debugging options which catch a use after free. Cc: Wolfgang Grandegger Cc: Oliver Hartkopp Signed-off-by: Marc Kleine-Budde diff --git a/drivers/net/can/sja1000/sja1000.c b/drivers/net/can/sja1000/sja1000.c index 32bd7f4..7b92e91 100644 --- a/drivers/net/can/sja1000/sja1000.c +++ b/drivers/net/can/sja1000/sja1000.c @@ -377,10 +377,9 @@ static void sja1000_rx(struct net_device *dev) /* release receive buffer */ sja1000_write_cmdreg(priv, CMD_RRB); - netif_rx(skb); - stats->rx_packets++; stats->rx_bytes += cf->can_dlc; + netif_rx(skb); can_led_event(dev, CAN_LED_EVENT_RX); } @@ -484,10 +483,9 @@ static int sja1000_err(struct net_device *dev, uint8_t isrc, uint8_t status) can_bus_off(dev); } - netif_rx(skb); - stats->rx_packets++; stats->rx_bytes += cf->can_dlc; + netif_rx(skb); return 0; } -- cgit v0.10.2 From 296decb6931fca0bbc0bfb26865837d52e287599 Mon Sep 17 00:00:00 2001 From: Marc Kleine-Budde Date: Sat, 11 Jul 2015 21:16:08 +0200 Subject: can: esd_usb2: don't touch skb after netif_rx() MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit There is no guarantee that the skb is in the same state after calling net_receive_skb() or netif_rx(). It might be freed or reused. Not really harmful as its a read access, except you turn on the proper debugging options which catch a use after free. Cc: Thomas Körper Signed-off-by: Marc Kleine-Budde diff --git a/drivers/net/can/usb/esd_usb2.c b/drivers/net/can/usb/esd_usb2.c index 411c1af..0e5a449 100644 --- a/drivers/net/can/usb/esd_usb2.c +++ b/drivers/net/can/usb/esd_usb2.c @@ -301,13 +301,12 @@ static void esd_usb2_rx_event(struct esd_usb2_net_priv *priv, cf->data[7] = rxerr; } - netif_rx(skb); - priv->bec.txerr = txerr; priv->bec.rxerr = rxerr; stats->rx_packets++; stats->rx_bytes += cf->can_dlc; + netif_rx(skb); } } @@ -347,10 +346,9 @@ static void esd_usb2_rx_can_msg(struct esd_usb2_net_priv *priv, cf->data[i] = msg->msg.rx.data[i]; } - netif_rx(skb); - stats->rx_packets++; stats->rx_bytes += cf->can_dlc; + netif_rx(skb); } return; -- cgit v0.10.2 From 43c021e8d6844c4b0f5cc9eda10d232e7b87a456 Mon Sep 17 00:00:00 2001 From: Marc Kleine-Budde Date: Sat, 11 Jul 2015 21:16:08 +0200 Subject: can: ems_usb: don't touch skb after netif_rx() There is no guarantee that the skb is in the same state after calling net_receive_skb() or netif_rx(). It might be freed or reused. Not really harmful as its a read access, except you turn on the proper debugging options which catch a use after free. Cc: Gerhard Uttenthaler Signed-off-by: Marc Kleine-Budde diff --git a/drivers/net/can/usb/ems_usb.c b/drivers/net/can/usb/ems_usb.c index 866bac0..2d39038 100644 --- a/drivers/net/can/usb/ems_usb.c +++ b/drivers/net/can/usb/ems_usb.c @@ -324,10 +324,9 @@ static void ems_usb_rx_can_msg(struct ems_usb *dev, struct ems_cpc_msg *msg) cf->data[i] = msg->msg.can_msg.msg[i]; } - netif_rx(skb); - stats->rx_packets++; stats->rx_bytes += cf->can_dlc; + netif_rx(skb); } static void ems_usb_rx_err(struct ems_usb *dev, struct ems_cpc_msg *msg) @@ -400,10 +399,9 @@ static void ems_usb_rx_err(struct ems_usb *dev, struct ems_cpc_msg *msg) stats->rx_errors++; } - netif_rx(skb); - stats->rx_packets++; stats->rx_bytes += cf->can_dlc; + netif_rx(skb); } /* -- cgit v0.10.2 From 9b721a4cefcbdedadbe72b5ec405046c139cf8ad Mon Sep 17 00:00:00 2001 From: Marc Kleine-Budde Date: Sat, 11 Jul 2015 21:16:08 +0200 Subject: can: usb_8dev: don't touch skb after netif_rx() There is no guarantee that the skb is in the same state after calling net_receive_skb() or netif_rx(). It might be freed or reused. Not really harmful as its a read access, except you turn on the proper debugging options which catch a use after free. Cc: Bernd Krumboeck Signed-off-by: Marc Kleine-Budde diff --git a/drivers/net/can/usb/usb_8dev.c b/drivers/net/can/usb/usb_8dev.c index dd52c7a..de95b1c 100644 --- a/drivers/net/can/usb/usb_8dev.c +++ b/drivers/net/can/usb/usb_8dev.c @@ -461,10 +461,9 @@ static void usb_8dev_rx_err_msg(struct usb_8dev_priv *priv, priv->bec.txerr = txerr; priv->bec.rxerr = rxerr; - netif_rx(skb); - stats->rx_packets++; stats->rx_bytes += cf->can_dlc; + netif_rx(skb); } /* Read data and status frames */ @@ -494,10 +493,9 @@ static void usb_8dev_rx_can_msg(struct usb_8dev_priv *priv, else memcpy(cf->data, msg->data, cf->can_dlc); - netif_rx(skb); - stats->rx_packets++; stats->rx_bytes += cf->can_dlc; + netif_rx(skb); can_led_event(priv->netdev, CAN_LED_EVENT_RX); } else { -- cgit v0.10.2 From 1c0ee046957648106b415df79038e4e62b144c19 Mon Sep 17 00:00:00 2001 From: Marc Kleine-Budde Date: Sat, 11 Jul 2015 21:16:08 +0200 Subject: can: pcan_usb: don't touch skb after netif_rx() There is no guarantee that the skb is in the same state after calling net_receive_skb() or netif_rx(). It might be freed or reused. Not really harmful as its a read access, except you turn on the proper debugging options which catch a use after free. Cc: Stephane Grosjean Signed-off-by: Marc Kleine-Budde diff --git a/drivers/net/can/usb/peak_usb/pcan_usb.c b/drivers/net/can/usb/peak_usb/pcan_usb.c index 72427f2..6b94007 100644 --- a/drivers/net/can/usb/peak_usb/pcan_usb.c +++ b/drivers/net/can/usb/peak_usb/pcan_usb.c @@ -526,9 +526,9 @@ static int pcan_usb_decode_error(struct pcan_usb_msg_context *mc, u8 n, hwts->hwtstamp = timeval_to_ktime(tv); } - netif_rx(skb); mc->netdev->stats.rx_packets++; mc->netdev->stats.rx_bytes += cf->can_dlc; + netif_rx(skb); return 0; } @@ -659,12 +659,11 @@ static int pcan_usb_decode_data(struct pcan_usb_msg_context *mc, u8 status_len) hwts = skb_hwtstamps(skb); hwts->hwtstamp = timeval_to_ktime(tv); - /* push the skb */ - netif_rx(skb); - /* update statistics */ mc->netdev->stats.rx_packets++; mc->netdev->stats.rx_bytes += cf->can_dlc; + /* push the skb */ + netif_rx(skb); return 0; diff --git a/drivers/net/can/usb/peak_usb/pcan_usb_pro.c b/drivers/net/can/usb/peak_usb/pcan_usb_pro.c index dec5171..7d61b32 100644 --- a/drivers/net/can/usb/peak_usb/pcan_usb_pro.c +++ b/drivers/net/can/usb/peak_usb/pcan_usb_pro.c @@ -553,9 +553,9 @@ static int pcan_usb_pro_handle_canmsg(struct pcan_usb_pro_interface *usb_if, hwts = skb_hwtstamps(skb); hwts->hwtstamp = timeval_to_ktime(tv); - netif_rx(skb); netdev->stats.rx_packets++; netdev->stats.rx_bytes += can_frame->can_dlc; + netif_rx(skb); return 0; } @@ -670,9 +670,9 @@ static int pcan_usb_pro_handle_error(struct pcan_usb_pro_interface *usb_if, peak_usb_get_ts_tv(&usb_if->time_ref, le32_to_cpu(er->ts32), &tv); hwts = skb_hwtstamps(skb); hwts->hwtstamp = timeval_to_ktime(tv); - netif_rx(skb); netdev->stats.rx_packets++; netdev->stats.rx_bytes += can_frame->can_dlc; + netif_rx(skb); return 0; } -- cgit v0.10.2 From 9f7e25edb1575a6d2363dc003f9cc09d840657e2 Mon Sep 17 00:00:00 2001 From: Mark Brown Date: Tue, 14 Jul 2015 11:17:26 +0100 Subject: regulator: core: Handle full constraints systems when resolving supplies When resolving device supplies if we fail to look up the regulator we substitute in the dummy supply instead if the system has fully specified constraints. When resolving supplies for regulators we do not have the equivalent code and instead just directly use the regulator_dev_lookup() result causing spurious failures. This does not affect DT systems since we are able to detect missing mappings directly as part of regulator_dev_lookup() and so have appropriate handling in the DT specific code. Reported-by: Christian Hartmann Signed-off-by: Mark Brown diff --git a/drivers/regulator/core.c b/drivers/regulator/core.c index c9f7201..bb8528d 100644 --- a/drivers/regulator/core.c +++ b/drivers/regulator/core.c @@ -1381,9 +1381,13 @@ static int regulator_resolve_supply(struct regulator_dev *rdev) } if (!r) { - dev_err(dev, "Failed to resolve %s-supply for %s\n", - rdev->supply_name, rdev->desc->name); - return -EPROBE_DEFER; + if (have_full_constraints()) { + r = dummy_regulator_rdev; + } else { + dev_err(dev, "Failed to resolve %s-supply for %s\n", + rdev->supply_name, rdev->desc->name); + return -EPROBE_DEFER; + } } /* Recursively resolve the supply of the supply */ -- cgit v0.10.2 From 4c62360d7562a20c996836d163259c87d9378120 Mon Sep 17 00:00:00 2001 From: "Luck, Tony" Date: Tue, 30 Jun 2015 15:57:51 -0700 Subject: efi: Handle memory error structures produced based on old versions of standard The memory error record structure includes as its first field a bitmask of which subsequent fields are valid. The allows new fields to be added to the structure while keeping compatibility with older software that parses these records. This mechanism was used between versions 2.2 and 2.3 to add four new fields, growing the size of the structure from 73 bytes to 80. But Linux just added all the new fields so this test: if (gdata->error_data_length >= sizeof(*mem_err)) cper_print_mem(newpfx, mem_err); else goto err_section_too_small; now make Linux complain about old format records being too short. Add a definition for the old format of the structure and use that for the minimum size check. Pass the actual size to cper_print_mem() so it can sanity check the validation_bits field to ensure that if a BIOS using the old format sets bits as if it were new, we won't access fields beyond the end of the structure. Signed-off-by: Tony Luck Cc: Signed-off-by: Matt Fleming diff --git a/drivers/firmware/efi/cper.c b/drivers/firmware/efi/cper.c index 4fd9961..d425374 100644 --- a/drivers/firmware/efi/cper.c +++ b/drivers/firmware/efi/cper.c @@ -305,10 +305,17 @@ const char *cper_mem_err_unpack(struct trace_seq *p, return ret; } -static void cper_print_mem(const char *pfx, const struct cper_sec_mem_err *mem) +static void cper_print_mem(const char *pfx, const struct cper_sec_mem_err *mem, + int len) { struct cper_mem_err_compact cmem; + /* Don't trust UEFI 2.1/2.2 structure with bad validation bits */ + if (len == sizeof(struct cper_sec_mem_err_old) && + (mem->validation_bits & ~(CPER_MEM_VALID_RANK_NUMBER - 1))) { + pr_err(FW_WARN "valid bits set for fields beyond structure\n"); + return; + } if (mem->validation_bits & CPER_MEM_VALID_ERROR_STATUS) printk("%s""error_status: 0x%016llx\n", pfx, mem->error_status); if (mem->validation_bits & CPER_MEM_VALID_PA) @@ -405,8 +412,10 @@ static void cper_estatus_print_section( } else if (!uuid_le_cmp(*sec_type, CPER_SEC_PLATFORM_MEM)) { struct cper_sec_mem_err *mem_err = (void *)(gdata + 1); printk("%s""section_type: memory error\n", newpfx); - if (gdata->error_data_length >= sizeof(*mem_err)) - cper_print_mem(newpfx, mem_err); + if (gdata->error_data_length >= + sizeof(struct cper_sec_mem_err_old)) + cper_print_mem(newpfx, mem_err, + gdata->error_data_length); else goto err_section_too_small; } else if (!uuid_le_cmp(*sec_type, CPER_SEC_PCIE)) { diff --git a/include/linux/cper.h b/include/linux/cper.h index 76abba4..dcacb1a 100644 --- a/include/linux/cper.h +++ b/include/linux/cper.h @@ -340,7 +340,27 @@ struct cper_ia_proc_ctx { __u64 mm_reg_addr; }; -/* Memory Error Section */ +/* Old Memory Error Section UEFI 2.1, 2.2 */ +struct cper_sec_mem_err_old { + __u64 validation_bits; + __u64 error_status; + __u64 physical_addr; + __u64 physical_addr_mask; + __u16 node; + __u16 card; + __u16 module; + __u16 bank; + __u16 device; + __u16 row; + __u16 column; + __u16 bit_pos; + __u64 requestor_id; + __u64 responder_id; + __u64 target_id; + __u8 error_type; +}; + +/* Memory Error Section UEFI >= 2.3 */ struct cper_sec_mem_err { __u64 validation_bits; __u64 error_status; -- cgit v0.10.2 From 0a0830feb2adce8c7234b8c166a32fe9e7616788 Mon Sep 17 00:00:00 2001 From: Roger Quadros Date: Tue, 2 Jun 2015 12:10:40 +0300 Subject: phy: ti-pipe3: fix suspend Relying on PM-ops for shutting down PHY clocks was a bad idea since the users (e.g. PCIe/SATA) might not have been suspended by then. The main culprit for not shutting down the clocks was the stray pm_runtime_get() call in probe. Fix the whole thing in the right way by getting rid of that pm_runtime_get() call from probe and removing all PM-ops. It is the sole responsibility of the PHY user to properly turn OFF and de-initialize the PHY as part of its suspend routine. As PHY core serializes init/exit we don't need to use a spinlock in this driver. So get rid of it. Signed-off-by: Roger Quadros Signed-off-by: Sekhar Nori Signed-off-by: Kishon Vijay Abraham I diff --git a/drivers/phy/phy-ti-pipe3.c b/drivers/phy/phy-ti-pipe3.c index 53f295c..3510b81 100644 --- a/drivers/phy/phy-ti-pipe3.c +++ b/drivers/phy/phy-ti-pipe3.c @@ -28,7 +28,6 @@ #include #include #include -#include #define PLL_STATUS 0x00000004 #define PLL_GO 0x00000008 @@ -83,10 +82,6 @@ struct ti_pipe3 { struct clk *refclk; struct clk *div_clk; struct pipe3_dpll_map *dpll_map; - bool enabled; - spinlock_t lock; /* serialize clock enable/disable */ - /* the below flag is needed specifically for SATA */ - bool refclk_enabled; }; static struct pipe3_dpll_map dpll_map_usb[] = { @@ -137,6 +132,9 @@ static struct pipe3_dpll_params *ti_pipe3_get_dpll_params(struct ti_pipe3 *phy) return NULL; } +static int ti_pipe3_enable_clocks(struct ti_pipe3 *phy); +static void ti_pipe3_disable_clocks(struct ti_pipe3 *phy); + static int ti_pipe3_power_off(struct phy *x) { struct ti_pipe3 *phy = phy_get_drvdata(x); @@ -217,6 +215,7 @@ static int ti_pipe3_init(struct phy *x) u32 val; int ret = 0; + ti_pipe3_enable_clocks(phy); /* * Set pcie_pcs register to 0x96 for proper functioning of phy * as recommended in AM572x TRM SPRUHZ6, section 18.5.2.2, table @@ -250,33 +249,35 @@ static int ti_pipe3_exit(struct phy *x) u32 val; unsigned long timeout; - /* SATA DPLL can't be powered down due to Errata i783 and PCIe - * does not have internal DPLL - */ - if (of_device_is_compatible(phy->dev->of_node, "ti,phy-pipe3-sata") || - of_device_is_compatible(phy->dev->of_node, "ti,phy-pipe3-pcie")) + /* SATA DPLL can't be powered down due to Errata i783 */ + if (of_device_is_compatible(phy->dev->of_node, "ti,phy-pipe3-sata")) return 0; - /* Put DPLL in IDLE mode */ - val = ti_pipe3_readl(phy->pll_ctrl_base, PLL_CONFIGURATION2); - val |= PLL_IDLE; - ti_pipe3_writel(phy->pll_ctrl_base, PLL_CONFIGURATION2, val); - - /* wait for LDO and Oscillator to power down */ - timeout = jiffies + msecs_to_jiffies(PLL_IDLE_TIME); - do { - cpu_relax(); - val = ti_pipe3_readl(phy->pll_ctrl_base, PLL_STATUS); - if ((val & PLL_TICOPWDN) && (val & PLL_LDOPWDN)) - break; - } while (!time_after(jiffies, timeout)); + /* PCIe doesn't have internal DPLL */ + if (!of_device_is_compatible(phy->dev->of_node, "ti,phy-pipe3-pcie")) { + /* Put DPLL in IDLE mode */ + val = ti_pipe3_readl(phy->pll_ctrl_base, PLL_CONFIGURATION2); + val |= PLL_IDLE; + ti_pipe3_writel(phy->pll_ctrl_base, PLL_CONFIGURATION2, val); - if (!(val & PLL_TICOPWDN) || !(val & PLL_LDOPWDN)) { - dev_err(phy->dev, "Failed to power down: PLL_STATUS 0x%x\n", - val); - return -EBUSY; + /* wait for LDO and Oscillator to power down */ + timeout = jiffies + msecs_to_jiffies(PLL_IDLE_TIME); + do { + cpu_relax(); + val = ti_pipe3_readl(phy->pll_ctrl_base, PLL_STATUS); + if ((val & PLL_TICOPWDN) && (val & PLL_LDOPWDN)) + break; + } while (!time_after(jiffies, timeout)); + + if (!(val & PLL_TICOPWDN) || !(val & PLL_LDOPWDN)) { + dev_err(phy->dev, "Failed to power down: PLL_STATUS 0x%x\n", + val); + return -EBUSY; + } } + ti_pipe3_disable_clocks(phy); + return 0; } static struct phy_ops ops = { @@ -306,7 +307,6 @@ static int ti_pipe3_probe(struct platform_device *pdev) return -ENOMEM; phy->dev = &pdev->dev; - spin_lock_init(&phy->lock); if (!of_device_is_compatible(node, "ti,phy-pipe3-pcie")) { match = of_match_device(ti_pipe3_id_table, &pdev->dev); @@ -402,6 +402,10 @@ static int ti_pipe3_probe(struct platform_device *pdev) platform_set_drvdata(pdev, phy); pm_runtime_enable(phy->dev); + /* Prevent auto-disable of refclk for SATA PHY due to Errata i783 */ + if (of_device_is_compatible(node, "ti,phy-pipe3-sata")) + if (!IS_ERR(phy->refclk)) + clk_prepare_enable(phy->refclk); generic_phy = devm_phy_create(phy->dev, NULL, &ops); if (IS_ERR(generic_phy)) @@ -413,63 +417,33 @@ static int ti_pipe3_probe(struct platform_device *pdev) if (IS_ERR(phy_provider)) return PTR_ERR(phy_provider); - pm_runtime_get(&pdev->dev); - return 0; } static int ti_pipe3_remove(struct platform_device *pdev) { - if (!pm_runtime_suspended(&pdev->dev)) - pm_runtime_put(&pdev->dev); pm_runtime_disable(&pdev->dev); return 0; } -#ifdef CONFIG_PM -static int ti_pipe3_enable_refclk(struct ti_pipe3 *phy) +static int ti_pipe3_enable_clocks(struct ti_pipe3 *phy) { - if (!IS_ERR(phy->refclk) && !phy->refclk_enabled) { - int ret; + int ret = 0; + if (!IS_ERR(phy->refclk)) { ret = clk_prepare_enable(phy->refclk); if (ret) { dev_err(phy->dev, "Failed to enable refclk %d\n", ret); return ret; } - phy->refclk_enabled = true; } - return 0; -} - -static void ti_pipe3_disable_refclk(struct ti_pipe3 *phy) -{ - if (!IS_ERR(phy->refclk)) - clk_disable_unprepare(phy->refclk); - - phy->refclk_enabled = false; -} - -static int ti_pipe3_enable_clocks(struct ti_pipe3 *phy) -{ - int ret = 0; - unsigned long flags; - - spin_lock_irqsave(&phy->lock, flags); - if (phy->enabled) - goto err1; - - ret = ti_pipe3_enable_refclk(phy); - if (ret) - goto err1; - if (!IS_ERR(phy->wkupclk)) { ret = clk_prepare_enable(phy->wkupclk); if (ret) { dev_err(phy->dev, "Failed to enable wkupclk %d\n", ret); - goto err2; + goto disable_refclk; } } @@ -477,96 +451,33 @@ static int ti_pipe3_enable_clocks(struct ti_pipe3 *phy) ret = clk_prepare_enable(phy->div_clk); if (ret) { dev_err(phy->dev, "Failed to enable div_clk %d\n", ret); - goto err3; + goto disable_wkupclk; } } - phy->enabled = true; - spin_unlock_irqrestore(&phy->lock, flags); return 0; -err3: +disable_wkupclk: if (!IS_ERR(phy->wkupclk)) clk_disable_unprepare(phy->wkupclk); -err2: +disable_refclk: if (!IS_ERR(phy->refclk)) clk_disable_unprepare(phy->refclk); - ti_pipe3_disable_refclk(phy); -err1: - spin_unlock_irqrestore(&phy->lock, flags); return ret; } static void ti_pipe3_disable_clocks(struct ti_pipe3 *phy) { - unsigned long flags; - - spin_lock_irqsave(&phy->lock, flags); - if (!phy->enabled) { - spin_unlock_irqrestore(&phy->lock, flags); - return; - } - if (!IS_ERR(phy->wkupclk)) clk_disable_unprepare(phy->wkupclk); - /* Don't disable refclk for SATA PHY due to Errata i783 */ - if (!of_device_is_compatible(phy->dev->of_node, "ti,phy-pipe3-sata")) - ti_pipe3_disable_refclk(phy); + if (!IS_ERR(phy->refclk)) + clk_disable_unprepare(phy->refclk); if (!IS_ERR(phy->div_clk)) clk_disable_unprepare(phy->div_clk); - phy->enabled = false; - spin_unlock_irqrestore(&phy->lock, flags); -} - -static int ti_pipe3_runtime_suspend(struct device *dev) -{ - struct ti_pipe3 *phy = dev_get_drvdata(dev); - - ti_pipe3_disable_clocks(phy); - return 0; } -static int ti_pipe3_runtime_resume(struct device *dev) -{ - struct ti_pipe3 *phy = dev_get_drvdata(dev); - int ret = 0; - - ret = ti_pipe3_enable_clocks(phy); - return ret; -} - -static int ti_pipe3_suspend(struct device *dev) -{ - struct ti_pipe3 *phy = dev_get_drvdata(dev); - - ti_pipe3_disable_clocks(phy); - return 0; -} - -static int ti_pipe3_resume(struct device *dev) -{ - struct ti_pipe3 *phy = dev_get_drvdata(dev); - int ret; - - ret = ti_pipe3_enable_clocks(phy); - if (ret) - return ret; - - pm_runtime_disable(dev); - pm_runtime_set_active(dev); - pm_runtime_enable(dev); - return 0; -} -#endif - -static const struct dev_pm_ops ti_pipe3_pm_ops = { - SET_RUNTIME_PM_OPS(ti_pipe3_runtime_suspend, - ti_pipe3_runtime_resume, NULL) - SET_SYSTEM_SLEEP_PM_OPS(ti_pipe3_suspend, ti_pipe3_resume) -}; - static const struct of_device_id ti_pipe3_id_table[] = { { .compatible = "ti,phy-usb3", @@ -592,7 +503,6 @@ static struct platform_driver ti_pipe3_driver = { .remove = ti_pipe3_remove, .driver = { .name = "ti-pipe3", - .pm = &ti_pipe3_pm_ops, .of_match_table = ti_pipe3_id_table, }, }; -- cgit v0.10.2 From 1c4b1d1da3d88ec3871459683d0e64fedbce0b16 Mon Sep 17 00:00:00 2001 From: Sebastian Ott Date: Mon, 29 Jun 2015 16:57:08 +0200 Subject: phy/pxa: add HAS_IOMEM dependency Fix this compile error: drivers/built-in.o: In function 'mv_usb2_phy_probe': phy-pxa-28nm-usb2.c:(.text+0x25ec): undefined reference to 'devm_ioremap_resource' drivers/built-in.o: In function 'mv_hsic_phy_probe': phy-pxa-28nm-hsic.c:(.text+0x3084): undefined reference to 'devm_ioremap_resource' Signed-off-by: Sebastian Ott Signed-off-by: Kishon Vijay Abraham I diff --git a/drivers/phy/Kconfig b/drivers/phy/Kconfig index c0e6ede..6b8dd16 100644 --- a/drivers/phy/Kconfig +++ b/drivers/phy/Kconfig @@ -56,6 +56,7 @@ config PHY_EXYNOS_MIPI_VIDEO config PHY_PXA_28NM_HSIC tristate "Marvell USB HSIC 28nm PHY Driver" + depends on HAS_IOMEM select GENERIC_PHY help Enable this to support Marvell USB HSIC PHY driver for Marvell @@ -66,6 +67,7 @@ config PHY_PXA_28NM_HSIC config PHY_PXA_28NM_USB2 tristate "Marvell USB 2.0 28nm PHY Driver" + depends on HAS_IOMEM select GENERIC_PHY help Enable this to support Marvell USB 2.0 PHY driver for Marvell -- cgit v0.10.2 From 96696a9df935d15fd2e89603454c20a692ec232a Mon Sep 17 00:00:00 2001 From: Thomas Hebb Date: Thu, 2 Jul 2015 01:04:18 -0400 Subject: phy: berlin-usb: fix divider for BG2CD The marvell,berlin2cd-usb-phy compatible incorrectly sets the PLL divider to BG2's value instead of BG2CD/BG2Q's. Change it to the right value. Signed-off-by: Thomas Hebb Cc: stable@vger.kernel.org Signed-off-by: Kishon Vijay Abraham I diff --git a/drivers/phy/phy-berlin-usb.c b/drivers/phy/phy-berlin-usb.c index c6fc95b..ab54f28 100644 --- a/drivers/phy/phy-berlin-usb.c +++ b/drivers/phy/phy-berlin-usb.c @@ -106,8 +106,8 @@ static const u32 phy_berlin_pll_dividers[] = { /* Berlin 2 */ CLK_REF_DIV(0xc) | FEEDBACK_CLK_DIV(0x54), - /* Berlin 2CD */ - CLK_REF_DIV(0x6) | FEEDBACK_CLK_DIV(0x55), + /* Berlin 2CD/Q */ + CLK_REF_DIV(0xc) | FEEDBACK_CLK_DIV(0x54), }; struct phy_berlin_usb_priv { -- cgit v0.10.2 From dcb54fcb3483862e993abdae99cec22fb3ae4099 Mon Sep 17 00:00:00 2001 From: Thomas Hebb Date: Thu, 2 Jul 2015 01:04:26 -0400 Subject: phy: berlin-usb: fix divider for BG2 The USB PLL divider set by the marvell,berlin2-usb-phy compatible is not correct for BG2. We couldn't change it before because BG2Q incorrectly used the same compatible string. Now that BG2Q's compatible is fixed, change BG2's divider to the correct value. Signed-off-by: Thomas Hebb Tested-by: Antoine Tenart Signed-off-by: Kishon Vijay Abraham I diff --git a/drivers/phy/phy-berlin-usb.c b/drivers/phy/phy-berlin-usb.c index ab54f28..335e06d 100644 --- a/drivers/phy/phy-berlin-usb.c +++ b/drivers/phy/phy-berlin-usb.c @@ -105,7 +105,7 @@ static const u32 phy_berlin_pll_dividers[] = { /* Berlin 2 */ - CLK_REF_DIV(0xc) | FEEDBACK_CLK_DIV(0x54), + CLK_REF_DIV(0x6) | FEEDBACK_CLK_DIV(0x55), /* Berlin 2CD/Q */ CLK_REF_DIV(0xc) | FEEDBACK_CLK_DIV(0x54), }; -- cgit v0.10.2 From 71d126fd28de2d4d9b7b2088dbccd7ca62fad6e0 Mon Sep 17 00:00:00 2001 From: Arne Fitzenreiter Date: Wed, 15 Jul 2015 13:54:36 +0200 Subject: libata: add ATA_HORKAGE_NOTRIM Some devices lose data on TRIM whether queued or not. This patch adds a horkage to disable TRIM. tj: Collapsed unnecessary if() nesting. Signed-off-by: Arne Fitzenreiter Signed-off-by: Tejun Heo Cc: stable@vger.kernel.org diff --git a/drivers/ata/libata-scsi.c b/drivers/ata/libata-scsi.c index 3131adc..641a61a 100644 --- a/drivers/ata/libata-scsi.c +++ b/drivers/ata/libata-scsi.c @@ -2568,7 +2568,8 @@ static unsigned int ata_scsiop_read_cap(struct ata_scsi_args *args, u8 *rbuf) rbuf[14] = (lowest_aligned >> 8) & 0x3f; rbuf[15] = lowest_aligned; - if (ata_id_has_trim(args->id)) { + if (ata_id_has_trim(args->id) && + !(dev->horkage & ATA_HORKAGE_NOTRIM)) { rbuf[14] |= 0x80; /* LBPME */ if (ata_id_has_zero_after_trim(args->id) && diff --git a/drivers/ata/libata-transport.c b/drivers/ata/libata-transport.c index d6c37bc..e2d9497 100644 --- a/drivers/ata/libata-transport.c +++ b/drivers/ata/libata-transport.c @@ -569,6 +569,8 @@ show_ata_dev_trim(struct device *dev, if (!ata_id_has_trim(ata_dev->id)) mode = "unsupported"; + else if (ata_dev->horkage & ATA_HORKAGE_NOTRIM) + mode = "forced_unsupported"; else if (ata_dev->horkage & ATA_HORKAGE_NO_NCQ_TRIM) mode = "forced_unqueued"; else if (ata_fpdma_dsm_supported(ata_dev)) diff --git a/include/linux/libata.h b/include/linux/libata.h index 36ce37b..5c8bac6 100644 --- a/include/linux/libata.h +++ b/include/linux/libata.h @@ -431,6 +431,8 @@ enum { ATA_HORKAGE_WD_BROKEN_LPM = (1 << 21), /* some WDs have broken LPM */ ATA_HORKAGE_ZERO_AFTER_TRIM = (1 << 22),/* guarantees zero after trim */ ATA_HORKAGE_NO_NCQ_LOG = (1 << 23), /* don't use NCQ for log read */ + ATA_HORKAGE_NOTRIM = (1 << 24), /* don't use TRIM */ + /* DMA mask for user DMA control: User visible values; DO NOT renumber */ -- cgit v0.10.2 From cda57b1b05cf7b8b99ab4b732bea0b05b6c015cc Mon Sep 17 00:00:00 2001 From: Arne Fitzenreiter Date: Wed, 15 Jul 2015 13:54:37 +0200 Subject: libata: force disable trim for SuperSSpeed S238 This device loses blocks, often the partition table area, on trim. Disable TRIM. http://pcengines.ch/msata16a.htm Signed-off-by: Arne Fitzenreiter Signed-off-by: Tejun Heo Cc: stable@vger.kernel.org diff --git a/drivers/ata/libata-core.c b/drivers/ata/libata-core.c index 58a6db8f..ed2b218 100644 --- a/drivers/ata/libata-core.c +++ b/drivers/ata/libata-core.c @@ -4239,6 +4239,9 @@ static const struct ata_blacklist_entry ata_device_blacklist [] = { { "Samsung SSD 8*", NULL, ATA_HORKAGE_NO_NCQ_TRIM | ATA_HORKAGE_ZERO_AFTER_TRIM, }, + /* devices that don't properly handle TRIM commands */ + { "SuperSSpeed S238*", NULL, ATA_HORKAGE_NOTRIM, }, + /* * As defined, the DRAT (Deterministic Read After Trim) and RZAT * (Return Zero After Trim) flags in the ATA Command Set are -- cgit v0.10.2 From af34d637637eabaf49406eb35c948cd51ba262a6 Mon Sep 17 00:00:00 2001 From: David Milburn Date: Mon, 13 Jul 2015 11:48:23 -0500 Subject: libata: add ATA_HORKAGE_MAX_SEC_1024 to revert back to previous max_sectors limit Since no longer limiting max_sectors to BLK_DEF_MAX_SECTORS (commit 34b48db66e08), data corruption may occur on ST380013AS drive configured on 82801JI (ICH10 Family) SATA controller. This patch will allow the driver to limit max_sectors as before # cat /sys/block/sdb/queue/max_sectors_kb 512 I was able to double the max_sectors_kb value up to 16384 on linux-4.2.0-rc2 before seeing corruption, but seems safer to use previous limit. Without this patch max_sectors_kb will be 32767. tj: Minor comment update. Reported-by: Jeff Moyer Signed-off-by: David Milburn Signed-off-by: Tejun Heo Cc: stable@vger.kernel.org # v3.19 and later Fixes: 34b48db66e08 ("block: remove artifical max_hw_sectors cap") diff --git a/drivers/ata/libata-core.c b/drivers/ata/libata-core.c index ed2b218..68202a8 100644 --- a/drivers/ata/libata-core.c +++ b/drivers/ata/libata-core.c @@ -2478,6 +2478,10 @@ int ata_dev_configure(struct ata_device *dev) dev->max_sectors = min_t(unsigned int, ATA_MAX_SECTORS_128, dev->max_sectors); + if (dev->horkage & ATA_HORKAGE_MAX_SEC_1024) + dev->max_sectors = min_t(unsigned int, ATA_MAX_SECTORS_1024, + dev->max_sectors); + if (dev->horkage & ATA_HORKAGE_MAX_SEC_LBA48) dev->max_sectors = ATA_MAX_SECTORS_LBA48; @@ -4146,6 +4150,12 @@ static const struct ata_blacklist_entry ata_device_blacklist [] = { { "Slimtype DVD A DS8A8SH", NULL, ATA_HORKAGE_MAX_SEC_LBA48 }, { "Slimtype DVD A DS8A9SH", NULL, ATA_HORKAGE_MAX_SEC_LBA48 }, + /* + * Causes silent data corruption with higher max sects. + * http://lkml.kernel.org/g/x49wpy40ysk.fsf@segfault.boston.devel.redhat.com + */ + { "ST380013AS", "3.20", ATA_HORKAGE_MAX_SEC_1024 }, + /* Devices we expect to fail diagnostics */ /* Devices where NCQ should be avoided */ diff --git a/include/linux/ata.h b/include/linux/ata.h index fed3641..6c78956 100644 --- a/include/linux/ata.h +++ b/include/linux/ata.h @@ -45,6 +45,7 @@ enum { ATA_SECT_SIZE = 512, ATA_MAX_SECTORS_128 = 128, ATA_MAX_SECTORS = 256, + ATA_MAX_SECTORS_1024 = 1024, ATA_MAX_SECTORS_LBA48 = 65535,/* TODO: 65536? */ ATA_MAX_SECTORS_TAPE = 65535, diff --git a/include/linux/libata.h b/include/linux/libata.h index 5c8bac6..c9cfbcd 100644 --- a/include/linux/libata.h +++ b/include/linux/libata.h @@ -432,7 +432,7 @@ enum { ATA_HORKAGE_ZERO_AFTER_TRIM = (1 << 22),/* guarantees zero after trim */ ATA_HORKAGE_NO_NCQ_LOG = (1 << 23), /* don't use NCQ for log read */ ATA_HORKAGE_NOTRIM = (1 << 24), /* don't use TRIM */ - + ATA_HORKAGE_MAX_SEC_1024 = (1 << 25), /* Limit max sects to 1024 */ /* DMA mask for user DMA control: User visible values; DO NOT renumber */ -- cgit v0.10.2 From d531be2ca2f27cca5f041b6a140504999144a617 Mon Sep 17 00:00:00 2001 From: Mikulas Patocka Date: Wed, 8 Jul 2015 13:06:12 -0400 Subject: libata: increase the timeout when setting transfer mode I have a ST4000DM000 disk. If Linux is booted while the disk is spun down, the command that sets transfer mode causes the disk to spin up. The spin-up takes longer than the default 5s timeout, so the command fails and timeout is reported. Fix this by increasing the timeout to 15s, which is enough for the disk to spin up. Signed-off-by: Mikulas Patocka Cc: stable@vger.kernel.org Signed-off-by: Tejun Heo diff --git a/drivers/ata/libata-core.c b/drivers/ata/libata-core.c index 68202a8..12b176a 100644 --- a/drivers/ata/libata-core.c +++ b/drivers/ata/libata-core.c @@ -4515,7 +4515,8 @@ static unsigned int ata_dev_set_xfermode(struct ata_device *dev) else /* In the ancient relic department - skip all of this */ return 0; - err_mask = ata_exec_internal(dev, &tf, NULL, DMA_NONE, NULL, 0, 0); + /* On some disks, this command causes spin-up, so we need longer timeout */ + err_mask = ata_exec_internal(dev, &tf, NULL, DMA_NONE, NULL, 0, 15000); DPRINTK("EXIT, err_mask=%x\n", err_mask); return err_mask; -- cgit v0.10.2 From 91a57731eed05714609855b1d72e6118b6b3ec6f Mon Sep 17 00:00:00 2001 From: Roger Quadros Date: Wed, 8 Jul 2015 17:34:43 +0300 Subject: ARM: DRA7: hwmod: fix gpmc hwmod GPMC smart idle is not really broken but it does not support smart idle with wakeup. Fixes: 556708fe8718 ("ARM: OMAP: DRA7: hwmod: Make gpmc software supervised as the smart idle is broken") Signed-off-by: Roger Quadros Reviewed-by: Paul Walmsley Signed-off-by: Paul Walmsley diff --git a/arch/arm/mach-omap2/omap_hwmod_7xx_data.c b/arch/arm/mach-omap2/omap_hwmod_7xx_data.c index 2606c66..562247b 100644 --- a/arch/arm/mach-omap2/omap_hwmod_7xx_data.c +++ b/arch/arm/mach-omap2/omap_hwmod_7xx_data.c @@ -827,8 +827,7 @@ static struct omap_hwmod_class_sysconfig dra7xx_gpmc_sysc = { .syss_offs = 0x0014, .sysc_flags = (SYSC_HAS_AUTOIDLE | SYSC_HAS_SIDLEMODE | SYSC_HAS_SOFTRESET | SYSS_HAS_RESET_STATUS), - .idlemodes = (SIDLE_FORCE | SIDLE_NO | SIDLE_SMART | - SIDLE_SMART_WKUP), + .idlemodes = (SIDLE_FORCE | SIDLE_NO | SIDLE_SMART), .sysc_fields = &omap_hwmod_sysc_type1, }; @@ -844,7 +843,7 @@ static struct omap_hwmod dra7xx_gpmc_hwmod = { .class = &dra7xx_gpmc_hwmod_class, .clkdm_name = "l3main1_clkdm", /* Skip reset for CONFIG_OMAP_GPMC_DEBUG for bootloader timings */ - .flags = HWMOD_SWSUP_SIDLE | DEBUG_OMAP_GPMC_HWMOD_FLAGS, + .flags = DEBUG_OMAP_GPMC_HWMOD_FLAGS, .main_clk = "l3_iclk_div", .prcm = { .omap4 = { -- cgit v0.10.2 From 035d210f928ce083435b4fd351a26d126c02c927 Mon Sep 17 00:00:00 2001 From: Daniel Borkmann Date: Mon, 13 Jul 2015 00:06:02 +0200 Subject: rtnetlink: reject non-IFLA_VF_PORT attributes inside IFLA_VF_PORTS Similarly as in commit 4f7d2cdfdde7 ("rtnetlink: verify IFLA_VF_INFO attributes before passing them to driver"), we have a double nesting of netlink attributes, i.e. IFLA_VF_PORTS only contains IFLA_VF_PORT that is nested itself. While IFLA_VF_PORTS is a verified attribute from ifla_policy[], we only check if the IFLA_VF_PORTS container has IFLA_VF_PORT attributes and then pass the attribute's content itself via nla_parse_nested(). It would be more correct to reject inner types other than IFLA_VF_PORT instead of continuing parsing and also similarly as in commit 4f7d2cdfdde7, to check for a minimum of NLA_HDRLEN. Signed-off-by: Daniel Borkmann Cc: Roopa Prabhu Cc: Scott Feldman Cc: Jason Gunthorpe Acked-by: Roopa Prabhu Signed-off-by: David S. Miller diff --git a/net/core/rtnetlink.c b/net/core/rtnetlink.c index 9e433d5..dc004b1 100644 --- a/net/core/rtnetlink.c +++ b/net/core/rtnetlink.c @@ -1804,10 +1804,13 @@ static int do_setlink(const struct sk_buff *skb, goto errout; nla_for_each_nested(attr, tb[IFLA_VF_PORTS], rem) { - if (nla_type(attr) != IFLA_VF_PORT) - continue; - err = nla_parse_nested(port, IFLA_PORT_MAX, - attr, ifla_port_policy); + if (nla_type(attr) != IFLA_VF_PORT || + nla_len(attr) < NLA_HDRLEN) { + err = -EINVAL; + goto errout; + } + err = nla_parse_nested(port, IFLA_PORT_MAX, attr, + ifla_port_policy); if (err < 0) goto errout; if (!port[IFLA_PORT_VF]) { -- cgit v0.10.2 From 738ac1ebb96d02e0d23bc320302a6ea94c612dec Mon Sep 17 00:00:00 2001 From: Herbert Xu Date: Mon, 13 Jul 2015 16:04:13 +0800 Subject: net: Clone skb before setting peeked flag Shared skbs must not be modified and this is crucial for broadcast and/or multicast paths where we use it as an optimisation to avoid unnecessary cloning. The function skb_recv_datagram breaks this rule by setting peeked without cloning the skb first. This causes funky races which leads to double-free. This patch fixes this by cloning the skb and replacing the skb in the list when setting skb->peeked. Fixes: a59322be07c9 ("[UDP]: Only increment counter on first peek/recv") Reported-by: Konstantin Khlebnikov Signed-off-by: Herbert Xu Signed-off-by: David S. Miller diff --git a/net/core/datagram.c b/net/core/datagram.c index b80fb91..4e9a3f6 100644 --- a/net/core/datagram.c +++ b/net/core/datagram.c @@ -131,6 +131,35 @@ out_noerr: goto out; } +static int skb_set_peeked(struct sk_buff *skb) +{ + struct sk_buff *nskb; + + if (skb->peeked) + return 0; + + /* We have to unshare an skb before modifying it. */ + if (!skb_shared(skb)) + goto done; + + nskb = skb_clone(skb, GFP_ATOMIC); + if (!nskb) + return -ENOMEM; + + skb->prev->next = nskb; + skb->next->prev = nskb; + nskb->prev = skb->prev; + nskb->next = skb->next; + + consume_skb(skb); + skb = nskb; + +done: + skb->peeked = 1; + + return 0; +} + /** * __skb_recv_datagram - Receive a datagram skbuff * @sk: socket @@ -165,7 +194,9 @@ out_noerr: struct sk_buff *__skb_recv_datagram(struct sock *sk, unsigned int flags, int *peeked, int *off, int *err) { + struct sk_buff_head *queue = &sk->sk_receive_queue; struct sk_buff *skb, *last; + unsigned long cpu_flags; long timeo; /* * Caller is allowed not to check sk->sk_err before skb_recv_datagram() @@ -184,8 +215,6 @@ struct sk_buff *__skb_recv_datagram(struct sock *sk, unsigned int flags, * Look at current nfs client by the way... * However, this function was correct in any case. 8) */ - unsigned long cpu_flags; - struct sk_buff_head *queue = &sk->sk_receive_queue; int _off = *off; last = (struct sk_buff *)queue; @@ -199,7 +228,11 @@ struct sk_buff *__skb_recv_datagram(struct sock *sk, unsigned int flags, _off -= skb->len; continue; } - skb->peeked = 1; + + error = skb_set_peeked(skb); + if (error) + goto unlock_err; + atomic_inc(&skb->users); } else __skb_unlink(skb, queue); @@ -223,6 +256,8 @@ struct sk_buff *__skb_recv_datagram(struct sock *sk, unsigned int flags, return NULL; +unlock_err: + spin_unlock_irqrestore(&queue->lock, cpu_flags); no_packet: *err = error; return NULL; -- cgit v0.10.2 From da278622bf04f8ddb14519a2b8214e108ef26101 Mon Sep 17 00:00:00 2001 From: Richard Stearn Date: Mon, 13 Jul 2015 11:38:24 +0200 Subject: NET: AX.25: Stop heartbeat timer on disconnect. This may result in a kernel panic. The bug has always existed but somehow we've run out of luck now and it bites. Signed-off-by: Richard Stearn Cc: stable@vger.kernel.org # all branches Signed-off-by: Ralf Baechle Signed-off-by: David S. Miller diff --git a/net/ax25/ax25_subr.c b/net/ax25/ax25_subr.c index 1997538..3b78e84 100644 --- a/net/ax25/ax25_subr.c +++ b/net/ax25/ax25_subr.c @@ -264,6 +264,7 @@ void ax25_disconnect(ax25_cb *ax25, int reason) { ax25_clear_queues(ax25); + ax25_stop_heartbeat(ax25); ax25_stop_t1timer(ax25); ax25_stop_t2timer(ax25); ax25_stop_t3timer(ax25); -- cgit v0.10.2 From c9805b9986ed88b7df1b14ea7d538d83f2149cf5 Mon Sep 17 00:00:00 2001 From: Fabio Estevam Date: Mon, 13 Jul 2015 08:13:52 -0300 Subject: Revert "net: fec: Ensure clocks are enabled while using mdio bus" This reverts commit 6c3e921b18edca290099adfddde8a50236bf2d80. commit 6c3e921b18ed ("net: fec: Ensure clocks are enabled while using mdio bus") prevents the kernel to boot on mx6 boards, so let's revert it. Reported-by: Tyler Baker Signed-off-by: Fabio Estevam Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/freescale/fec_main.c b/drivers/net/ethernet/freescale/fec_main.c index 42e20e5..1f89c59 100644 --- a/drivers/net/ethernet/freescale/fec_main.c +++ b/drivers/net/ethernet/freescale/fec_main.c @@ -24,7 +24,6 @@ #include #include #include -#include #include #include #include @@ -78,7 +77,6 @@ static void fec_enet_itr_coal_init(struct net_device *ndev); #define FEC_ENET_RAEM_V 0x8 #define FEC_ENET_RAFL_V 0x8 #define FEC_ENET_OPD_V 0xFFF0 -#define FEC_MDIO_PM_TIMEOUT 100 /* ms */ static struct platform_device_id fec_devtype[] = { { @@ -1769,13 +1767,7 @@ static void fec_enet_adjust_link(struct net_device *ndev) static int fec_enet_mdio_read(struct mii_bus *bus, int mii_id, int regnum) { struct fec_enet_private *fep = bus->priv; - struct device *dev = &fep->pdev->dev; unsigned long time_left; - int ret = 0; - - ret = pm_runtime_get_sync(dev); - if (IS_ERR_VALUE(ret)) - return ret; fep->mii_timeout = 0; init_completion(&fep->mdio_done); @@ -1791,30 +1783,18 @@ static int fec_enet_mdio_read(struct mii_bus *bus, int mii_id, int regnum) if (time_left == 0) { fep->mii_timeout = 1; netdev_err(fep->netdev, "MDIO read timeout\n"); - ret = -ETIMEDOUT; - goto out; + return -ETIMEDOUT; } - ret = FEC_MMFR_DATA(readl(fep->hwp + FEC_MII_DATA)); - -out: - pm_runtime_mark_last_busy(dev); - pm_runtime_put_autosuspend(dev); - - return ret; + /* return value */ + return FEC_MMFR_DATA(readl(fep->hwp + FEC_MII_DATA)); } static int fec_enet_mdio_write(struct mii_bus *bus, int mii_id, int regnum, u16 value) { struct fec_enet_private *fep = bus->priv; - struct device *dev = &fep->pdev->dev; unsigned long time_left; - int ret = 0; - - ret = pm_runtime_get_sync(dev); - if (IS_ERR_VALUE(ret)) - return ret; fep->mii_timeout = 0; init_completion(&fep->mdio_done); @@ -1831,13 +1811,10 @@ static int fec_enet_mdio_write(struct mii_bus *bus, int mii_id, int regnum, if (time_left == 0) { fep->mii_timeout = 1; netdev_err(fep->netdev, "MDIO write timeout\n"); - ret = -ETIMEDOUT; + return -ETIMEDOUT; } - pm_runtime_mark_last_busy(dev); - pm_runtime_put_autosuspend(dev); - - return ret; + return 0; } static int fec_enet_clk_enable(struct net_device *ndev, bool enable) @@ -1849,6 +1826,9 @@ static int fec_enet_clk_enable(struct net_device *ndev, bool enable) ret = clk_prepare_enable(fep->clk_ahb); if (ret) return ret; + ret = clk_prepare_enable(fep->clk_ipg); + if (ret) + goto failed_clk_ipg; if (fep->clk_enet_out) { ret = clk_prepare_enable(fep->clk_enet_out); if (ret) @@ -1872,6 +1852,7 @@ static int fec_enet_clk_enable(struct net_device *ndev, bool enable) } } else { clk_disable_unprepare(fep->clk_ahb); + clk_disable_unprepare(fep->clk_ipg); if (fep->clk_enet_out) clk_disable_unprepare(fep->clk_enet_out); if (fep->clk_ptp) { @@ -1893,6 +1874,8 @@ failed_clk_ptp: if (fep->clk_enet_out) clk_disable_unprepare(fep->clk_enet_out); failed_clk_enet_out: + clk_disable_unprepare(fep->clk_ipg); +failed_clk_ipg: clk_disable_unprepare(fep->clk_ahb); return ret; @@ -2864,14 +2847,10 @@ fec_enet_open(struct net_device *ndev) struct fec_enet_private *fep = netdev_priv(ndev); int ret; - ret = pm_runtime_get_sync(&fep->pdev->dev); - if (IS_ERR_VALUE(ret)) - return ret; - pinctrl_pm_select_default_state(&fep->pdev->dev); ret = fec_enet_clk_enable(ndev, true); if (ret) - goto clk_enable; + return ret; /* I should reset the ring buffers here, but I don't yet know * a simple way to do that. @@ -2902,9 +2881,6 @@ err_enet_mii_probe: fec_enet_free_buffers(ndev); err_enet_alloc: fec_enet_clk_enable(ndev, false); -clk_enable: - pm_runtime_mark_last_busy(&fep->pdev->dev); - pm_runtime_put_autosuspend(&fep->pdev->dev); pinctrl_pm_select_sleep_state(&fep->pdev->dev); return ret; } @@ -2927,9 +2903,6 @@ fec_enet_close(struct net_device *ndev) fec_enet_clk_enable(ndev, false); pinctrl_pm_select_sleep_state(&fep->pdev->dev); - pm_runtime_mark_last_busy(&fep->pdev->dev); - pm_runtime_put_autosuspend(&fep->pdev->dev); - fec_enet_free_buffers(ndev); return 0; @@ -3415,10 +3388,6 @@ fec_probe(struct platform_device *pdev) if (ret) goto failed_clk; - ret = clk_prepare_enable(fep->clk_ipg); - if (ret) - goto failed_clk_ipg; - fep->reg_phy = devm_regulator_get(&pdev->dev, "phy"); if (!IS_ERR(fep->reg_phy)) { ret = regulator_enable(fep->reg_phy); @@ -3465,8 +3434,6 @@ fec_probe(struct platform_device *pdev) netif_carrier_off(ndev); fec_enet_clk_enable(ndev, false); pinctrl_pm_select_sleep_state(&pdev->dev); - pm_runtime_set_active(&pdev->dev); - pm_runtime_enable(&pdev->dev); ret = register_netdev(ndev); if (ret) @@ -3480,12 +3447,6 @@ fec_probe(struct platform_device *pdev) fep->rx_copybreak = COPYBREAK_DEFAULT; INIT_WORK(&fep->tx_timeout_work, fec_enet_timeout_work); - - pm_runtime_set_autosuspend_delay(&pdev->dev, FEC_MDIO_PM_TIMEOUT); - pm_runtime_use_autosuspend(&pdev->dev); - pm_runtime_mark_last_busy(&pdev->dev); - pm_runtime_put_autosuspend(&pdev->dev); - return 0; failed_register: @@ -3496,8 +3457,6 @@ failed_init: if (fep->reg_phy) regulator_disable(fep->reg_phy); failed_regulator: - clk_disable_unprepare(fep->clk_ipg); -failed_clk_ipg: fec_enet_clk_enable(ndev, false); failed_clk: failed_phy: @@ -3609,28 +3568,7 @@ failed_clk: return ret; } -static int __maybe_unused fec_runtime_suspend(struct device *dev) -{ - struct net_device *ndev = dev_get_drvdata(dev); - struct fec_enet_private *fep = netdev_priv(ndev); - - clk_disable_unprepare(fep->clk_ipg); - - return 0; -} - -static int __maybe_unused fec_runtime_resume(struct device *dev) -{ - struct net_device *ndev = dev_get_drvdata(dev); - struct fec_enet_private *fep = netdev_priv(ndev); - - return clk_prepare_enable(fep->clk_ipg); -} - -static const struct dev_pm_ops fec_pm_ops = { - SET_SYSTEM_SLEEP_PM_OPS(fec_suspend, fec_resume) - SET_RUNTIME_PM_OPS(fec_runtime_suspend, fec_runtime_resume, NULL) -}; +static SIMPLE_DEV_PM_OPS(fec_pm_ops, fec_suspend, fec_resume); static struct platform_driver fec_driver = { .driver = { -- cgit v0.10.2 From 89c22d8c3b278212eef6a8cc66b570bc840a6f5a Mon Sep 17 00:00:00 2001 From: Herbert Xu Date: Mon, 13 Jul 2015 20:01:42 +0800 Subject: net: Fix skb csum races when peeking When we calculate the checksum on the recv path, we store the result in the skb as an optimisation in case we need the checksum again down the line. This is in fact bogus for the MSG_PEEK case as this is done without any locking. So multiple threads can peek and then store the result to the same skb, potentially resulting in bogus skb states. This patch fixes this by only storing the result if the skb is not shared. This preserves the optimisations for the few cases where it can be done safely due to locking or other reasons, e.g., SIOCINQ. Signed-off-by: Herbert Xu Acked-by: Eric Dumazet Signed-off-by: David S. Miller diff --git a/net/core/datagram.c b/net/core/datagram.c index 4e9a3f6..4967262 100644 --- a/net/core/datagram.c +++ b/net/core/datagram.c @@ -657,7 +657,8 @@ __sum16 __skb_checksum_complete_head(struct sk_buff *skb, int len) !skb->csum_complete_sw) netdev_rx_csum_fault(skb->dev); } - skb->csum_valid = !sum; + if (!skb_shared(skb)) + skb->csum_valid = !sum; return sum; } EXPORT_SYMBOL(__skb_checksum_complete_head); @@ -677,11 +678,13 @@ __sum16 __skb_checksum_complete(struct sk_buff *skb) netdev_rx_csum_fault(skb->dev); } - /* Save full packet checksum */ - skb->csum = csum; - skb->ip_summed = CHECKSUM_COMPLETE; - skb->csum_complete_sw = 1; - skb->csum_valid = !sum; + if (!skb_shared(skb)) { + /* Save full packet checksum */ + skb->csum = csum; + skb->ip_summed = CHECKSUM_COMPLETE; + skb->csum_complete_sw = 1; + skb->csum_valid = !sum; + } return sum; } -- cgit v0.10.2 From bc8c20acaea154efc558f5f4122ed65d396f6156 Mon Sep 17 00:00:00 2001 From: Satish Ashok Date: Mon, 13 Jul 2015 05:28:37 -0700 Subject: bridge: multicast: treat igmpv3 report with INCLUDE and no sources as a leave A report with INCLUDE/Change_to_include and empty source list should be treated as a leave, specified by RFC 3376, section 3.1: "If the requested filter mode is INCLUDE *and* the requested source list is empty, then the entry corresponding to the requested interface and multicast address is deleted if present. If no such entry is present, the request is ignored." Signed-off-by: Satish Ashok Signed-off-by: Nikolay Aleksandrov Signed-off-by: David S. Miller diff --git a/net/bridge/br_multicast.c b/net/bridge/br_multicast.c index 742a6c2..79db489 100644 --- a/net/bridge/br_multicast.c +++ b/net/bridge/br_multicast.c @@ -39,6 +39,16 @@ static void br_multicast_start_querier(struct net_bridge *br, struct bridge_mcast_own_query *query); static void br_multicast_add_router(struct net_bridge *br, struct net_bridge_port *port); +static void br_ip4_multicast_leave_group(struct net_bridge *br, + struct net_bridge_port *port, + __be32 group, + __u16 vid); +#if IS_ENABLED(CONFIG_IPV6) +static void br_ip6_multicast_leave_group(struct net_bridge *br, + struct net_bridge_port *port, + const struct in6_addr *group, + __u16 vid); +#endif unsigned int br_mdb_rehash_seq; static inline int br_ip_equal(const struct br_ip *a, const struct br_ip *b) @@ -1010,9 +1020,15 @@ static int br_ip4_multicast_igmp3_report(struct net_bridge *br, continue; } - err = br_ip4_multicast_add_group(br, port, group, vid); - if (err) - break; + if ((type == IGMPV3_CHANGE_TO_INCLUDE || + type == IGMPV3_MODE_IS_INCLUDE) && + ntohs(grec->grec_nsrcs) == 0) { + br_ip4_multicast_leave_group(br, port, group, vid); + } else { + err = br_ip4_multicast_add_group(br, port, group, vid); + if (err) + break; + } } return err; @@ -1071,10 +1087,17 @@ static int br_ip6_multicast_mld2_report(struct net_bridge *br, continue; } - err = br_ip6_multicast_add_group(br, port, &grec->grec_mca, - vid); - if (err) - break; + if ((grec->grec_type == MLD2_CHANGE_TO_INCLUDE || + grec->grec_type == MLD2_MODE_IS_INCLUDE) && + ntohs(*nsrcs) == 0) { + br_ip6_multicast_leave_group(br, port, &grec->grec_mca, + vid); + } else { + err = br_ip6_multicast_add_group(br, port, + &grec->grec_mca, vid); + if (!err) + break; + } } return err; -- cgit v0.10.2 From 5ebc784625ea68a9570d1f70557e7932988cd1b4 Mon Sep 17 00:00:00 2001 From: Nikolay Aleksandrov Date: Mon, 13 Jul 2015 06:36:19 -0700 Subject: bridge: mdb: fix double add notification Since the mdb add/del code was introduced there have been 2 br_mdb_notify calls when doing br_mdb_add() resulting in 2 notifications on each add. Example: Command: bridge mdb add dev br0 port eth1 grp 239.0.0.1 permanent Before patch: root@debian:~# bridge monitor all [MDB]dev br0 port eth1 grp 239.0.0.1 permanent [MDB]dev br0 port eth1 grp 239.0.0.1 permanent After patch: root@debian:~# bridge monitor all [MDB]dev br0 port eth1 grp 239.0.0.1 permanent Signed-off-by: Nikolay Aleksandrov Fixes: cfd567543590 ("bridge: add support of adding and deleting mdb entries") Signed-off-by: David S. Miller diff --git a/net/bridge/br_mdb.c b/net/bridge/br_mdb.c index c11cf26..1198a3d 100644 --- a/net/bridge/br_mdb.c +++ b/net/bridge/br_mdb.c @@ -351,7 +351,6 @@ static int br_mdb_add_group(struct net_bridge *br, struct net_bridge_port *port, if (state == MDB_TEMPORARY) mod_timer(&p->timer, now + br->multicast_membership_interval); - br_mdb_notify(br->dev, port, group, RTM_NEWMDB); return 0; } -- cgit v0.10.2 From f82b681a511f4d61069e9586a9cf97bdef371ef3 Mon Sep 17 00:00:00 2001 From: Yuchung Cheng Date: Mon, 13 Jul 2015 12:10:20 -0700 Subject: tcp: don't use F-RTO on non-recurring timeouts Currently F-RTO may repeatedly send new data packets on non-recurring timeouts in CA_Loss mode. This is a bug because F-RTO (RFC5682) should only be used on either new recovery or recurring timeouts. This exacerbates the recovery progress during frequent timeout & repair, because we prioritize sending new data packets instead of repairing the holes when the bandwidth is already scarce. Fix it by correcting the test of a new recovery episode. Signed-off-by: Yuchung Cheng Signed-off-by: Neal Cardwell Signed-off-by: David S. Miller diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c index 684f095..728f5b3 100644 --- a/net/ipv4/tcp_input.c +++ b/net/ipv4/tcp_input.c @@ -1917,14 +1917,13 @@ void tcp_enter_loss(struct sock *sk) const struct inet_connection_sock *icsk = inet_csk(sk); struct tcp_sock *tp = tcp_sk(sk); struct sk_buff *skb; - bool new_recovery = false; + bool new_recovery = icsk->icsk_ca_state < TCP_CA_Recovery; bool is_reneg; /* is receiver reneging on SACKs? */ /* Reduce ssthresh if it has not yet been made inside this window. */ if (icsk->icsk_ca_state <= TCP_CA_Disorder || !after(tp->high_seq, tp->snd_una) || (icsk->icsk_ca_state == TCP_CA_Loss && !icsk->icsk_retransmits)) { - new_recovery = true; tp->prior_ssthresh = tcp_current_ssthresh(sk); tp->snd_ssthresh = icsk->icsk_ca_ops->ssthresh(sk); tcp_ca_event(sk, CA_EVENT_LOSS); -- cgit v0.10.2 From 052cbda41fdc243a8d40cce7ab3a6327b4b2887e Mon Sep 17 00:00:00 2001 From: WANG Cong Date: Mon, 13 Jul 2015 12:30:07 -0700 Subject: fq_codel: fix a use-after-free Fixes: 25331d6ce42b ("net: sched: implement qstat helper routines") Cc: John Fastabend Signed-off-by: Cong Wang Signed-off-by: Cong Wang Acked-by: Eric Dumazet Signed-off-by: David S. Miller diff --git a/net/sched/sch_fq_codel.c b/net/sched/sch_fq_codel.c index d75993f..06e7c84 100644 --- a/net/sched/sch_fq_codel.c +++ b/net/sched/sch_fq_codel.c @@ -155,10 +155,10 @@ static unsigned int fq_codel_drop(struct Qdisc *sch) skb = dequeue_head(flow); len = qdisc_pkt_len(skb); q->backlogs[idx] -= len; - kfree_skb(skb); sch->q.qlen--; qdisc_qstats_drop(sch); qdisc_qstats_backlog_dec(sch, skb); + kfree_skb(skb); flow->dropped++; return idx; } -- cgit v0.10.2 From fd98e9419d8d622a4de91f76b306af6aa627aa9c Mon Sep 17 00:00:00 2001 From: Tilman Schmidt Date: Tue, 14 Jul 2015 00:37:13 +0200 Subject: isdn/gigaset: reset tty->receive_room when attaching ser_gigaset Commit 79901317ce80 ("n_tty: Don't flush buffer when closing ldisc"), first merged in kernel release 3.10, caused the following regression in the Gigaset M101 driver: Before that commit, when closing the N_TTY line discipline in preparation to switching to N_GIGASET_M101, receive_room would be reset to a non-zero value by the call to n_tty_flush_buffer() in n_tty's close method. With the removal of that call, receive_room might be left at zero, blocking data reception on the serial line. The present patch fixes that regression by setting receive_room to an appropriate value in the ldisc open method. Fixes: 79901317ce80 ("n_tty: Don't flush buffer when closing ldisc") Signed-off-by: Tilman Schmidt Signed-off-by: David S. Miller diff --git a/drivers/isdn/gigaset/ser-gigaset.c b/drivers/isdn/gigaset/ser-gigaset.c index 8c91fd5..3ac9c41 100644 --- a/drivers/isdn/gigaset/ser-gigaset.c +++ b/drivers/isdn/gigaset/ser-gigaset.c @@ -524,9 +524,18 @@ gigaset_tty_open(struct tty_struct *tty) cs->hw.ser->tty = tty; atomic_set(&cs->hw.ser->refcnt, 1); init_completion(&cs->hw.ser->dead_cmp); - tty->disc_data = cs; + /* Set the amount of data we're willing to receive per call + * from the hardware driver to half of the input buffer size + * to leave some reserve. + * Note: We don't do flow control towards the hardware driver. + * If more data is received than will fit into the input buffer, + * it will be dropped and an error will be logged. This should + * never happen as the device is slow and the buffer size ample. + */ + tty->receive_room = RBUFSIZE/2; + /* OK.. Initialization of the datastructures and the HW is done.. Now * startup system and notify the LL that we are ready to run */ -- cgit v0.10.2 From 62c13bad39211387afd4354e8e127a535d13e70d Mon Sep 17 00:00:00 2001 From: Tilman Schmidt Date: Tue, 14 Jul 2015 00:37:13 +0200 Subject: isdn/gigaset: drop unused ldisc methods The line discipline read and write methods are optional so the dummy methods in ser_gigaset are unnecessary and can be removed. Signed-off-by: Tilman Schmidt Signed-off-by: David S. Miller diff --git a/drivers/isdn/gigaset/ser-gigaset.c b/drivers/isdn/gigaset/ser-gigaset.c index 3ac9c41..375be50 100644 --- a/drivers/isdn/gigaset/ser-gigaset.c +++ b/drivers/isdn/gigaset/ser-gigaset.c @@ -607,28 +607,6 @@ static int gigaset_tty_hangup(struct tty_struct *tty) } /* - * Read on the tty. - * Unused, received data goes only to the Gigaset driver. - */ -static ssize_t -gigaset_tty_read(struct tty_struct *tty, struct file *file, - unsigned char __user *buf, size_t count) -{ - return -EAGAIN; -} - -/* - * Write on the tty. - * Unused, transmit data comes only from the Gigaset driver. - */ -static ssize_t -gigaset_tty_write(struct tty_struct *tty, struct file *file, - const unsigned char *buf, size_t count) -{ - return -EAGAIN; -} - -/* * Ioctl on the tty. * Called in process context only. * May be re-entered by multiple ioctl calling threads. @@ -761,8 +739,6 @@ static struct tty_ldisc_ops gigaset_ldisc = { .open = gigaset_tty_open, .close = gigaset_tty_close, .hangup = gigaset_tty_hangup, - .read = gigaset_tty_read, - .write = gigaset_tty_write, .ioctl = gigaset_tty_ioctl, .receive_buf = gigaset_tty_receive, .write_wakeup = gigaset_tty_wakeup, -- cgit v0.10.2 From 03645a11a570d52e70631838cb786eb4253eb463 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Tue, 14 Jul 2015 08:10:22 +0200 Subject: ipv6: lock socket in ip6_datagram_connect() ip6_datagram_connect() is doing a lot of socket changes without socket being locked. This looks wrong, at least for udp_lib_rehash() which could corrupt lists because of concurrent udp_sk(sk)->udp_portaddr_hash accesses. Signed-off-by: Eric Dumazet Acked-by: Herbert Xu Signed-off-by: David S. Miller diff --git a/include/net/ip.h b/include/net/ip.h index 0750a18..d5fe9f2 100644 --- a/include/net/ip.h +++ b/include/net/ip.h @@ -161,6 +161,7 @@ static inline __u8 get_rtconn_flags(struct ipcm_cookie* ipc, struct sock* sk) } /* datagram.c */ +int __ip4_datagram_connect(struct sock *sk, struct sockaddr *uaddr, int addr_len); int ip4_datagram_connect(struct sock *sk, struct sockaddr *uaddr, int addr_len); void ip4_datagram_release_cb(struct sock *sk); diff --git a/net/ipv4/datagram.c b/net/ipv4/datagram.c index 90c0e83..574fad9 100644 --- a/net/ipv4/datagram.c +++ b/net/ipv4/datagram.c @@ -20,7 +20,7 @@ #include #include -int ip4_datagram_connect(struct sock *sk, struct sockaddr *uaddr, int addr_len) +int __ip4_datagram_connect(struct sock *sk, struct sockaddr *uaddr, int addr_len) { struct inet_sock *inet = inet_sk(sk); struct sockaddr_in *usin = (struct sockaddr_in *) uaddr; @@ -39,8 +39,6 @@ int ip4_datagram_connect(struct sock *sk, struct sockaddr *uaddr, int addr_len) sk_dst_reset(sk); - lock_sock(sk); - oif = sk->sk_bound_dev_if; saddr = inet->inet_saddr; if (ipv4_is_multicast(usin->sin_addr.s_addr)) { @@ -82,9 +80,19 @@ int ip4_datagram_connect(struct sock *sk, struct sockaddr *uaddr, int addr_len) sk_dst_set(sk, &rt->dst); err = 0; out: - release_sock(sk); return err; } +EXPORT_SYMBOL(__ip4_datagram_connect); + +int ip4_datagram_connect(struct sock *sk, struct sockaddr *uaddr, int addr_len) +{ + int res; + + lock_sock(sk); + res = __ip4_datagram_connect(sk, uaddr, addr_len); + release_sock(sk); + return res; +} EXPORT_SYMBOL(ip4_datagram_connect); /* Because UDP xmit path can manipulate sk_dst_cache without holding diff --git a/net/ipv6/datagram.c b/net/ipv6/datagram.c index 62d908e..b10a889 100644 --- a/net/ipv6/datagram.c +++ b/net/ipv6/datagram.c @@ -40,7 +40,7 @@ static bool ipv6_mapped_addr_any(const struct in6_addr *a) return ipv6_addr_v4mapped(a) && (a->s6_addr32[3] == 0); } -int ip6_datagram_connect(struct sock *sk, struct sockaddr *uaddr, int addr_len) +static int __ip6_datagram_connect(struct sock *sk, struct sockaddr *uaddr, int addr_len) { struct sockaddr_in6 *usin = (struct sockaddr_in6 *) uaddr; struct inet_sock *inet = inet_sk(sk); @@ -56,7 +56,7 @@ int ip6_datagram_connect(struct sock *sk, struct sockaddr *uaddr, int addr_len) if (usin->sin6_family == AF_INET) { if (__ipv6_only_sock(sk)) return -EAFNOSUPPORT; - err = ip4_datagram_connect(sk, uaddr, addr_len); + err = __ip4_datagram_connect(sk, uaddr, addr_len); goto ipv4_connected; } @@ -98,9 +98,9 @@ int ip6_datagram_connect(struct sock *sk, struct sockaddr *uaddr, int addr_len) sin.sin_addr.s_addr = daddr->s6_addr32[3]; sin.sin_port = usin->sin6_port; - err = ip4_datagram_connect(sk, - (struct sockaddr *) &sin, - sizeof(sin)); + err = __ip4_datagram_connect(sk, + (struct sockaddr *) &sin, + sizeof(sin)); ipv4_connected: if (err) @@ -204,6 +204,16 @@ out: fl6_sock_release(flowlabel); return err; } + +int ip6_datagram_connect(struct sock *sk, struct sockaddr *uaddr, int addr_len) +{ + int res; + + lock_sock(sk); + res = __ip6_datagram_connect(sk, uaddr, addr_len); + release_sock(sk); + return res; +} EXPORT_SYMBOL_GPL(ip6_datagram_connect); int ip6_datagram_connect_v6_only(struct sock *sk, struct sockaddr *uaddr, -- cgit v0.10.2 From 515866f8185b92fb18a782408c53839f003c7669 Mon Sep 17 00:00:00 2001 From: Konstantin Khlebnikov Date: Tue, 14 Jul 2015 16:35:50 +0300 Subject: ipvlan: remove counters of ipv4 and ipv6 addresses They are unused after commit f631c44bbe15 ("ipvlan: Always set broadcast bit in multicast filter"). Signed-off-by: Konstantin Khlebnikov Signed-off-by: David S. Miller diff --git a/drivers/net/ipvlan/ipvlan.h b/drivers/net/ipvlan/ipvlan.h index 953a974..68e2549 100644 --- a/drivers/net/ipvlan/ipvlan.h +++ b/drivers/net/ipvlan/ipvlan.h @@ -67,8 +67,6 @@ struct ipvl_dev { struct ipvl_port *port; struct net_device *phy_dev; struct list_head addrs; - int ipv4cnt; - int ipv6cnt; struct ipvl_pcpu_stats __percpu *pcpu_stats; DECLARE_BITMAP(mac_filters, IPVLAN_MAC_FILTER_SIZE); netdev_features_t sfeatures; diff --git a/drivers/net/ipvlan/ipvlan_main.c b/drivers/net/ipvlan/ipvlan_main.c index 1acc283..048ecf0 100644 --- a/drivers/net/ipvlan/ipvlan_main.c +++ b/drivers/net/ipvlan/ipvlan_main.c @@ -153,10 +153,9 @@ static int ipvlan_open(struct net_device *dev) else dev->flags &= ~IFF_NOARP; - if (ipvlan->ipv6cnt > 0 || ipvlan->ipv4cnt > 0) { - list_for_each_entry(addr, &ipvlan->addrs, anode) - ipvlan_ht_addr_add(ipvlan, addr); - } + list_for_each_entry(addr, &ipvlan->addrs, anode) + ipvlan_ht_addr_add(ipvlan, addr); + return dev_uc_add(phy_dev, phy_dev->dev_addr); } @@ -171,10 +170,9 @@ static int ipvlan_stop(struct net_device *dev) dev_uc_del(phy_dev, phy_dev->dev_addr); - if (ipvlan->ipv6cnt > 0 || ipvlan->ipv4cnt > 0) { - list_for_each_entry(addr, &ipvlan->addrs, anode) - ipvlan_ht_addr_del(addr, !dev->dismantle); - } + list_for_each_entry(addr, &ipvlan->addrs, anode) + ipvlan_ht_addr_del(addr, !dev->dismantle); + return 0; } @@ -471,8 +469,6 @@ static int ipvlan_link_new(struct net *src_net, struct net_device *dev, ipvlan->port = port; ipvlan->sfeatures = IPVLAN_FEATURES; INIT_LIST_HEAD(&ipvlan->addrs); - ipvlan->ipv4cnt = 0; - ipvlan->ipv6cnt = 0; /* TODO Probably put random address here to be presented to the * world but keep using the physical-dev address for the outgoing @@ -508,12 +504,11 @@ static void ipvlan_link_delete(struct net_device *dev, struct list_head *head) struct ipvl_dev *ipvlan = netdev_priv(dev); struct ipvl_addr *addr, *next; - if (ipvlan->ipv6cnt > 0 || ipvlan->ipv4cnt > 0) { - list_for_each_entry_safe(addr, next, &ipvlan->addrs, anode) { - ipvlan_ht_addr_del(addr, !dev->dismantle); - list_del(&addr->anode); - } + list_for_each_entry_safe(addr, next, &ipvlan->addrs, anode) { + ipvlan_ht_addr_del(addr, !dev->dismantle); + list_del(&addr->anode); } + list_del_rcu(&ipvlan->pnode); unregister_netdevice_queue(dev, head); netdev_upper_dev_unlink(ipvlan->phy_dev, dev); @@ -627,7 +622,7 @@ static int ipvlan_add_addr6(struct ipvl_dev *ipvlan, struct in6_addr *ip6_addr) memcpy(&addr->ip6addr, ip6_addr, sizeof(struct in6_addr)); addr->atype = IPVL_IPV6; list_add_tail(&addr->anode, &ipvlan->addrs); - ipvlan->ipv6cnt++; + /* If the interface is not up, the address will be added to the hash * list by ipvlan_open. */ @@ -647,8 +642,6 @@ static void ipvlan_del_addr6(struct ipvl_dev *ipvlan, struct in6_addr *ip6_addr) ipvlan_ht_addr_del(addr, true); list_del(&addr->anode); - ipvlan->ipv6cnt--; - WARN_ON(ipvlan->ipv6cnt < 0); kfree_rcu(addr, rcu); return; @@ -699,7 +692,7 @@ static int ipvlan_add_addr4(struct ipvl_dev *ipvlan, struct in_addr *ip4_addr) memcpy(&addr->ip4addr, ip4_addr, sizeof(struct in_addr)); addr->atype = IPVL_IPV4; list_add_tail(&addr->anode, &ipvlan->addrs); - ipvlan->ipv4cnt++; + /* If the interface is not up, the address will be added to the hash * list by ipvlan_open. */ @@ -719,8 +712,6 @@ static void ipvlan_del_addr4(struct ipvl_dev *ipvlan, struct in_addr *ip4_addr) ipvlan_ht_addr_del(addr, true); list_del(&addr->anode); - ipvlan->ipv4cnt--; - WARN_ON(ipvlan->ipv4cnt < 0); kfree_rcu(addr, rcu); return; -- cgit v0.10.2 From 6a725497318545aae246232ed05a8df9cffb0a02 Mon Sep 17 00:00:00 2001 From: Konstantin Khlebnikov Date: Tue, 14 Jul 2015 16:35:51 +0300 Subject: ipvlan: plug memory leak in ipvlan_link_delete Add missing kfree_rcu(addr, rcu); Signed-off-by: Konstantin Khlebnikov Signed-off-by: David S. Miller diff --git a/drivers/net/ipvlan/ipvlan_main.c b/drivers/net/ipvlan/ipvlan_main.c index 048ecf0..7d81e37 100644 --- a/drivers/net/ipvlan/ipvlan_main.c +++ b/drivers/net/ipvlan/ipvlan_main.c @@ -507,6 +507,7 @@ static void ipvlan_link_delete(struct net_device *dev, struct list_head *head) list_for_each_entry_safe(addr, next, &ipvlan->addrs, anode) { ipvlan_ht_addr_del(addr, !dev->dismantle); list_del(&addr->anode); + kfree_rcu(addr, rcu); } list_del_rcu(&ipvlan->pnode); -- cgit v0.10.2 From 6640e673c6f3dbaace085ca2686a8a343dc23a71 Mon Sep 17 00:00:00 2001 From: Konstantin Khlebnikov Date: Tue, 14 Jul 2015 16:35:53 +0300 Subject: ipvlan: unhash addresses without synchronize_rcu All structures used in traffic forwarding are rcu-protected: ipvl_addr, ipvl_dev and ipvl_port. Thus we can unhash addresses without synchronization. We'll anyway hash it back into the same bucket: in worst case lockless lookup will scan hash once again. Signed-off-by: Konstantin Khlebnikov Signed-off-by: David S. Miller diff --git a/drivers/net/ipvlan/ipvlan.h b/drivers/net/ipvlan/ipvlan.h index 68e2549..40f9d7e 100644 --- a/drivers/net/ipvlan/ipvlan.h +++ b/drivers/net/ipvlan/ipvlan.h @@ -122,5 +122,5 @@ struct ipvl_addr *ipvlan_find_addr(const struct ipvl_dev *ipvlan, bool ipvlan_addr_busy(struct ipvl_port *port, void *iaddr, bool is_v6); struct ipvl_addr *ipvlan_ht_addr_lookup(const struct ipvl_port *port, const void *iaddr, bool is_v6); -void ipvlan_ht_addr_del(struct ipvl_addr *addr, bool sync); +void ipvlan_ht_addr_del(struct ipvl_addr *addr); #endif /* __IPVLAN_H */ diff --git a/drivers/net/ipvlan/ipvlan_core.c b/drivers/net/ipvlan/ipvlan_core.c index 8afbeda..8f8628a 100644 --- a/drivers/net/ipvlan/ipvlan_core.c +++ b/drivers/net/ipvlan/ipvlan_core.c @@ -85,11 +85,9 @@ void ipvlan_ht_addr_add(struct ipvl_dev *ipvlan, struct ipvl_addr *addr) hlist_add_head_rcu(&addr->hlnode, &port->hlhead[hash]); } -void ipvlan_ht_addr_del(struct ipvl_addr *addr, bool sync) +void ipvlan_ht_addr_del(struct ipvl_addr *addr) { hlist_del_init_rcu(&addr->hlnode); - if (sync) - synchronize_rcu(); } struct ipvl_addr *ipvlan_find_addr(const struct ipvl_dev *ipvlan, diff --git a/drivers/net/ipvlan/ipvlan_main.c b/drivers/net/ipvlan/ipvlan_main.c index 7d81e37..e995bc5 100644 --- a/drivers/net/ipvlan/ipvlan_main.c +++ b/drivers/net/ipvlan/ipvlan_main.c @@ -171,7 +171,7 @@ static int ipvlan_stop(struct net_device *dev) dev_uc_del(phy_dev, phy_dev->dev_addr); list_for_each_entry(addr, &ipvlan->addrs, anode) - ipvlan_ht_addr_del(addr, !dev->dismantle); + ipvlan_ht_addr_del(addr); return 0; } @@ -505,7 +505,7 @@ static void ipvlan_link_delete(struct net_device *dev, struct list_head *head) struct ipvl_addr *addr, *next; list_for_each_entry_safe(addr, next, &ipvlan->addrs, anode) { - ipvlan_ht_addr_del(addr, !dev->dismantle); + ipvlan_ht_addr_del(addr); list_del(&addr->anode); kfree_rcu(addr, rcu); } @@ -641,7 +641,7 @@ static void ipvlan_del_addr6(struct ipvl_dev *ipvlan, struct in6_addr *ip6_addr) if (!addr) return; - ipvlan_ht_addr_del(addr, true); + ipvlan_ht_addr_del(addr); list_del(&addr->anode); kfree_rcu(addr, rcu); @@ -711,7 +711,7 @@ static void ipvlan_del_addr4(struct ipvl_dev *ipvlan, struct in_addr *ip4_addr) if (!addr) return; - ipvlan_ht_addr_del(addr, true); + ipvlan_ht_addr_del(addr); list_del(&addr->anode); kfree_rcu(addr, rcu); -- cgit v0.10.2 From 0fba37a3af03a7e74bf9e75473729adb98da49c3 Mon Sep 17 00:00:00 2001 From: WANG Cong Date: Tue, 14 Jul 2015 16:35:54 +0300 Subject: ipvlan: use rcu_deference_bh() in ipvlan_queue_xmit() In tx path rcu_read_lock_bh() is held, so we need rcu_deference_bh(). This fixes the following warning: =============================== [ INFO: suspicious RCU usage. ] 4.1.0-rc1+ #1007 Not tainted ------------------------------- drivers/net/ipvlan/ipvlan.h:106 suspicious rcu_dereference_check() usage! other info that might help us debug this: rcu_scheduler_active = 1, debug_locks = 0 1 lock held by dhclient/1076: #0: (rcu_read_lock_bh){......}, at: [] rcu_lock_acquire+0x0/0x26 stack backtrace: CPU: 2 PID: 1076 Comm: dhclient Not tainted 4.1.0-rc1+ #1007 Hardware name: Bochs Bochs, BIOS Bochs 01/01/2011 0000000000000001 ffff8800d381bac8 ffffffff81a4154f 000000003c1a3c19 ffff8800d4d0a690 ffff8800d381baf8 ffffffff810b849f ffff880117d41148 ffff880117d40000 ffff880117d40068 0000000000000156 ffff8800d381bb18 Call Trace: [] dump_stack+0x4c/0x65 [] lockdep_rcu_suspicious+0x107/0x110 [] ipvlan_port_get_rcu+0x47/0x4e [] ipvlan_queue_xmit+0x35/0x450 [] ? rcu_read_unlock+0x3e/0x5f [] ? local_clock+0x19/0x22 [] ? __lock_is_held+0x39/0x52 [] ipvlan_start_xmit+0x1b/0x44 [] dev_hard_start_xmit+0x2ae/0x467 [] __dev_queue_xmit+0x50a/0x60c [] dev_queue_xmit_sk+0x13/0x15 [] dev_queue_xmit+0x10/0x12 [] packet_sendmsg+0xb6b/0xbdf [] ? mark_lock+0x2e/0x226 [] ? sched_clock_cpu+0x9e/0xb7 [] sock_sendmsg_nosec+0x12/0x1d [] sock_sendmsg+0x29/0x2e [] sock_write_iter+0x70/0x91 [] __vfs_write+0x7e/0xa7 [] vfs_write+0x92/0xe8 [] SyS_write+0x47/0x7e [] system_call_fastpath+0x12/0x6f Fixes: 2ad7bf363841 ("ipvlan: Initial check-in of the IPVLAN driver.") Cc: Mahesh Bandewar Signed-off-by: Cong Wang Acked-by: Mahesh Bandewar Acked-by: Konstantin Khlebnikov Signed-off-by: David S. Miller diff --git a/drivers/net/ipvlan/ipvlan.h b/drivers/net/ipvlan/ipvlan.h index 40f9d7e..9542b7b 100644 --- a/drivers/net/ipvlan/ipvlan.h +++ b/drivers/net/ipvlan/ipvlan.h @@ -104,6 +104,11 @@ static inline struct ipvl_port *ipvlan_port_get_rcu(const struct net_device *d) return rcu_dereference(d->rx_handler_data); } +static inline struct ipvl_port *ipvlan_port_get_rcu_bh(const struct net_device *d) +{ + return rcu_dereference_bh(d->rx_handler_data); +} + static inline struct ipvl_port *ipvlan_port_get_rtnl(const struct net_device *d) { return rtnl_dereference(d->rx_handler_data); diff --git a/drivers/net/ipvlan/ipvlan_core.c b/drivers/net/ipvlan/ipvlan_core.c index 8f8628a..207f62e 100644 --- a/drivers/net/ipvlan/ipvlan_core.c +++ b/drivers/net/ipvlan/ipvlan_core.c @@ -529,7 +529,7 @@ static int ipvlan_xmit_mode_l2(struct sk_buff *skb, struct net_device *dev) int ipvlan_queue_xmit(struct sk_buff *skb, struct net_device *dev) { struct ipvl_dev *ipvlan = netdev_priv(dev); - struct ipvl_port *port = ipvlan_port_get_rcu(ipvlan->phy_dev); + struct ipvl_port *port = ipvlan_port_get_rcu_bh(ipvlan->phy_dev); if (!port) goto out; -- cgit v0.10.2 From 23a5a49c83dd8a7201a42e96d24238bde3547c11 Mon Sep 17 00:00:00 2001 From: Konstantin Khlebnikov Date: Tue, 14 Jul 2015 16:35:55 +0300 Subject: ipvlan: ignore addresses from ipv6 autoconfiguration Inet6addr notifier is atomic and runs in bh context without RTNL when ipv6 receives router advertisement packet and performs autoconfiguration. Proper fix still in discussion. Let's at least plug the bug. v1: http://lkml.kernel.org/r/20150514135618.14062.1969.stgit@buzz v2: http://lkml.kernel.org/r/20150703125840.24121.91556.stgit@buzz Signed-off-by: Konstantin Khlebnikov Signed-off-by: David S. Miller diff --git a/drivers/net/ipvlan/ipvlan_main.c b/drivers/net/ipvlan/ipvlan_main.c index e995bc5..20b58bd 100644 --- a/drivers/net/ipvlan/ipvlan_main.c +++ b/drivers/net/ipvlan/ipvlan_main.c @@ -655,6 +655,10 @@ static int ipvlan_addr6_event(struct notifier_block *unused, struct net_device *dev = (struct net_device *)if6->idev->dev; struct ipvl_dev *ipvlan = netdev_priv(dev); + /* FIXME IPv6 autoconf calls us from bh without RTNL */ + if (in_softirq()) + return NOTIFY_DONE; + if (!netif_is_ipvlan(dev)) return NOTIFY_DONE; -- cgit v0.10.2 From e8d092aafd9e68c04d7b468e95ff7a617998a796 Mon Sep 17 00:00:00 2001 From: WANG Cong Date: Tue, 14 Jul 2015 11:21:57 -0700 Subject: net_sched: fix a use-after-free in sfq Fixes: 25331d6ce42b ("net: sched: implement qstat helper routines") Cc: John Fastabend Signed-off-by: Cong Wang Signed-off-by: Cong Wang Signed-off-by: David S. Miller diff --git a/net/sched/sch_sfq.c b/net/sched/sch_sfq.c index 7d14926..52f75a5 100644 --- a/net/sched/sch_sfq.c +++ b/net/sched/sch_sfq.c @@ -306,10 +306,10 @@ drop: len = qdisc_pkt_len(skb); slot->backlog -= len; sfq_dec(q, x); - kfree_skb(skb); sch->q.qlen--; qdisc_qstats_drop(sch); qdisc_qstats_backlog_dec(sch, skb); + kfree_skb(skb); return len; } -- cgit v0.10.2 From c0afd9ce4d6a646fb6433536f95a418bb348fab1 Mon Sep 17 00:00:00 2001 From: WANG Cong Date: Tue, 14 Jul 2015 11:21:58 -0700 Subject: fq_codel: fix return value of fq_codel_drop() The ->drop() is supposed to return the number of bytes it dropped, however fq_codel_drop() returns the index of the flow where it drops a packet from. Fix this by introducing a helper to wrap fq_codel_drop(). Cc: Eric Dumazet Signed-off-by: Cong Wang Signed-off-by: Cong Wang Acked-by: Eric Dumazet Signed-off-by: David S. Miller diff --git a/net/sched/sch_fq_codel.c b/net/sched/sch_fq_codel.c index 06e7c84..21ca33c 100644 --- a/net/sched/sch_fq_codel.c +++ b/net/sched/sch_fq_codel.c @@ -163,6 +163,15 @@ static unsigned int fq_codel_drop(struct Qdisc *sch) return idx; } +static unsigned int fq_codel_qdisc_drop(struct Qdisc *sch) +{ + unsigned int prev_backlog; + + prev_backlog = sch->qstats.backlog; + fq_codel_drop(sch); + return prev_backlog - sch->qstats.backlog; +} + static int fq_codel_enqueue(struct sk_buff *skb, struct Qdisc *sch) { struct fq_codel_sched_data *q = qdisc_priv(sch); @@ -604,7 +613,7 @@ static struct Qdisc_ops fq_codel_qdisc_ops __read_mostly = { .enqueue = fq_codel_enqueue, .dequeue = fq_codel_dequeue, .peek = qdisc_peek_dequeued, - .drop = fq_codel_drop, + .drop = fq_codel_qdisc_drop, .init = fq_codel_init, .reset = fq_codel_reset, .destroy = fq_codel_destroy, -- cgit v0.10.2 From ddf06c1e569a64a44c4c750ae45b2604f19e45f0 Mon Sep 17 00:00:00 2001 From: Alexei Starovoitov Date: Tue, 14 Jul 2015 12:15:19 -0700 Subject: tc: act_bpf: fix memory leak prog->bpf_ops is populated when act_bpf is used with classic BPF and prog->bpf_name is optionally used with extended BPF. Fix memory leak when act_bpf is released. Fixes: d23b8ad8ab23 ("tc: add BPF based action") Fixes: a8cb5f556b56 ("act_bpf: add initial eBPF support for actions") Acked-by: Daniel Borkmann Signed-off-by: Alexei Starovoitov Signed-off-by: David S. Miller diff --git a/net/sched/act_bpf.c b/net/sched/act_bpf.c index 1d56903..1df7828 100644 --- a/net/sched/act_bpf.c +++ b/net/sched/act_bpf.c @@ -339,6 +339,9 @@ static void tcf_bpf_cleanup(struct tc_action *act, int bind) bpf_prog_put(prog->filter); else bpf_prog_destroy(prog->filter); + + kfree(prog->bpf_ops); + kfree(prog->bpf_name); } static struct tc_action_ops act_bpf_ops __read_mostly = { -- cgit v0.10.2 From 25b401c1816ae64bcc5dcb1d39ab41812522a0ce Mon Sep 17 00:00:00 2001 From: Stefan Agner Date: Mon, 18 May 2015 18:33:27 +0200 Subject: can: mcp251x: fix resume when device is down If a valid power regulator or a dummy regulator is used (which happens to be the case when no regulator is specified), restart_work is queued no matter whether the device was running or not at suspend time. Since work queues get initialized in the ndo_open callback, resuming leads to a NULL pointer exception. Reverse exactly the steps executed at suspend time: - Enable the power regulator in any case - Enable the transceiver regulator if the device was running, even in case we have a power regulator - Queue restart_work only in case the device was running Fixes: bf66f3736a94 ("can: mcp251x: Move to threaded interrupts instead of workqueues.") Signed-off-by: Stefan Agner Cc: linux-stable Signed-off-by: Marc Kleine-Budde diff --git a/drivers/net/can/spi/mcp251x.c b/drivers/net/can/spi/mcp251x.c index c1a95a3..3b2f34e 100644 --- a/drivers/net/can/spi/mcp251x.c +++ b/drivers/net/can/spi/mcp251x.c @@ -1222,17 +1222,16 @@ static int __maybe_unused mcp251x_can_resume(struct device *dev) struct spi_device *spi = to_spi_device(dev); struct mcp251x_priv *priv = spi_get_drvdata(spi); - if (priv->after_suspend & AFTER_SUSPEND_POWER) { + if (priv->after_suspend & AFTER_SUSPEND_POWER) mcp251x_power_enable(priv->power, 1); + + if (priv->after_suspend & AFTER_SUSPEND_UP) { + mcp251x_power_enable(priv->transceiver, 1); queue_work(priv->wq, &priv->restart_work); } else { - if (priv->after_suspend & AFTER_SUSPEND_UP) { - mcp251x_power_enable(priv->transceiver, 1); - queue_work(priv->wq, &priv->restart_work); - } else { - priv->after_suspend = 0; - } + priv->after_suspend = 0; } + priv->force_quit = 0; enable_irq(spi->irq); return 0; -- cgit v0.10.2 From 69da3f2ac528642acbd06ed14564ac1b9a918394 Mon Sep 17 00:00:00 2001 From: Stefan Agner Date: Mon, 18 May 2015 18:33:28 +0200 Subject: can: mcp251x: get regulators optionally The regulators power and transceiver are optional. If those are not present, the pointer (or error pointer) is correctly handled by the driver, hence we can use devm_regulator_get_optional safely, which avoids regulators getting created. Signed-off-by: Stefan Agner Signed-off-by: Marc Kleine-Budde diff --git a/drivers/net/can/spi/mcp251x.c b/drivers/net/can/spi/mcp251x.c index 3b2f34e..b7e83c2 100644 --- a/drivers/net/can/spi/mcp251x.c +++ b/drivers/net/can/spi/mcp251x.c @@ -1086,8 +1086,8 @@ static int mcp251x_can_probe(struct spi_device *spi) if (ret) goto out_clk; - priv->power = devm_regulator_get(&spi->dev, "vdd"); - priv->transceiver = devm_regulator_get(&spi->dev, "xceiver"); + priv->power = devm_regulator_get_optional(&spi->dev, "vdd"); + priv->transceiver = devm_regulator_get_optional(&spi->dev, "xceiver"); if ((PTR_ERR(priv->power) == -EPROBE_DEFER) || (PTR_ERR(priv->transceiver) == -EPROBE_DEFER)) { ret = -EPROBE_DEFER; -- cgit v0.10.2 From 9051bd393cf25e76dfb45409792719a854661500 Mon Sep 17 00:00:00 2001 From: "Martin K. Petersen" Date: Wed, 15 Jul 2015 21:03:23 -0400 Subject: libata: Do not blacklist M510DC A new Micron drive was just announced, once again recycling the first part of the model string. Add an underscore to the M510/M550 pattern to avoid picking up the new DC drive. Signed-off-by: Martin K. Petersen Cc: Signed-off-by: Tejun Heo diff --git a/drivers/ata/libata-core.c b/drivers/ata/libata-core.c index 12b176a..db5d9f7 100644 --- a/drivers/ata/libata-core.c +++ b/drivers/ata/libata-core.c @@ -4240,7 +4240,7 @@ static const struct ata_blacklist_entry ata_device_blacklist [] = { ATA_HORKAGE_ZERO_AFTER_TRIM, }, { "Crucial_CT*M500*", NULL, ATA_HORKAGE_NO_NCQ_TRIM | ATA_HORKAGE_ZERO_AFTER_TRIM, }, - { "Micron_M5[15]0*", "MU01", ATA_HORKAGE_NO_NCQ_TRIM | + { "Micron_M5[15]0_*", "MU01", ATA_HORKAGE_NO_NCQ_TRIM | ATA_HORKAGE_ZERO_AFTER_TRIM, }, { "Crucial_CT*M550*", "MU01", ATA_HORKAGE_NO_NCQ_TRIM | ATA_HORKAGE_ZERO_AFTER_TRIM, }, -- cgit v0.10.2 From 7d01cd261c76f95913c81554a751968a1d282d3a Mon Sep 17 00:00:00 2001 From: Oleksij Rempel Date: Mon, 13 Jul 2015 09:54:42 -0700 Subject: Input: zforce - don't overwrite the stack If we get a corrupted packet with PAYLOAD_LENGTH > FRAME_MAXSIZE, we will silently overwrite the stack. Cc: stable@vger.kernel.org Signed-off-by: Oleksij Rempel Signed-off-by: Dirk Behme Signed-off-by: Dmitry Torokhov diff --git a/drivers/input/touchscreen/zforce_ts.c b/drivers/input/touchscreen/zforce_ts.c index f58a196..80285c7 100644 --- a/drivers/input/touchscreen/zforce_ts.c +++ b/drivers/input/touchscreen/zforce_ts.c @@ -429,7 +429,7 @@ static int zforce_read_packet(struct zforce_ts *ts, u8 *buf) goto unlock; } - if (buf[PAYLOAD_LENGTH] == 0) { + if (buf[PAYLOAD_LENGTH] == 0 || buf[PAYLOAD_LENGTH] > FRAME_MAXSIZE) { dev_err(&client->dev, "invalid payload length: %d\n", buf[PAYLOAD_LENGTH]); ret = -EIO; -- cgit v0.10.2 From d98399e688f8c4f42f4270c2dfe1293f69247c5b Mon Sep 17 00:00:00 2001 From: Peter Hutterer Date: Thu, 16 Jul 2015 10:38:15 -0700 Subject: Input: elantech - force resolution of 31 u/mm All Elantech touchpads pre-v4 with dynamic resolution queries have a fixed resolution of 800dpi -> 31.49 units/mm. Set this statically, so userspace does not have to guess. Signed-off-by: Peter Hutterer Reviewed-by: Hans de Goede Reviewed-by: David Herrmann Signed-off-by: Dmitry Torokhov diff --git a/drivers/input/mouse/elantech.c b/drivers/input/mouse/elantech.c index ce3d400..22b9ca9 100644 --- a/drivers/input/mouse/elantech.c +++ b/drivers/input/mouse/elantech.c @@ -1167,7 +1167,7 @@ static int elantech_set_input_params(struct psmouse *psmouse) struct input_dev *dev = psmouse->dev; struct elantech_data *etd = psmouse->private; unsigned int x_min = 0, y_min = 0, x_max = 0, y_max = 0, width = 0; - unsigned int x_res = 0, y_res = 0; + unsigned int x_res = 31, y_res = 31; if (elantech_set_range(psmouse, &x_min, &y_min, &x_max, &y_max, &width)) return -1; @@ -1232,8 +1232,6 @@ static int elantech_set_input_params(struct psmouse *psmouse) /* For X to recognize me as touchpad. */ input_set_abs_params(dev, ABS_X, x_min, x_max, 0, 0); input_set_abs_params(dev, ABS_Y, y_min, y_max, 0, 0); - input_abs_set_res(dev, ABS_X, x_res); - input_abs_set_res(dev, ABS_Y, y_res); /* * range of pressure and width is the same as v2, * report ABS_PRESSURE, ABS_TOOL_WIDTH for compatibility. @@ -1246,8 +1244,6 @@ static int elantech_set_input_params(struct psmouse *psmouse) input_mt_init_slots(dev, ETP_MAX_FINGERS, 0); input_set_abs_params(dev, ABS_MT_POSITION_X, x_min, x_max, 0, 0); input_set_abs_params(dev, ABS_MT_POSITION_Y, y_min, y_max, 0, 0); - input_abs_set_res(dev, ABS_MT_POSITION_X, x_res); - input_abs_set_res(dev, ABS_MT_POSITION_Y, y_res); input_set_abs_params(dev, ABS_MT_PRESSURE, ETP_PMIN_V2, ETP_PMAX_V2, 0, 0); /* @@ -1259,6 +1255,13 @@ static int elantech_set_input_params(struct psmouse *psmouse) break; } + input_abs_set_res(dev, ABS_X, x_res); + input_abs_set_res(dev, ABS_Y, y_res); + if (etd->hw_version > 1) { + input_abs_set_res(dev, ABS_MT_POSITION_X, x_res); + input_abs_set_res(dev, ABS_MT_POSITION_Y, y_res); + } + etd->y_max = y_max; etd->width = width; -- cgit v0.10.2 From e2c09ae7a74d94222187edbe8f5cf1fa9364efcd Mon Sep 17 00:00:00 2001 From: Javier Martinez Canillas Date: Wed, 15 Jul 2015 16:10:28 +0200 Subject: regulator: core: Increase refcount for regulator supply's module When a regulator is unregistered with regulator_unregister(), a call to regulator_put() is made for its input supply if there is one. This does a module_put() to decrement the refcount of the module that owns the supply but there isn't a corresponding try_module_get() in set_supply() to make the calls balanced. Signed-off-by: Javier Martinez Canillas Signed-off-by: Mark Brown diff --git a/drivers/regulator/core.c b/drivers/regulator/core.c index c9f7201..934fde4 100644 --- a/drivers/regulator/core.c +++ b/drivers/regulator/core.c @@ -1105,6 +1105,9 @@ static int set_supply(struct regulator_dev *rdev, rdev_info(rdev, "supplied by %s\n", rdev_get_name(supply_rdev)); + if (!try_module_get(supply_rdev->owner)) + return -ENODEV; + rdev->supply = create_regulator(supply_rdev, &rdev->dev, "SUPPLY"); if (rdev->supply == NULL) { err = -ENOMEM; -- cgit v0.10.2 From 36a1f1b6ddc6d1442424e29548e790633ca39c7b Mon Sep 17 00:00:00 2001 From: Javier Martinez Canillas Date: Wed, 15 Jul 2015 16:10:29 +0200 Subject: regulator: core: Fix memory leak in regulator_resolve_supply() The regulator_resolve_supply() function calls set_supply() which in turn calls create_regulator() to allocate a supply regulator. If an error occurs after set_supply() succeeded, the allocated regulator has to be freed before propagating the error code. Signed-off-by: Javier Martinez Canillas Signed-off-by: Mark Brown diff --git a/drivers/regulator/core.c b/drivers/regulator/core.c index 934fde4..80a123e 100644 --- a/drivers/regulator/core.c +++ b/drivers/regulator/core.c @@ -109,6 +109,7 @@ static int _regulator_do_set_voltage(struct regulator_dev *rdev, static struct regulator *create_regulator(struct regulator_dev *rdev, struct device *dev, const char *supply_name); +static void _regulator_put(struct regulator *regulator); static const char *rdev_get_name(struct regulator_dev *rdev) { @@ -1401,8 +1402,11 @@ static int regulator_resolve_supply(struct regulator_dev *rdev) /* Cascade always-on state to supply */ if (_regulator_is_enabled(rdev)) { ret = regulator_enable(rdev->supply); - if (ret < 0) + if (ret < 0) { + if (rdev->supply) + _regulator_put(rdev->supply); return ret; + } } return 0; -- cgit v0.10.2 From e2370f07cf703c0920c99bc24de3d152262738f0 Mon Sep 17 00:00:00 2001 From: Sergei Shtylyov Date: Wed, 15 Jul 2015 00:56:52 +0300 Subject: ravb: do not invalidate cache for RX buffer twice First, dma_sync_single_for_cpu() shouldn't have been called in the first place (it's a streaming DMA API), dma_unmap_single() should have been called instead. Second, dma_unmap_single() call after handing the buffer to napi_gro_receive() makes little sense. Moreover desc->dptr might not be valid at this point. Signed-off-by: Sergei Shtylyov Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/renesas/ravb_main.c b/drivers/net/ethernet/renesas/ravb_main.c index fd97457..0e5fde3 100644 --- a/drivers/net/ethernet/renesas/ravb_main.c +++ b/drivers/net/ethernet/renesas/ravb_main.c @@ -543,10 +543,9 @@ static bool ravb_rx(struct net_device *ndev, int *quota, int q) skb = priv->rx_skb[q][entry]; priv->rx_skb[q][entry] = NULL; - dma_sync_single_for_cpu(&ndev->dev, - le32_to_cpu(desc->dptr), - ALIGN(PKT_BUF_SZ, 16), - DMA_FROM_DEVICE); + dma_unmap_single(&ndev->dev, le32_to_cpu(desc->dptr), + ALIGN(PKT_BUF_SZ, 16), + DMA_FROM_DEVICE); get_ts &= (q == RAVB_NC) ? RAVB_RXTSTAMP_TYPE_V2_L2_EVENT : ~RAVB_RXTSTAMP_TYPE_V2_L2_EVENT; @@ -584,9 +583,6 @@ static bool ravb_rx(struct net_device *ndev, int *quota, int q) if (!skb) break; /* Better luck next round. */ ravb_set_buffer_align(skb); - dma_unmap_single(&ndev->dev, le32_to_cpu(desc->dptr), - ALIGN(PKT_BUF_SZ, 16), - DMA_FROM_DEVICE); dma_addr = dma_map_single(&ndev->dev, skb->data, le16_to_cpu(desc->ds_cc), DMA_FROM_DEVICE); -- cgit v0.10.2 From 5cde07abcbc57b3430b044cb2cf028d97d7a7254 Mon Sep 17 00:00:00 2001 From: Krzysztof Kozlowski Date: Mon, 15 Jun 2015 13:54:29 +0900 Subject: pinctrl: samsung: Remove old unused defines Since 9a2c1c3b91aa ("pinctrl: samsung: Allow grouping multiple pinmux/pinconf nodes") the defines for GPIO group and function names are not used anywhere in the driver. Signed-off-by: Krzysztof Kozlowski Inspired-by: Dan Carpenter Signed-off-by: Linus Walleij diff --git a/drivers/pinctrl/samsung/pinctrl-samsung.c b/drivers/pinctrl/samsung/pinctrl-samsung.c index 3dd5a3b..c760bf4 100644 --- a/drivers/pinctrl/samsung/pinctrl-samsung.c +++ b/drivers/pinctrl/samsung/pinctrl-samsung.c @@ -33,11 +33,6 @@ #include "../core.h" #include "pinctrl-samsung.h" -#define GROUP_SUFFIX "-grp" -#define GSUFFIX_LEN sizeof(GROUP_SUFFIX) -#define FUNCTION_SUFFIX "-mux" -#define FSUFFIX_LEN sizeof(FUNCTION_SUFFIX) - /* list of all possible config options supported */ static struct pin_config { const char *property; -- cgit v0.10.2 From 4479004e6409087d1b4986881dc98c6c15dffb28 Mon Sep 17 00:00:00 2001 From: Tom Hughes Date: Mon, 29 Jun 2015 19:41:49 +0100 Subject: mac80211: clear subdir_stations when removing debugfs If we don't do this, and we then fail to recreate the debugfs directory during a mode change, then we will fail later trying to add stations to this now bogus directory: BUG: unable to handle kernel NULL pointer dereference at 0000006c IP: [] mutex_lock+0x12/0x30 Call Trace: [] start_creating+0x44/0xc0 [] debugfs_create_dir+0x13/0xf0 [] ieee80211_sta_debugfs_add+0x6e/0x490 [mac80211] Cc: stable@kernel.org Signed-off-by: Tom Hughes Signed-off-by: Johannes Berg diff --git a/net/mac80211/debugfs_netdev.c b/net/mac80211/debugfs_netdev.c index 29236e8..c09c013 100644 --- a/net/mac80211/debugfs_netdev.c +++ b/net/mac80211/debugfs_netdev.c @@ -723,6 +723,7 @@ void ieee80211_debugfs_remove_netdev(struct ieee80211_sub_if_data *sdata) debugfs_remove_recursive(sdata->vif.debugfs_dir); sdata->vif.debugfs_dir = NULL; + sdata->debugfs.subdir_stations = NULL; } void ieee80211_debugfs_rename_netdev(struct ieee80211_sub_if_data *sdata) -- cgit v0.10.2 From e9de01907e3d1957591113daa3857c0bf01067ef Mon Sep 17 00:00:00 2001 From: Michal Kazior Date: Thu, 2 Jul 2015 09:59:56 +0200 Subject: mac80211: don't clear all tx flags when requeing When acting as AP and a PS-Poll frame is received associated station is marked as one in a Service Period. This state is kept until Tx status for released frame is reported. While a station is in Service Period PS-Poll frames are ignored. However if PS-Poll was received during A-MPDU teardown it was possible to have the to-be released frame re-queued back to pending queue. In such case the frame was stripped of 2 important flags: (a) IEEE80211_TX_CTL_NO_PS_BUFFER (b) IEEE80211_TX_STATUS_EOSP Stripping of (a) led to the frame that was to be released to be queued back to ps_tx_buf queue. If station remained to use only PS-Poll frames the re-queued frame (and new ones) was never actually transmitted because mac80211 would ignore subsequent PS-Poll frames due to station being in Service Period. There was nothing left to clear the Service Period bit (no xmit -> no tx status -> no SP end), i.e. the AP would have the station stuck in Service Period. Beacon TIM would repeatedly prompt station to poll for frames but it would get none. Once (a) is not stripped (b) becomes important because it's the main condition to clear the Service Period bit of the station when Tx status for the released frame is reported back. This problem was observed with ath9k acting as P2P GO in some testing scenarios but isn't limited to it. AP operation with mac80211 based Tx A-MPDU control combined with clients using PS-Poll frames is subject to this race. Signed-off-by: Michal Kazior Signed-off-by: Johannes Berg diff --git a/net/mac80211/tx.c b/net/mac80211/tx.c index 8410bb3..b823350 100644 --- a/net/mac80211/tx.c +++ b/net/mac80211/tx.c @@ -1117,7 +1117,9 @@ static bool ieee80211_tx_prep_agg(struct ieee80211_tx_data *tx, queued = true; info->control.vif = &tx->sdata->vif; info->flags |= IEEE80211_TX_INTFL_NEED_TXPROCESSING; - info->flags &= ~IEEE80211_TX_TEMPORARY_FLAGS; + info->flags &= ~IEEE80211_TX_TEMPORARY_FLAGS | + IEEE80211_TX_CTL_NO_PS_BUFFER | + IEEE80211_TX_STATUS_EOSP; __skb_queue_tail(&tid_tx->pending, skb); if (skb_queue_len(&tid_tx->pending) > STA_MAX_TX_BUFFER) purge_skb = __skb_dequeue(&tid_tx->pending); -- cgit v0.10.2 From 541b6ed7cee1ec7c8f525f51a0ff097776111aeb Mon Sep 17 00:00:00 2001 From: Chaitanya T K Date: Wed, 10 Jun 2015 19:12:31 +0530 Subject: mac80211: wowlan: enable powersave if suspend while ps-polling If for any reason we're in the middle of PS-polling or awake after TX due to dynamic powersave while going to suspend, go back to save power. This might cause a response frame to get lost, but since we can't really wait for it while going to suspend that's still better than not enabling powersave which would cause higher power usage during (and possibly even after) suspend. Note that this really only affects the very few drivers that use the powersave implementation in mac80211. Signed-off-by: Chaitanya T K [rewrite misleading commit log] Signed-off-by: Johannes Berg diff --git a/net/mac80211/pm.c b/net/mac80211/pm.c index 06b60980..b676b9f 100644 --- a/net/mac80211/pm.c +++ b/net/mac80211/pm.c @@ -76,6 +76,22 @@ int __ieee80211_suspend(struct ieee80211_hw *hw, struct cfg80211_wowlan *wowlan) if (sdata->vif.type != NL80211_IFTYPE_STATION) continue; ieee80211_mgd_quiesce(sdata); + /* If suspended during TX in progress, and wowlan + * is enabled (connection will be active) there + * can be a race where the driver is put out + * of power-save due to TX and during suspend + * dynamic_ps_timer is cancelled and TX packet + * is flushed, leaving the driver in ACTIVE even + * after resuming until dynamic_ps_timer puts + * driver back in DOZE. + */ + if (sdata->u.mgd.associated && + sdata->u.mgd.powersave && + !(local->hw.conf.flags & IEEE80211_CONF_PS)) { + local->hw.conf.flags |= IEEE80211_CONF_PS; + ieee80211_hw_config(local, + IEEE80211_CONF_CHANGE_PS); + } } err = drv_suspend(local, wowlan); -- cgit v0.10.2 From d8d9008cfb919db48d95f96b05e81f84b3774318 Mon Sep 17 00:00:00 2001 From: Johannes Berg Date: Wed, 8 Jul 2015 15:41:50 +0300 Subject: mac80211: shut down interfaces before destroying interface list If the hardware is unregistered while interfaces are up, mac80211 will unregister all interfaces, which in turns causes mac80211 to be called again to remove them all from the driver and eventually shut down the hardware. During this shutdown, however, it's currently already unsafe to iterate the list of interfaces atomically, as the list is manipulated in an unsafe manner. This puts an undue burden on the driver - it must stop all its activities before calling ieee80211_unregister_hw(), while in the normal stop path it can do all cleanup in the stop method. If, for example, it's using the iteration during RX for some reason, it would have to stop RX before unregistering to avoid crashes. Fix this problem by closing all interfaces before unregistering them. This will cause the driver stop to have completed before we manipulate the interface list, and after the driver is stopped *and* has called ieee80211_unregister_hw() it really musn't be iterating any more as the memory will be freed as well. Signed-off-by: Johannes Berg diff --git a/net/mac80211/iface.c b/net/mac80211/iface.c index ed1edac..553ac6d 100644 --- a/net/mac80211/iface.c +++ b/net/mac80211/iface.c @@ -1863,10 +1863,6 @@ void ieee80211_sdata_stop(struct ieee80211_sub_if_data *sdata) ieee80211_teardown_sdata(sdata); } -/* - * Remove all interfaces, may only be called at hardware unregistration - * time because it doesn't do RCU-safe list removals. - */ void ieee80211_remove_interfaces(struct ieee80211_local *local) { struct ieee80211_sub_if_data *sdata, *tmp; @@ -1875,14 +1871,21 @@ void ieee80211_remove_interfaces(struct ieee80211_local *local) ASSERT_RTNL(); - /* - * Close all AP_VLAN interfaces first, as otherwise they - * might be closed while the AP interface they belong to - * is closed, causing unregister_netdevice_many() to crash. + /* Before destroying the interfaces, make sure they're all stopped so + * that the hardware is stopped. Otherwise, the driver might still be + * iterating the interfaces during the shutdown, e.g. from a worker + * or from RX processing or similar, and if it does so (using atomic + * iteration) while we're manipulating the list, the iteration will + * crash. + * + * After this, the hardware should be stopped and the driver should + * have stopped all of its activities, so that we can do RCU-unaware + * manipulations of the interface list below. */ - list_for_each_entry(sdata, &local->interfaces, list) - if (sdata->vif.type == NL80211_IFTYPE_AP_VLAN) - dev_close(sdata->dev); + cfg80211_shutdown_all_interfaces(local->hw.wiphy); + + WARN(local->open_count, "%s: open count remains %d\n", + wiphy_name(local->hw.wiphy), local->open_count); mutex_lock(&local->iflist_mtx); list_for_each_entry_safe(sdata, tmp, &local->interfaces, list) { -- cgit v0.10.2 From 042ab5fc7a80b934032fcc673a125feb36645b33 Mon Sep 17 00:00:00 2001 From: Thomas Petazzoni Date: Thu, 9 Jul 2015 15:35:15 +0200 Subject: wireless: regulatory: reduce log level of CRDA related messages With a basic Linux userspace, the messages "Calling CRDA to update world regulatory domain" appears 10 times after boot every second or so, followed by a final "Exceeded CRDA call max attempts. Not calling CRDA". For those of us not having the corresponding userspace parts, having those messages repeatedly displayed at boot time is a bit annoying, so this commit reduces their log level to pr_debug(). Signed-off-by: Thomas Petazzoni Signed-off-by: Johannes Berg diff --git a/net/wireless/reg.c b/net/wireless/reg.c index d359e06..29134c8 100644 --- a/net/wireless/reg.c +++ b/net/wireless/reg.c @@ -544,15 +544,15 @@ static int call_crda(const char *alpha2) reg_regdb_query(alpha2); if (reg_crda_timeouts > REG_MAX_CRDA_TIMEOUTS) { - pr_info("Exceeded CRDA call max attempts. Not calling CRDA\n"); + pr_debug("Exceeded CRDA call max attempts. Not calling CRDA\n"); return -EINVAL; } if (!is_world_regdom((char *) alpha2)) - pr_info("Calling CRDA for country: %c%c\n", + pr_debug("Calling CRDA for country: %c%c\n", alpha2[0], alpha2[1]); else - pr_info("Calling CRDA to update world regulatory domain\n"); + pr_debug("Calling CRDA to update world regulatory domain\n"); return kobject_uevent_env(®_pdev->dev.kobj, KOBJ_CHANGE, env); } -- cgit v0.10.2 From 2ea752cd2ce066c5d8c1807b5310ef329885cecb Mon Sep 17 00:00:00 2001 From: Bob Copeland Date: Tue, 14 Jul 2015 08:31:55 -0400 Subject: mac80211: correct aid location in peering frames According to 802.11-2012 8.5.16.3.2 AID comes directly after the capability bytes in mesh peering confirm frames. The existing code, however, was adding a 2 byte offset to this location, resulting in garbage data going out over the air. Remove the offset to fix it. Signed-off-by: Bob Copeland Signed-off-by: Johannes Berg diff --git a/net/mac80211/mesh_plink.c b/net/mac80211/mesh_plink.c index 5438d13..f17127e 100644 --- a/net/mac80211/mesh_plink.c +++ b/net/mac80211/mesh_plink.c @@ -306,7 +306,7 @@ static int mesh_plink_frame_tx(struct ieee80211_sub_if_data *sdata, if (action == WLAN_SP_MESH_PEERING_CONFIRM) { /* AID */ pos = skb_put(skb, 2); - put_unaligned_le16(plid, pos + 2); + put_unaligned_le16(plid, pos); } if (ieee80211_add_srates_ie(sdata, skb, true, band) || ieee80211_add_ext_srates_ie(sdata, skb, true, band) || -- cgit v0.10.2 From b3e7de873df77c1fa0bc2cfaf3eaff757b80e773 Mon Sep 17 00:00:00 2001 From: Bob Copeland Date: Tue, 14 Jul 2015 08:31:56 -0400 Subject: mac80211: add missing length check for confirm frames Although mesh_rx_plink_frame() already checks that frames have enough bytes for the action code plus another two bytes for capability/reason code, it doesn't take into account that confirm frames also have an additional two-byte aid. As a result, a corrupt frame could cause a subsequent subtraction to wrap around to ill effect. Add another check for this case. Signed-off-by: Bob Copeland Signed-off-by: Johannes Berg diff --git a/net/mac80211/mesh_plink.c b/net/mac80211/mesh_plink.c index f17127e..3b59099 100644 --- a/net/mac80211/mesh_plink.c +++ b/net/mac80211/mesh_plink.c @@ -1122,6 +1122,9 @@ void mesh_rx_plink_frame(struct ieee80211_sub_if_data *sdata, WLAN_SP_MESH_PEERING_CONFIRM) { baseaddr += 4; baselen += 4; + + if (baselen > len) + return; } ieee802_11_parse_elems(baseaddr, len - baselen, true, &elems); mesh_process_plink_frame(sdata, mgmt, &elems); -- cgit v0.10.2 From 923b352f19d9ea971ae2536eab55f5fc9e95fedf Mon Sep 17 00:00:00 2001 From: Arik Nemtsov Date: Wed, 8 Jul 2015 15:41:44 +0300 Subject: cfg80211: use RTNL locked reg_can_beacon for IR-relaxation The RTNL is required to check for IR-relaxation conditions that allow more channels to beacon. Export an RTNL locked version of reg_can_beacon and use it where possible in AP/STA interface type flows, where IR-relaxation may be applicable. Fixes: 06f207fc5418 ("cfg80211: change GO_CONCURRENT to IR_CONCURRENT for STA") Signed-off-by: Arik Nemtsov Signed-off-by: Emmanuel Grumbach Signed-off-by: Johannes Berg diff --git a/include/net/cfg80211.h b/include/net/cfg80211.h index a741678..883fe1e 100644 --- a/include/net/cfg80211.h +++ b/include/net/cfg80211.h @@ -4868,6 +4868,23 @@ bool cfg80211_reg_can_beacon(struct wiphy *wiphy, struct cfg80211_chan_def *chandef, enum nl80211_iftype iftype); +/** + * cfg80211_reg_can_beacon_relax - check if beaconing is allowed with relaxation + * @wiphy: the wiphy + * @chandef: the channel definition + * @iftype: interface type + * + * Return: %true if there is no secondary channel or the secondary channel(s) + * can be used for beaconing (i.e. is not a radar channel etc.). This version + * also checks if IR-relaxation conditions apply, to allow beaconing under + * more permissive conditions. + * + * Requires the RTNL to be held. + */ +bool cfg80211_reg_can_beacon_relax(struct wiphy *wiphy, + struct cfg80211_chan_def *chandef, + enum nl80211_iftype iftype); + /* * cfg80211_ch_switch_notify - update wdev channel and notify userspace * @dev: the device which switched channels diff --git a/net/mac80211/tdls.c b/net/mac80211/tdls.c index ad31b2d..8db6e29 100644 --- a/net/mac80211/tdls.c +++ b/net/mac80211/tdls.c @@ -60,6 +60,7 @@ ieee80211_tdls_add_subband(struct ieee80211_sub_if_data *sdata, struct ieee80211_channel *ch; struct cfg80211_chan_def chandef; int i, subband_start; + struct wiphy *wiphy = sdata->local->hw.wiphy; for (i = start; i <= end; i += spacing) { if (!ch_cnt) @@ -70,9 +71,8 @@ ieee80211_tdls_add_subband(struct ieee80211_sub_if_data *sdata, /* we will be active on the channel */ cfg80211_chandef_create(&chandef, ch, NL80211_CHAN_NO_HT); - if (cfg80211_reg_can_beacon(sdata->local->hw.wiphy, - &chandef, - sdata->wdev.iftype)) { + if (cfg80211_reg_can_beacon_relax(wiphy, &chandef, + sdata->wdev.iftype)) { ch_cnt++; /* * check if the next channel is also part of diff --git a/net/wireless/chan.c b/net/wireless/chan.c index 915b328..59cabc9 100644 --- a/net/wireless/chan.c +++ b/net/wireless/chan.c @@ -797,23 +797,18 @@ static bool cfg80211_ir_permissive_chan(struct wiphy *wiphy, return false; } -bool cfg80211_reg_can_beacon(struct wiphy *wiphy, - struct cfg80211_chan_def *chandef, - enum nl80211_iftype iftype) +static bool _cfg80211_reg_can_beacon(struct wiphy *wiphy, + struct cfg80211_chan_def *chandef, + enum nl80211_iftype iftype, + bool check_no_ir) { bool res; u32 prohibited_flags = IEEE80211_CHAN_DISABLED | IEEE80211_CHAN_RADAR; - trace_cfg80211_reg_can_beacon(wiphy, chandef, iftype); + trace_cfg80211_reg_can_beacon(wiphy, chandef, iftype, check_no_ir); - /* - * Under certain conditions suggested by some regulatory bodies a - * GO/STA can IR on channels marked with IEEE80211_NO_IR. Set this flag - * only if such relaxations are not enabled and the conditions are not - * met. - */ - if (!cfg80211_ir_permissive_chan(wiphy, iftype, chandef->chan)) + if (check_no_ir) prohibited_flags |= IEEE80211_CHAN_NO_IR; if (cfg80211_chandef_dfs_required(wiphy, chandef, iftype) > 0 && @@ -827,8 +822,36 @@ bool cfg80211_reg_can_beacon(struct wiphy *wiphy, trace_cfg80211_return_bool(res); return res; } + +bool cfg80211_reg_can_beacon(struct wiphy *wiphy, + struct cfg80211_chan_def *chandef, + enum nl80211_iftype iftype) +{ + return _cfg80211_reg_can_beacon(wiphy, chandef, iftype, true); +} EXPORT_SYMBOL(cfg80211_reg_can_beacon); +bool cfg80211_reg_can_beacon_relax(struct wiphy *wiphy, + struct cfg80211_chan_def *chandef, + enum nl80211_iftype iftype) +{ + bool check_no_ir; + + ASSERT_RTNL(); + + /* + * Under certain conditions suggested by some regulatory bodies a + * GO/STA can IR on channels marked with IEEE80211_NO_IR. Set this flag + * only if such relaxations are not enabled and the conditions are not + * met. + */ + check_no_ir = !cfg80211_ir_permissive_chan(wiphy, iftype, + chandef->chan); + + return _cfg80211_reg_can_beacon(wiphy, chandef, iftype, check_no_ir); +} +EXPORT_SYMBOL(cfg80211_reg_can_beacon_relax); + int cfg80211_set_monitor_channel(struct cfg80211_registered_device *rdev, struct cfg80211_chan_def *chandef) { diff --git a/net/wireless/nl80211.c b/net/wireless/nl80211.c index c264eff..76b4157 100644 --- a/net/wireless/nl80211.c +++ b/net/wireless/nl80211.c @@ -2003,7 +2003,8 @@ static int __nl80211_set_channel(struct cfg80211_registered_device *rdev, switch (iftype) { case NL80211_IFTYPE_AP: case NL80211_IFTYPE_P2P_GO: - if (!cfg80211_reg_can_beacon(&rdev->wiphy, &chandef, iftype)) { + if (!cfg80211_reg_can_beacon_relax(&rdev->wiphy, &chandef, + iftype)) { result = -EINVAL; break; } @@ -3403,8 +3404,8 @@ static int nl80211_start_ap(struct sk_buff *skb, struct genl_info *info) } else if (!nl80211_get_ap_channel(rdev, ¶ms)) return -EINVAL; - if (!cfg80211_reg_can_beacon(&rdev->wiphy, ¶ms.chandef, - wdev->iftype)) + if (!cfg80211_reg_can_beacon_relax(&rdev->wiphy, ¶ms.chandef, + wdev->iftype)) return -EINVAL; if (info->attrs[NL80211_ATTR_ACL_POLICY]) { @@ -6492,8 +6493,8 @@ skip_beacons: if (err) return err; - if (!cfg80211_reg_can_beacon(&rdev->wiphy, ¶ms.chandef, - wdev->iftype)) + if (!cfg80211_reg_can_beacon_relax(&rdev->wiphy, ¶ms.chandef, + wdev->iftype)) return -EINVAL; err = cfg80211_chandef_dfs_required(wdev->wiphy, @@ -10170,7 +10171,8 @@ static int nl80211_tdls_channel_switch(struct sk_buff *skb, return -EINVAL; /* we will be active on the TDLS link */ - if (!cfg80211_reg_can_beacon(&rdev->wiphy, &chandef, wdev->iftype)) + if (!cfg80211_reg_can_beacon_relax(&rdev->wiphy, &chandef, + wdev->iftype)) return -EINVAL; /* don't allow switching to DFS channels */ diff --git a/net/wireless/reg.c b/net/wireless/reg.c index 29134c8..aa2d754 100644 --- a/net/wireless/reg.c +++ b/net/wireless/reg.c @@ -1589,7 +1589,7 @@ static bool reg_wdev_chan_valid(struct wiphy *wiphy, struct wireless_dev *wdev) case NL80211_IFTYPE_AP: case NL80211_IFTYPE_P2P_GO: case NL80211_IFTYPE_ADHOC: - return cfg80211_reg_can_beacon(wiphy, &chandef, iftype); + return cfg80211_reg_can_beacon_relax(wiphy, &chandef, iftype); case NL80211_IFTYPE_STATION: case NL80211_IFTYPE_P2P_CLIENT: return cfg80211_chandef_usable(wiphy, &chandef, diff --git a/net/wireless/trace.h b/net/wireless/trace.h index af3617c..a808279 100644 --- a/net/wireless/trace.h +++ b/net/wireless/trace.h @@ -2358,20 +2358,23 @@ TRACE_EVENT(cfg80211_cqm_rssi_notify, TRACE_EVENT(cfg80211_reg_can_beacon, TP_PROTO(struct wiphy *wiphy, struct cfg80211_chan_def *chandef, - enum nl80211_iftype iftype), - TP_ARGS(wiphy, chandef, iftype), + enum nl80211_iftype iftype, bool check_no_ir), + TP_ARGS(wiphy, chandef, iftype, check_no_ir), TP_STRUCT__entry( WIPHY_ENTRY CHAN_DEF_ENTRY __field(enum nl80211_iftype, iftype) + __field(bool, check_no_ir) ), TP_fast_assign( WIPHY_ASSIGN; CHAN_DEF_ASSIGN(chandef); __entry->iftype = iftype; + __entry->check_no_ir = check_no_ir; ), - TP_printk(WIPHY_PR_FMT ", " CHAN_DEF_PR_FMT ", iftype=%d", - WIPHY_PR_ARG, CHAN_DEF_PR_ARG, __entry->iftype) + TP_printk(WIPHY_PR_FMT ", " CHAN_DEF_PR_FMT ", iftype=%d check_no_ir=%s", + WIPHY_PR_ARG, CHAN_DEF_PR_ARG, __entry->iftype, + BOOL_TO_STR(__entry->check_no_ir)) ); TRACE_EVENT(cfg80211_chandef_dfs_required, -- cgit v0.10.2 From d6726c8145290bef950ae2538ea6ae1d96a1944b Mon Sep 17 00:00:00 2001 From: "Steven Rostedt (Red Hat)" Date: Fri, 17 Jul 2015 14:03:26 -0400 Subject: tracing: Fix sample output of dynamic arrays He Kuang noticed that the trace event samples for arrays was broken: "The output result of trace_foo_bar event in traceevent samples is wrong. This problem can be reproduced as following: (Build kernel with SAMPLE_TRACE_EVENTS=m) $ insmod trace-events-sample.ko $ echo 1 > /sys/kernel/debug/tracing/events/sample-trace/foo_bar/enable $ cat /sys/kernel/debug/tracing/trace event-sample-980 [000] .... 43.649559: foo_bar: foo hello 21 0x15 BIT1|BIT3|0x10 {0x1,0x6f6f6e53,0xff007970,0xffffffff} Snoopy ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ The array length is not right, should be {0x1}. (ffffffff,ffffffff) event-sample-980 [000] .... 44.653827: foo_bar: foo hello 22 0x16 BIT2|BIT3|0x10 {0x1,0x2,0x646e6147,0x666c61,0xffffffff,0xffffffff,0x750aeffe,0x7} ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ The array length is not right, should be {0x1,0x2}. Gandalf (ffffffff,ffffffff)" This was caused by an update to have __print_array()'s second parameter be the count of items in the array and not the size of the array. As there is already users of __print_array(), it can not change. But the sample code can and we can also improve on the documentation about __print_array() and __get_dynamic_array_len(). Link: http://lkml.kernel.org/r/1436839171-31527-2-git-send-email-hekuang@huawei.com Fixes: ac01ce1410fc2 ("tracing: Make ftrace_print_array_seq compute buf_len") Reported-by: He Kuang Signed-off-by: Steven Rostedt diff --git a/samples/trace_events/trace-events-sample.h b/samples/trace_events/trace-events-sample.h index 8965d1b..125d640 100644 --- a/samples/trace_events/trace-events-sample.h +++ b/samples/trace_events/trace-events-sample.h @@ -168,7 +168,10 @@ * * For __dynamic_array(int, foo, bar) use __get_dynamic_array(foo) * Use __get_dynamic_array_len(foo) to get the length of the array - * saved. + * saved. Note, __get_dynamic_array_len() returns the total allocated + * length of the dynamic array; __print_array() expects the second + * parameter to be the number of elements. To get that, the array length + * needs to be divided by the element size. * * For __string(foo, bar) use __get_str(foo) * @@ -288,7 +291,7 @@ TRACE_EVENT(foo_bar, * This prints out the array that is defined by __array in a nice format. */ __print_array(__get_dynamic_array(list), - __get_dynamic_array_len(list), + __get_dynamic_array_len(list) / sizeof(int), sizeof(int)), __get_str(str), __get_bitmask(cpus)) ); -- cgit v0.10.2 From 88cc7b4eee1e7b9bca1a64dae5adaa044cf72312 Mon Sep 17 00:00:00 2001 From: Srinivas Pandruvada Date: Mon, 1 Jun 2015 16:36:27 -0700 Subject: hid-sensor: Fix suspend/resume delay By default all the sensors are runtime suspended state (lowest power state). During Linux suspend process, all the run time suspended devices are resumed and then suspended. This caused all sensors to power up and introduced delay in suspend time, when we introduced runtime PM for HID sensors. The opposite process happens during resume process. To fix this, we do powerup process of the sensors only when the request is issued from user (raw or tiggerred). In this way when runtime, resume calls for powerup it will simply return as this will not match user requested state. Note this is a regression fix as the increase in suspend / resume times can be substantial (report of 8 seconds on Len's laptop!) Signed-off-by: Srinivas Pandruvada Tested-by: Len Brown Cc: Signed-off-by: Jonathan Cameron diff --git a/drivers/iio/common/hid-sensors/hid-sensor-trigger.c b/drivers/iio/common/hid-sensors/hid-sensor-trigger.c index 610fc98..5955110 100644 --- a/drivers/iio/common/hid-sensors/hid-sensor-trigger.c +++ b/drivers/iio/common/hid-sensors/hid-sensor-trigger.c @@ -36,6 +36,8 @@ static int _hid_sensor_power_state(struct hid_sensor_common *st, bool state) s32 poll_value = 0; if (state) { + if (!atomic_read(&st->user_requested_state)) + return 0; if (sensor_hub_device_open(st->hsdev)) return -EIO; @@ -52,8 +54,12 @@ static int _hid_sensor_power_state(struct hid_sensor_common *st, bool state) poll_value = hid_sensor_read_poll_value(st); } else { - if (!atomic_dec_and_test(&st->data_ready)) + int val; + + val = atomic_dec_if_positive(&st->data_ready); + if (val < 0) return 0; + sensor_hub_device_close(st->hsdev); state_val = hid_sensor_get_usage_index(st->hsdev, st->power_state.report_id, @@ -92,9 +98,11 @@ EXPORT_SYMBOL(hid_sensor_power_state); int hid_sensor_power_state(struct hid_sensor_common *st, bool state) { + #ifdef CONFIG_PM int ret; + atomic_set(&st->user_requested_state, state); if (state) ret = pm_runtime_get_sync(&st->pdev->dev); else { @@ -109,6 +117,7 @@ int hid_sensor_power_state(struct hid_sensor_common *st, bool state) return 0; #else + atomic_set(&st->user_requested_state, state); return _hid_sensor_power_state(st, state); #endif } diff --git a/include/linux/hid-sensor-hub.h b/include/linux/hid-sensor-hub.h index 0042bf3..c02b5ce 100644 --- a/include/linux/hid-sensor-hub.h +++ b/include/linux/hid-sensor-hub.h @@ -230,6 +230,7 @@ struct hid_sensor_common { struct platform_device *pdev; unsigned usage_id; atomic_t data_ready; + atomic_t user_requested_state; struct iio_trigger *trigger; struct hid_sensor_hub_attribute_info poll; struct hid_sensor_hub_attribute_info report_state; -- cgit v0.10.2 From c11e28f959a6e700d19a952e078cb368c2e781a4 Mon Sep 17 00:00:00 2001 From: Heiko Stuebner Date: Mon, 25 May 2015 22:36:58 +0200 Subject: iio: adc: rockchip_saradc: add missing MODULE_* data The module-data is currently missing. This includes the license-information which makes the driver taint the kernel and miss symbols when compiled as module. Fixes: 44d6f2ef94f9 ("iio: adc: add driver for Rockchip saradc") Signed-off-by: Heiko Stuebner Cc: Signed-off-by: Jonathan Cameron diff --git a/drivers/iio/adc/rockchip_saradc.c b/drivers/iio/adc/rockchip_saradc.c index 8d4e019..9c311c1 100644 --- a/drivers/iio/adc/rockchip_saradc.c +++ b/drivers/iio/adc/rockchip_saradc.c @@ -349,3 +349,7 @@ static struct platform_driver rockchip_saradc_driver = { }; module_platform_driver(rockchip_saradc_driver); + +MODULE_AUTHOR("Heiko Stuebner "); +MODULE_DESCRIPTION("Rockchip SARADC driver"); +MODULE_LICENSE("GPL v2"); -- cgit v0.10.2 From b220da654d268c79f420fbfac6a35eb420b9cc4b Mon Sep 17 00:00:00 2001 From: Fabio Estevam Date: Sun, 24 May 2015 17:39:04 -0300 Subject: iio: twl4030-madc: Pass the IRQF_ONESHOT flag Since commit 1c6c69525b40 ("genirq: Reject bogus threaded irq requests") threaded IRQs without a primary handler need to be requested with IRQF_ONESHOT, otherwise the request will fail. So pass the IRQF_ONESHOT flag in this case. The semantic patch that makes this change is available in scripts/coccinelle/misc/irqf_oneshot.cocci. Signed-off-by: Fabio Estevam Cc: Signed-off-by: Jonathan Cameron diff --git a/drivers/iio/adc/twl4030-madc.c b/drivers/iio/adc/twl4030-madc.c index 06f4792..ebe415f 100644 --- a/drivers/iio/adc/twl4030-madc.c +++ b/drivers/iio/adc/twl4030-madc.c @@ -833,7 +833,8 @@ static int twl4030_madc_probe(struct platform_device *pdev) irq = platform_get_irq(pdev, 0); ret = devm_request_threaded_irq(&pdev->dev, irq, NULL, twl4030_madc_threaded_irq_handler, - IRQF_TRIGGER_RISING, "twl4030_madc", madc); + IRQF_TRIGGER_RISING | IRQF_ONESHOT, + "twl4030_madc", madc); if (ret) { dev_err(&pdev->dev, "could not request irq\n"); goto err_i2c; -- cgit v0.10.2 From 2acbe15f8ae539809ef8169a65a8d859cf932530 Mon Sep 17 00:00:00 2001 From: Adriana Reus Date: Fri, 12 Jun 2015 18:10:23 +0300 Subject: iio: inv-mpu: Specify the expected format/precision for write channels The gyroscope needs IIO_VAL_INT_PLUS_NANO for the scale channel and unless specified write returns MICRO by default. This needs to be properly specified so that write operations into scale have the expected behaviour. Signed-off-by: Adriana Reus Cc: Signed-off-by: Jonathan Cameron diff --git a/drivers/iio/imu/inv_mpu6050/inv_mpu_core.c b/drivers/iio/imu/inv_mpu6050/inv_mpu_core.c index 17d4bb1..65ce868 100644 --- a/drivers/iio/imu/inv_mpu6050/inv_mpu_core.c +++ b/drivers/iio/imu/inv_mpu6050/inv_mpu_core.c @@ -431,6 +431,23 @@ static int inv_mpu6050_write_gyro_scale(struct inv_mpu6050_state *st, int val) return -EINVAL; } +static int inv_write_raw_get_fmt(struct iio_dev *indio_dev, + struct iio_chan_spec const *chan, long mask) +{ + switch (mask) { + case IIO_CHAN_INFO_SCALE: + switch (chan->type) { + case IIO_ANGL_VEL: + return IIO_VAL_INT_PLUS_NANO; + default: + return IIO_VAL_INT_PLUS_MICRO; + } + default: + return IIO_VAL_INT_PLUS_MICRO; + } + + return -EINVAL; +} static int inv_mpu6050_write_accel_scale(struct inv_mpu6050_state *st, int val) { int result, i; @@ -696,6 +713,7 @@ static const struct iio_info mpu_info = { .driver_module = THIS_MODULE, .read_raw = &inv_mpu6050_read_raw, .write_raw = &inv_mpu6050_write_raw, + .write_raw_get_fmt = &inv_write_raw_get_fmt, .attrs = &inv_attribute_group, .validate_trigger = inv_mpu6050_validate_trigger, }; -- cgit v0.10.2 From 63fc444967591c3849438aa0f21d5de96a928311 Mon Sep 17 00:00:00 2001 From: Daniel Baluta Date: Thu, 11 Jun 2015 18:49:33 +0300 Subject: iio: ABI: Clarify proximity output value Current description for proximity measurement is ambiguous. While the first part says that proximity is measured by observing reflectivity, the second part incorrectly infers that reported values should behave like a distance. This is because of AS3935 lightning sensor which uses the proximity API, while not being a true proximity sensor. Note this is marked for stable as it accompanies a fix in ABI usage to the sx9500 driver which would otherwise appear to be correct. Fixes: 614e8842ddf ("iio: ABI: add clarification for proximity") Signed-off-by: Daniel Baluta Cc: Signed-off-by: Jonathan Cameron diff --git a/Documentation/ABI/testing/sysfs-bus-iio b/Documentation/ABI/testing/sysfs-bus-iio index bbed111..70c9b1a 100644 --- a/Documentation/ABI/testing/sysfs-bus-iio +++ b/Documentation/ABI/testing/sysfs-bus-iio @@ -1234,10 +1234,8 @@ Description: object is near the sensor, usually be observing reflectivity of infrared or ultrasound emitted. Often these sensors are unit less and as such conversion - to SI units is not possible. Where it is, the units should - be meters. If such a conversion is not possible, the reported - values should behave in the same way as a distance, i.e. lower - values indicate something is closer to the sensor. + to SI units is not possible. Higher proximity measurements + indicate closer objects, and vice versa. What: /sys/.../iio:deviceX/in_illuminance_input What: /sys/.../iio:deviceX/in_illuminance_raw -- cgit v0.10.2 From 86d24f04f9a0a7bad8e1719e5aa464eb5779e678 Mon Sep 17 00:00:00 2001 From: Daniel Baluta Date: Thu, 11 Jun 2015 18:49:34 +0300 Subject: iio: proximity: sx9500: Fix proximity value Because of the ABI confusion proximity value exposed by SX9500 was inverted. Signed-off-by: Daniel Baluta Reviewed-by: Vlad Dogaru Cc: Signed-off-by: Jonathan Cameron diff --git a/drivers/iio/proximity/sx9500.c b/drivers/iio/proximity/sx9500.c index 2042e37..01494cd 100644 --- a/drivers/iio/proximity/sx9500.c +++ b/drivers/iio/proximity/sx9500.c @@ -281,7 +281,7 @@ static int sx9500_read_prox_data(struct sx9500_data *data, if (ret < 0) return ret; - *val = 32767 - (s16)be16_to_cpu(regval); + *val = be16_to_cpu(regval); return IIO_VAL_INT; } -- cgit v0.10.2 From 09f4dcc9443c4731b6669066bd1bc4312ef17bcc Mon Sep 17 00:00:00 2001 From: JM Friedt Date: Fri, 19 Jun 2015 14:48:06 +0200 Subject: iio: DAC: ad5624r_spi: fix bit shift of output data value The value sent on the SPI bus is shifted by an erroneous number of bits. The shift value was already computed in the iio_chan_spec structure and hence subtracting this argument to 16 yields an erroneous data position in the SPI stream. Signed-off-by: JM Friedt Acked-by: Lars-Peter Clausen Cc: Signed-off-by: Jonathan Cameron diff --git a/drivers/iio/dac/ad5624r_spi.c b/drivers/iio/dac/ad5624r_spi.c index 61bb9d4..e98428d 100644 --- a/drivers/iio/dac/ad5624r_spi.c +++ b/drivers/iio/dac/ad5624r_spi.c @@ -22,7 +22,7 @@ #include "ad5624r.h" static int ad5624r_spi_write(struct spi_device *spi, - u8 cmd, u8 addr, u16 val, u8 len) + u8 cmd, u8 addr, u16 val, u8 shift) { u32 data; u8 msg[3]; @@ -35,7 +35,7 @@ static int ad5624r_spi_write(struct spi_device *spi, * 14-, 12-bit input code followed by 0, 2, or 4 don't care bits, * for the AD5664R, AD5644R, and AD5624R, respectively. */ - data = (0 << 22) | (cmd << 19) | (addr << 16) | (val << (16 - len)); + data = (0 << 22) | (cmd << 19) | (addr << 16) | (val << shift); msg[0] = data >> 16; msg[1] = data >> 8; msg[2] = data; -- cgit v0.10.2 From 815983e9afcb21552e1b9d77e3755d394cbf5249 Mon Sep 17 00:00:00 2001 From: Jan Leupold Date: Wed, 17 Jun 2015 18:21:36 +0200 Subject: iio: adc: at91_adc: allow to use full range of startup time The DT-Property "atmel,adc-startup-time" is stored in an u8 for a microsecond value. When trying to increase the value of STARTUP in Register AT91_ADC_MR some higher values can't be reached. Change the type in function parameter and private structure field from u8 to u32. Signed-off-by: Jan Leupold [nicolas.ferre@atmel.com: change commit message, increase u16 to u32 for startup time] Signed-off-by: Nicolas Ferre Acked-by: Alexandre Belloni Cc: Signed-off-by: Jonathan Cameron diff --git a/drivers/iio/adc/at91_adc.c b/drivers/iio/adc/at91_adc.c index 8a0eb4a..7b40925 100644 --- a/drivers/iio/adc/at91_adc.c +++ b/drivers/iio/adc/at91_adc.c @@ -182,7 +182,7 @@ struct at91_adc_caps { u8 ts_pen_detect_sensitivity; /* startup time calculate function */ - u32 (*calc_startup_ticks)(u8 startup_time, u32 adc_clk_khz); + u32 (*calc_startup_ticks)(u32 startup_time, u32 adc_clk_khz); u8 num_channels; struct at91_adc_reg_desc registers; @@ -201,7 +201,7 @@ struct at91_adc_state { u8 num_channels; void __iomem *reg_base; struct at91_adc_reg_desc *registers; - u8 startup_time; + u32 startup_time; u8 sample_hold_time; bool sleep_mode; struct iio_trigger **trig; @@ -779,7 +779,7 @@ ret: return ret; } -static u32 calc_startup_ticks_9260(u8 startup_time, u32 adc_clk_khz) +static u32 calc_startup_ticks_9260(u32 startup_time, u32 adc_clk_khz) { /* * Number of ticks needed to cover the startup time of the ADC @@ -790,7 +790,7 @@ static u32 calc_startup_ticks_9260(u8 startup_time, u32 adc_clk_khz) return round_up((startup_time * adc_clk_khz / 1000) - 1, 8) / 8; } -static u32 calc_startup_ticks_9x5(u8 startup_time, u32 adc_clk_khz) +static u32 calc_startup_ticks_9x5(u32 startup_time, u32 adc_clk_khz) { /* * For sama5d3x and at91sam9x5, the formula changes to: -- cgit v0.10.2 From bd7bd0cc3a60e082b17a4f64ce76831aa1c6dffe Mon Sep 17 00:00:00 2001 From: Hartmut Knaack Date: Thu, 18 Jun 2015 00:31:59 +0200 Subject: iio:light:cm3323: clear bitmask before set When setting the bits for integration time, the appropriate bitmask needs to be cleared first. Signed-off-by: Hartmut Knaack Cc: Signed-off-by: Jonathan Cameron diff --git a/drivers/iio/light/cm3323.c b/drivers/iio/light/cm3323.c index 869033e..a1d4905 100644 --- a/drivers/iio/light/cm3323.c +++ b/drivers/iio/light/cm3323.c @@ -123,7 +123,7 @@ static int cm3323_set_it_bits(struct cm3323_data *data, int val, int val2) for (i = 0; i < ARRAY_SIZE(cm3323_int_time); i++) { if (val == cm3323_int_time[i].val && val2 == cm3323_int_time[i].val2) { - reg_conf = data->reg_conf; + reg_conf = data->reg_conf & ~CM3323_CONF_IT_MASK; reg_conf |= i << CM3323_CONF_IT_SHIFT; ret = i2c_smbus_write_word_data(data->client, -- cgit v0.10.2 From c13c9da6d86fa9cb6449deb6c71b18ffcf32ae53 Mon Sep 17 00:00:00 2001 From: Hartmut Knaack Date: Mon, 15 Jun 2015 23:48:24 +0200 Subject: iio:accel:bmc150-accel: fix counting direction In bmc150_accel_unregister_triggers() triggers should be unregistered in reverse order of registration. Trigger registration starts with number 0, counting up. In consequence, trigger number needs to be count down here. Signed-off-by: Hartmut Knaack Reviewed-by: Octavian Purdila Cc: Signed-off-by: Jonathan Cameron diff --git a/drivers/iio/accel/bmc150-accel.c b/drivers/iio/accel/bmc150-accel.c index 4e70f51..cc5a357 100644 --- a/drivers/iio/accel/bmc150-accel.c +++ b/drivers/iio/accel/bmc150-accel.c @@ -1464,7 +1464,7 @@ static void bmc150_accel_unregister_triggers(struct bmc150_accel_data *data, { int i; - for (i = from; i >= 0; i++) { + for (i = from; i >= 0; i--) { if (data->triggers[i].indio_trig) { iio_trigger_unregister(data->triggers[i].indio_trig); data->triggers[i].indio_trig = NULL; -- cgit v0.10.2 From 40f34b0c75c2a633ba647afec41947d8dd3bc920 Mon Sep 17 00:00:00 2001 From: Peter Meerwald Date: Sun, 14 Jun 2015 23:09:35 +0200 Subject: iio: light: tcs3414: Fix bug preventing to set integration time the millisecond values in tcs3414_times should be checked against val2, not val, which is always zero. Signed-off-by: Peter Meerwald Reported-by: Stephan Kleisinger Cc: Signed-off-by: Jonathan Cameron diff --git a/drivers/iio/light/tcs3414.c b/drivers/iio/light/tcs3414.c index 71c2bde..f8b1df0 100644 --- a/drivers/iio/light/tcs3414.c +++ b/drivers/iio/light/tcs3414.c @@ -185,7 +185,7 @@ static int tcs3414_write_raw(struct iio_dev *indio_dev, if (val != 0) return -EINVAL; for (i = 0; i < ARRAY_SIZE(tcs3414_times); i++) { - if (val == tcs3414_times[i] * 1000) { + if (val2 == tcs3414_times[i] * 1000) { data->timing &= ~TCS3414_INTEG_MASK; data->timing |= i; return i2c_smbus_write_byte_data( -- cgit v0.10.2 From 5646e856db1e3e2b260c6ba732524e5a886770d5 Mon Sep 17 00:00:00 2001 From: Hartmut Knaack Date: Sun, 21 Jun 2015 12:15:50 +0200 Subject: iio:adc:cc10001_adc: fix Kconfig dependency The Cosmic Circuits 10001 ADC driver depends on HAS_IOMEM, HAVE_CLK and REGULATOR together, not just any of these. Signed-off-by: Hartmut Knaack Cc: Signed-off-by: Jonathan Cameron diff --git a/drivers/iio/adc/Kconfig b/drivers/iio/adc/Kconfig index 7c55658..eb0cd89 100644 --- a/drivers/iio/adc/Kconfig +++ b/drivers/iio/adc/Kconfig @@ -153,8 +153,7 @@ config DA9150_GPADC config CC10001_ADC tristate "Cosmic Circuits 10001 ADC driver" - depends on HAVE_CLK || REGULATOR - depends on HAS_IOMEM + depends on HAS_IOMEM && HAVE_CLK && REGULATOR select IIO_BUFFER select IIO_TRIGGERED_BUFFER help -- cgit v0.10.2 From 0f16fc8bb38156c842f6976ddbdf4f266e893a1b Mon Sep 17 00:00:00 2001 From: Tiberiu Breana Date: Wed, 1 Jul 2015 15:32:00 +0300 Subject: iio: light: STK3310: un-invert proximity values In accordance with the recent proximity ABI changes, STK3310's proximity readings should be un-inversed in order to return low values for far-away objects and high values for close ones. As consequences of this change, iio event directions have been switched and maximum proximity sensor reference values have also been adjusted in accordance with the real readings. Signed-off-by: Tiberiu Breana Acked-by: Hartmut Knaack Signed-off-by: Jonathan Cameron diff --git a/drivers/iio/light/stk3310.c b/drivers/iio/light/stk3310.c index fee4297..c1a2182 100644 --- a/drivers/iio/light/stk3310.c +++ b/drivers/iio/light/stk3310.c @@ -43,7 +43,6 @@ #define STK3311_CHIP_ID_VAL 0x1D #define STK3310_PSINT_EN 0x01 #define STK3310_PS_MAX_VAL 0xFFFF -#define STK3310_THRESH_MAX 0xFFFF #define STK3310_DRIVER_NAME "stk3310" #define STK3310_REGMAP_NAME "stk3310_regmap" @@ -84,15 +83,13 @@ static const struct reg_field stk3310_reg_field_flag_psint = REG_FIELD(STK3310_REG_FLAG, 4, 4); static const struct reg_field stk3310_reg_field_flag_nf = REG_FIELD(STK3310_REG_FLAG, 0, 0); -/* - * Maximum PS values with regard to scale. Used to export the 'inverse' - * PS value (high values for far objects, low values for near objects). - */ + +/* Estimate maximum proximity values with regard to measurement scale. */ static const int stk3310_ps_max[4] = { - STK3310_PS_MAX_VAL / 64, - STK3310_PS_MAX_VAL / 16, - STK3310_PS_MAX_VAL / 4, - STK3310_PS_MAX_VAL, + STK3310_PS_MAX_VAL / 640, + STK3310_PS_MAX_VAL / 160, + STK3310_PS_MAX_VAL / 40, + STK3310_PS_MAX_VAL / 10 }; static const int stk3310_scale_table[][2] = { @@ -128,14 +125,14 @@ static const struct iio_event_spec stk3310_events[] = { /* Proximity event */ { .type = IIO_EV_TYPE_THRESH, - .dir = IIO_EV_DIR_FALLING, + .dir = IIO_EV_DIR_RISING, .mask_separate = BIT(IIO_EV_INFO_VALUE) | BIT(IIO_EV_INFO_ENABLE), }, /* Out-of-proximity event */ { .type = IIO_EV_TYPE_THRESH, - .dir = IIO_EV_DIR_RISING, + .dir = IIO_EV_DIR_FALLING, .mask_separate = BIT(IIO_EV_INFO_VALUE) | BIT(IIO_EV_INFO_ENABLE), }, @@ -205,23 +202,16 @@ static int stk3310_read_event(struct iio_dev *indio_dev, u8 reg; u16 buf; int ret; - unsigned int index; struct stk3310_data *data = iio_priv(indio_dev); if (info != IIO_EV_INFO_VALUE) return -EINVAL; - /* - * Only proximity interrupts are implemented at the moment. - * Since we're inverting proximity values, the sensor's 'high' - * threshold will become our 'low' threshold, associated with - * 'near' events. Similarly, the sensor's 'low' threshold will - * be our 'high' threshold, associated with 'far' events. - */ + /* Only proximity interrupts are implemented at the moment. */ if (dir == IIO_EV_DIR_RISING) - reg = STK3310_REG_THDL_PS; - else if (dir == IIO_EV_DIR_FALLING) reg = STK3310_REG_THDH_PS; + else if (dir == IIO_EV_DIR_FALLING) + reg = STK3310_REG_THDL_PS; else return -EINVAL; @@ -232,8 +222,7 @@ static int stk3310_read_event(struct iio_dev *indio_dev, dev_err(&data->client->dev, "register read failed\n"); return ret; } - regmap_field_read(data->reg_ps_gain, &index); - *val = swab16(stk3310_ps_max[index] - buf); + *val = swab16(buf); return IIO_VAL_INT; } @@ -257,13 +246,13 @@ static int stk3310_write_event(struct iio_dev *indio_dev, return -EINVAL; if (dir == IIO_EV_DIR_RISING) - reg = STK3310_REG_THDL_PS; - else if (dir == IIO_EV_DIR_FALLING) reg = STK3310_REG_THDH_PS; + else if (dir == IIO_EV_DIR_FALLING) + reg = STK3310_REG_THDL_PS; else return -EINVAL; - buf = swab16(stk3310_ps_max[index] - val); + buf = swab16(val); ret = regmap_bulk_write(data->regmap, reg, &buf, 2); if (ret < 0) dev_err(&client->dev, "failed to set PS threshold!\n"); @@ -334,14 +323,6 @@ static int stk3310_read_raw(struct iio_dev *indio_dev, return ret; } *val = swab16(buf); - if (chan->type == IIO_PROXIMITY) { - /* - * Invert the proximity data so we return low values - * for close objects and high values for far ones. - */ - regmap_field_read(data->reg_ps_gain, &index); - *val = stk3310_ps_max[index] - *val; - } mutex_unlock(&data->lock); return IIO_VAL_INT; case IIO_CHAN_INFO_INT_TIME: @@ -581,8 +562,8 @@ static irqreturn_t stk3310_irq_event_handler(int irq, void *private) } event = IIO_UNMOD_EVENT_CODE(IIO_PROXIMITY, 1, IIO_EV_TYPE_THRESH, - (dir ? IIO_EV_DIR_RISING : - IIO_EV_DIR_FALLING)); + (dir ? IIO_EV_DIR_FALLING : + IIO_EV_DIR_RISING)); iio_push_event(indio_dev, event, data->timestamp); /* Reset the interrupt flag */ -- cgit v0.10.2 From 53c8eccb710fbe10ad5ad1e056486f953c4cba18 Mon Sep 17 00:00:00 2001 From: Hartmut Knaack Date: Sat, 4 Jul 2015 17:56:57 +0200 Subject: iio:light:stk3310: Fix REGMAP_I2C dependency The stk3310 driver makes use of regmap_i2c, so this dependency needs to be reflected in Kconfig. Signed-off-by: Hartmut Knaack Signed-off-by: Jonathan Cameron diff --git a/drivers/iio/light/Kconfig b/drivers/iio/light/Kconfig index e6198b7..df995a4 100644 --- a/drivers/iio/light/Kconfig +++ b/drivers/iio/light/Kconfig @@ -201,6 +201,7 @@ config LTR501 config STK3310 tristate "STK3310 ALS and proximity sensor" depends on I2C + select REGMAP_I2C help Say yes here to get support for the Sensortek STK3310 ambient light and proximity sensor. The STK3311 model is also supported by this -- cgit v0.10.2 From c99389ad3d80eeef0d0980a17e07f41822bda687 Mon Sep 17 00:00:00 2001 From: Teodora Baluta Date: Mon, 29 Jun 2015 15:44:50 +0300 Subject: iio: magnetometer: mmc35240: fix available sampling frequencies Fix the sampling frequencies according to the datasheet (page 8). The datasheet specifies the following available frequencies for continuous mode: 1.5 Hz, 13 Hz, 25 Hz, and 50 Hz. Also fix comments about the ODR to comply with datasheet. Signed-off-by: Teodora Baluta Signed-off-by: Jonathan Cameron diff --git a/drivers/iio/magnetometer/mmc35240.c b/drivers/iio/magnetometer/mmc35240.c index 7a2ea71..d927397 100644 --- a/drivers/iio/magnetometer/mmc35240.c +++ b/drivers/iio/magnetometer/mmc35240.c @@ -84,10 +84,10 @@ #define MMC35240_OTP_START_ADDR 0x1B enum mmc35240_resolution { - MMC35240_16_BITS_SLOW = 0, /* 100 Hz */ - MMC35240_16_BITS_FAST, /* 200 Hz */ - MMC35240_14_BITS, /* 333 Hz */ - MMC35240_12_BITS, /* 666 Hz */ + MMC35240_16_BITS_SLOW = 0, /* 7.92 ms */ + MMC35240_16_BITS_FAST, /* 4.08 ms */ + MMC35240_14_BITS, /* 2.16 ms */ + MMC35240_12_BITS, /* 1.20 ms */ }; enum mmc35240_axis { @@ -100,22 +100,22 @@ static const struct { int sens[3]; /* sensitivity per X, Y, Z axis */ int nfo; /* null field output */ } mmc35240_props_table[] = { - /* 16 bits, 100Hz ODR */ + /* 16 bits, 125Hz ODR */ { {1024, 1024, 1024}, 32768, }, - /* 16 bits, 200Hz ODR */ + /* 16 bits, 250Hz ODR */ { {1024, 1024, 770}, 32768, }, - /* 14 bits, 333Hz ODR */ + /* 14 bits, 450Hz ODR */ { {256, 256, 193}, 8192, }, - /* 12 bits, 666Hz ODR */ + /* 12 bits, 800Hz ODR */ { {64, 64, 48}, 2048, @@ -133,9 +133,15 @@ struct mmc35240_data { int axis_scale[3]; }; -static const int mmc35240_samp_freq[] = {100, 200, 333, 666}; +static const struct { + int val; + int val2; +} mmc35240_samp_freq[] = { {1, 500000}, + {13, 0}, + {25, 0}, + {50, 0} }; -static IIO_CONST_ATTR_SAMP_FREQ_AVAIL("100 200 333 666"); +static IIO_CONST_ATTR_SAMP_FREQ_AVAIL("1.5 13 25 50"); #define MMC35240_CHANNEL(_axis) { \ .type = IIO_MAGN, \ @@ -168,7 +174,8 @@ static int mmc35240_get_samp_freq_index(struct mmc35240_data *data, int i; for (i = 0; i < ARRAY_SIZE(mmc35240_samp_freq); i++) - if (mmc35240_samp_freq[i] == val) + if (mmc35240_samp_freq[i].val == val && + mmc35240_samp_freq[i].val2 == val2) return i; return -EINVAL; } @@ -378,9 +385,9 @@ static int mmc35240_read_raw(struct iio_dev *indio_dev, if (i < 0 || i >= ARRAY_SIZE(mmc35240_samp_freq)) return -EINVAL; - *val = mmc35240_samp_freq[i]; - *val2 = 0; - return IIO_VAL_INT; + *val = mmc35240_samp_freq[i].val; + *val2 = mmc35240_samp_freq[i].val2; + return IIO_VAL_INT_PLUS_MICRO; default: return -EINVAL; } -- cgit v0.10.2 From 498adaeb898c4eff190d598799bd9b0d607ce051 Mon Sep 17 00:00:00 2001 From: Vlad Dogaru Date: Tue, 30 Jun 2015 14:20:58 +0300 Subject: iio: sx9500: rework error handling of raw readings Fix error handling so that we can power the chip down even if a raw read fails. Reported-by: Hartmut Knaack Signed-off-by: Vlad Dogaru Acked-by: Hartmut Knaack Signed-off-by: Jonathan Cameron diff --git a/drivers/iio/proximity/sx9500.c b/drivers/iio/proximity/sx9500.c index 01494cd..21eaa16 100644 --- a/drivers/iio/proximity/sx9500.c +++ b/drivers/iio/proximity/sx9500.c @@ -329,27 +329,29 @@ static int sx9500_read_proximity(struct sx9500_data *data, else ret = sx9500_wait_for_sample(data); - if (ret < 0) - return ret; - mutex_lock(&data->mutex); - ret = sx9500_read_prox_data(data, chan, val); if (ret < 0) - goto out; + goto out_dec_data_rdy; - ret = sx9500_dec_chan_users(data, chan->channel); + ret = sx9500_read_prox_data(data, chan, val); if (ret < 0) - goto out; + goto out_dec_data_rdy; ret = sx9500_dec_data_rdy_users(data); if (ret < 0) + goto out_dec_chan; + + ret = sx9500_dec_chan_users(data, chan->channel); + if (ret < 0) goto out; ret = IIO_VAL_INT; goto out; +out_dec_data_rdy: + sx9500_dec_data_rdy_users(data); out_dec_chan: sx9500_dec_chan_users(data, chan->channel); out: -- cgit v0.10.2 From 0d1462de0b24fd09284411a3753f24a3d7c67577 Mon Sep 17 00:00:00 2001 From: Vlad Dogaru Date: Tue, 30 Jun 2015 14:20:59 +0300 Subject: iio: sx9500: fix bug in compensation code The initial compensation was mistakingly toggling an extra bit in the control register. Fix this and make sure it's less likely to happen by introducing an additional macro. Reported-by: Hartmut Knaack Signed-off-by: Vlad Dogaru Acked-by: Hartmut Knaack Signed-off-by: Jonathan Cameron diff --git a/drivers/iio/proximity/sx9500.c b/drivers/iio/proximity/sx9500.c index 21eaa16..d260509 100644 --- a/drivers/iio/proximity/sx9500.c +++ b/drivers/iio/proximity/sx9500.c @@ -80,6 +80,7 @@ #define SX9500_COMPSTAT_MASK GENMASK(3, 0) #define SX9500_NUM_CHANNELS 4 +#define SX9500_CHAN_MASK GENMASK(SX9500_NUM_CHANNELS - 1, 0) struct sx9500_data { struct mutex mutex; @@ -802,8 +803,7 @@ static int sx9500_init_compensation(struct iio_dev *indio_dev) unsigned int val; ret = regmap_update_bits(data->regmap, SX9500_REG_PROX_CTRL0, - GENMASK(SX9500_NUM_CHANNELS, 0), - GENMASK(SX9500_NUM_CHANNELS, 0)); + SX9500_CHAN_MASK, SX9500_CHAN_MASK); if (ret < 0) return ret; @@ -823,7 +823,7 @@ static int sx9500_init_compensation(struct iio_dev *indio_dev) out: regmap_update_bits(data->regmap, SX9500_REG_PROX_CTRL0, - GENMASK(SX9500_NUM_CHANNELS, 0), 0); + SX9500_CHAN_MASK, 0); return ret; } -- cgit v0.10.2 From fe5adb917431aafe1324a2455634ed59147df807 Mon Sep 17 00:00:00 2001 From: Hartmut Knaack Date: Sun, 28 Jun 2015 12:31:53 +0200 Subject: iio:light:ltr501: fix variable in ltr501_init When filling data->als_contr, the register content read into status needs to be used, instead of the return status value of regmap_read. Fixes: 8592a7eefa540 ("iio: ltr501: Add support for ltr559 chip") Signed-off-by: Hartmut Knaack Acked-by: Daniel Baluta Signed-off-by: Jonathan Cameron diff --git a/drivers/iio/light/ltr501.c b/drivers/iio/light/ltr501.c index 1ef7d37..b5a0e66 100644 --- a/drivers/iio/light/ltr501.c +++ b/drivers/iio/light/ltr501.c @@ -1302,7 +1302,7 @@ static int ltr501_init(struct ltr501_data *data) if (ret < 0) return ret; - data->als_contr = ret | data->chip_info->als_mode_active; + data->als_contr = status | data->chip_info->als_mode_active; ret = regmap_read(data->regmap, LTR501_PS_CONTR, &status); if (ret < 0) -- cgit v0.10.2 From c42b9e13f98239ec77acf7c4a20003e1cdd9bace Mon Sep 17 00:00:00 2001 From: Hartmut Knaack Date: Sun, 28 Jun 2015 12:31:52 +0200 Subject: iio:light:ltr501: fix regmap dependency The use of regmap in commit 2f2c96338afc requires REGMAP_I2C to be selected, in order to meet all dependencies. Fixes: 2f2c96338afc ("iio: ltr501: Add regmap support.") Signed-off-by: Hartmut Knaack Acked-by: Daniel Baluta Signed-off-by: Jonathan Cameron diff --git a/drivers/iio/light/Kconfig b/drivers/iio/light/Kconfig index df995a4..a5c5925 100644 --- a/drivers/iio/light/Kconfig +++ b/drivers/iio/light/Kconfig @@ -188,6 +188,7 @@ config SENSORS_LM3533 config LTR501 tristate "LTR-501ALS-01 light sensor" depends on I2C + select REGMAP_I2C select IIO_BUFFER select IIO_TRIGGERED_BUFFER help -- cgit v0.10.2 From 06b00f99ca8a0ec79f5d428dfb128abd14162c57 Mon Sep 17 00:00:00 2001 From: Geert Uytterhoeven Date: Mon, 29 Jun 2015 09:14:33 +0200 Subject: iio: sx9500: Add missing init in sx9500_buffer_pre{en,dis}able() MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit drivers/iio/proximity/sx9500.c: In function ‘sx9500_buffer_preenable’: drivers/iio/proximity/sx9500.c:682: warning: ‘ret’ may be used uninitialized in this function drivers/iio/proximity/sx9500.c: In function ‘sx9500_buffer_predisable’: drivers/iio/proximity/sx9500.c:706: warning: ‘ret’ may be used uninitialized in this function If active_scan_mask is empty, it will loop once more over all channels, doing nothing. Signed-off-by: Geert Uytterhoeven Reviewed-by: Vlad Dogaru Signed-off-by: Jonathan Cameron diff --git a/drivers/iio/proximity/sx9500.c b/drivers/iio/proximity/sx9500.c index d260509..3d756bd 100644 --- a/drivers/iio/proximity/sx9500.c +++ b/drivers/iio/proximity/sx9500.c @@ -682,7 +682,7 @@ out: static int sx9500_buffer_preenable(struct iio_dev *indio_dev) { struct sx9500_data *data = iio_priv(indio_dev); - int ret, i; + int ret = 0, i; mutex_lock(&data->mutex); @@ -706,7 +706,7 @@ static int sx9500_buffer_preenable(struct iio_dev *indio_dev) static int sx9500_buffer_predisable(struct iio_dev *indio_dev) { struct sx9500_data *data = iio_priv(indio_dev); - int ret, i; + int ret = 0, i; iio_triggered_buffer_predisable(indio_dev); -- cgit v0.10.2 From f451957dafe712ab2cd4b7ca8ba9197798115fbe Mon Sep 17 00:00:00 2001 From: Peter Meerwald Date: Sun, 21 Jun 2015 23:50:21 +0200 Subject: iio: tmp006: Check channel info on write only SAMP_FREQ is writable Will lead to SAMP_FREQ being written by any attempt to write to the other exported attributes and hence a rather unexpected result! Signed-off-by: Peter Meerwald Cc: Signed-off-by: Jonathan Cameron diff --git a/drivers/iio/temperature/tmp006.c b/drivers/iio/temperature/tmp006.c index fcc49f8..8f21f32 100644 --- a/drivers/iio/temperature/tmp006.c +++ b/drivers/iio/temperature/tmp006.c @@ -132,6 +132,9 @@ static int tmp006_write_raw(struct iio_dev *indio_dev, struct tmp006_data *data = iio_priv(indio_dev); int i; + if (mask != IIO_CHAN_INFO_SAMP_FREQ) + return -EINVAL; + for (i = 0; i < ARRAY_SIZE(tmp006_freqs); i++) if ((val == tmp006_freqs[i][0]) && (val2 == tmp006_freqs[i][1])) { -- cgit v0.10.2 From 42e0bb4db0ee3727de8ec1e5e0eb5f2dd686be00 Mon Sep 17 00:00:00 2001 From: Malcolm Priestley Date: Thu, 9 Jul 2015 17:01:24 +0100 Subject: staging: vt6655: check ieee80211_bss_conf bssid not NULL Sometimes bssid can go null on failed association. Signed-off-by: Malcolm Priestley Cc: # v3.19+ Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/staging/vt6655/device_main.c b/drivers/staging/vt6655/device_main.c index ed040fb..b0c8e23 100644 --- a/drivers/staging/vt6655/device_main.c +++ b/drivers/staging/vt6655/device_main.c @@ -1418,7 +1418,7 @@ static void vnt_bss_info_changed(struct ieee80211_hw *hw, priv->current_aid = conf->aid; - if (changed & BSS_CHANGED_BSSID) { + if (changed & BSS_CHANGED_BSSID && conf->bssid) { unsigned long flags; spin_lock_irqsave(&priv->lock, flags); -- cgit v0.10.2 From 7eb7ee30559c9c88ca23fa3b95bd06629669d0a8 Mon Sep 17 00:00:00 2001 From: Malcolm Priestley Date: Thu, 9 Jul 2015 17:03:57 +0100 Subject: staging: vt6656: check ieee80211_bss_conf bssid not NULL Sometimes bssid can go null on failed association. Signed-off-by: Malcolm Priestley Cc: # v3.17+ Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/staging/vt6656/main_usb.c b/drivers/staging/vt6656/main_usb.c index f97323f..af572d7 100644 --- a/drivers/staging/vt6656/main_usb.c +++ b/drivers/staging/vt6656/main_usb.c @@ -701,7 +701,7 @@ static void vnt_bss_info_changed(struct ieee80211_hw *hw, priv->current_aid = conf->aid; - if (changed & BSS_CHANGED_BSSID) + if (changed & BSS_CHANGED_BSSID && conf->bssid) vnt_mac_set_bssid_addr(priv, (u8 *)conf->bssid); -- cgit v0.10.2 From 2e187a028496ffde88c9441a5dbb4e2cd58abb98 Mon Sep 17 00:00:00 2001 From: Hartmut Knaack Date: Fri, 17 Jul 2015 10:52:24 +0200 Subject: iio:magnetometer:bmc150_magn: add regmap dependency bmc150_magn makes use of REGMAP_I2C, so select it to build always without errors. Fixes: c91746a2361d ("iio: magn: Add support for BMC150 magnetometer") Signed-off-by: Hartmut Knaack Signed-off-by: Jonathan Cameron diff --git a/drivers/iio/magnetometer/Kconfig b/drivers/iio/magnetometer/Kconfig index dcadfc4..efb9350 100644 --- a/drivers/iio/magnetometer/Kconfig +++ b/drivers/iio/magnetometer/Kconfig @@ -90,6 +90,7 @@ config IIO_ST_MAGN_SPI_3AXIS config BMC150_MAGN tristate "Bosch BMC150 Magnetometer Driver" depends on I2C + select REGMAP_I2C select IIO_BUFFER select IIO_TRIGGERED_BUFFER help -- cgit v0.10.2 From 6a14925ef22bf6e9e3f5209db50708210d5ee451 Mon Sep 17 00:00:00 2001 From: Hartmut Knaack Date: Fri, 17 Jul 2015 10:52:26 +0200 Subject: iio:magnetometer:bmc150_magn: output intended variable According to the debug/error string, the content of chip_id is supposed to be output, rather than the return value of the previous operation. Fixes: c91746a2361d ("iio: magn: Add support for BMC150 magnetometer") Signed-off-by: Hartmut Knaack Signed-off-by: Jonathan Cameron diff --git a/drivers/iio/magnetometer/bmc150_magn.c b/drivers/iio/magnetometer/bmc150_magn.c index d4c1788..1347a1f 100644 --- a/drivers/iio/magnetometer/bmc150_magn.c +++ b/drivers/iio/magnetometer/bmc150_magn.c @@ -706,11 +706,11 @@ static int bmc150_magn_init(struct bmc150_magn_data *data) goto err_poweroff; } if (chip_id != BMC150_MAGN_CHIP_ID_VAL) { - dev_err(&data->client->dev, "Invalid chip id 0x%x\n", ret); + dev_err(&data->client->dev, "Invalid chip id 0x%x\n", chip_id); ret = -ENODEV; goto err_poweroff; } - dev_dbg(&data->client->dev, "Chip id %x\n", ret); + dev_dbg(&data->client->dev, "Chip id %x\n", chip_id); preset = bmc150_magn_presets_table[BMC150_MAGN_DEFAULT_PRESET]; ret = bmc150_magn_set_odr(data, preset.odr); -- cgit v0.10.2 From 8b14821a5c45e3472082c60723ca54cf2fe9c2f3 Mon Sep 17 00:00:00 2001 From: Daniel Baluta Date: Tue, 14 Jul 2015 17:56:52 +0300 Subject: iio: magnetometer: mmc35240: Fix crash in pm suspend We must set i2c client private data at probe in order to correctly retrieve it in pm suspend/resume, preventing the following crash: [ 321.790582] PM: Syncing filesystems ... done. [ 322.364440] PM: Preparing system for mem sleep [ 322.400047] PM: Entering mem sleep [ 322.462178] BUG: unable to handle kernel NULL pointer dereference at 0000036c [ 322.469119] IP: [<80e0b3d2>] mmc35240_suspend+0x12/0x30 [ 322.474291] *pdpt = 000000002fd6f001 *pde = 0000000000000000 [ 322.479967] Oops: 0000 1 PREEMPT SMP [ 322.496516] task: a86d0df0 ti: a8766000 task.ti: a8766000 [ 322.570744] Call Trace: [ 322.573217] [<80c0d2d1>] pm_generic_suspend+0x21/0x30 [ 322.578284] [<80d042ab>] i2c_device_pm_suspend+0x1b/0x30 Signed-off-by: Daniel Baluta Signed-off-by: Jonathan Cameron diff --git a/drivers/iio/magnetometer/mmc35240.c b/drivers/iio/magnetometer/mmc35240.c index d927397..36a4b53 100644 --- a/drivers/iio/magnetometer/mmc35240.c +++ b/drivers/iio/magnetometer/mmc35240.c @@ -503,6 +503,7 @@ static int mmc35240_probe(struct i2c_client *client, } data = iio_priv(indio_dev); + i2c_set_clientdata(client, indio_dev); data->client = client; data->regmap = regmap; data->res = MMC35240_16_BITS_SLOW; -- cgit v0.10.2 From 3ceaa2c207ccf6933affc8358c0a04beb93e9aa7 Mon Sep 17 00:00:00 2001 From: Daniel Baluta Date: Tue, 14 Jul 2015 17:56:53 +0300 Subject: iio: magnetometer: mmc35240: Fix SET/RESET mask This fixes setting the SET/RESET bit in the REG_CTRL0 register. Signed-off-by: Daniel Baluta Signed-off-by: Jonathan Cameron diff --git a/drivers/iio/magnetometer/mmc35240.c b/drivers/iio/magnetometer/mmc35240.c index 36a4b53..e9dfa0e 100644 --- a/drivers/iio/magnetometer/mmc35240.c +++ b/drivers/iio/magnetometer/mmc35240.c @@ -202,8 +202,8 @@ static int mmc35240_hw_set(struct mmc35240_data *data, bool set) coil_bit = MMC35240_CTRL0_RESET_BIT; return regmap_update_bits(data->regmap, MMC35240_REG_CTRL0, - MMC35240_CTRL0_REFILL_BIT, - coil_bit); + coil_bit, coil_bit); + } static int mmc35240_init(struct mmc35240_data *data) -- cgit v0.10.2 From 354c879dbd9393ac411049f968474f2bf6c2a495 Mon Sep 17 00:00:00 2001 From: Viorel Suman Date: Tue, 14 Jul 2015 17:56:54 +0300 Subject: iio: magnetometer: mmc35240: fix SET/RESET sequence The RESET operation invoked in the last instance will align in the natural way all 3 axis and the chip top view. Without this, north and south are swapped. Signed-off-by: Viorel Suman Signed-off-by: Daniel Baluta Signed-off-by: Jonathan Cameron diff --git a/drivers/iio/magnetometer/mmc35240.c b/drivers/iio/magnetometer/mmc35240.c index e9dfa0e..706ebfd 100644 --- a/drivers/iio/magnetometer/mmc35240.c +++ b/drivers/iio/magnetometer/mmc35240.c @@ -222,14 +222,15 @@ static int mmc35240_init(struct mmc35240_data *data) /* * make sure we restore sensor characteristics, by doing - * a RESET/SET sequence + * a SET/RESET sequence, the axis polarity being naturally + * aligned after RESET */ - ret = mmc35240_hw_set(data, false); + ret = mmc35240_hw_set(data, true); if (ret < 0) return ret; usleep_range(MMC53240_WAIT_SET_RESET, MMC53240_WAIT_SET_RESET + 1); - ret = mmc35240_hw_set(data, true); + ret = mmc35240_hw_set(data, false); if (ret < 0) return ret; -- cgit v0.10.2 From c68a67b7adeec240bd79780635dc0ce74d0cb61a Mon Sep 17 00:00:00 2001 From: Crt Mori Date: Sun, 5 Jul 2015 20:03:53 +0200 Subject: iio: mlx96014: Replace offset sign Changed the offset to negative as usual equation is: (raw + offset)*scale and in this case offset should be negative (as we deduct 273.15 Kelvin to get temperature in Celsius). Signed-off-by: Crt Mori Acked-by: Peter Meerwald Signed-off-by: Jonathan Cameron diff --git a/drivers/iio/temperature/mlx90614.c b/drivers/iio/temperature/mlx90614.c index cb2e8ad..7a2b639 100644 --- a/drivers/iio/temperature/mlx90614.c +++ b/drivers/iio/temperature/mlx90614.c @@ -204,7 +204,7 @@ static int mlx90614_read_raw(struct iio_dev *indio_dev, *val = ret; return IIO_VAL_INT; case IIO_CHAN_INFO_OFFSET: - *val = 13657; + *val = -13657; *val2 = 500000; return IIO_VAL_INT_PLUS_MICRO; case IIO_CHAN_INFO_SCALE: -- cgit v0.10.2 From bf604a4c44a91cc2ceb60d4643a091b6b32cc999 Mon Sep 17 00:00:00 2001 From: Fugang Duan Date: Thu, 16 Jul 2015 14:49:09 +0800 Subject: iio: adc: vf610: fix the adc register read fail issue Read the register only when the adc register address is 4 byte aligned. (rather than the other way around). Signed-off-by: Haibo Chen Signed-off-by: Fugang Duan Cc: Signed-off-by: Jonathan Cameron diff --git a/drivers/iio/adc/vf610_adc.c b/drivers/iio/adc/vf610_adc.c index 480f335..819632b 100644 --- a/drivers/iio/adc/vf610_adc.c +++ b/drivers/iio/adc/vf610_adc.c @@ -635,7 +635,7 @@ static int vf610_adc_reg_access(struct iio_dev *indio_dev, struct vf610_adc *info = iio_priv(indio_dev); if ((readval == NULL) || - (!(reg % 4) || (reg > VF610_REG_ADC_PCTL))) + ((reg % 4) || (reg > VF610_REG_ADC_PCTL))) return -EINVAL; *readval = readl(info->regs + reg); -- cgit v0.10.2 From 41be6a0d5aa3c4fcb472f3d44251683b085b00a3 Mon Sep 17 00:00:00 2001 From: Manfred Schlaegl Date: Fri, 10 Jul 2015 22:55:30 +0200 Subject: iio: mcp320x: Fix NULL pointer dereference On reading in_voltage_scale of we got an NULL pointer dereference Oops. The reason for this is, that mcp320x_read_raw tries to access chip_info->resolution from struct mcp320x, but chip_info is never set. chip_info was never set since the driver was added, but there was no acute problem, because it was not referenced. The acute problem exists since b12206e917ac34bec41b9ff93d37d8bd53a2b3bc iio: adc: mcp320x. Add support for more ADCs This patch fixes the issue by setting chip_info in mcp320x_probe. Signed-off-by: Manfred Schlaegl Reviewed-by: Michael Welling Signed-off-by: Jonathan Cameron diff --git a/drivers/iio/adc/mcp320x.c b/drivers/iio/adc/mcp320x.c index 8d9c9b9..d819823 100644 --- a/drivers/iio/adc/mcp320x.c +++ b/drivers/iio/adc/mcp320x.c @@ -299,6 +299,8 @@ static int mcp320x_probe(struct spi_device *spi) indio_dev->channels = chip_info->channels; indio_dev->num_channels = chip_info->num_channels; + adc->chip_info = chip_info; + adc->transfer[0].tx_buf = &adc->tx_buf; adc->transfer[0].len = sizeof(adc->tx_buf); adc->transfer[1].rx_buf = adc->rx_buf; -- cgit v0.10.2 From c5d0db0690ff0a963dc082e7645268c466bf9a84 Mon Sep 17 00:00:00 2001 From: Martin Kepplinger Date: Sun, 5 Jul 2015 19:50:18 +0200 Subject: iio: mma8452: use iio event type IIO_EV_TYPE_MAG IIO_EV_TYPE_THRESH in rising direction describes an event where the threshold is crossed in rising direction, positive or negative values being possible. This is not the case here. Since the threshold is no signed value and only the magnitude is compared, IIO_EV_TYPE_MAG is what describes the behaviour of these devices, see the sysfs-bus-iio ABI Documentation. Signed-off-by: Martin Kepplinger Signed-off-by: Christoph Muellner Cc: Signed-off-by: Jonathan Cameron diff --git a/drivers/iio/accel/mma8452.c b/drivers/iio/accel/mma8452.c index e8e2077..13ea1ea 100644 --- a/drivers/iio/accel/mma8452.c +++ b/drivers/iio/accel/mma8452.c @@ -557,21 +557,21 @@ static void mma8452_transient_interrupt(struct iio_dev *indio_dev) if (src & MMA8452_TRANSIENT_SRC_XTRANSE) iio_push_event(indio_dev, IIO_MOD_EVENT_CODE(IIO_ACCEL, 0, IIO_MOD_X, - IIO_EV_TYPE_THRESH, + IIO_EV_TYPE_MAG, IIO_EV_DIR_RISING), ts); if (src & MMA8452_TRANSIENT_SRC_YTRANSE) iio_push_event(indio_dev, IIO_MOD_EVENT_CODE(IIO_ACCEL, 0, IIO_MOD_Y, - IIO_EV_TYPE_THRESH, + IIO_EV_TYPE_MAG, IIO_EV_DIR_RISING), ts); if (src & MMA8452_TRANSIENT_SRC_ZTRANSE) iio_push_event(indio_dev, IIO_MOD_EVENT_CODE(IIO_ACCEL, 0, IIO_MOD_Z, - IIO_EV_TYPE_THRESH, + IIO_EV_TYPE_MAG, IIO_EV_DIR_RISING), ts); } @@ -644,7 +644,7 @@ static int mma8452_reg_access_dbg(struct iio_dev *indio_dev, static const struct iio_event_spec mma8452_transient_event[] = { { - .type = IIO_EV_TYPE_THRESH, + .type = IIO_EV_TYPE_MAG, .dir = IIO_EV_DIR_RISING, .mask_separate = BIT(IIO_EV_INFO_ENABLE), .mask_shared_by_type = BIT(IIO_EV_INFO_VALUE) | -- cgit v0.10.2 From 037e966f2d63896129220112644f32962af6d115 Mon Sep 17 00:00:00 2001 From: Hartmut Knaack Date: Thu, 9 Jul 2015 23:51:29 +0200 Subject: iio:light:stk3310: move device register to end of probe iio_device_register should be the last operation during probe. Therefor move up interrupt setup code and while at it, change the check for invalid values of client->irq to be smaller than zero. Fixes: 3dd477acbdd1 ("iio: light: Add threshold interrupt support for STK3310") As the device_register makes the userspace interfaces of the device available it is possible for requests to come in before the probe sequence has finished. This can lead to unhandled interrupts and similar. Signed-off-by: Hartmut Knaack Reviewed-by: Tiberiu Breana Signed-off-by: Jonathan Cameron diff --git a/drivers/iio/light/stk3310.c b/drivers/iio/light/stk3310.c index c1a2182..e7e6e5a 100644 --- a/drivers/iio/light/stk3310.c +++ b/drivers/iio/light/stk3310.c @@ -608,13 +608,7 @@ static int stk3310_probe(struct i2c_client *client, if (ret < 0) return ret; - ret = iio_device_register(indio_dev); - if (ret < 0) { - dev_err(&client->dev, "device_register failed\n"); - stk3310_set_state(data, STK3310_STATE_STANDBY); - } - - if (client->irq <= 0) + if (client->irq < 0) client->irq = stk3310_gpio_probe(client); if (client->irq >= 0) { @@ -629,6 +623,12 @@ static int stk3310_probe(struct i2c_client *client, client->irq); } + ret = iio_device_register(indio_dev); + if (ret < 0) { + dev_err(&client->dev, "device_register failed\n"); + stk3310_set_state(data, STK3310_STATE_STANDBY); + } + return ret; } -- cgit v0.10.2 From 423ad0c405d87868bd1ab19e8b503e6a41c86caf Mon Sep 17 00:00:00 2001 From: Hartmut Knaack Date: Thu, 9 Jul 2015 23:51:30 +0200 Subject: iio:light:stk3310: make endianness independent of host Data is stored in the device in be16 format. Make use of be16_to_cpu and cpu_to_be16 to have correct endianness on any host architecture. Signed-off-by: Hartmut Knaack Reviewed-by: Tiberiu Breana Signed-off-by: Jonathan Cameron diff --git a/drivers/iio/light/stk3310.c b/drivers/iio/light/stk3310.c index e7e6e5a..11a027a 100644 --- a/drivers/iio/light/stk3310.c +++ b/drivers/iio/light/stk3310.c @@ -200,7 +200,7 @@ static int stk3310_read_event(struct iio_dev *indio_dev, int *val, int *val2) { u8 reg; - u16 buf; + __be16 buf; int ret; struct stk3310_data *data = iio_priv(indio_dev); @@ -222,7 +222,7 @@ static int stk3310_read_event(struct iio_dev *indio_dev, dev_err(&data->client->dev, "register read failed\n"); return ret; } - *val = swab16(buf); + *val = be16_to_cpu(buf); return IIO_VAL_INT; } @@ -235,7 +235,7 @@ static int stk3310_write_event(struct iio_dev *indio_dev, int val, int val2) { u8 reg; - u16 buf; + __be16 buf; int ret; unsigned int index; struct stk3310_data *data = iio_priv(indio_dev); @@ -252,7 +252,7 @@ static int stk3310_write_event(struct iio_dev *indio_dev, else return -EINVAL; - buf = swab16(val); + buf = cpu_to_be16(val); ret = regmap_bulk_write(data->regmap, reg, &buf, 2); if (ret < 0) dev_err(&client->dev, "failed to set PS threshold!\n"); @@ -301,7 +301,7 @@ static int stk3310_read_raw(struct iio_dev *indio_dev, int *val, int *val2, long mask) { u8 reg; - u16 buf; + __be16 buf; int ret; unsigned int index; struct stk3310_data *data = iio_priv(indio_dev); @@ -322,7 +322,7 @@ static int stk3310_read_raw(struct iio_dev *indio_dev, mutex_unlock(&data->lock); return ret; } - *val = swab16(buf); + *val = be16_to_cpu(buf); mutex_unlock(&data->lock); return IIO_VAL_INT; case IIO_CHAN_INFO_INT_TIME: -- cgit v0.10.2 From 27aa2e3a3c99d3319de12ada0d292e33f5831e60 Mon Sep 17 00:00:00 2001 From: Linus Walleij Date: Wed, 8 Jul 2015 15:12:06 +0200 Subject: pinctrl: abx500: remove strict mode Commit a21763a0b1e5a5ab8310f581886d04beadc16616 "pinctrl: nomadik: activate strict mux mode" put all Nomadik pin controllers to strict mode. This was not good on the Snowball platform: the muxing of GPIOs to different pins is done with hogs in the DTS file, and then these GPIOs are used by offset, relying on hogs to mux the pins. Since that means the pin controller "owns" the pins and at the same time we have a GPIO user, this pin controller is by definition not strict. Signed-off-by: Linus Walleij diff --git a/drivers/pinctrl/nomadik/pinctrl-abx500.c b/drivers/pinctrl/nomadik/pinctrl-abx500.c index 557d0f2..97681fa 100644 --- a/drivers/pinctrl/nomadik/pinctrl-abx500.c +++ b/drivers/pinctrl/nomadik/pinctrl-abx500.c @@ -787,7 +787,6 @@ static const struct pinmux_ops abx500_pinmux_ops = { .set_mux = abx500_pmx_set, .gpio_request_enable = abx500_gpio_request_enable, .gpio_disable_free = abx500_gpio_disable_free, - .strict = true, }; static int abx500_get_groups_cnt(struct pinctrl_dev *pctldev) -- cgit v0.10.2 From 61bb3aef92a4d102382f399eafccd5c72be6fdf2 Mon Sep 17 00:00:00 2001 From: Sergei Shtylyov Date: Fri, 26 Jun 2015 01:40:56 +0300 Subject: sh-pfc: fix sparse GPIOs for R-Car SoCs The PFC driver causes the kernel to hang on the R-Car gen2 SoC based boards when the CPU_ALL_PORT() macro is fixed to reflect the reality, i.e. when the GPIO space becomes actually sparse. This happens because the _GP_GPIO() macro includes an indexed initializer which causes the "holes" (array entries filled with all 0s) between the groups of the existing GPIOs; and the driver can't cope with that. There seems to be no reason to use the indexed initializer, so we can remove the index specifier and so avoid the "holes". Signed-off-by: Sergei Shtylyov Acked-by: Laurent Pinchart Tested-by: Geert Uytterhoeven Signed-off-by: Linus Walleij diff --git a/drivers/pinctrl/sh-pfc/sh_pfc.h b/drivers/pinctrl/sh-pfc/sh_pfc.h index c7508d5..0874cfe 100644 --- a/drivers/pinctrl/sh-pfc/sh_pfc.h +++ b/drivers/pinctrl/sh-pfc/sh_pfc.h @@ -224,7 +224,7 @@ struct sh_pfc_soc_info { /* PINMUX_GPIO_GP_ALL - Expand to a list of sh_pfc_pin entries */ #define _GP_GPIO(bank, _pin, _name, sfx) \ - [(bank * 32) + _pin] = { \ + { \ .pin = (bank * 32) + _pin, \ .name = __stringify(_name), \ .enum_id = _name##_DATA, \ -- cgit v0.10.2 From c10372e615b8f790d30cbfcf59e43908ca42bf1a Mon Sep 17 00:00:00 2001 From: Grygorii Strashko Date: Mon, 6 Jul 2015 18:13:37 +0300 Subject: pinctrl: single: ensure pcs irq will not be forced threaded The PSC IRQ is requested using request_irq() API and as result it can be forced to be threaded IRQ in RT-Kernel if PCS_QUIRK_HAS_SHARED_IRQ is enabled for pinctrl domain. As result, following 'possible irq lock inversion dependency' report can be seen: ========================================================= [ INFO: possible irq lock inversion dependency detected ] 3.14.43-rt42-00360-g96ff499-dirty #24 Not tainted --------------------------------------------------------- irq/369-pinctrl/927 just changed the state of lock: (&pcs->lock){+.....}, at: [] pcs_irq_handle+0x48/0x9c but this lock was taken by another, HARDIRQ-safe lock in the past: (&irq_desc_lock_class){-.....} and interrupts could create inverse lock ordering between them. other info that might help us debug this: Possible interrupt unsafe locking scenario: CPU0 CPU1 ---- ---- lock(&pcs->lock); local_irq_disable(); lock(&irq_desc_lock_class); lock(&pcs->lock); lock(&irq_desc_lock_class); *** DEADLOCK *** no locks held by irq/369-pinctrl/927. the shortest dependencies between 2nd lock and 1st lock: -> (&irq_desc_lock_class){-.....} ops: 58724 { IN-HARDIRQ-W at: [] lock_acquire+0x9c/0x158 [] _raw_spin_lock+0x48/0x58 [] handle_fasteoi_irq+0x24/0x15c [] generic_handle_irq+0x3c/0x4c [] handle_IRQ+0x50/0xa0 [] gic_handle_irq+0x3c/0x6c [] __irq_svc+0x44/0x8c [] arch_cpu_idle+0x40/0x4c [] cpu_startup_entry+0x270/0x2e0 [] rest_init+0xd4/0xe4 [] start_kernel+0x3d0/0x3dc [<80008084>] 0x80008084 INITIAL USE at: [] lock_acquire+0x9c/0x158 [] _raw_spin_lock_irqsave+0x54/0x68 [] __irq_get_desc_lock+0x64/0xa4 [] irq_set_chip+0x30/0x78 [] irq_set_chip_and_handler_name+0x24/0x3c [] gic_irq_domain_map+0x48/0xb4 [] irq_domain_associate+0x84/0x1d4 [] irq_create_mapping+0x80/0x11c [] irq_create_of_mapping+0x80/0x120 [] irq_of_parse_and_map+0x34/0x3c [] omap_dm_timer_init_one+0x90/0x30c [] omap5_realtime_timer_init+0x8c/0x48c [] time_init+0x28/0x38 [] start_kernel+0x240/0x3dc [<80008084>] 0x80008084 } ... key at: [] irq_desc_lock_class+0x0/0x8 ... acquired at: [] _raw_spin_lock+0x48/0x58 [] pcs_irq_unmask+0x58/0xa0 [] irq_enable+0x38/0x48 [] irq_startup+0x78/0x7c [] __setup_irq+0x4a8/0x4f4 [] request_threaded_irq+0xb8/0x138 [] omap_8250_startup+0x4c/0x148 [] serial8250_startup+0x24/0x30 [] uart_startup.part.9+0x5c/0x1b4 [] uart_open+0xf4/0x16c [] tty_open+0x170/0x61c [] chrdev_open+0xbc/0x1b4 [] do_dentry_open+0x1e8/0x2bc [] finish_open+0x44/0x5c [] do_last.isra.47+0x710/0xca0 [] path_openat+0xc4/0x640 [] do_filp_open+0x3c/0x98 [] do_sys_open+0x114/0x1d8 [] SyS_open+0x28/0x2c [] kernel_init_freeable+0x168/0x1e4 [] kernel_init+0x1c/0xf8 [] ret_from_fork+0x14/0x20 -> (&pcs->lock){+.....} ops: 65 { HARDIRQ-ON-W at: [] lock_acquire+0x9c/0x158 [] _raw_spin_lock+0x48/0x58 [] pcs_irq_handle+0x48/0x9c [] pcs_irq_handler+0x1c/0x28 [] irq_forced_thread_fn+0x30/0x74 [] irq_thread+0x158/0x1c4 [] kthread+0xd4/0xe8 [] ret_from_fork+0x14/0x20 INITIAL USE at: [] lock_acquire+0x9c/0x158 [] _raw_spin_lock_irqsave+0x54/0x68 [] pcs_enable+0x7c/0xe8 [] pinmux_enable_setting+0x178/0x220 [] pinctrl_select_state+0x110/0x194 [] pinctrl_bind_pins+0x7c/0x108 [] driver_probe_device+0x70/0x254 [] __driver_attach+0x9c/0xa0 [] bus_for_each_dev+0x78/0xac [] driver_attach+0x2c/0x30 [] bus_add_driver+0x15c/0x204 [] driver_register+0x88/0x108 [] __platform_driver_register+0x64/0x6c [] omap_hsmmc_driver_init+0x1c/0x20 [] do_one_initcall+0x110/0x170 [] kernel_init_freeable+0x140/0x1e4 [] kernel_init+0x1c/0xf8 [] ret_from_fork+0x14/0x20 } ... key at: [] __key.18572+0x0/0x8 ... acquired at: [] mark_lock+0x388/0x76c [] __lock_acquire+0x6d0/0x1f98 [] lock_acquire+0x9c/0x158 [] _raw_spin_lock+0x48/0x58 [] pcs_irq_handle+0x48/0x9c [] pcs_irq_handler+0x1c/0x28 [] irq_forced_thread_fn+0x30/0x74 [] irq_thread+0x158/0x1c4 [] kthread+0xd4/0xe8 [] ret_from_fork+0x14/0x20 stack backtrace: CPU: 1 PID: 927 Comm: irq/369-pinctrl Not tainted 3.14.43-rt42-00360-g96ff499-dirty #24 [] (unwind_backtrace) from [] (show_stack+0x20/0x24) [] (show_stack) from [] (dump_stack+0x84/0xd0) [] (dump_stack) from [] (print_irq_inversion_bug+0x1d0/0x21c) [] (print_irq_inversion_bug) from [] (check_usage_backwards+0xb4/0x11c) [] (check_usage_backwards) from [] (mark_lock+0x388/0x76c) [] (mark_lock) from [] (__lock_acquire+0x6d0/0x1f98) [] (__lock_acquire) from [] (lock_acquire+0x9c/0x158) [] (lock_acquire) from [] (_raw_spin_lock+0x48/0x58) [] (_raw_spin_lock) from [] (pcs_irq_handle+0x48/0x9c) [] (pcs_irq_handle) from [] (pcs_irq_handler+0x1c/0x28) [] (pcs_irq_handler) from [] (irq_forced_thread_fn+0x30/0x74) [] (irq_forced_thread_fn) from [] (irq_thread+0x158/0x1c4) [] (irq_thread) from [] (kthread+0xd4/0xe8) [] (kthread) from [] (ret_from_fork+0x14/0x20) To fix it use IRQF_NO_THREAD to ensure that pcs irq will not be forced threaded. Cc: Tony Lindgren Cc: Sebastian Andrzej Siewior Signed-off-by: Grygorii Strashko Acked-by: Tony Lindgren Signed-off-by: Linus Walleij diff --git a/drivers/pinctrl/pinctrl-single.c b/drivers/pinctrl/pinctrl-single.c index b2de09d..0b8d480 100644 --- a/drivers/pinctrl/pinctrl-single.c +++ b/drivers/pinctrl/pinctrl-single.c @@ -1760,7 +1760,8 @@ static int pcs_irq_init_chained_handler(struct pcs_device *pcs, int res; res = request_irq(pcs_soc->irq, pcs_irq_handler, - IRQF_SHARED | IRQF_NO_SUSPEND, + IRQF_SHARED | IRQF_NO_SUSPEND | + IRQF_NO_THREAD, name, pcs_soc); if (res) { pcs_soc->irq = -1; -- cgit v0.10.2 From 714b1dd8f72e39ef4bc0f38f7f341bb1d57d98bf Mon Sep 17 00:00:00 2001 From: Jonathan Bell Date: Tue, 30 Jun 2015 12:35:39 +0100 Subject: pinctrl: bcm2835: Clear the event latch register when disabling interrupts It's possible to hit a race condition if interrupts are generated on a GPIO pin when the IRQ line in question is being disabled. If the interrupt is freed, bcm2835_gpio_irq_disable() is called which disables the event generation sources (edge, level). If an event occurred between the last disabling of hard IRQs and the write to the event source registers, a bit would be set in the GPIO event detect register (GPEDSn) which goes unacknowledged by bcm2835_gpio_irq_handler() so Linux complains loudly. There is no per-GPIO mask register, so when disabling GPIO interrupts write 1 to the relevant bit in GPEDSn to clear out any stale events. Signed-off-by: Jonathan Bell Acked-by: Stephen Warren Signed-off-by: Linus Walleij diff --git a/drivers/pinctrl/bcm/pinctrl-bcm2835.c b/drivers/pinctrl/bcm/pinctrl-bcm2835.c index efcf2a2..6177315 100644 --- a/drivers/pinctrl/bcm/pinctrl-bcm2835.c +++ b/drivers/pinctrl/bcm/pinctrl-bcm2835.c @@ -473,6 +473,8 @@ static void bcm2835_gpio_irq_disable(struct irq_data *data) spin_lock_irqsave(&pc->irq_lock[bank], flags); bcm2835_gpio_irq_config(pc, gpio, false); + /* Clear events that were latched prior to clearing event sources */ + bcm2835_gpio_set_bit(pc, GPEDS0, gpio); clear_bit(offset, &pc->enabled_irq_map[bank]); spin_unlock_irqrestore(&pc->irq_lock[bank], flags); } -- cgit v0.10.2 From 9571b25df1dbf4db17191b54f59734e8b77fd591 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Uwe=20Kleine-K=C3=B6nig?= Date: Fri, 17 Jul 2015 09:38:43 +0200 Subject: Subject: pinctrl: imx1-core: Fix debug output in .pin_config_set callback MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit imx1_pinconf_set assumes that the array of pins in struct imx1_pinctrl_soc_info can be indexed by pin id to get the pinctrl_pin_desc for a pin. This used to be correct up to commit 607af165c047 which removed some entries from the array and so made it wrong to access the array by pin id. The result of this bug is a wrong pin name in the output for small pin ids and an oops for the bigger ones. This patch is the result of a discussion that includes patches by Markus Pargmann and Chris Ruehl. Fixes: 607af165c047 ("pinctrl: i.MX27: Remove nonexistent pad definitions") Cc: stable@vger.kernel.org Reported-by: Chris Ruehl Signed-off-by: Uwe Kleine-König Reviewed-by: Markus Pargmann Signed-off-by: Linus Walleij diff --git a/drivers/pinctrl/freescale/pinctrl-imx1-core.c b/drivers/pinctrl/freescale/pinctrl-imx1-core.c index 5fd4437..88a7fac 100644 --- a/drivers/pinctrl/freescale/pinctrl-imx1-core.c +++ b/drivers/pinctrl/freescale/pinctrl-imx1-core.c @@ -403,14 +403,13 @@ static int imx1_pinconf_set(struct pinctrl_dev *pctldev, unsigned num_configs) { struct imx1_pinctrl *ipctl = pinctrl_dev_get_drvdata(pctldev); - const struct imx1_pinctrl_soc_info *info = ipctl->info; int i; for (i = 0; i != num_configs; ++i) { imx1_write_bit(ipctl, pin_id, configs[i] & 0x01, MX1_PUEN); dev_dbg(ipctl->dev, "pinconf set pullup pin %s\n", - info->pins[pin_id].name); + pin_desc_get(pctldev, pin_id)->name); } return 0; -- cgit v0.10.2 From 681ccdcc756f17f847beba5ac4cd3d03e0949489 Mon Sep 17 00:00:00 2001 From: Joachim Eastwood Date: Wed, 15 Jul 2015 00:25:26 +0200 Subject: pinctrl: lpc18xx: fix schmitt trigger setup The param_val variable is what determines if schmitt trigger is enabled on a pin or not. A typo here mean that schmitt trigger was always enabled for standard and i2c pins. Signed-off-by: Joachim Eastwood Signed-off-by: Linus Walleij diff --git a/drivers/pinctrl/pinctrl-lpc18xx.c b/drivers/pinctrl/pinctrl-lpc18xx.c index ef0b697..347c763 100644 --- a/drivers/pinctrl/pinctrl-lpc18xx.c +++ b/drivers/pinctrl/pinctrl-lpc18xx.c @@ -823,7 +823,7 @@ static int lpc18xx_pconf_set_i2c0(struct pinctrl_dev *pctldev, break; case PIN_CONFIG_INPUT_SCHMITT_ENABLE: - if (param) + if (param_val) *reg &= ~(LPC18XX_SCU_I2C0_ZIF << shift); else *reg |= (LPC18XX_SCU_I2C0_ZIF << shift); @@ -876,7 +876,7 @@ static int lpc18xx_pconf_set_pin(struct pinctrl_dev *pctldev, break; case PIN_CONFIG_INPUT_SCHMITT_ENABLE: - if (param) + if (param_val) *reg &= ~LPC18XX_SCU_PIN_ZIF; else *reg |= LPC18XX_SCU_PIN_ZIF; -- cgit v0.10.2 From 09adcdf212d777ace8bb31bed8ca5c1dbd56bd0f Mon Sep 17 00:00:00 2001 From: Viresh Kumar Date: Thu, 16 Jul 2015 16:56:15 +0530 Subject: AVR32/time: Migrate to new 'set-state' interface Migrate avr32 driver to the new 'set-state' interface provided by clockevents core, the earlier 'set-mode' interface is marked obsolete now. This also enables us to implement callbacks for new states of clockevent devices, for example: ONESHOT_STOPPED. We want to call cpu_idle_poll_ctrl() in shutdown only if we were in oneshot or resume state earlier. Create another variable to save this information and check that in shutdown callback. Cc: Haavard Skinnemoen Cc: Hans-Christian Egtvedt Signed-off-by: Viresh Kumar Acked-by: Hans-Christian Egtvedt diff --git a/arch/avr32/kernel/time.c b/arch/avr32/kernel/time.c index d0f771b..a124c55 100644 --- a/arch/avr32/kernel/time.c +++ b/arch/avr32/kernel/time.c @@ -18,6 +18,7 @@ #include +static bool disable_cpu_idle_poll; static cycle_t read_cycle_count(struct clocksource *cs) { @@ -80,45 +81,45 @@ static int comparator_next_event(unsigned long delta, return 0; } -static void comparator_mode(enum clock_event_mode mode, - struct clock_event_device *evdev) +static int comparator_shutdown(struct clock_event_device *evdev) { - switch (mode) { - case CLOCK_EVT_MODE_ONESHOT: - pr_debug("%s: start\n", evdev->name); - /* FALLTHROUGH */ - case CLOCK_EVT_MODE_RESUME: + pr_debug("%s: %s\n", __func__, evdev->name); + sysreg_write(COMPARE, 0); + + if (disable_cpu_idle_poll) { + disable_cpu_idle_poll = false; /* - * If we're using the COUNT and COMPARE registers we - * need to force idle poll. + * Only disable idle poll if we have forced that + * in a previous call. */ - cpu_idle_poll_ctrl(true); - break; - case CLOCK_EVT_MODE_UNUSED: - case CLOCK_EVT_MODE_SHUTDOWN: - sysreg_write(COMPARE, 0); - pr_debug("%s: stop\n", evdev->name); - if (evdev->mode == CLOCK_EVT_MODE_ONESHOT || - evdev->mode == CLOCK_EVT_MODE_RESUME) { - /* - * Only disable idle poll if we have forced that - * in a previous call. - */ - cpu_idle_poll_ctrl(false); - } - break; - default: - BUG(); + cpu_idle_poll_ctrl(false); } + return 0; +} + +static int comparator_set_oneshot(struct clock_event_device *evdev) +{ + pr_debug("%s: %s\n", __func__, evdev->name); + + disable_cpu_idle_poll = true; + /* + * If we're using the COUNT and COMPARE registers we + * need to force idle poll. + */ + cpu_idle_poll_ctrl(true); + + return 0; } static struct clock_event_device comparator = { - .name = "avr32_comparator", - .features = CLOCK_EVT_FEAT_ONESHOT, - .shift = 16, - .rating = 50, - .set_next_event = comparator_next_event, - .set_mode = comparator_mode, + .name = "avr32_comparator", + .features = CLOCK_EVT_FEAT_ONESHOT, + .shift = 16, + .rating = 50, + .set_next_event = comparator_next_event, + .set_state_shutdown = comparator_shutdown, + .set_state_oneshot = comparator_set_oneshot, + .tick_resume = comparator_set_oneshot, }; void read_persistent_clock(struct timespec *ts) -- cgit v0.10.2 From 3827ec3d8fd51aef8352b0282b14f0f3ab615930 Mon Sep 17 00:00:00 2001 From: Martin Schwidefsky Date: Mon, 20 Jul 2015 10:01:46 +0200 Subject: s390: adapt entry.S to the move of thread_struct git commit 0c8c0f03e3a292e031596484275c14cf39c0ab7a "x86/fpu, sched: Dynamically allocate 'struct fpu'" moved the thread_struct to the end of the task_struct. This causes some of the offsets used in entry.S to overflow their instruction operand field. To fix this use aghi to create a dedicated pointer for the thread_struct. Signed-off-by: Martin Schwidefsky diff --git a/arch/s390/kernel/asm-offsets.c b/arch/s390/kernel/asm-offsets.c index c7d1b9d..a2da259 100644 --- a/arch/s390/kernel/asm-offsets.c +++ b/arch/s390/kernel/asm-offsets.c @@ -23,15 +23,15 @@ int main(void) { - DEFINE(__THREAD_info, offsetof(struct task_struct, stack)); - DEFINE(__THREAD_ksp, offsetof(struct task_struct, thread.ksp)); - DEFINE(__THREAD_mm_segment, offsetof(struct task_struct, thread.mm_segment)); - BLANK(); + DEFINE(__TASK_thread_info, offsetof(struct task_struct, stack)); + DEFINE(__TASK_thread, offsetof(struct task_struct, thread)); DEFINE(__TASK_pid, offsetof(struct task_struct, pid)); BLANK(); - DEFINE(__THREAD_per_cause, offsetof(struct task_struct, thread.per_event.cause)); - DEFINE(__THREAD_per_address, offsetof(struct task_struct, thread.per_event.address)); - DEFINE(__THREAD_per_paid, offsetof(struct task_struct, thread.per_event.paid)); + DEFINE(__THREAD_ksp, offsetof(struct thread_struct, ksp)); + DEFINE(__THREAD_per_cause, offsetof(struct thread_struct, per_event.cause)); + DEFINE(__THREAD_per_address, offsetof(struct thread_struct, per_event.address)); + DEFINE(__THREAD_per_paid, offsetof(struct thread_struct, per_event.paid)); + DEFINE(__THREAD_trap_tdb, offsetof(struct thread_struct, trap_tdb)); BLANK(); DEFINE(__TI_task, offsetof(struct thread_info, task)); DEFINE(__TI_flags, offsetof(struct thread_info, flags)); @@ -176,7 +176,6 @@ int main(void) DEFINE(__LC_VDSO_PER_CPU, offsetof(struct _lowcore, vdso_per_cpu_data)); DEFINE(__LC_GMAP, offsetof(struct _lowcore, gmap)); DEFINE(__LC_PGM_TDB, offsetof(struct _lowcore, pgm_tdb)); - DEFINE(__THREAD_trap_tdb, offsetof(struct task_struct, thread.trap_tdb)); DEFINE(__GMAP_ASCE, offsetof(struct gmap, asce)); DEFINE(__SIE_PROG0C, offsetof(struct kvm_s390_sie_block, prog0c)); DEFINE(__SIE_PROG20, offsetof(struct kvm_s390_sie_block, prog20)); diff --git a/arch/s390/kernel/entry.S b/arch/s390/kernel/entry.S index 3238893..84062e7 100644 --- a/arch/s390/kernel/entry.S +++ b/arch/s390/kernel/entry.S @@ -178,17 +178,21 @@ _PIF_WORK = (_PIF_PER_TRAP) */ ENTRY(__switch_to) stmg %r6,%r15,__SF_GPRS(%r15) # store gprs of prev task - stg %r15,__THREAD_ksp(%r2) # store kernel stack of prev - lg %r4,__THREAD_info(%r2) # get thread_info of prev - lg %r5,__THREAD_info(%r3) # get thread_info of next + lgr %r1,%r2 + aghi %r1,__TASK_thread # thread_struct of prev task + lg %r4,__TASK_thread_info(%r2) # get thread_info of prev + lg %r5,__TASK_thread_info(%r3) # get thread_info of next + stg %r15,__THREAD_ksp(%r1) # store kernel stack of prev + lgr %r1,%r3 + aghi %r1,__TASK_thread # thread_struct of next task lgr %r15,%r5 aghi %r15,STACK_INIT # end of kernel stack of next stg %r3,__LC_CURRENT # store task struct of next stg %r5,__LC_THREAD_INFO # store thread info of next stg %r15,__LC_KERNEL_STACK # store end of kernel stack + lg %r15,__THREAD_ksp(%r1) # load kernel stack of next lctl %c4,%c4,__TASK_pid(%r3) # load pid to control reg. 4 mvc __LC_CURRENT_PID+4(4,%r0),__TASK_pid(%r3) # store pid of next - lg %r15,__THREAD_ksp(%r3) # load kernel stack of next lmg %r6,%r15,__SF_GPRS(%r15) # load gprs of next task br %r14 @@ -417,6 +421,7 @@ ENTRY(pgm_check_handler) LAST_BREAK %r14 lg %r15,__LC_KERNEL_STACK lg %r14,__TI_task(%r12) + aghi %r14,__TASK_thread # pointer to thread_struct lghi %r13,__LC_PGM_TDB tm __LC_PGM_ILC+2,0x02 # check for transaction abort jz 2f diff --git a/arch/s390/kernel/traps.c b/arch/s390/kernel/traps.c index 4d96c9f..7bea81d 100644 --- a/arch/s390/kernel/traps.c +++ b/arch/s390/kernel/traps.c @@ -259,7 +259,7 @@ void vector_exception(struct pt_regs *regs) } /* get vector interrupt code from fpc */ - asm volatile("stfpc %0" : "=m" (current->thread.fp_regs.fpc)); + asm volatile("stfpc %0" : "=Q" (current->thread.fp_regs.fpc)); vic = (current->thread.fp_regs.fpc & 0xf00) >> 8; switch (vic) { case 1: /* invalid vector operation */ @@ -297,7 +297,7 @@ void data_exception(struct pt_regs *regs) location = get_trap_ip(regs); - asm volatile("stfpc %0" : "=m" (current->thread.fp_regs.fpc)); + asm volatile("stfpc %0" : "=Q" (current->thread.fp_regs.fpc)); /* Check for vector register enablement */ if (MACHINE_HAS_VX && !current->thread.vxrs && (current->thread.fp_regs.fpc & FPC_DXC_MASK) == 0xfe00) { -- cgit v0.10.2 From 0838aa7fcfcd875caa7bcc5dab0c3fd40444553d Mon Sep 17 00:00:00 2001 From: Pablo Neira Ayuso Date: Mon, 13 Jul 2015 15:11:48 +0200 Subject: netfilter: fix netns dependencies with conntrack templates Quoting Daniel Borkmann: "When adding connection tracking template rules to a netns, f.e. to configure netfilter zones, the kernel will endlessly busy-loop as soon as we try to delete the given netns in case there's at least one template present, which is problematic i.e. if there is such bravery that the priviledged user inside the netns is assumed untrusted. Minimal example: ip netns add foo ip netns exec foo iptables -t raw -A PREROUTING -d 1.2.3.4 -j CT --zone 1 ip netns del foo What happens is that when nf_ct_iterate_cleanup() is being called from nf_conntrack_cleanup_net_list() for a provided netns, we always end up with a net->ct.count > 0 and thus jump back to i_see_dead_people. We don't get a soft-lockup as we still have a schedule() point, but the serving CPU spins on 100% from that point onwards. Since templates are normally allocated with nf_conntrack_alloc(), we also bump net->ct.count. The issue why they are not yet nf_ct_put() is because the per netns .exit() handler from x_tables (which would eventually invoke xt_CT's xt_ct_tg_destroy() that drops reference on info->ct) is called in the dependency chain at a *later* point in time than the per netns .exit() handler for the connection tracker. This is clearly a chicken'n'egg problem: after the connection tracker .exit() handler, we've teared down all the connection tracking infrastructure already, so rightfully, xt_ct_tg_destroy() cannot be invoked at a later point in time during the netns cleanup, as that would lead to a use-after-free. At the same time, we cannot make x_tables depend on the connection tracker module, so that the xt_ct_tg_destroy() would be invoked earlier in the cleanup chain." Daniel confirms this has to do with the order in which modules are loaded or having compiled nf_conntrack as modules while x_tables built-in. So we have no guarantees regarding the order in which netns callbacks are executed. Fix this by allocating the templates through kmalloc() from the respective SYNPROXY and CT targets, so they don't depend on the conntrack kmem cache. Then, release then via nf_ct_tmpl_free() from destroy_conntrack(). This branch is marked as unlikely since conntrack templates are rarely allocated and only from the configuration plane path. Note that templates are not kept in any list to avoid further dependencies with nf_conntrack anymore, thus, the tmpl larval list is removed. Reported-by: Daniel Borkmann Signed-off-by: Pablo Neira Ayuso Tested-by: Daniel Borkmann diff --git a/include/net/netfilter/nf_conntrack.h b/include/net/netfilter/nf_conntrack.h index 095433b..37cd391 100644 --- a/include/net/netfilter/nf_conntrack.h +++ b/include/net/netfilter/nf_conntrack.h @@ -291,7 +291,7 @@ extern unsigned int nf_conntrack_max; extern unsigned int nf_conntrack_hash_rnd; void init_nf_conntrack_hash_rnd(void); -void nf_conntrack_tmpl_insert(struct net *net, struct nf_conn *tmpl); +struct nf_conn *nf_ct_tmpl_alloc(struct net *net, u16 zone, gfp_t flags); #define NF_CT_STAT_INC(net, count) __this_cpu_inc((net)->ct.stat->count) #define NF_CT_STAT_INC_ATOMIC(net, count) this_cpu_inc((net)->ct.stat->count) diff --git a/include/net/netns/conntrack.h b/include/net/netns/conntrack.h index 29d6a94..723b61c 100644 --- a/include/net/netns/conntrack.h +++ b/include/net/netns/conntrack.h @@ -68,7 +68,6 @@ struct ct_pcpu { spinlock_t lock; struct hlist_nulls_head unconfirmed; struct hlist_nulls_head dying; - struct hlist_nulls_head tmpl; }; struct netns_ct { diff --git a/net/netfilter/nf_conntrack_core.c b/net/netfilter/nf_conntrack_core.c index 13fad86..651039a 100644 --- a/net/netfilter/nf_conntrack_core.c +++ b/net/netfilter/nf_conntrack_core.c @@ -287,6 +287,46 @@ static void nf_ct_del_from_dying_or_unconfirmed_list(struct nf_conn *ct) spin_unlock(&pcpu->lock); } +/* Released via destroy_conntrack() */ +struct nf_conn *nf_ct_tmpl_alloc(struct net *net, u16 zone, gfp_t flags) +{ + struct nf_conn *tmpl; + + tmpl = kzalloc(sizeof(struct nf_conn), GFP_KERNEL); + if (tmpl == NULL) + return NULL; + + tmpl->status = IPS_TEMPLATE; + write_pnet(&tmpl->ct_net, net); + +#ifdef CONFIG_NF_CONNTRACK_ZONES + if (zone) { + struct nf_conntrack_zone *nf_ct_zone; + + nf_ct_zone = nf_ct_ext_add(tmpl, NF_CT_EXT_ZONE, GFP_ATOMIC); + if (!nf_ct_zone) + goto out_free; + nf_ct_zone->id = zone; + } +#endif + atomic_set(&tmpl->ct_general.use, 0); + + return tmpl; +#ifdef CONFIG_NF_CONNTRACK_ZONES +out_free: + kfree(tmpl); + return NULL; +#endif +} +EXPORT_SYMBOL_GPL(nf_ct_tmpl_alloc); + +static void nf_ct_tmpl_free(struct nf_conn *tmpl) +{ + nf_ct_ext_destroy(tmpl); + nf_ct_ext_free(tmpl); + kfree(tmpl); +} + static void destroy_conntrack(struct nf_conntrack *nfct) { @@ -298,6 +338,10 @@ destroy_conntrack(struct nf_conntrack *nfct) NF_CT_ASSERT(atomic_read(&nfct->use) == 0); NF_CT_ASSERT(!timer_pending(&ct->timeout)); + if (unlikely(nf_ct_is_template(ct))) { + nf_ct_tmpl_free(ct); + return; + } rcu_read_lock(); l4proto = __nf_ct_l4proto_find(nf_ct_l3num(ct), nf_ct_protonum(ct)); if (l4proto && l4proto->destroy) @@ -540,28 +584,6 @@ out: } EXPORT_SYMBOL_GPL(nf_conntrack_hash_check_insert); -/* deletion from this larval template list happens via nf_ct_put() */ -void nf_conntrack_tmpl_insert(struct net *net, struct nf_conn *tmpl) -{ - struct ct_pcpu *pcpu; - - __set_bit(IPS_TEMPLATE_BIT, &tmpl->status); - __set_bit(IPS_CONFIRMED_BIT, &tmpl->status); - nf_conntrack_get(&tmpl->ct_general); - - /* add this conntrack to the (per cpu) tmpl list */ - local_bh_disable(); - tmpl->cpu = smp_processor_id(); - pcpu = per_cpu_ptr(nf_ct_net(tmpl)->ct.pcpu_lists, tmpl->cpu); - - spin_lock(&pcpu->lock); - /* Overload tuple linked list to put us in template list. */ - hlist_nulls_add_head_rcu(&tmpl->tuplehash[IP_CT_DIR_ORIGINAL].hnnode, - &pcpu->tmpl); - spin_unlock_bh(&pcpu->lock); -} -EXPORT_SYMBOL_GPL(nf_conntrack_tmpl_insert); - /* Confirm a connection given skb; places it in hash table */ int __nf_conntrack_confirm(struct sk_buff *skb) @@ -1751,7 +1773,6 @@ int nf_conntrack_init_net(struct net *net) spin_lock_init(&pcpu->lock); INIT_HLIST_NULLS_HEAD(&pcpu->unconfirmed, UNCONFIRMED_NULLS_VAL); INIT_HLIST_NULLS_HEAD(&pcpu->dying, DYING_NULLS_VAL); - INIT_HLIST_NULLS_HEAD(&pcpu->tmpl, TEMPLATE_NULLS_VAL); } net->ct.stat = alloc_percpu(struct ip_conntrack_stat); diff --git a/net/netfilter/nf_synproxy_core.c b/net/netfilter/nf_synproxy_core.c index 789feea..71f1e9f 100644 --- a/net/netfilter/nf_synproxy_core.c +++ b/net/netfilter/nf_synproxy_core.c @@ -349,12 +349,10 @@ static void __net_exit synproxy_proc_exit(struct net *net) static int __net_init synproxy_net_init(struct net *net) { struct synproxy_net *snet = synproxy_pernet(net); - struct nf_conntrack_tuple t; struct nf_conn *ct; int err = -ENOMEM; - memset(&t, 0, sizeof(t)); - ct = nf_conntrack_alloc(net, 0, &t, &t, GFP_KERNEL); + ct = nf_ct_tmpl_alloc(net, 0, GFP_KERNEL); if (IS_ERR(ct)) { err = PTR_ERR(ct); goto err1; @@ -365,7 +363,8 @@ static int __net_init synproxy_net_init(struct net *net) if (!nfct_synproxy_ext_add(ct)) goto err2; - nf_conntrack_tmpl_insert(net, ct); + __set_bit(IPS_CONFIRMED_BIT, &ct->status); + nf_conntrack_get(&ct->ct_general); snet->tmpl = ct; snet->stats = alloc_percpu(struct synproxy_stats); diff --git a/net/netfilter/xt_CT.c b/net/netfilter/xt_CT.c index 75747ae..c663003 100644 --- a/net/netfilter/xt_CT.c +++ b/net/netfilter/xt_CT.c @@ -184,7 +184,6 @@ out: static int xt_ct_tg_check(const struct xt_tgchk_param *par, struct xt_ct_target_info_v1 *info) { - struct nf_conntrack_tuple t; struct nf_conn *ct; int ret = -EOPNOTSUPP; @@ -202,8 +201,7 @@ static int xt_ct_tg_check(const struct xt_tgchk_param *par, if (ret < 0) goto err1; - memset(&t, 0, sizeof(t)); - ct = nf_conntrack_alloc(par->net, info->zone, &t, &t, GFP_KERNEL); + ct = nf_ct_tmpl_alloc(par->net, info->zone, GFP_KERNEL); ret = PTR_ERR(ct); if (IS_ERR(ct)) goto err2; @@ -227,8 +225,8 @@ static int xt_ct_tg_check(const struct xt_tgchk_param *par, if (ret < 0) goto err3; } - - nf_conntrack_tmpl_insert(par->net, ct); + __set_bit(IPS_CONFIRMED_BIT, &ct->status); + nf_conntrack_get(&ct->ct_general); out: info->ct = ct; return 0; -- cgit v0.10.2 From 4696b8874d7de39850931888bc9f2aa12d29fb46 Mon Sep 17 00:00:00 2001 From: Lu Baolu Date: Thu, 4 Jun 2015 20:44:14 +0800 Subject: usb: ulpi: ulpi_init should be executed in subsys_initcall Phy drivers and the ulpi interface providers depend on the registration of the ulpi bus. Ulpi registers the bus in module_init(). This could cause unnecessary bus users' probe delays. i.e. unnecessary -EPROBE_DEFER happening on ulpi_drivers in case they're registered before ulpi bus itself. Reported-by: Zhuo Qiuxu Signed-off-by: Lu Baolu Acked-by: Heikki Krogerus Signed-off-by: Felipe Balbi diff --git a/drivers/usb/common/ulpi.c b/drivers/usb/common/ulpi.c index 0e6f968..01c0c04 100644 --- a/drivers/usb/common/ulpi.c +++ b/drivers/usb/common/ulpi.c @@ -242,7 +242,7 @@ static int __init ulpi_init(void) { return bus_register(&ulpi_bus); } -module_init(ulpi_init); +subsys_initcall(ulpi_init); static void __exit ulpi_exit(void) { -- cgit v0.10.2 From 53e20f2eb161fbe9eea28b54dccc870cec94eca2 Mon Sep 17 00:00:00 2001 From: Alexey Khoroshilov Date: Sun, 19 Jul 2015 23:13:28 +0700 Subject: usb: gadget: mv_udc_core: fix phy_regs I/O memory leak There was an omission in transition to devm_xxx resource handling. iounmap(udc->phy_regs) were removed, but ioremap() was left without devm_. Found by Linux Driver Verification project (linuxtesting.org). Signed-off-by: Alexey Khoroshilov Fixes: 3517c31a8ece6 ("usb: gadget: mv_udc: use devm_xxx for probe") Signed-off-by: Felipe Balbi diff --git a/drivers/usb/gadget/udc/mv_udc_core.c b/drivers/usb/gadget/udc/mv_udc_core.c index d32160d..5da37c9 100644 --- a/drivers/usb/gadget/udc/mv_udc_core.c +++ b/drivers/usb/gadget/udc/mv_udc_core.c @@ -2167,7 +2167,7 @@ static int mv_udc_probe(struct platform_device *pdev) return -ENODEV; } - udc->phy_regs = ioremap(r->start, resource_size(r)); + udc->phy_regs = devm_ioremap(&pdev->dev, r->start, resource_size(r)); if (udc->phy_regs == NULL) { dev_err(&pdev->dev, "failed to map phy I/O memory\n"); return -EBUSY; -- cgit v0.10.2 From 7ace8fc8219e4cbbfd5b4790390d9a01a2541cdf Mon Sep 17 00:00:00 2001 From: Yoshihiro Shimoda Date: Mon, 13 Jul 2015 18:10:05 +0900 Subject: usb: gadget: udc: core: Fix argument of dma_map_single for IOMMU The dma_map_single and dma_unmap_single should set "gadget->dev.parent" instead of "&gadget->dev" in the first argument because the parent has a udc controller's device pointer. Otherwise, iommu functions are not called in ARM environment. Signed-off-by: Yoshihiro Shimoda Signed-off-by: Felipe Balbi diff --git a/drivers/usb/gadget/udc/udc-core.c b/drivers/usb/gadget/udc/udc-core.c index d69c355..362ee8a 100644 --- a/drivers/usb/gadget/udc/udc-core.c +++ b/drivers/usb/gadget/udc/udc-core.c @@ -60,13 +60,15 @@ static DEFINE_MUTEX(udc_lock); int usb_gadget_map_request(struct usb_gadget *gadget, struct usb_request *req, int is_in) { + struct device *dev = gadget->dev.parent; + if (req->length == 0) return 0; if (req->num_sgs) { int mapped; - mapped = dma_map_sg(&gadget->dev, req->sg, req->num_sgs, + mapped = dma_map_sg(dev, req->sg, req->num_sgs, is_in ? DMA_TO_DEVICE : DMA_FROM_DEVICE); if (mapped == 0) { dev_err(&gadget->dev, "failed to map SGs\n"); @@ -75,11 +77,11 @@ int usb_gadget_map_request(struct usb_gadget *gadget, req->num_mapped_sgs = mapped; } else { - req->dma = dma_map_single(&gadget->dev, req->buf, req->length, + req->dma = dma_map_single(dev, req->buf, req->length, is_in ? DMA_TO_DEVICE : DMA_FROM_DEVICE); - if (dma_mapping_error(&gadget->dev, req->dma)) { - dev_err(&gadget->dev, "failed to map buffer\n"); + if (dma_mapping_error(dev, req->dma)) { + dev_err(dev, "failed to map buffer\n"); return -EFAULT; } } @@ -95,12 +97,12 @@ void usb_gadget_unmap_request(struct usb_gadget *gadget, return; if (req->num_mapped_sgs) { - dma_unmap_sg(&gadget->dev, req->sg, req->num_mapped_sgs, + dma_unmap_sg(gadget->dev.parent, req->sg, req->num_mapped_sgs, is_in ? DMA_TO_DEVICE : DMA_FROM_DEVICE); req->num_mapped_sgs = 0; } else { - dma_unmap_single(&gadget->dev, req->dma, req->length, + dma_unmap_single(gadget->dev.parent, req->dma, req->length, is_in ? DMA_TO_DEVICE : DMA_FROM_DEVICE); } } -- cgit v0.10.2 From 75993300d008f418ee2569a632185fc1d7d50674 Mon Sep 17 00:00:00 2001 From: "Michael S. Tsirkin" Date: Wed, 15 Jul 2015 15:26:19 +0300 Subject: virtio_net: don't require ANY_LAYOUT with VERSION_1 ANY_LAYOUT is a compatibility feature. It's implied for VERSION_1 devices, and non-transitional devices might not offer it. Change code to behave accordingly. Signed-off-by: Michael S. Tsirkin Reviewed-by: Paolo Bonzini Reviewed-by: Stefan Hajnoczi Signed-off-by: David S. Miller diff --git a/drivers/net/virtio_net.c b/drivers/net/virtio_net.c index 63c7810..7fbca37 100644 --- a/drivers/net/virtio_net.c +++ b/drivers/net/virtio_net.c @@ -1828,7 +1828,8 @@ static int virtnet_probe(struct virtio_device *vdev) else vi->hdr_len = sizeof(struct virtio_net_hdr); - if (virtio_has_feature(vdev, VIRTIO_F_ANY_LAYOUT)) + if (virtio_has_feature(vdev, VIRTIO_F_ANY_LAYOUT) || + virtio_has_feature(vdev, VIRTIO_F_VERSION_1)) vi->any_header_sg = true; if (virtio_has_feature(vdev, VIRTIO_NET_F_CTRL_VQ)) -- cgit v0.10.2 From 40a7166044a6fce3dbcaa8b5c89845980c218c98 Mon Sep 17 00:00:00 2001 From: Vivien Didelot Date: Wed, 15 Jul 2015 10:07:07 -0400 Subject: net: dsa: mv88e6xxx: fix fid_mask when leaving bridge The mv88e6xxx_priv_state structure contains an fid_mask, where 1 means the FID is free to use, 0 means the FID is in use. This patch fixes the bit clear in mv88e6xxx_leave_bridge() when assigning a new FID to a port. Example scenario: I have 7 ports, port 5 is CPU, port 6 is unused (no PHY). After setting the ports 0, 1 and 2 in bridge br0, and ports 3 and 4 in bridge br1, I have the following fid_mask: 0b111110010110 (0xf96). Indeed, br0 uses FID 0, and br1 uses FID 3. After setting nomaster for port 0, I get the wrong fid_mask: 0b10 (0x2). With this patch we correctly get 0b111110010100 (0xf94), meaning port 0 uses FID 1, br0 uses FID 0, and br1 uses FID 3. Signed-off-by: Vivien Didelot Reviewed-by: Guenter Roeck Signed-off-by: David S. Miller diff --git a/drivers/net/dsa/mv88e6xxx.c b/drivers/net/dsa/mv88e6xxx.c index fd8547c..5613424 100644 --- a/drivers/net/dsa/mv88e6xxx.c +++ b/drivers/net/dsa/mv88e6xxx.c @@ -1163,7 +1163,7 @@ int mv88e6xxx_leave_bridge(struct dsa_switch *ds, int port, u32 br_port_mask) newfid = __ffs(ps->fid_mask); ps->fid[port] = newfid; - ps->fid_mask &= (1 << newfid); + ps->fid_mask &= ~(1 << newfid); ps->bridge_mask[fid] &= ~(1 << port); ps->bridge_mask[newfid] = 1 << port; -- cgit v0.10.2 From 06f6d1094aa0992432b1e2a0920b0ee86ccd83bf Mon Sep 17 00:00:00 2001 From: Nikolay Aleksandrov Date: Wed, 15 Jul 2015 21:52:51 +0200 Subject: bonding: fix destruction of bond with devices different from arphrd_ether When the bonding is being unloaded and the netdevice notifier is unregistered it executes NETDEV_UNREGISTER for each device which should remove the bond's proc entry but if the device enslaved is not of ARPHRD_ETHER type and is in front of the bonding, it may execute bond_release_and_destroy() first which would release the last slave and destroy the bond device leaving the proc entry and thus we will get the following error (with dynamic debug on for bond_netdev_event to see the events order): [ 908.963051] eql: event: 9 [ 908.963052] eql: IFF_SLAVE [ 908.963054] eql: event: 2 [ 908.963056] eql: IFF_SLAVE [ 908.963058] eql: event: 6 [ 908.963059] eql: IFF_SLAVE [ 908.963110] bond0: Releasing active interface eql [ 908.976168] bond0: Destroying bond bond0 [ 908.976266] bond0 (unregistering): Released all slaves [ 908.984097] ------------[ cut here ]------------ [ 908.984107] WARNING: CPU: 0 PID: 1787 at fs/proc/generic.c:575 remove_proc_entry+0x112/0x160() [ 908.984110] remove_proc_entry: removing non-empty directory 'net/bonding', leaking at least 'bond0' [ 908.984111] Modules linked in: bonding(-) eql(O) 9p nfsd auth_rpcgss oid_registry nfs_acl nfs lockd grace fscache sunrpc crct10dif_pclmul crc32_pclmul crc32c_intel ghash_clmulni_intel ppdev qxl drm_kms_helper snd_hda_codec_generic aesni_intel ttm aes_x86_64 glue_helper pcspkr lrw gf128mul ablk_helper cryptd snd_hda_intel virtio_console snd_hda_codec psmouse serio_raw snd_hwdep snd_hda_core 9pnet_virtio 9pnet evdev joydev drm virtio_balloon snd_pcm snd_timer snd soundcore i2c_piix4 i2c_core pvpanic acpi_cpufreq parport_pc parport processor thermal_sys button autofs4 ext4 crc16 mbcache jbd2 hid_generic usbhid hid sg sr_mod cdrom ata_generic virtio_blk virtio_net floppy ata_piix e1000 libata ehci_pci virtio_pci scsi_mod uhci_hcd ehci_hcd virtio_ring virtio usbcore usb_common [last unloaded: bonding] [ 908.984168] CPU: 0 PID: 1787 Comm: rmmod Tainted: G W O 4.2.0-rc2+ #8 [ 908.984170] Hardware name: Bochs Bochs, BIOS Bochs 01/01/2011 [ 908.984172] 0000000000000000 ffffffff81732d41 ffffffff81525b34 ffff8800358dfda8 [ 908.984175] ffffffff8106c521 ffff88003595af78 ffff88003595af40 ffff88003e3a4280 [ 908.984178] ffffffffa058d040 0000000000000000 ffffffff8106c59a ffffffff8172ebd0 [ 908.984181] Call Trace: [ 908.984188] [] ? dump_stack+0x40/0x50 [ 908.984193] [] ? warn_slowpath_common+0x81/0xb0 [ 908.984196] [] ? warn_slowpath_fmt+0x4a/0x50 [ 908.984199] [] ? remove_proc_entry+0x112/0x160 [ 908.984205] [] ? bond_destroy_proc_dir+0x26/0x30 [bonding] [ 908.984208] [] ? bond_net_exit+0x8e/0xa0 [bonding] [ 908.984217] [] ? ops_exit_list.isra.4+0x37/0x70 [ 908.984225] [] ? unregister_pernet_operations+0x8d/0xd0 [ 908.984228] [] ? unregister_pernet_subsys+0x1d/0x30 [ 908.984232] [] ? bonding_exit+0x23/0xdba [bonding] [ 908.984236] [] ? SyS_delete_module+0x18a/0x250 [ 908.984241] [] ? task_work_run+0x89/0xc0 [ 908.984244] [] ? entry_SYSCALL_64_fastpath+0x16/0x75 [ 908.984247] ---[ end trace 7c006ed4abbef24b ]--- Thus remove the proc entry manually if bond_release_and_destroy() is used. Because of the checks in bond_remove_proc_entry() it's not a problem for a bond device to change namespaces (the bug fixed by the Fixes commit) but since commit f9399814927ad ("bonding: Don't allow bond devices to change network namespaces.") that can't happen anyway. Reported-by: Carol Soto Signed-off-by: Nikolay Aleksandrov Fixes: a64d49c3dd50 ("bonding: Manage /proc/net/bonding/ entries from the netdev events") Tested-by: Carol L Soto Signed-off-by: David S. Miller diff --git a/drivers/net/bonding/bond_main.c b/drivers/net/bonding/bond_main.c index 317a494..ec1404e 100644 --- a/drivers/net/bonding/bond_main.c +++ b/drivers/net/bonding/bond_main.c @@ -1916,6 +1916,7 @@ static int bond_release_and_destroy(struct net_device *bond_dev, bond_dev->priv_flags |= IFF_DISABLE_NETPOLL; netdev_info(bond_dev, "Destroying bond %s\n", bond_dev->name); + bond_remove_proc_entry(bond); unregister_netdevice(bond_dev); } return ret; -- cgit v0.10.2 From 8c4b4b0d19880354864f7720ee5e4e7ab11859d2 Mon Sep 17 00:00:00 2001 From: Boris Brezillon Date: Thu, 16 Jul 2015 20:55:34 +0200 Subject: drm: atmel-hlcdc: fix vblank initial state drm_vblank_on() now warns on nested use or if vblank is not properly initialized. This patch fixes Atmel HLCDC vblank initial state. Signed-off-by: Boris Brezillon Reported-by: Sylvain Rochet diff --git a/drivers/gpu/drm/atmel-hlcdc/atmel_hlcdc_crtc.c b/drivers/gpu/drm/atmel-hlcdc/atmel_hlcdc_crtc.c index f69b925..5ae5c69 100644 --- a/drivers/gpu/drm/atmel-hlcdc/atmel_hlcdc_crtc.c +++ b/drivers/gpu/drm/atmel-hlcdc/atmel_hlcdc_crtc.c @@ -355,6 +355,7 @@ int atmel_hlcdc_crtc_create(struct drm_device *dev) planes->overlays[i]->base.possible_crtcs = 1 << crtc->id; drm_crtc_helper_add(&crtc->base, &lcdc_crtc_helper_funcs); + drm_crtc_vblank_reset(&crtc->base); dc->crtc = &crtc->base; diff --git a/drivers/gpu/drm/atmel-hlcdc/atmel_hlcdc_dc.c b/drivers/gpu/drm/atmel-hlcdc/atmel_hlcdc_dc.c index 60b0c13..6fad1f9 100644 --- a/drivers/gpu/drm/atmel-hlcdc/atmel_hlcdc_dc.c +++ b/drivers/gpu/drm/atmel-hlcdc/atmel_hlcdc_dc.c @@ -313,20 +313,20 @@ static int atmel_hlcdc_dc_load(struct drm_device *dev) pm_runtime_enable(dev->dev); - ret = atmel_hlcdc_dc_modeset_init(dev); + ret = drm_vblank_init(dev, 1); if (ret < 0) { - dev_err(dev->dev, "failed to initialize mode setting\n"); + dev_err(dev->dev, "failed to initialize vblank\n"); goto err_periph_clk_disable; } - drm_mode_config_reset(dev); - - ret = drm_vblank_init(dev, 1); + ret = atmel_hlcdc_dc_modeset_init(dev); if (ret < 0) { - dev_err(dev->dev, "failed to initialize vblank\n"); + dev_err(dev->dev, "failed to initialize mode setting\n"); goto err_periph_clk_disable; } + drm_mode_config_reset(dev); + pm_runtime_get_sync(dev->dev); ret = drm_irq_install(dev, dc->hlcdc->irq); pm_runtime_put_sync(dev->dev); -- cgit v0.10.2 From 7d5cd2ce5292b45e555de776cb9e72975a07460d Mon Sep 17 00:00:00 2001 From: Nikolay Aleksandrov Date: Wed, 15 Jul 2015 22:57:01 +0200 Subject: bonding: correctly handle bonding type change on enslave failure If the bond is enslaving a device with different type it will be setup by it, but if after being setup the enslave fails the bond doesn't switch back its type and also keeps pointers to foreign structures that can be long gone. Thus revert back any type changes if the enslave failed and the bond had to change its type. Example: Before patch: $ echo lo > bond0/bonding/slaves -bash: echo: write error: Cannot assign requested address $ ip l sh bond0 20: bond0: mtu 1500 qdisc noop state DOWN mode DEFAULT group default link/loopback 16:54:78:34:bd:41 brd 00:00:00:00:00:00 $ echo +eth1 > bond0/bonding/slaves $ ip l sh bond0 20: bond0: mtu 1500 qdisc noop state DOWN mode DEFAULT group default qlen 1000 link/ether 52:54:00:3f:47:69 brd ff:ff:ff:ff:ff:ff (notice the MASTER flag is gone) After patch: $ echo lo > bond0/bonding/slaves -bash: echo: write error: Cannot assign requested address $ ip l sh bond0 21: bond0: mtu 1500 qdisc noop state DOWN mode DEFAULT group default qlen 1000 link/ether 6e:66:94:f6:07:fc brd ff:ff:ff:ff:ff:ff $ echo +eth1 > bond0/bonding/slaves $ ip l sh bond0 21: bond0: mtu 1500 qdisc noop state DOWN mode DEFAULT group default qlen 1000 link/ether 52:54:00:3f:47:69 brd ff:ff:ff:ff:ff:ff Signed-off-by: Nikolay Aleksandrov Fixes: e36b9d16c6a6 ("bonding: clean muticast addresses when device changes type") Signed-off-by: David S. Miller diff --git a/drivers/net/bonding/bond_main.c b/drivers/net/bonding/bond_main.c index ec1404e..1d26d67 100644 --- a/drivers/net/bonding/bond_main.c +++ b/drivers/net/bonding/bond_main.c @@ -1725,9 +1725,16 @@ err_free: err_undo_flags: /* Enslave of first slave has failed and we need to fix master's mac */ - if (!bond_has_slaves(bond) && - ether_addr_equal_64bits(bond_dev->dev_addr, slave_dev->dev_addr)) - eth_hw_addr_random(bond_dev); + if (!bond_has_slaves(bond)) { + if (ether_addr_equal_64bits(bond_dev->dev_addr, + slave_dev->dev_addr)) + eth_hw_addr_random(bond_dev); + if (bond_dev->type != ARPHRD_ETHER) { + ether_setup(bond_dev); + bond_dev->flags |= IFF_MASTER; + bond_dev->priv_flags &= ~IFF_TX_SKB_SHARING; + } + } return res; } -- cgit v0.10.2 From b8c6cd1d316f3b01ae578d8e29179f6396c0eaa2 Mon Sep 17 00:00:00 2001 From: Florian Fainelli Date: Wed, 15 Jul 2015 16:09:32 -0700 Subject: net: dsa: bcm_sf2: do not use indirect reads and writes for 7445E0 7445E0 contains an ECO which disconnected the internal SF2 pseudo-PHY which was known to conflict with the external pseudo-PHY of BCM53125 switches. This motivated the need to utilize the internal SF2 MDIO controller via indirect register reads/writes to control external Broadcom switches due to this address conflict (both responded at address 30d). For 7445E0, the internal pseudo-PHY of the SF2 switch got disconnected, and as a consequence this prevents the internal SF2 MDIO bus controller from reading data (reads back everything as 0) since the MDI line is tied low. Fix this by making the indirect register reads and writes conditional to 7445D0, on 7445E0 we can utilize the SWITCH_MDIO controller (backed by mdio-unimac and not the DSA created slave MII bus). We utilize of_machine_is_compatible() here since this is the only way for use to differentiate between these two chips in a way that does not violate layers or becomes (too) vendor-specific. Signed-off-by: Florian Fainelli Signed-off-by: David S. Miller diff --git a/drivers/net/dsa/bcm_sf2.c b/drivers/net/dsa/bcm_sf2.c index 972982f..079897b 100644 --- a/drivers/net/dsa/bcm_sf2.c +++ b/drivers/net/dsa/bcm_sf2.c @@ -696,9 +696,20 @@ static int bcm_sf2_sw_setup(struct dsa_switch *ds) } /* Include the pseudo-PHY address and the broadcast PHY address to - * divert reads towards our workaround + * divert reads towards our workaround. This is only required for + * 7445D0, since 7445E0 disconnects the internal switch pseudo-PHY such + * that we can use the regular SWITCH_MDIO master controller instead. + * + * By default, DSA initializes ds->phys_mii_mask to ds->phys_port_mask + * to have a 1:1 mapping between Port address and PHY address in order + * to utilize the slave_mii_bus instance to read from Port PHYs. This is + * not what we want here, so we initialize phys_mii_mask 0 to always + * utilize the "master" MDIO bus backed by the "mdio-unimac" driver. */ - ds->phys_mii_mask |= ((1 << BRCM_PSEUDO_PHY_ADDR) | (1 << 0)); + if (of_machine_is_compatible("brcm,bcm7445d0")) + ds->phys_mii_mask |= ((1 << BRCM_PSEUDO_PHY_ADDR) | (1 << 0)); + else + ds->phys_mii_mask = 0; rev = reg_readl(priv, REG_SWITCH_REVISION); priv->hw_params.top_rev = (rev >> SWITCH_TOP_REV_SHIFT) & -- cgit v0.10.2 From fdbf5b097bbd9693a86c0b8bfdd071a9a2117cfc Mon Sep 17 00:00:00 2001 From: Herbert Xu Date: Mon, 20 Jul 2015 17:55:38 +0800 Subject: Revert "sit: Add gro callbacks to sit_offload" This patch reverts 19424e052fb44da2f00d1a868cbb51f3e9f4bbb5 ("sit: Add gro callbacks to sit_offload") because it generates packets that cannot be handled even by our own GSO. Reported-by: Wolfgang Walter Signed-off-by: Herbert Xu Signed-off-by: David S. Miller diff --git a/net/ipv6/ip6_offload.c b/net/ipv6/ip6_offload.c index e893cd1..08b6204 100644 --- a/net/ipv6/ip6_offload.c +++ b/net/ipv6/ip6_offload.c @@ -292,8 +292,6 @@ static struct packet_offload ipv6_packet_offload __read_mostly = { static const struct net_offload sit_offload = { .callbacks = { .gso_segment = ipv6_gso_segment, - .gro_receive = ipv6_gro_receive, - .gro_complete = ipv6_gro_complete, }, }; -- cgit v0.10.2 From fee50f3c8427aacabfb603229bf0a9056c3f2638 Mon Sep 17 00:00:00 2001 From: Darren Hart Date: Mon, 20 Jul 2015 15:48:37 -0700 Subject: selftests/futex: Fix futex_cmp_requeue_pi() error handling An earlier (pre-kernel-integration) refactoring of this code mistakenly replaced the error condition, <, with a >. Use < to detect an error as opposed to a successful requeue or signal race. Reported-by: David Binderman Cc: Shuah Khan Signed-off-by: Darren Hart Signed-off-by: Shuah Khan diff --git a/tools/testing/selftests/futex/functional/futex_requeue_pi_signal_restart.c b/tools/testing/selftests/futex/functional/futex_requeue_pi_signal_restart.c index 7f0c756..3d7dc6a 100644 --- a/tools/testing/selftests/futex/functional/futex_requeue_pi_signal_restart.c +++ b/tools/testing/selftests/futex/functional/futex_requeue_pi_signal_restart.c @@ -191,7 +191,7 @@ int main(int argc, char *argv[]) if (res > 0) { atomic_set(&requeued, 1); break; - } else if (res > 0) { + } else if (res < 0) { error("FUTEX_CMP_REQUEUE_PI failed\n", errno); ret = RET_ERROR; break; -- cgit v0.10.2 From a951bc1e6ba58f11df5ed5ddc41311e10f5fd20b Mon Sep 17 00:00:00 2001 From: dingtianhong Date: Thu, 16 Jul 2015 16:30:02 +0800 Subject: bonding: correct the MAC address for "follow" fail_over_mac policy The "follow" fail_over_mac policy is useful for multiport devices that either become confused or incur a performance penalty when multiple ports are programmed with the same MAC address, but the same MAC address still may happened by this steps for this policy: 1) echo +eth0 > /sys/class/net/bond0/bonding/slaves bond0 has the same mac address with eth0, it is MAC1. 2) echo +eth1 > /sys/class/net/bond0/bonding/slaves eth1 is backup, eth1 has MAC2. 3) ifconfig eth0 down eth1 became active slave, bond will swap MAC for eth0 and eth1, so eth1 has MAC1, and eth0 has MAC2. 4) ifconfig eth1 down there is no active slave, and eth1 still has MAC1, eth2 has MAC2. 5) ifconfig eth0 up the eth0 became active slave again, the bond set eth0 to MAC1. Something wrong here, then if you set eth1 up, the eth0 and eth1 will have the same MAC address, it will break this policy for ACTIVE_BACKUP mode. This patch will fix this problem by finding the old active slave and swap them MAC address before change active slave. Signed-off-by: Ding Tianhong Tested-by: Nikolay Aleksandrov Signed-off-by: David S. Miller diff --git a/drivers/net/bonding/bond_main.c b/drivers/net/bonding/bond_main.c index 1d26d67..e1ccefc 100644 --- a/drivers/net/bonding/bond_main.c +++ b/drivers/net/bonding/bond_main.c @@ -625,6 +625,23 @@ static void bond_set_dev_addr(struct net_device *bond_dev, call_netdevice_notifiers(NETDEV_CHANGEADDR, bond_dev); } +static struct slave *bond_get_old_active(struct bonding *bond, + struct slave *new_active) +{ + struct slave *slave; + struct list_head *iter; + + bond_for_each_slave(bond, slave, iter) { + if (slave == new_active) + continue; + + if (ether_addr_equal(bond->dev->dev_addr, slave->dev->dev_addr)) + return slave; + } + + return NULL; +} + /* bond_do_fail_over_mac * * Perform special MAC address swapping for fail_over_mac settings @@ -652,6 +669,9 @@ static void bond_do_fail_over_mac(struct bonding *bond, if (!new_active) return; + if (!old_active) + old_active = bond_get_old_active(bond, new_active); + if (old_active) { ether_addr_copy(tmp_mac, new_active->dev->dev_addr); ether_addr_copy(saddr.sa_data, -- cgit v0.10.2 From 1342ff45bd4a43d4ef058aa419399437461f3c88 Mon Sep 17 00:00:00 2001 From: Beata Michalska Date: Mon, 15 Jun 2015 10:12:46 +0200 Subject: ARM: dts: Update video-phy node with syscon phandle for exynos3250 As a follow-up to recent changes to Exynos mipi video phy driver, introducing support for PMU regmap in commit e4b3d38088df ("phy: exynos-video-mipi: Fix regression by adding support for PMU regmap") add a syscon phandle to video-phy node to bring back to life both MIPI DSI display and MIPI CSIS-2 camera sensor on Exynos3250. Signed-off-by: Beata Michalska Reviewed-by: Javier Martinez Canillas Signed-off-by: Krzysztof Kozlowski Signed-off-by: Kukjin Kim diff --git a/arch/arm/boot/dts/exynos3250.dtsi b/arch/arm/boot/dts/exynos3250.dtsi index d720133..2db9943 100644 --- a/arch/arm/boot/dts/exynos3250.dtsi +++ b/arch/arm/boot/dts/exynos3250.dtsi @@ -138,8 +138,8 @@ mipi_phy: video-phy@10020710 { compatible = "samsung,s5pv210-mipi-video-phy"; - reg = <0x10020710 8>; #phy-cells = <1>; + syscon = <&pmu_system_controller>; }; pd_cam: cam-power-domain@10023C00 { -- cgit v0.10.2 From 300bde79f3d86b1af6fcde7b7b2439ecffccbff4 Mon Sep 17 00:00:00 2001 From: Thomas Abraham Date: Fri, 3 Apr 2015 18:43:47 +0200 Subject: ARM: dts: add CPU OPP and regulator supply property for exynos4210 For Exynos4210 platforms, add CPU operating points and CPU regulator supply properties for migrating from Exynos specific cpufreq driver to using generic cpufreq driver. Cc: Doug Anderson Cc: Andreas Faerber Cc: Sachin Kamat Cc: Andreas Farber Cc: Javier Martinez Canillas Signed-off-by: Thomas Abraham [b.zolnierkie: removed exynos5250 and exynos5420 support for now] Signed-off-by: Bartlomiej Zolnierkiewicz [k.kozlowski: Rebased, moved cpu nodes to alphabetical position] Signed-off-by: Krzysztof Kozlowski Signed-off-by: Kukjin Kim diff --git a/arch/arm/boot/dts/exynos4210-origen.dts b/arch/arm/boot/dts/exynos4210-origen.dts index e0abfc3..e050d85 100644 --- a/arch/arm/boot/dts/exynos4210-origen.dts +++ b/arch/arm/boot/dts/exynos4210-origen.dts @@ -127,6 +127,10 @@ }; }; +&cpu0 { + cpu0-supply = <&buck1_reg>; +}; + &fimd { pinctrl-0 = <&lcd_en &lcd_clk &lcd_data24 &pwm0_out>; pinctrl-names = "default"; diff --git a/arch/arm/boot/dts/exynos4210-trats.dts b/arch/arm/boot/dts/exynos4210-trats.dts index 98f3ce6..ba34886 100644 --- a/arch/arm/boot/dts/exynos4210-trats.dts +++ b/arch/arm/boot/dts/exynos4210-trats.dts @@ -188,6 +188,10 @@ }; }; +&cpu0 { + cpu0-supply = <&varm_breg>; +}; + &dsi_0 { vddcore-supply = <&vusb_reg>; vddio-supply = <&vmipi_reg>; diff --git a/arch/arm/boot/dts/exynos4210-universal_c210.dts b/arch/arm/boot/dts/exynos4210-universal_c210.dts index d4f2b11..775892b 100644 --- a/arch/arm/boot/dts/exynos4210-universal_c210.dts +++ b/arch/arm/boot/dts/exynos4210-universal_c210.dts @@ -548,6 +548,10 @@ }; }; +&cpu0 { + cpu0-supply = <&vdd_arm_reg>; +}; + &pinctrl_1 { hdmi_hpd: hdmi-hpd { samsung,pins = "gpx3-7"; diff --git a/arch/arm/boot/dts/exynos4210.dtsi b/arch/arm/boot/dts/exynos4210.dtsi index 10d3c17..3e5ba66 100644 --- a/arch/arm/boot/dts/exynos4210.dtsi +++ b/arch/arm/boot/dts/exynos4210.dtsi @@ -40,6 +40,18 @@ device_type = "cpu"; compatible = "arm,cortex-a9"; reg = <0x900>; + clocks = <&clock CLK_ARM_CLK>; + clock-names = "cpu"; + clock-latency = <160000>; + + operating-points = < + 1200000 1250000 + 1000000 1150000 + 800000 1075000 + 500000 975000 + 400000 975000 + 200000 950000 + >; cooling-min-level = <4>; cooling-max-level = <2>; #cooling-cells = <2>; /* min followed by max */ -- cgit v0.10.2 From 194ac06e39238685abc0eeb436efa82e6571b90f Mon Sep 17 00:00:00 2001 From: "Karicheri, Muralidharan" Date: Thu, 16 Jul 2015 15:32:14 -0400 Subject: net: netcp: fix improper initialization in netcp_ndo_open() The keystone qmss will raise interrupt when packet arrive at the receive queue. Only control available to avoid interrupt from happening is to keep the free descriptor queue (FDQ) empty in the receive side. So the filling of descriptors into the FDQ has to happen after request_irq() call is made as part of knav_queue_enable_notify(). So move the function netcp_rxpool_refill() after this call. Signed-off-by: Murali Karicheri Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/ti/netcp_core.c b/drivers/net/ethernet/ti/netcp_core.c index 5ec4ed3..ec8ed30 100644 --- a/drivers/net/ethernet/ti/netcp_core.c +++ b/drivers/net/ethernet/ti/netcp_core.c @@ -1617,11 +1617,11 @@ static int netcp_ndo_open(struct net_device *ndev) } mutex_unlock(&netcp_modules_lock); - netcp_rxpool_refill(netcp); napi_enable(&netcp->rx_napi); napi_enable(&netcp->tx_napi); knav_queue_enable_notify(netcp->tx_compl_q); knav_queue_enable_notify(netcp->rx_queue); + netcp_rxpool_refill(netcp); netif_tx_wake_all_queues(ndev); dev_dbg(netcp->ndev_dev, "netcp device %s opened\n", ndev->name); return 0; -- cgit v0.10.2 From 06613e38f1f2b098d46e9549756c0d5c040f2ef8 Mon Sep 17 00:00:00 2001 From: Sergei Shtylyov Date: Fri, 17 Jul 2015 00:28:38 +0300 Subject: ravb: fix race updating TCCR The TCCR.TSRQn bit may get clearead after TCCR gets read, so that TCCR write would get skipped. We don't need to check this bit before setting. Signed-off-by: Sergei Shtylyov Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/renesas/ravb_main.c b/drivers/net/ethernet/renesas/ravb_main.c index 0e5fde3..d08c250 100644 --- a/drivers/net/ethernet/renesas/ravb_main.c +++ b/drivers/net/ethernet/renesas/ravb_main.c @@ -1275,7 +1275,6 @@ static netdev_tx_t ravb_start_xmit(struct sk_buff *skb, struct net_device *ndev) u32 dma_addr; void *buffer; u32 entry; - u32 tccr; spin_lock_irqsave(&priv->lock, flags); if (priv->cur_tx[q] - priv->dirty_tx[q] >= priv->num_tx_ring[q]) { @@ -1324,9 +1323,7 @@ static netdev_tx_t ravb_start_xmit(struct sk_buff *skb, struct net_device *ndev) dma_wmb(); desc->die_dt = DT_FSINGLE; - tccr = ravb_read(ndev, TCCR); - if (!(tccr & (TCCR_TSRQ0 << q))) - ravb_write(ndev, tccr | (TCCR_TSRQ0 << q), TCCR); + ravb_write(ndev, ravb_read(ndev, TCCR) | (TCCR_TSRQ0 << q), TCCR); priv->cur_tx[q]++; if (priv->cur_tx[q] - priv->dirty_tx[q] >= priv->num_tx_ring[q] && -- cgit v0.10.2 From e3426ca7bc2957ee072f61360c2b81b4adb629ad Mon Sep 17 00:00:00 2001 From: Reinhard Speyerer Date: Thu, 16 Jul 2015 23:28:14 +0200 Subject: qmi_wwan: add the second QMI/network interface for Sierra Wireless MC7305/MC7355 Sierra Wireless MC7305/MC7355 with USB ID 1199:9041 also provide a second QMI/network interface like the MC73xx with USB ID 1199:68c0 on USB interface #10 when used in the appropriate USB configuration. Add the corresponding QMI_FIXED_INTF entry to the qmi_wwan driver. Please note that the second QMI/network interface is not working for early MC73xx firmware versions like 01.08.x as the device does not respond to QMI messages on the second /dev/cdc-wdm port. Signed-off-by: Reinhard Speyerer Signed-off-by: David S. Miller diff --git a/drivers/net/usb/qmi_wwan.c b/drivers/net/usb/qmi_wwan.c index f603f36..9d43460 100644 --- a/drivers/net/usb/qmi_wwan.c +++ b/drivers/net/usb/qmi_wwan.c @@ -757,6 +757,7 @@ static const struct usb_device_id products[] = { {QMI_FIXED_INTF(0x1199, 0x901c, 8)}, /* Sierra Wireless EM7700 */ {QMI_FIXED_INTF(0x1199, 0x901f, 8)}, /* Sierra Wireless EM7355 */ {QMI_FIXED_INTF(0x1199, 0x9041, 8)}, /* Sierra Wireless MC7305/MC7355 */ + {QMI_FIXED_INTF(0x1199, 0x9041, 10)}, /* Sierra Wireless MC7305/MC7355 */ {QMI_FIXED_INTF(0x1199, 0x9051, 8)}, /* Netgear AirCard 340U */ {QMI_FIXED_INTF(0x1199, 0x9053, 8)}, /* Sierra Wireless Modem */ {QMI_FIXED_INTF(0x1199, 0x9054, 8)}, /* Sierra Wireless Modem */ -- cgit v0.10.2 From bbc03778b9954a2ec93baed63718e4df0192f130 Mon Sep 17 00:00:00 2001 From: Dave Hansen Date: Mon, 20 Jul 2015 16:01:53 -0700 Subject: x86/mm: Add parenthesis for TLB tracepoint size calculation flush_tlb_info->flush_start/end are both normal virtual addresses. When calculating 'nr_pages' (only used for the tracepoint), I neglected to put parenthesis in. Thanks to David Koufaty for pointing this out. Signed-off-by: Dave Hansen Cc: Andrew Morton Cc: Linus Torvalds Cc: Peter Zijlstra Cc: Thomas Gleixner Cc: dave@sr71.net Cc: Link: http://lkml.kernel.org/r/20150720230153.9E834081@viggo.jf.intel.com Signed-off-by: Ingo Molnar diff --git a/arch/x86/mm/tlb.c b/arch/x86/mm/tlb.c index 3250f23..90b924a 100644 --- a/arch/x86/mm/tlb.c +++ b/arch/x86/mm/tlb.c @@ -117,7 +117,7 @@ static void flush_tlb_func(void *info) } else { unsigned long addr; unsigned long nr_pages = - f->flush_end - f->flush_start / PAGE_SIZE; + (f->flush_end - f->flush_start) / PAGE_SIZE; addr = f->flush_start; while (addr < f->flush_end) { __flush_tlb_single(addr); -- cgit v0.10.2 From a89652769470d12cd484ee3d3f7bde0742be8d96 Mon Sep 17 00:00:00 2001 From: "Kirill A. Shutemov" Date: Mon, 20 Jul 2015 14:29:58 -0700 Subject: x86/mpx: Do not set ->vm_ops on MPX VMAs MPX setups private anonymous mapping, but uses vma->vm_ops too. This can confuse core VM, as it relies on vm->vm_ops to distinguish file VMAs from anonymous. As result we will get SIGBUS, because handle_pte_fault() thinks it's file VMA without vm_ops->fault and it doesn't know how to handle the situation properly. Let's fix that by not setting ->vm_ops. We don't really need ->vm_ops here: MPX VMA can be detected with VM_MPX flag. And vma_merge() will not merge MPX VMA with non-MPX VMA, because ->vm_flags won't match. The only thing left is name of VMA. I'm not sure if it's part of ABI, or we can just drop it. The patch keep it by providing arch_vma_name() on x86. Signed-off-by: Kirill A. Shutemov Signed-off-by: Dave Hansen Cc: # Fixes: 6b7339f4 (mm: avoid setting up anonymous pages into file mapping) Cc: Andy Lutomirski Cc: Linus Torvalds Cc: Peter Zijlstra Cc: Thomas Gleixner Cc: dave@sr71.net Link: http://lkml.kernel.org/r/20150720212958.305CC3E9@viggo.jf.intel.com Signed-off-by: Ingo Molnar diff --git a/arch/x86/mm/mmap.c b/arch/x86/mm/mmap.c index 9d518d6..844b06d 100644 --- a/arch/x86/mm/mmap.c +++ b/arch/x86/mm/mmap.c @@ -126,3 +126,10 @@ void arch_pick_mmap_layout(struct mm_struct *mm) mm->get_unmapped_area = arch_get_unmapped_area_topdown; } } + +const char *arch_vma_name(struct vm_area_struct *vma) +{ + if (vma->vm_flags & VM_MPX) + return "[mpx]"; + return NULL; +} diff --git a/arch/x86/mm/mpx.c b/arch/x86/mm/mpx.c index 7a657f5..db1b0bc 100644 --- a/arch/x86/mm/mpx.c +++ b/arch/x86/mm/mpx.c @@ -20,20 +20,6 @@ #define CREATE_TRACE_POINTS #include -static const char *mpx_mapping_name(struct vm_area_struct *vma) -{ - return "[mpx]"; -} - -static struct vm_operations_struct mpx_vma_ops = { - .name = mpx_mapping_name, -}; - -static int is_mpx_vma(struct vm_area_struct *vma) -{ - return (vma->vm_ops == &mpx_vma_ops); -} - static inline unsigned long mpx_bd_size_bytes(struct mm_struct *mm) { if (is_64bit_mm(mm)) @@ -53,9 +39,6 @@ static inline unsigned long mpx_bt_size_bytes(struct mm_struct *mm) /* * This is really a simplified "vm_mmap". it only handles MPX * bounds tables (the bounds directory is user-allocated). - * - * Later on, we use the vma->vm_ops to uniquely identify these - * VMAs. */ static unsigned long mpx_mmap(unsigned long len) { @@ -101,7 +84,6 @@ static unsigned long mpx_mmap(unsigned long len) ret = -ENOMEM; goto out; } - vma->vm_ops = &mpx_vma_ops; if (vm_flags & VM_LOCKED) { up_write(&mm->mmap_sem); @@ -812,7 +794,7 @@ static noinline int zap_bt_entries_mapping(struct mm_struct *mm, * so stop immediately and return an error. This * probably results in a SIGSEGV. */ - if (!is_mpx_vma(vma)) + if (!(vma->vm_flags & VM_MPX)) return -EINVAL; len = min(vma->vm_end, end) - addr; @@ -945,9 +927,9 @@ static int try_unmap_single_bt(struct mm_struct *mm, * lots of tables even though we have no actual table * entries in use. */ - while (next && is_mpx_vma(next)) + while (next && (next->vm_flags & VM_MPX)) next = next->vm_next; - while (prev && is_mpx_vma(prev)) + while (prev && (prev->vm_flags & VM_MPX)) prev = prev->vm_prev; /* * We know 'start' and 'end' lie within an area controlled -- cgit v0.10.2 From 5bc016f1abaa1c5ac0e3af23aa79faec4634a074 Mon Sep 17 00:00:00 2001 From: Jan Beulich Date: Mon, 20 Jul 2015 08:49:01 +0100 Subject: x86/fpu: Disable dependent CPU features on "noxsave" Complete the set of dependent features that need disabling at once: XSAVEC, AVX-512 and all currently known to the kernel extensions to it, as well as MPX need to be disabled too. Signed-off-by: Jan Beulich Cc: Dave Hansen Cc: Linus Torvalds Cc: Peter Zijlstra Cc: Thomas Gleixner Link: http://lkml.kernel.org/r/55ACC40D0200007800092E6C@mail.emea.novell.com Signed-off-by: Ingo Molnar diff --git a/arch/x86/kernel/fpu/init.c b/arch/x86/kernel/fpu/init.c index 0b39173..1e173f6 100644 --- a/arch/x86/kernel/fpu/init.c +++ b/arch/x86/kernel/fpu/init.c @@ -351,9 +351,15 @@ static int __init x86_noxsave_setup(char *s) setup_clear_cpu_cap(X86_FEATURE_XSAVE); setup_clear_cpu_cap(X86_FEATURE_XSAVEOPT); + setup_clear_cpu_cap(X86_FEATURE_XSAVEC); setup_clear_cpu_cap(X86_FEATURE_XSAVES); setup_clear_cpu_cap(X86_FEATURE_AVX); setup_clear_cpu_cap(X86_FEATURE_AVX2); + setup_clear_cpu_cap(X86_FEATURE_AVX512F); + setup_clear_cpu_cap(X86_FEATURE_AVX512PF); + setup_clear_cpu_cap(X86_FEATURE_AVX512ER); + setup_clear_cpu_cap(X86_FEATURE_AVX512CD); + setup_clear_cpu_cap(X86_FEATURE_MPX); return 1; } -- cgit v0.10.2 From ca1fec58bc6a90be96a59b4769e951156846c6ca Mon Sep 17 00:00:00 2001 From: Jan Beulich Date: Mon, 20 Jul 2015 08:46:14 +0100 Subject: x86/mm/pat: Adjust default caching mode translation tables Make WT really mean WT (rather than UC). I can't see why commit 9cd25aac1f ("x86/mm/pat: Emulate PAT when it is disabled") didn't make this to match its changes to pat_init(). Signed-off-by: Jan Beulich Cc: Andy Lutomirski Cc: Borislav Petkov Cc: Linus Torvalds Cc: Peter Zijlstra Cc: Thomas Gleixner Cc: Toshi Kani Link: http://lkml.kernel.org/r/55ACC3660200007800092E62@mail.emea.novell.com Signed-off-by: Ingo Molnar diff --git a/arch/x86/mm/init.c b/arch/x86/mm/init.c index 8533b46..7a45322 100644 --- a/arch/x86/mm/init.c +++ b/arch/x86/mm/init.c @@ -43,18 +43,18 @@ uint16_t __cachemode2pte_tbl[_PAGE_CACHE_MODE_NUM] = { [_PAGE_CACHE_MODE_WC ] = 0 | _PAGE_PCD, [_PAGE_CACHE_MODE_UC_MINUS] = 0 | _PAGE_PCD, [_PAGE_CACHE_MODE_UC ] = _PAGE_PWT | _PAGE_PCD, - [_PAGE_CACHE_MODE_WT ] = 0 | _PAGE_PCD, + [_PAGE_CACHE_MODE_WT ] = _PAGE_PWT | 0, [_PAGE_CACHE_MODE_WP ] = 0 | _PAGE_PCD, }; EXPORT_SYMBOL(__cachemode2pte_tbl); uint8_t __pte2cachemode_tbl[8] = { [__pte2cm_idx( 0 | 0 | 0 )] = _PAGE_CACHE_MODE_WB, - [__pte2cm_idx(_PAGE_PWT | 0 | 0 )] = _PAGE_CACHE_MODE_UC_MINUS, + [__pte2cm_idx(_PAGE_PWT | 0 | 0 )] = _PAGE_CACHE_MODE_WT, [__pte2cm_idx( 0 | _PAGE_PCD | 0 )] = _PAGE_CACHE_MODE_UC_MINUS, [__pte2cm_idx(_PAGE_PWT | _PAGE_PCD | 0 )] = _PAGE_CACHE_MODE_UC, [__pte2cm_idx( 0 | 0 | _PAGE_PAT)] = _PAGE_CACHE_MODE_WB, - [__pte2cm_idx(_PAGE_PWT | 0 | _PAGE_PAT)] = _PAGE_CACHE_MODE_UC_MINUS, + [__pte2cm_idx(_PAGE_PWT | 0 | _PAGE_PAT)] = _PAGE_CACHE_MODE_WT, [__pte2cm_idx(0 | _PAGE_PCD | _PAGE_PAT)] = _PAGE_CACHE_MODE_UC_MINUS, [__pte2cm_idx(_PAGE_PWT | _PAGE_PCD | _PAGE_PAT)] = _PAGE_CACHE_MODE_UC, }; -- cgit v0.10.2 From b8a23e8d8e31abeda2f6cfa36a772414b2a86ffc Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Fri, 17 Jul 2015 10:19:23 +0200 Subject: caif: fix leaks and race in caif_queue_rcv_skb() 1) If sk_filter() is applied, skb was leaked (not freed) 2) Testing SOCK_DEAD twice is racy : packet could be freed while already queued. 3) Remove obsolete comment about caching skb->len Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller diff --git a/net/caif/caif_socket.c b/net/caif/caif_socket.c index 3cc71b9..cc85891 100644 --- a/net/caif/caif_socket.c +++ b/net/caif/caif_socket.c @@ -121,12 +121,13 @@ static void caif_flow_ctrl(struct sock *sk, int mode) * Copied from sock.c:sock_queue_rcv_skb(), but changed so packets are * not dropped, but CAIF is sending flow off instead. */ -static int caif_queue_rcv_skb(struct sock *sk, struct sk_buff *skb) +static void caif_queue_rcv_skb(struct sock *sk, struct sk_buff *skb) { int err; unsigned long flags; struct sk_buff_head *list = &sk->sk_receive_queue; struct caifsock *cf_sk = container_of(sk, struct caifsock, sk); + bool queued = false; if (atomic_read(&sk->sk_rmem_alloc) + skb->truesize >= (unsigned int)sk->sk_rcvbuf && rx_flow_is_on(cf_sk)) { @@ -139,7 +140,8 @@ static int caif_queue_rcv_skb(struct sock *sk, struct sk_buff *skb) err = sk_filter(sk, skb); if (err) - return err; + goto out; + if (!sk_rmem_schedule(sk, skb, skb->truesize) && rx_flow_is_on(cf_sk)) { set_rx_flow_off(cf_sk); net_dbg_ratelimited("sending flow OFF due to rmem_schedule\n"); @@ -147,21 +149,16 @@ static int caif_queue_rcv_skb(struct sock *sk, struct sk_buff *skb) } skb->dev = NULL; skb_set_owner_r(skb, sk); - /* Cache the SKB length before we tack it onto the receive - * queue. Once it is added it no longer belongs to us and - * may be freed by other threads of control pulling packets - * from the queue. - */ spin_lock_irqsave(&list->lock, flags); - if (!sock_flag(sk, SOCK_DEAD)) + queued = !sock_flag(sk, SOCK_DEAD); + if (queued) __skb_queue_tail(list, skb); spin_unlock_irqrestore(&list->lock, flags); - - if (!sock_flag(sk, SOCK_DEAD)) +out: + if (queued) sk->sk_data_ready(sk); else kfree_skb(skb); - return 0; } /* Packet Receive Callback function called from CAIF Stack */ -- cgit v0.10.2 From 8bf4ada2e21378816b28205427ee6b0e1ca4c5f1 Mon Sep 17 00:00:00 2001 From: Konstantin Khlebnikov Date: Fri, 17 Jul 2015 14:01:11 +0300 Subject: net: ratelimit warnings about dst entry refcount underflow or overflow Kernel generates a lot of warnings when dst entry reference counter overflows and becomes negative. That bug was seen several times at machines with outdated 3.10.y kernels. Most like it's already fixed in upstream. Anyway that flood completely kills machine and makes further debugging impossible. Signed-off-by: Konstantin Khlebnikov Signed-off-by: David S. Miller diff --git a/net/core/dst.c b/net/core/dst.c index e956ce6..002144be 100644 --- a/net/core/dst.c +++ b/net/core/dst.c @@ -284,7 +284,9 @@ void dst_release(struct dst_entry *dst) int newrefcnt; newrefcnt = atomic_dec_return(&dst->__refcnt); - WARN_ON(newrefcnt < 0); + if (unlikely(newrefcnt < 0)) + net_warn_ratelimited("%s: dst:%p refcnt:%d\n", + __func__, dst, newrefcnt); if (unlikely(dst->flags & DST_NOCACHE) && !newrefcnt) call_rcu(&dst->rcu_head, dst_destroy_rcu); } -- cgit v0.10.2 From e0536cd910d5bc98300d7830f1d9317df9ab8afa Mon Sep 17 00:00:00 2001 From: Shaohui Xie Date: Fri, 17 Jul 2015 18:07:19 +0800 Subject: net/mdio: fix mdio_bus_match for c45 PHY We store c45 PHY's id information in c45_ids, so it should be used to check the matching between PHY driver and PHY device for c45 PHY. Signed-off-by: Shaohui Xie Signed-off-by: David S. Miller diff --git a/drivers/net/phy/mdio_bus.c b/drivers/net/phy/mdio_bus.c index 095ef3f..46a14cb 100644 --- a/drivers/net/phy/mdio_bus.c +++ b/drivers/net/phy/mdio_bus.c @@ -421,6 +421,8 @@ static int mdio_bus_match(struct device *dev, struct device_driver *drv) { struct phy_device *phydev = to_phy_device(dev); struct phy_driver *phydrv = to_phy_driver(drv); + const int num_ids = ARRAY_SIZE(phydev->c45_ids.device_ids); + int i; if (of_driver_match_device(dev, drv)) return 1; @@ -428,8 +430,21 @@ static int mdio_bus_match(struct device *dev, struct device_driver *drv) if (phydrv->match_phy_device) return phydrv->match_phy_device(phydev); - return (phydrv->phy_id & phydrv->phy_id_mask) == - (phydev->phy_id & phydrv->phy_id_mask); + if (phydev->is_c45) { + for (i = 1; i < num_ids; i++) { + if (!(phydev->c45_ids.devices_in_package & (1 << i))) + continue; + + if ((phydrv->phy_id & phydrv->phy_id_mask) == + (phydev->c45_ids.device_ids[i] & + phydrv->phy_id_mask)) + return 1; + } + return 0; + } else { + return (phydrv->phy_id & phydrv->phy_id_mask) == + (phydev->phy_id & phydrv->phy_id_mask); + } } #ifdef CONFIG_PM -- cgit v0.10.2 From f6bfc46da6292b630ba389592123f0dd02066172 Mon Sep 17 00:00:00 2001 From: Daniel Borkmann Date: Fri, 17 Jul 2015 22:38:43 +0200 Subject: sched: cls_bpf: fix panic on filter replace The following test case causes a NULL pointer dereference in cls_bpf: FOO="1,6 0 0 4294967295," tc filter add dev foo parent 1: bpf bytecode "$FOO" flowid 1:1 action ok tc filter replace dev foo parent 1: pref 49152 handle 0x1 \ bpf bytecode "$FOO" flowid 1:1 action drop The problem is that commit 1f947bf151e9 ("net: sched: rcu'ify cls_bpf") accidentally swapped the arguments of list_replace_rcu(), the old element needs to be the first argument and the new element the second. Fixes: 1f947bf151e9 ("net: sched: rcu'ify cls_bpf") Signed-off-by: Daniel Borkmann Acked-by: John Fastabend Acked-by: Alexei Starovoitov Signed-off-by: David S. Miller diff --git a/net/sched/cls_bpf.c b/net/sched/cls_bpf.c index c79ecfd..e5168f8 100644 --- a/net/sched/cls_bpf.c +++ b/net/sched/cls_bpf.c @@ -378,7 +378,7 @@ static int cls_bpf_change(struct net *net, struct sk_buff *in_skb, goto errout; if (oldprog) { - list_replace_rcu(&prog->link, &oldprog->link); + list_replace_rcu(&oldprog->link, &prog->link); tcf_unbind_filter(tp, &oldprog->res); call_rcu(&oldprog->rcu, __cls_bpf_delete_prog); } else { -- cgit v0.10.2 From ff3532f2655b79058fec035ca54fced69a483084 Mon Sep 17 00:00:00 2001 From: Daniel Borkmann Date: Fri, 17 Jul 2015 22:38:44 +0200 Subject: sched: cls_flower: fix panic on filter replace The following test case causes a NULL pointer dereference in cls_flower: tc filter add dev foo parent 1: flower eth_type ipv4 action ok flowid 1:1 tc filter replace dev foo parent 1: pref 49152 handle 0x1 \ flower eth_type ipv6 action ok flowid 1:1 The problem is that commit 77b9900ef53a ("tc: introduce Flower classifier") accidentally swapped the arguments of list_replace_rcu(), the old element needs to be the first argument and the new element the second. Fixes: 77b9900ef53a ("tc: introduce Flower classifier") Signed-off-by: Daniel Borkmann Acked-by: Jiri Pirko Signed-off-by: David S. Miller diff --git a/net/sched/cls_flower.c b/net/sched/cls_flower.c index 9d37ccd..2f3d03f 100644 --- a/net/sched/cls_flower.c +++ b/net/sched/cls_flower.c @@ -499,7 +499,7 @@ static int fl_change(struct net *net, struct sk_buff *in_skb, *arg = (unsigned long) fnew; if (fold) { - list_replace_rcu(&fnew->list, &fold->list); + list_replace_rcu(&fold->list, &fnew->list); tcf_unbind_filter(tp, &fold->res); call_rcu(&fold->rcu, fl_destroy_filter); } else { -- cgit v0.10.2 From 32b2f4b196b37695fdb42b31afcbc15399d6ef91 Mon Sep 17 00:00:00 2001 From: Daniel Borkmann Date: Fri, 17 Jul 2015 22:38:45 +0200 Subject: sched: cls_flow: fix panic on filter replace The following test case causes a NULL pointer dereference in cls_flow: tc filter add dev foo parent 1: handle 0x1 flow hash keys dst action ok tc filter replace dev foo parent 1: pref 49152 handle 0x1 \ flow hash keys mark action drop To be more precise, actually two different panics are fixed, the first occurs because tcf_exts_init() is not called on the newly allocated filter when we do a replace. And the second panic uncovered after that happens since the arguments of list_replace_rcu() are swapped, the old element needs to be the first argument and the new element the second. Fixes: 70da9f0bf999 ("net: sched: cls_flow use RCU") Signed-off-by: Daniel Borkmann Acked-by: John Fastabend Signed-off-by: David S. Miller diff --git a/net/sched/cls_flow.c b/net/sched/cls_flow.c index 76bc3a2..bb2a0f5 100644 --- a/net/sched/cls_flow.c +++ b/net/sched/cls_flow.c @@ -425,6 +425,8 @@ static int flow_change(struct net *net, struct sk_buff *in_skb, if (!fnew) goto err2; + tcf_exts_init(&fnew->exts, TCA_FLOW_ACT, TCA_FLOW_POLICE); + fold = (struct flow_filter *)*arg; if (fold) { err = -EINVAL; @@ -486,7 +488,6 @@ static int flow_change(struct net *net, struct sk_buff *in_skb, fnew->mask = ~0U; fnew->tp = tp; get_random_bytes(&fnew->hashrnd, 4); - tcf_exts_init(&fnew->exts, TCA_FLOW_ACT, TCA_FLOW_POLICE); } fnew->perturb_timer.function = flow_perturbation; @@ -526,7 +527,7 @@ static int flow_change(struct net *net, struct sk_buff *in_skb, if (*arg == 0) list_add_tail_rcu(&fnew->list, &head->filters); else - list_replace_rcu(&fnew->list, &fold->list); + list_replace_rcu(&fold->list, &fnew->list); *arg = (unsigned long)fnew; -- cgit v0.10.2 From a7a6268590bdc6760df52570a1df41654e3096ba Mon Sep 17 00:00:00 2001 From: Joachim Eastwood Date: Fri, 17 Jul 2015 23:48:17 +0200 Subject: stmmac: fix setting of driver data in stmmac_dvr_probe Commit 803f8fc46274b ("stmmac: move driver data setting into stmmac_dvr_probe") mistakenly set priv and not priv->dev as driver data. This meant that the remove, resume and suspend callbacks that fetched and tried to use this data would most likely explode. Fix the issue by using the correct variable. Fixes: 803f8fc46274b ("stmmac: move driver data setting into stmmac_dvr_probe") Signed-off-by: Joachim Eastwood Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c index 50f7a7a..864b476 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c @@ -2843,7 +2843,7 @@ int stmmac_dvr_probe(struct device *device, if (res->mac) memcpy(priv->dev->dev_addr, res->mac, ETH_ALEN); - dev_set_drvdata(device, priv); + dev_set_drvdata(device, priv->dev); /* Verify driver arguments */ stmmac_verify_args(); -- cgit v0.10.2 From a84e32894191cfcbffa54180d78d7d4654d56c20 Mon Sep 17 00:00:00 2001 From: Simon Guinot Date: Sun, 19 Jul 2015 13:00:53 +0200 Subject: net: mvneta: fix refilling for Rx DMA buffers With the actual code, if a memory allocation error happens while refilling a Rx descriptor, then the original Rx buffer is both passed to the networking stack (in a SKB) and let in the Rx ring. This leads to various kernel oops and crashes. As a fix, this patch moves Rx descriptor refilling ahead of building SKB with the associated Rx buffer. In case of a memory allocation failure, data is dropped and the original DMA buffer is put back into the Rx ring. Signed-off-by: Simon Guinot Fixes: c5aff18204da ("net: mvneta: driver for Marvell Armada 370/XP network unit") Cc: # v3.8+ Tested-by: Yoann Sculo Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/marvell/mvneta.c b/drivers/net/ethernet/marvell/mvneta.c index 370e20ed..62e48bc 100644 --- a/drivers/net/ethernet/marvell/mvneta.c +++ b/drivers/net/ethernet/marvell/mvneta.c @@ -1462,7 +1462,7 @@ static int mvneta_rx(struct mvneta_port *pp, int rx_todo, struct mvneta_rx_queue *rxq) { struct net_device *dev = pp->dev; - int rx_done, rx_filled; + int rx_done; u32 rcvd_pkts = 0; u32 rcvd_bytes = 0; @@ -1473,7 +1473,6 @@ static int mvneta_rx(struct mvneta_port *pp, int rx_todo, rx_todo = rx_done; rx_done = 0; - rx_filled = 0; /* Fairness NAPI loop */ while (rx_done < rx_todo) { @@ -1484,7 +1483,6 @@ static int mvneta_rx(struct mvneta_port *pp, int rx_todo, int rx_bytes, err; rx_done++; - rx_filled++; rx_status = rx_desc->status; rx_bytes = rx_desc->data_size - (ETH_FCS_LEN + MVNETA_MH_SIZE); data = (unsigned char *)rx_desc->buf_cookie; @@ -1524,6 +1522,14 @@ static int mvneta_rx(struct mvneta_port *pp, int rx_todo, continue; } + /* Refill processing */ + err = mvneta_rx_refill(pp, rx_desc); + if (err) { + netdev_err(dev, "Linux processing - Can't refill\n"); + rxq->missed++; + goto err_drop_frame; + } + skb = build_skb(data, pp->frag_size > PAGE_SIZE ? 0 : pp->frag_size); if (!skb) goto err_drop_frame; @@ -1543,14 +1549,6 @@ static int mvneta_rx(struct mvneta_port *pp, int rx_todo, mvneta_rx_csum(pp, rx_status, skb); napi_gro_receive(&pp->napi, skb); - - /* Refill processing */ - err = mvneta_rx_refill(pp, rx_desc); - if (err) { - netdev_err(dev, "Linux processing - Can't refill\n"); - rxq->missed++; - rx_filled--; - } } if (rcvd_pkts) { @@ -1563,7 +1561,7 @@ static int mvneta_rx(struct mvneta_port *pp, int rx_todo, } /* Update rxq management counters */ - mvneta_rxq_desc_num_update(pp, rxq, rx_done, rx_filled); + mvneta_rxq_desc_num_update(pp, rxq, rx_done, rx_done); return rx_done; } -- cgit v0.10.2 From fd0a1b8607ef311a2c800dd54c9a4a3583756ea6 Mon Sep 17 00:00:00 2001 From: "Luis R. Rodriguez" Date: Fri, 17 Jul 2015 14:07:24 -0700 Subject: x86/mm/pat, drivers/infiniband/ipath: Replace WARN() with pr_warn() WARN() may confuse users, fix that. ipath_init_one() is part the device's probe so this would only be triggered if a corresponding device was found. Signed-off-by: Luis R. Rodriguez Acked-by: Doug Ledford Cc: Linus Torvalds Cc: Peter Zijlstra Cc: Thomas Gleixner Cc: andy@silverblocksystems.net Cc: benh@kernel.crashing.org Cc: bp@suse.de Cc: dan.j.williams@intel.com Cc: jkosina@suse.cz Cc: julia.lawall@lip6.fr Cc: luto@amacapital.net Cc: mchehab@osg.samsung.com Link: http://lkml.kernel.org/r/1437167245-28273-2-git-send-email-mcgrof@do-not-panic.com Signed-off-by: Ingo Molnar diff --git a/drivers/infiniband/hw/ipath/ipath_driver.c b/drivers/infiniband/hw/ipath/ipath_driver.c index 2d7e503..871dbe5 100644 --- a/drivers/infiniband/hw/ipath/ipath_driver.c +++ b/drivers/infiniband/hw/ipath/ipath_driver.c @@ -31,6 +31,8 @@ * SOFTWARE. */ +#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt + #include #include #include @@ -399,8 +401,8 @@ static int ipath_init_one(struct pci_dev *pdev, const struct pci_device_id *ent) u32 bar0 = 0, bar1 = 0; #ifdef CONFIG_X86_64 - if (WARN(pat_enabled(), - "ipath needs PAT disabled, boot with nopat kernel parameter\n")) { + if (pat_enabled()) { + pr_warn("ipath needs PAT disabled, boot with nopat kernel parameter\n"); ret = -ENODEV; goto bail; } -- cgit v0.10.2 From f5530d5af835ffa82a0607f5f1977d63ac02551f Mon Sep 17 00:00:00 2001 From: "Luis R. Rodriguez" Date: Fri, 17 Jul 2015 14:07:25 -0700 Subject: x86/mm/pat, drivers/media/ivtv: Move the PAT warning and replace WARN() with pr_warn() On built-in kernels this warning will always splat, even if no ivtvfb hardware is present, as this is part of the module init: if (WARN(pat_enabled(), "ivtvfb needs PAT disabled, boot with nopat kernel parameter\n")) { Fix that by shifting the PAT requirement check out under the code that does the "quasi-probe" for the device. This device driver relies on an existing driver to find its own devices, it looks for that device driver and its own found devices, then uses driver_for_each_device() to try to see if it can probe each of those devices as a frambuffer device with ivtvfb_init_card(). We tuck the PAT requiremenet check then on the ivtvfb_init_card() call making the check at least require an ivtv device present before complaining. Reported-by: Fengguang Wu [0-day test robot] Signed-off-by: Luis R. Rodriguez Cc: Linus Torvalds Cc: Peter Zijlstra Cc: Thomas Gleixner Cc: andy@silverblocksystems.net Cc: benh@kernel.crashing.org Cc: bp@suse.de Cc: dan.j.williams@intel.com Cc: dledford@redhat.com Cc: jkosina@suse.cz Cc: julia.lawall@lip6.fr Cc: luto@amacapital.net Cc: mchehab@osg.samsung.com Link: http://lkml.kernel.org/r/1437167245-28273-3-git-send-email-mcgrof@do-not-panic.com Signed-off-by: Ingo Molnar diff --git a/drivers/media/pci/ivtv/ivtvfb.c b/drivers/media/pci/ivtv/ivtvfb.c index 4cb365d..8b95eef 100644 --- a/drivers/media/pci/ivtv/ivtvfb.c +++ b/drivers/media/pci/ivtv/ivtvfb.c @@ -38,6 +38,8 @@ Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA */ +#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt + #include #include #include @@ -1171,6 +1173,13 @@ static int ivtvfb_init_card(struct ivtv *itv) { int rc; +#ifdef CONFIG_X86_64 + if (pat_enabled()) { + pr_warn("ivtvfb needs PAT disabled, boot with nopat kernel parameter\n"); + return -ENODEV; + } +#endif + if (itv->osd_info) { IVTVFB_ERR("Card %d already initialised\n", ivtvfb_card_id); return -EBUSY; @@ -1265,12 +1274,6 @@ static int __init ivtvfb_init(void) int registered = 0; int err; -#ifdef CONFIG_X86_64 - if (WARN(pat_enabled(), - "ivtvfb needs PAT disabled, boot with nopat kernel parameter\n")) { - return -ENODEV; - } -#endif if (ivtvfb_card_id < -1 || ivtvfb_card_id >= IVTV_MAX_CARDS) { printk(KERN_ERR "ivtvfb: ivtvfb_card_id parameter is out of range (valid range: -1 - %d)\n", -- cgit v0.10.2 From 648a9bc5308d952f2c80772301b339f73026f013 Mon Sep 17 00:00:00 2001 From: Chris Wilson Date: Thu, 16 Jul 2015 12:37:56 +0100 Subject: drm/i915: Use two 32bit reads for select 64bit REG_READ ioctls MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Since the hardware sometimes mysteriously totally flummoxes the 64bit read of a 64bit register when read using a single instruction, split the read into two instructions. Since the read here is of automatically incrementing timestamp counters, we also have to be very careful in order to make sure that it does not increment between the two instructions. However, since userspace tried to workaround this issue and so enshrined this ABI for a broken hardware read and in the process neglected that the read only fails in some environments, we have to introduce a new uABI flag for userspace to request the 2x32 bit accurate read of the timestamp. v2: Fix alignment check and include details of the workaround for userspace. Reported-by: Karol Herbst Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=91317 Testcase: igt/gem_reg_read Signed-off-by: Chris Wilson Cc: MichaÅ‚ Winiarski Cc: stable@vger.kernel.org Tested-by: MichaÅ‚ Winiarski Signed-off-by: Daniel Vetter diff --git a/drivers/gpu/drm/i915/intel_uncore.c b/drivers/gpu/drm/i915/intel_uncore.c index a6d8a3e..260389a 100644 --- a/drivers/gpu/drm/i915/intel_uncore.c +++ b/drivers/gpu/drm/i915/intel_uncore.c @@ -1274,10 +1274,12 @@ int i915_reg_read_ioctl(struct drm_device *dev, struct drm_i915_private *dev_priv = dev->dev_private; struct drm_i915_reg_read *reg = data; struct register_whitelist const *entry = whitelist; + unsigned size; + u64 offset; int i, ret = 0; for (i = 0; i < ARRAY_SIZE(whitelist); i++, entry++) { - if (entry->offset == reg->offset && + if (entry->offset == (reg->offset & -entry->size) && (1 << INTEL_INFO(dev)->gen & entry->gen_bitmask)) break; } @@ -1285,23 +1287,33 @@ int i915_reg_read_ioctl(struct drm_device *dev, if (i == ARRAY_SIZE(whitelist)) return -EINVAL; + /* We use the low bits to encode extra flags as the register should + * be naturally aligned (and those that are not so aligned merely + * limit the available flags for that register). + */ + offset = entry->offset; + size = entry->size; + size |= reg->offset ^ offset; + intel_runtime_pm_get(dev_priv); - switch (entry->size) { + switch (size) { + case 8 | 1: + reg->val = I915_READ64_2x32(offset, offset+4); + break; case 8: - reg->val = I915_READ64(reg->offset); + reg->val = I915_READ64(offset); break; case 4: - reg->val = I915_READ(reg->offset); + reg->val = I915_READ(offset); break; case 2: - reg->val = I915_READ16(reg->offset); + reg->val = I915_READ16(offset); break; case 1: - reg->val = I915_READ8(reg->offset); + reg->val = I915_READ8(offset); break; default: - MISSING_CASE(entry->size); ret = -EINVAL; goto out; } diff --git a/include/uapi/drm/i915_drm.h b/include/uapi/drm/i915_drm.h index 6e1a2ed..db809b7 100644 --- a/include/uapi/drm/i915_drm.h +++ b/include/uapi/drm/i915_drm.h @@ -1070,6 +1070,14 @@ struct drm_i915_reg_read { __u64 offset; __u64 val; /* Return value */ }; +/* Known registers: + * + * Render engine timestamp - 0x2358 + 64bit - gen7+ + * - Note this register returns an invalid value if using the default + * single instruction 8byte read, in order to workaround that use + * offset (0x2538 | 1) instead. + * + */ struct drm_i915_reset_stats { __u32 ctx_id; -- cgit v0.10.2 From 0848f6428ba3a2e42db124d41ac6f548655735bf Mon Sep 17 00:00:00 2001 From: Edward Hyunkoo Jee Date: Tue, 21 Jul 2015 09:43:59 +0200 Subject: inet: frags: fix defragmented packet's IP header for af_packet When ip_frag_queue() computes positions, it assumes that the passed sk_buff does not contain L2 headers. However, when PACKET_FANOUT_FLAG_DEFRAG is used, IP reassembly functions can be called on outgoing packets that contain L2 headers. Also, IPv4 checksum is not corrected after reassembly. Fixes: 7736d33f4262 ("packet: Add pre-defragmentation support for ipv4 fanouts.") Signed-off-by: Edward Hyunkoo Jee Signed-off-by: Eric Dumazet Cc: Willem de Bruijn Cc: Jerry Chu Signed-off-by: David S. Miller diff --git a/net/ipv4/ip_fragment.c b/net/ipv4/ip_fragment.c index a50dc6d..31f71b1 100644 --- a/net/ipv4/ip_fragment.c +++ b/net/ipv4/ip_fragment.c @@ -351,7 +351,7 @@ static int ip_frag_queue(struct ipq *qp, struct sk_buff *skb) ihl = ip_hdrlen(skb); /* Determine the position of this fragment. */ - end = offset + skb->len - ihl; + end = offset + skb->len - skb_network_offset(skb) - ihl; err = -EINVAL; /* Is this the final fragment? */ @@ -381,7 +381,7 @@ static int ip_frag_queue(struct ipq *qp, struct sk_buff *skb) goto err; err = -ENOMEM; - if (!pskb_pull(skb, ihl)) + if (!pskb_pull(skb, skb_network_offset(skb) + ihl)) goto err; err = pskb_trim_rcsum(skb, end - offset); @@ -641,6 +641,8 @@ static int ip_frag_reasm(struct ipq *qp, struct sk_buff *prev, iph->frag_off = 0; } + ip_send_check(iph); + IP_INC_STATS_BH(net, IPSTATS_MIB_REASMOKS); qp->q.fragments = NULL; qp->q.fragments_tail = NULL; -- cgit v0.10.2 From 1e353cddcf81fbb79dd637cab9a22c837e94c205 Mon Sep 17 00:00:00 2001 From: Mugunthan V N Date: Tue, 21 Jul 2015 16:00:42 +0530 Subject: drivers: net: cpsw: remove tx event processing in rx napi poll With commit c03abd84634d ("net: ethernet: cpsw: don't requests IRQs we don't use") common isr and napi are separated into separate tx isr and rx isr/napi, but still in rx napi tx events are handled. So removing the tx event handling in rx napi. Signed-off-by: Mugunthan V N Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/ti/cpsw.c b/drivers/net/ethernet/ti/cpsw.c index f335bf1..d155bf2 100644 --- a/drivers/net/ethernet/ti/cpsw.c +++ b/drivers/net/ethernet/ti/cpsw.c @@ -793,9 +793,7 @@ static irqreturn_t cpsw_rx_interrupt(int irq, void *dev_id) static int cpsw_poll(struct napi_struct *napi, int budget) { struct cpsw_priv *priv = napi_to_priv(napi); - int num_tx, num_rx; - - num_tx = cpdma_chan_process(priv->txch, 128); + int num_rx; num_rx = cpdma_chan_process(priv->rxch, budget); if (num_rx < budget) { @@ -810,9 +808,8 @@ static int cpsw_poll(struct napi_struct *napi, int budget) } } - if (num_rx || num_tx) - cpsw_dbg(priv, intr, "poll %d rx, %d tx pkts\n", - num_rx, num_tx); + if (num_rx) + cpsw_dbg(priv, intr, "poll %d rx pkts\n", num_rx); return num_rx; } -- cgit v0.10.2 From d725e66c06ab440032f49ef17e960896d0ec6d49 Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Tue, 21 Jul 2015 16:06:53 -0700 Subject: Revert "fsnotify: fix oops in fsnotify_clear_marks_by_group_flags()" This reverts commit a2673b6e040663bf16a552f8619e6bde9f4b9acf. Kinglong Mee reports a memory leak with that patch, and Jan Kara confirms: "Thanks for report! You are right that my patch introduces a race between fsnotify kthread and fsnotify_destroy_group() which can result in leaking inotify event on group destruction. I haven't yet decided whether the right fix is not to queue events for dying notification group (as that is pointless anyway) or whether we should just fix the original problem differently... Whenever I look at fsnotify code mark handling I get lost in the maze of locks, lists, and subtle differences between how different notification systems handle notification marks :( I'll think about it over night" and after thinking about it, Jan says: "OK, I have looked into the code some more and I found another relatively simple way of fixing the original oops. It will be IMHO better than trying to fixup this issue which has more potential for breakage. I'll ask Linus to revert the fsnotify fix he already merged and send a new fix" Reported-by: Kinglong Mee Requested-by: Jan Kara Cc: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/notify/mark.c b/fs/notify/mark.c index 3e594ce4..92e48c7 100644 --- a/fs/notify/mark.c +++ b/fs/notify/mark.c @@ -152,15 +152,31 @@ void fsnotify_destroy_mark_locked(struct fsnotify_mark *mark, BUG(); list_del_init(&mark->g_list); + spin_unlock(&mark->lock); if (inode && (mark->flags & FSNOTIFY_MARK_FLAG_OBJECT_PINNED)) iput(inode); + /* release lock temporarily */ + mutex_unlock(&group->mark_mutex); spin_lock(&destroy_lock); list_add(&mark->g_list, &destroy_list); spin_unlock(&destroy_lock); wake_up(&destroy_waitq); + /* + * We don't necessarily have a ref on mark from caller so the above destroy + * may have actually freed it, unless this group provides a 'freeing_mark' + * function which must be holding a reference. + */ + + /* + * Some groups like to know that marks are being freed. This is a + * callback to the group function to let it know that this mark + * is being freed. + */ + if (group->ops->freeing_mark) + group->ops->freeing_mark(mark, group); /* * __fsnotify_update_child_dentry_flags(inode); @@ -175,6 +191,8 @@ void fsnotify_destroy_mark_locked(struct fsnotify_mark *mark, */ atomic_dec(&group->num_marks); + + mutex_lock_nested(&group->mark_mutex, SINGLE_DEPTH_NESTING); } void fsnotify_destroy_mark(struct fsnotify_mark *mark, @@ -187,10 +205,7 @@ void fsnotify_destroy_mark(struct fsnotify_mark *mark, /* * Destroy all marks in the given list. The marks must be already detached from - * the original inode / vfsmount. Note that we can race with - * fsnotify_clear_marks_by_group_flags(). However we hold a reference to each - * mark so they won't get freed from under us and nobody else touches our - * free_list list_head. + * the original inode / vfsmount. */ void fsnotify_destroy_marks(struct list_head *to_free) { @@ -391,7 +406,7 @@ struct fsnotify_mark *fsnotify_find_mark(struct hlist_head *head, } /* - * Clear any marks in a group in which mark->flags & flags is true. + * clear any marks in a group in which mark->flags & flags is true */ void fsnotify_clear_marks_by_group_flags(struct fsnotify_group *group, unsigned int flags) @@ -445,7 +460,6 @@ static int fsnotify_mark_destroy(void *ignored) { struct fsnotify_mark *mark, *next; struct list_head private_destroy_list; - struct fsnotify_group *group; for (;;) { spin_lock(&destroy_lock); @@ -457,14 +471,6 @@ static int fsnotify_mark_destroy(void *ignored) list_for_each_entry_safe(mark, next, &private_destroy_list, g_list) { list_del_init(&mark->g_list); - group = mark->group; - /* - * Some groups like to know that marks are being freed. - * This is a callback to the group function to let it - * know that this mark is being freed. - */ - if (group && group->ops->freeing_mark) - group->ops->freeing_mark(mark, group); fsnotify_put_mark(mark); } -- cgit v0.10.2 From 2d5b569b665ea6d0b15c52529ff06300de81a7ce Mon Sep 17 00:00:00 2001 From: NeilBrown Date: Mon, 6 Jul 2015 12:49:23 +1000 Subject: md/raid5: avoid races when changing cache size. Cache size can grow or shrink due to various pressures at any time. So when we resize the cache as part of a 'grow' operation (i.e. change the size to allow more devices) we need to blocks that automatic growing/shrinking. So introduce a mutex. auto grow/shrink uses mutex_trylock() and just doesn't bother if there is a blockage. Resizing the whole cache holds the mutex to ensure that the correct number of new stripes is allocated. This bug can result in some stripes not being freed when an array is stopped. This leads to the kmem_cache not being freed and a subsequent array can try to use the same kmem_cache and get confused. Fixes: edbe83ab4c27 ("md/raid5: allow the stripe_cache to grow and shrink.") Cc: stable@vger.kernel.org (4.1 - please delay until 2 weeks after release of 4.2) Signed-off-by: NeilBrown diff --git a/drivers/md/raid5.c b/drivers/md/raid5.c index 59e44e9..38300ee 100644 --- a/drivers/md/raid5.c +++ b/drivers/md/raid5.c @@ -2162,6 +2162,9 @@ static int resize_stripes(struct r5conf *conf, int newsize) if (!sc) return -ENOMEM; + /* Need to ensure auto-resizing doesn't interfere */ + mutex_lock(&conf->cache_size_mutex); + for (i = conf->max_nr_stripes; i; i--) { nsh = alloc_stripe(sc, GFP_KERNEL); if (!nsh) @@ -2178,6 +2181,7 @@ static int resize_stripes(struct r5conf *conf, int newsize) kmem_cache_free(sc, nsh); } kmem_cache_destroy(sc); + mutex_unlock(&conf->cache_size_mutex); return -ENOMEM; } /* Step 2 - Must use GFP_NOIO now. @@ -2224,6 +2228,7 @@ static int resize_stripes(struct r5conf *conf, int newsize) } else err = -ENOMEM; + mutex_unlock(&conf->cache_size_mutex); /* Step 4, return new stripes to service */ while(!list_empty(&newstripes)) { nsh = list_entry(newstripes.next, struct stripe_head, lru); @@ -5857,12 +5862,14 @@ static void raid5d(struct md_thread *thread) pr_debug("%d stripes handled\n", handled); spin_unlock_irq(&conf->device_lock); - if (test_and_clear_bit(R5_ALLOC_MORE, &conf->cache_state)) { + if (test_and_clear_bit(R5_ALLOC_MORE, &conf->cache_state) && + mutex_trylock(&conf->cache_size_mutex)) { grow_one_stripe(conf, __GFP_NOWARN); /* Set flag even if allocation failed. This helps * slow down allocation requests when mem is short */ set_bit(R5_DID_ALLOC, &conf->cache_state); + mutex_unlock(&conf->cache_size_mutex); } async_tx_issue_pending_all(); @@ -5894,18 +5901,22 @@ raid5_set_cache_size(struct mddev *mddev, int size) return -EINVAL; conf->min_nr_stripes = size; + mutex_lock(&conf->cache_size_mutex); while (size < conf->max_nr_stripes && drop_one_stripe(conf)) ; + mutex_unlock(&conf->cache_size_mutex); err = md_allow_write(mddev); if (err) return err; + mutex_lock(&conf->cache_size_mutex); while (size > conf->max_nr_stripes) if (!grow_one_stripe(conf, GFP_KERNEL)) break; + mutex_unlock(&conf->cache_size_mutex); return 0; } @@ -6371,11 +6382,18 @@ static unsigned long raid5_cache_scan(struct shrinker *shrink, struct shrink_control *sc) { struct r5conf *conf = container_of(shrink, struct r5conf, shrinker); - int ret = 0; - while (ret < sc->nr_to_scan) { - if (drop_one_stripe(conf) == 0) - return SHRINK_STOP; - ret++; + unsigned long ret = SHRINK_STOP; + + if (mutex_trylock(&conf->cache_size_mutex)) { + ret= 0; + while (ret < sc->nr_to_scan) { + if (drop_one_stripe(conf) == 0) { + ret = SHRINK_STOP; + break; + } + ret++; + } + mutex_unlock(&conf->cache_size_mutex); } return ret; } @@ -6444,6 +6462,7 @@ static struct r5conf *setup_conf(struct mddev *mddev) goto abort; spin_lock_init(&conf->device_lock); seqcount_init(&conf->gen_lock); + mutex_init(&conf->cache_size_mutex); init_waitqueue_head(&conf->wait_for_quiescent); for (i = 0; i < NR_STRIPE_HASH_LOCKS; i++) { init_waitqueue_head(&conf->wait_for_stripe[i]); diff --git a/drivers/md/raid5.h b/drivers/md/raid5.h index 02c3bf8..d051442 100644 --- a/drivers/md/raid5.h +++ b/drivers/md/raid5.h @@ -482,7 +482,8 @@ struct r5conf { */ int active_name; char cache_name[2][32]; - struct kmem_cache *slab_cache; /* for allocating stripes */ + struct kmem_cache *slab_cache; /* for allocating stripes */ + struct mutex cache_size_mutex; /* Protect changes to cache size */ int seq_flush, seq_write; int quiesce; -- cgit v0.10.2 From 299b0685e31c9f3dcc2d58ee3beca761a40b44b3 Mon Sep 17 00:00:00 2001 From: NeilBrown Date: Mon, 6 Jul 2015 17:37:49 +1000 Subject: md/raid10: always set reshape_safe when initializing reshape_position. 'reshape_position' tracks where in the reshape we have reached. 'reshape_safe' tracks where in the reshape we have safely recorded in the metadata. These are compared to determine when to update the metadata. So it is important that reshape_safe is initialised properly. Currently it isn't. When starting a reshape from the beginning it usually has the correct value by luck. But when reducing the number of devices in a RAID10, it has the wrong value and this leads to the metadata not being updated correctly. This can lead to corruption if the reshape is not allowed to complete. This patch is suitable for any -stable kernel which supports RAID10 reshape, which is 3.5 and later. Fixes: 3ea7daa5d7fd ("md/raid10: add reshape support") Cc: stable@vger.kernel.org (v3.5+ please wait for -final to be out for 2 weeks) Signed-off-by: NeilBrown diff --git a/drivers/md/raid10.c b/drivers/md/raid10.c index 940f2f3..38c58e1 100644 --- a/drivers/md/raid10.c +++ b/drivers/md/raid10.c @@ -3556,6 +3556,7 @@ static struct r10conf *setup_conf(struct mddev *mddev) /* far_copies must be 1 */ conf->prev.stride = conf->dev_sectors; } + conf->reshape_safe = conf->reshape_progress; spin_lock_init(&conf->device_lock); INIT_LIST_HEAD(&conf->retry_list); @@ -3760,7 +3761,6 @@ static int run(struct mddev *mddev) } conf->offset_diff = min_offset_diff; - conf->reshape_safe = conf->reshape_progress; clear_bit(MD_RECOVERY_SYNC, &mddev->recovery); clear_bit(MD_RECOVERY_CHECK, &mddev->recovery); set_bit(MD_RECOVERY_RESHAPE, &mddev->recovery); @@ -4103,6 +4103,7 @@ static int raid10_start_reshape(struct mddev *mddev) conf->reshape_progress = size; } else conf->reshape_progress = 0; + conf->reshape_safe = conf->reshape_progress; spin_unlock_irq(&conf->device_lock); if (mddev->delta_disks && mddev->bitmap) { @@ -4170,6 +4171,7 @@ abort: rdev->new_data_offset = rdev->data_offset; smp_wmb(); conf->reshape_progress = MaxSector; + conf->reshape_safe = MaxSector; mddev->reshape_position = MaxSector; spin_unlock_irq(&conf->device_lock); return ret; @@ -4524,6 +4526,7 @@ static void end_reshape(struct r10conf *conf) md_finish_reshape(conf->mddev); smp_wmb(); conf->reshape_progress = MaxSector; + conf->reshape_safe = MaxSector; spin_unlock_irq(&conf->device_lock); /* read-ahead size must cover two whole stripes, which is -- cgit v0.10.2 From ee5d004fd0591536a061451eba2b187092e9127c Mon Sep 17 00:00:00 2001 From: NeilBrown Date: Wed, 22 Jul 2015 10:20:07 +1000 Subject: md: flush ->event_work before stopping array. The 'event_work' worker used by dm-raid may still be running when the array is stopped. This can result in an oops. So flush the workqueue on which it is run after detaching and before destroying the device. Reported-by: Heinz Mauelshagen Signed-off-by: NeilBrown Cc: stable@vger.kernel.org (2.6.38+ please delay 2 weeks after -final release) Fixes: 9d09e663d550 ("dm: raid456 basic support") diff --git a/drivers/md/md.c b/drivers/md/md.c index df92d30..5025b3e 100644 --- a/drivers/md/md.c +++ b/drivers/md/md.c @@ -5382,6 +5382,8 @@ static void __md_stop(struct mddev *mddev) { struct md_personality *pers = mddev->pers; mddev_detach(mddev); + /* Ensure ->event_work is done */ + flush_workqueue(md_misc_wq); spin_lock(&mddev->lock); mddev->ready = 0; mddev->pers = NULL; -- cgit v0.10.2 From 89e478a2aa58af2548b7f316e4d5b6bcc9eade5b Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Wed, 22 Jul 2015 07:02:00 +0200 Subject: tcp: suppress a division by zero warning Andrew Morton reported following warning on one ARM build with gcc-4.4 : net/ipv4/inet_hashtables.c: In function 'inet_ehash_locks_alloc': net/ipv4/inet_hashtables.c:617: warning: division by zero Even guarded with a test on sizeof(spinlock_t), compiler does not like current construct on a !CONFIG_SMP build. Remove the warning by using a temporary variable. Fixes: 095dc8e0c368 ("tcp: fix/cleanup inet_ehash_locks_alloc()") Reported-by: Andrew Morton Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller diff --git a/net/ipv4/inet_hashtables.c b/net/ipv4/inet_hashtables.c index 5f9b063..0cb9165 100644 --- a/net/ipv4/inet_hashtables.c +++ b/net/ipv4/inet_hashtables.c @@ -624,22 +624,21 @@ EXPORT_SYMBOL_GPL(inet_hashinfo_init); int inet_ehash_locks_alloc(struct inet_hashinfo *hashinfo) { + unsigned int locksz = sizeof(spinlock_t); unsigned int i, nblocks = 1; - if (sizeof(spinlock_t) != 0) { + if (locksz != 0) { /* allocate 2 cache lines or at least one spinlock per cpu */ - nblocks = max_t(unsigned int, - 2 * L1_CACHE_BYTES / sizeof(spinlock_t), - 1); + nblocks = max(2U * L1_CACHE_BYTES / locksz, 1U); nblocks = roundup_pow_of_two(nblocks * num_possible_cpus()); /* no more locks than number of hash buckets */ nblocks = min(nblocks, hashinfo->ehash_mask + 1); - hashinfo->ehash_locks = kmalloc_array(nblocks, sizeof(spinlock_t), + hashinfo->ehash_locks = kmalloc_array(nblocks, locksz, GFP_KERNEL | __GFP_NOWARN); if (!hashinfo->ehash_locks) - hashinfo->ehash_locks = vmalloc(nblocks * sizeof(spinlock_t)); + hashinfo->ehash_locks = vmalloc(nblocks * locksz); if (!hashinfo->ehash_locks) return -ENOMEM; -- cgit v0.10.2 From 7aed35cb65348fc8b9ce0c2394ff675e5fc750da Mon Sep 17 00:00:00 2001 From: Nicolas Schichan Date: Tue, 21 Jul 2015 14:14:12 +0200 Subject: ARM: net: fix condition for load_order > 0 when translating load instructions. To check whether the load should take the fast path or not, the code would check that (r_skb_hlen - load_order) is greater than the offset of the access using an "Unsigned higher or same" condition. For halfword accesses and an skb length of 1 at offset 0, that test is valid, as we end up comparing 0xffffffff(-1) and 0, so the fast path is taken and the filter allows the load to wrongly succeed. A similar issue exists for word loads at offset 0 and an skb length of less than 4. Fix that by using the condition "Signed greater than or equal" condition for the fast path code for load orders greater than 0. Signed-off-by: Nicolas Schichan Signed-off-by: David S. Miller diff --git a/arch/arm/net/bpf_jit_32.c b/arch/arm/net/bpf_jit_32.c index 4550d24..21f5ace 100644 --- a/arch/arm/net/bpf_jit_32.c +++ b/arch/arm/net/bpf_jit_32.c @@ -547,7 +547,7 @@ load_common: emit(ARM_SUB_I(r_scratch, r_skb_hl, 1 << load_order), ctx); emit(ARM_CMP_R(r_scratch, r_off), ctx); - condt = ARM_COND_HS; + condt = ARM_COND_GE; } else { emit(ARM_CMP_R(r_skb_hl, r_off), ctx); condt = ARM_COND_HI; -- cgit v0.10.2 From 6d715e301e950e3314d590bdbabf0c26e4fed94b Mon Sep 17 00:00:00 2001 From: Nicolas Schichan Date: Tue, 21 Jul 2015 14:14:13 +0200 Subject: ARM: net: handle negative offsets in BPF JIT. Previously, the JIT would reject negative offsets known during code generation and mishandle negative offsets provided at runtime. Fix that by calling bpf_internal_load_pointer_neg_helper() appropriately in the jit_get_skb_{b,h,w} slow path helpers and by forcing the execution flow to the slow path helpers when the offset is negative. Signed-off-by: Nicolas Schichan Signed-off-by: David S. Miller diff --git a/arch/arm/net/bpf_jit_32.c b/arch/arm/net/bpf_jit_32.c index 21f5ace..d9b2524 100644 --- a/arch/arm/net/bpf_jit_32.c +++ b/arch/arm/net/bpf_jit_32.c @@ -74,32 +74,52 @@ struct jit_ctx { int bpf_jit_enable __read_mostly; -static u64 jit_get_skb_b(struct sk_buff *skb, unsigned offset) +static inline int call_neg_helper(struct sk_buff *skb, int offset, void *ret, + unsigned int size) +{ + void *ptr = bpf_internal_load_pointer_neg_helper(skb, offset, size); + + if (!ptr) + return -EFAULT; + memcpy(ret, ptr, size); + return 0; +} + +static u64 jit_get_skb_b(struct sk_buff *skb, int offset) { u8 ret; int err; - err = skb_copy_bits(skb, offset, &ret, 1); + if (offset < 0) + err = call_neg_helper(skb, offset, &ret, 1); + else + err = skb_copy_bits(skb, offset, &ret, 1); return (u64)err << 32 | ret; } -static u64 jit_get_skb_h(struct sk_buff *skb, unsigned offset) +static u64 jit_get_skb_h(struct sk_buff *skb, int offset) { u16 ret; int err; - err = skb_copy_bits(skb, offset, &ret, 2); + if (offset < 0) + err = call_neg_helper(skb, offset, &ret, 2); + else + err = skb_copy_bits(skb, offset, &ret, 2); return (u64)err << 32 | ntohs(ret); } -static u64 jit_get_skb_w(struct sk_buff *skb, unsigned offset) +static u64 jit_get_skb_w(struct sk_buff *skb, int offset) { u32 ret; int err; - err = skb_copy_bits(skb, offset, &ret, 4); + if (offset < 0) + err = call_neg_helper(skb, offset, &ret, 4); + else + err = skb_copy_bits(skb, offset, &ret, 4); return (u64)err << 32 | ntohl(ret); } @@ -536,9 +556,6 @@ static int build_body(struct jit_ctx *ctx) case BPF_LD | BPF_B | BPF_ABS: load_order = 0; load: - /* the interpreter will deal with the negative K */ - if ((int)k < 0) - return -ENOTSUPP; emit_mov_i(r_off, k, ctx); load_common: ctx->seen |= SEEN_DATA | SEEN_CALL; @@ -553,6 +570,18 @@ load_common: condt = ARM_COND_HI; } + /* + * test for negative offset, only if we are + * currently scheduled to take the fast + * path. this will update the flags so that + * the slowpath instruction are ignored if the + * offset is negative. + * + * for loard_order == 0 the HI condition will + * make loads at offset 0 take the slow path too. + */ + _emit(condt, ARM_CMP_I(r_off, 0), ctx); + _emit(condt, ARM_ADD_R(r_scratch, r_off, r_skb_data), ctx); -- cgit v0.10.2 From c18fe54b3f8d7daa6a43270f82676d6a563582a4 Mon Sep 17 00:00:00 2001 From: Nicolas Schichan Date: Tue, 21 Jul 2015 14:14:14 +0200 Subject: ARM: net: fix vlan access instructions in ARM JIT. This makes BPF_ANC | SKF_AD_VLAN_TAG and BPF_ANC | SKF_AD_VLAN_TAG_PRESENT have the same behaviour as the in kernel VM and makes the test_bpf LD_VLAN_TAG and LD_VLAN_TAG_PRESENT tests pass. Signed-off-by: Nicolas Schichan Signed-off-by: David S. Miller diff --git a/arch/arm/net/bpf_jit_32.c b/arch/arm/net/bpf_jit_32.c index d9b2524..c011e22 100644 --- a/arch/arm/net/bpf_jit_32.c +++ b/arch/arm/net/bpf_jit_32.c @@ -889,9 +889,11 @@ b_epilogue: off = offsetof(struct sk_buff, vlan_tci); emit(ARM_LDRH_I(r_A, r_skb, off), ctx); if (code == (BPF_ANC | SKF_AD_VLAN_TAG)) - OP_IMM3(ARM_AND, r_A, r_A, VLAN_VID_MASK, ctx); - else - OP_IMM3(ARM_AND, r_A, r_A, VLAN_TAG_PRESENT, ctx); + OP_IMM3(ARM_AND, r_A, r_A, ~VLAN_TAG_PRESENT, ctx); + else { + OP_IMM3(ARM_LSR, r_A, r_A, 12, ctx); + OP_IMM3(ARM_AND, r_A, r_A, 0x1, ctx); + } break; case BPF_ANC | SKF_AD_QUEUE: ctx->seen |= SEEN_SKB; -- cgit v0.10.2 From 0470eb99b4721586ccac954faac3fa4472da0845 Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Tue, 21 Jul 2015 16:33:50 +0200 Subject: netlink: don't hold mutex in rcu callback when releasing mmapd ring Kirill A. Shutemov says: This simple test-case trigers few locking asserts in kernel: int main(int argc, char **argv) { unsigned int block_size = 16 * 4096; struct nl_mmap_req req = { .nm_block_size = block_size, .nm_block_nr = 64, .nm_frame_size = 16384, .nm_frame_nr = 64 * block_size / 16384, }; unsigned int ring_size; int fd; fd = socket(AF_NETLINK, SOCK_RAW, NETLINK_GENERIC); if (setsockopt(fd, SOL_NETLINK, NETLINK_RX_RING, &req, sizeof(req)) < 0) exit(1); if (setsockopt(fd, SOL_NETLINK, NETLINK_TX_RING, &req, sizeof(req)) < 0) exit(1); ring_size = req.nm_block_nr * req.nm_block_size; mmap(NULL, 2 * ring_size, PROT_READ|PROT_WRITE, MAP_SHARED, fd, 0); return 0; } +++ exited with 0 +++ BUG: sleeping function called from invalid context at /home/kas/git/public/linux-mm/kernel/locking/mutex.c:616 in_atomic(): 1, irqs_disabled(): 0, pid: 1, name: init 3 locks held by init/1: #0: (reboot_mutex){+.+...}, at: [] SyS_reboot+0xa9/0x220 #1: ((reboot_notifier_list).rwsem){.+.+..}, at: [] __blocking_notifier_call_chain+0x39/0x70 #2: (rcu_callback){......}, at: [] rcu_do_batch.isra.49+0x160/0x10c0 Preemption disabled at:[] __delay+0xf/0x20 CPU: 1 PID: 1 Comm: init Not tainted 4.1.0-00009-gbddf4c4818e0 #253 Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS Debian-1.8.2-1 04/01/2014 ffff88017b3d8000 ffff88027bc03c38 ffffffff81929ceb 0000000000000102 0000000000000000 ffff88027bc03c68 ffffffff81085a9d 0000000000000002 ffffffff81ca2a20 0000000000000268 0000000000000000 ffff88027bc03c98 Call Trace: [] dump_stack+0x4f/0x7b [] ___might_sleep+0x16d/0x270 [] __might_sleep+0x4d/0x90 [] mutex_lock_nested+0x2f/0x430 [] ? _raw_spin_unlock_irqrestore+0x5d/0x80 [] ? __this_cpu_preempt_check+0x13/0x20 [] netlink_set_ring+0x1ed/0x350 [] ? netlink_undo_bind+0x70/0x70 [] netlink_sock_destruct+0x80/0x150 [] __sk_free+0x1d/0x160 [] sk_free+0x19/0x20 [..] Cong Wang says: We can't hold mutex lock in a rcu callback, [..] Thomas Graf says: The socket should be dead at this point. It might be simpler to add a netlink_release_ring() function which doesn't require locking at all. Reported-by: "Kirill A. Shutemov" Diagnosed-by: Cong Wang Suggested-by: Thomas Graf Signed-off-by: Florian Westphal Signed-off-by: David S. Miller diff --git a/net/netlink/af_netlink.c b/net/netlink/af_netlink.c index 9a0ae71..d8e2e39 100644 --- a/net/netlink/af_netlink.c +++ b/net/netlink/af_netlink.c @@ -357,25 +357,52 @@ err1: return NULL; } + +static void +__netlink_set_ring(struct sock *sk, struct nl_mmap_req *req, bool tx_ring, void **pg_vec, + unsigned int order) +{ + struct netlink_sock *nlk = nlk_sk(sk); + struct sk_buff_head *queue; + struct netlink_ring *ring; + + queue = tx_ring ? &sk->sk_write_queue : &sk->sk_receive_queue; + ring = tx_ring ? &nlk->tx_ring : &nlk->rx_ring; + + spin_lock_bh(&queue->lock); + + ring->frame_max = req->nm_frame_nr - 1; + ring->head = 0; + ring->frame_size = req->nm_frame_size; + ring->pg_vec_pages = req->nm_block_size / PAGE_SIZE; + + swap(ring->pg_vec_len, req->nm_block_nr); + swap(ring->pg_vec_order, order); + swap(ring->pg_vec, pg_vec); + + __skb_queue_purge(queue); + spin_unlock_bh(&queue->lock); + + WARN_ON(atomic_read(&nlk->mapped)); + + if (pg_vec) + free_pg_vec(pg_vec, order, req->nm_block_nr); +} + static int netlink_set_ring(struct sock *sk, struct nl_mmap_req *req, - bool closing, bool tx_ring) + bool tx_ring) { struct netlink_sock *nlk = nlk_sk(sk); struct netlink_ring *ring; - struct sk_buff_head *queue; void **pg_vec = NULL; unsigned int order = 0; - int err; ring = tx_ring ? &nlk->tx_ring : &nlk->rx_ring; - queue = tx_ring ? &sk->sk_write_queue : &sk->sk_receive_queue; - if (!closing) { - if (atomic_read(&nlk->mapped)) - return -EBUSY; - if (atomic_read(&ring->pending)) - return -EBUSY; - } + if (atomic_read(&nlk->mapped)) + return -EBUSY; + if (atomic_read(&ring->pending)) + return -EBUSY; if (req->nm_block_nr) { if (ring->pg_vec != NULL) @@ -407,31 +434,19 @@ static int netlink_set_ring(struct sock *sk, struct nl_mmap_req *req, return -EINVAL; } - err = -EBUSY; mutex_lock(&nlk->pg_vec_lock); - if (closing || atomic_read(&nlk->mapped) == 0) { - err = 0; - spin_lock_bh(&queue->lock); - - ring->frame_max = req->nm_frame_nr - 1; - ring->head = 0; - ring->frame_size = req->nm_frame_size; - ring->pg_vec_pages = req->nm_block_size / PAGE_SIZE; - - swap(ring->pg_vec_len, req->nm_block_nr); - swap(ring->pg_vec_order, order); - swap(ring->pg_vec, pg_vec); - - __skb_queue_purge(queue); - spin_unlock_bh(&queue->lock); - - WARN_ON(atomic_read(&nlk->mapped)); + if (atomic_read(&nlk->mapped) == 0) { + __netlink_set_ring(sk, req, tx_ring, pg_vec, order); + mutex_unlock(&nlk->pg_vec_lock); + return 0; } + mutex_unlock(&nlk->pg_vec_lock); if (pg_vec) free_pg_vec(pg_vec, order, req->nm_block_nr); - return err; + + return -EBUSY; } static void netlink_mm_open(struct vm_area_struct *vma) @@ -900,10 +915,10 @@ static void netlink_sock_destruct(struct sock *sk) memset(&req, 0, sizeof(req)); if (nlk->rx_ring.pg_vec) - netlink_set_ring(sk, &req, true, false); + __netlink_set_ring(sk, &req, false, NULL, 0); memset(&req, 0, sizeof(req)); if (nlk->tx_ring.pg_vec) - netlink_set_ring(sk, &req, true, true); + __netlink_set_ring(sk, &req, true, NULL, 0); } #endif /* CONFIG_NETLINK_MMAP */ @@ -2223,7 +2238,7 @@ static int netlink_setsockopt(struct socket *sock, int level, int optname, return -EINVAL; if (copy_from_user(&req, optval, sizeof(req))) return -EFAULT; - err = netlink_set_ring(sk, &req, false, + err = netlink_set_ring(sk, &req, optname == NETLINK_TX_RING); break; } -- cgit v0.10.2 From bac541e46319c2dc6ff9fd5ad6df59cd38686bdb Mon Sep 17 00:00:00 2001 From: Chris J Arges Date: Tue, 21 Jul 2015 12:36:33 -0500 Subject: openvswitch: allocate nr_node_ids flow_stats instead of num_possible_nodes Some architectures like POWER can have a NUMA node_possible_map that contains sparse entries. This causes memory corruption with openvswitch since it allocates flow_cache with a multiple of num_possible_nodes() and assumes the node variable returned by for_each_node will index into flow->stats[node]. Use nr_node_ids to allocate a maximal sparse array instead of num_possible_nodes(). The crash was noticed after 3af229f2 was applied as it changed the node_possible_map to match node_online_map on boot. Fixes: 3af229f2071f5b5cb31664be6109561fbe19c861 Signed-off-by: Chris J Arges Acked-by: Pravin B Shelar Acked-by: Nishanth Aravamudan Signed-off-by: David S. Miller diff --git a/net/openvswitch/flow_table.c b/net/openvswitch/flow_table.c index 4613df8..6552394 100644 --- a/net/openvswitch/flow_table.c +++ b/net/openvswitch/flow_table.c @@ -752,7 +752,7 @@ int ovs_flow_init(void) BUILD_BUG_ON(sizeof(struct sw_flow_key) % sizeof(long)); flow_cache = kmem_cache_create("sw_flow", sizeof(struct sw_flow) - + (num_possible_nodes() + + (nr_node_ids * sizeof(struct flow_stats *)), 0, 0, NULL); if (flow_cache == NULL) -- cgit v0.10.2 From a46fa260f6f5e8f80a725b28e4aee5a04d1bd79e Mon Sep 17 00:00:00 2001 From: Dan Murphy Date: Tue, 21 Jul 2015 12:06:45 -0500 Subject: net: phy: dp83867: Fix warning check for setting the internal delay MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Fix warning: logical ‘or’ of collectively exhaustive tests is always true Change the internal delay check from an 'or' condition to an 'and' condition. Reported-by: David Binderman Signed-off-by: Dan Murphy Acked-by: Florian Fainelli Signed-off-by: David S. Miller diff --git a/drivers/net/phy/dp83867.c b/drivers/net/phy/dp83867.c index c7a12e2..8a3bf54 100644 --- a/drivers/net/phy/dp83867.c +++ b/drivers/net/phy/dp83867.c @@ -164,7 +164,7 @@ static int dp83867_config_init(struct phy_device *phydev) return ret; } - if ((phydev->interface >= PHY_INTERFACE_MODE_RGMII_ID) || + if ((phydev->interface >= PHY_INTERFACE_MODE_RGMII_ID) && (phydev->interface <= PHY_INTERFACE_MODE_RGMII_RXID)) { val = phy_read_mmd_indirect(phydev, DP83867_RGMIICTL, DP83867_DEVADDR, phydev->addr); -- cgit v0.10.2 From d8b48911fd249bc1a3431a9515619403c96d6af3 Mon Sep 17 00:00:00 2001 From: Sergei Shtylyov Date: Wed, 22 Jul 2015 01:31:59 +0300 Subject: ravb: fix ring memory allocation The driver is written as if it can adapt to a low memory situation allocating less RX skbs and TX aligned buffers than the respective RX/TX ring sizes. In reality though the driver would malfunction in this case. Stop being overly smart and just fail in such situation -- this is achieved by moving the memory allocation from ravb_ring_format() to ravb_ring_init(). We leave dma_map_single() calls in place but make their failure non-fatal by marking the corresponding RX descriptors with zero data size which should prevent DMA to an invalid addresses. Signed-off-by: Sergei Shtylyov Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/renesas/ravb_main.c b/drivers/net/ethernet/renesas/ravb_main.c index d08c250..78849dd 100644 --- a/drivers/net/ethernet/renesas/ravb_main.c +++ b/drivers/net/ethernet/renesas/ravb_main.c @@ -228,9 +228,7 @@ static void ravb_ring_format(struct net_device *ndev, int q) struct ravb_desc *desc = NULL; int rx_ring_size = sizeof(*rx_desc) * priv->num_rx_ring[q]; int tx_ring_size = sizeof(*tx_desc) * priv->num_tx_ring[q]; - struct sk_buff *skb; dma_addr_t dma_addr; - void *buffer; int i; priv->cur_rx[q] = 0; @@ -241,41 +239,28 @@ static void ravb_ring_format(struct net_device *ndev, int q) memset(priv->rx_ring[q], 0, rx_ring_size); /* Build RX ring buffer */ for (i = 0; i < priv->num_rx_ring[q]; i++) { - priv->rx_skb[q][i] = NULL; - skb = netdev_alloc_skb(ndev, PKT_BUF_SZ + RAVB_ALIGN - 1); - if (!skb) - break; - ravb_set_buffer_align(skb); /* RX descriptor */ rx_desc = &priv->rx_ring[q][i]; /* The size of the buffer should be on 16-byte boundary. */ rx_desc->ds_cc = cpu_to_le16(ALIGN(PKT_BUF_SZ, 16)); - dma_addr = dma_map_single(&ndev->dev, skb->data, + dma_addr = dma_map_single(&ndev->dev, priv->rx_skb[q][i]->data, ALIGN(PKT_BUF_SZ, 16), DMA_FROM_DEVICE); - if (dma_mapping_error(&ndev->dev, dma_addr)) { - dev_kfree_skb(skb); - break; - } - priv->rx_skb[q][i] = skb; + /* We just set the data size to 0 for a failed mapping which + * should prevent DMA from happening... + */ + if (dma_mapping_error(&ndev->dev, dma_addr)) + rx_desc->ds_cc = cpu_to_le16(0); rx_desc->dptr = cpu_to_le32(dma_addr); rx_desc->die_dt = DT_FEMPTY; } rx_desc = &priv->rx_ring[q][i]; rx_desc->dptr = cpu_to_le32((u32)priv->rx_desc_dma[q]); rx_desc->die_dt = DT_LINKFIX; /* type */ - priv->dirty_rx[q] = (u32)(i - priv->num_rx_ring[q]); memset(priv->tx_ring[q], 0, tx_ring_size); /* Build TX ring buffer */ for (i = 0; i < priv->num_tx_ring[q]; i++) { - priv->tx_skb[q][i] = NULL; - priv->tx_buffers[q][i] = NULL; - buffer = kmalloc(PKT_BUF_SZ + RAVB_ALIGN - 1, GFP_KERNEL); - if (!buffer) - break; - /* Aligned TX buffer */ - priv->tx_buffers[q][i] = buffer; tx_desc = &priv->tx_ring[q][i]; tx_desc->die_dt = DT_EEMPTY; } @@ -298,7 +283,10 @@ static void ravb_ring_format(struct net_device *ndev, int q) static int ravb_ring_init(struct net_device *ndev, int q) { struct ravb_private *priv = netdev_priv(ndev); + struct sk_buff *skb; int ring_size; + void *buffer; + int i; /* Allocate RX and TX skb rings */ priv->rx_skb[q] = kcalloc(priv->num_rx_ring[q], @@ -308,12 +296,28 @@ static int ravb_ring_init(struct net_device *ndev, int q) if (!priv->rx_skb[q] || !priv->tx_skb[q]) goto error; + for (i = 0; i < priv->num_rx_ring[q]; i++) { + skb = netdev_alloc_skb(ndev, PKT_BUF_SZ + RAVB_ALIGN - 1); + if (!skb) + goto error; + ravb_set_buffer_align(skb); + priv->rx_skb[q][i] = skb; + } + /* Allocate rings for the aligned buffers */ priv->tx_buffers[q] = kcalloc(priv->num_tx_ring[q], sizeof(*priv->tx_buffers[q]), GFP_KERNEL); if (!priv->tx_buffers[q]) goto error; + for (i = 0; i < priv->num_tx_ring[q]; i++) { + buffer = kmalloc(PKT_BUF_SZ + RAVB_ALIGN - 1, GFP_KERNEL); + if (!buffer) + goto error; + /* Aligned TX buffer */ + priv->tx_buffers[q][i] = buffer; + } + /* Allocate all RX descriptors. */ ring_size = sizeof(struct ravb_ex_rx_desc) * (priv->num_rx_ring[q] + 1); priv->rx_ring[q] = dma_alloc_coherent(NULL, ring_size, @@ -524,6 +528,10 @@ static bool ravb_rx(struct net_device *ndev, int *quota, int q) if (--boguscnt < 0) break; + /* We use 0-byte descriptors to mark the DMA mapping errors */ + if (!pkt_len) + continue; + if (desc_status & MSC_MC) stats->multicast++; @@ -587,10 +595,11 @@ static bool ravb_rx(struct net_device *ndev, int *quota, int q) le16_to_cpu(desc->ds_cc), DMA_FROM_DEVICE); skb_checksum_none_assert(skb); - if (dma_mapping_error(&ndev->dev, dma_addr)) { - dev_kfree_skb_any(skb); - break; - } + /* We just set the data size to 0 for a failed mapping + * which should prevent DMA from happening... + */ + if (dma_mapping_error(&ndev->dev, dma_addr)) + desc->ds_cc = cpu_to_le16(0); desc->dptr = cpu_to_le32(dma_addr); priv->rx_skb[q][entry] = skb; } -- cgit v0.10.2 From c0d3fce5e192c6b45a9d8e06aecfcec546f73884 Mon Sep 17 00:00:00 2001 From: Will Deacon Date: Mon, 20 Jul 2015 15:14:53 +0100 Subject: arm64: switch_to: calculate cpu context pointer using separate register Commit 0c8c0f03e3a2 ("x86/fpu, sched: Dynamically allocate 'struct fpu'") moved the thread_struct to the bottom of task_struct. As a result, the offset is now too large to be used in an immediate add on arm64 with some kernel configs: arch/arm64/kernel/entry.S: Assembler messages: arch/arm64/kernel/entry.S:588: Error: immediate out of range arch/arm64/kernel/entry.S:597: Error: immediate out of range This patch calculates the offset using an additional register instead of an immediate offset. Fixes: 0c8c0f03e3a2 ("x86/fpu, sched: Dynamically allocate 'struct fpu'") Cc: Dave Hansen Cc: Olof Johansson Cc: Ingo Molnar Signed-off-by: Will Deacon Tested-by: Guenter Roeck Signed-off-by: Catalin Marinas diff --git a/arch/arm64/kernel/entry.S b/arch/arm64/kernel/entry.S index f860bfd..e163518 100644 --- a/arch/arm64/kernel/entry.S +++ b/arch/arm64/kernel/entry.S @@ -585,7 +585,8 @@ ENDPROC(el0_irq) * */ ENTRY(cpu_switch_to) - add x8, x0, #THREAD_CPU_CONTEXT + mov x10, #THREAD_CPU_CONTEXT + add x8, x0, x10 mov x9, sp stp x19, x20, [x8], #16 // store callee-saved registers stp x21, x22, [x8], #16 @@ -594,7 +595,7 @@ ENTRY(cpu_switch_to) stp x27, x28, [x8], #16 stp x29, x9, [x8], #16 str lr, [x8] - add x8, x1, #THREAD_CPU_CONTEXT + add x8, x1, x10 ldp x19, x20, [x8], #16 // restore callee-saved registers ldp x21, x22, [x8], #16 ldp x23, x24, [x8], #16 -- cgit v0.10.2 From 3bc38fc110c642c0a455a5214465bb4d62805a75 Mon Sep 17 00:00:00 2001 From: Jiang Liu Date: Mon, 13 Jul 2015 20:30:04 +0000 Subject: ARM64/irq: Use access helper irq_data_get_affinity_mask() This is a preparatory patch for moving irq_data struct members. Signed-off-by: Jiang Liu Reviewed-by: Hanjun Guo Cc: linux-arm-kernel@lists.infradead.org Signed-off-by: Thomas Gleixner Signed-off-by: Catalin Marinas diff --git a/arch/arm64/kernel/irq.c b/arch/arm64/kernel/irq.c index 240b75c..463fa2e 100644 --- a/arch/arm64/kernel/irq.c +++ b/arch/arm64/kernel/irq.c @@ -61,7 +61,7 @@ void __init init_IRQ(void) static bool migrate_one_irq(struct irq_desc *desc) { struct irq_data *d = irq_desc_get_irq_data(desc); - const struct cpumask *affinity = d->affinity; + const struct cpumask *affinity = irq_data_get_affinity_mask(d); struct irq_chip *c; bool ret = false; @@ -81,7 +81,7 @@ static bool migrate_one_irq(struct irq_desc *desc) if (!c->irq_set_affinity) pr_debug("IRQ%u: unable to set affinity\n", d->irq); else if (c->irq_set_affinity(d, affinity, false) == IRQ_SET_MASK_OK && ret) - cpumask_copy(d->affinity, affinity); + cpumask_copy(irq_data_get_affinity_mask(d), affinity); return ret; } -- cgit v0.10.2 From 5677d67ae3949f09f57357241b88222d49b8c782 Mon Sep 17 00:00:00 2001 From: Daniel Vetter Date: Thu, 16 Jul 2015 16:47:50 +0200 Subject: drm: Stop resetting connector state to unknown MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit It's causing piles of issues since we've stopped forcing full detect cycles in the sysfs interfaces with commit c484f02d0f02fbbfc6decc945a69aae011041a27 Author: Chris Wilson Date: Fri Mar 6 12:36:42 2015 +0000 drm: Lighten sysfs connector 'status' The original justification for this was that the hpd handlers could use the unknown state as a hint to force a full detection. But current i915 code isn't doing that any more, and no one else really uses reset on resume. So instead just keep the old state around. References: http://article.gmane.org/gmane.comp.freedesktop.xorg.drivers.intel/62584 Bugzilla: https://bugzilla.kernel.org/show_bug.cgi?id=100641 Cc: Rui Matos Cc: Julien Wajsberg Cc: kuddel.mail@gmx.de Cc: Lennart Poettering Cc: stable@vger.kernel.org Acked-by: Rob Clark Tested-by: Rui Tiago Cação Matos Signed-off-by: Daniel Vetter diff --git a/drivers/gpu/drm/drm_crtc.c b/drivers/gpu/drm/drm_crtc.c index 357bd04..fed7483 100644 --- a/drivers/gpu/drm/drm_crtc.c +++ b/drivers/gpu/drm/drm_crtc.c @@ -5398,12 +5398,9 @@ void drm_mode_config_reset(struct drm_device *dev) if (encoder->funcs->reset) encoder->funcs->reset(encoder); - list_for_each_entry(connector, &dev->mode_config.connector_list, head) { - connector->status = connector_status_unknown; - + list_for_each_entry(connector, &dev->mode_config.connector_list, head) if (connector->funcs->reset) connector->funcs->reset(connector); - } } EXPORT_SYMBOL(drm_mode_config_reset); -- cgit v0.10.2 From aebda618718157a69c0dc0adb978d69bc2b8723c Mon Sep 17 00:00:00 2001 From: John Youn Date: Mon, 17 Sep 2001 00:00:00 -0700 Subject: usb: dwc3: Reset the transfer resource index on SET_INTERFACE This fixes an issue introduced in commit b23c843992b6 (usb: dwc3: gadget: fix DEPSTARTCFG for non-EP0 EPs) that made sure we would only use DEPSTARTCFG once per SetConfig. The trick is that we should use one DEPSTARTCFG per SetConfig *OR* SetInterface. SetInterface was completely missed from the original patch. This problem became aparent after commit 76e838c9f776 (usb: dwc3: gadget: return error if command sent to DEPCMD register fails) added checking of the return status of device endpoint commands. 'Set Endpoint Transfer Resource' command was caught failing occasionally. This is because the Transfer Resource Index was not getting reset during a SET_INTERFACE request. Finally, to fix the issue, was we have to do is make sure that our start_config_issued flag gets reset whenever we receive a SetInterface request. To verify the problem (and its fix), all we have to do is run test 9 from testusb with 'testusb -t 9 -s 2048 -a -c 5000'. Tested-by: Huang Rui Tested-by: Subbaraya Sundeep Bhatta Fixes: b23c843992b6 (usb: dwc3: gadget: fix DEPSTARTCFG for non-EP0 EPs) Cc: # v3.2+ Signed-off-by: John Youn Signed-off-by: Felipe Balbi diff --git a/drivers/usb/dwc3/ep0.c b/drivers/usb/dwc3/ep0.c index 2ef3c8d..69e769c 100644 --- a/drivers/usb/dwc3/ep0.c +++ b/drivers/usb/dwc3/ep0.c @@ -727,6 +727,10 @@ static int dwc3_ep0_std_request(struct dwc3 *dwc, struct usb_ctrlrequest *ctrl) dwc3_trace(trace_dwc3_ep0, "USB_REQ_SET_ISOCH_DELAY"); ret = dwc3_ep0_set_isoch_delay(dwc, ctrl); break; + case USB_REQ_SET_INTERFACE: + dwc3_trace(trace_dwc3_ep0, "USB_REQ_SET_INTERFACE"); + dwc->start_config_issued = false; + /* Fall through */ default: dwc3_trace(trace_dwc3_ep0, "Forwarding to gadget driver"); ret = dwc3_ep0_delegate_req(dwc, ctrl); -- cgit v0.10.2 From 4b31814d20cbe5cd4ccf18089751e77a04afe4f2 Mon Sep 17 00:00:00 2001 From: Joe Stringer Date: Tue, 21 Jul 2015 21:37:31 -0700 Subject: netfilter: nf_conntrack: Support expectations in different zones When zones were originally introduced, the expectation functions were all extended to perform lookup using the zone. However, insertion was not modified to check the zone. This means that two expectations which are intended to apply for different connections that have the same tuple but exist in different zones cannot both be tracked. Fixes: 5d0aa2ccd4 (netfilter: nf_conntrack: add support for "conntrack zones") Signed-off-by: Joe Stringer Signed-off-by: Pablo Neira Ayuso diff --git a/net/netfilter/nf_conntrack_expect.c b/net/netfilter/nf_conntrack_expect.c index 7a17070..b45a422 100644 --- a/net/netfilter/nf_conntrack_expect.c +++ b/net/netfilter/nf_conntrack_expect.c @@ -219,7 +219,8 @@ static inline int expect_clash(const struct nf_conntrack_expect *a, a->mask.src.u3.all[count] & b->mask.src.u3.all[count]; } - return nf_ct_tuple_mask_cmp(&a->tuple, &b->tuple, &intersect_mask); + return nf_ct_tuple_mask_cmp(&a->tuple, &b->tuple, &intersect_mask) && + nf_ct_zone(a->master) == nf_ct_zone(b->master); } static inline int expect_matches(const struct nf_conntrack_expect *a, -- cgit v0.10.2 From 1c9cf9b211030a454a84cbc1cb15b82d9aa49011 Mon Sep 17 00:00:00 2001 From: Toshi Kani Date: Thu, 16 Jul 2015 17:23:14 -0600 Subject: x86/mm: Move warning from __ioremap_check_ram() to the call site __ioremap_check_ram() has a WARN_ONCE() which is emitted when the given pfn range is not RAM. The warning is bogus in two aspects: - it never triggers since walk_system_ram_range() only calls __ioremap_check_ram() for RAM ranges. - the warning message is wrong as it says: "ioremap on RAM' after it established that the pfn range is not RAM. Move the WARN_ONCE() to __ioremap_caller(), and update the message to include the address range so we get an actual warning when something tries to ioremap system RAM. [ tglx: Massaged changelog ] Signed-off-by: Toshi Kani Reviewed-by: Dan Williams Cc: Roland Dreier Cc: Luis R. Rodriguez Cc: H. Peter Anvin Cc: Ingo Molnar Cc: Borislav Petkov Cc: linux-mm@kvack.org Link: http://lkml.kernel.org/r/1437088996-28511-2-git-send-email-toshi.kani@hp.com Signed-off-by: Thomas Gleixner diff --git a/arch/x86/mm/ioremap.c b/arch/x86/mm/ioremap.c index cc5ccc4..fd3df0d 100644 --- a/arch/x86/mm/ioremap.c +++ b/arch/x86/mm/ioremap.c @@ -63,8 +63,6 @@ static int __ioremap_check_ram(unsigned long start_pfn, unsigned long nr_pages, !PageReserved(pfn_to_page(start_pfn + i))) return 1; - WARN_ONCE(1, "ioremap on RAM pfn 0x%lx\n", start_pfn); - return 0; } @@ -131,8 +129,11 @@ static void __iomem *__ioremap_caller(resource_size_t phys_addr, pfn = phys_addr >> PAGE_SHIFT; last_pfn = last_addr >> PAGE_SHIFT; if (walk_system_ram_range(pfn, last_pfn - pfn + 1, NULL, - __ioremap_check_ram) == 1) + __ioremap_check_ram) == 1) { + WARN_ONCE(1, "ioremap on RAM at 0x%llx - 0x%llx\n", + phys_addr, last_addr); return NULL; + } } /* * Mappings have to be page-aligned -- cgit v0.10.2 From 9a58eebe1ace609bedf8c5a65e70a097459f5696 Mon Sep 17 00:00:00 2001 From: Toshi Kani Date: Thu, 16 Jul 2015 17:23:15 -0600 Subject: x86/mm: Remove region_is_ram() call from ioremap __ioremap_caller() calls region_is_ram() to walk through the iomem_resource table to check if a target range is in RAM, which was added to improve the lookup performance over page_is_ram() (commit 906e36c5c717 "x86: use optimized ioresource lookup in ioremap function"). page_is_ram() was no longer used when this change was added, though. __ioremap_caller() then calls walk_system_ram_range(), which had replaced page_is_ram() to improve the lookup performance (commit c81c8a1eeede "x86, ioremap: Speed up check for RAM pages"). Since both checks walk through the same iomem_resource table for the same purpose, there is no need to call both functions. Aside of that walk_system_ram_range() is the only useful check at the moment because region_is_ram() always returns -1 due to an implementation bug. That bug in region_is_ram() cannot be fixed without breaking existing ioremap callers, which rely on the subtle difference of walk_system_ram_range() versus non page aligned ranges. Once these offending callers are fixed we can use region_is_ram() and remove walk_system_ram_range(). [ tglx: Massaged changelog ] Signed-off-by: Toshi Kani Reviewed-by: Dan Williams Cc: Roland Dreier Cc: Mike Travis Cc: Luis R. Rodriguez Cc: H. Peter Anvin Cc: Ingo Molnar Cc: Borislav Petkov Cc: linux-mm@kvack.org Link: http://lkml.kernel.org/r/1437088996-28511-3-git-send-email-toshi.kani@hp.com Signed-off-by: Thomas Gleixner diff --git a/arch/x86/mm/ioremap.c b/arch/x86/mm/ioremap.c index fd3df0d..b9d4a33 100644 --- a/arch/x86/mm/ioremap.c +++ b/arch/x86/mm/ioremap.c @@ -92,7 +92,6 @@ static void __iomem *__ioremap_caller(resource_size_t phys_addr, pgprot_t prot; int retval; void __iomem *ret_addr; - int ram_region; /* Don't allow wraparound or zero size */ last_addr = phys_addr + size - 1; @@ -115,26 +114,15 @@ static void __iomem *__ioremap_caller(resource_size_t phys_addr, /* * Don't allow anybody to remap normal RAM that we're using.. */ - /* First check if whole region can be identified as RAM or not */ - ram_region = region_is_ram(phys_addr, size); - if (ram_region > 0) { - WARN_ONCE(1, "ioremap on RAM at 0x%lx - 0x%lx\n", - (unsigned long int)phys_addr, - (unsigned long int)last_addr); - return NULL; - } - - /* If could not be identified(-1), check page by page */ - if (ram_region < 0) { - pfn = phys_addr >> PAGE_SHIFT; - last_pfn = last_addr >> PAGE_SHIFT; - if (walk_system_ram_range(pfn, last_pfn - pfn + 1, NULL, + pfn = phys_addr >> PAGE_SHIFT; + last_pfn = last_addr >> PAGE_SHIFT; + if (walk_system_ram_range(pfn, last_pfn - pfn + 1, NULL, __ioremap_check_ram) == 1) { - WARN_ONCE(1, "ioremap on RAM at 0x%llx - 0x%llx\n", + WARN_ONCE(1, "ioremap on RAM at 0x%llx - 0x%llx\n", phys_addr, last_addr); - return NULL; - } + return NULL; } + /* * Mappings have to be page-aligned */ -- cgit v0.10.2 From 8c38de992be9aed0b34c4fab8f972c83d3b00dc4 Mon Sep 17 00:00:00 2001 From: Toshi Kani Date: Thu, 16 Jul 2015 17:23:16 -0600 Subject: mm: Fix bugs in region_is_ram() region_is_ram() looks up the iomem_resource table to check if a target range is in RAM. However, it always returns with -1 due to invalid range checks. It always breaks the loop at the first entry of the table. Another issue is that it compares p->flags and flags, but it always fails. flags is declared as int, which makes it as a negative value with IORESOURCE_BUSY (0x80000000) set while p->flags is unsigned long. Fix the range check and flags so that region_is_ram() works as advertised. Signed-off-by: Toshi Kani Reviewed-by: Dan Williams Cc: Mike Travis Cc: Luis R. Rodriguez Cc: Andrew Morton Cc: Roland Dreier Cc: linux-mm@kvack.org Link: http://lkml.kernel.org/r/1437088996-28511-4-git-send-email-toshi.kani@hp.com Signed-off-by: Thomas Gleixner diff --git a/kernel/resource.c b/kernel/resource.c index 90552aa..fed052a 100644 --- a/kernel/resource.c +++ b/kernel/resource.c @@ -504,13 +504,13 @@ int region_is_ram(resource_size_t start, unsigned long size) { struct resource *p; resource_size_t end = start + size - 1; - int flags = IORESOURCE_MEM | IORESOURCE_BUSY; + unsigned long flags = IORESOURCE_MEM | IORESOURCE_BUSY; const char *name = "System RAM"; int ret = -1; read_lock(&resource_lock); for (p = iomem_resource.child; p ; p = p->sibling) { - if (end < p->start) + if (p->end < start) continue; if (p->start <= start && end <= p->end) { @@ -521,7 +521,7 @@ int region_is_ram(resource_size_t start, unsigned long size) ret = 1; break; } - if (p->end < start) + if (end < p->start) break; /* not found */ } read_unlock(&resource_lock); -- cgit v0.10.2 From 21974061cfb3c4b0b1a83447fb5e7cdcd06e56dc Mon Sep 17 00:00:00 2001 From: Mike Krinkin Date: Sun, 19 Jul 2015 09:53:17 +0300 Subject: null_blk: fix use-after-free problem end_cmd finishes request associated with nullb_cmd struct, so we should save pointer to request_queue in a local variable before calling end_cmd. The problem was causes general protection fault with slab poisoning enabled. Fixes: 8b70f45e2eb2 ("null_blk: restart request processing on completion handler") Tested-by: Akinobu Mita Signed-off-by: Mike Krinkin Signed-off-by: Jens Axboe diff --git a/drivers/block/null_blk.c b/drivers/block/null_blk.c index 69de41a..3177b24 100644 --- a/drivers/block/null_blk.c +++ b/drivers/block/null_blk.c @@ -240,19 +240,19 @@ static enum hrtimer_restart null_cmd_timer_expired(struct hrtimer *timer) while ((entry = llist_del_all(&cq->list)) != NULL) { entry = llist_reverse_order(entry); do { + struct request_queue *q = NULL; + cmd = container_of(entry, struct nullb_cmd, ll_list); entry = entry->next; + if (cmd->rq) + q = cmd->rq->q; end_cmd(cmd); - if (cmd->rq) { - struct request_queue *q = cmd->rq->q; - - if (!q->mq_ops && blk_queue_stopped(q)) { - spin_lock(q->queue_lock); - if (blk_queue_stopped(q)) - blk_start_queue(q); - spin_unlock(q->queue_lock); - } + if (q && !q->mq_ops && blk_queue_stopped(q)) { + spin_lock(q->queue_lock); + if (blk_queue_stopped(q)) + blk_start_queue(q); + spin_unlock(q->queue_lock); } } while (entry); } -- cgit v0.10.2 From 68514471ceceac63c7fa9ad684d882f41be5b2d8 Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Wed, 22 Jul 2015 16:31:17 -0400 Subject: SUNRPC: Fix a backchannel deadlock xprt_alloc_bc_request() cannot call xprt_free_bc_request() without deadlocking, since it already holds the xprt->bc_pa_lock. Reported-by: Chuck Lever Fixes: 0d2a970d0ae55 ("SUNRPC: Fix a backchannel race") Signed-off-by: Trond Myklebust diff --git a/net/sunrpc/backchannel_rqst.c b/net/sunrpc/backchannel_rqst.c index 9825ff0..5a3b50a 100644 --- a/net/sunrpc/backchannel_rqst.c +++ b/net/sunrpc/backchannel_rqst.c @@ -240,8 +240,8 @@ static struct rpc_rqst *xprt_alloc_bc_request(struct rpc_xprt *xprt, __be32 xid) req = xprt_alloc_bc_req(xprt, GFP_ATOMIC); if (!req) goto not_found; - /* Note: this 'free' request adds it to xprt->bc_pa_list */ - xprt_free_bc_request(req); + list_add_tail(&req->rq_bc_pa_list, &xprt->bc_pa_list); + xprt->bc_alloc_count++; } req = list_first_entry(&xprt->bc_pa_list, struct rpc_rqst, rq_bc_pa_list); -- cgit v0.10.2 From 1980bd4d829a87ccd21b949f8a11ff1b426f5b0b Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Wed, 22 Jul 2015 17:05:32 -0400 Subject: SUNRPC: xprt_complete_bc_request must also decrement the free slot count Calling xprt_complete_bc_request() effectively causes the slot to be allocated, so it needs to decrement the backchannel free slot count as well. Fixes: 0d2a970d0ae5 ("SUNRPC: Fix a backchannel race") Signed-off-by: Trond Myklebust diff --git a/net/sunrpc/backchannel_rqst.c b/net/sunrpc/backchannel_rqst.c index 5a3b50a..6255d14 100644 --- a/net/sunrpc/backchannel_rqst.c +++ b/net/sunrpc/backchannel_rqst.c @@ -336,7 +336,7 @@ void xprt_complete_bc_request(struct rpc_rqst *req, uint32_t copied) spin_lock(&xprt->bc_pa_lock); list_del(&req->rq_bc_pa_list); - xprt->bc_alloc_count--; + xprt_dec_alloc_count(xprt, 1); spin_unlock(&xprt->bc_pa_lock); req->rq_private_buf.len = copied; -- cgit v0.10.2 From 3c38cbe2ade88240fabb585b408f779ad3b9a31b Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Wed, 22 Jul 2015 13:46:13 -0400 Subject: NFSv4: We must set NFS_OPEN_STATE flag in nfs_resync_open_stateid_locked Otherwise, nfs4_select_rw_stateid() will always return the zero stateid instead of the correct open stateid. Fixes: f95549cf24660 ("NFSv4: More CLOSE/OPEN races") Cc: stable@vger.kernel.org # 4.0+ Signed-off-by: Trond Myklebust diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c index c5c9e0d..9264994 100644 --- a/fs/nfs/nfs4proc.c +++ b/fs/nfs/nfs4proc.c @@ -1200,12 +1200,15 @@ static bool nfs_need_update_open_stateid(struct nfs4_state *state, static void nfs_resync_open_stateid_locked(struct nfs4_state *state) { + if (!(state->n_wronly || state->n_rdonly || state->n_rdwr)) + return; if (state->n_wronly) set_bit(NFS_O_WRONLY_STATE, &state->flags); if (state->n_rdonly) set_bit(NFS_O_RDONLY_STATE, &state->flags); if (state->n_rdwr) set_bit(NFS_O_RDWR_STATE, &state->flags); + set_bit(NFS_OPEN_STATE, &state->flags); } static void nfs_clear_open_stateid_locked(struct nfs4_state *state, -- cgit v0.10.2 From 2b83d3de4c18af49800e0b26ae013db4fcf43a4a Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Sun, 5 Jul 2015 20:06:38 -0400 Subject: NFSv4/pnfs: Ensure we don't miss a file extension pNFS writes don't return attributes, however that doesn't mean that we should ignore the fact that they may be extending the file. This patch ensures that if a write is seen to extend the file, then we always set an attribute barrier, and update the cached file size. Signed-off-by: Trond Myklebust diff --git a/fs/nfs/write.c b/fs/nfs/write.c index 359e9ad..0e6a2b8 100644 --- a/fs/nfs/write.c +++ b/fs/nfs/write.c @@ -1378,24 +1378,27 @@ static void nfs_writeback_check_extend(struct nfs_pgio_header *hdr, { struct nfs_pgio_args *argp = &hdr->args; struct nfs_pgio_res *resp = &hdr->res; + u64 size = argp->offset + resp->count; if (!(fattr->valid & NFS_ATTR_FATTR_SIZE)) + fattr->size = size; + if (nfs_size_to_loff_t(fattr->size) < i_size_read(hdr->inode)) { + fattr->valid &= ~NFS_ATTR_FATTR_SIZE; return; - if (argp->offset + resp->count != fattr->size) - return; - if (nfs_size_to_loff_t(fattr->size) < i_size_read(hdr->inode)) + } + if (size != fattr->size) return; /* Set attribute barrier */ nfs_fattr_set_barrier(fattr); + /* ...and update size */ + fattr->valid |= NFS_ATTR_FATTR_SIZE; } void nfs_writeback_update_inode(struct nfs_pgio_header *hdr) { - struct nfs_fattr *fattr = hdr->res.fattr; + struct nfs_fattr *fattr = &hdr->fattr; struct inode *inode = hdr->inode; - if (fattr == NULL) - return; spin_lock(&inode->i_lock); nfs_writeback_check_extend(hdr, fattr); nfs_post_op_update_inode_force_wcc_locked(inode, fattr); -- cgit v0.10.2 From 85a23cee3f2c928475f31777ead5a71340a12fc3 Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Sun, 5 Jul 2015 11:02:53 -0400 Subject: NFS: Don't revalidate the mapping if both size and change attr are up to date If we've ensured that the size and the change attribute are both correct, then there is no point in marking those attributes as needing revalidation again. Only do so if we know the size is incorrect and was not updated. Fixes: f2467b6f64da ("NFS: Clear NFS_INO_REVAL_PAGECACHE when...") Signed-off-by: Trond Myklebust diff --git a/fs/nfs/inode.c b/fs/nfs/inode.c index b77b328..d654661 100644 --- a/fs/nfs/inode.c +++ b/fs/nfs/inode.c @@ -1244,9 +1244,11 @@ static int nfs_check_inode_attributes(struct inode *inode, struct nfs_fattr *fat if (fattr->valid & NFS_ATTR_FATTR_SIZE) { cur_size = i_size_read(inode); new_isize = nfs_size_to_loff_t(fattr->size); - if (cur_size != new_isize && nfsi->nrequests == 0) + if (cur_size != new_isize) invalid |= NFS_INO_INVALID_ATTR|NFS_INO_REVAL_PAGECACHE; } + if (nfsi->nrequests != 0) + invalid &= ~NFS_INO_REVAL_PAGECACHE; /* Have any file permissions changed? */ if ((fattr->valid & NFS_ATTR_FATTR_MODE) && (inode->i_mode & S_IALLUGO) != (fattr->mode & S_IALLUGO)) @@ -1684,8 +1686,7 @@ static int nfs_update_inode(struct inode *inode, struct nfs_fattr *fattr) invalid |= NFS_INO_INVALID_ATTR | NFS_INO_INVALID_DATA | NFS_INO_INVALID_ACCESS - | NFS_INO_INVALID_ACL - | NFS_INO_REVAL_PAGECACHE; + | NFS_INO_INVALID_ACL; if (S_ISDIR(inode->i_mode)) nfs_force_lookup_revalidate(inode); inode->i_version = fattr->change_attr; @@ -1717,7 +1718,6 @@ static int nfs_update_inode(struct inode *inode, struct nfs_fattr *fattr) if ((nfsi->nrequests == 0) || new_isize > cur_isize) { i_size_write(inode, new_isize); invalid |= NFS_INO_INVALID_ATTR|NFS_INO_INVALID_DATA; - invalid &= ~NFS_INO_REVAL_PAGECACHE; } dprintk("NFS: isize change on server for file %s/%ld " "(%Ld to %Ld)\n", -- cgit v0.10.2 From 5c675d6420511e035c150e420ab26d0306bbb736 Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Sun, 5 Jul 2015 16:07:04 -0400 Subject: NFS: Set NFS_INO_REVAL_PAGECACHE if the change attribute is uninitialised We can't allow caching of data until the change attribute has been initialised correctly. Signed-off-by: Trond Myklebust diff --git a/fs/nfs/inode.c b/fs/nfs/inode.c index d654661..426e4f8 100644 --- a/fs/nfs/inode.c +++ b/fs/nfs/inode.c @@ -443,7 +443,8 @@ nfs_fhget(struct super_block *sb, struct nfs_fh *fh, struct nfs_fattr *fattr, st if (fattr->valid & NFS_ATTR_FATTR_CHANGE) inode->i_version = fattr->change_attr; else if (nfs_server_capable(inode, NFS_CAP_CHANGE_ATTR)) - nfs_set_cache_invalid(inode, NFS_INO_INVALID_ATTR); + nfs_set_cache_invalid(inode, NFS_INO_INVALID_ATTR + | NFS_INO_REVAL_PAGECACHE); if (fattr->valid & NFS_ATTR_FATTR_SIZE) inode->i_size = nfs_size_to_loff_t(fattr->size); else -- cgit v0.10.2 From cd812599796f500b042f5464b6665755eca21137 Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Sun, 5 Jul 2015 11:12:07 -0400 Subject: NFS: Remove the "NFS_CAP_CHANGE_ATTR" capability Setting the change attribute has been mandatory for all NFS versions, since commit 3a1556e8662c ("NFSv2/v3: Simulate the change attribute"). We should therefore not have anything be conditional on it being set/unset. Signed-off-by: Trond Myklebust diff --git a/fs/nfs/client.c b/fs/nfs/client.c index ecebb40..4a90c9b 100644 --- a/fs/nfs/client.c +++ b/fs/nfs/client.c @@ -775,7 +775,7 @@ static int nfs_init_server(struct nfs_server *server, server->options = data->options; server->caps |= NFS_CAP_HARDLINKS|NFS_CAP_SYMLINKS|NFS_CAP_FILEID| NFS_CAP_MODE|NFS_CAP_NLINK|NFS_CAP_OWNER|NFS_CAP_OWNER_GROUP| - NFS_CAP_ATIME|NFS_CAP_CTIME|NFS_CAP_MTIME|NFS_CAP_CHANGE_ATTR; + NFS_CAP_ATIME|NFS_CAP_CTIME|NFS_CAP_MTIME; if (data->rsize) server->rsize = nfs_block_size(data->rsize, NULL); diff --git a/fs/nfs/inode.c b/fs/nfs/inode.c index 426e4f8..0adc7d2 100644 --- a/fs/nfs/inode.c +++ b/fs/nfs/inode.c @@ -442,7 +442,7 @@ nfs_fhget(struct super_block *sb, struct nfs_fh *fh, struct nfs_fattr *fattr, st nfs_set_cache_invalid(inode, NFS_INO_INVALID_ATTR); if (fattr->valid & NFS_ATTR_FATTR_CHANGE) inode->i_version = fattr->change_attr; - else if (nfs_server_capable(inode, NFS_CAP_CHANGE_ATTR)) + else nfs_set_cache_invalid(inode, NFS_INO_INVALID_ATTR | NFS_INO_REVAL_PAGECACHE); if (fattr->valid & NFS_ATTR_FATTR_SIZE) @@ -1692,7 +1692,7 @@ static int nfs_update_inode(struct inode *inode, struct nfs_fattr *fattr) nfs_force_lookup_revalidate(inode); inode->i_version = fattr->change_attr; } - } else if (server->caps & NFS_CAP_CHANGE_ATTR) + } else nfsi->cache_validity |= save_cache_validity; if (fattr->valid & NFS_ATTR_FATTR_MTIME) { diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c index 9264994..c85ffe6 100644 --- a/fs/nfs/nfs4proc.c +++ b/fs/nfs/nfs4proc.c @@ -8591,7 +8591,6 @@ static const struct nfs4_minor_version_ops nfs_v4_0_minor_ops = { .minor_version = 0, .init_caps = NFS_CAP_READDIRPLUS | NFS_CAP_ATOMIC_OPEN - | NFS_CAP_CHANGE_ATTR | NFS_CAP_POSIX_LOCK, .init_client = nfs40_init_client, .shutdown_client = nfs40_shutdown_client, @@ -8617,7 +8616,6 @@ static const struct nfs4_minor_version_ops nfs_v4_1_minor_ops = { .minor_version = 1, .init_caps = NFS_CAP_READDIRPLUS | NFS_CAP_ATOMIC_OPEN - | NFS_CAP_CHANGE_ATTR | NFS_CAP_POSIX_LOCK | NFS_CAP_STATEID_NFSV41 | NFS_CAP_ATOMIC_OPEN_V1, @@ -8640,7 +8638,6 @@ static const struct nfs4_minor_version_ops nfs_v4_2_minor_ops = { .minor_version = 2, .init_caps = NFS_CAP_READDIRPLUS | NFS_CAP_ATOMIC_OPEN - | NFS_CAP_CHANGE_ATTR | NFS_CAP_POSIX_LOCK | NFS_CAP_STATEID_NFSV41 | NFS_CAP_ATOMIC_OPEN_V1 diff --git a/include/linux/nfs_fs_sb.h b/include/linux/nfs_fs_sb.h index a2ea149..20bc8e5 100644 --- a/include/linux/nfs_fs_sb.h +++ b/include/linux/nfs_fs_sb.h @@ -220,7 +220,7 @@ struct nfs_server { #define NFS_CAP_SYMLINKS (1U << 2) #define NFS_CAP_ACLS (1U << 3) #define NFS_CAP_ATOMIC_OPEN (1U << 4) -#define NFS_CAP_CHANGE_ATTR (1U << 5) +/* #define NFS_CAP_CHANGE_ATTR (1U << 5) */ #define NFS_CAP_FILEID (1U << 6) #define NFS_CAP_MODE (1U << 7) #define NFS_CAP_NLINK (1U << 8) -- cgit v0.10.2 From 115c48d7a5351abeadd0c8a3dc87eca3d66a6475 Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Sun, 5 Jul 2015 12:36:34 -0400 Subject: NFS: nfs_mark_for_revalidate should always set NFS_INO_REVAL_PAGECACHE I'm not aware of any existing bugs around this, but the expectation is that nfs_mark_for_revalidate() should always force a revalidation of the cached metadata. Signed-off-by: Trond Myklebust diff --git a/include/linux/nfs_fs.h b/include/linux/nfs_fs.h index f91b5ad..874b772 100644 --- a/include/linux/nfs_fs.h +++ b/include/linux/nfs_fs.h @@ -292,9 +292,12 @@ static inline void nfs_mark_for_revalidate(struct inode *inode) struct nfs_inode *nfsi = NFS_I(inode); spin_lock(&inode->i_lock); - nfsi->cache_validity |= NFS_INO_INVALID_ATTR|NFS_INO_INVALID_ACCESS; + nfsi->cache_validity |= NFS_INO_INVALID_ATTR | + NFS_INO_REVAL_PAGECACHE | + NFS_INO_INVALID_ACCESS | + NFS_INO_INVALID_ACL; if (S_ISDIR(inode->i_mode)) - nfsi->cache_validity |= NFS_INO_REVAL_PAGECACHE|NFS_INO_INVALID_DATA; + nfsi->cache_validity |= NFS_INO_INVALID_DATA; spin_unlock(&inode->i_lock); } -- cgit v0.10.2 From c3c5819a350952439c3198aa46581f9e4c46557f Mon Sep 17 00:00:00 2001 From: Mathias Nyman Date: Tue, 21 Jul 2015 17:20:25 +0300 Subject: xhci: call BIOS workaround to enable runtime suspend on Intel Braswell Intel xhci hw that require XHCI_PME_STUCK quirk have as default disabled xhci from going to D3 state in runtime suspend. Driver needs to verify it can deal with the hw by calling an ACPI _DSM method to get D3 enabled. Signed-off-by: Mathias Nyman Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/usb/host/xhci-pci.c b/drivers/usb/host/xhci-pci.c index 4a4cb1d..da10dc7 100644 --- a/drivers/usb/host/xhci-pci.c +++ b/drivers/usb/host/xhci-pci.c @@ -23,6 +23,7 @@ #include #include #include +#include #include "xhci.h" #include "xhci-trace.h" @@ -190,6 +191,19 @@ static void xhci_pme_quirk(struct xhci_hcd *xhci) readl(reg); } +#ifdef CONFIG_ACPI +static void xhci_pme_acpi_rtd3_enable(struct pci_dev *dev) +{ + static const u8 intel_dsm_uuid[] = { + 0xb7, 0x0c, 0x34, 0xac, 0x01, 0xe9, 0xbf, 0x45, + 0xb7, 0xe6, 0x2b, 0x34, 0xec, 0x93, 0x1e, 0x23, + }; + acpi_evaluate_dsm(ACPI_HANDLE(&dev->dev), intel_dsm_uuid, 3, 1, NULL); +} +#else + static void xhci_pme_acpi_rtd3_enable(struct pci_dev *dev) { } +#endif /* CONFIG_ACPI */ + /* called during probe() after chip reset completes */ static int xhci_pci_setup(struct usb_hcd *hcd) { @@ -263,6 +277,9 @@ static int xhci_pci_probe(struct pci_dev *dev, const struct pci_device_id *id) HCC_MAX_PSA(xhci->hcc_params) >= 4) xhci->shared_hcd->can_do_streams = 1; + if (xhci->quirks & XHCI_PME_STUCK_QUIRK) + xhci_pme_acpi_rtd3_enable(dev); + /* USB-2 and USB-3 roothubs initialized, allow runtime pm suspend */ pm_runtime_put_noidle(&dev->dev); -- cgit v0.10.2 From abce329c27b315cfc01be1a305ee976ee13ed4cf Mon Sep 17 00:00:00 2001 From: Rajmohan Mani Date: Tue, 21 Jul 2015 17:20:26 +0300 Subject: xhci: Workaround to get D3 working in Intel xHCI The xHCI in Intel CherryView / Braswell Platform requires a driver workaround to get xHCI D3 working. Without this workaround, xHCI might not enter D3. Workaround is to configure SSIC PORT as "unused" before D3 entry and "used" after D3 exit. This is done through a vendor specific register (PORT2_SSIC_CONFIG_REG2 at offset 0x883c), in xhci suspend / resume callbacks. Verified xHCI D3 works fine in CherryView / Braswell platform. Signed-off-by: Rajmohan Mani Signed-off-by: Mathias Nyman Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/usb/host/xhci-pci.c b/drivers/usb/host/xhci-pci.c index da10dc7..5590eac 100644 --- a/drivers/usb/host/xhci-pci.c +++ b/drivers/usb/host/xhci-pci.c @@ -28,6 +28,10 @@ #include "xhci.h" #include "xhci-trace.h" +#define PORT2_SSIC_CONFIG_REG2 0x883c +#define PROG_DONE (1 << 30) +#define SSIC_PORT_UNUSED (1 << 31) + /* Device for a quirk */ #define PCI_VENDOR_ID_FRESCO_LOGIC 0x1b73 #define PCI_DEVICE_ID_FRESCO_LOGIC_PDK 0x1000 @@ -177,14 +181,44 @@ static void xhci_pci_quirks(struct device *dev, struct xhci_hcd *xhci) } /* + * In some Intel xHCI controllers, in order to get D3 working, + * through a vendor specific SSIC CONFIG register at offset 0x883c, + * SSIC PORT need to be marked as "unused" before putting xHCI + * into D3. After D3 exit, the SSIC port need to be marked as "used". + * Without this change, xHCI might not enter D3 state. * Make sure PME works on some Intel xHCI controllers by writing 1 to clear * the Internal PME flag bit in vendor specific PMCTRL register at offset 0x80a4 */ -static void xhci_pme_quirk(struct xhci_hcd *xhci) +static void xhci_pme_quirk(struct usb_hcd *hcd, bool suspend) { + struct xhci_hcd *xhci = hcd_to_xhci(hcd); + struct pci_dev *pdev = to_pci_dev(hcd->self.controller); u32 val; void __iomem *reg; + if (pdev->vendor == PCI_VENDOR_ID_INTEL && + pdev->device == PCI_DEVICE_ID_INTEL_CHERRYVIEW_XHCI) { + + reg = (void __iomem *) xhci->cap_regs + PORT2_SSIC_CONFIG_REG2; + + /* Notify SSIC that SSIC profile programming is not done */ + val = readl(reg) & ~PROG_DONE; + writel(val, reg); + + /* Mark SSIC port as unused(suspend) or used(resume) */ + val = readl(reg); + if (suspend) + val |= SSIC_PORT_UNUSED; + else + val &= ~SSIC_PORT_UNUSED; + writel(val, reg); + + /* Notify SSIC that SSIC profile programming is done */ + val = readl(reg) | PROG_DONE; + writel(val, reg); + readl(reg); + } + reg = (void __iomem *) xhci->cap_regs + 0x80a4; val = readl(reg); writel(val | BIT(28), reg); @@ -324,7 +358,7 @@ static int xhci_pci_suspend(struct usb_hcd *hcd, bool do_wakeup) pdev->no_d3cold = true; if (xhci->quirks & XHCI_PME_STUCK_QUIRK) - xhci_pme_quirk(xhci); + xhci_pme_quirk(hcd, true); return xhci_suspend(xhci, do_wakeup); } @@ -357,7 +391,7 @@ static int xhci_pci_resume(struct usb_hcd *hcd, bool hibernated) usb_enable_intel_xhci_ports(pdev); if (xhci->quirks & XHCI_PME_STUCK_QUIRK) - xhci_pme_quirk(xhci); + xhci_pme_quirk(hcd, false); retval = xhci_resume(xhci, hibernated); return retval; -- cgit v0.10.2 From 3496810663922617d4b706ef2780c279252ddd6a Mon Sep 17 00:00:00 2001 From: AMAN DEEP Date: Tue, 21 Jul 2015 17:20:27 +0300 Subject: usb: xhci: Bugfix for NULL pointer deference in xhci_endpoint_init() function virt_dev->num_cached_rings counts on freed ring and is not updated correctly. In xhci_free_or_cache_endpoint_ring() function, the free ring is added into cache and then num_rings_cache is incremented as below: virt_dev->ring_cache[rings_cached] = virt_dev->eps[ep_index].ring; virt_dev->num_rings_cached++; here, free ring pointer is added to a current index and then index is incremented. So current index always points to empty location in the ring cache. For getting available free ring, current index should be decremented first and then corresponding ring buffer value should be taken from ring cache. But In function xhci_endpoint_init(), the num_rings_cached index is accessed before decrement. virt_dev->eps[ep_index].new_ring = virt_dev->ring_cache[virt_dev->num_rings_cached]; virt_dev->ring_cache[virt_dev->num_rings_cached] = NULL; virt_dev->num_rings_cached--; This is bug in manipulating the index of ring cache. And it should be as below: virt_dev->num_rings_cached--; virt_dev->eps[ep_index].new_ring = virt_dev->ring_cache[virt_dev->num_rings_cached]; virt_dev->ring_cache[virt_dev->num_rings_cached] = NULL; Cc: Signed-off-by: Aman Deep Signed-off-by: Mathias Nyman Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/usb/host/xhci-mem.c b/drivers/usb/host/xhci-mem.c index f833640..3e442f7 100644 --- a/drivers/usb/host/xhci-mem.c +++ b/drivers/usb/host/xhci-mem.c @@ -1427,10 +1427,10 @@ int xhci_endpoint_init(struct xhci_hcd *xhci, /* Attempt to use the ring cache */ if (virt_dev->num_rings_cached == 0) return -ENOMEM; + virt_dev->num_rings_cached--; virt_dev->eps[ep_index].new_ring = virt_dev->ring_cache[virt_dev->num_rings_cached]; virt_dev->ring_cache[virt_dev->num_rings_cached] = NULL; - virt_dev->num_rings_cached--; xhci_reinit_cached_ring(xhci, virt_dev->eps[ep_index].new_ring, 1, type); } -- cgit v0.10.2 From 326124a027abc9a7f43f72dc94f6f0f7a55b02b3 Mon Sep 17 00:00:00 2001 From: Brian Campbell Date: Tue, 21 Jul 2015 17:20:28 +0300 Subject: xhci: Calculate old endpoints correctly on device reset When resetting a device the number of active TTs may need to be corrected by xhci_update_tt_active_eps, but the number of old active endpoints supplied to it was always zero, so the number of TTs and the bandwidth reserved for them was not updated, and could rise unnecessarily. This affected systems using Intel's Patherpoint chipset, which rely on software bandwidth checking. For example, a Lenovo X230 would lose the ability to use ports on the docking station after enough suspend/resume cycles because the bandwidth calculated would rise with every cycle when a suitable device is attached. The correct number of active endpoints is calculated in the same way as in xhci_reserve_bandwidth. Cc: Signed-off-by: Brian Campbell Signed-off-by: Mathias Nyman Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/usb/host/xhci.c b/drivers/usb/host/xhci.c index 7da0d60..526ebc0 100644 --- a/drivers/usb/host/xhci.c +++ b/drivers/usb/host/xhci.c @@ -3453,6 +3453,9 @@ int xhci_discover_or_reset_device(struct usb_hcd *hcd, struct usb_device *udev) return -EINVAL; } + if (virt_dev->tt_info) + old_active_eps = virt_dev->tt_info->active_eps; + if (virt_dev->udev != udev) { /* If the virt_dev and the udev does not match, this virt_dev * may belong to another udev. -- cgit v0.10.2 From 243292a2ad3dc365849b820a64868927168894ac Mon Sep 17 00:00:00 2001 From: Zhuang Jin Can Date: Tue, 21 Jul 2015 17:20:29 +0300 Subject: xhci: report U3 when link is in resume state xhci_hub_report_usb3_link_state() returns pls as U0 when the link is in resume state, and this causes usb core to think the link is in U0 while actually it's in resume state. When usb core transfers control request on the link, it fails with TRB error as the link is not ready for transfer. To fix the issue, report U3 when the link is in resume state, thus usb core knows the link it's not ready for transfer. Cc: Signed-off-by: Zhuang Jin Can Signed-off-by: Mathias Nyman Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/usb/host/xhci-hub.c b/drivers/usb/host/xhci-hub.c index e75c565..7df8878 100644 --- a/drivers/usb/host/xhci-hub.c +++ b/drivers/usb/host/xhci-hub.c @@ -484,10 +484,13 @@ static void xhci_hub_report_usb3_link_state(struct xhci_hcd *xhci, u32 pls = status_reg & PORT_PLS_MASK; /* resume state is a xHCI internal state. - * Do not report it to usb core. + * Do not report it to usb core, instead, pretend to be U3, + * thus usb core knows it's not ready for transfer */ - if (pls == XDEV_RESUME) + if (pls == XDEV_RESUME) { + *status |= USB_SS_PORT_LS_U3; return; + } /* When the CAS bit is set then warm reset * should be performed on port -- cgit v0.10.2 From fac4271d1126c45ceaceb7f4a336317b771eb121 Mon Sep 17 00:00:00 2001 From: Zhuang Jin Can Date: Tue, 21 Jul 2015 17:20:30 +0300 Subject: xhci: prevent bus_suspend if SS port resuming in phase 1 When the link is just waken, it's in Resume state, and driver sets PLS to U0. This refers to Phase 1. Phase 2 refers to when the link has completed the transition from Resume state to U0. With the fix of xhci: report U3 when link is in resume state, it also exposes an issue that usb3 roothub and controller can suspend right after phase 1, and this causes a hard hang in controller. To fix the issue, we need to prevent usb3 bus suspend if any port is resuming in phase 1. [merge separate USB2 and USB3 port resume checking to one -Mathias] Cc: Signed-off-by: Zhuang Jin Can Signed-off-by: Mathias Nyman Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/usb/host/xhci-hub.c b/drivers/usb/host/xhci-hub.c index 7df8878..fdca8ed 100644 --- a/drivers/usb/host/xhci-hub.c +++ b/drivers/usb/host/xhci-hub.c @@ -1123,10 +1123,10 @@ int xhci_bus_suspend(struct usb_hcd *hcd) spin_lock_irqsave(&xhci->lock, flags); if (hcd->self.root_hub->do_remote_wakeup) { - if (bus_state->resuming_ports) { + if (bus_state->resuming_ports || /* USB2 */ + bus_state->port_remote_wakeup) { /* USB3 */ spin_unlock_irqrestore(&xhci->lock, flags); - xhci_dbg(xhci, "suspend failed because " - "a port is resuming\n"); + xhci_dbg(xhci, "suspend failed because a port is resuming\n"); return -EBUSY; } } diff --git a/drivers/usb/host/xhci-ring.c b/drivers/usb/host/xhci-ring.c index 94416ff..6a8fc52 100644 --- a/drivers/usb/host/xhci-ring.c +++ b/drivers/usb/host/xhci-ring.c @@ -1546,6 +1546,9 @@ static void handle_port_status(struct xhci_hcd *xhci, usb_hcd_resume_root_hub(hcd); } + if (hcd->speed == HCD_USB3 && (temp & PORT_PLS_MASK) == XDEV_INACTIVE) + bus_state->port_remote_wakeup &= ~(1 << faked_port_index); + if ((temp & PORT_PLC) && (temp & PORT_PLS_MASK) == XDEV_RESUME) { xhci_dbg(xhci, "port resume event for port %d\n", port_id); diff --git a/drivers/usb/host/xhci.h b/drivers/usb/host/xhci.h index 31e46cc..ed2ebf6 100644 --- a/drivers/usb/host/xhci.h +++ b/drivers/usb/host/xhci.h @@ -285,6 +285,7 @@ struct xhci_op_regs { #define XDEV_U0 (0x0 << 5) #define XDEV_U2 (0x2 << 5) #define XDEV_U3 (0x3 << 5) +#define XDEV_INACTIVE (0x6 << 5) #define XDEV_RESUME (0xf << 5) /* true: port has power (see HCC_PPC) */ #define PORT_POWER (1 << 9) -- cgit v0.10.2 From aca3a0489ac019b58cf32794d5362bb284cb9b94 Mon Sep 17 00:00:00 2001 From: Zhuang Jin Can Date: Tue, 21 Jul 2015 17:20:31 +0300 Subject: xhci: do not report PLC when link is in internal resume state Port link change with port in resume state should not be reported to usbcore, as this is an internal state to be handled by xhci driver. Reporting PLC to usbcore may cause usbcore clearing PLC first and port change event irq won't be generated. Cc: Signed-off-by: Zhuang Jin Can Signed-off-by: Mathias Nyman Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/usb/host/xhci-hub.c b/drivers/usb/host/xhci-hub.c index fdca8ed..78241b5 100644 --- a/drivers/usb/host/xhci-hub.c +++ b/drivers/usb/host/xhci-hub.c @@ -591,7 +591,14 @@ static u32 xhci_get_port_status(struct usb_hcd *hcd, status |= USB_PORT_STAT_C_RESET << 16; /* USB3.0 only */ if (hcd->speed == HCD_USB3) { - if ((raw_port_status & PORT_PLC)) + /* Port link change with port in resume state should not be + * reported to usbcore, as this is an internal state to be + * handled by xhci driver. Reporting PLC to usbcore may + * cause usbcore clearing PLC first and port change event + * irq won't be generated. + */ + if ((raw_port_status & PORT_PLC) && + (raw_port_status & PORT_PLS_MASK) != XDEV_RESUME) status |= USB_PORT_STAT_C_LINK_STATE << 16; if ((raw_port_status & PORT_WRC)) status |= USB_PORT_STAT_C_BH_RESET << 16; -- cgit v0.10.2 From 2d2a316765d956bc5cb6bb367b2ec52ca59ab8e9 Mon Sep 17 00:00:00 2001 From: Lu Baolu Date: Tue, 16 Jun 2015 09:08:26 +0800 Subject: usb: core: lpm: set lpm_capable for root hub device Commit 25cd2882e2fc ("usb/xhci: Change how we indicate a host supports Link PM.") removed the code to set lpm_capable for USB 3.0 super-speed root hub. The intention of that change was to avoid touching usb core internal field, a.k.a. lpm_capable, and let usb core to set it by checking U1 and U2 exit latency values in the descriptor. Usb core checks and sets lpm_capable in hub_port_init(). Unfortunately, root hub is a special usb device as it has no parent. Hub_port_init() will never be called for a root hub device. That means lpm_capable will by no means be set for the root hub. As the result, lpm isn't functional at all in Linux kernel. This patch add the code to check and set lpm_capable when registering a root hub device. It could be back-ported to kernels as old as v3.15, that contains the Commit 25cd2882e2fc ("usb/xhci: Change how we indicate a host supports Link PM."). Cc: stable@vger.kernel.org # 3.15 Reported-by: Kevin Strasser Signed-off-by: Lu Baolu Acked-by: Alan Stern Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/usb/core/hcd.c b/drivers/usb/core/hcd.c index be5b207..cbcd092 100644 --- a/drivers/usb/core/hcd.c +++ b/drivers/usb/core/hcd.c @@ -1022,9 +1022,12 @@ static int register_root_hub(struct usb_hcd *hcd) dev_name(&usb_dev->dev), retval); return (retval < 0) ? retval : -EMSGSIZE; } - if (usb_dev->speed == USB_SPEED_SUPER) { + + if (le16_to_cpu(usb_dev->descriptor.bcdUSB) >= 0x0201) { retval = usb_get_bos_descriptor(usb_dev); - if (retval < 0) { + if (!retval) { + usb_dev->lpm_capable = usb_device_supports_lpm(usb_dev); + } else if (usb_dev->speed == USB_SPEED_SUPER) { mutex_unlock(&usb_bus_list_lock); dev_dbg(parent_dev, "can't read %s bos descriptor %d\n", dev_name(&usb_dev->dev), retval); diff --git a/drivers/usb/core/hub.c b/drivers/usb/core/hub.c index 43cb2f2..73dfa19 100644 --- a/drivers/usb/core/hub.c +++ b/drivers/usb/core/hub.c @@ -122,7 +122,7 @@ struct usb_hub *usb_hub_to_struct_hub(struct usb_device *hdev) return usb_get_intfdata(hdev->actconfig->interface[0]); } -static int usb_device_supports_lpm(struct usb_device *udev) +int usb_device_supports_lpm(struct usb_device *udev) { /* USB 2.1 (and greater) devices indicate LPM support through * their USB 2.0 Extended Capabilities BOS descriptor. diff --git a/drivers/usb/core/usb.h b/drivers/usb/core/usb.h index 7eb1e26..457255a 100644 --- a/drivers/usb/core/usb.h +++ b/drivers/usb/core/usb.h @@ -65,6 +65,7 @@ extern int usb_hub_init(void); extern void usb_hub_cleanup(void); extern int usb_major_init(void); extern void usb_major_cleanup(void); +extern int usb_device_supports_lpm(struct usb_device *udev); #ifdef CONFIG_PM -- cgit v0.10.2 From 7d8021c967648accd1b78e5e1ddaad655cd2c61f Mon Sep 17 00:00:00 2001 From: Alan Stern Date: Tue, 30 Jun 2015 11:25:54 -0400 Subject: USB: OHCI: Fix race between ED unlink and URB submission This patch fixes a bug introduced by commit 977dcfdc6031 ("USB: OHCI: don't lose track of EDs when a controller dies"). The commit changed ed_state from ED_UNLINK to ED_IDLE too early, before finish_urb() had been called. The user-visible consequence is that the driver occasionally crashes or locks up when an URB is submitted while another URB for the same endpoint is being unlinked. This patch moves the ED state change later, to the right place. The drawback is that now we may unnecessarily execute some instructions multiple times when a controller dies. Since controllers dying is an exceptional occurrence, a little wasted time won't matter. Signed-off-by: Alan Stern Reported-by: Heiko Przybyl Tested-by: Heiko Przybyl Fixes: 977dcfdc60311e7aa571cabf6f39c36dde13339e CC: Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/usb/host/ohci-q.c b/drivers/usb/host/ohci-q.c index f7d561e..d029bbe 100644 --- a/drivers/usb/host/ohci-q.c +++ b/drivers/usb/host/ohci-q.c @@ -981,10 +981,6 @@ rescan_all: int completed, modified; __hc32 *prev; - /* Is this ED already invisible to the hardware? */ - if (ed->state == ED_IDLE) - goto ed_idle; - /* only take off EDs that the HC isn't using, accounting for * frame counter wraps and EDs with partially retired TDs */ @@ -1012,12 +1008,10 @@ skip_ed: } /* ED's now officially unlinked, hc doesn't see */ - ed->state = ED_IDLE; ed->hwHeadP &= ~cpu_to_hc32(ohci, ED_H); ed->hwNextED = 0; wmb(); ed->hwINFO &= ~cpu_to_hc32(ohci, ED_SKIP | ED_DEQUEUE); -ed_idle: /* reentrancy: if we drop the schedule lock, someone might * have modified this list. normally it's just prepending @@ -1088,6 +1082,7 @@ rescan_this: if (list_empty(&ed->td_list)) { *last = ed->ed_next; ed->ed_next = NULL; + ed->state = ED_IDLE; list_del(&ed->in_use_list); } else if (ohci->rh_state == OHCI_RH_RUNNING) { *last = ed->ed_next; -- cgit v0.10.2 From 5fb2c782f451a4fb9c19c076e2c442839faf0f76 Mon Sep 17 00:00:00 2001 From: Oliver Neukum Date: Mon, 6 Jul 2015 13:12:32 +0200 Subject: usb-storage: ignore ZTE MF 823 card reader in mode 0x1225 This device automatically switches itself to another mode (0x1405) unless the specific access pattern of Windows is followed in its initial mode. That makes a dirty unmount of the internal storage devices inevitable if they are mounted. So the card reader of such a device should be ignored, lest an unclean removal become inevitable. This replaces an earlier patch that ignored all LUNs of this device. That patch was overly broad. Signed-off-by: Oliver Neukum CC: stable@vger.kernel.org Reviewed-by: Lars Melin Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/usb/storage/unusual_devs.h b/drivers/usb/storage/unusual_devs.h index caf1888..87898ca 100644 --- a/drivers/usb/storage/unusual_devs.h +++ b/drivers/usb/storage/unusual_devs.h @@ -2065,6 +2065,18 @@ UNUSUAL_DEV( 0x1908, 0x3335, 0x0200, 0x0200, USB_SC_DEVICE, USB_PR_DEVICE, NULL, US_FL_NO_READ_DISC_INFO ), +/* Reported by Oliver Neukum + * This device morphes spontaneously into another device if the access + * pattern of Windows isn't followed. Thus writable media would be dirty + * if the initial instance is used. So the device is limited to its + * virtual CD. + * And yes, the concept that BCD goes up to 9 is not heeded */ +UNUSUAL_DEV( 0x19d2, 0x1225, 0x0000, 0xffff, + "ZTE,Incorporated", + "ZTE WCDMA Technologies MSM", + USB_SC_DEVICE, USB_PR_DEVICE, NULL, + US_FL_SINGLE_LUN ), + /* Reported by Sven Geggus * This encrypted pen drive returns bogus data for the initial READ(10). */ -- cgit v0.10.2 From 2d64f31bdfbb2ea66cbd4271c079f11808c79c28 Mon Sep 17 00:00:00 2001 From: Hans de Goede Date: Wed, 8 Jul 2015 14:03:41 +0200 Subject: usb-storage: Add ignore-device quirk for gm12u320 based usb mini projectors Grain-media GM12U320 based devices are mini video projectors using USB for both power and video data transport. Their usb-storage interface is a virtual windows driver CD. The gm12u320 kms driver needs these interfaces to talk to the device and export it as framebuffer & kms dri device nodes, so make sure that the usb-storage driver does not bind to it. Signed-off-by: Hans de Goede Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/usb/storage/unusual_devs.h b/drivers/usb/storage/unusual_devs.h index 87898ca..6b24791 100644 --- a/drivers/usb/storage/unusual_devs.h +++ b/drivers/usb/storage/unusual_devs.h @@ -2086,6 +2086,17 @@ UNUSUAL_DEV( 0x1b1c, 0x1ab5, 0x0200, 0x0200, USB_SC_DEVICE, USB_PR_DEVICE, NULL, US_FL_INITIAL_READ10 ), +/* Reported by Hans de Goede + * These are mini projectors using USB for both power and video data transport + * The usb-storage interface is a virtual windows driver CD, which the gm12u320 + * driver automatically converts into framebuffer & kms dri device nodes. + */ +UNUSUAL_DEV( 0x1de1, 0xc102, 0x0000, 0xffff, + "Grain-media Technology Corp.", + "USB3.0 Device GM12U320", + USB_SC_DEVICE, USB_PR_DEVICE, NULL, + US_FL_IGNORE_DEVICE ), + /* Patch by Richard Schütz * This external hard drive enclosure uses a JMicron chip which * needs the US_FL_IGNORE_RESIDUE flag to work properly. */ -- cgit v0.10.2 From 91b725600d8f27d759d211e36dda6182c6f93ec6 Mon Sep 17 00:00:00 2001 From: Johannes Thumshirn Date: Wed, 8 Jul 2015 17:25:42 +0200 Subject: cdc-acm: Destroy acm_minors IDR on module exit Destroy acm_minors IDR on module exit, reclaiming the allocated memory. This was detected by the following semantic patch (written by Luis Rodriguez ) @ defines_module_init @ declarer name module_init, module_exit; declarer name DEFINE_IDR; identifier init; @@ module_init(init); @ defines_module_exit @ identifier exit; @@ module_exit(exit); @ declares_idr depends on defines_module_init && defines_module_exit @ identifier idr; @@ DEFINE_IDR(idr); @ on_exit_calls_destroy depends on declares_idr && defines_module_exit @ identifier declares_idr.idr, defines_module_exit.exit; @@ exit(void) { ... idr_destroy(&idr); ... } @ missing_module_idr_destroy depends on declares_idr && defines_module_exit && !on_exit_calls_destroy @ identifier declares_idr.idr, defines_module_exit.exit; @@ exit(void) { ... +idr_destroy(&idr); } Signed-off-by: Johannes Thumshirn Acked-by: Oliver Neukum Acked-by: Johan Hovold Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/usb/class/cdc-acm.c b/drivers/usb/class/cdc-acm.c index 519a77b..b30e742 100644 --- a/drivers/usb/class/cdc-acm.c +++ b/drivers/usb/class/cdc-acm.c @@ -1944,6 +1944,7 @@ static void __exit acm_exit(void) usb_deregister(&acm_driver); tty_unregister_driver(acm_tty_driver); put_tty_driver(acm_tty_driver); + idr_destroy(&acm_minors); } module_init(acm_init); -- cgit v0.10.2 From 1209544d8a2a6084f58625ca66f5cd77580df53f Mon Sep 17 00:00:00 2001 From: Alan Stern Date: Wed, 8 Jul 2015 12:14:56 -0400 Subject: USB: OHCI: fix bad #define in ohci-tmio.c An incorrect definition of CCR_PM_USBPW3 in ohci-tmio.c is a perennial source of invalid diagnoses from static scanners, such as in . This patch fixes the definition. Signed-off-by: Alan Stern Reported-by: David Binderman CC: Dmitry Eremin-Solenikov Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/usb/host/ohci-tmio.c b/drivers/usb/host/ohci-tmio.c index e9a6eec..cfcfadf 100644 --- a/drivers/usb/host/ohci-tmio.c +++ b/drivers/usb/host/ohci-tmio.c @@ -58,7 +58,7 @@ #define CCR_PM_CKRNEN 0x0002 #define CCR_PM_USBPW1 0x0004 #define CCR_PM_USBPW2 0x0008 -#define CCR_PM_USBPW3 0x0008 +#define CCR_PM_USBPW3 0x0010 #define CCR_PM_PMEE 0x0100 #define CCR_PM_PMES 0x8000 -- cgit v0.10.2 From 5f6c2d2b7dbb541c1e922538c49fa04c494ae3d7 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Wed, 22 Jul 2015 18:05:53 -0400 Subject: blkcg: fix gendisk reference leak in blkg_conf_prep() When a blkcg configuration is targeted to a partition rather than a whole device, blkg_conf_prep fails with -EINVAL; unfortunately, it forgets to put the gendisk ref in that case. Fix it. Signed-off-by: Tejun Heo Cc: stable@vger.kernel.org Signed-off-by: Jens Axboe diff --git a/block/blk-cgroup.c b/block/blk-cgroup.c index 9da02c0..d6283b3 100644 --- a/block/blk-cgroup.c +++ b/block/blk-cgroup.c @@ -718,8 +718,12 @@ int blkg_conf_prep(struct blkcg *blkcg, const struct blkcg_policy *pol, return -EINVAL; disk = get_gendisk(MKDEV(major, minor), &part); - if (!disk || part) + if (!disk) return -EINVAL; + if (part) { + put_disk(disk); + return -EINVAL; + } rcu_read_lock(); spin_lock_irq(disk->queue->queue_lock); -- cgit v0.10.2 From d3b178adb3a3adf54ecf77758138b654c3ee7f09 Mon Sep 17 00:00:00 2001 From: Goldwyn Rodrigues Date: Wed, 22 Jul 2015 12:09:17 -0500 Subject: md: Skip cluster setup for dm-raid There is a bug that the bitmap superblock isn't initialised properly for dm-raid, so a new field can have garbage in new fields. (dm-raid does initialisation in the kernel - md initialised the superblock in mdadm). This means that for dm-raid we cannot currently trust the new ->nodes field. So: - use __GFP_ZERO to initialise the superblock properly for all new arrays - initialise all fields in bitmap_info in bitmap_new_disk_sb - ignore ->nodes for dm arrays (yes, this is a hack) This bug exposes dm-raid to bug in the (still experimental) md-cluster code, so it is suitable for -stable. It does cause crashes. References: https://bugzilla.kernel.org/show_bug.cgi?id=100491 Cc: stable@vger.kernel.org (v4.1) Signed-off-By: Goldwyn Rodrigues Signed-off-by: NeilBrown diff --git a/drivers/md/bitmap.c b/drivers/md/bitmap.c index 135a090..c90118e 100644 --- a/drivers/md/bitmap.c +++ b/drivers/md/bitmap.c @@ -494,7 +494,7 @@ static int bitmap_new_disk_sb(struct bitmap *bitmap) bitmap_super_t *sb; unsigned long chunksize, daemon_sleep, write_behind; - bitmap->storage.sb_page = alloc_page(GFP_KERNEL); + bitmap->storage.sb_page = alloc_page(GFP_KERNEL | __GFP_ZERO); if (bitmap->storage.sb_page == NULL) return -ENOMEM; bitmap->storage.sb_page->index = 0; @@ -541,6 +541,7 @@ static int bitmap_new_disk_sb(struct bitmap *bitmap) sb->state = cpu_to_le32(bitmap->flags); bitmap->events_cleared = bitmap->mddev->events; sb->events_cleared = cpu_to_le64(bitmap->mddev->events); + bitmap->mddev->bitmap_info.nodes = 0; kunmap_atomic(sb); @@ -611,8 +612,16 @@ re_read: daemon_sleep = le32_to_cpu(sb->daemon_sleep) * HZ; write_behind = le32_to_cpu(sb->write_behind); sectors_reserved = le32_to_cpu(sb->sectors_reserved); - nodes = le32_to_cpu(sb->nodes); - strlcpy(bitmap->mddev->bitmap_info.cluster_name, sb->cluster_name, 64); + /* XXX: This is a hack to ensure that we don't use clustering + * in case: + * - dm-raid is in use and + * - the nodes written in bitmap_sb is erroneous. + */ + if (!bitmap->mddev->sync_super) { + nodes = le32_to_cpu(sb->nodes); + strlcpy(bitmap->mddev->bitmap_info.cluster_name, + sb->cluster_name, 64); + } /* verify that the bitmap-specific fields are valid */ if (sb->magic != cpu_to_le32(BITMAP_MAGIC)) -- cgit v0.10.2 From 8a7330130470e173832385861c1aa6e37a4e50d2 Mon Sep 17 00:00:00 2001 From: Zhao Lei Date: Wed, 15 Jul 2015 11:48:14 +0800 Subject: btrfs: Fix lockdep warning of btrfs_run_delayed_iputs() Liu Bo reported a lockdep warning of delayed_iput_sem in xfstests generic/241: [ 2061.345955] ============================================= [ 2061.346027] [ INFO: possible recursive locking detected ] [ 2061.346027] 4.1.0+ #268 Tainted: G W [ 2061.346027] --------------------------------------------- [ 2061.346027] btrfs-cleaner/3045 is trying to acquire lock: [ 2061.346027] (&fs_info->delayed_iput_sem){++++..}, at: [] btrfs_run_delayed_iputs+0x6b/0x100 [ 2061.346027] but task is already holding lock: [ 2061.346027] (&fs_info->delayed_iput_sem){++++..}, at: [] btrfs_run_delayed_iputs+0x6b/0x100 [ 2061.346027] other info that might help us debug this: [ 2061.346027] Possible unsafe locking scenario: [ 2061.346027] CPU0 [ 2061.346027] ---- [ 2061.346027] lock(&fs_info->delayed_iput_sem); [ 2061.346027] lock(&fs_info->delayed_iput_sem); [ 2061.346027] *** DEADLOCK *** It is rarely happened, about 1/400 in my test env. The reason is recursion of btrfs_run_delayed_iputs(): cleaner_kthread -> btrfs_run_delayed_iputs() *1 -> get delayed_iput_sem lock *2 -> iput() -> ... -> btrfs_commit_transaction() -> btrfs_run_delayed_iputs() *1 -> get delayed_iput_sem lock (dead lock) *2 *1: recursion of btrfs_run_delayed_iputs() *2: warning of lockdep about delayed_iput_sem When fs is in high stress, new iputs may added into fs_info->delayed_iputs list when btrfs_run_delayed_iputs() is running, which cause second btrfs_run_delayed_iputs() run into down_read(&fs_info->delayed_iput_sem) again, and cause above lockdep warning. Actually, it will not cause real problem because both locks are read lock, but to avoid lockdep warning, we can do a fix. Fix: Don't do btrfs_run_delayed_iputs() in btrfs_commit_transaction() for cleaner_kthread thread to break above recursion path. cleaner_kthread is calling btrfs_run_delayed_iputs() explicitly in code, and don't need to call btrfs_run_delayed_iputs() again in btrfs_commit_transaction(), it also give us a bonus to avoid stack overflow. Test: No above lockdep warning after patch in 1200 generic/241 tests. Reported-by: Liu Bo Signed-off-by: Zhao Lei Reviewed-by: Liu Bo Signed-off-by: Chris Mason diff --git a/fs/btrfs/transaction.c b/fs/btrfs/transaction.c index 51e0f0d..f5021fc 100644 --- a/fs/btrfs/transaction.c +++ b/fs/btrfs/transaction.c @@ -2152,7 +2152,8 @@ int btrfs_commit_transaction(struct btrfs_trans_handle *trans, kmem_cache_free(btrfs_trans_handle_cachep, trans); - if (current != root->fs_info->transaction_kthread) + if (current != root->fs_info->transaction_kthread && + current != root->fs_info->cleaner_kthread) btrfs_run_delayed_iputs(root); return ret; -- cgit v0.10.2 From 95ab1f64908795a2edd6b847eca94a0c63a44be4 Mon Sep 17 00:00:00 2001 From: Zhao Lei Date: Wed, 15 Jul 2015 21:02:09 +0800 Subject: btrfs: Avoid NULL pointer dereference of free_extent_buffer when read_tree_block() fail When read_tree_block() failed, we can see following dmesg: [ 134.371389] BUG: unable to handle kernel NULL pointer dereference at 0000000000000063 [ 134.372236] IP: [] free_extent_buffer+0x21/0x90 [ 134.372236] PGD 0 [ 134.372236] Oops: 0000 [#1] SMP [ 134.372236] Modules linked in: [ 134.372236] CPU: 0 PID: 2289 Comm: mount Not tainted 4.2.0-rc1_HEAD_c65b99f046843d2455aa231747b5a07a999a9f3d_+ #115 [ 134.372236] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS rel-1.7.5.1-0-g8936dbb-20141113_115728-nilsson.home.kraxel.org 04/01/2014 [ 134.372236] task: ffff88003b6e1a00 ti: ffff880011e60000 task.ti: ffff880011e60000 [ 134.372236] RIP: 0010:[] [] free_extent_buffer+0x21/0x90 ... [ 134.372236] Call Trace: [ 134.372236] [] free_root_extent_buffers+0x91/0xb0 [ 134.372236] [] free_root_pointers+0x17d/0x190 [ 134.372236] [] open_ctree+0x1ca0/0x25b0 [ 134.372236] [] ? disk_name+0x97/0xb0 [ 134.372236] [] btrfs_mount+0x8fa/0xab0 ... Reason: read_tree_block() changed to return error number on fail, and this value(not NULL) is set to tree_root->node, then subsequent code will run to: free_root_pointers() ->free_root_extent_buffers() ->free_extent_buffer() ->atomic_read((extent_buffer *)(-E_XXX)->refs); and trigger above error. Fix: Set tree_root->node to NULL on fail to make error_handle code happy. Signed-off-by: Zhao Lei Signed-off-by: Chris Mason diff --git a/fs/btrfs/disk-io.c b/fs/btrfs/disk-io.c index e5aad7f..84cbbb2 100644 --- a/fs/btrfs/disk-io.c +++ b/fs/btrfs/disk-io.c @@ -2842,6 +2842,7 @@ int open_ctree(struct super_block *sb, !extent_buffer_uptodate(chunk_root->node)) { printk(KERN_ERR "BTRFS: failed to read chunk root on %s\n", sb->s_id); + chunk_root->node = NULL; goto fail_tree_roots; } btrfs_set_root_node(&chunk_root->root_item, chunk_root->node); @@ -2879,7 +2880,7 @@ retry_root_backup: !extent_buffer_uptodate(tree_root->node)) { printk(KERN_WARNING "BTRFS: failed to read tree root on %s\n", sb->s_id); - + tree_root->node = NULL; goto recovery_tree_root; } -- cgit v0.10.2 From 3e303ea60db3f0222c25a13f39a7cca7bf860df0 Mon Sep 17 00:00:00 2001 From: Anand Jain Date: Fri, 17 Jul 2015 23:45:23 +0800 Subject: btrfs: its btrfs_err() instead of btrfs_error() sorry I indented to use btrfs_err() and I have no idea how btrfs_error() got there. infact I was thinking about these kind of oversights since these two func are too closely named. Signed-off-by: Anand Jain Reviewed-by: Liu Bo Reviewed-by: David Sterba Signed-off-by: Chris Mason diff --git a/fs/btrfs/dev-replace.c b/fs/btrfs/dev-replace.c index 862fbc2..564a7de 100644 --- a/fs/btrfs/dev-replace.c +++ b/fs/btrfs/dev-replace.c @@ -378,7 +378,7 @@ int btrfs_dev_replace_start(struct btrfs_root *root, ret = btrfs_kobj_add_device(tgt_device->fs_devices, tgt_device); if (ret) - btrfs_error(root->fs_info, ret, "kobj add dev failed"); + btrfs_err(root->fs_info, "kobj add dev failed %d\n", ret); printk_in_rcu(KERN_INFO "BTRFS: dev_replace from %s (devid %llu) to %s started\n", -- cgit v0.10.2 From 00d80e342c0f4f1990ab69f594ee1e2348e51dd9 Mon Sep 17 00:00:00 2001 From: Filipe Manana Date: Mon, 20 Jul 2015 14:56:20 +0100 Subject: Btrfs: fix quick exhaustion of the system array in the superblock Omar reported that after commit 4fbcdf669454 ("Btrfs: fix -ENOSPC when finishing block group creation"), introduced in 4.2-rc1, the following test was failing due to exhaustion of the system array in the superblock: #!/bin/bash truncate -s 100T big.img mkfs.btrfs big.img mount -o loop big.img /mnt/loop num=5 sz=10T for ((i = 0; i < $num; i++)); do echo fallocate $i $sz fallocate -l $sz /mnt/loop/testfile$i done btrfs filesystem sync /mnt/loop for ((i = 0; i < $num; i++)); do echo rm $i rm /mnt/loop/testfile$i btrfs filesystem sync /mnt/loop done umount /mnt/loop This made btrfs_add_system_chunk() fail with -EFBIG due to excessive allocation of system block groups. This happened because the test creates a large number of data block groups per transaction and when committing the transaction we start the writeout of the block group caches for all the new new (dirty) block groups, which results in pre-allocating space for each block group's free space cache using the same transaction handle. That in turn often leads to creation of more block groups, and all get attached to the new_bgs list of the same transaction handle to the point of getting a list with over 1500 elements, and creation of new block groups leads to the need of reserving space in the chunk block reserve and often creating a new system block group too. So that made us quickly exhaust the chunk block reserve/system space info, because as of the commit mentioned before, we do reserve space for each new block group in the chunk block reserve, unlike before where we would not and would at most allocate one new system block group and therefore would only ensure that there was enough space in the system space info to allocate 1 new block group even if we ended up allocating thousands of new block groups using the same transaction handle. That worked most of the time because the computed required space at check_system_chunk() is very pessimistic (assumes a chunk tree height of BTRFS_MAX_LEVEL/8 and that all nodes/leafs in a path will be COWed and split) and since the updates to the chunk tree all happen at btrfs_create_pending_block_groups it is unlikely that a path needs to be COWed more than once (unless writepages() for the btree inode is called by mm in between) and that compensated for the need of creating any new nodes/leads in the chunk tree. So fix this by ensuring we don't accumulate a too large list of new block groups in a transaction's handles new_bgs list, inserting/updating the chunk tree for all accumulated new block groups and releasing the unused space from the chunk block reserve whenever the list becomes sufficiently large. This is a generic solution even though the problem currently can only happen when starting the writeout of the free space caches for all dirty block groups (btrfs_start_dirty_block_groups()). Reported-by: Omar Sandoval Signed-off-by: Filipe Manana Tested-by: Omar Sandoval Signed-off-by: Chris Mason diff --git a/fs/btrfs/extent-tree.c b/fs/btrfs/extent-tree.c index 171312d..07204bf 100644 --- a/fs/btrfs/extent-tree.c +++ b/fs/btrfs/extent-tree.c @@ -4227,6 +4227,24 @@ out: space_info->chunk_alloc = 0; spin_unlock(&space_info->lock); mutex_unlock(&fs_info->chunk_mutex); + /* + * When we allocate a new chunk we reserve space in the chunk block + * reserve to make sure we can COW nodes/leafs in the chunk tree or + * add new nodes/leafs to it if we end up needing to do it when + * inserting the chunk item and updating device items as part of the + * second phase of chunk allocation, performed by + * btrfs_finish_chunk_alloc(). So make sure we don't accumulate a + * large number of new block groups to create in our transaction + * handle's new_bgs list to avoid exhausting the chunk block reserve + * in extreme cases - like having a single transaction create many new + * block groups when starting to write out the free space caches of all + * the block groups that were made dirty during the lifetime of the + * transaction. + */ + if (trans->chunk_bytes_reserved >= (2 * 1024 * 1024ull)) { + btrfs_create_pending_block_groups(trans, trans->root); + btrfs_trans_release_chunk_metadata(trans); + } return ret; } -- cgit v0.10.2 From f2d0a123bcf16d1a9cf7942ddc98e0ef77862c2b Mon Sep 17 00:00:00 2001 From: "Eric W. Biederman" Date: Fri, 17 Jul 2015 14:15:30 -0500 Subject: mnt: Clarify and correct the disconnect logic in umount_tree rmdir mntpoint will result in an infinite loop when there is a mount locked on the mountpoint in another mount namespace. This is because the logic to test to see if a mount should be disconnected in umount_tree is buggy. Move the logic to decide if a mount should remain connected to it's mountpoint into it's own function disconnect_mount so that clarity of expression instead of terseness of expression becomes a virtue. When the conditions where it is invalid to leave a mount connected are first ruled out, the logic for deciding if a mount should be disconnected becomes much clearer and simpler. Fixes: e0c9c0afd2fc958ffa34b697972721d81df8a56f mnt: Update detach_mounts to leave mounts connected Fixes: ce07d891a0891d3c0d0c2d73d577490486b809e1 mnt: Honor MNT_LOCKED when detaching mounts Cc: stable@vger.kernel.org Signed-off-by: "Eric W. Biederman" diff --git a/fs/namespace.c b/fs/namespace.c index c7cb8a5..7829eab 100644 --- a/fs/namespace.c +++ b/fs/namespace.c @@ -1361,6 +1361,36 @@ enum umount_tree_flags { UMOUNT_PROPAGATE = 2, UMOUNT_CONNECTED = 4, }; + +static bool disconnect_mount(struct mount *mnt, enum umount_tree_flags how) +{ + /* Leaving mounts connected is only valid for lazy umounts */ + if (how & UMOUNT_SYNC) + return true; + + /* A mount without a parent has nothing to be connected to */ + if (!mnt_has_parent(mnt)) + return true; + + /* Because the reference counting rules change when mounts are + * unmounted and connected, umounted mounts may not be + * connected to mounted mounts. + */ + if (!(mnt->mnt_parent->mnt.mnt_flags & MNT_UMOUNT)) + return true; + + /* Has it been requested that the mount remain connected? */ + if (how & UMOUNT_CONNECTED) + return false; + + /* Is the mount locked such that it needs to remain connected? */ + if (IS_MNT_LOCKED(mnt)) + return false; + + /* By default disconnect the mount */ + return true; +} + /* * mount_lock must be held * namespace_sem must be held for write @@ -1398,10 +1428,7 @@ static void umount_tree(struct mount *mnt, enum umount_tree_flags how) if (how & UMOUNT_SYNC) p->mnt.mnt_flags |= MNT_SYNC_UMOUNT; - disconnect = !(((how & UMOUNT_CONNECTED) && - mnt_has_parent(p) && - (p->mnt_parent->mnt.mnt_flags & MNT_UMOUNT)) || - IS_MNT_LOCKED_AND_LAZY(p)); + disconnect = disconnect_mount(p, how); pin_insert_group(&p->mnt_umount, &p->mnt_parent->mnt, disconnect ? &unmounted : NULL); diff --git a/fs/pnode.h b/fs/pnode.h index 7114ce6..0fcdbe7 100644 --- a/fs/pnode.h +++ b/fs/pnode.h @@ -20,8 +20,6 @@ #define SET_MNT_MARK(m) ((m)->mnt.mnt_flags |= MNT_MARKED) #define CLEAR_MNT_MARK(m) ((m)->mnt.mnt_flags &= ~MNT_MARKED) #define IS_MNT_LOCKED(m) ((m)->mnt.mnt_flags & MNT_LOCKED) -#define IS_MNT_LOCKED_AND_LAZY(m) \ - (((m)->mnt.mnt_flags & (MNT_LOCKED|MNT_SYNC_UMOUNT)) == MNT_LOCKED) #define CL_EXPIRE 0x01 #define CL_SLAVE 0x02 -- cgit v0.10.2 From cabea695875e3a07313c205a9753c7416126dfa2 Mon Sep 17 00:00:00 2001 From: Sudip Mukherjee Date: Mon, 15 Jun 2015 20:05:49 +0530 Subject: parport: fix error handling After registering the device if exclusive access fails for any reason then we need to unregister the device to remove all references. Signed-off-by: Sudip Mukherjee Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/parport/share.c b/drivers/parport/share.c index 8067f54..d8079e3 100644 --- a/drivers/parport/share.c +++ b/drivers/parport/share.c @@ -907,7 +907,8 @@ parport_register_dev_model(struct parport *port, const char *name, spin_unlock(&port->physport->pardevice_lock); pr_debug("%s: cannot grant exclusive access for device %s\n", port->name, name); - goto err_put_dev; + device_unregister(&par_dev->dev); + goto err_put_port; } port->flags |= PARPORT_FLAG_EXCL; } -- cgit v0.10.2 From 23c405912b881e3ca516554efde852c2ad550b31 Mon Sep 17 00:00:00 2001 From: Sudip Mukherjee Date: Mon, 15 Jun 2015 20:05:50 +0530 Subject: parport: fix memory leak After the reference count becomes 0 when put_device() is called, it will execute the release callback where we are freeing all the allocated memory associated with the device. We missed freeing par_dev->state. Signed-off-by: Sudip Mukherjee Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/parport/share.c b/drivers/parport/share.c index d8079e3..1efec44 100644 --- a/drivers/parport/share.c +++ b/drivers/parport/share.c @@ -816,6 +816,7 @@ static void free_pardevice(struct device *dev) struct pardevice *par_dev = to_pardevice(dev); kfree(par_dev->name); + kfree(par_dev->state); kfree(par_dev); } -- cgit v0.10.2 From 68d35c7b3b1aa686e3039eb2626bf5e3ea8dbe81 Mon Sep 17 00:00:00 2001 From: Sudip Mukherjee Date: Mon, 15 Jun 2015 20:05:51 +0530 Subject: parport: fix freeing freed memory After the reference count becomes 0 when put_device() is called, it will execute the release callback where we are freeing all the allocated memory associated with the device. So if we just continue on the error path then we are again freeing devname and trying to dereference par_dev which has already been free-ed in the release callback. Signed-off-by: Sudip Mukherjee Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/parport/share.c b/drivers/parport/share.c index 1efec44..c02b5f2 100644 --- a/drivers/parport/share.c +++ b/drivers/parport/share.c @@ -892,8 +892,10 @@ parport_register_dev_model(struct parport *port, const char *name, par_dev->dev.release = free_pardevice; par_dev->devmodel = true; ret = device_register(&par_dev->dev); - if (ret) - goto err_put_dev; + if (ret) { + put_device(&par_dev->dev); + goto err_put_port; + } /* Chain this onto the list */ par_dev->prev = NULL; @@ -940,8 +942,6 @@ parport_register_dev_model(struct parport *port, const char *name, return par_dev; -err_put_dev: - put_device(&par_dev->dev); err_free_devname: kfree(devname); err_free_par_dev: -- cgit v0.10.2 From a63444375f08a18a6b2286351606f08d6e9fa63d Mon Sep 17 00:00:00 2001 From: Sudeep Dutt Date: Wed, 22 Jul 2015 11:50:10 -0700 Subject: misc: mic: scif bug fix for vmalloc_to_page crash v4.2-rc1 enabled huge page support for ioremap(..). Calling vmalloc_to_page after v4.2-rc1 results in the crash shown below on the host upon booting X100 coprocessors: BUG: unable to handle kernel paging request at ffffc47c00000000 IP: [] vmalloc_to_page+0x6c/0xb0 This patch fixes this crash by obtaining the fake struct page pointer which is required to be passed into dma_map_sg(..) by calling pfn_to_page(..) instead of vmalloc_to_page(..). Link: https://lkml.org/lkml/2015/7/18/110 Reviewed-by: Nikhil Rao Suggested-by: Toshi Kani Signed-off-by: Ashutosh Dixit Signed-off-by: Sudeep Dutt Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/misc/mic/scif/scif_nodeqp.c b/drivers/misc/mic/scif/scif_nodeqp.c index 41e3bdb..6dfdae3 100644 --- a/drivers/misc/mic/scif/scif_nodeqp.c +++ b/drivers/misc/mic/scif/scif_nodeqp.c @@ -357,7 +357,7 @@ static void scif_p2p_freesg(struct scatterlist *sg) } static struct scatterlist * -scif_p2p_setsg(void __iomem *va, int page_size, int page_cnt) +scif_p2p_setsg(phys_addr_t pa, int page_size, int page_cnt) { struct scatterlist *sg; struct page *page; @@ -368,16 +368,11 @@ scif_p2p_setsg(void __iomem *va, int page_size, int page_cnt) return NULL; sg_init_table(sg, page_cnt); for (i = 0; i < page_cnt; i++) { - page = vmalloc_to_page((void __force *)va); - if (!page) - goto p2p_sg_err; + page = pfn_to_page(pa >> PAGE_SHIFT); sg_set_page(&sg[i], page, page_size, 0); - va += page_size; + pa += page_size; } return sg; -p2p_sg_err: - kfree(sg); - return NULL; } /* Init p2p mappings required to access peerdev from scifdev */ @@ -395,14 +390,14 @@ scif_init_p2p_info(struct scif_dev *scifdev, struct scif_dev *peerdev) p2p = kzalloc(sizeof(*p2p), GFP_KERNEL); if (!p2p) return NULL; - p2p->ppi_sg[SCIF_PPI_MMIO] = scif_p2p_setsg(psdev->mmio->va, + p2p->ppi_sg[SCIF_PPI_MMIO] = scif_p2p_setsg(psdev->mmio->pa, PAGE_SIZE, num_mmio_pages); if (!p2p->ppi_sg[SCIF_PPI_MMIO]) goto free_p2p; p2p->sg_nentries[SCIF_PPI_MMIO] = num_mmio_pages; sg_page_shift = get_order(min(psdev->aper->len, (u64)(1 << 30))); num_aper_chunks = num_aper_pages >> (sg_page_shift - PAGE_SHIFT); - p2p->ppi_sg[SCIF_PPI_APER] = scif_p2p_setsg(psdev->aper->va, + p2p->ppi_sg[SCIF_PPI_APER] = scif_p2p_setsg(psdev->aper->pa, 1 << sg_page_shift, num_aper_chunks); p2p->sg_nentries[SCIF_PPI_APER] = num_aper_chunks; -- cgit v0.10.2 From 154322f47376fed6ab1e4b350aa45fffa15a61aa Mon Sep 17 00:00:00 2001 From: Tomas Winkler Date: Thu, 18 Jun 2015 11:41:03 +0300 Subject: mei: prevent unloading mei hw modules while the device is opened. chrdev_open() increases reference counter on cdev->owner. Instead of assigning the owner to mei subsystem, the owner has to be set to the underlaying HW module (mei_me or mei_txe), so once the device is opened the HW module cannot be unloaded. Cc: #3.17+ Signed-off-by: Tomas Winkler Signed-off-by: Alexander Usyskin Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/misc/mei/main.c b/drivers/misc/mei/main.c index 8eb0a95..e9513d6 100644 --- a/drivers/misc/mei/main.c +++ b/drivers/misc/mei/main.c @@ -682,7 +682,7 @@ int mei_register(struct mei_device *dev, struct device *parent) /* Fill in the data structures */ devno = MKDEV(MAJOR(mei_devt), dev->minor); cdev_init(&dev->cdev, &mei_fops); - dev->cdev.owner = mei_fops.owner; + dev->cdev.owner = parent->driver->owner; /* Add the device */ ret = cdev_add(&dev->cdev, devno, 1); -- cgit v0.10.2 From 10dc331ff5e7e4668c0f0c95b1a873aba9b70826 Mon Sep 17 00:00:00 2001 From: Xiao Guangrong Date: Thu, 16 Jul 2015 03:25:54 +0800 Subject: KVM: MTRR: fix memory type handling if MTRR is completely disabled Currently code uses default memory type if MTRR is fully disabled, fix it by using UC instead. Signed-off-by: Xiao Guangrong Tested-by: Alex Williamson Signed-off-by: Paolo Bonzini diff --git a/arch/x86/kvm/mtrr.c b/arch/x86/kvm/mtrr.c index de1d2d8..e275013 100644 --- a/arch/x86/kvm/mtrr.c +++ b/arch/x86/kvm/mtrr.c @@ -120,6 +120,16 @@ static u8 mtrr_default_type(struct kvm_mtrr *mtrr_state) return mtrr_state->deftype & IA32_MTRR_DEF_TYPE_TYPE_MASK; } +static u8 mtrr_disabled_type(void) +{ + /* + * Intel SDM 11.11.2.2: all MTRRs are disabled when + * IA32_MTRR_DEF_TYPE.E bit is cleared, and the UC + * memory type is applied to all of physical memory. + */ + return MTRR_TYPE_UNCACHABLE; +} + /* * Three terms are used in the following code: * - segment, it indicates the address segments covered by fixed MTRRs. @@ -434,6 +444,8 @@ struct mtrr_iter { /* output fields. */ int mem_type; + /* mtrr is completely disabled? */ + bool mtrr_disabled; /* [start, end) is not fully covered in MTRRs? */ bool partial_map; @@ -549,7 +561,7 @@ static void mtrr_lookup_var_next(struct mtrr_iter *iter) static void mtrr_lookup_start(struct mtrr_iter *iter) { if (!mtrr_is_enabled(iter->mtrr_state)) { - iter->partial_map = true; + iter->mtrr_disabled = true; return; } @@ -563,6 +575,7 @@ static void mtrr_lookup_init(struct mtrr_iter *iter, iter->mtrr_state = mtrr_state; iter->start = start; iter->end = end; + iter->mtrr_disabled = false; iter->partial_map = false; iter->fixed = false; iter->range = NULL; @@ -656,6 +669,9 @@ u8 kvm_mtrr_get_guest_memory_type(struct kvm_vcpu *vcpu, gfn_t gfn) return MTRR_TYPE_WRBACK; } + if (iter.mtrr_disabled) + return mtrr_disabled_type(); + /* It is not covered by MTRRs. */ if (iter.partial_map) { /* @@ -689,6 +705,9 @@ bool kvm_mtrr_check_gfn_range_consistency(struct kvm_vcpu *vcpu, gfn_t gfn, return false; } + if (iter.mtrr_disabled) + return true; + if (!iter.partial_map) return true; -- cgit v0.10.2 From 3e5d2fdceda172554e681b68c853bf5d08205bbf Mon Sep 17 00:00:00 2001 From: Xiao Guangrong Date: Thu, 16 Jul 2015 03:25:55 +0800 Subject: KVM: MTRR: simplify kvm_mtrr_get_guest_memory_type kvm_mtrr_get_guest_memory_type never returns -1 which is implied in the current code since if @type = -1 (means no MTRR contains the range), iter.partial_map must be true Simplify the code to indicate this fact Signed-off-by: Xiao Guangrong Tested-by: Alex Williamson Signed-off-by: Paolo Bonzini diff --git a/arch/x86/kvm/mtrr.c b/arch/x86/kvm/mtrr.c index e275013..dc0a84a 100644 --- a/arch/x86/kvm/mtrr.c +++ b/arch/x86/kvm/mtrr.c @@ -672,15 +672,16 @@ u8 kvm_mtrr_get_guest_memory_type(struct kvm_vcpu *vcpu, gfn_t gfn) if (iter.mtrr_disabled) return mtrr_disabled_type(); - /* It is not covered by MTRRs. */ - if (iter.partial_map) { - /* - * We just check one page, partially covered by MTRRs is - * impossible. - */ - WARN_ON(type != -1); - type = mtrr_default_type(mtrr_state); - } + /* + * We just check one page, partially covered by MTRRs is + * impossible. + */ + WARN_ON(iter.partial_map); + + /* not contained in any MTRRs. */ + if (type == -1) + return mtrr_default_type(mtrr_state); + return type; } EXPORT_SYMBOL_GPL(kvm_mtrr_get_guest_memory_type); -- cgit v0.10.2 From 41dbc6bcd9722bba8f09256655d060af30c63d34 Mon Sep 17 00:00:00 2001 From: Paolo Bonzini Date: Thu, 23 Jul 2015 08:22:45 +0200 Subject: KVM: x86: introduce kvm_check_has_quirk The logic of the disabled_quirks field usually results in a double negation. Wrap it in a simple function that checks the bit and negates it. Based on a patch from Xiao Guangrong. Signed-off-by: Paolo Bonzini diff --git a/arch/x86/kvm/lapic.c b/arch/x86/kvm/lapic.c index 954e98a..1c42544 100644 --- a/arch/x86/kvm/lapic.c +++ b/arch/x86/kvm/lapic.c @@ -1595,7 +1595,7 @@ void kvm_lapic_reset(struct kvm_vcpu *vcpu, bool init_event) for (i = 0; i < APIC_LVT_NUM; i++) apic_set_reg(apic, APIC_LVTT + 0x10 * i, APIC_LVT_MASKED); apic_update_lvtt(apic); - if (!(vcpu->kvm->arch.disabled_quirks & KVM_QUIRK_LINT0_REENABLED)) + if (kvm_check_has_quirk(vcpu->kvm, KVM_QUIRK_LINT0_REENABLED)) apic_set_reg(apic, APIC_LVT0, SET_APIC_DELIVERY_MODE(0, APIC_MODE_EXTINT)); apic_manage_nmi_watchdog(apic, kvm_apic_get_reg(apic, APIC_LVT0)); diff --git a/arch/x86/kvm/svm.c b/arch/x86/kvm/svm.c index bbc678a..8cbec76 100644 --- a/arch/x86/kvm/svm.c +++ b/arch/x86/kvm/svm.c @@ -1672,7 +1672,7 @@ static void svm_set_cr0(struct kvm_vcpu *vcpu, unsigned long cr0) * does not do it - this results in some delay at * reboot */ - if (!(vcpu->kvm->arch.disabled_quirks & KVM_QUIRK_CD_NW_CLEARED)) + if (kvm_check_has_quirk(vcpu->kvm, KVM_QUIRK_CD_NW_CLEARED)) cr0 &= ~(X86_CR0_CD | X86_CR0_NW); svm->vmcb->save.cr0 = cr0; mark_dirty(svm->vmcb, VMCB_CR); diff --git a/arch/x86/kvm/x86.h b/arch/x86/kvm/x86.h index edc8cdc..0ca2f3e 100644 --- a/arch/x86/kvm/x86.h +++ b/arch/x86/kvm/x86.h @@ -147,6 +147,11 @@ static inline void kvm_register_writel(struct kvm_vcpu *vcpu, return kvm_register_write(vcpu, reg, val); } +static inline bool kvm_check_has_quirk(struct kvm *kvm, u64 quirk) +{ + return !(kvm->arch.disabled_quirks & quirk); +} + void kvm_before_handle_nmi(struct kvm_vcpu *vcpu); void kvm_after_handle_nmi(struct kvm_vcpu *vcpu); void kvm_set_pending_timer(struct kvm_vcpu *vcpu); -- cgit v0.10.2 From fb279950ba02e3210a16b11ecfa8871f3ee0ca49 Mon Sep 17 00:00:00 2001 From: Xiao Guangrong Date: Thu, 16 Jul 2015 03:25:56 +0800 Subject: KVM: vmx: obey KVM_QUIRK_CD_NW_CLEARED OVMF depends on WB to boot fast, because it only clears caches after it has set up MTRRs---which is too late. Let's do writeback if CR0.CD is set to make it happy, similar to what SVM is already doing. Signed-off-by: Xiao Guangrong Tested-by: Alex Williamson Signed-off-by: Paolo Bonzini diff --git a/arch/x86/kvm/vmx.c b/arch/x86/kvm/vmx.c index 5b4e938..e9e1fc5 100644 --- a/arch/x86/kvm/vmx.c +++ b/arch/x86/kvm/vmx.c @@ -8650,7 +8650,10 @@ static u64 vmx_get_mt_mask(struct kvm_vcpu *vcpu, gfn_t gfn, bool is_mmio) if (kvm_read_cr0(vcpu) & X86_CR0_CD) { ipat = VMX_EPT_IPAT_BIT; - cache = MTRR_TYPE_UNCACHABLE; + if (kvm_check_has_quirk(vcpu->kvm, KVM_QUIRK_CD_NW_CLEARED)) + cache = MTRR_TYPE_WRBACK; + else + cache = MTRR_TYPE_UNCACHABLE; goto exit; } -- cgit v0.10.2 From 0da029ed7ee5fdf49a2a0e14160c3e9999be9292 Mon Sep 17 00:00:00 2001 From: Paolo Bonzini Date: Thu, 23 Jul 2015 08:24:42 +0200 Subject: KVM: x86: rename quirk constants to KVM_X86_QUIRK_* Make them clearly architecture-dependent; the capability is valid for all architectures, but the argument is not. Signed-off-by: Paolo Bonzini diff --git a/arch/x86/include/uapi/asm/kvm.h b/arch/x86/include/uapi/asm/kvm.h index a4ae82e..cd54147 100644 --- a/arch/x86/include/uapi/asm/kvm.h +++ b/arch/x86/include/uapi/asm/kvm.h @@ -354,7 +354,7 @@ struct kvm_xcrs { struct kvm_sync_regs { }; -#define KVM_QUIRK_LINT0_REENABLED (1 << 0) -#define KVM_QUIRK_CD_NW_CLEARED (1 << 1) +#define KVM_X86_QUIRK_LINT0_REENABLED (1 << 0) +#define KVM_X86_QUIRK_CD_NW_CLEARED (1 << 1) #endif /* _ASM_X86_KVM_H */ diff --git a/arch/x86/kvm/lapic.c b/arch/x86/kvm/lapic.c index 1c42544..2a5ca97 100644 --- a/arch/x86/kvm/lapic.c +++ b/arch/x86/kvm/lapic.c @@ -1595,7 +1595,7 @@ void kvm_lapic_reset(struct kvm_vcpu *vcpu, bool init_event) for (i = 0; i < APIC_LVT_NUM; i++) apic_set_reg(apic, APIC_LVTT + 0x10 * i, APIC_LVT_MASKED); apic_update_lvtt(apic); - if (kvm_check_has_quirk(vcpu->kvm, KVM_QUIRK_LINT0_REENABLED)) + if (kvm_check_has_quirk(vcpu->kvm, KVM_X86_QUIRK_LINT0_REENABLED)) apic_set_reg(apic, APIC_LVT0, SET_APIC_DELIVERY_MODE(0, APIC_MODE_EXTINT)); apic_manage_nmi_watchdog(apic, kvm_apic_get_reg(apic, APIC_LVT0)); diff --git a/arch/x86/kvm/svm.c b/arch/x86/kvm/svm.c index 8cbec76..8e0c084 100644 --- a/arch/x86/kvm/svm.c +++ b/arch/x86/kvm/svm.c @@ -1672,7 +1672,7 @@ static void svm_set_cr0(struct kvm_vcpu *vcpu, unsigned long cr0) * does not do it - this results in some delay at * reboot */ - if (kvm_check_has_quirk(vcpu->kvm, KVM_QUIRK_CD_NW_CLEARED)) + if (kvm_check_has_quirk(vcpu->kvm, KVM_X86_QUIRK_CD_NW_CLEARED)) cr0 &= ~(X86_CR0_CD | X86_CR0_NW); svm->vmcb->save.cr0 = cr0; mark_dirty(svm->vmcb, VMCB_CR); diff --git a/arch/x86/kvm/vmx.c b/arch/x86/kvm/vmx.c index e9e1fc5..83b7b5c 100644 --- a/arch/x86/kvm/vmx.c +++ b/arch/x86/kvm/vmx.c @@ -8650,7 +8650,7 @@ static u64 vmx_get_mt_mask(struct kvm_vcpu *vcpu, gfn_t gfn, bool is_mmio) if (kvm_read_cr0(vcpu) & X86_CR0_CD) { ipat = VMX_EPT_IPAT_BIT; - if (kvm_check_has_quirk(vcpu->kvm, KVM_QUIRK_CD_NW_CLEARED)) + if (kvm_check_has_quirk(vcpu->kvm, KVM_X86_QUIRK_CD_NW_CLEARED)) cache = MTRR_TYPE_WRBACK; else cache = MTRR_TYPE_UNCACHABLE; -- cgit v0.10.2 From 3ba3a73e9f204ce7904cae7d14be399b467c4fef Mon Sep 17 00:00:00 2001 From: Alexey Kardashevskiy Date: Mon, 20 Jul 2015 20:45:51 +1000 Subject: powerpc/powernv/ioda2: Fix calculation for memory allocated for TCE table The existing code stores the amount of memory allocated for a TCE table. At the moment it uses @offset which is a virtual offset in the TCE table which is only correct for a one level tables and it does not include memory allocated for intermediate levels. When multilevel TCE table is requested, WARN_ON in tce_iommu_create_table() prints a warning. This adds an additional counter to pnv_pci_ioda2_table_do_alloc_pages() to count actually allocated memory. Signed-off-by: Alexey Kardashevskiy Reviewed-by: David Gibson Signed-off-by: Michael Ellerman diff --git a/arch/powerpc/platforms/powernv/pci-ioda.c b/arch/powerpc/platforms/powernv/pci-ioda.c index 5738d31..85cbc96 100644 --- a/arch/powerpc/platforms/powernv/pci-ioda.c +++ b/arch/powerpc/platforms/powernv/pci-ioda.c @@ -2220,7 +2220,7 @@ static void pnv_pci_ioda_setup_opal_tce_kill(struct pnv_phb *phb) static __be64 *pnv_pci_ioda2_table_do_alloc_pages(int nid, unsigned shift, unsigned levels, unsigned long limit, - unsigned long *current_offset) + unsigned long *current_offset, unsigned long *total_allocated) { struct page *tce_mem = NULL; __be64 *addr, *tmp; @@ -2236,6 +2236,7 @@ static __be64 *pnv_pci_ioda2_table_do_alloc_pages(int nid, unsigned shift, } addr = page_address(tce_mem); memset(addr, 0, allocated); + *total_allocated += allocated; --levels; if (!levels) { @@ -2245,7 +2246,7 @@ static __be64 *pnv_pci_ioda2_table_do_alloc_pages(int nid, unsigned shift, for (i = 0; i < entries; ++i) { tmp = pnv_pci_ioda2_table_do_alloc_pages(nid, shift, - levels, limit, current_offset); + levels, limit, current_offset, total_allocated); if (!tmp) break; @@ -2267,7 +2268,7 @@ static long pnv_pci_ioda2_table_alloc_pages(int nid, __u64 bus_offset, struct iommu_table *tbl) { void *addr; - unsigned long offset = 0, level_shift; + unsigned long offset = 0, level_shift, total_allocated = 0; const unsigned window_shift = ilog2(window_size); unsigned entries_shift = window_shift - page_shift; unsigned table_shift = max_t(unsigned, entries_shift + 3, PAGE_SHIFT); @@ -2286,7 +2287,7 @@ static long pnv_pci_ioda2_table_alloc_pages(int nid, __u64 bus_offset, /* Allocate TCE table */ addr = pnv_pci_ioda2_table_do_alloc_pages(nid, level_shift, - levels, tce_table_size, &offset); + levels, tce_table_size, &offset, &total_allocated); /* addr==NULL means that the first level allocation failed */ if (!addr) @@ -2308,7 +2309,7 @@ static long pnv_pci_ioda2_table_alloc_pages(int nid, __u64 bus_offset, page_shift); tbl->it_level_size = 1ULL << (level_shift - 3); tbl->it_indirect_levels = levels - 1; - tbl->it_allocated_size = offset; + tbl->it_allocated_size = total_allocated; pr_devel("Created TCE table: ws=%08llx ts=%lx @%08llx\n", window_size, tce_table_size, bus_offset); -- cgit v0.10.2 From 120d200a86273d1840d091959dca65617cd9eff2 Mon Sep 17 00:00:00 2001 From: Luis Henriques Date: Fri, 17 Jul 2015 14:20:31 +0100 Subject: macintosh/ans-lcd: fix build failure after module_init/exit relocation After commit 0fd972a7d91d ("module: relocate module_init from init.h to module.h") ans-lcd module fails to build with: drivers/macintosh/ans-lcd.c:201:1: warning: data definition has no type or storage class [enabled by default] module_init(anslcd_init); ^ drivers/macintosh/ans-lcd.c:201:1: error: type defaults to 'int' in declaration of 'module_init' [-Werror=implicit-int] drivers/macintosh/ans-lcd.c:201:1: warning: parameter names (without types) in function declaration [enabled by default] drivers/macintosh/ans-lcd.c:202:1: warning: data definition has no type or storage class [enabled by default] module_exit(anslcd_exit); ^ drivers/macintosh/ans-lcd.c:202:1: error: type defaults to 'int' in declaration of 'module_exit' [-Werror=implicit-int] drivers/macintosh/ans-lcd.c:202:1: warning: parameter names (without types) in function declaration [enabled by default] drivers/macintosh/ans-lcd.c:155:1: warning: 'anslcd_init' defined but not used [-Wunused-function] anslcd_init(void) ^ drivers/macintosh/ans-lcd.c:195:1: warning: 'anslcd_exit' defined but not used [-Wunused-function] anslcd_exit(void) ^ This commit fixes it by replacing linux/init.h by linux/module.h. Fixes: 0fd972a7d91d ("module: relocate module_init from init.h to module.h") Signed-off-by: Luis Henriques Acked-by: Paul Gortmaker Signed-off-by: Michael Ellerman diff --git a/drivers/macintosh/ans-lcd.c b/drivers/macintosh/ans-lcd.c index 1a57e88..cd35079 100644 --- a/drivers/macintosh/ans-lcd.c +++ b/drivers/macintosh/ans-lcd.c @@ -7,7 +7,7 @@ #include #include #include -#include +#include #include #include -- cgit v0.10.2 From 46ebb7af7b93792de65e124e1ab8b89a108a41f2 Mon Sep 17 00:00:00 2001 From: Alex Williamson Date: Tue, 14 Jul 2015 14:48:53 -0600 Subject: iommu/vt-d: Fix VM domain ID leak This continues the attempt to fix commit fb170fb4c548 ("iommu/vt-d: Introduce helper functions to make code symmetric for readability"). The previous attempt in commit 71684406905f ("iommu/vt-d: Detach domain *only* from attached iommus") overlooked the fact that dmar_domain.iommu_bmp gets cleared for VM domains when devices are detached: intel_iommu_detach_device domain_remove_one_dev_info domain_detach_iommu The domain is detached from the iommu, but the iommu is still attached to the domain, for whatever reason. Thus when we get to domain_exit(), we can't rely on iommu_bmp for VM domains to find the active iommus, we must check them all. Without that, the corresponding bit in intel_iommu.domain_ids doesn't get cleared and repeated VM domain creation and destruction will run out of domain IDs. Meanwhile we still can't call iommu_detach_domain() on arbitrary non-VM domains or we risk clearing in-use domain IDs, as 71684406905f attempted to address. It's tempting to modify iommu_detach_domain() to test the domain iommu_bmp, but the call ordering from domain_remove_one_dev_info() prevents it being able to work as fb170fb4c548 seems to have intended. Caching of unused VM domains on the iommu object seems to be the root of the problem, but this code is far too fragile for that kind of rework to be proposed for stable, so we simply revert this chunk to its state prior to fb170fb4c548. Fixes: fb170fb4c548 ("iommu/vt-d: Introduce helper functions to make code symmetric for readability") Fixes: 71684406905f ("iommu/vt-d: Detach domain *only* from attached iommus") Signed-off-by: Alex Williamson Cc: Jiang Liu Cc: stable@vger.kernel.org # v3.17+ Signed-off-by: Joerg Roedel diff --git a/drivers/iommu/intel-iommu.c b/drivers/iommu/intel-iommu.c index a98a7b2..0649b94 100644 --- a/drivers/iommu/intel-iommu.c +++ b/drivers/iommu/intel-iommu.c @@ -1830,8 +1830,9 @@ static int domain_init(struct dmar_domain *domain, int guest_width) static void domain_exit(struct dmar_domain *domain) { + struct dmar_drhd_unit *drhd; + struct intel_iommu *iommu; struct page *freelist = NULL; - int i; /* Domain 0 is reserved, so dont process it */ if (!domain) @@ -1851,8 +1852,10 @@ static void domain_exit(struct dmar_domain *domain) /* clear attached or cached domains */ rcu_read_lock(); - for_each_set_bit(i, domain->iommu_bmp, g_num_of_iommus) - iommu_detach_domain(domain, g_iommus[i]); + for_each_active_iommu(iommu, drhd) + if (domain_type_is_vm(domain) || + test_bit(iommu->seq_id, domain->iommu_bmp)) + iommu_detach_domain(domain, iommu); rcu_read_unlock(); dma_free_pagelist(freelist); -- cgit v0.10.2 From 9a258afa928b45e6dd2efcac46ccf7eea705d35a Mon Sep 17 00:00:00 2001 From: Roger Quadros Date: Thu, 16 Jul 2015 16:16:44 +0300 Subject: ARM: OMAP2+: hwmod: Fix _wait_target_ready() for hwmods without sysc For hwmods without sysc, _init_mpu_rt_base(oh) won't be called and so _find_mpu_rt_port(oh) will return NULL thus preventing ready state check on those modules after the module is enabled. This can potentially cause a bus access error if the module is accessed before the module is ready. Fix this by unconditionally calling _init_mpu_rt_base() during hwmod _init(). Do ioremap only if we need SYSC access. Eventhough _wait_target_ready() check doesn't really need MPU RT port but just the PRCM registers, we still mandate that the hwmod must have an MPU RT port if ready state check needs to be done. Else it would mean that the module is not accessible by MPU so there is no point in waiting for target to be ready. e.g. this fixes the below DCAN bus access error on AM437x-gp-evm. [ 16.672978] ------------[ cut here ]------------ [ 16.677885] WARNING: CPU: 0 PID: 1580 at drivers/bus/omap_l3_noc.c:147 l3_interrupt_handler+0x234/0x35c() [ 16.687946] 44000000.ocp:L3 Custom Error: MASTER M2 (64-bit) TARGET L4_PER_0 (Read): Data Access in User mode during Functional access [ 16.700654] Modules linked in: xhci_hcd btwilink ti_vpfe dwc3 videobuf2_core ov2659 bluetooth v4l2_common videodev ti_am335x_adc kfifo_buf industrialio c_can_platform videobuf2_dma_contig media snd_soc_tlv320aic3x pixcir_i2c_ts c_can dc [ 16.731144] CPU: 0 PID: 1580 Comm: rpc.statd Not tainted 3.14.26-02561-gf733aa036398 #180 [ 16.739747] Backtrace: [ 16.742336] [] (dump_backtrace) from [] (show_stack+0x18/0x1c) [ 16.750285] r6:00000093 r5:00000009 r4:eab5b8a8 r3:00000000 [ 16.756252] [] (show_stack) from [] (dump_stack+0x20/0x28) [ 16.763870] [] (dump_stack) from [] (warn_slowpath_common+0x6c/0x8c) [ 16.772408] [] (warn_slowpath_common) from [] (warn_slowpath_fmt+0x38/0x40) [ 16.781550] r8:c05d1f90 r7:c0730844 r6:c0730448 r5:80080003 r4:ed0cd210 [ 16.788626] [] (warn_slowpath_fmt) from [] (l3_interrupt_handler+0x234/0x35c) [ 16.797968] r3:ed0cd480 r2:c0730508 [ 16.801747] [] (l3_interrupt_handler) from [] (handle_irq_event_percpu+0x54/0x1bc) [ 16.811533] r10:ed005600 r9:c084855b r8:0000002a r7:00000000 r6:00000000 r5:0000002a [ 16.819780] r4:ed0e6d80 [ 16.822453] [] (handle_irq_event_percpu) from [] (handle_irq_event+0x30/0x40) [ 16.831789] r10:eb2b6938 r9:eb2b6960 r8:bf011420 r7:fa240100 r6:00000000 r5:0000002a [ 16.840052] r4:ed005600 [ 16.842744] [] (handle_irq_event) from [] (handle_fasteoi_irq+0x74/0x128) [ 16.851702] r4:ed005600 r3:00000000 [ 16.855479] [] (handle_fasteoi_irq) from [] (generic_handle_irq+0x28/0x38) [ 16.864523] r4:0000002a r3:c0066164 [ 16.868294] [] (generic_handle_irq) from [] (handle_IRQ+0x38/0x8c) [ 16.876612] r4:c081c640 r3:00000202 [ 16.880380] [] (handle_IRQ) from [] (gic_handle_irq+0x30/0x5c) [ 16.888328] r6:eab5ba38 r5:c0804460 r4:fa24010c r3:00000100 [ 16.894303] [] (gic_handle_irq) from [] (__irq_svc+0x40/0x50) [ 16.902193] Exception stack(0xeab5ba38 to 0xeab5ba80) [ 16.907499] ba20: 00000000 00000006 [ 16.916108] ba40: fa1d0000 fa1d0008 ed3d3000 eab5bab4 ed3d3460 c0842af4 bf011420 eb2b6960 [ 16.924716] ba60: eb2b6938 eab5ba8c eab5ba90 eab5ba80 bf035220 bf07702c 600f0013 ffffffff [ 16.933317] r7:eab5ba6c r6:ffffffff r5:600f0013 r4:bf07702c [ 16.939317] [] (c_can_plat_read_reg_aligned_to_16bit [c_can_platform]) from [] (c_can_get_berr_counter+0x38/0x64 [c_can]) [ 16.952696] [] (c_can_get_berr_counter [c_can]) from [] (can_fill_info+0x124/0x15c [can_dev]) [ 16.963480] r5:ec8c9740 r4:ed3d3000 [ 16.967253] [] (can_fill_info [can_dev]) from [] (rtnl_fill_ifinfo+0x58c/0x8fc) [ 16.976749] r6:ec8c9740 r5:ed3d3000 r4:eb2b6780 [ 16.981613] [] (rtnl_fill_ifinfo) from [] (rtnl_dump_ifinfo+0xf0/0x1dc) [ 16.990401] r10:ec8c9740 r9:00000000 r8:00000000 r7:00000000 r6:ebd4d1b4 r5:ed3d3000 [ 16.998671] r4:00000000 [ 17.001342] [] (rtnl_dump_ifinfo) from [] (netlink_dump+0xa8/0x1e0) [ 17.009772] r10:00000000 r9:00000000 r8:c0503318 r7:ebf3e6c0 r6:ebd4d1b4 r5:ec8c9740 [ 17.018050] r4:ebd4d000 [ 17.020714] [] (netlink_dump) from [] (__netlink_dump_start+0x104/0x154) [ 17.029591] r6:eab5bd34 r5:ec8c9980 r4:ebd4d000 [ 17.034454] [] (__netlink_dump_start) from [] (rtnetlink_rcv_msg+0x110/0x1f4) [ 17.043778] r7:00000000 r6:ec8c9980 r5:00000f40 r4:ebf3e6c0 [ 17.049743] [] (rtnetlink_rcv_msg) from [] (netlink_rcv_skb+0xb4/0xc8) [ 17.058449] r8:eab5bdac r7:ec8c9980 r6:c05054f4 r5:ec8c9980 r4:ebf3e6c0 [ 17.065534] [] (netlink_rcv_skb) from [] (rtnetlink_rcv+0x24/0x2c) [ 17.073854] r6:ebd4d000 r5:00000014 r4:ec8c9980 r3:c0504110 [ 17.079846] [] (rtnetlink_rcv) from [] (netlink_unicast+0x180/0x1ec) [ 17.088363] r4:ed0c6800 r3:c0504110 [ 17.092113] [] (netlink_unicast) from [] (netlink_sendmsg+0x2ac/0x380) [ 17.100813] r10:00000000 r8:00000008 r7:ec8c9980 r6:ebd4d000 r5:eab5be70 r4:eab5bee4 [ 17.109083] [] (netlink_sendmsg) from [] (sock_sendmsg+0x90/0xb0) [ 17.117305] r10:00000000 r9:eab5a000 r8:becdda3c r7:0000000c r6:ea978400 r5:eab5be70 [ 17.125563] r4:c05103c4 [ 17.128225] [] (sock_sendmsg) from [] (SyS_sendto+0xb8/0xdc) [ 17.136001] r6:becdda5c r5:00000014 r4:ecd37040 [ 17.140876] [] (SyS_sendto) from [] (ret_fast_syscall+0x0/0x30) [ 17.148923] r10:00000000 r8:c000e804 r7:00000122 r6:becdda5c r5:0000000c r4:becdda5c [ 17.157169] ---[ end trace 2b71e15b38f58bad ]--- Fixes: 6423d6df1440 ("ARM: OMAP2+: hwmod: check for module address space during init") Signed-off-by: Roger Quadros Signed-off-by: Paul Walmsley Cc: diff --git a/arch/arm/mach-omap2/omap_hwmod.c b/arch/arm/mach-omap2/omap_hwmod.c index d78c12e..486cc4d 100644 --- a/arch/arm/mach-omap2/omap_hwmod.c +++ b/arch/arm/mach-omap2/omap_hwmod.c @@ -2373,6 +2373,9 @@ static int of_dev_hwmod_lookup(struct device_node *np, * registers. This address is needed early so the OCP registers that * are part of the device's address space can be ioremapped properly. * + * If SYSC access is not needed, the registers will not be remapped + * and non-availability of MPU access is not treated as an error. + * * Returns 0 on success, -EINVAL if an invalid hwmod is passed, and * -ENXIO on absent or invalid register target address space. */ @@ -2387,6 +2390,11 @@ static int __init _init_mpu_rt_base(struct omap_hwmod *oh, void *data, _save_mpu_port_index(oh); + /* if we don't need sysc access we don't need to ioremap */ + if (!oh->class->sysc) + return 0; + + /* we can't continue without MPU PORT if we need sysc access */ if (oh->_int_flags & _HWMOD_NO_MPU_PORT) return -ENXIO; @@ -2396,8 +2404,10 @@ static int __init _init_mpu_rt_base(struct omap_hwmod *oh, void *data, oh->name); /* Extract the IO space from device tree blob */ - if (!np) + if (!np) { + pr_err("omap_hwmod: %s: no dt node\n", oh->name); return -ENXIO; + } va_start = of_iomap(np, index + oh->mpu_rt_idx); } else { @@ -2456,13 +2466,11 @@ static int __init _init(struct omap_hwmod *oh, void *data) oh->name, np->name); } - if (oh->class->sysc) { - r = _init_mpu_rt_base(oh, NULL, index, np); - if (r < 0) { - WARN(1, "omap_hwmod: %s: doesn't have mpu register target base\n", - oh->name); - return 0; - } + r = _init_mpu_rt_base(oh, NULL, index, np); + if (r < 0) { + WARN(1, "omap_hwmod: %s: doesn't have mpu register target base\n", + oh->name); + return 0; } r = _init_clocks(oh, NULL); -- cgit v0.10.2 From 25ba265390c09b0a2b2f3fd9ba82e37248b7a371 Mon Sep 17 00:00:00 2001 From: Johan Hedberg Date: Mon, 20 Jul 2015 20:31:25 +0300 Subject: Bluetooth: Fix NULL pointer dereference in smp_conn_security The l2cap_conn->smp pointer may be NULL for various valid reasons where SMP has failed to initialize properly. One such scenario is when crypto support is missing, another when the adapter has been powered on through a legacy method. The smp_conn_security() function should have the appropriate check for this situation to avoid NULL pointer dereferences. Signed-off-by: Johan Hedberg Signed-off-by: Marcel Holtmann Cc: stable@vger.kernel.org # 4.0+ diff --git a/net/bluetooth/smp.c b/net/bluetooth/smp.c index 3d0f7d2..ad82324 100644 --- a/net/bluetooth/smp.c +++ b/net/bluetooth/smp.c @@ -2312,6 +2312,10 @@ int smp_conn_security(struct hci_conn *hcon, __u8 sec_level) return 1; chan = conn->smp; + if (!chan) { + BT_ERR("SMP security requested but not available"); + return 1; + } if (!hci_dev_test_flag(hcon->hdev, HCI_LE_ENABLED)) return 1; -- cgit v0.10.2 From d84b272a12a6aa96897372449e47e57f72f2730f Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Michel=20D=C3=A4nzer?= Date: Thu, 16 Jul 2015 10:45:10 +0900 Subject: drm/amdgpu/dce11: Re-set VBLANK interrupt state when enabling a CRTC MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Something (ATOM BIOS?) seems to be clobbering the LB_INTERRUPT_MASK register while the CRTC is off, which caused e.g. glxgears or gnome-shell to hang after a modeset. Reviewed-and-Tested-by: Alex Deucher Tested-by: Sonny Jiang Signed-off-by: Michel Dänzer diff --git a/drivers/gpu/drm/amd/amdgpu/dce_v11_0.c b/drivers/gpu/drm/amd/amdgpu/dce_v11_0.c index 7f7abb0..dcb402e 100644 --- a/drivers/gpu/drm/amd/amdgpu/dce_v11_0.c +++ b/drivers/gpu/drm/amd/amdgpu/dce_v11_0.c @@ -2631,6 +2631,7 @@ static void dce_v11_0_crtc_dpms(struct drm_crtc *crtc, int mode) struct drm_device *dev = crtc->dev; struct amdgpu_device *adev = dev->dev_private; struct amdgpu_crtc *amdgpu_crtc = to_amdgpu_crtc(crtc); + unsigned type; switch (mode) { case DRM_MODE_DPMS_ON: @@ -2639,6 +2640,9 @@ static void dce_v11_0_crtc_dpms(struct drm_crtc *crtc, int mode) dce_v11_0_vga_enable(crtc, true); amdgpu_atombios_crtc_blank(crtc, ATOM_DISABLE); dce_v11_0_vga_enable(crtc, false); + /* Make sure VBLANK interrupt is still enabled */ + type = amdgpu_crtc_idx_to_irq_type(adev, amdgpu_crtc->crtc_id); + amdgpu_irq_update(adev, &adev->crtc_irq, type); drm_vblank_post_modeset(dev, amdgpu_crtc->crtc_id); dce_v11_0_crtc_load_lut(crtc); break; -- cgit v0.10.2 From 5e6775abb5ce7ada76fa17c67857576176d0b60f Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Michel=20D=C3=A4nzer?= Date: Fri, 17 Jul 2015 11:20:18 +0900 Subject: drm/amdgpu/dce10: Re-set VBLANK interrupt state when enabling a CRTC MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit This doesn't seem strictly necessary with Tonga right now, but that might change with future power management enhancements. Reviewed-by: Alex Deucher Signed-off-by: Michel Dänzer diff --git a/drivers/gpu/drm/amd/amdgpu/dce_v10_0.c b/drivers/gpu/drm/amd/amdgpu/dce_v10_0.c index 6e77964..e70a26f 100644 --- a/drivers/gpu/drm/amd/amdgpu/dce_v10_0.c +++ b/drivers/gpu/drm/amd/amdgpu/dce_v10_0.c @@ -2632,6 +2632,7 @@ static void dce_v10_0_crtc_dpms(struct drm_crtc *crtc, int mode) struct drm_device *dev = crtc->dev; struct amdgpu_device *adev = dev->dev_private; struct amdgpu_crtc *amdgpu_crtc = to_amdgpu_crtc(crtc); + unsigned type; switch (mode) { case DRM_MODE_DPMS_ON: @@ -2640,6 +2641,9 @@ static void dce_v10_0_crtc_dpms(struct drm_crtc *crtc, int mode) dce_v10_0_vga_enable(crtc, true); amdgpu_atombios_crtc_blank(crtc, ATOM_DISABLE); dce_v10_0_vga_enable(crtc, false); + /* Make sure VBLANK interrupt is still enabled */ + type = amdgpu_crtc_idx_to_irq_type(adev, amdgpu_crtc->crtc_id); + amdgpu_irq_update(adev, &adev->crtc_irq, type); drm_vblank_post_modeset(dev, amdgpu_crtc->crtc_id); dce_v10_0_crtc_load_lut(crtc); break; -- cgit v0.10.2 From 6a585777c832f45c05d1aa8a4872b2b292a908c2 Mon Sep 17 00:00:00 2001 From: Alex Deucher Date: Fri, 10 Jul 2015 14:16:24 -0400 Subject: drm/amdgpu: implement VCE 3.0 harvesting support (v4) MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit For boards with bad VCE blocks, only configure the working block. v2: use the harvest info for pipe setup v3: fix mask check as noted by Leo v4: add dGPU support Reviewed-by: Christian König (v2) Signed-off-by: Alex Deucher diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu.h b/drivers/gpu/drm/amd/amdgpu/amdgpu.h index 0165783..e9fde72 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu.h @@ -1614,6 +1614,9 @@ struct amdgpu_uvd { #define AMDGPU_MAX_VCE_HANDLES 16 #define AMDGPU_VCE_FIRMWARE_OFFSET 256 +#define AMDGPU_VCE_HARVEST_VCE0 (1 << 0) +#define AMDGPU_VCE_HARVEST_VCE1 (1 << 1) + struct amdgpu_vce { struct amdgpu_bo *vcpu_bo; uint64_t gpu_addr; @@ -1626,6 +1629,7 @@ struct amdgpu_vce { const struct firmware *fw; /* VCE firmware */ struct amdgpu_ring ring[AMDGPU_MAX_VCE_RINGS]; struct amdgpu_irq_src irq; + unsigned harvest_config; }; /* diff --git a/drivers/gpu/drm/amd/amdgpu/vce_v3_0.c b/drivers/gpu/drm/amd/amdgpu/vce_v3_0.c index d62c4002..d1064ca 100644 --- a/drivers/gpu/drm/amd/amdgpu/vce_v3_0.c +++ b/drivers/gpu/drm/amd/amdgpu/vce_v3_0.c @@ -35,6 +35,8 @@ #include "oss/oss_2_0_d.h" #include "oss/oss_2_0_sh_mask.h" #include "gca/gfx_8_0_d.h" +#include "smu/smu_7_1_2_d.h" +#include "smu/smu_7_1_2_sh_mask.h" #define GRBM_GFX_INDEX__VCE_INSTANCE__SHIFT 0x04 #define GRBM_GFX_INDEX__VCE_INSTANCE_MASK 0x10 @@ -112,6 +114,10 @@ static int vce_v3_0_start(struct amdgpu_device *adev) mutex_lock(&adev->grbm_idx_mutex); for (idx = 0; idx < 2; ++idx) { + + if (adev->vce.harvest_config & (1 << idx)) + continue; + if(idx == 0) WREG32_P(mmGRBM_GFX_INDEX, 0, ~GRBM_GFX_INDEX__VCE_INSTANCE_MASK); @@ -190,10 +196,52 @@ static int vce_v3_0_start(struct amdgpu_device *adev) return 0; } +#define ixVCE_HARVEST_FUSE_MACRO__ADDRESS 0xC0014074 +#define VCE_HARVEST_FUSE_MACRO__SHIFT 27 +#define VCE_HARVEST_FUSE_MACRO__MASK 0x18000000 + +static unsigned vce_v3_0_get_harvest_config(struct amdgpu_device *adev) +{ + u32 tmp; + unsigned ret; + + if (adev->flags & AMDGPU_IS_APU) + tmp = (RREG32_SMC(ixVCE_HARVEST_FUSE_MACRO__ADDRESS) & + VCE_HARVEST_FUSE_MACRO__MASK) >> + VCE_HARVEST_FUSE_MACRO__SHIFT; + else + tmp = (RREG32_SMC(ixCC_HARVEST_FUSES) & + CC_HARVEST_FUSES__VCE_DISABLE_MASK) >> + CC_HARVEST_FUSES__VCE_DISABLE__SHIFT; + + switch (tmp) { + case 1: + ret = AMDGPU_VCE_HARVEST_VCE0; + break; + case 2: + ret = AMDGPU_VCE_HARVEST_VCE1; + break; + case 3: + ret = AMDGPU_VCE_HARVEST_VCE0 | AMDGPU_VCE_HARVEST_VCE1; + break; + default: + ret = 0; + } + + return ret; +} + static int vce_v3_0_early_init(void *handle) { struct amdgpu_device *adev = (struct amdgpu_device *)handle; + adev->vce.harvest_config = vce_v3_0_get_harvest_config(adev); + + if ((adev->vce.harvest_config & + (AMDGPU_VCE_HARVEST_VCE0 | AMDGPU_VCE_HARVEST_VCE1)) == + (AMDGPU_VCE_HARVEST_VCE0 | AMDGPU_VCE_HARVEST_VCE1)) + return -ENOENT; + vce_v3_0_set_ring_funcs(adev); vce_v3_0_set_irq_funcs(adev); -- cgit v0.10.2 From fa92754e9c47cf3e5607c0865f4cf59d090cda37 Mon Sep 17 00:00:00 2001 From: Leo Liu Date: Mon, 13 Jul 2015 12:46:23 -0400 Subject: drm/amdgpu: add VCE harvesting instance query MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Signed-off-by: Leo Liu Reviewed-by: Alex Deucher Reviewed-by: Christian König diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c index 5533434..31ad444 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c @@ -459,6 +459,7 @@ static int amdgpu_info_ioctl(struct drm_device *dev, void *data, struct drm_file memcpy(&dev_info.cu_bitmap[0], &cu_info.bitmap[0], sizeof(cu_info.bitmap)); dev_info.vram_type = adev->mc.vram_type; dev_info.vram_bit_width = adev->mc.vram_width; + dev_info.vce_harvest_config = adev->vce.harvest_config; return copy_to_user(out, &dev_info, min((size_t)size, sizeof(dev_info))) ? -EFAULT : 0; diff --git a/include/uapi/drm/amdgpu_drm.h b/include/uapi/drm/amdgpu_drm.h index b6fce90..d708a53 100644 --- a/include/uapi/drm/amdgpu_drm.h +++ b/include/uapi/drm/amdgpu_drm.h @@ -614,6 +614,8 @@ struct drm_amdgpu_info_device { uint32_t vram_type; /** video memory bit width*/ uint32_t vram_bit_width; + /* vce harvesting instance */ + uint32_t vce_harvest_config; }; struct drm_amdgpu_info_hw_ip { -- cgit v0.10.2 From fe78fcc85a2046c51f1535710996860557eeec20 Mon Sep 17 00:00:00 2001 From: "Eric W. Biederman" Date: Fri, 17 Jul 2015 14:54:27 -0500 Subject: mnt: In detach_mounts detach the appropriate unmounted mount The handling of in detach_mounts of unmounted but connected mounts is buggy and can lead to an infinite loop. Correct the handling of unmounted mounts in detach_mount. When the mountpoint of an unmounted but connected mount is connected to a dentry, and that dentry is deleted we need to disconnect that mount from the parent mount and the deleted dentry. Nothing changes for the unmounted and connected children. They can be safely ignored. Cc: stable@vger.kernel.org Fixes: ce07d891a0891d3c0d0c2d73d577490486b809e1 mnt: Honor MNT_LOCKED when detaching mounts Signed-off-by: "Eric W. Biederman" diff --git a/fs/namespace.c b/fs/namespace.c index 7829eab..2b8aa15 100644 --- a/fs/namespace.c +++ b/fs/namespace.c @@ -1565,11 +1565,8 @@ void __detach_mounts(struct dentry *dentry) while (!hlist_empty(&mp->m_list)) { mnt = hlist_entry(mp->m_list.first, struct mount, mnt_mp_list); if (mnt->mnt.mnt_flags & MNT_UMOUNT) { - struct mount *p, *tmp; - list_for_each_entry_safe(p, tmp, &mnt->mnt_mounts, mnt_child) { - hlist_add_head(&p->mnt_umount.s_list, &unmounted); - umount_mnt(p); - } + hlist_add_head(&mnt->mnt_umount.s_list, &unmounted); + umount_mnt(mnt); } else umount_tree(mnt, UMOUNT_CONNECTED); } -- cgit v0.10.2 From 3f81d2447b37ac697b3c600039f2c6b628c06e21 Mon Sep 17 00:00:00 2001 From: Chris Metcalf Date: Thu, 23 Jul 2015 14:11:09 -0400 Subject: tile: use free_bootmem_late() for initrd We were previously using free_bootmem() and just getting lucky that nothing too bad happened. Signed-off-by: Chris Metcalf Cc: stable@vger.kernel.org diff --git a/arch/tile/kernel/setup.c b/arch/tile/kernel/setup.c index 99c9ff8..6b755d1 100644 --- a/arch/tile/kernel/setup.c +++ b/arch/tile/kernel/setup.c @@ -1139,7 +1139,7 @@ static void __init load_hv_initrd(void) void __init free_initrd_mem(unsigned long begin, unsigned long end) { - free_bootmem(__pa(begin), end - begin); + free_bootmem_late(__pa(begin), end - begin); } static int __init setup_initrd(char *str) -- cgit v0.10.2 From f2d52cd4db08db06200176cfebead9778878d4fc Mon Sep 17 00:00:00 2001 From: Alex Deucher Date: Tue, 14 Jul 2015 16:16:29 -0400 Subject: drm/amdgpu/cz: implement voltage validation properly CZ uses a different set of registers compared to previous asics and supports separate NB and GFX planes. Reviewed-by: Jammy Zhou Signed-off-by: Alex Deucher diff --git a/drivers/gpu/drm/amd/amdgpu/cz_dpm.c b/drivers/gpu/drm/amd/amdgpu/cz_dpm.c index 1a2d419..1316d54 100644 --- a/drivers/gpu/drm/amd/amdgpu/cz_dpm.c +++ b/drivers/gpu/drm/amd/amdgpu/cz_dpm.c @@ -494,6 +494,13 @@ static void cz_dpm_fini(struct amdgpu_device *adev) amdgpu_free_extended_power_table(adev); } +#define ixSMUSVI_NB_CURRENTVID 0xD8230044 +#define CURRENT_NB_VID_MASK 0xff000000 +#define CURRENT_NB_VID__SHIFT 24 +#define ixSMUSVI_GFX_CURRENTVID 0xD8230048 +#define CURRENT_GFX_VID_MASK 0xff000000 +#define CURRENT_GFX_VID__SHIFT 24 + static void cz_dpm_debugfs_print_current_performance_level(struct amdgpu_device *adev, struct seq_file *m) @@ -505,18 +512,20 @@ cz_dpm_debugfs_print_current_performance_level(struct amdgpu_device *adev, TARGET_AND_CURRENT_PROFILE_INDEX__CURR_SCLK_INDEX_MASK) >> TARGET_AND_CURRENT_PROFILE_INDEX__CURR_SCLK_INDEX__SHIFT; u32 sclk, tmp; - u16 vddc; + u16 vddnb, vddgfx; if (current_index >= NUM_SCLK_LEVELS) { seq_printf(m, "invalid dpm profile %d\n", current_index); } else { sclk = table->entries[current_index].clk; - tmp = (RREG32_SMC(ixSMU_VOLTAGE_STATUS) & - SMU_VOLTAGE_STATUS__SMU_VOLTAGE_CURRENT_LEVEL_MASK) >> - SMU_VOLTAGE_STATUS__SMU_VOLTAGE_CURRENT_LEVEL__SHIFT; - vddc = cz_convert_8bit_index_to_voltage(adev, (u16)tmp); - seq_printf(m, "power level %d sclk: %u vddc: %u\n", - current_index, sclk, vddc); + tmp = (RREG32_SMC(ixSMUSVI_NB_CURRENTVID) & + CURRENT_NB_VID_MASK) >> CURRENT_NB_VID__SHIFT; + vddnb = cz_convert_8bit_index_to_voltage(adev, (u16)tmp); + tmp = (RREG32_SMC(ixSMUSVI_GFX_CURRENTVID) & + CURRENT_GFX_VID_MASK) >> CURRENT_GFX_VID__SHIFT; + vddgfx = cz_convert_8bit_index_to_voltage(adev, (u16)tmp); + seq_printf(m, "power level %d sclk: %u vddnb: %u vddgfx: %u\n", + current_index, sclk, vddnb, vddgfx); } } -- cgit v0.10.2 From acc6a1a69b79fad70c4794a925dbfffa9fd6b21b Mon Sep 17 00:00:00 2001 From: Alex Deucher Date: Wed, 22 Jul 2015 12:03:50 -0400 Subject: drm/amdgpu/cz/dpm: properly report UVD and VCE clock levels VCE, UVD DPM work similarly to SCLK DPM. Report the current clock levels for UVD and VCE via debugfs. Signed-off-by: Alex Deucher diff --git a/drivers/gpu/drm/amd/amdgpu/cz_dpm.c b/drivers/gpu/drm/amd/amdgpu/cz_dpm.c index 1316d54..ace870a 100644 --- a/drivers/gpu/drm/amd/amdgpu/cz_dpm.c +++ b/drivers/gpu/drm/amd/amdgpu/cz_dpm.c @@ -505,27 +505,56 @@ static void cz_dpm_debugfs_print_current_performance_level(struct amdgpu_device *adev, struct seq_file *m) { + struct cz_power_info *pi = cz_get_pi(adev); struct amdgpu_clock_voltage_dependency_table *table = &adev->pm.dpm.dyn_state.vddc_dependency_on_sclk; - u32 current_index = - (RREG32_SMC(ixTARGET_AND_CURRENT_PROFILE_INDEX) & - TARGET_AND_CURRENT_PROFILE_INDEX__CURR_SCLK_INDEX_MASK) >> - TARGET_AND_CURRENT_PROFILE_INDEX__CURR_SCLK_INDEX__SHIFT; - u32 sclk, tmp; + struct amdgpu_uvd_clock_voltage_dependency_table *uvd_table = + &adev->pm.dpm.dyn_state.uvd_clock_voltage_dependency_table; + struct amdgpu_vce_clock_voltage_dependency_table *vce_table = + &adev->pm.dpm.dyn_state.vce_clock_voltage_dependency_table; + u32 sclk_index = REG_GET_FIELD(RREG32_SMC(ixTARGET_AND_CURRENT_PROFILE_INDEX), + TARGET_AND_CURRENT_PROFILE_INDEX, CURR_SCLK_INDEX); + u32 uvd_index = REG_GET_FIELD(RREG32_SMC(ixTARGET_AND_CURRENT_PROFILE_INDEX_2), + TARGET_AND_CURRENT_PROFILE_INDEX_2, CURR_UVD_INDEX); + u32 vce_index = REG_GET_FIELD(RREG32_SMC(ixTARGET_AND_CURRENT_PROFILE_INDEX_2), + TARGET_AND_CURRENT_PROFILE_INDEX_2, CURR_VCE_INDEX); + u32 sclk, vclk, dclk, ecclk, tmp; u16 vddnb, vddgfx; - if (current_index >= NUM_SCLK_LEVELS) { - seq_printf(m, "invalid dpm profile %d\n", current_index); + if (sclk_index >= NUM_SCLK_LEVELS) { + seq_printf(m, "invalid sclk dpm profile %d\n", sclk_index); } else { - sclk = table->entries[current_index].clk; - tmp = (RREG32_SMC(ixSMUSVI_NB_CURRENTVID) & - CURRENT_NB_VID_MASK) >> CURRENT_NB_VID__SHIFT; - vddnb = cz_convert_8bit_index_to_voltage(adev, (u16)tmp); - tmp = (RREG32_SMC(ixSMUSVI_GFX_CURRENTVID) & - CURRENT_GFX_VID_MASK) >> CURRENT_GFX_VID__SHIFT; - vddgfx = cz_convert_8bit_index_to_voltage(adev, (u16)tmp); - seq_printf(m, "power level %d sclk: %u vddnb: %u vddgfx: %u\n", - current_index, sclk, vddnb, vddgfx); + sclk = table->entries[sclk_index].clk; + seq_printf(m, "%u sclk: %u\n", sclk_index, sclk); + } + + tmp = (RREG32_SMC(ixSMUSVI_NB_CURRENTVID) & + CURRENT_NB_VID_MASK) >> CURRENT_NB_VID__SHIFT; + vddnb = cz_convert_8bit_index_to_voltage(adev, (u16)tmp); + tmp = (RREG32_SMC(ixSMUSVI_GFX_CURRENTVID) & + CURRENT_GFX_VID_MASK) >> CURRENT_GFX_VID__SHIFT; + vddgfx = cz_convert_8bit_index_to_voltage(adev, (u16)tmp); + seq_printf(m, "vddnb: %u vddgfx: %u\n", vddnb, vddgfx); + + seq_printf(m, "uvd %sabled\n", pi->uvd_power_gated ? "dis" : "en"); + if (!pi->uvd_power_gated) { + if (uvd_index >= CZ_MAX_HARDWARE_POWERLEVELS) { + seq_printf(m, "invalid uvd dpm level %d\n", uvd_index); + } else { + vclk = uvd_table->entries[uvd_index].vclk; + dclk = uvd_table->entries[uvd_index].dclk; + seq_printf(m, "%u uvd vclk: %u dclk: %u\n", uvd_index, vclk, dclk); + } + } + + seq_printf(m, "vce %sabled\n", pi->vce_power_gated ? "dis" : "en"); + if (!pi->vce_power_gated) { + if (vce_index >= CZ_MAX_HARDWARE_POWERLEVELS) { + seq_printf(m, "invalid vce dpm level %d\n", vce_index); + } else { + ecclk = vce_table->entries[vce_index].ecclk; + seq_printf(m, "%u vce ecclk: %u\n", vce_index, ecclk); + } } } -- cgit v0.10.2 From 5aa2a96b34073fb775917fbc7d5578a04164b9f7 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Thu, 23 Jul 2015 14:27:09 -0400 Subject: block: export bio_associate_*() and wbc_account_io() bio_associate_blkcg(), bio_associate_current() and wbc_account_io() are used to implement cgroup writeback support for filesystems and thus need to be exported. Export them. Signed-off-by: Tejun Heo Reported-by: Stephen Rothwell Signed-off-by: Jens Axboe diff --git a/block/bio.c b/block/bio.c index 2a00d34..b671a0f 100644 --- a/block/bio.c +++ b/block/bio.c @@ -2009,6 +2009,7 @@ int bio_associate_blkcg(struct bio *bio, struct cgroup_subsys_state *blkcg_css) bio->bi_css = blkcg_css; return 0; } +EXPORT_SYMBOL_GPL(bio_associate_blkcg); /** * bio_associate_current - associate a bio with %current @@ -2039,6 +2040,7 @@ int bio_associate_current(struct bio *bio) bio->bi_css = task_get_css(current, blkio_cgrp_id); return 0; } +EXPORT_SYMBOL_GPL(bio_associate_current); /** * bio_disassociate_task - undo bio_associate_current() diff --git a/fs/fs-writeback.c b/fs/fs-writeback.c index f0520bc..518c629 100644 --- a/fs/fs-writeback.c +++ b/fs/fs-writeback.c @@ -702,6 +702,7 @@ void wbc_account_io(struct writeback_control *wbc, struct page *page, else wbc->wb_tcand_bytes -= min(bytes, wbc->wb_tcand_bytes); } +EXPORT_SYMBOL_GPL(wbc_account_io); /** * inode_congested - test whether an inode is congested -- cgit v0.10.2 From 968491709e5b1aaf429428814fff3d932fa90b60 Mon Sep 17 00:00:00 2001 From: Bernhard Bender Date: Thu, 23 Jul 2015 13:58:08 -0700 Subject: Input: usbtouchscreen - avoid unresponsive TSC-30 touch screen This patch fixes a problem in the usbtouchscreen driver for DMC TSC-30 touch screen. Due to a missing delay between the RESET and SET_RATE commands, the touch screen may become unresponsive during system startup or driver loading. According to the DMC documentation, a delay is needed after the RESET command to allow the chip to complete its internal initialization. As this delay is not guaranteed, we had a system where the touch screen occasionally did not send any touch data. There was no other indication of the problem. The patch fixes the problem by adding a 150ms delay between the RESET and SET_RATE commands. Cc: stable@vger.kernel.org Suggested-by: Jakob Mustafa Signed-off-by: Bernhard Bender Signed-off-by: Dmitry Torokhov diff --git a/drivers/input/touchscreen/usbtouchscreen.c b/drivers/input/touchscreen/usbtouchscreen.c index f2c6c35..2c41107 100644 --- a/drivers/input/touchscreen/usbtouchscreen.c +++ b/drivers/input/touchscreen/usbtouchscreen.c @@ -627,6 +627,9 @@ static int dmc_tsc10_init(struct usbtouch_usb *usbtouch) goto err_out; } + /* TSC-25 data sheet specifies a delay after the RESET command */ + msleep(150); + /* set coordinate output rate */ buf[0] = buf[1] = 0xFF; ret = usb_control_msg(dev, usb_rcvctrlpipe (dev, 0), -- cgit v0.10.2 From 3b19e032295647b7be2aa3be62510db4aaeda759 Mon Sep 17 00:00:00 2001 From: Peter Hurley Date: Sat, 27 Jun 2015 09:21:32 -0400 Subject: n_tty: signal and flush atomically When handling signalling char, claim the termios write lock before signalling waiting readers and writers to prevent further i/o before flushing the echo and output buffers. This prevents a userspace signal handler which may output from racing the terminal flush. Reference: Bugzilla #99351 ("Output truncated in ssh session after...") Fixes: commit d2b6f44779d3 ("n_tty: Fix signal handling flushes") Reported-by: Filipe Brandenburger Signed-off-by: Peter Hurley Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/tty/n_tty.c b/drivers/tty/n_tty.c index c9c27f6..ee8bfac 100644 --- a/drivers/tty/n_tty.c +++ b/drivers/tty/n_tty.c @@ -1108,19 +1108,29 @@ static void eraser(unsigned char c, struct tty_struct *tty) * Locking: ctrl_lock */ -static void isig(int sig, struct tty_struct *tty) +static void __isig(int sig, struct tty_struct *tty) { - struct n_tty_data *ldata = tty->disc_data; struct pid *tty_pgrp = tty_get_pgrp(tty); if (tty_pgrp) { kill_pgrp(tty_pgrp, sig, 1); put_pid(tty_pgrp); } +} - if (!L_NOFLSH(tty)) { +static void isig(int sig, struct tty_struct *tty) +{ + struct n_tty_data *ldata = tty->disc_data; + + if (L_NOFLSH(tty)) { + /* signal only */ + __isig(sig, tty); + + } else { /* signal and flush */ up_read(&tty->termios_rwsem); down_write(&tty->termios_rwsem); + __isig(sig, tty); + /* clear echo buffer */ mutex_lock(&ldata->output_lock); ldata->echo_head = ldata->echo_tail = 0; -- cgit v0.10.2 From 97a60eac33f663d2150ee1b3c69253d87ba53dff Mon Sep 17 00:00:00 2001 From: Krzysztof Kozlowski Date: Thu, 9 Jul 2015 22:21:41 +0900 Subject: serial: amba-pl011: Fix devm_ioremap_resource return value check Value returned by devm_ioremap_resource() was checked for non-NULL but devm_ioremap_resource() returns IOMEM_ERR_PTR, not NULL. In case of error this could lead to dereference of ERR_PTR. Signed-off-by: Krzysztof Kozlowski Cc: Fixes: 3873e2d7f63a ("drivers: PL011: refactor pl011_probe()") Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/tty/serial/amba-pl011.c b/drivers/tty/serial/amba-pl011.c index 50cf5b1..fd27e98 100644 --- a/drivers/tty/serial/amba-pl011.c +++ b/drivers/tty/serial/amba-pl011.c @@ -2310,8 +2310,8 @@ static int pl011_setup_port(struct device *dev, struct uart_amba_port *uap, void __iomem *base; base = devm_ioremap_resource(dev, mmiobase); - if (!base) - return -ENOMEM; + if (IS_ERR(base)) + return PTR_ERR(base); index = pl011_probe_dt_alias(index, dev); -- cgit v0.10.2 From 08b33249d89700ba555d4ab5cc88714192b8ee46 Mon Sep 17 00:00:00 2001 From: Dongxing Zhang Date: Wed, 10 Jun 2015 15:21:10 +0800 Subject: tty/vt: Fix the memory leak in visual_init If vc->vc_uni_pagedir_loc is not NULL, its refcount needs to be decreased before vc_uni_pagedir_loc is re-assigned. unreferenced object 0xffff88002cdd13b0 (size 512): comm "setfont", pid 503, jiffies 4294896503 (age 722.828s) hex dump (first 32 bytes): 40 92 61 2b 00 88 ff ff 00 00 00 00 00 00 00 00 @.a+............ 00 00 00 00 00 00 00 00 a0 ad 61 2b 00 88 ff ff ..........a+.... backtrace: [] kmemleak_alloc+0x4e/0xb0 [] kmem_cache_alloc_trace+0x1c8/0x240 [] con_do_clear_unimap.isra.2+0x83/0xe0 [] con_clear_unimap+0x22/0x40 [] vt_ioctl+0xeb8/0x1170 [] tty_ioctl+0x208/0xca0 [] do_vfs_ioctl+0x2f8/0x510 [] SyS_ioctl+0x81/0xa0 [] system_call_fastpath+0x16/0x75 [] 0xffffffffffffffff unreferenced object 0xffff88002b619240 (size 256): comm "setfont", pid 503, jiffies 4294896503 (age 722.828s) hex dump (first 32 bytes): 90 bc 84 d5 00 88 ff ff 58 85 84 d5 00 88 ff ff ........X....... 88 ac 84 d5 00 88 ff ff e0 b1 84 d5 00 88 ff ff ................ backtrace: [] kmemleak_alloc+0x4e/0xb0 [] kmem_cache_alloc_trace+0x1c8/0x240 [] con_insert_unipair+0x86/0x170 [] con_set_unimap+0x1b7/0x280 [] vt_ioctl+0xe65/0x1170 [] tty_ioctl+0x208/0xca0 [] do_vfs_ioctl+0x2f8/0x510 [] SyS_ioctl+0x81/0xa0 [] system_call_fastpath+0x16/0x75 [] 0xffffffffffffffff Signed-off-by: Dongxing Zhang Signed-off-by: Xiaoming Wang Reviewed-by: Peter Hurley Tested-by: Konstantin Khlebnikov Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/tty/vt/vt.c b/drivers/tty/vt/vt.c index 8fe5298..4462d16 100644 --- a/drivers/tty/vt/vt.c +++ b/drivers/tty/vt/vt.c @@ -742,6 +742,8 @@ static void visual_init(struct vc_data *vc, int num, int init) __module_get(vc->vc_sw->owner); vc->vc_num = num; vc->vc_display_fg = &master_display_fg; + if (vc->vc_uni_pagedir_loc) + con_free_unimap(vc); vc->vc_uni_pagedir_loc = &vc->vc_uni_pagedir; vc->vc_uni_pagedir = NULL; vc->vc_hi_font_mask = 0; -- cgit v0.10.2 From f3f5da624e0a891c34d8cd513c57f1d9b0c7dadc Mon Sep 17 00:00:00 2001 From: "Martin K. Petersen" Date: Wed, 22 Jul 2015 07:57:12 -0400 Subject: block: Do a full clone when splitting discard bios This fixes a data corruption bug when using discard on top of MD linear, raid0 and raid10 personalities. Commit 20d0189b1012 "block: Introduce new bio_split()" permits sharing the bio_vec between the two resulting bios. That is fine for read/write requests where the bio_vec is immutable. For discards, however, we need to be able to attach a payload and update the bio_vec so the page can get mapped to a scatterlist entry. Therefore the bio_vec can not be shared when splitting discards and we must do a full clone. Signed-off-by: Martin K. Petersen Reported-by: Seunguk Shin Tested-by: Seunguk Shin Cc: Seunguk Shin Cc: Jens Axboe Cc: Kent Overstreet Cc: # v3.14+ Reviewed-by: Christoph Hellwig Signed-off-by: Jens Axboe diff --git a/block/bio.c b/block/bio.c index b671a0f..d6e5ba3 100644 --- a/block/bio.c +++ b/block/bio.c @@ -1831,8 +1831,9 @@ EXPORT_SYMBOL(bio_endio); * Allocates and returns a new bio which represents @sectors from the start of * @bio, and updates @bio to represent the remaining sectors. * - * The newly allocated bio will point to @bio's bi_io_vec; it is the caller's - * responsibility to ensure that @bio is not freed before the split. + * Unless this is a discard request the newly allocated bio will point + * to @bio's bi_io_vec; it is the caller's responsibility to ensure that + * @bio is not freed before the split. */ struct bio *bio_split(struct bio *bio, int sectors, gfp_t gfp, struct bio_set *bs) @@ -1842,7 +1843,15 @@ struct bio *bio_split(struct bio *bio, int sectors, BUG_ON(sectors <= 0); BUG_ON(sectors >= bio_sectors(bio)); - split = bio_clone_fast(bio, gfp, bs); + /* + * Discards need a mutable bio_vec to accommodate the payload + * required by the DSM TRIM and UNMAP commands. + */ + if (bio->bi_rw & REQ_DISCARD) + split = bio_clone_bioset(bio, gfp, bs); + else + split = bio_clone_fast(bio, gfp, bs); + if (!split) return NULL; -- cgit v0.10.2 From d8c2c0d89205a3538c5fd77dc9d4767ee6bebc70 Mon Sep 17 00:00:00 2001 From: Axel Lin Date: Sun, 28 Jun 2015 09:45:08 +0800 Subject: serial: etraxfs-uart: Fix release etraxfs_uart_ports In probe, we use dev_id as array index of etraxfs_uart_ports and store the index in port->line. So etraxfs_uart_ports[port->line] should be released when unload the module. Signed-off-by: Axel Lin Acked-by: Niklas Cassel Acked-by: Jesper Nilsson Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/tty/serial/etraxfs-uart.c b/drivers/tty/serial/etraxfs-uart.c index a57301a..679709f 100644 --- a/drivers/tty/serial/etraxfs-uart.c +++ b/drivers/tty/serial/etraxfs-uart.c @@ -950,7 +950,7 @@ static int etraxfs_uart_remove(struct platform_device *pdev) port = platform_get_drvdata(pdev); uart_remove_one_port(&etraxfs_uart_driver, port); - etraxfs_uart_ports[pdev->id] = NULL; + etraxfs_uart_ports[port->line] = NULL; return 0; } -- cgit v0.10.2 From 32aa6339d9c5156de5d881b103e0e360934d6a74 Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Wed, 1 Jul 2015 14:19:52 +0200 Subject: sc16is7xx: fix Kconfig dependencies When I2C=m and SPI=y or-ing them will produce =y while what we need is the lower bound, i.e. =m. Fortunately SPI is a boolean so we need to handle only one special case. Reported-by: kbuild test robot Signed-off-by: Jakub Kicinski Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/tty/serial/Kconfig b/drivers/tty/serial/Kconfig index 76e65b7..15b4079 100644 --- a/drivers/tty/serial/Kconfig +++ b/drivers/tty/serial/Kconfig @@ -1185,7 +1185,7 @@ config SERIAL_SC16IS7XX_CORE config SERIAL_SC16IS7XX tristate "SC16IS7xx serial support" select SERIAL_CORE - depends on I2C || SPI_MASTER + depends on (SPI_MASTER && !I2C) || I2C help This selects support for SC16IS7xx serial ports. Supported ICs are SC16IS740, SC16IS741, SC16IS750, SC16IS752, -- cgit v0.10.2 From dec273ecc1169ade46999bb0feca3814915f5c11 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Bo=20Svang=C3=A5rd?= Date: Sat, 13 Jun 2015 13:40:20 +0200 Subject: sc16is7xx: fix FIFO address of secondary UART MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Calls to regmap_raw_read/write needed register rewrite in a similar way as function calls to regmap_read/write already had. This enables reading/writing the serial datastream to the device. Signed-off-by: Bo SvangÃ¥rd Signed-off-by: Jakub Kicinski Signed-off-by: Jon Ringle Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/tty/serial/sc16is7xx.c b/drivers/tty/serial/sc16is7xx.c index 9e65760..5ccc698 100644 --- a/drivers/tty/serial/sc16is7xx.c +++ b/drivers/tty/serial/sc16is7xx.c @@ -354,6 +354,26 @@ static void sc16is7xx_port_write(struct uart_port *port, u8 reg, u8 val) (reg << SC16IS7XX_REG_SHIFT) | port->line, val); } +static void sc16is7xx_fifo_read(struct uart_port *port, unsigned int rxlen) +{ + struct sc16is7xx_port *s = dev_get_drvdata(port->dev); + u8 addr = (SC16IS7XX_RHR_REG << SC16IS7XX_REG_SHIFT) | port->line; + + regcache_cache_bypass(s->regmap, true); + regmap_raw_read(s->regmap, addr, s->buf, rxlen); + regcache_cache_bypass(s->regmap, false); +} + +static void sc16is7xx_fifo_write(struct uart_port *port, u8 to_send) +{ + struct sc16is7xx_port *s = dev_get_drvdata(port->dev); + u8 addr = (SC16IS7XX_THR_REG << SC16IS7XX_REG_SHIFT) | port->line; + + regcache_cache_bypass(s->regmap, true); + regmap_raw_write(s->regmap, addr, s->buf, to_send); + regcache_cache_bypass(s->regmap, false); +} + static void sc16is7xx_port_update(struct uart_port *port, u8 reg, u8 mask, u8 val) { @@ -508,10 +528,7 @@ static void sc16is7xx_handle_rx(struct uart_port *port, unsigned int rxlen, s->buf[0] = sc16is7xx_port_read(port, SC16IS7XX_RHR_REG); bytes_read = 1; } else { - regcache_cache_bypass(s->regmap, true); - regmap_raw_read(s->regmap, SC16IS7XX_RHR_REG, - s->buf, rxlen); - regcache_cache_bypass(s->regmap, false); + sc16is7xx_fifo_read(port, rxlen); bytes_read = rxlen; } @@ -591,9 +608,8 @@ static void sc16is7xx_handle_tx(struct uart_port *port) s->buf[i] = xmit->buf[xmit->tail]; xmit->tail = (xmit->tail + 1) & (UART_XMIT_SIZE - 1); } - regcache_cache_bypass(s->regmap, true); - regmap_raw_write(s->regmap, SC16IS7XX_THR_REG, s->buf, to_send); - regcache_cache_bypass(s->regmap, false); + + sc16is7xx_fifo_write(port, to_send); } if (uart_circ_chars_pending(xmit) < WAKEUP_CHARS) -- cgit v0.10.2 From 907eda32a36fcdb979bdb91ea097abb3dd2c23c9 Mon Sep 17 00:00:00 2001 From: David Jander Date: Fri, 26 Jun 2015 08:11:30 +0200 Subject: Revert "serial: imx: initialized DMA w/o HW flow enabled" This reverts commit 068500e08dc87ea9a453cc4a500cf3ab28d0f936. According to some tests, SDMA support is broken at least for i.MX6 without HW flow control. Different forms of data-corruption appear either with the ROM firmware for the SDMA controller as well as when loading Freescale provided SDMA firmware versions 1.1 or 3.1. Signed-off-by: David Jander Cc: stable Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/tty/serial/imx.c b/drivers/tty/serial/imx.c index 2c90dc3..54fdc78 100644 --- a/drivers/tty/serial/imx.c +++ b/drivers/tty/serial/imx.c @@ -1121,11 +1121,6 @@ static int imx_startup(struct uart_port *port) writel(temp & ~UCR4_DREN, sport->port.membase + UCR4); - /* Can we enable the DMA support? */ - if (is_imx6q_uart(sport) && !uart_console(port) && - !sport->dma_is_inited) - imx_uart_dma_init(sport); - spin_lock_irqsave(&sport->port.lock, flags); /* Reset fifo's and state machines */ i = 100; @@ -1143,9 +1138,6 @@ static int imx_startup(struct uart_port *port) writel(USR1_RTSD, sport->port.membase + USR1); writel(USR2_ORE, sport->port.membase + USR2); - if (sport->dma_is_inited && !sport->dma_is_enabled) - imx_enable_dma(sport); - temp = readl(sport->port.membase + UCR1); temp |= UCR1_RRDYEN | UCR1_RTSDEN | UCR1_UARTEN; @@ -1316,6 +1308,11 @@ imx_set_termios(struct uart_port *port, struct ktermios *termios, } else { ucr2 |= UCR2_CTSC; } + + /* Can we enable the DMA support? */ + if (is_imx6q_uart(sport) && !uart_console(port) + && !sport->dma_is_inited) + imx_uart_dma_init(sport); } else { termios->c_cflag &= ~CRTSCTS; } @@ -1432,6 +1429,8 @@ imx_set_termios(struct uart_port *port, struct ktermios *termios, if (UART_ENABLE_MS(&sport->port, termios->c_cflag)) imx_enable_ms(&sport->port); + if (sport->dma_is_inited && !sport->dma_is_enabled) + imx_enable_dma(sport); spin_unlock_irqrestore(&sport->port.lock, flags); } -- cgit v0.10.2 From 7525a9901b5ee3c84933ad069e41506c37404ddc Mon Sep 17 00:00:00 2001 From: Peter Hurley Date: Wed, 8 Jul 2015 23:10:06 -0400 Subject: m32r: Add ioreadXX/iowriteXX big-endian mmio accessors commit c627f2ceb692 ("serial: 8250: Add support for big-endian MMIO accesses") added support for 32-bit big-endian mmio to the 8250 driver. Support for ioreadXXbe/iowriteXXbe io accessors was missing from m32r arch, which caused build errors. Add trivial macro mmio accessors. Reported-by: Fengguang Wu Cc: Kevin Cernekee Signed-off-by: Peter Hurley Signed-off-by: Greg Kroah-Hartman diff --git a/arch/m32r/include/asm/io.h b/arch/m32r/include/asm/io.h index 0c3f25e..f8de767 100644 --- a/arch/m32r/include/asm/io.h +++ b/arch/m32r/include/asm/io.h @@ -174,6 +174,11 @@ static inline void _writel(unsigned long l, unsigned long addr) #define iowrite16 writew #define iowrite32 writel +#define ioread16be(addr) be16_to_cpu(readw(addr)) +#define ioread32be(addr) be32_to_cpu(readl(addr)) +#define iowrite16be(v, addr) writew(cpu_to_be16(v), (addr)) +#define iowrite32be(v, addr) writel(cpu_to_be32(v), (addr)) + #define mmiowb() #define flush_write_buffers() do { } while (0) /* M32R_FIXME */ -- cgit v0.10.2 From e144c58cad6667876173dd76977e9e6557e34941 Mon Sep 17 00:00:00 2001 From: Peter Hurley Date: Sun, 12 Jul 2015 21:05:26 -0400 Subject: serial: core: Fix crashes while echoing when closing While closing, new rx data may be received after the input buffers have been flushed but before stop_rx() halts receiving [1]. The new data might not be processed by flush_to_ldisc() until after uart_shutdown() and normal input processing is re-enabled (ie., tty->closing = 0). The race is outlined below: CPU 0 | CPU 1 | uart_close() | tty_port_close_start() | tty->closing = 1 | tty_ldisc_flush() | | => IRQ | while (LSR & data ready) | uart_insert_char() | tty_flip_buffer_push() | <= EOI stop_rx() | . uart_shutdown() | . free xmit.buf | . tty_port_tty_set(NULL) | . tty->closing = 0 | . | flush_to_ldisc() | n_tty_receive_buf_common() | __receive_buf() | ... | commit_echoes() | uart_flush_chars() | __uart_start() | ** OOPS on port.tty deref ** tty_ldisc_flush() | Input processing must be prevented from echoing (tty->closing = 1) until _after_ the input buffers have been flushed again at the end of uart_close(). [1] In fact, some input may actually be buffered _after_ stop_rx() since the rx interrupt may have already triggered but not yet been handled when stop_rx() disables rx interrupts. Fixes: 2e758910832d ("serial: core: Flush ldisc after dropping port mutex in uart_close()") Reported-by: Robert Elliott Signed-off-by: Peter Hurley Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/tty/serial/serial_core.c b/drivers/tty/serial/serial_core.c index 7ae1592..f368520 100644 --- a/drivers/tty/serial/serial_core.c +++ b/drivers/tty/serial/serial_core.c @@ -1418,7 +1418,7 @@ static void uart_close(struct tty_struct *tty, struct file *filp) mutex_lock(&port->mutex); uart_shutdown(tty, state); tty_port_tty_set(port, NULL); - tty->closing = 0; + spin_lock_irqsave(&port->lock, flags); if (port->blocked_open) { @@ -1444,6 +1444,7 @@ static void uart_close(struct tty_struct *tty, struct file *filp) mutex_unlock(&port->mutex); tty_ldisc_flush(tty); + tty->closing = 0; } static void uart_wait_until_sent(struct tty_struct *tty, int timeout) -- cgit v0.10.2 From 61e86cc90af49cecef9c54ccea1f572fbcb695ac Mon Sep 17 00:00:00 2001 From: Peter Hurley Date: Sun, 12 Jul 2015 20:47:35 -0400 Subject: tty: vt: Fix !TASK_RUNNING diagnostic warning from paste_selection() Pasting text with gpm on a VC produced warning [1]. Reset task state to TASK_RUNNING in the paste_selection() loop, if the loop did not sleep. [1] WARNING: CPU: 6 PID: 1960 at /home/peter/src/kernels/mainline/kernel/sched/core.c:7286 __might_sleep+0x7f/0x90() do not call blocking ops when !TASK_RUNNING; state=1 set at [] paste_selection+0x9e/0x1a0 Modules linked in: btrfs xor raid6_pq ufs qnx4 hfsplus hfs minix ntfs msdos jfs xfs libcrc32c ..... CPU: 6 PID: 1960 Comm: gpm Not tainted 4.1.0-rc7+tty-xeon+debug #rc7+tty Hardware name: Dell Inc. Precision WorkStation T5400 /0RW203, BIOS A11 04/30/2012 ffffffff81c9c0a0 ffff8802b0fd3ac8 ffffffff8185778a 0000000000000001 ffff8802b0fd3b18 ffff8802b0fd3b08 ffffffff8108039a ffffffff82ae8510 ffffffff81c9ce00 0000000000000015 0000000000000000 0000000000000000 Call Trace: [] dump_stack+0x4f/0x7b [] warn_slowpath_common+0x8a/0xc0 [] warn_slowpath_fmt+0x46/0x50 [] ? __lock_acquire+0xe2d/0x13a0 [] ? paste_selection+0x9e/0x1a0 [] ? paste_selection+0x9e/0x1a0 [] __might_sleep+0x7f/0x90 [] down_read+0x2a/0xa0 [] ? sched_clock_cpu+0xb8/0xe0 [] n_tty_receive_buf_common+0x4c/0xba0 [] ? mark_held_locks+0x75/0xa0 [] ? _raw_spin_unlock_irqrestore+0x65/0x80 [] ? get_parent_ip+0x11/0x50 [] n_tty_receive_buf2+0x14/0x20 [] paste_selection+0x157/0x1a0 [] ? wake_up_state+0x20/0x20 [] tioclinux+0xb8/0x2c0 [] vt_ioctl+0xaee/0x11a0 [] ? sched_clock_local+0x25/0x90 [] ? vtime_account_user+0x91/0xa0 [] tty_ioctl+0x20c/0xe20 [] ? vtime_account_user+0x91/0xa0 [] ? get_parent_ip+0x11/0x50 [] ? preempt_count_sub+0x49/0x50 [] ? context_tracking_exit+0x5c/0x290 [] ? context_tracking_exit+0x5c/0x290 [] do_vfs_ioctl+0x318/0x570 [] ? trace_hardirqs_on+0xd/0x10 [] ? trace_hardirqs_on_caller+0x115/0x1e0 [] ? __fget_light+0x6c/0xa0 [] SyS_ioctl+0x81/0xa0 [] system_call_fastpath+0x16/0x7a Signed-off-by: Peter Hurley Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/tty/vt/selection.c b/drivers/tty/vt/selection.c index ea27804..381a2b1 100644 --- a/drivers/tty/vt/selection.c +++ b/drivers/tty/vt/selection.c @@ -356,6 +356,7 @@ int paste_selection(struct tty_struct *tty) schedule(); continue; } + __set_current_state(TASK_RUNNING); count = sel_buffer_lth - pasted; count = tty_ldisc_receive_buf(ld, sel_buffer + pasted, NULL, count); -- cgit v0.10.2 From f135b978c2dab3d439eacb8353d3c0aac6af072a Mon Sep 17 00:00:00 2001 From: Alexandre Courbot Date: Wed, 1 Jul 2015 17:32:29 +0900 Subject: drm/ttm: recognize ARM64 arch in ioprot handler Return proper pgprot for ARM64. This is required for objects like Nouveau fences to be mapped with expected coherency. Signed-off-by: Alexandre Courbot Signed-off-by: Dave Airlie diff --git a/drivers/gpu/drm/ttm/ttm_bo_util.c b/drivers/gpu/drm/ttm/ttm_bo_util.c index 882cccd..ac6fe40 100644 --- a/drivers/gpu/drm/ttm/ttm_bo_util.c +++ b/drivers/gpu/drm/ttm/ttm_bo_util.c @@ -490,7 +490,8 @@ pgprot_t ttm_io_prot(uint32_t caching_flags, pgprot_t tmp) else if (boot_cpu_data.x86 > 3) tmp = pgprot_noncached(tmp); #endif -#if defined(__ia64__) || defined(__arm__) || defined(__powerpc__) +#if defined(__ia64__) || defined(__arm__) || defined(__aarch64__) || \ + defined(__powerpc__) if (caching_flags & TTM_PL_FLAG_WC) tmp = pgprot_writecombine(tmp); else -- cgit v0.10.2 From 34cab6f42003cb06f48f86a86652984dec338ae9 Mon Sep 17 00:00:00 2001 From: NeilBrown Date: Fri, 24 Jul 2015 09:22:16 +1000 Subject: md/raid1: fix test for 'was read error from last working device'. When we get a read error from the last working device, we don't try to repair it, and don't fail the device. We simple report a read error to the caller. However the current test for 'is this the last working device' is wrong. When there is only one fully working device, it assumes that a non-faulty device is that device. However a spare which is rebuilding would be non-faulty but so not the only working device. So change the test from "!Faulty" to "In_sync". If ->degraded says there is only one fully working device and this device is in_sync, this must be the one. This bug has existed since we allowed read_balance to read from a recovering spare in v3.0 Reported-and-tested-by: Alexander Lyakas Fixes: 76073054c95b ("md/raid1: clean up read_balance.") Cc: stable@vger.kernel.org (v3.0+) Signed-off-by: NeilBrown diff --git a/drivers/md/raid1.c b/drivers/md/raid1.c index f80f1af..50cf0c8 100644 --- a/drivers/md/raid1.c +++ b/drivers/md/raid1.c @@ -336,7 +336,7 @@ static void raid1_end_read_request(struct bio *bio, int error) spin_lock_irqsave(&conf->device_lock, flags); if (r1_bio->mddev->degraded == conf->raid_disks || (r1_bio->mddev->degraded == conf->raid_disks-1 && - !test_bit(Faulty, &conf->mirrors[mirror].rdev->flags))) + test_bit(In_sync, &conf->mirrors[mirror].rdev->flags))) uptodate = 1; spin_unlock_irqrestore(&conf->device_lock, flags); } -- cgit v0.10.2 From f7357273198adc86fe11c2a7be8a0816f44103bb Mon Sep 17 00:00:00 2001 From: Goldwyn Rodrigues Date: Wed, 22 Jul 2015 12:09:16 -0500 Subject: md: Skip cluster setup in case of error while reading bitmap If the bitmap read fails, the error code set is -EINVAL. However, we don't check for errors and go ahead with cluster_setup. Skip the cluster setup in case of error. Signed-off-by: Goldwyn Rodrigues Signed-off-by: NeilBrown diff --git a/drivers/md/bitmap.c b/drivers/md/bitmap.c index c90118e..a4ab513 100644 --- a/drivers/md/bitmap.c +++ b/drivers/md/bitmap.c @@ -680,7 +680,7 @@ out: kunmap_atomic(sb); /* Assiging chunksize is required for "re_read" */ bitmap->mddev->bitmap_info.chunksize = chunksize; - if (nodes && (bitmap->cluster_slot < 0)) { + if (err == 0 && nodes && (bitmap->cluster_slot < 0)) { err = md_setup_cluster(bitmap->mddev, nodes); if (err) { pr_err("%s: Could not setup cluster service (%d)\n", -- cgit v0.10.2 From b0c26a79d6993b280931f8e2b406ca4b220bb58f Mon Sep 17 00:00:00 2001 From: Goldwyn Rodrigues Date: Wed, 22 Jul 2015 12:09:15 -0500 Subject: md: Return error if request_module fails and returns positive value request_module() can return 256 (process exited) in some cases, which is not as specified in the documentation before the request_module() definition. Convert the error to -ENOENT. The positive error number results in bitmap_create() returning a value that is meant to be an error but doesn't look like one, so it is dereferenced as a point and causes a crash. (not needed for stable as this is "experimental" code) Fixes: edb39c9deda8 ("Introduce md_cluster_operations to handle cluster functions") Signed-off-By: Goldwyn Rodrigues Signed-off-by: NeilBrown diff --git a/drivers/md/md.c b/drivers/md/md.c index 5025b3e..0fcf6de 100644 --- a/drivers/md/md.c +++ b/drivers/md/md.c @@ -7439,7 +7439,7 @@ int md_setup_cluster(struct mddev *mddev, int nodes) err = request_module("md-cluster"); if (err) { pr_err("md-cluster module not found.\n"); - return err; + return -ENOENT; } spin_lock(&pers_lock); -- cgit v0.10.2 From 33e38ac6887d975fe2635c7fcaefb6d5495cb2e1 Mon Sep 17 00:00:00 2001 From: Goldwyn Rodrigues Date: Wed, 1 Jul 2015 12:19:56 +1000 Subject: md-cluster: fix bitmap sub-offset in bitmap_read_sb bitmap_read_sb is modifying mddev->bitmap_info.offset. This works for the first bitmap read. However, when multiple bitmaps need to be opened by the same node, it ends up corrupting the offset. Fix it by using a local variable. Also, bitmap_read_sb is not required in bitmap_copy_from_slot since it is called in bitmap_create. Remove bitmap_read_sb(). Signed-off-by: Goldwyn Rodrigues Signed-off-by: NeilBrown diff --git a/drivers/md/bitmap.c b/drivers/md/bitmap.c index a4ab513..1b9c2df 100644 --- a/drivers/md/bitmap.c +++ b/drivers/md/bitmap.c @@ -559,6 +559,7 @@ static int bitmap_read_sb(struct bitmap *bitmap) unsigned long sectors_reserved = 0; int err = -EINVAL; struct page *sb_page; + loff_t offset = bitmap->mddev->bitmap_info.offset; if (!bitmap->storage.file && !bitmap->mddev->bitmap_info.offset) { chunksize = 128 * 1024 * 1024; @@ -585,9 +586,9 @@ re_read: bm_blocks = ((bm_blocks+7) >> 3) + sizeof(bitmap_super_t); /* to 4k blocks */ bm_blocks = DIV_ROUND_UP_SECTOR_T(bm_blocks, 4096); - bitmap->mddev->bitmap_info.offset += bitmap->cluster_slot * (bm_blocks << 3); + offset = bitmap->mddev->bitmap_info.offset + (bitmap->cluster_slot * (bm_blocks << 3)); pr_info("%s:%d bm slot: %d offset: %llu\n", __func__, __LINE__, - bitmap->cluster_slot, (unsigned long long)bitmap->mddev->bitmap_info.offset); + bitmap->cluster_slot, offset); } if (bitmap->storage.file) { @@ -598,7 +599,7 @@ re_read: bitmap, bytes, sb_page); } else { err = read_sb_page(bitmap->mddev, - bitmap->mddev->bitmap_info.offset, + offset, sb_page, 0, sizeof(bitmap_super_t)); } @@ -1875,10 +1876,6 @@ int bitmap_copy_from_slot(struct mddev *mddev, int slot, if (IS_ERR(bitmap)) return PTR_ERR(bitmap); - rv = bitmap_read_sb(bitmap); - if (rv) - goto err; - rv = bitmap_init_from_disk(bitmap, 0); if (rv) goto err; -- cgit v0.10.2 From 90382ed9afeafd42ef193f0eadc6b2a252d6c24d Mon Sep 17 00:00:00 2001 From: Goldwyn Rodrigues Date: Wed, 24 Jun 2015 09:30:32 -0500 Subject: Fix read-balancing during node failure During a node failure, We need to suspend read balancing so that the reads are directed to the first device and stale data is not read. Suspending writes is not required because these would be recorded and synced eventually. A new flag MD_CLUSTER_SUSPEND_READ_BALANCING is set in recover_prep(). area_resyncing() will respond true for the entire devices if this flag is set and the request type is READ. The flag is cleared in recover_done(). Signed-off-by: Goldwyn Rodrigues Reported-By: David Teigland Signed-off-by: NeilBrown diff --git a/drivers/md/md-cluster.c b/drivers/md/md-cluster.c index fcfc4b9..0072190 100644 --- a/drivers/md/md-cluster.c +++ b/drivers/md/md-cluster.c @@ -44,6 +44,7 @@ struct resync_info { /* md_cluster_info flags */ #define MD_CLUSTER_WAITING_FOR_NEWDISK 1 +#define MD_CLUSTER_SUSPEND_READ_BALANCING 2 struct md_cluster_info { @@ -275,6 +276,9 @@ clear_bit: static void recover_prep(void *arg) { + struct mddev *mddev = arg; + struct md_cluster_info *cinfo = mddev->cluster_info; + set_bit(MD_CLUSTER_SUSPEND_READ_BALANCING, &cinfo->state); } static void recover_slot(void *arg, struct dlm_slot *slot) @@ -307,6 +311,7 @@ static void recover_done(void *arg, struct dlm_slot *slots, cinfo->slot_number = our_slot; complete(&cinfo->completion); + clear_bit(MD_CLUSTER_SUSPEND_READ_BALANCING, &cinfo->state); } static const struct dlm_lockspace_ops md_ls_ops = { @@ -816,12 +821,17 @@ static void resync_finish(struct mddev *mddev) resync_send(mddev, RESYNCING, 0, 0); } -static int area_resyncing(struct mddev *mddev, sector_t lo, sector_t hi) +static int area_resyncing(struct mddev *mddev, int direction, + sector_t lo, sector_t hi) { struct md_cluster_info *cinfo = mddev->cluster_info; int ret = 0; struct suspend_info *s; + if ((direction == READ) && + test_bit(MD_CLUSTER_SUSPEND_READ_BALANCING, &cinfo->state)) + return 1; + spin_lock_irq(&cinfo->suspend_lock); if (list_empty(&cinfo->suspend_list)) goto out; diff --git a/drivers/md/md-cluster.h b/drivers/md/md-cluster.h index 6817ee0..00defe2 100644 --- a/drivers/md/md-cluster.h +++ b/drivers/md/md-cluster.h @@ -18,7 +18,7 @@ struct md_cluster_operations { int (*metadata_update_start)(struct mddev *mddev); int (*metadata_update_finish)(struct mddev *mddev); int (*metadata_update_cancel)(struct mddev *mddev); - int (*area_resyncing)(struct mddev *mddev, sector_t lo, sector_t hi); + int (*area_resyncing)(struct mddev *mddev, int direction, sector_t lo, sector_t hi); int (*add_new_disk_start)(struct mddev *mddev, struct md_rdev *rdev); int (*add_new_disk_finish)(struct mddev *mddev); int (*new_disk_ack)(struct mddev *mddev, bool ack); diff --git a/drivers/md/raid1.c b/drivers/md/raid1.c index 50cf0c8..94f5b55 100644 --- a/drivers/md/raid1.c +++ b/drivers/md/raid1.c @@ -541,7 +541,7 @@ static int read_balance(struct r1conf *conf, struct r1bio *r1_bio, int *max_sect if ((conf->mddev->recovery_cp < this_sector + sectors) || (mddev_is_clustered(conf->mddev) && - md_cluster_ops->area_resyncing(conf->mddev, this_sector, + md_cluster_ops->area_resyncing(conf->mddev, READ, this_sector, this_sector + sectors))) choose_first = 1; else @@ -1111,7 +1111,8 @@ static void make_request(struct mddev *mddev, struct bio * bio) ((bio_end_sector(bio) > mddev->suspend_lo && bio->bi_iter.bi_sector < mddev->suspend_hi) || (mddev_is_clustered(mddev) && - md_cluster_ops->area_resyncing(mddev, bio->bi_iter.bi_sector, bio_end_sector(bio))))) { + md_cluster_ops->area_resyncing(mddev, WRITE, + bio->bi_iter.bi_sector, bio_end_sector(bio))))) { /* As the suspend_* range is controlled by * userspace, we want an interruptible * wait. @@ -1124,7 +1125,7 @@ static void make_request(struct mddev *mddev, struct bio * bio) if (bio_end_sector(bio) <= mddev->suspend_lo || bio->bi_iter.bi_sector >= mddev->suspend_hi || (mddev_is_clustered(mddev) && - !md_cluster_ops->area_resyncing(mddev, + !md_cluster_ops->area_resyncing(mddev, WRITE, bio->bi_iter.bi_sector, bio_end_sector(bio)))) break; schedule(); -- cgit v0.10.2 From e6030cb06c40e4ab4e8c712f13f494a09638ed2c Mon Sep 17 00:00:00 2001 From: NeilBrown Date: Fri, 17 Jul 2015 13:26:23 +1000 Subject: md/raid5: clear R5_NeedReplace when no longer needed. This flag is currently never cleared, which can in rare cases trigger a warn-on if it is still set but the block isn't InSync. So clear it when it isn't need, which includes if the replacement device has failed. Signed-off-by: NeilBrown diff --git a/drivers/md/raid5.c b/drivers/md/raid5.c index 38300ee..643d217 100644 --- a/drivers/md/raid5.c +++ b/drivers/md/raid5.c @@ -4066,8 +4066,10 @@ static void analyse_stripe(struct stripe_head *sh, struct stripe_head_state *s) &first_bad, &bad_sectors)) set_bit(R5_ReadRepl, &dev->flags); else { - if (rdev) + if (rdev && !test_bit(Faulty, &rdev->flags)) set_bit(R5_NeedReplace, &dev->flags); + else + clear_bit(R5_NeedReplace, &dev->flags); rdev = rcu_dereference(conf->disks[i].rdev); clear_bit(R5_ReadRepl, &dev->flags); } -- cgit v0.10.2 From fd4e1393c414275d28577a2242f4e424cf80a605 Mon Sep 17 00:00:00 2001 From: Nicholas Mc Guire Date: Mon, 6 Jul 2015 18:04:40 +0200 Subject: tcm_qla2xxx: pass timeout as HZ independent value API compliance scanning with coccinelle flagged: ./drivers/scsi/qla2xxx/tcm_qla2xxx.c:407:2-29: WARNING: timeout is HZ dependent This was introduced in 'commit 75f8c1f693ee ("[SCSI] tcm_qla2xxx: Add >= 24xx series fabric module for target-core")'. wait_for_completion_timeout() expects a timeout in jiffies so the numeric constant makes the effective timeout HZ dependent. Resolved by converting it to CONST * HZ. Signed-off-by: Nicholas Mc Guire Acked-by: Nilesh Javali Signed-off-by: Nicholas Bellinger diff --git a/drivers/scsi/qla2xxx/tcm_qla2xxx.c b/drivers/scsi/qla2xxx/tcm_qla2xxx.c index d9a8c60..33364b6 100644 --- a/drivers/scsi/qla2xxx/tcm_qla2xxx.c +++ b/drivers/scsi/qla2xxx/tcm_qla2xxx.c @@ -405,7 +405,7 @@ static int tcm_qla2xxx_write_pending_status(struct se_cmd *se_cmd) se_cmd->t_state == TRANSPORT_COMPLETE_QF_WP) { spin_unlock_irqrestore(&se_cmd->t_state_lock, flags); wait_for_completion_timeout(&se_cmd->t_transport_stop_comp, - 3000); + 3 * HZ); return 0; } spin_unlock_irqrestore(&se_cmd->t_state_lock, flags); -- cgit v0.10.2 From bc1a7d6aff763ea29db6319185327f86b0fe93b6 Mon Sep 17 00:00:00 2001 From: Andy Grover Date: Thu, 9 Jul 2015 09:56:47 -0700 Subject: target: Indicate success if writing 0 to pi_prot_type See https://bugzilla.redhat.com/show_bug.cgi?id=1240687 Returning 0 from a configfs store function results in infinite retries. Reported-by: Yanko Kaneti Signed-off-by: Andy Grover Signed-off-by: Nicholas Bellinger diff --git a/drivers/target/target_core_configfs.c b/drivers/target/target_core_configfs.c index 0b0de36..cb09e69 100644 --- a/drivers/target/target_core_configfs.c +++ b/drivers/target/target_core_configfs.c @@ -747,7 +747,7 @@ static ssize_t store_pi_prot_type(struct se_dev_attrib *da, if (!dev->transport->init_prot || !dev->transport->free_prot) { /* 0 is only allowed value for non-supporting backends */ if (flag == 0) - return 0; + return count; pr_err("DIF protection not supported by backend: %s\n", dev->transport->name); -- cgit v0.10.2 From 9105bfc038ca5a506404ce37cd3c0e85f76351e3 Mon Sep 17 00:00:00 2001 From: Andy Grover Date: Thu, 9 Jul 2015 09:56:48 -0700 Subject: target: Do not return 0 from aptpl and alua configfs store functions Here are some more instances where we are returning 0 from a configfs store function, the unintended result of which is likely infinite retries from userspace. Signed-off-by: Andy Grover Signed-off-by: Nicholas Bellinger diff --git a/drivers/target/target_core_configfs.c b/drivers/target/target_core_configfs.c index cb09e69..7f3cb3a 100644 --- a/drivers/target/target_core_configfs.c +++ b/drivers/target/target_core_configfs.c @@ -1590,9 +1590,9 @@ static ssize_t target_core_dev_pr_store_attr_res_aptpl_metadata( u8 type = 0; if (dev->transport->transport_flags & TRANSPORT_FLAG_PASSTHROUGH) - return 0; + return count; if (dev->dev_reservation_flags & DRF_SPC2_RESERVATIONS) - return 0; + return count; if (dev->export_count) { pr_debug("Unable to process APTPL metadata while" @@ -1985,7 +1985,7 @@ static ssize_t target_core_store_alua_lu_gp( lu_gp_mem = dev->dev_alua_lu_gp_mem; if (!lu_gp_mem) - return 0; + return count; if (count > LU_GROUP_NAME_BUF) { pr_err("ALUA LU Group Alias too large!\n"); -- cgit v0.10.2 From c20910264c367a5dfbf6c09e8ec2ff0c5c52857a Mon Sep 17 00:00:00 2001 From: David Disseldorp Date: Sun, 12 Jul 2015 18:49:18 +0200 Subject: target/configfs: handle match_int() errors As a follow up to ce31c1b0dc4038a1dec64585d892adb73d9c45f4 - there are still a few LIO match_int() calls that don't check the return value. Propagate errors rather than using the potentially uninitialised result. Signed-off-by: David Disseldorp Signed-off-by: Nicholas Bellinger diff --git a/drivers/target/target_core_configfs.c b/drivers/target/target_core_configfs.c index 7f3cb3a..c2e9fea 100644 --- a/drivers/target/target_core_configfs.c +++ b/drivers/target/target_core_configfs.c @@ -1658,22 +1658,32 @@ static ssize_t target_core_dev_pr_store_attr_res_aptpl_metadata( * PR APTPL Metadata for Reservation */ case Opt_res_holder: - match_int(args, &arg); + ret = match_int(args, &arg); + if (ret) + goto out; res_holder = arg; break; case Opt_res_type: - match_int(args, &arg); + ret = match_int(args, &arg); + if (ret) + goto out; type = (u8)arg; break; case Opt_res_scope: - match_int(args, &arg); + ret = match_int(args, &arg); + if (ret) + goto out; break; case Opt_res_all_tg_pt: - match_int(args, &arg); + ret = match_int(args, &arg); + if (ret) + goto out; all_tg_pt = (int)arg; break; case Opt_mapped_lun: - match_int(args, &arg); + ret = match_int(args, &arg); + if (ret) + goto out; mapped_lun = (u64)arg; break; /* @@ -1701,14 +1711,20 @@ static ssize_t target_core_dev_pr_store_attr_res_aptpl_metadata( } break; case Opt_tpgt: - match_int(args, &arg); + ret = match_int(args, &arg); + if (ret) + goto out; tpgt = (u16)arg; break; case Opt_port_rtpi: - match_int(args, &arg); + ret = match_int(args, &arg); + if (ret) + goto out; break; case Opt_target_lun: - match_int(args, &arg); + ret = match_int(args, &arg); + if (ret) + goto out; target_lun = (u64)arg; break; default: -- cgit v0.10.2 From 408806f740497c5d71f9c305b3d6aad260ff186d Mon Sep 17 00:00:00 2001 From: Kishon Vijay Abraham I Date: Tue, 16 Jun 2015 16:07:17 +0530 Subject: mmc: omap_hsmmc: Fix DTO and DCRC handling DTO/DCRC errors were not being informed to the mmc core since commit ae4bf788ee9b ("mmc: omap_hsmmc: consolidate error report handling of HSMMC IRQ"). This commit made sure 'end_trans' is never set on DTO/DCRC errors. This is because after this commit 'host->data' is checked after it has been cleared to NULL by omap_hsmmc_dma_cleanup(). Because 'end_trans' is never set, omap_hsmmc_xfer_done() is never invoked making core layer not to be aware of DTO/DCRC errors. Because of this any command invoked after DTO/DCRC error leads to a hang. Fix this by checking for 'host->data' before it is actually cleared. Fixes: ae4bf788ee9b ("mmc: omap_hsmmc: consolidate error report handling of HSMMC IRQ") CC: stable@vger.kernel.org Signed-off-by: Kishon Vijay Abraham I Signed-off-by: Vignesh R Tested-by: Andreas Fenkart Signed-off-by: Ulf Hansson diff --git a/drivers/mmc/host/omap_hsmmc.c b/drivers/mmc/host/omap_hsmmc.c index b2b411d..0132ba7 100644 --- a/drivers/mmc/host/omap_hsmmc.c +++ b/drivers/mmc/host/omap_hsmmc.c @@ -1062,6 +1062,10 @@ static void omap_hsmmc_do_irq(struct omap_hsmmc_host *host, int status) if (status & (CTO_EN | CCRC_EN)) end_cmd = 1; + if (host->data || host->response_busy) { + end_trans = !end_cmd; + host->response_busy = 0; + } if (status & (CTO_EN | DTO_EN)) hsmmc_command_incomplete(host, -ETIMEDOUT, end_cmd); else if (status & (CCRC_EN | DCRC_EN)) @@ -1081,10 +1085,6 @@ static void omap_hsmmc_do_irq(struct omap_hsmmc_host *host, int status) } dev_dbg(mmc_dev(host->mmc), "AC12 err: 0x%x\n", ac12); } - if (host->data || host->response_busy) { - end_trans = !end_cmd; - host->response_busy = 0; - } } OMAP_HSMMC_WRITE(host->base, STAT, status); -- cgit v0.10.2 From 5027cd1e323d2f8136abb32ac7c8e1e1a94eac05 Mon Sep 17 00:00:00 2001 From: Vignesh R Date: Tue, 16 Jun 2015 16:07:18 +0530 Subject: mmc: omap_hsmmc: Handle BADA, DEB and CEB interrupts Sometimes BADA, DEB or CEB error interrupts occur when sd card is unplugged during data transfer. These interrupts are currently ignored by the interrupt handler. But, this results in card not being recognised on subsequent insertion. This is because mmcqd is waiting forever for the data transfer(for which error occurred) to complete. Fix this, by reporting BADA, DEB, CEB errors to mmc-core as -EILSEQ, so that the core can do appropriate handling. Signed-off-by: Vignesh R Tested-by: Andreas Fenkart Signed-off-by: Ulf Hansson diff --git a/drivers/mmc/host/omap_hsmmc.c b/drivers/mmc/host/omap_hsmmc.c index 0132ba7..4d12032 100644 --- a/drivers/mmc/host/omap_hsmmc.c +++ b/drivers/mmc/host/omap_hsmmc.c @@ -1068,7 +1068,8 @@ static void omap_hsmmc_do_irq(struct omap_hsmmc_host *host, int status) } if (status & (CTO_EN | DTO_EN)) hsmmc_command_incomplete(host, -ETIMEDOUT, end_cmd); - else if (status & (CCRC_EN | DCRC_EN)) + else if (status & (CCRC_EN | DCRC_EN | DEB_EN | CEB_EN | + BADA_EN)) hsmmc_command_incomplete(host, -EILSEQ, end_cmd); if (status & ACE_EN) { -- cgit v0.10.2 From 7ac020366b0a436d726408841160b5dc32c19214 Mon Sep 17 00:00:00 2001 From: Peng Fan Date: Mon, 22 Jun 2015 11:41:23 +0800 Subject: mmc: sdhci check parameters before call dma_free_coherent We should not call dma_free_coherent if host->adma_table is NULL, otherwise may trigger panic. Fixes: d1e49f77d7c7 ("mmc: sdhci: convert ADMA descriptors to a...") Signed-off-by: Peng Fan Acked-by: Adrian Hunter Signed-off-by: Ulf Hansson diff --git a/drivers/mmc/host/sdhci.c b/drivers/mmc/host/sdhci.c index bc14452..655900f 100644 --- a/drivers/mmc/host/sdhci.c +++ b/drivers/mmc/host/sdhci.c @@ -2978,8 +2978,11 @@ int sdhci_add_host(struct sdhci_host *host) GFP_KERNEL); host->align_buffer = kmalloc(host->align_buffer_sz, GFP_KERNEL); if (!host->adma_table || !host->align_buffer) { - dma_free_coherent(mmc_dev(mmc), host->adma_table_sz, - host->adma_table, host->adma_addr); + if (host->adma_table) + dma_free_coherent(mmc_dev(mmc), + host->adma_table_sz, + host->adma_table, + host->adma_addr); kfree(host->align_buffer); pr_warn("%s: Unable to allocate ADMA buffers - falling back to standard DMA\n", mmc_hostname(mmc)); -- cgit v0.10.2 From c2b22fff7186f8a6bd6dbfe960ad00421ba82a52 Mon Sep 17 00:00:00 2001 From: Geert Uytterhoeven Date: Fri, 26 Jun 2015 14:09:26 +0200 Subject: mmc: MMC_MTK should depend on HAS_DMA If NO_DMA=y: ERROR: "dma_alloc_coherent" [drivers/mmc/host/mtk-sd.ko] undefined! ERROR: "dma_unmap_sg" [drivers/mmc/host/mtk-sd.ko] undefined! ERROR: "dma_map_sg" [drivers/mmc/host/mtk-sd.ko] undefined! ERROR: "dma_free_coherent" [drivers/mmc/host/mtk-sd.ko] undefined! Add a dependency on HAS_DMA to fix this. Signed-off-by: Geert Uytterhoeven Signed-off-by: Ulf Hansson diff --git a/drivers/mmc/host/Kconfig b/drivers/mmc/host/Kconfig index fd9a58e..6a0f9c7 100644 --- a/drivers/mmc/host/Kconfig +++ b/drivers/mmc/host/Kconfig @@ -779,6 +779,7 @@ config MMC_TOSHIBA_PCI config MMC_MTK tristate "MediaTek SD/MMC Card Interface support" + depends on HAS_DMA help This selects the MediaTek(R) Secure digital and Multimedia card Interface. If you have a machine with a integrated SD/MMC card reader, say Y or M here. -- cgit v0.10.2 From 9098f84cced870f54d8c410dd2444cfa61467fa0 Mon Sep 17 00:00:00 2001 From: Tomas Winkler Date: Thu, 16 Jul 2015 15:50:45 +0200 Subject: mmc: block: Add missing mmc_blk_put() in power_ro_lock_show() Enclosing mmc_blk_put() is missing in power_ro_lock_show() sysfs handler, let's add it. Fixes: add710eaa886 ("mmc: boot partition ro lock support") Signed-off-by: Tomas Winkler Signed-off-by: Ulf Hansson diff --git a/drivers/mmc/card/block.c b/drivers/mmc/card/block.c index c9c3d20..a1b820f 100644 --- a/drivers/mmc/card/block.c +++ b/drivers/mmc/card/block.c @@ -208,6 +208,8 @@ static ssize_t power_ro_lock_show(struct device *dev, ret = snprintf(buf, PAGE_SIZE, "%d\n", locked); + mmc_blk_put(md); + return ret; } -- cgit v0.10.2 From 8e91125ff3f57f15c6568e2a6d32743b3f7815e4 Mon Sep 17 00:00:00 2001 From: Joakim Tjernlund Date: Wed, 22 Jul 2015 16:44:26 +0200 Subject: mmc: sdhci-esdhc: Make 8BIT bus work Support for 8BIT bus with was added some time ago to sdhci-esdhc but then missed to remove the 8BIT from the reserved bit mask which made 8BIT non functional. Fixes: 66b50a00992d ("mmc: esdhc: Add support for 8-bit bus width and..") Signed-off-by: Joakim Tjernlund Signed-off-by: Ulf Hansson diff --git a/drivers/mmc/host/sdhci-esdhc.h b/drivers/mmc/host/sdhci-esdhc.h index 3497cfa..a870c42 100644 --- a/drivers/mmc/host/sdhci-esdhc.h +++ b/drivers/mmc/host/sdhci-esdhc.h @@ -45,6 +45,6 @@ #define ESDHC_DMA_SYSCTL 0x40c #define ESDHC_DMA_SNOOP 0x00000040 -#define ESDHC_HOST_CONTROL_RES 0x05 +#define ESDHC_HOST_CONTROL_RES 0x01 #endif /* _DRIVERS_MMC_SDHCI_ESDHC_H */ -- cgit v0.10.2 From aca45c0e95dad1c4ba4d38da192756b0e10cbbbd Mon Sep 17 00:00:00 2001 From: Dong Aisheng Date: Wed, 22 Jul 2015 20:53:00 +0800 Subject: dts: imx51: fix sd card gpio polarity specified in device tree cd-gpios polarity should be changed to GPIO_ACTIVE_LOW and wp-gpios should be changed to GPIO_ACTIVE_HIGH. Otherwise, the SD may not work properly due to wrong polarity inversion specified in DT after switch to common parsing function mmc_of_parse(). Signed-off-by: Dong Aisheng Acked-by: Shawn Guo Signed-off-by: Ulf Hansson diff --git a/arch/arm/boot/dts/imx51-apf51dev.dts b/arch/arm/boot/dts/imx51-apf51dev.dts index 93d3ea1..0f3fe29 100644 --- a/arch/arm/boot/dts/imx51-apf51dev.dts +++ b/arch/arm/boot/dts/imx51-apf51dev.dts @@ -98,7 +98,7 @@ &esdhc1 { pinctrl-names = "default"; pinctrl-0 = <&pinctrl_esdhc1>; - cd-gpios = <&gpio2 29 GPIO_ACTIVE_HIGH>; + cd-gpios = <&gpio2 29 GPIO_ACTIVE_LOW>; bus-width = <4>; status = "okay"; }; -- cgit v0.10.2 From 94d76946859b4bcfa0da373357f14feda2af0622 Mon Sep 17 00:00:00 2001 From: Dong Aisheng Date: Wed, 22 Jul 2015 20:53:01 +0800 Subject: dts: imx53: fix sd card gpio polarity specified in device tree cd-gpios polarity should be changed to GPIO_ACTIVE_LOW and wp-gpios should be changed to GPIO_ACTIVE_HIGH. Otherwise, the SD may not work properly due to wrong polarity inversion specified in DT after switch to common parsing function mmc_of_parse(). Signed-off-by: Dong Aisheng Acked-by: Shawn Guo Signed-off-by: Ulf Hansson diff --git a/arch/arm/boot/dts/imx53-ard.dts b/arch/arm/boot/dts/imx53-ard.dts index e9337ad..3bc1883 100644 --- a/arch/arm/boot/dts/imx53-ard.dts +++ b/arch/arm/boot/dts/imx53-ard.dts @@ -103,8 +103,8 @@ &esdhc1 { pinctrl-names = "default"; pinctrl-0 = <&pinctrl_esdhc1>; - cd-gpios = <&gpio1 1 0>; - wp-gpios = <&gpio1 9 0>; + cd-gpios = <&gpio1 1 GPIO_ACTIVE_LOW>; + wp-gpios = <&gpio1 9 GPIO_ACTIVE_HIGH>; status = "okay"; }; diff --git a/arch/arm/boot/dts/imx53-m53evk.dts b/arch/arm/boot/dts/imx53-m53evk.dts index d0e0f57..53f4088 100644 --- a/arch/arm/boot/dts/imx53-m53evk.dts +++ b/arch/arm/boot/dts/imx53-m53evk.dts @@ -124,8 +124,8 @@ &esdhc1 { pinctrl-names = "default"; pinctrl-0 = <&pinctrl_esdhc1>; - cd-gpios = <&gpio1 1 0>; - wp-gpios = <&gpio1 9 0>; + cd-gpios = <&gpio1 1 GPIO_ACTIVE_LOW>; + wp-gpios = <&gpio1 9 GPIO_ACTIVE_HIGH>; status = "okay"; }; diff --git a/arch/arm/boot/dts/imx53-qsb-common.dtsi b/arch/arm/boot/dts/imx53-qsb-common.dtsi index ab4ba39..b0d5542 100644 --- a/arch/arm/boot/dts/imx53-qsb-common.dtsi +++ b/arch/arm/boot/dts/imx53-qsb-common.dtsi @@ -147,8 +147,8 @@ &esdhc3 { pinctrl-names = "default"; pinctrl-0 = <&pinctrl_esdhc3>; - cd-gpios = <&gpio3 11 0>; - wp-gpios = <&gpio3 12 0>; + cd-gpios = <&gpio3 11 GPIO_ACTIVE_LOW>; + wp-gpios = <&gpio3 12 GPIO_ACTIVE_HIGH>; bus-width = <8>; status = "okay"; }; diff --git a/arch/arm/boot/dts/imx53-smd.dts b/arch/arm/boot/dts/imx53-smd.dts index 1d32557..fc89ce1 100644 --- a/arch/arm/boot/dts/imx53-smd.dts +++ b/arch/arm/boot/dts/imx53-smd.dts @@ -41,8 +41,8 @@ &esdhc1 { pinctrl-names = "default"; pinctrl-0 = <&pinctrl_esdhc1>; - cd-gpios = <&gpio3 13 0>; - wp-gpios = <&gpio4 11 0>; + cd-gpios = <&gpio3 13 GPIO_ACTIVE_LOW>; + wp-gpios = <&gpio4 11 GPIO_ACTIVE_HIGH>; status = "okay"; }; diff --git a/arch/arm/boot/dts/imx53-tqma53.dtsi b/arch/arm/boot/dts/imx53-tqma53.dtsi index 4f1f0e2..e03373a 100644 --- a/arch/arm/boot/dts/imx53-tqma53.dtsi +++ b/arch/arm/boot/dts/imx53-tqma53.dtsi @@ -41,8 +41,8 @@ pinctrl-0 = <&pinctrl_esdhc2>, <&pinctrl_esdhc2_cdwp>; vmmc-supply = <®_3p3v>; - wp-gpios = <&gpio1 2 0>; - cd-gpios = <&gpio1 4 0>; + wp-gpios = <&gpio1 2 GPIO_ACTIVE_HIGH>; + cd-gpios = <&gpio1 4 GPIO_ACTIVE_LOW>; status = "disabled"; }; diff --git a/arch/arm/boot/dts/imx53-tx53.dtsi b/arch/arm/boot/dts/imx53-tx53.dtsi index 704bd72..d3e50b2 100644 --- a/arch/arm/boot/dts/imx53-tx53.dtsi +++ b/arch/arm/boot/dts/imx53-tx53.dtsi @@ -183,7 +183,7 @@ }; &esdhc1 { - cd-gpios = <&gpio3 24 GPIO_ACTIVE_HIGH>; + cd-gpios = <&gpio3 24 GPIO_ACTIVE_LOW>; fsl,wp-controller; pinctrl-names = "default"; pinctrl-0 = <&pinctrl_esdhc1>; @@ -191,7 +191,7 @@ }; &esdhc2 { - cd-gpios = <&gpio3 25 GPIO_ACTIVE_HIGH>; + cd-gpios = <&gpio3 25 GPIO_ACTIVE_LOW>; fsl,wp-controller; pinctrl-names = "default"; pinctrl-0 = <&pinctrl_esdhc2>; diff --git a/arch/arm/boot/dts/imx53-voipac-bsb.dts b/arch/arm/boot/dts/imx53-voipac-bsb.dts index c17d3ad..fc51b87 100644 --- a/arch/arm/boot/dts/imx53-voipac-bsb.dts +++ b/arch/arm/boot/dts/imx53-voipac-bsb.dts @@ -119,8 +119,8 @@ &esdhc2 { pinctrl-names = "default"; pinctrl-0 = <&pinctrl_esdhc2>; - cd-gpios = <&gpio3 25 0>; - wp-gpios = <&gpio2 19 0>; + cd-gpios = <&gpio3 25 GPIO_ACTIVE_LOW>; + wp-gpios = <&gpio2 19 GPIO_ACTIVE_HIGH>; vmmc-supply = <®_3p3v>; status = "okay"; }; -- cgit v0.10.2 From 89c1a8cf63f8c69dfddb6e377c04df8b25ab1c88 Mon Sep 17 00:00:00 2001 From: Dong Aisheng Date: Wed, 22 Jul 2015 20:53:02 +0800 Subject: dts: imx6: fix sd card gpio polarity specified in device tree cd-gpios polarity should be changed to GPIO_ACTIVE_LOW and wp-gpios should be changed to GPIO_ACTIVE_HIGH. Otherwise, the SD may not work properly due to wrong polarity inversion specified in DT after switch to common parsing function mmc_of_parse(). Signed-off-by: Dong Aisheng Acked-by: Shawn Guo Signed-off-by: Ulf Hansson diff --git a/arch/arm/boot/dts/imx6dl-riotboard.dts b/arch/arm/boot/dts/imx6dl-riotboard.dts index 43cb3fd..5111f51 100644 --- a/arch/arm/boot/dts/imx6dl-riotboard.dts +++ b/arch/arm/boot/dts/imx6dl-riotboard.dts @@ -305,8 +305,8 @@ &usdhc2 { pinctrl-names = "default"; pinctrl-0 = <&pinctrl_usdhc2>; - cd-gpios = <&gpio1 4 0>; - wp-gpios = <&gpio1 2 0>; + cd-gpios = <&gpio1 4 GPIO_ACTIVE_LOW>; + wp-gpios = <&gpio1 2 GPIO_ACTIVE_HIGH>; vmmc-supply = <®_3p3v>; status = "okay"; }; @@ -314,8 +314,8 @@ &usdhc3 { pinctrl-names = "default"; pinctrl-0 = <&pinctrl_usdhc3>; - cd-gpios = <&gpio7 0 0>; - wp-gpios = <&gpio7 1 0>; + cd-gpios = <&gpio7 0 GPIO_ACTIVE_LOW>; + wp-gpios = <&gpio7 1 GPIO_ACTIVE_HIGH>; vmmc-supply = <®_3p3v>; status = "okay"; }; diff --git a/arch/arm/boot/dts/imx6q-arm2.dts b/arch/arm/boot/dts/imx6q-arm2.dts index 78df05e..d6515f7 100644 --- a/arch/arm/boot/dts/imx6q-arm2.dts +++ b/arch/arm/boot/dts/imx6q-arm2.dts @@ -11,6 +11,7 @@ */ /dts-v1/; +#include #include "imx6q.dtsi" / { @@ -196,8 +197,8 @@ }; &usdhc3 { - cd-gpios = <&gpio6 11 0>; - wp-gpios = <&gpio6 14 0>; + cd-gpios = <&gpio6 11 GPIO_ACTIVE_LOW>; + wp-gpios = <&gpio6 14 GPIO_ACTIVE_HIGH>; vmmc-supply = <®_3p3v>; pinctrl-names = "default"; pinctrl-0 = <&pinctrl_usdhc3 diff --git a/arch/arm/boot/dts/imx6q-gk802.dts b/arch/arm/boot/dts/imx6q-gk802.dts index 703539c..00bd63e 100644 --- a/arch/arm/boot/dts/imx6q-gk802.dts +++ b/arch/arm/boot/dts/imx6q-gk802.dts @@ -7,6 +7,7 @@ */ /dts-v1/; +#include #include "imx6q.dtsi" / { @@ -161,7 +162,7 @@ pinctrl-names = "default"; pinctrl-0 = <&pinctrl_usdhc3>; bus-width = <4>; - cd-gpios = <&gpio6 11 0>; + cd-gpios = <&gpio6 11 GPIO_ACTIVE_LOW>; vmmc-supply = <®_3p3v>; status = "okay"; }; diff --git a/arch/arm/boot/dts/imx6q-tbs2910.dts b/arch/arm/boot/dts/imx6q-tbs2910.dts index a43abfa..5645d52 100644 --- a/arch/arm/boot/dts/imx6q-tbs2910.dts +++ b/arch/arm/boot/dts/imx6q-tbs2910.dts @@ -251,7 +251,7 @@ pinctrl-names = "default"; pinctrl-0 = <&pinctrl_usdhc2>; bus-width = <4>; - cd-gpios = <&gpio2 2 GPIO_ACTIVE_HIGH>; + cd-gpios = <&gpio2 2 GPIO_ACTIVE_LOW>; vmmc-supply = <®_3p3v>; status = "okay"; }; @@ -260,7 +260,7 @@ pinctrl-names = "default"; pinctrl-0 = <&pinctrl_usdhc3>; bus-width = <4>; - cd-gpios = <&gpio2 0 GPIO_ACTIVE_HIGH>; + cd-gpios = <&gpio2 0 GPIO_ACTIVE_LOW>; wp-gpios = <&gpio2 1 GPIO_ACTIVE_HIGH>; vmmc-supply = <®_3p3v>; status = "okay"; diff --git a/arch/arm/boot/dts/imx6qdl-aristainetos.dtsi b/arch/arm/boot/dts/imx6qdl-aristainetos.dtsi index e6d9195..f4d6ae5 100644 --- a/arch/arm/boot/dts/imx6qdl-aristainetos.dtsi +++ b/arch/arm/boot/dts/imx6qdl-aristainetos.dtsi @@ -173,7 +173,7 @@ pinctrl-names = "default"; pinctrl-0 = <&pinctrl_usdhc1>; vmmc-supply = <®_3p3v>; - cd-gpios = <&gpio4 7 GPIO_ACTIVE_HIGH>; + cd-gpios = <&gpio4 7 GPIO_ACTIVE_LOW>; status = "okay"; }; @@ -181,7 +181,7 @@ pinctrl-names = "default"; pinctrl-0 = <&pinctrl_usdhc2>; vmmc-supply = <®_3p3v>; - cd-gpios = <&gpio4 8 GPIO_ACTIVE_HIGH>; + cd-gpios = <&gpio4 8 GPIO_ACTIVE_LOW>; status = "okay"; }; diff --git a/arch/arm/boot/dts/imx6qdl-aristainetos2.dtsi b/arch/arm/boot/dts/imx6qdl-aristainetos2.dtsi index 1d85de2..a47a039 100644 --- a/arch/arm/boot/dts/imx6qdl-aristainetos2.dtsi +++ b/arch/arm/boot/dts/imx6qdl-aristainetos2.dtsi @@ -392,7 +392,7 @@ &usdhc1 { pinctrl-names = "default"; pinctrl-0 = <&pinctrl_usdhc1>; - cd-gpios = <&gpio1 27 GPIO_ACTIVE_HIGH>; + cd-gpios = <&gpio1 27 GPIO_ACTIVE_LOW>; no-1-8-v; status = "okay"; }; @@ -400,7 +400,7 @@ &usdhc2 { pinctrl-names = "default"; pinctrl-0 = <&pinctrl_usdhc2>; - cd-gpios = <&gpio4 5 GPIO_ACTIVE_HIGH>; + cd-gpios = <&gpio4 5 GPIO_ACTIVE_LOW>; wp-gpios = <&gpio2 10 GPIO_ACTIVE_HIGH>; no-1-8-v; status = "okay"; diff --git a/arch/arm/boot/dts/imx6qdl-cubox-i.dtsi b/arch/arm/boot/dts/imx6qdl-cubox-i.dtsi index 59e5d15..ff41f83 100644 --- a/arch/arm/boot/dts/imx6qdl-cubox-i.dtsi +++ b/arch/arm/boot/dts/imx6qdl-cubox-i.dtsi @@ -258,6 +258,6 @@ pinctrl-names = "default"; pinctrl-0 = <&pinctrl_cubox_i_usdhc2_aux &pinctrl_cubox_i_usdhc2>; vmmc-supply = <®_3p3v>; - cd-gpios = <&gpio1 4 0>; + cd-gpios = <&gpio1 4 GPIO_ACTIVE_LOW>; status = "okay"; }; diff --git a/arch/arm/boot/dts/imx6qdl-dfi-fs700-m60.dtsi b/arch/arm/boot/dts/imx6qdl-dfi-fs700-m60.dtsi index 2c253d6..45e7c39 100644 --- a/arch/arm/boot/dts/imx6qdl-dfi-fs700-m60.dtsi +++ b/arch/arm/boot/dts/imx6qdl-dfi-fs700-m60.dtsi @@ -1,3 +1,5 @@ +#include + / { regulators { compatible = "simple-bus"; @@ -181,7 +183,7 @@ &usdhc2 { /* module slot */ pinctrl-names = "default"; pinctrl-0 = <&pinctrl_usdhc2>; - cd-gpios = <&gpio2 2 0>; + cd-gpios = <&gpio2 2 GPIO_ACTIVE_LOW>; status = "okay"; }; diff --git a/arch/arm/boot/dts/imx6qdl-gw52xx.dtsi b/arch/arm/boot/dts/imx6qdl-gw52xx.dtsi index b5756c2..4493f6e 100644 --- a/arch/arm/boot/dts/imx6qdl-gw52xx.dtsi +++ b/arch/arm/boot/dts/imx6qdl-gw52xx.dtsi @@ -318,7 +318,7 @@ &usdhc3 { pinctrl-names = "default"; pinctrl-0 = <&pinctrl_usdhc3>; - cd-gpios = <&gpio7 0 GPIO_ACTIVE_HIGH>; + cd-gpios = <&gpio7 0 GPIO_ACTIVE_LOW>; vmmc-supply = <®_3p3v>; status = "okay"; }; diff --git a/arch/arm/boot/dts/imx6qdl-gw53xx.dtsi b/arch/arm/boot/dts/imx6qdl-gw53xx.dtsi index 86f03c1..a857d12 100644 --- a/arch/arm/boot/dts/imx6qdl-gw53xx.dtsi +++ b/arch/arm/boot/dts/imx6qdl-gw53xx.dtsi @@ -324,7 +324,7 @@ &usdhc3 { pinctrl-names = "default"; pinctrl-0 = <&pinctrl_usdhc3>; - cd-gpios = <&gpio7 0 GPIO_ACTIVE_HIGH>; + cd-gpios = <&gpio7 0 GPIO_ACTIVE_LOW>; vmmc-supply = <®_3p3v>; status = "okay"; }; diff --git a/arch/arm/boot/dts/imx6qdl-gw54xx.dtsi b/arch/arm/boot/dts/imx6qdl-gw54xx.dtsi index 4a8d97f..1afe338 100644 --- a/arch/arm/boot/dts/imx6qdl-gw54xx.dtsi +++ b/arch/arm/boot/dts/imx6qdl-gw54xx.dtsi @@ -417,7 +417,7 @@ &usdhc3 { pinctrl-names = "default"; pinctrl-0 = <&pinctrl_usdhc3>; - cd-gpios = <&gpio7 0 GPIO_ACTIVE_HIGH>; + cd-gpios = <&gpio7 0 GPIO_ACTIVE_LOW>; vmmc-supply = <®_3p3v>; status = "okay"; }; diff --git a/arch/arm/boot/dts/imx6qdl-hummingboard.dtsi b/arch/arm/boot/dts/imx6qdl-hummingboard.dtsi index 62a82f3..6dd0b76 100644 --- a/arch/arm/boot/dts/imx6qdl-hummingboard.dtsi +++ b/arch/arm/boot/dts/imx6qdl-hummingboard.dtsi @@ -299,6 +299,6 @@ &pinctrl_hummingboard_usdhc2 >; vmmc-supply = <®_3p3v>; - cd-gpios = <&gpio1 4 0>; + cd-gpios = <&gpio1 4 GPIO_ACTIVE_LOW>; status = "okay"; }; diff --git a/arch/arm/boot/dts/imx6qdl-nitrogen6x.dtsi b/arch/arm/boot/dts/imx6qdl-nitrogen6x.dtsi index 3af16df..d7fe667 100644 --- a/arch/arm/boot/dts/imx6qdl-nitrogen6x.dtsi +++ b/arch/arm/boot/dts/imx6qdl-nitrogen6x.dtsi @@ -453,7 +453,7 @@ &usdhc3 { pinctrl-names = "default"; pinctrl-0 = <&pinctrl_usdhc3>; - cd-gpios = <&gpio7 0 0>; + cd-gpios = <&gpio7 0 GPIO_ACTIVE_LOW>; vmmc-supply = <®_3p3v>; status = "okay"; }; @@ -461,7 +461,7 @@ &usdhc4 { pinctrl-names = "default"; pinctrl-0 = <&pinctrl_usdhc4>; - cd-gpios = <&gpio2 6 0>; + cd-gpios = <&gpio2 6 GPIO_ACTIVE_LOW>; vmmc-supply = <®_3p3v>; status = "okay"; }; diff --git a/arch/arm/boot/dts/imx6qdl-phytec-pfla02.dtsi b/arch/arm/boot/dts/imx6qdl-phytec-pfla02.dtsi index 1ce6133..9e6ecd9 100644 --- a/arch/arm/boot/dts/imx6qdl-phytec-pfla02.dtsi +++ b/arch/arm/boot/dts/imx6qdl-phytec-pfla02.dtsi @@ -409,8 +409,8 @@ &usdhc2 { pinctrl-names = "default"; pinctrl-0 = <&pinctrl_usdhc2>; - cd-gpios = <&gpio1 4 0>; - wp-gpios = <&gpio1 2 0>; + cd-gpios = <&gpio1 4 GPIO_ACTIVE_LOW>; + wp-gpios = <&gpio1 2 GPIO_ACTIVE_HIGH>; status = "disabled"; }; @@ -418,7 +418,7 @@ pinctrl-names = "default"; pinctrl-0 = <&pinctrl_usdhc3 &pinctrl_usdhc3_cdwp>; - cd-gpios = <&gpio1 27 0>; - wp-gpios = <&gpio1 29 0>; + cd-gpios = <&gpio1 27 GPIO_ACTIVE_LOW>; + wp-gpios = <&gpio1 29 GPIO_ACTIVE_HIGH>; status = "disabled"; }; diff --git a/arch/arm/boot/dts/imx6qdl-rex.dtsi b/arch/arm/boot/dts/imx6qdl-rex.dtsi index 488a640..3373fd9 100644 --- a/arch/arm/boot/dts/imx6qdl-rex.dtsi +++ b/arch/arm/boot/dts/imx6qdl-rex.dtsi @@ -342,7 +342,7 @@ pinctrl-0 = <&pinctrl_usdhc2>; bus-width = <4>; cd-gpios = <&gpio2 2 GPIO_ACTIVE_LOW>; - wp-gpios = <&gpio2 3 GPIO_ACTIVE_LOW>; + wp-gpios = <&gpio2 3 GPIO_ACTIVE_HIGH>; status = "okay"; }; @@ -351,6 +351,6 @@ pinctrl-0 = <&pinctrl_usdhc3>; bus-width = <4>; cd-gpios = <&gpio2 0 GPIO_ACTIVE_LOW>; - wp-gpios = <&gpio2 1 GPIO_ACTIVE_LOW>; + wp-gpios = <&gpio2 1 GPIO_ACTIVE_HIGH>; status = "okay"; }; diff --git a/arch/arm/boot/dts/imx6qdl-sabreauto.dtsi b/arch/arm/boot/dts/imx6qdl-sabreauto.dtsi index 3b24b126..e329ca5 100644 --- a/arch/arm/boot/dts/imx6qdl-sabreauto.dtsi +++ b/arch/arm/boot/dts/imx6qdl-sabreauto.dtsi @@ -467,8 +467,8 @@ pinctrl-0 = <&pinctrl_usdhc3>; pinctrl-1 = <&pinctrl_usdhc3_100mhz>; pinctrl-2 = <&pinctrl_usdhc3_200mhz>; - cd-gpios = <&gpio6 15 0>; - wp-gpios = <&gpio1 13 0>; + cd-gpios = <&gpio6 15 GPIO_ACTIVE_LOW>; + wp-gpios = <&gpio1 13 GPIO_ACTIVE_HIGH>; status = "okay"; }; diff --git a/arch/arm/boot/dts/imx6qdl-sabrelite.dtsi b/arch/arm/boot/dts/imx6qdl-sabrelite.dtsi index e00c44f..7823793 100644 --- a/arch/arm/boot/dts/imx6qdl-sabrelite.dtsi +++ b/arch/arm/boot/dts/imx6qdl-sabrelite.dtsi @@ -448,8 +448,8 @@ &usdhc3 { pinctrl-names = "default"; pinctrl-0 = <&pinctrl_usdhc3>; - cd-gpios = <&gpio7 0 0>; - wp-gpios = <&gpio7 1 0>; + cd-gpios = <&gpio7 0 GPIO_ACTIVE_LOW>; + wp-gpios = <&gpio7 1 GPIO_ACTIVE_HIGH>; vmmc-supply = <®_3p3v>; status = "okay"; }; @@ -457,7 +457,7 @@ &usdhc4 { pinctrl-names = "default"; pinctrl-0 = <&pinctrl_usdhc4>; - cd-gpios = <&gpio2 6 0>; + cd-gpios = <&gpio2 6 GPIO_ACTIVE_LOW>; vmmc-supply = <®_3p3v>; status = "okay"; }; diff --git a/arch/arm/boot/dts/imx6qdl-sabresd.dtsi b/arch/arm/boot/dts/imx6qdl-sabresd.dtsi index a626e6d..944eb81 100644 --- a/arch/arm/boot/dts/imx6qdl-sabresd.dtsi +++ b/arch/arm/boot/dts/imx6qdl-sabresd.dtsi @@ -562,8 +562,8 @@ pinctrl-names = "default"; pinctrl-0 = <&pinctrl_usdhc2>; bus-width = <8>; - cd-gpios = <&gpio2 2 0>; - wp-gpios = <&gpio2 3 0>; + cd-gpios = <&gpio2 2 GPIO_ACTIVE_LOW>; + wp-gpios = <&gpio2 3 GPIO_ACTIVE_HIGH>; status = "okay"; }; @@ -571,8 +571,8 @@ pinctrl-names = "default"; pinctrl-0 = <&pinctrl_usdhc3>; bus-width = <8>; - cd-gpios = <&gpio2 0 0>; - wp-gpios = <&gpio2 1 0>; + cd-gpios = <&gpio2 0 GPIO_ACTIVE_LOW>; + wp-gpios = <&gpio2 1 GPIO_ACTIVE_HIGH>; status = "okay"; }; diff --git a/arch/arm/boot/dts/imx6qdl-tx6.dtsi b/arch/arm/boot/dts/imx6qdl-tx6.dtsi index f02b80b..da08de3 100644 --- a/arch/arm/boot/dts/imx6qdl-tx6.dtsi +++ b/arch/arm/boot/dts/imx6qdl-tx6.dtsi @@ -680,7 +680,7 @@ pinctrl-0 = <&pinctrl_usdhc1>; bus-width = <4>; no-1-8-v; - cd-gpios = <&gpio7 2 0>; + cd-gpios = <&gpio7 2 GPIO_ACTIVE_LOW>; fsl,wp-controller; status = "okay"; }; @@ -690,7 +690,7 @@ pinctrl-0 = <&pinctrl_usdhc2>; bus-width = <4>; no-1-8-v; - cd-gpios = <&gpio7 3 0>; + cd-gpios = <&gpio7 3 GPIO_ACTIVE_LOW>; fsl,wp-controller; status = "okay"; }; diff --git a/arch/arm/boot/dts/imx6qdl-wandboard.dtsi b/arch/arm/boot/dts/imx6qdl-wandboard.dtsi index 5fb0916..9e096d8 100644 --- a/arch/arm/boot/dts/imx6qdl-wandboard.dtsi +++ b/arch/arm/boot/dts/imx6qdl-wandboard.dtsi @@ -9,6 +9,8 @@ * */ +#include + / { regulators { compatible = "simple-bus"; @@ -250,13 +252,13 @@ &usdhc1 { pinctrl-names = "default"; pinctrl-0 = <&pinctrl_usdhc1>; - cd-gpios = <&gpio1 2 0>; + cd-gpios = <&gpio1 2 GPIO_ACTIVE_LOW>; status = "okay"; }; &usdhc3 { pinctrl-names = "default"; pinctrl-0 = <&pinctrl_usdhc3>; - cd-gpios = <&gpio3 9 0>; + cd-gpios = <&gpio3 9 GPIO_ACTIVE_LOW>; status = "okay"; }; diff --git a/arch/arm/boot/dts/imx6sl-evk.dts b/arch/arm/boot/dts/imx6sl-evk.dts index 945887d..b84dff2 100644 --- a/arch/arm/boot/dts/imx6sl-evk.dts +++ b/arch/arm/boot/dts/imx6sl-evk.dts @@ -617,8 +617,8 @@ pinctrl-1 = <&pinctrl_usdhc1_100mhz>; pinctrl-2 = <&pinctrl_usdhc1_200mhz>; bus-width = <8>; - cd-gpios = <&gpio4 7 0>; - wp-gpios = <&gpio4 6 0>; + cd-gpios = <&gpio4 7 GPIO_ACTIVE_LOW>; + wp-gpios = <&gpio4 6 GPIO_ACTIVE_HIGH>; status = "okay"; }; @@ -627,8 +627,8 @@ pinctrl-0 = <&pinctrl_usdhc2>; pinctrl-1 = <&pinctrl_usdhc2_100mhz>; pinctrl-2 = <&pinctrl_usdhc2_200mhz>; - cd-gpios = <&gpio5 0 0>; - wp-gpios = <&gpio4 29 0>; + cd-gpios = <&gpio5 0 GPIO_ACTIVE_LOW>; + wp-gpios = <&gpio4 29 GPIO_ACTIVE_HIGH>; status = "okay"; }; @@ -637,6 +637,6 @@ pinctrl-0 = <&pinctrl_usdhc3>; pinctrl-1 = <&pinctrl_usdhc3_100mhz>; pinctrl-2 = <&pinctrl_usdhc3_200mhz>; - cd-gpios = <&gpio3 22 0>; + cd-gpios = <&gpio3 22 GPIO_ACTIVE_LOW>; status = "okay"; }; diff --git a/arch/arm/boot/dts/imx6sx-sabreauto.dts b/arch/arm/boot/dts/imx6sx-sabreauto.dts index e3c0b63..115f3fd 100644 --- a/arch/arm/boot/dts/imx6sx-sabreauto.dts +++ b/arch/arm/boot/dts/imx6sx-sabreauto.dts @@ -49,7 +49,7 @@ pinctrl-1 = <&pinctrl_usdhc3_100mhz>; pinctrl-2 = <&pinctrl_usdhc3_200mhz>; bus-width = <8>; - cd-gpios = <&gpio7 10 GPIO_ACTIVE_HIGH>; + cd-gpios = <&gpio7 10 GPIO_ACTIVE_LOW>; wp-gpios = <&gpio3 19 GPIO_ACTIVE_HIGH>; keep-power-in-suspend; enable-sdio-wakeup; @@ -61,7 +61,7 @@ pinctrl-names = "default"; pinctrl-0 = <&pinctrl_usdhc4>; bus-width = <8>; - cd-gpios = <&gpio7 11 GPIO_ACTIVE_HIGH>; + cd-gpios = <&gpio7 11 GPIO_ACTIVE_LOW>; no-1-8-v; keep-power-in-suspend; enable-sdio-wakup; diff --git a/arch/arm/boot/dts/imx6sx-sdb.dtsi b/arch/arm/boot/dts/imx6sx-sdb.dtsi index cef04ce..ac88c34 100644 --- a/arch/arm/boot/dts/imx6sx-sdb.dtsi +++ b/arch/arm/boot/dts/imx6sx-sdb.dtsi @@ -293,7 +293,7 @@ pinctrl-1 = <&pinctrl_usdhc3_100mhz>; pinctrl-2 = <&pinctrl_usdhc3_200mhz>; bus-width = <8>; - cd-gpios = <&gpio2 10 GPIO_ACTIVE_HIGH>; + cd-gpios = <&gpio2 10 GPIO_ACTIVE_LOW>; wp-gpios = <&gpio2 15 GPIO_ACTIVE_HIGH>; keep-power-in-suspend; enable-sdio-wakeup; @@ -304,7 +304,7 @@ &usdhc4 { pinctrl-names = "default"; pinctrl-0 = <&pinctrl_usdhc4>; - cd-gpios = <&gpio6 21 GPIO_ACTIVE_HIGH>; + cd-gpios = <&gpio6 21 GPIO_ACTIVE_LOW>; wp-gpios = <&gpio6 20 GPIO_ACTIVE_HIGH>; status = "okay"; }; -- cgit v0.10.2 From cf75eb15be2bdd054a76aeef6458beaa4a6ab770 Mon Sep 17 00:00:00 2001 From: Dong Aisheng Date: Wed, 22 Jul 2015 20:53:03 +0800 Subject: dts: imx25: fix sd card gpio polarity specified in device tree cd-gpios polarity should be changed to GPIO_ACTIVE_LOW and wp-gpios should be changed to GPIO_ACTIVE_HIGH. Otherwise, the SD may not work properly due to wrong polarity inversion specified in DT after switch to common parsing function mmc_of_parse(). Signed-off-by: Dong Aisheng Acked-by: Shawn Guo Signed-off-by: Ulf Hansson diff --git a/arch/arm/boot/dts/imx25-pdk.dts b/arch/arm/boot/dts/imx25-pdk.dts index dd45e69..9351296 100644 --- a/arch/arm/boot/dts/imx25-pdk.dts +++ b/arch/arm/boot/dts/imx25-pdk.dts @@ -10,6 +10,7 @@ */ /dts-v1/; +#include #include #include "imx25.dtsi" @@ -114,8 +115,8 @@ &esdhc1 { pinctrl-names = "default"; pinctrl-0 = <&pinctrl_esdhc1>; - cd-gpios = <&gpio2 1 0>; - wp-gpios = <&gpio2 0 0>; + cd-gpios = <&gpio2 1 GPIO_ACTIVE_LOW>; + wp-gpios = <&gpio2 0 GPIO_ACTIVE_HIGH>; status = "okay"; }; -- cgit v0.10.2 From 1cd55947f2bfdb99a2a741d65c7463d106ef9132 Mon Sep 17 00:00:00 2001 From: Dong Aisheng Date: Wed, 22 Jul 2015 20:53:04 +0800 Subject: dts: imx7: fix sd card gpio polarity specified in device tree cd-gpios polarity should be changed to GPIO_ACTIVE_LOW and wp-gpios should be changed to GPIO_ACTIVE_HIGH. Otherwise, the SD may not work properly due to wrong polarity inversion specified in DT after switch to common parsing function mmc_of_parse(). Signed-off-by: Dong Aisheng Acked-by: Shawn Guo Signed-off-by: Ulf Hansson diff --git a/arch/arm/boot/dts/imx7d-sdb.dts b/arch/arm/boot/dts/imx7d-sdb.dts index 4d1a4b9..fdd1d7c 100644 --- a/arch/arm/boot/dts/imx7d-sdb.dts +++ b/arch/arm/boot/dts/imx7d-sdb.dts @@ -234,8 +234,8 @@ &usdhc1 { pinctrl-names = "default"; pinctrl-0 = <&pinctrl_usdhc1>; - cd-gpios = <&gpio5 0 0>; - wp-gpios = <&gpio5 1 0>; + cd-gpios = <&gpio5 0 GPIO_ACTIVE_LOW>; + wp-gpios = <&gpio5 1 GPIO_ACTIVE_HIGH>; enable-sdio-wakeup; keep-power-in-suspend; status = "okay"; -- cgit v0.10.2 From 4800e87a2ee886c1972deb73f057d1a6541edb36 Mon Sep 17 00:00:00 2001 From: Dong Aisheng Date: Wed, 22 Jul 2015 20:53:05 +0800 Subject: mmc: sdhci-esdhc-imx: fix cd regression for dt platform Current card detect probe process is that when driver finds a valid ESDHC_CD_GPIO, it will clear the quirk SDHCI_QUIRK_BROKEN_CARD_DETECTION which is set by default for all esdhc/usdhc controllers. Then host driver will know there's a valid card detect function. Commit 8d86e4fcccf6 ("mmc: sdhci-esdhc-imx: Call mmc_of_parse()") breaks GPIO CD function for dt platform that it will return directly when find ESDHC_CD_GPIO for dt platform which result in the later wrongly to keep SDHCI_QUIRK_BROKEN_CARD_DETECTION for all dt platforms. Then MMC_CAP_NEEDS_POLL will be used instead even there's a valid GPIO card detect. This patch adds back this function and follows the original approach to clear the quirk if find an valid CD GPIO for dt platforms. Fixes: 8d86e4fcccf6 ("mmc: sdhci-esdhc-imx: Call mmc_of_parse()") Signed-off-by: Dong Aisheng Reviewed-by: Johan Derycke Signed-off-by: Ulf Hansson diff --git a/drivers/mmc/host/sdhci-esdhc-imx.c b/drivers/mmc/host/sdhci-esdhc-imx.c index faf0cb9..4815391 100644 --- a/drivers/mmc/host/sdhci-esdhc-imx.c +++ b/drivers/mmc/host/sdhci-esdhc-imx.c @@ -881,6 +881,7 @@ sdhci_esdhc_imx_probe_dt(struct platform_device *pdev, struct esdhc_platform_data *boarddata) { struct device_node *np = pdev->dev.of_node; + int ret; if (!np) return -ENODEV; @@ -917,7 +918,14 @@ sdhci_esdhc_imx_probe_dt(struct platform_device *pdev, mmc_of_parse_voltage(np, &host->ocr_mask); /* call to generic mmc_of_parse to support additional capabilities */ - return mmc_of_parse(host->mmc); + ret = mmc_of_parse(host->mmc); + if (ret) + return ret; + + if (!IS_ERR_VALUE(mmc_gpio_get_cd(host->mmc))) + host->quirks &= ~SDHCI_QUIRK_BROKEN_CARD_DETECTION; + + return 0; } #else static inline int -- cgit v0.10.2 From 91fa4252878afccc9e75edd84f31047899ddd1b7 Mon Sep 17 00:00:00 2001 From: Dong Aisheng Date: Wed, 22 Jul 2015 20:53:06 +0800 Subject: mmc: sdhci-esdhc-imx: move all non dt probe code into one function This is an incremental fix of commit e62bd351b("mmc: sdhci-esdhc-imx: Do not break platform data boards"). After commit 8d86e4fcccf6 ("mmc: sdhci-esdhc-imx: Call mmc_of_parse()"), we do not need to run the check of boarddata->wp_type/cd_type/max_bus_width again for dt platform since those are already handled by mmc_of_parse(). Current code only exclude the checking of wp_type for dt platform which does not make sense. This patch moves all non dt probe code into one function. Besides, since we only support SD3.0/eMMC HS200 for dt platform, the support_vsel checking and ultra high speed pinctrl state are also merged into sdhci_esdhc_imx_probe_dt. Then we have two separately probe function for dt and non dt type. This can make the driver probe more clearly. Signed-off-by: Dong Aisheng Reviewed-by: Johan Derycke Signed-off-by: Ulf Hansson diff --git a/drivers/mmc/host/sdhci-esdhc-imx.c b/drivers/mmc/host/sdhci-esdhc-imx.c index 4815391..10f03ee 100644 --- a/drivers/mmc/host/sdhci-esdhc-imx.c +++ b/drivers/mmc/host/sdhci-esdhc-imx.c @@ -878,14 +878,12 @@ static const struct sdhci_pltfm_data sdhci_esdhc_imx_pdata = { static int sdhci_esdhc_imx_probe_dt(struct platform_device *pdev, struct sdhci_host *host, - struct esdhc_platform_data *boarddata) + struct pltfm_imx_data *imx_data) { struct device_node *np = pdev->dev.of_node; + struct esdhc_platform_data *boarddata = &imx_data->boarddata; int ret; - if (!np) - return -ENODEV; - if (of_get_property(np, "non-removable", NULL)) boarddata->cd_type = ESDHC_CD_PERMANENT; @@ -917,6 +915,26 @@ sdhci_esdhc_imx_probe_dt(struct platform_device *pdev, mmc_of_parse_voltage(np, &host->ocr_mask); + /* sdr50 and sdr104 needs work on 1.8v signal voltage */ + if ((boarddata->support_vsel) && esdhc_is_usdhc(imx_data) && + !IS_ERR(imx_data->pins_default)) { + imx_data->pins_100mhz = pinctrl_lookup_state(imx_data->pinctrl, + ESDHC_PINCTRL_STATE_100MHZ); + imx_data->pins_200mhz = pinctrl_lookup_state(imx_data->pinctrl, + ESDHC_PINCTRL_STATE_200MHZ); + if (IS_ERR(imx_data->pins_100mhz) || + IS_ERR(imx_data->pins_200mhz)) { + dev_warn(mmc_dev(host->mmc), + "could not get ultra high speed state, work on normal mode\n"); + /* + * fall back to not support uhs by specify no 1.8v quirk + */ + host->quirks2 |= SDHCI_QUIRK2_NO_1_8_V; + } + } else { + host->quirks2 |= SDHCI_QUIRK2_NO_1_8_V; + } + /* call to generic mmc_of_parse to support additional capabilities */ ret = mmc_of_parse(host->mmc); if (ret) @@ -931,22 +949,85 @@ sdhci_esdhc_imx_probe_dt(struct platform_device *pdev, static inline int sdhci_esdhc_imx_probe_dt(struct platform_device *pdev, struct sdhci_host *host, - struct esdhc_platform_data *boarddata) + struct pltfm_imx_data *imx_data) { return -ENODEV; } #endif +static int sdhci_esdhc_imx_probe_nondt(struct platform_device *pdev, + struct sdhci_host *host, + struct pltfm_imx_data *imx_data) +{ + struct esdhc_platform_data *boarddata = &imx_data->boarddata; + int err; + + if (!host->mmc->parent->platform_data) { + dev_err(mmc_dev(host->mmc), "no board data!\n"); + return -EINVAL; + } + + imx_data->boarddata = *((struct esdhc_platform_data *) + host->mmc->parent->platform_data); + /* write_protect */ + if (boarddata->wp_type == ESDHC_WP_GPIO) { + err = mmc_gpio_request_ro(host->mmc, boarddata->wp_gpio); + if (err) { + dev_err(mmc_dev(host->mmc), + "failed to request write-protect gpio!\n"); + return err; + } + host->mmc->caps2 |= MMC_CAP2_RO_ACTIVE_HIGH; + } + + /* card_detect */ + switch (boarddata->cd_type) { + case ESDHC_CD_GPIO: + err = mmc_gpio_request_cd(host->mmc, boarddata->cd_gpio, 0); + if (err) { + dev_err(mmc_dev(host->mmc), + "failed to request card-detect gpio!\n"); + return err; + } + /* fall through */ + + case ESDHC_CD_CONTROLLER: + /* we have a working card_detect back */ + host->quirks &= ~SDHCI_QUIRK_BROKEN_CARD_DETECTION; + break; + + case ESDHC_CD_PERMANENT: + host->mmc->caps |= MMC_CAP_NONREMOVABLE; + break; + + case ESDHC_CD_NONE: + break; + } + + switch (boarddata->max_bus_width) { + case 8: + host->mmc->caps |= MMC_CAP_8_BIT_DATA | MMC_CAP_4_BIT_DATA; + break; + case 4: + host->mmc->caps |= MMC_CAP_4_BIT_DATA; + break; + case 1: + default: + host->quirks |= SDHCI_QUIRK_FORCE_1_BIT_DATA; + break; + } + + return 0; +} + static int sdhci_esdhc_imx_probe(struct platform_device *pdev) { const struct of_device_id *of_id = of_match_device(imx_esdhc_dt_ids, &pdev->dev); struct sdhci_pltfm_host *pltfm_host; struct sdhci_host *host; - struct esdhc_platform_data *boarddata; int err; struct pltfm_imx_data *imx_data; - bool dt = true; host = sdhci_pltfm_init(pdev, &sdhci_esdhc_imx_pdata, 0); if (IS_ERR(host)) @@ -1038,84 +1119,12 @@ static int sdhci_esdhc_imx_probe(struct platform_device *pdev) if (imx_data->socdata->flags & ESDHC_FLAG_ERR004536) host->quirks |= SDHCI_QUIRK_BROKEN_ADMA; - boarddata = &imx_data->boarddata; - if (sdhci_esdhc_imx_probe_dt(pdev, host, boarddata) < 0) { - if (!host->mmc->parent->platform_data) { - dev_err(mmc_dev(host->mmc), "no board data!\n"); - err = -EINVAL; - goto disable_clk; - } - imx_data->boarddata = *((struct esdhc_platform_data *) - host->mmc->parent->platform_data); - dt = false; - } - /* write_protect */ - if (boarddata->wp_type == ESDHC_WP_GPIO && !dt) { - err = mmc_gpio_request_ro(host->mmc, boarddata->wp_gpio); - if (err) { - dev_err(mmc_dev(host->mmc), - "failed to request write-protect gpio!\n"); - goto disable_clk; - } - host->mmc->caps2 |= MMC_CAP2_RO_ACTIVE_HIGH; - } - - /* card_detect */ - switch (boarddata->cd_type) { - case ESDHC_CD_GPIO: - if (dt) - break; - err = mmc_gpio_request_cd(host->mmc, boarddata->cd_gpio, 0); - if (err) { - dev_err(mmc_dev(host->mmc), - "failed to request card-detect gpio!\n"); - goto disable_clk; - } - /* fall through */ - - case ESDHC_CD_CONTROLLER: - /* we have a working card_detect back */ - host->quirks &= ~SDHCI_QUIRK_BROKEN_CARD_DETECTION; - break; - - case ESDHC_CD_PERMANENT: - host->mmc->caps |= MMC_CAP_NONREMOVABLE; - break; - - case ESDHC_CD_NONE: - break; - } - - switch (boarddata->max_bus_width) { - case 8: - host->mmc->caps |= MMC_CAP_8_BIT_DATA | MMC_CAP_4_BIT_DATA; - break; - case 4: - host->mmc->caps |= MMC_CAP_4_BIT_DATA; - break; - case 1: - default: - host->quirks |= SDHCI_QUIRK_FORCE_1_BIT_DATA; - break; - } - - /* sdr50 and sdr104 needs work on 1.8v signal voltage */ - if ((boarddata->support_vsel) && esdhc_is_usdhc(imx_data) && - !IS_ERR(imx_data->pins_default)) { - imx_data->pins_100mhz = pinctrl_lookup_state(imx_data->pinctrl, - ESDHC_PINCTRL_STATE_100MHZ); - imx_data->pins_200mhz = pinctrl_lookup_state(imx_data->pinctrl, - ESDHC_PINCTRL_STATE_200MHZ); - if (IS_ERR(imx_data->pins_100mhz) || - IS_ERR(imx_data->pins_200mhz)) { - dev_warn(mmc_dev(host->mmc), - "could not get ultra high speed state, work on normal mode\n"); - /* fall back to not support uhs by specify no 1.8v quirk */ - host->quirks2 |= SDHCI_QUIRK2_NO_1_8_V; - } - } else { - host->quirks2 |= SDHCI_QUIRK2_NO_1_8_V; - } + if (of_id) + err = sdhci_esdhc_imx_probe_dt(pdev, host, imx_data); + else + err = sdhci_esdhc_imx_probe_nondt(pdev, host, imx_data); + if (err) + goto disable_clk; err = sdhci_add_host(host); if (err) -- cgit v0.10.2 From 5924175755a0ed902d91f2f2660e914032fa63e5 Mon Sep 17 00:00:00 2001 From: Dong Aisheng Date: Wed, 22 Jul 2015 20:53:07 +0800 Subject: mmc: sdhci: make max-frequency property in device tree work Device tree provides option to specify the max freqency with property "max-frequency" in dts and common parse function mmc_of_parse() will parse it and use this value to set host->f_max to tell the MMC core the maxinum frequency the host works. However, current sdhci driver will finally overwrite this value with host->max_clk regardless of the max-frequency property. This patch makes sure not overwrite the max-frequency set from device tree and do basic sanity check. Signed-off-by: Dong Aisheng Reviewed-by: Johan Derycke Signed-off-by: Ulf Hansson diff --git a/drivers/mmc/host/sdhci.c b/drivers/mmc/host/sdhci.c index 655900f..1dbe932 100644 --- a/drivers/mmc/host/sdhci.c +++ b/drivers/mmc/host/sdhci.c @@ -2866,6 +2866,7 @@ int sdhci_add_host(struct sdhci_host *host) u32 max_current_caps; unsigned int ocr_avail; unsigned int override_timeout_clk; + u32 max_clk; int ret; WARN_ON(host == NULL); @@ -3050,18 +3051,22 @@ int sdhci_add_host(struct sdhci_host *host) * Set host parameters. */ mmc->ops = &sdhci_ops; - mmc->f_max = host->max_clk; + max_clk = host->max_clk; + if (host->ops->get_min_clock) mmc->f_min = host->ops->get_min_clock(host); else if (host->version >= SDHCI_SPEC_300) { if (host->clk_mul) { mmc->f_min = (host->max_clk * host->clk_mul) / 1024; - mmc->f_max = host->max_clk * host->clk_mul; + max_clk = host->max_clk * host->clk_mul; } else mmc->f_min = host->max_clk / SDHCI_MAX_DIV_SPEC_300; } else mmc->f_min = host->max_clk / SDHCI_MAX_DIV_SPEC_200; + if (!mmc->f_max || (mmc->f_max && (mmc->f_max > max_clk))) + mmc->f_max = max_clk; + if (!(host->quirks & SDHCI_QUIRK_DATA_TIMEOUT_USES_SDCLK)) { host->timeout_clk = (caps[0] & SDHCI_TIMEOUT_CLK_MASK) >> SDHCI_TIMEOUT_CLK_SHIFT; -- cgit v0.10.2 From 2e05d66b6b6cbe6018c8dcc15a379d913efdea83 Mon Sep 17 00:00:00 2001 From: Dong Aisheng Date: Wed, 22 Jul 2015 20:53:08 +0800 Subject: mmc: sdhci-esdhc-imx: remove duplicated dts parsing After commit 8d86e4fcccf6 ("mmc: sdhci-esdhc-imx: Call mmc_of_parse()"), we do not need those duplicated parsing anymore. Note: fsl,cd-controller is also deleted due to the driver does not support controller card detection anymore after switch to runtime pm. And there's no user of it right now in device tree. wp-gpios is kept because we're still support fsl,wp-controller, so we need a way to check if it's gpio wp or controller wp. Signed-off-by: Dong Aisheng Reviewed-by: Johan Derycke Signed-off-by: Ulf Hansson diff --git a/drivers/mmc/host/sdhci-esdhc-imx.c b/drivers/mmc/host/sdhci-esdhc-imx.c index 10f03ee..1b0e618 100644 --- a/drivers/mmc/host/sdhci-esdhc-imx.c +++ b/drivers/mmc/host/sdhci-esdhc-imx.c @@ -884,27 +884,13 @@ sdhci_esdhc_imx_probe_dt(struct platform_device *pdev, struct esdhc_platform_data *boarddata = &imx_data->boarddata; int ret; - if (of_get_property(np, "non-removable", NULL)) - boarddata->cd_type = ESDHC_CD_PERMANENT; - - if (of_get_property(np, "fsl,cd-controller", NULL)) - boarddata->cd_type = ESDHC_CD_CONTROLLER; - if (of_get_property(np, "fsl,wp-controller", NULL)) boarddata->wp_type = ESDHC_WP_CONTROLLER; - boarddata->cd_gpio = of_get_named_gpio(np, "cd-gpios", 0); - if (gpio_is_valid(boarddata->cd_gpio)) - boarddata->cd_type = ESDHC_CD_GPIO; - boarddata->wp_gpio = of_get_named_gpio(np, "wp-gpios", 0); if (gpio_is_valid(boarddata->wp_gpio)) boarddata->wp_type = ESDHC_WP_GPIO; - of_property_read_u32(np, "bus-width", &boarddata->max_bus_width); - - of_property_read_u32(np, "max-frequency", &boarddata->f_max); - if (of_find_property(np, "no-1-8-v", NULL)) boarddata->support_vsel = false; else -- cgit v0.10.2 From a3bd4f989f532694337dd30538b635d5213ab86a Mon Sep 17 00:00:00 2001 From: Dong Aisheng Date: Wed, 22 Jul 2015 20:53:09 +0800 Subject: mmc: sdhci-esdhc-imx: clear f_max in boarddata After commit 8d86e4fcccf6 ("mmc: sdhci-esdhc-imx: Call mmc_of_parse()"), it's not used anymore. Signed-off-by: Dong Aisheng Reviewed-by: Johan Derycke Signed-off-by: Ulf Hansson diff --git a/drivers/mmc/host/sdhci-esdhc-imx.c b/drivers/mmc/host/sdhci-esdhc-imx.c index 1b0e618..c6b9f64 100644 --- a/drivers/mmc/host/sdhci-esdhc-imx.c +++ b/drivers/mmc/host/sdhci-esdhc-imx.c @@ -581,13 +581,8 @@ static void esdhc_writeb_le(struct sdhci_host *host, u8 val, int reg) static unsigned int esdhc_pltfm_get_max_clock(struct sdhci_host *host) { struct sdhci_pltfm_host *pltfm_host = sdhci_priv(host); - struct pltfm_imx_data *imx_data = pltfm_host->priv; - struct esdhc_platform_data *boarddata = &imx_data->boarddata; - if (boarddata->f_max && (boarddata->f_max < pltfm_host->clock)) - return boarddata->f_max; - else - return pltfm_host->clock; + return pltfm_host->clock; } static unsigned int esdhc_pltfm_get_min_clock(struct sdhci_host *host) diff --git a/include/linux/platform_data/mmc-esdhc-imx.h b/include/linux/platform_data/mmc-esdhc-imx.h index 75f70f6..e1571ef 100644 --- a/include/linux/platform_data/mmc-esdhc-imx.h +++ b/include/linux/platform_data/mmc-esdhc-imx.h @@ -43,7 +43,6 @@ struct esdhc_platform_data { enum wp_types wp_type; enum cd_types cd_type; int max_bus_width; - unsigned int f_max; bool support_vsel; unsigned int delay_line; }; -- cgit v0.10.2 From cae13e792f8f863a60cee2cfdbe704ea84cf2a07 Mon Sep 17 00:00:00 2001 From: Dong Aisheng Date: Wed, 22 Jul 2015 20:53:10 +0800 Subject: dts: mmc: fsl-imx-esdhc: remove fsl,cd-controller support It's not supported by driver anymore after using runtime pm and there's no user of it, so delete it now. Signed-off-by: Dong Aisheng Reviewed-by: Johan Derycke Signed-off-by: Ulf Hansson diff --git a/Documentation/devicetree/bindings/mmc/fsl-imx-esdhc.txt b/Documentation/devicetree/bindings/mmc/fsl-imx-esdhc.txt index 5d0376b..211e778 100644 --- a/Documentation/devicetree/bindings/mmc/fsl-imx-esdhc.txt +++ b/Documentation/devicetree/bindings/mmc/fsl-imx-esdhc.txt @@ -17,7 +17,6 @@ Required properties: "fsl,imx6sx-usdhc" Optional properties: -- fsl,cd-controller : Indicate to use controller internal card detection - fsl,wp-controller : Indicate to use controller internal write protection - fsl,delay-line : Specify the number of delay cells for override mode. This is used to set the clock delay for DLL(Delay Line) on override mode @@ -35,7 +34,6 @@ esdhc@70004000 { compatible = "fsl,imx51-esdhc"; reg = <0x70004000 0x4000>; interrupts = <1>; - fsl,cd-controller; fsl,wp-controller; }; -- cgit v0.10.2 From 9cd76049f0d90ae241f5ad80e311489824527000 Mon Sep 17 00:00:00 2001 From: Jingju Hou Date: Thu, 23 Jul 2015 17:56:23 +0800 Subject: mmc: sdhci-pxav3: fix platform_data is not initialized pdev->dev.platform_data is not initialized if match is true in function sdhci_pxav3_probe. Just local variable pdata is assigned the return value from function pxav3_get_mmc_pdata(). static int sdhci_pxav3_probe(struct platform_device *pdev) { struct sdhci_pxa_platdata *pdata = pdev->dev.platform_data; ... if (match) { ret = mmc_of_parse(host->mmc); if (ret) goto err_of_parse; sdhci_get_of_property(pdev); pdata = pxav3_get_mmc_pdata(dev); } ... } Signed-off-by: Jingju Hou Fixes: b650352dd3df("mmc: sdhci-pxa: Add device tree support") Signed-off-by: Ulf Hansson diff --git a/drivers/mmc/host/sdhci-pxav3.c b/drivers/mmc/host/sdhci-pxav3.c index 9cd5fc6..946d37f 100644 --- a/drivers/mmc/host/sdhci-pxav3.c +++ b/drivers/mmc/host/sdhci-pxav3.c @@ -411,6 +411,7 @@ static int sdhci_pxav3_probe(struct platform_device *pdev) goto err_of_parse; sdhci_get_of_property(pdev); pdata = pxav3_get_mmc_pdata(dev); + pdev->dev.platform_data = pdata; } else if (pdata) { /* on-chip device */ if (pdata->flags & PXA_FLAG_CARD_PERMANENT) -- cgit v0.10.2 From 8a0a5da6d9cbf1b142115ff6e6b253a82633c3d9 Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Fri, 24 Jul 2015 16:13:43 +0200 Subject: x86/mm: Fix newly introduced printk format warnings Signed-off-by: Thomas Gleixner diff --git a/arch/x86/mm/ioremap.c b/arch/x86/mm/ioremap.c index b9d4a33..b9c78f3 100644 --- a/arch/x86/mm/ioremap.c +++ b/arch/x86/mm/ioremap.c @@ -118,8 +118,8 @@ static void __iomem *__ioremap_caller(resource_size_t phys_addr, last_pfn = last_addr >> PAGE_SHIFT; if (walk_system_ram_range(pfn, last_pfn - pfn + 1, NULL, __ioremap_check_ram) == 1) { - WARN_ONCE(1, "ioremap on RAM at 0x%llx - 0x%llx\n", - phys_addr, last_addr); + WARN_ONCE(1, "ioremap on RAM at %pa - %pa\n", + &phys_addr, &last_addr); return NULL; } -- cgit v0.10.2 From c0c3322e98029e752526d906d9e3a680ed213c03 Mon Sep 17 00:00:00 2001 From: Denys Vlasenko Date: Fri, 24 Jul 2015 14:16:43 +0200 Subject: x86/asm/entry/32: Revert 'Do not use R9 in SYSCALL32' commit This change reverts most of commit 53e9accf0f 'Do not use R9 in SYSCALL32'. I don't yet understand how, but code in that commit sometimes fails to preserve EBP. See https://bugzilla.kernel.org/show_bug.cgi?id=101061 "Problems while executing 32-bit code on AMD64" Reported-and-tested-by: Krzysztof A. Sobiecki Signed-off-by: Denys Vlasenko Cc: Linus Torvalds Cc: Steven Rostedt Cc: Borislav Petkov Cc: Andy Lutomirski Cc: Oleg Nesterov Cc: Frederic Weisbecker Cc: Alexei Starovoitov Cc: Will Drewry Cc: Kees Cook CC: x86@kernel.org Link: http://lkml.kernel.org/r/1437740203-11552-1-git-send-email-dvlasenk@redhat.com Signed-off-by: Thomas Gleixner diff --git a/arch/x86/entry/entry_64_compat.S b/arch/x86/entry/entry_64_compat.S index bb187a6..5a18447 100644 --- a/arch/x86/entry/entry_64_compat.S +++ b/arch/x86/entry/entry_64_compat.S @@ -205,7 +205,6 @@ sysexit_from_sys_call: movl RDX(%rsp), %edx /* arg3 */ movl RSI(%rsp), %ecx /* arg4 */ movl RDI(%rsp), %r8d /* arg5 */ - movl %ebp, %r9d /* arg6 */ .endm .macro auditsys_exit exit @@ -236,6 +235,7 @@ sysexit_from_sys_call: sysenter_auditsys: auditsys_entry_common + movl %ebp, %r9d /* reload 6th syscall arg */ jmp sysenter_dispatch sysexit_audit: @@ -336,7 +336,7 @@ ENTRY(entry_SYSCALL_compat) * 32-bit zero extended: */ ASM_STAC -1: movl (%r8), %ebp +1: movl (%r8), %r9d _ASM_EXTABLE(1b, ia32_badarg) ASM_CLAC orl $TS_COMPAT, ASM_THREAD_INFO(TI_status, %rsp, SIZEOF_PTREGS) @@ -346,7 +346,7 @@ ENTRY(entry_SYSCALL_compat) cstar_do_call: /* 32-bit syscall -> 64-bit C ABI argument conversion */ movl %edi, %r8d /* arg5 */ - movl %ebp, %r9d /* arg6 */ + /* r9 already loaded */ /* arg6 */ xchg %ecx, %esi /* rsi:arg2, rcx:arg4 */ movl %ebx, %edi /* arg1 */ movl %edx, %edx /* arg3 (zero extension) */ @@ -358,7 +358,6 @@ cstar_dispatch: call *ia32_sys_call_table(, %rax, 8) movq %rax, RAX(%rsp) 1: - movl RCX(%rsp), %ebp DISABLE_INTERRUPTS(CLBR_NONE) TRACE_IRQS_OFF testl $_TIF_ALLWORK_MASK, ASM_THREAD_INFO(TI_flags, %rsp, SIZEOF_PTREGS) @@ -392,7 +391,9 @@ sysretl_from_sys_call: #ifdef CONFIG_AUDITSYSCALL cstar_auditsys: + movl %r9d, R9(%rsp) /* register to be clobbered by call */ auditsys_entry_common + movl R9(%rsp), %r9d /* reload 6th syscall arg */ jmp cstar_dispatch sysretl_audit: @@ -404,14 +405,16 @@ cstar_tracesys: testl $(_TIF_WORK_SYSCALL_ENTRY & ~_TIF_SYSCALL_AUDIT), ASM_THREAD_INFO(TI_flags, %rsp, SIZEOF_PTREGS) jz cstar_auditsys #endif + xchgl %r9d, %ebp SAVE_EXTRA_REGS xorl %eax, %eax /* Do not leak kernel information */ movq %rax, R11(%rsp) movq %rax, R10(%rsp) - movq %rax, R9(%rsp) + movq %r9, R9(%rsp) movq %rax, R8(%rsp) movq %rsp, %rdi /* &pt_regs -> arg1 */ call syscall_trace_enter + movl R9(%rsp), %r9d /* Reload arg registers from stack. (see sysenter_tracesys) */ movl RCX(%rsp), %ecx @@ -421,6 +424,7 @@ cstar_tracesys: movl %eax, %eax /* zero extension */ RESTORE_EXTRA_REGS + xchgl %ebp, %r9d jmp cstar_do_call END(entry_SYSCALL_compat) -- cgit v0.10.2 From f6ee9b582d2db652497b73c1f117591dfb6d3a90 Mon Sep 17 00:00:00 2001 From: Sascha Hauer Date: Fri, 24 Jul 2015 15:01:08 +0200 Subject: spi: imx: Fix small DMA transfers DMA transfers must be greater than the watermark level size. spi_imx->rx_wml and spi_imx->tx_wml contain the watermark level in 32bit words whereas struct spi_transfer contains the transfer len in bytes. Fix the check if DMA is possible for a transfer accordingly. This fixes transfers with sizes between 33 and 128 bytes for which previously was claimed that DMA is possible. Fixes: f62caccd12c17e4 (spi: spi-imx: add DMA support) Signed-off-by: Sascha Hauer Cc: stable@vger.kernel.org Signed-off-by: Mark Brown diff --git a/drivers/spi/spi-imx.c b/drivers/spi/spi-imx.c index eb7d3a6..f9deb84 100644 --- a/drivers/spi/spi-imx.c +++ b/drivers/spi/spi-imx.c @@ -201,8 +201,9 @@ static bool spi_imx_can_dma(struct spi_master *master, struct spi_device *spi, { struct spi_imx_data *spi_imx = spi_master_get_devdata(master); - if (spi_imx->dma_is_inited && (transfer->len > spi_imx->rx_wml) - && (transfer->len > spi_imx->tx_wml)) + if (spi_imx->dma_is_inited + && transfer->len > spi_imx->rx_wml * sizeof(u32) + && transfer->len > spi_imx->tx_wml * sizeof(u32)) return true; return false; } -- cgit v0.10.2 From 49fa63d8c2f95b636cc31a7164aeac0309dfa3c9 Mon Sep 17 00:00:00 2001 From: Steve Wise Date: Wed, 22 Jul 2015 14:14:17 -0500 Subject: RDMA/cxgb3: fail get_dma_mr on 64 bit arches T3 HW only supports 32 bit MRs. If the system uses 64 bit memory addresses, then a registered 32 bit MR will wrap and write to the wrong memory when used with addresses > 4GB. To prevent this, simply fail to allocate an MR on 64 bit machines (other means of registering memory are still available and software can still work, we just don't allow this means of memory registration). Signed-off-by: Steve Wise Signed-off-by: Doug Ledford diff --git a/drivers/infiniband/hw/cxgb3/iwch_provider.c b/drivers/infiniband/hw/cxgb3/iwch_provider.c index b1b7323..bbbe018 100644 --- a/drivers/infiniband/hw/cxgb3/iwch_provider.c +++ b/drivers/infiniband/hw/cxgb3/iwch_provider.c @@ -736,6 +736,10 @@ static struct ib_mr *iwch_get_dma_mr(struct ib_pd *pd, int acc) /* * T3 only supports 32 bits of size. */ + if (sizeof(phys_addr_t) > 4) { + pr_warn_once(MOD "Cannot support dma_mrs on this platform.\n"); + return ERR_PTR(-ENOTSUPP); + } bl.size = 0xffffffff; bl.addr = 0; kva = 0; -- cgit v0.10.2 From efc1eedbf63a194b3b576fc25776f3f1fa55a4d4 Mon Sep 17 00:00:00 2001 From: Jason Gunthorpe Date: Wed, 22 Jul 2015 14:30:03 -0600 Subject: IB/ipoib: Fix CONFIG_INFINIBAND_IPOIB_CM If the above is turned off then ipoib_cm_dev_init unconditionally returns ENOSYS, and the newly added error handling in 0b3957 prevents ipoib from coming up at all: kernel: mlx4_0: ipoib_transport_dev_init failed kernel: mlx4_0: failed to initialize port 1 (ret = -12) Fixes: 0b39578bcde4 (IB/ipoib: Use dedicated workqueues per interface) Signed-off-by: Jason Gunthorpe Signed-off-by: Doug Ledford diff --git a/drivers/infiniband/ulp/ipoib/ipoib_verbs.c b/drivers/infiniband/ulp/ipoib/ipoib_verbs.c index 9e6ee82..851c821 100644 --- a/drivers/infiniband/ulp/ipoib/ipoib_verbs.c +++ b/drivers/infiniband/ulp/ipoib/ipoib_verbs.c @@ -177,7 +177,8 @@ int ipoib_transport_dev_init(struct net_device *dev, struct ib_device *ca) else size += ipoib_recvq_size * ipoib_max_conn_qp; } else - goto out_free_wq; + if (ret != -ENOSYS) + goto out_free_wq; cq_attr.cqe = size; priv->recv_cq = ib_create_cq(priv->ca, ipoib_ib_completion, NULL, -- cgit v0.10.2 From 71ee67306ecbdfc0c94ed93c77ff99d29e961d69 Mon Sep 17 00:00:00 2001 From: Devesh Sharma Date: Fri, 24 Jul 2015 05:03:59 +0530 Subject: RDMA/ocrdma: update ocrdma license to dual-license Change of license from GPLv2 to dual-license (GPLv2 and BSD 2-Clause) All contributors were contacted off-list and permission to make this change was received. The complete list of contributors are Cc:ed here. Cc: Tejun Heo Cc: Duan Jiong Cc: Roland Dreier Cc: Jes Sorensen Cc: Sasha Levin Cc: Dan Carpenter Cc: Prarit Bhargava Cc: Colin Ian King Cc: Wei Yongjun Cc: Moni Shoua Cc: Rasmus Villemoes Cc: Li RongQing Cc: Devendra Naga Signed-off-by: Devesh Sharma Signed-off-by: Doug Ledford diff --git a/drivers/infiniband/hw/ocrdma/ocrdma.h b/drivers/infiniband/hw/ocrdma/ocrdma.h index b396344..6a36338 100644 --- a/drivers/infiniband/hw/ocrdma/ocrdma.h +++ b/drivers/infiniband/hw/ocrdma/ocrdma.h @@ -1,21 +1,36 @@ -/******************************************************************* - * This file is part of the Emulex RoCE Device Driver for * - * RoCE (RDMA over Converged Ethernet) adapters. * - * Copyright (C) 2008-2012 Emulex. All rights reserved. * - * EMULEX and SLI are trademarks of Emulex. * - * www.emulex.com * - * * - * This program is free software; you can redistribute it and/or * - * modify it under the terms of version 2 of the GNU General * - * Public License as published by the Free Software Foundation. * - * This program is distributed in the hope that it will be useful. * - * ALL EXPRESS OR IMPLIED CONDITIONS, REPRESENTATIONS AND * - * WARRANTIES, INCLUDING ANY IMPLIED WARRANTY OF MERCHANTABILITY, * - * FITNESS FOR A PARTICULAR PURPOSE, OR NON-INFRINGEMENT, ARE * - * DISCLAIMED, EXCEPT TO THE EXTENT THAT SUCH DISCLAIMERS ARE HELD * - * TO BE LEGALLY INVALID. See the GNU General Public License for * - * more details, a copy of which can be found in the file COPYING * - * included with this package. * +/* This file is part of the Emulex RoCE Device Driver for + * RoCE (RDMA over Converged Ethernet) adapters. + * Copyright (C) 2012-2015 Emulex. All rights reserved. + * EMULEX and SLI are trademarks of Emulex. + * www.emulex.com + * + * This software is available to you under a choice of one of two licenses. + * You may choose to be licensed under the terms of the GNU General Public + * License (GPL) Version 2, available from the file COPYING in the main + * directory of this source tree, or the BSD license below: + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions + * are met: + * + * - Redistributions of source code must retain the above copyright notice, + * this list of conditions and the following disclaimer. + * + * - Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in + * the documentation and/or other materials provided with the distribution. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" + * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO,THE + * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE + * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE + * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR + * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF + * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR + * BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, + * WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR + * OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF + * ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. * * Contact Information: * linux-drivers@emulex.com @@ -23,7 +38,7 @@ * Emulex * 3333 Susan Street * Costa Mesa, CA 92626 - *******************************************************************/ + */ #ifndef __OCRDMA_H__ #define __OCRDMA_H__ diff --git a/drivers/infiniband/hw/ocrdma/ocrdma_abi.h b/drivers/infiniband/hw/ocrdma/ocrdma_abi.h index 1554cca..430b135 100644 --- a/drivers/infiniband/hw/ocrdma/ocrdma_abi.h +++ b/drivers/infiniband/hw/ocrdma/ocrdma_abi.h @@ -1,21 +1,36 @@ -/******************************************************************* - * This file is part of the Emulex RoCE Device Driver for * - * RoCE (RDMA over Converged Ethernet) adapters. * - * Copyright (C) 2008-2012 Emulex. All rights reserved. * - * EMULEX and SLI are trademarks of Emulex. * - * www.emulex.com * - * * - * This program is free software; you can redistribute it and/or * - * modify it under the terms of version 2 of the GNU General * - * Public License as published by the Free Software Foundation. * - * This program is distributed in the hope that it will be useful. * - * ALL EXPRESS OR IMPLIED CONDITIONS, REPRESENTATIONS AND * - * WARRANTIES, INCLUDING ANY IMPLIED WARRANTY OF MERCHANTABILITY, * - * FITNESS FOR A PARTICULAR PURPOSE, OR NON-INFRINGEMENT, ARE * - * DISCLAIMED, EXCEPT TO THE EXTENT THAT SUCH DISCLAIMERS ARE HELD * - * TO BE LEGALLY INVALID. See the GNU General Public License for * - * more details, a copy of which can be found in the file COPYING * - * included with this package. * +/* This file is part of the Emulex RoCE Device Driver for + * RoCE (RDMA over Converged Ethernet) adapters. + * Copyright (C) 2012-2015 Emulex. All rights reserved. + * EMULEX and SLI are trademarks of Emulex. + * www.emulex.com + * + * This software is available to you under a choice of one of two licenses. + * You may choose to be licensed under the terms of the GNU General Public + * License (GPL) Version 2, available from the file COPYING in the main + * directory of this source tree, or the BSD license below: + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions + * are met: + * + * - Redistributions of source code must retain the above copyright notice, + * this list of conditions and the following disclaimer. + * + * - Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in + * the documentation and/or other materials provided with the distribution. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" + * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO,THE + * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE + * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE + * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR + * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF + * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR + * BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, + * WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR + * OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF + * ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. * * Contact Information: * linux-drivers@emulex.com @@ -23,7 +38,7 @@ * Emulex * 3333 Susan Street * Costa Mesa, CA 92626 - *******************************************************************/ + */ #ifndef __OCRDMA_ABI_H__ #define __OCRDMA_ABI_H__ diff --git a/drivers/infiniband/hw/ocrdma/ocrdma_ah.c b/drivers/infiniband/hw/ocrdma/ocrdma_ah.c index 29b2767..44766fe 100644 --- a/drivers/infiniband/hw/ocrdma/ocrdma_ah.c +++ b/drivers/infiniband/hw/ocrdma/ocrdma_ah.c @@ -1,21 +1,36 @@ -/******************************************************************* - * This file is part of the Emulex RoCE Device Driver for * - * RoCE (RDMA over Converged Ethernet) adapters. * - * Copyright (C) 2008-2012 Emulex. All rights reserved. * - * EMULEX and SLI are trademarks of Emulex. * - * www.emulex.com * - * * - * This program is free software; you can redistribute it and/or * - * modify it under the terms of version 2 of the GNU General * - * Public License as published by the Free Software Foundation. * - * This program is distributed in the hope that it will be useful. * - * ALL EXPRESS OR IMPLIED CONDITIONS, REPRESENTATIONS AND * - * WARRANTIES, INCLUDING ANY IMPLIED WARRANTY OF MERCHANTABILITY, * - * FITNESS FOR A PARTICULAR PURPOSE, OR NON-INFRINGEMENT, ARE * - * DISCLAIMED, EXCEPT TO THE EXTENT THAT SUCH DISCLAIMERS ARE HELD * - * TO BE LEGALLY INVALID. See the GNU General Public License for * - * more details, a copy of which can be found in the file COPYING * - * included with this package. * +/* This file is part of the Emulex RoCE Device Driver for + * RoCE (RDMA over Converged Ethernet) adapters. + * Copyright (C) 2012-2015 Emulex. All rights reserved. + * EMULEX and SLI are trademarks of Emulex. + * www.emulex.com + * + * This software is available to you under a choice of one of two licenses. + * You may choose to be licensed under the terms of the GNU General Public + * License (GPL) Version 2, available from the file COPYING in the main + * directory of this source tree, or the BSD license below: + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions + * are met: + * + * - Redistributions of source code must retain the above copyright notice, + * this list of conditions and the following disclaimer. + * + * - Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in + * the documentation and/or other materials provided with the distribution. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" + * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO,THE + * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE + * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE + * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR + * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF + * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR + * BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, + * WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR + * OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF + * ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. * * Contact Information: * linux-drivers@emulex.com @@ -23,7 +38,7 @@ * Emulex * 3333 Susan Street * Costa Mesa, CA 92626 - *******************************************************************/ + */ #include #include diff --git a/drivers/infiniband/hw/ocrdma/ocrdma_ah.h b/drivers/infiniband/hw/ocrdma/ocrdma_ah.h index cf366fe..04a30ae 100644 --- a/drivers/infiniband/hw/ocrdma/ocrdma_ah.h +++ b/drivers/infiniband/hw/ocrdma/ocrdma_ah.h @@ -1,21 +1,36 @@ -/******************************************************************* - * This file is part of the Emulex RoCE Device Driver for * - * RoCE (RDMA over Converged Ethernet) adapters. * - * Copyright (C) 2008-2012 Emulex. All rights reserved. * - * EMULEX and SLI are trademarks of Emulex. * - * www.emulex.com * - * * - * This program is free software; you can redistribute it and/or * - * modify it under the terms of version 2 of the GNU General * - * Public License as published by the Free Software Foundation. * - * This program is distributed in the hope that it will be useful. * - * ALL EXPRESS OR IMPLIED CONDITIONS, REPRESENTATIONS AND * - * WARRANTIES, INCLUDING ANY IMPLIED WARRANTY OF MERCHANTABILITY, * - * FITNESS FOR A PARTICULAR PURPOSE, OR NON-INFRINGEMENT, ARE * - * DISCLAIMED, EXCEPT TO THE EXTENT THAT SUCH DISCLAIMERS ARE HELD * - * TO BE LEGALLY INVALID. See the GNU General Public License for * - * more details, a copy of which can be found in the file COPYING * - * included with this package. * +/* This file is part of the Emulex RoCE Device Driver for + * RoCE (RDMA over Converged Ethernet) adapters. + * Copyright (C) 2012-2015 Emulex. All rights reserved. + * EMULEX and SLI are trademarks of Emulex. + * www.emulex.com + * + * This software is available to you under a choice of one of two licenses. + * You may choose to be licensed under the terms of the GNU General Public + * License (GPL) Version 2, available from the file COPYING in the main + * directory of this source tree, or the BSD license below: + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions + * are met: + * + * - Redistributions of source code must retain the above copyright notice, + * this list of conditions and the following disclaimer. + * + * - Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in + * the documentation and/or other materials provided with the distribution. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" + * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO,THE + * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE + * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE + * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR + * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF + * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR + * BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, + * WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR + * OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF + * ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. * * Contact Information: * linux-drivers@emulex.com @@ -23,7 +38,7 @@ * Emulex * 3333 Susan Street * Costa Mesa, CA 92626 - *******************************************************************/ + */ #ifndef __OCRDMA_AH_H__ #define __OCRDMA_AH_H__ diff --git a/drivers/infiniband/hw/ocrdma/ocrdma_hw.c b/drivers/infiniband/hw/ocrdma/ocrdma_hw.c index 47615ff..aab391a 100644 --- a/drivers/infiniband/hw/ocrdma/ocrdma_hw.c +++ b/drivers/infiniband/hw/ocrdma/ocrdma_hw.c @@ -1,21 +1,36 @@ -/******************************************************************* - * This file is part of the Emulex RoCE Device Driver for * - * RoCE (RDMA over Converged Ethernet) CNA Adapters. * - * Copyright (C) 2008-2012 Emulex. All rights reserved. * - * EMULEX and SLI are trademarks of Emulex. * - * www.emulex.com * - * * - * This program is free software; you can redistribute it and/or * - * modify it under the terms of version 2 of the GNU General * - * Public License as published by the Free Software Foundation. * - * This program is distributed in the hope that it will be useful. * - * ALL EXPRESS OR IMPLIED CONDITIONS, REPRESENTATIONS AND * - * WARRANTIES, INCLUDING ANY IMPLIED WARRANTY OF MERCHANTABILITY, * - * FITNESS FOR A PARTICULAR PURPOSE, OR NON-INFRINGEMENT, ARE * - * DISCLAIMED, EXCEPT TO THE EXTENT THAT SUCH DISCLAIMERS ARE HELD * - * TO BE LEGALLY INVALID. See the GNU General Public License for * - * more details, a copy of which can be found in the file COPYING * - * included with this package. * +/* This file is part of the Emulex RoCE Device Driver for + * RoCE (RDMA over Converged Ethernet) adapters. + * Copyright (C) 2012-2015 Emulex. All rights reserved. + * EMULEX and SLI are trademarks of Emulex. + * www.emulex.com + * + * This software is available to you under a choice of one of two licenses. + * You may choose to be licensed under the terms of the GNU General Public + * License (GPL) Version 2, available from the file COPYING in the main + * directory of this source tree, or the BSD license below: + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions + * are met: + * + * - Redistributions of source code must retain the above copyright notice, + * this list of conditions and the following disclaimer. + * + * - Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in + * the documentation and/or other materials provided with the distribution. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" + * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO,THE + * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE + * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE + * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR + * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF + * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR + * BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, + * WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR + * OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF + * ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. * * Contact Information: * linux-drivers@emulex.com @@ -23,7 +38,7 @@ * Emulex * 3333 Susan Street * Costa Mesa, CA 92626 - *******************************************************************/ + */ #include #include diff --git a/drivers/infiniband/hw/ocrdma/ocrdma_hw.h b/drivers/infiniband/hw/ocrdma/ocrdma_hw.h index e905972..7ed885c 100644 --- a/drivers/infiniband/hw/ocrdma/ocrdma_hw.h +++ b/drivers/infiniband/hw/ocrdma/ocrdma_hw.h @@ -1,21 +1,36 @@ -/******************************************************************* - * This file is part of the Emulex RoCE Device Driver for * - * RoCE (RDMA over Converged Ethernet) CNA Adapters. * - * Copyright (C) 2008-2012 Emulex. All rights reserved. * - * EMULEX and SLI are trademarks of Emulex. * - * www.emulex.com * - * * - * This program is free software; you can redistribute it and/or * - * modify it under the terms of version 2 of the GNU General * - * Public License as published by the Free Software Foundation. * - * This program is distributed in the hope that it will be useful. * - * ALL EXPRESS OR IMPLIED CONDITIONS, REPRESENTATIONS AND * - * WARRANTIES, INCLUDING ANY IMPLIED WARRANTY OF MERCHANTABILITY, * - * FITNESS FOR A PARTICULAR PURPOSE, OR NON-INFRINGEMENT, ARE * - * DISCLAIMED, EXCEPT TO THE EXTENT THAT SUCH DISCLAIMERS ARE HELD * - * TO BE LEGALLY INVALID. See the GNU General Public License for * - * more details, a copy of which can be found in the file COPYING * - * included with this package. * +/* This file is part of the Emulex RoCE Device Driver for + * RoCE (RDMA over Converged Ethernet) adapters. + * Copyright (C) 2012-2015 Emulex. All rights reserved. + * EMULEX and SLI are trademarks of Emulex. + * www.emulex.com + * + * This software is available to you under a choice of one of two licenses. + * You may choose to be licensed under the terms of the GNU General Public + * License (GPL) Version 2, available from the file COPYING in the main + * directory of this source tree, or the BSD license below: + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions + * are met: + * + * - Redistributions of source code must retain the above copyright notice, + * this list of conditions and the following disclaimer. + * + * - Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in + * the documentation and/or other materials provided with the distribution. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" + * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO,THE + * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE + * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE + * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR + * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF + * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR + * BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, + * WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR + * OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF + * ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. * * Contact Information: * linux-drivers@emulex.com @@ -23,7 +38,7 @@ * Emulex * 3333 Susan Street * Costa Mesa, CA 92626 - *******************************************************************/ + */ #ifndef __OCRDMA_HW_H__ #define __OCRDMA_HW_H__ diff --git a/drivers/infiniband/hw/ocrdma/ocrdma_main.c b/drivers/infiniband/hw/ocrdma/ocrdma_main.c index d98a707..6ded95a 100644 --- a/drivers/infiniband/hw/ocrdma/ocrdma_main.c +++ b/drivers/infiniband/hw/ocrdma/ocrdma_main.c @@ -1,21 +1,36 @@ -/******************************************************************* - * This file is part of the Emulex RoCE Device Driver for * - * RoCE (RDMA over Converged Ethernet) adapters. * - * Copyright (C) 2008-2012 Emulex. All rights reserved. * - * EMULEX and SLI are trademarks of Emulex. * - * www.emulex.com * - * * - * This program is free software; you can redistribute it and/or * - * modify it under the terms of version 2 of the GNU General * - * Public License as published by the Free Software Foundation. * - * This program is distributed in the hope that it will be useful. * - * ALL EXPRESS OR IMPLIED CONDITIONS, REPRESENTATIONS AND * - * WARRANTIES, INCLUDING ANY IMPLIED WARRANTY OF MERCHANTABILITY, * - * FITNESS FOR A PARTICULAR PURPOSE, OR NON-INFRINGEMENT, ARE * - * DISCLAIMED, EXCEPT TO THE EXTENT THAT SUCH DISCLAIMERS ARE HELD * - * TO BE LEGALLY INVALID. See the GNU General Public License for * - * more details, a copy of which can be found in the file COPYING * - * included with this package. * +/* This file is part of the Emulex RoCE Device Driver for + * RoCE (RDMA over Converged Ethernet) adapters. + * Copyright (C) 2012-2015 Emulex. All rights reserved. + * EMULEX and SLI are trademarks of Emulex. + * www.emulex.com + * + * This software is available to you under a choice of one of two licenses. + * You may choose to be licensed under the terms of the GNU General Public + * License (GPL) Version 2, available from the file COPYING in the main + * directory of this source tree, or the BSD license below: + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions + * are met: + * + * - Redistributions of source code must retain the above copyright notice, + * this list of conditions and the following disclaimer. + * + * - Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in + * the documentation and/or other materials provided with the distribution. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" + * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO,THE + * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE + * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE + * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR + * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF + * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR + * BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, + * WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR + * OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF + * ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. * * Contact Information: * linux-drivers@emulex.com @@ -23,7 +38,7 @@ * Emulex * 3333 Susan Street * Costa Mesa, CA 92626 - *******************************************************************/ + */ #include #include diff --git a/drivers/infiniband/hw/ocrdma/ocrdma_sli.h b/drivers/infiniband/hw/ocrdma/ocrdma_sli.h index 02ad0ae..80006b2 100644 --- a/drivers/infiniband/hw/ocrdma/ocrdma_sli.h +++ b/drivers/infiniband/hw/ocrdma/ocrdma_sli.h @@ -1,21 +1,36 @@ -/******************************************************************* - * This file is part of the Emulex RoCE Device Driver for * - * RoCE (RDMA over Converged Ethernet) adapters. * - * Copyright (C) 2008-2012 Emulex. All rights reserved. * - * EMULEX and SLI are trademarks of Emulex. * - * www.emulex.com * - * * - * This program is free software; you can redistribute it and/or * - * modify it under the terms of version 2 of the GNU General * - * Public License as published by the Free Software Foundation. * - * This program is distributed in the hope that it will be useful. * - * ALL EXPRESS OR IMPLIED CONDITIONS, REPRESENTATIONS AND * - * WARRANTIES, INCLUDING ANY IMPLIED WARRANTY OF MERCHANTABILITY, * - * FITNESS FOR A PARTICULAR PURPOSE, OR NON-INFRINGEMENT, ARE * - * DISCLAIMED, EXCEPT TO THE EXTENT THAT SUCH DISCLAIMERS ARE HELD * - * TO BE LEGALLY INVALID. See the GNU General Public License for * - * more details, a copy of which can be found in the file COPYING * - * included with this package. * +/* This file is part of the Emulex RoCE Device Driver for + * RoCE (RDMA over Converged Ethernet) adapters. + * Copyright (C) 2012-2015 Emulex. All rights reserved. + * EMULEX and SLI are trademarks of Emulex. + * www.emulex.com + * + * This software is available to you under a choice of one of two licenses. + * You may choose to be licensed under the terms of the GNU General Public + * License (GPL) Version 2, available from the file COPYING in the main + * directory of this source tree, or the BSD license below: + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions + * are met: + * + * - Redistributions of source code must retain the above copyright notice, + * this list of conditions and the following disclaimer. + * + * - Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in + * the documentation and/or other materials provided with the distribution. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" + * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO,THE + * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE + * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE + * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR + * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF + * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR + * BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, + * WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR + * OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF + * ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. * * Contact Information: * linux-drivers@emulex.com @@ -23,7 +38,7 @@ * Emulex * 3333 Susan Street * Costa Mesa, CA 92626 - *******************************************************************/ + */ #ifndef __OCRDMA_SLI_H__ #define __OCRDMA_SLI_H__ diff --git a/drivers/infiniband/hw/ocrdma/ocrdma_stats.c b/drivers/infiniband/hw/ocrdma/ocrdma_stats.c index 48d7ef5..69334e2 100644 --- a/drivers/infiniband/hw/ocrdma/ocrdma_stats.c +++ b/drivers/infiniband/hw/ocrdma/ocrdma_stats.c @@ -1,21 +1,36 @@ -/******************************************************************* - * This file is part of the Emulex RoCE Device Driver for * - * RoCE (RDMA over Converged Ethernet) adapters. * - * Copyright (C) 2008-2014 Emulex. All rights reserved. * - * EMULEX and SLI are trademarks of Emulex. * - * www.emulex.com * - * * - * This program is free software; you can redistribute it and/or * - * modify it under the terms of version 2 of the GNU General * - * Public License as published by the Free Software Foundation. * - * This program is distributed in the hope that it will be useful. * - * ALL EXPRESS OR IMPLIED CONDITIONS, REPRESENTATIONS AND * - * WARRANTIES, INCLUDING ANY IMPLIED WARRANTY OF MERCHANTABILITY, * - * FITNESS FOR A PARTICULAR PURPOSE, OR NON-INFRINGEMENT, ARE * - * DISCLAIMED, EXCEPT TO THE EXTENT THAT SUCH DISCLAIMERS ARE HELD * - * TO BE LEGALLY INVALID. See the GNU General Public License for * - * more details, a copy of which can be found in the file COPYING * - * included with this package. * +/* This file is part of the Emulex RoCE Device Driver for + * RoCE (RDMA over Converged Ethernet) adapters. + * Copyright (C) 2012-2015 Emulex. All rights reserved. + * EMULEX and SLI are trademarks of Emulex. + * www.emulex.com + * + * This software is available to you under a choice of one of two licenses. + * You may choose to be licensed under the terms of the GNU General Public + * License (GPL) Version 2, available from the file COPYING in the main + * directory of this source tree, or the BSD license below: + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions + * are met: + * + * - Redistributions of source code must retain the above copyright notice, + * this list of conditions and the following disclaimer. + * + * - Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in + * the documentation and/or other materials provided with the distribution. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" + * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO,THE + * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE + * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE + * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR + * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF + * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR + * BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, + * WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR + * OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF + * ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. * * Contact Information: * linux-drivers@emulex.com @@ -23,7 +38,7 @@ * Emulex * 3333 Susan Street * Costa Mesa, CA 92626 - *******************************************************************/ + */ #include #include diff --git a/drivers/infiniband/hw/ocrdma/ocrdma_stats.h b/drivers/infiniband/hw/ocrdma/ocrdma_stats.h index 091edd6..c9e58d0 100644 --- a/drivers/infiniband/hw/ocrdma/ocrdma_stats.h +++ b/drivers/infiniband/hw/ocrdma/ocrdma_stats.h @@ -1,21 +1,36 @@ -/******************************************************************* - * This file is part of the Emulex RoCE Device Driver for * - * RoCE (RDMA over Converged Ethernet) adapters. * - * Copyright (C) 2008-2014 Emulex. All rights reserved. * - * EMULEX and SLI are trademarks of Emulex. * - * www.emulex.com * - * * - * This program is free software; you can redistribute it and/or * - * modify it under the terms of version 2 of the GNU General * - * Public License as published by the Free Software Foundation. * - * This program is distributed in the hope that it will be useful. * - * ALL EXPRESS OR IMPLIED CONDITIONS, REPRESENTATIONS AND * - * WARRANTIES, INCLUDING ANY IMPLIED WARRANTY OF MERCHANTABILITY, * - * FITNESS FOR A PARTICULAR PURPOSE, OR NON-INFRINGEMENT, ARE * - * DISCLAIMED, EXCEPT TO THE EXTENT THAT SUCH DISCLAIMERS ARE HELD * - * TO BE LEGALLY INVALID. See the GNU General Public License for * - * more details, a copy of which can be found in the file COPYING * - * included with this package. * +/* This file is part of the Emulex RoCE Device Driver for + * RoCE (RDMA over Converged Ethernet) adapters. + * Copyright (C) 2012-2015 Emulex. All rights reserved. + * EMULEX and SLI are trademarks of Emulex. + * www.emulex.com + * + * This software is available to you under a choice of one of two licenses. + * You may choose to be licensed under the terms of the GNU General Public + * License (GPL) Version 2, available from the file COPYING in the main + * directory of this source tree, or the BSD license below: + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions + * are met: + * + * - Redistributions of source code must retain the above copyright notice, + * this list of conditions and the following disclaimer. + * + * - Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in + * the documentation and/or other materials provided with the distribution. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" + * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO,THE + * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE + * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE + * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR + * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF + * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR + * BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, + * WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR + * OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF + * ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. * * Contact Information: * linux-drivers@emulex.com @@ -23,7 +38,7 @@ * Emulex * 3333 Susan Street * Costa Mesa, CA 92626 - *******************************************************************/ + */ #ifndef __OCRDMA_STATS_H__ #define __OCRDMA_STATS_H__ diff --git a/drivers/infiniband/hw/ocrdma/ocrdma_verbs.c b/drivers/infiniband/hw/ocrdma/ocrdma_verbs.c index 5bb61eb..bc84cd4 100644 --- a/drivers/infiniband/hw/ocrdma/ocrdma_verbs.c +++ b/drivers/infiniband/hw/ocrdma/ocrdma_verbs.c @@ -1,21 +1,36 @@ -/******************************************************************* - * This file is part of the Emulex RoCE Device Driver for * - * RoCE (RDMA over Converged Ethernet) adapters. * - * Copyright (C) 2008-2012 Emulex. All rights reserved. * - * EMULEX and SLI are trademarks of Emulex. * - * www.emulex.com * - * * - * This program is free software; you can redistribute it and/or * - * modify it under the terms of version 2 of the GNU General * - * Public License as published by the Free Software Foundation. * - * This program is distributed in the hope that it will be useful. * - * ALL EXPRESS OR IMPLIED CONDITIONS, REPRESENTATIONS AND * - * WARRANTIES, INCLUDING ANY IMPLIED WARRANTY OF MERCHANTABILITY, * - * FITNESS FOR A PARTICULAR PURPOSE, OR NON-INFRINGEMENT, ARE * - * DISCLAIMED, EXCEPT TO THE EXTENT THAT SUCH DISCLAIMERS ARE HELD * - * TO BE LEGALLY INVALID. See the GNU General Public License for * - * more details, a copy of which can be found in the file COPYING * - * included with this package. * +/* This file is part of the Emulex RoCE Device Driver for + * RoCE (RDMA over Converged Ethernet) adapters. + * Copyright (C) 2012-2015 Emulex. All rights reserved. + * EMULEX and SLI are trademarks of Emulex. + * www.emulex.com + * + * This software is available to you under a choice of one of two licenses. + * You may choose to be licensed under the terms of the GNU General Public + * License (GPL) Version 2, available from the file COPYING in the main + * directory of this source tree, or the BSD license below: + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions + * are met: + * + * - Redistributions of source code must retain the above copyright notice, + * this list of conditions and the following disclaimer. + * + * - Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in + * the documentation and/or other materials provided with the distribution. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" + * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO,THE + * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE + * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE + * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR + * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF + * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR + * BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, + * WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR + * OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF + * ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. * * Contact Information: * linux-drivers@emulex.com @@ -23,7 +38,7 @@ * Emulex * 3333 Susan Street * Costa Mesa, CA 92626 - *******************************************************************/ + */ #include #include diff --git a/drivers/infiniband/hw/ocrdma/ocrdma_verbs.h b/drivers/infiniband/hw/ocrdma/ocrdma_verbs.h index b15c608..eaccb2d 100644 --- a/drivers/infiniband/hw/ocrdma/ocrdma_verbs.h +++ b/drivers/infiniband/hw/ocrdma/ocrdma_verbs.h @@ -1,21 +1,36 @@ -/******************************************************************* - * This file is part of the Emulex RoCE Device Driver for * - * RoCE (RDMA over Converged Ethernet) adapters. * - * Copyright (C) 2008-2012 Emulex. All rights reserved. * - * EMULEX and SLI are trademarks of Emulex. * - * www.emulex.com * - * * - * This program is free software; you can redistribute it and/or * - * modify it under the terms of version 2 of the GNU General * - * Public License as published by the Free Software Foundation. * - * This program is distributed in the hope that it will be useful. * - * ALL EXPRESS OR IMPLIED CONDITIONS, REPRESENTATIONS AND * - * WARRANTIES, INCLUDING ANY IMPLIED WARRANTY OF MERCHANTABILITY, * - * FITNESS FOR A PARTICULAR PURPOSE, OR NON-INFRINGEMENT, ARE * - * DISCLAIMED, EXCEPT TO THE EXTENT THAT SUCH DISCLAIMERS ARE HELD * - * TO BE LEGALLY INVALID. See the GNU General Public License for * - * more details, a copy of which can be found in the file COPYING * - * included with this package. * +/* This file is part of the Emulex RoCE Device Driver for + * RoCE (RDMA over Converged Ethernet) adapters. + * Copyright (C) 2012-2015 Emulex. All rights reserved. + * EMULEX and SLI are trademarks of Emulex. + * www.emulex.com + * + * This software is available to you under a choice of one of two licenses. + * You may choose to be licensed under the terms of the GNU General Public + * License (GPL) Version 2, available from the file COPYING in the main + * directory of this source tree, or the BSD license below: + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions + * are met: + * + * - Redistributions of source code must retain the above copyright notice, + * this list of conditions and the following disclaimer. + * + * - Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in + * the documentation and/or other materials provided with the distribution. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" + * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO,THE + * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE + * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE + * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR + * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF + * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR + * BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, + * WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR + * OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF + * ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. * * Contact Information: * linux-drivers@emulex.com @@ -23,7 +38,7 @@ * Emulex * 3333 Susan Street * Costa Mesa, CA 92626 - *******************************************************************/ + */ #ifndef __OCRDMA_VERBS_H__ #define __OCRDMA_VERBS_H__ -- cgit v0.10.2 From b8f5595eb96c9fce1c907d13e89581e5061edf2e Mon Sep 17 00:00:00 2001 From: Devesh Sharma Date: Fri, 24 Jul 2015 05:04:00 +0530 Subject: RDMA/ocrdma: update ocrdma module license string Change module_license from "GPL" to "Dual BSD/GPL" Cc: Tejun Heo Cc: Duan Jiong Cc: Roland Dreier Cc: Jes Sorensen Cc: Sasha Levin Cc: Dan Carpenter Cc: Prarit Bhargava Cc: Colin Ian King Cc: Wei Yongjun Cc: Moni Shoua Cc: Rasmus Villemoes Cc: Li RongQing Cc: Devendra Naga Signed-off-by: Devesh Sharma Signed-off-by: Doug Ledford diff --git a/drivers/infiniband/hw/ocrdma/ocrdma_main.c b/drivers/infiniband/hw/ocrdma/ocrdma_main.c index 6ded95a..b119a34 100644 --- a/drivers/infiniband/hw/ocrdma/ocrdma_main.c +++ b/drivers/infiniband/hw/ocrdma/ocrdma_main.c @@ -61,7 +61,7 @@ MODULE_VERSION(OCRDMA_ROCE_DRV_VERSION); MODULE_DESCRIPTION(OCRDMA_ROCE_DRV_DESC " " OCRDMA_ROCE_DRV_VERSION); MODULE_AUTHOR("Emulex Corporation"); -MODULE_LICENSE("GPL"); +MODULE_LICENSE("Dual BSD/GPL"); static LIST_HEAD(ocrdma_dev_list); static DEFINE_SPINLOCK(ocrdma_devlist_lock); -- cgit v0.10.2 From b38ebd1d4b6656582b8c16358bb88d059d28b794 Mon Sep 17 00:00:00 2001 From: Dmitry Torokhov Date: Wed, 22 Jul 2015 14:56:39 -0700 Subject: Input: LEDs - skip unnamed LEDs Devices may declare more LEDs than what is known to input-leds (HID does this for some devices). Instead of showing ugly warnings on connect and, even worse, oopsing on disconnect, let's simply ignore LEDs that are not known to us. Reported-and-tested-by: Vlastimil Babka Signed-off-by: Dmitry Torokhov diff --git a/drivers/input/input-leds.c b/drivers/input/input-leds.c index 074a65e..766bf26 100644 --- a/drivers/input/input-leds.c +++ b/drivers/input/input-leds.c @@ -71,6 +71,18 @@ static void input_leds_event(struct input_handle *handle, unsigned int type, { } +static int input_leds_get_count(struct input_dev *dev) +{ + unsigned int led_code; + int count = 0; + + for_each_set_bit(led_code, dev->ledbit, LED_CNT) + if (input_led_info[led_code].name) + count++; + + return count; +} + static int input_leds_connect(struct input_handler *handler, struct input_dev *dev, const struct input_device_id *id) @@ -81,7 +93,7 @@ static int input_leds_connect(struct input_handler *handler, int led_no; int error; - num_leds = bitmap_weight(dev->ledbit, LED_CNT); + num_leds = input_leds_get_count(dev); if (!num_leds) return -ENXIO; @@ -112,7 +124,7 @@ static int input_leds_connect(struct input_handler *handler, led->handle = &leds->handle; led->code = led_code; - if (WARN_ON(!input_led_info[led_code].name)) + if (!input_led_info[led_code].name) continue; led->cdev.name = kasprintf(GFP_KERNEL, "%s::%s", -- cgit v0.10.2 From 8b5a359c5b3e631f17eeb1dcb930474000d33d49 Mon Sep 17 00:00:00 2001 From: Bastien Nocera Date: Fri, 24 Jul 2015 09:08:53 -0700 Subject: Input: goodix - fix touch coordinates on WinBook TW100 and TW700 The touchscreen on the WinBook TW100 and TW700 don't match the default display, with 0,0 touches being reported when touching at the bottom right of the screen. 1280,800 0,800 +-------------+ | | | | | | +-------------+ 1280,0 0,0 It's unfortunately impossible to detect this problem with data from the DSDT, or other auxiliary metadata, so fallback to quirking this specific model of tablet instead. Signed-off-by: Bastien Nocera Reviewed-by: Benjamin Tissoires Signed-off-by: Dmitry Torokhov diff --git a/drivers/input/touchscreen/goodix.c b/drivers/input/touchscreen/goodix.c index b4d12e2..e36162b 100644 --- a/drivers/input/touchscreen/goodix.c +++ b/drivers/input/touchscreen/goodix.c @@ -15,6 +15,7 @@ */ #include +#include #include #include #include @@ -34,6 +35,7 @@ struct goodix_ts_data { int abs_y_max; unsigned int max_touch_num; unsigned int int_trigger_type; + bool rotated_screen; }; #define GOODIX_MAX_HEIGHT 4096 @@ -60,6 +62,30 @@ static const unsigned long goodix_irq_flags[] = { IRQ_TYPE_LEVEL_HIGH, }; +/* + * Those tablets have their coordinates origin at the bottom right + * of the tablet, as if rotated 180 degrees + */ +static const struct dmi_system_id rotated_screen[] = { +#if defined(CONFIG_DMI) && defined(CONFIG_X86) + { + .ident = "WinBook TW100", + .matches = { + DMI_MATCH(DMI_SYS_VENDOR, "WinBook"), + DMI_MATCH(DMI_PRODUCT_NAME, "TW100") + } + }, + { + .ident = "WinBook TW700", + .matches = { + DMI_MATCH(DMI_SYS_VENDOR, "WinBook"), + DMI_MATCH(DMI_PRODUCT_NAME, "TW700") + }, + }, +#endif + {} +}; + /** * goodix_i2c_read - read data from a register of the i2c slave device. * @@ -129,6 +155,11 @@ static void goodix_ts_report_touch(struct goodix_ts_data *ts, u8 *coor_data) int input_y = get_unaligned_le16(&coor_data[3]); int input_w = get_unaligned_le16(&coor_data[5]); + if (ts->rotated_screen) { + input_x = ts->abs_x_max - input_x; + input_y = ts->abs_y_max - input_y; + } + input_mt_slot(ts->input_dev, id); input_mt_report_slot_state(ts->input_dev, MT_TOOL_FINGER, true); input_report_abs(ts->input_dev, ABS_MT_POSITION_X, input_x); @@ -223,6 +254,11 @@ static void goodix_read_config(struct goodix_ts_data *ts) ts->abs_y_max = GOODIX_MAX_HEIGHT; ts->max_touch_num = GOODIX_MAX_CONTACTS; } + + ts->rotated_screen = dmi_check_system(rotated_screen); + if (ts->rotated_screen) + dev_dbg(&ts->client->dev, + "Applying '180 degrees rotated screen' quirk\n"); } /** -- cgit v0.10.2 From bffc4496886683fac86d31e5d0cf9a22f8044e3d Mon Sep 17 00:00:00 2001 From: Axel Lin Date: Sun, 19 Jul 2015 12:09:16 +0800 Subject: ASoC: cs4265: Fix setting dai format for Left/Right Justified MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit The settings in current code does not match the datasheet, fix it. DAC Control - Address 03h DAC Digital Interface Format (Bits 5:4) DAC_DIF1 DAC_DIF0 Description 0 0 Left Justified, up to 24-bit data (default) 0 1 I²S, up to 24-bit data 1 0 Right-Justified, 16-bit Data 1 1 Right-Justified, 24-bit Data Transmitter Control 2 - Address 12h Transmitter Digital Interface Format (Bits 7:6) Tx_DIF1 Tx_DIF0 Description Format Figure 0 0 Left Justified, up to 24-bit data (default) 0 1 I²S, up to 24-bit data 1 0 Right-Justified, 16-bit Data 1 1 Right-Justified, 24-bit Data Signed-off-by: Axel Lin Signed-off-by: Mark Brown diff --git a/sound/soc/codecs/cs4265.c b/sound/soc/codecs/cs4265.c index d7ec4756..8e36198 100644 --- a/sound/soc/codecs/cs4265.c +++ b/sound/soc/codecs/cs4265.c @@ -457,14 +457,14 @@ static int cs4265_pcm_hw_params(struct snd_pcm_substream *substream, case SND_SOC_DAIFMT_RIGHT_J: if (params_width(params) == 16) { snd_soc_update_bits(codec, CS4265_DAC_CTL, - CS4265_DAC_CTL_DIF, (1 << 5)); + CS4265_DAC_CTL_DIF, (2 << 4)); snd_soc_update_bits(codec, CS4265_SPDIF_CTL2, - CS4265_SPDIF_CTL2_DIF, (1 << 7)); + CS4265_SPDIF_CTL2_DIF, (2 << 6)); } else { snd_soc_update_bits(codec, CS4265_DAC_CTL, - CS4265_DAC_CTL_DIF, (3 << 5)); + CS4265_DAC_CTL_DIF, (3 << 4)); snd_soc_update_bits(codec, CS4265_SPDIF_CTL2, - CS4265_SPDIF_CTL2_DIF, (1 << 7)); + CS4265_SPDIF_CTL2_DIF, (3 << 6)); } break; case SND_SOC_DAIFMT_LEFT_J: @@ -473,7 +473,7 @@ static int cs4265_pcm_hw_params(struct snd_pcm_substream *substream, snd_soc_update_bits(codec, CS4265_ADC_CTL, CS4265_ADC_DIF, 0); snd_soc_update_bits(codec, CS4265_SPDIF_CTL2, - CS4265_SPDIF_CTL2_DIF, (1 << 6)); + CS4265_SPDIF_CTL2_DIF, 0); break; default: -- cgit v0.10.2 From e3eea1404f5ff7a2ceb7b5e7ba412a6fd94f2935 Mon Sep 17 00:00:00 2001 From: "Steven Rostedt (Red Hat)" Date: Fri, 24 Jul 2015 10:38:12 -0400 Subject: ftrace: Fix breakage of set_ftrace_pid Commit 4104d326b670 ("ftrace: Remove global function list and call function directly") simplified the ftrace code by removing the global_ops list with a new design. But this cleanup also broke the filtering of PIDs that are added to the set_ftrace_pid file. Add back the proper hooks to have pid filtering working once again. Cc: stable@vger.kernel.org # 3.16+ Reported-by: Matt Fleming Reported-by: Richard Weinberger Tested-by: Matt Fleming Signed-off-by: Steven Rostedt diff --git a/include/linux/ftrace.h b/include/linux/ftrace.h index 1da6029..6cd8c0e 100644 --- a/include/linux/ftrace.h +++ b/include/linux/ftrace.h @@ -116,6 +116,7 @@ ftrace_func_t ftrace_ops_get_func(struct ftrace_ops *ops); * SAVE_REGS. If another ops with this flag set is already registered * for any of the functions that this ops will be registered for, then * this ops will fail to register or set_filter_ip. + * PID - Is affected by set_ftrace_pid (allows filtering on those pids) */ enum { FTRACE_OPS_FL_ENABLED = 1 << 0, @@ -132,6 +133,7 @@ enum { FTRACE_OPS_FL_MODIFYING = 1 << 11, FTRACE_OPS_FL_ALLOC_TRAMP = 1 << 12, FTRACE_OPS_FL_IPMODIFY = 1 << 13, + FTRACE_OPS_FL_PID = 1 << 14, }; #ifdef CONFIG_DYNAMIC_FTRACE @@ -159,6 +161,7 @@ struct ftrace_ops { struct ftrace_ops *next; unsigned long flags; void *private; + ftrace_func_t saved_func; int __percpu *disabled; #ifdef CONFIG_DYNAMIC_FTRACE int nr_trampolines; diff --git a/kernel/trace/ftrace.c b/kernel/trace/ftrace.c index 02bece4..eb11011 100644 --- a/kernel/trace/ftrace.c +++ b/kernel/trace/ftrace.c @@ -98,6 +98,13 @@ struct ftrace_pid { struct pid *pid; }; +static bool ftrace_pids_enabled(void) +{ + return !list_empty(&ftrace_pids); +} + +static void ftrace_update_trampoline(struct ftrace_ops *ops); + /* * ftrace_disabled is set when an anomaly is discovered. * ftrace_disabled is much stronger than ftrace_enabled. @@ -109,7 +116,6 @@ static DEFINE_MUTEX(ftrace_lock); static struct ftrace_ops *ftrace_control_list __read_mostly = &ftrace_list_end; static struct ftrace_ops *ftrace_ops_list __read_mostly = &ftrace_list_end; ftrace_func_t ftrace_trace_function __read_mostly = ftrace_stub; -ftrace_func_t ftrace_pid_function __read_mostly = ftrace_stub; static struct ftrace_ops global_ops; static struct ftrace_ops control_ops; @@ -183,14 +189,7 @@ static void ftrace_pid_func(unsigned long ip, unsigned long parent_ip, if (!test_tsk_trace_trace(current)) return; - ftrace_pid_function(ip, parent_ip, op, regs); -} - -static void set_ftrace_pid_function(ftrace_func_t func) -{ - /* do not set ftrace_pid_function to itself! */ - if (func != ftrace_pid_func) - ftrace_pid_function = func; + op->saved_func(ip, parent_ip, op, regs); } /** @@ -202,7 +201,6 @@ static void set_ftrace_pid_function(ftrace_func_t func) void clear_ftrace_function(void) { ftrace_trace_function = ftrace_stub; - ftrace_pid_function = ftrace_stub; } static void control_ops_disable_all(struct ftrace_ops *ops) @@ -436,6 +434,12 @@ static int __register_ftrace_function(struct ftrace_ops *ops) } else add_ftrace_ops(&ftrace_ops_list, ops); + /* Always save the function, and reset at unregistering */ + ops->saved_func = ops->func; + + if (ops->flags & FTRACE_OPS_FL_PID && ftrace_pids_enabled()) + ops->func = ftrace_pid_func; + ftrace_update_trampoline(ops); if (ftrace_enabled) @@ -463,15 +467,28 @@ static int __unregister_ftrace_function(struct ftrace_ops *ops) if (ftrace_enabled) update_ftrace_function(); + ops->func = ops->saved_func; + return 0; } static void ftrace_update_pid_func(void) { + bool enabled = ftrace_pids_enabled(); + struct ftrace_ops *op; + /* Only do something if we are tracing something */ if (ftrace_trace_function == ftrace_stub) return; + do_for_each_ftrace_op(op, ftrace_ops_list) { + if (op->flags & FTRACE_OPS_FL_PID) { + op->func = enabled ? ftrace_pid_func : + op->saved_func; + ftrace_update_trampoline(op); + } + } while_for_each_ftrace_op(op); + update_ftrace_function(); } @@ -1133,7 +1150,8 @@ static struct ftrace_ops global_ops = { .local_hash.filter_hash = EMPTY_HASH, INIT_OPS_HASH(global_ops) .flags = FTRACE_OPS_FL_RECURSION_SAFE | - FTRACE_OPS_FL_INITIALIZED, + FTRACE_OPS_FL_INITIALIZED | + FTRACE_OPS_FL_PID, }; /* @@ -5023,7 +5041,9 @@ static void ftrace_update_trampoline(struct ftrace_ops *ops) static struct ftrace_ops global_ops = { .func = ftrace_stub, - .flags = FTRACE_OPS_FL_RECURSION_SAFE | FTRACE_OPS_FL_INITIALIZED, + .flags = FTRACE_OPS_FL_RECURSION_SAFE | + FTRACE_OPS_FL_INITIALIZED | + FTRACE_OPS_FL_PID, }; static int __init ftrace_nodyn_init(void) @@ -5080,11 +5100,6 @@ void ftrace_init_array_ops(struct trace_array *tr, ftrace_func_t func) if (WARN_ON(tr->ops->func != ftrace_stub)) printk("ftrace ops had %pS for function\n", tr->ops->func); - /* Only the top level instance does pid tracing */ - if (!list_empty(&ftrace_pids)) { - set_ftrace_pid_function(func); - func = ftrace_pid_func; - } } tr->ops->func = func; tr->ops->private = tr; @@ -5371,7 +5386,7 @@ static void *fpid_start(struct seq_file *m, loff_t *pos) { mutex_lock(&ftrace_lock); - if (list_empty(&ftrace_pids) && (!*pos)) + if (!ftrace_pids_enabled() && (!*pos)) return (void *) 1; return seq_list_start(&ftrace_pids, *pos); @@ -5610,6 +5625,7 @@ static struct ftrace_ops graph_ops = { .func = ftrace_stub, .flags = FTRACE_OPS_FL_RECURSION_SAFE | FTRACE_OPS_FL_INITIALIZED | + FTRACE_OPS_FL_PID | FTRACE_OPS_FL_STUB, #ifdef FTRACE_GRAPH_TRAMP_ADDR .trampoline = FTRACE_GRAPH_TRAMP_ADDR, -- cgit v0.10.2 From 4bc94d5dc95d3a2159d821b39277477e69507f67 Mon Sep 17 00:00:00 2001 From: Alex Williamson Date: Fri, 24 Jul 2015 15:14:04 -0600 Subject: vfio: Fix lockdep issue When we open a device file descriptor, we currently have the following: vfio_group_get_device_fd() mutex_lock(&group->device_lock); open() ... if (ret) release() If we hit that error case, we call the backend driver release path, which for vfio-pci looks like this: vfio_pci_release() vfio_pci_disable() vfio_pci_try_bus_reset() vfio_pci_get_devs() vfio_device_get_from_dev() vfio_group_get_device() mutex_lock(&group->device_lock); Whoops, we've stumbled back onto group.device_lock and created a deadlock. There's a low likelihood of ever seeing this play out, but obviously it needs to be fixed. To do that we can use a reference to the vfio_device for vfio_group_get_device_fd() rather than holding the lock. There was a loop in this function, theoretically allowing multiple devices with the same name, but in practice we don't expect such a thing to happen and the code is already aborting from the loop with break on any sort of error rather than continuing and only parsing the first match anyway, so the loop was effectively unused already. Signed-off-by: Alex Williamson Fixes: 20f300175a1e ("vfio/pci: Fix racy vfio_device_get_from_dev() call") Reported-by: Joerg Roedel Tested-by: Joerg Roedel diff --git a/drivers/vfio/vfio.c b/drivers/vfio/vfio.c index 2fb29df..563c510 100644 --- a/drivers/vfio/vfio.c +++ b/drivers/vfio/vfio.c @@ -689,6 +689,23 @@ struct vfio_device *vfio_device_get_from_dev(struct device *dev) } EXPORT_SYMBOL_GPL(vfio_device_get_from_dev); +static struct vfio_device *vfio_device_get_from_name(struct vfio_group *group, + char *buf) +{ + struct vfio_device *device; + + mutex_lock(&group->device_lock); + list_for_each_entry(device, &group->device_list, group_next) { + if (!strcmp(dev_name(device->dev), buf)) { + vfio_device_get(device); + break; + } + } + mutex_unlock(&group->device_lock); + + return device; +} + /* * Caller must hold a reference to the vfio_device */ @@ -1198,53 +1215,53 @@ static int vfio_group_get_device_fd(struct vfio_group *group, char *buf) { struct vfio_device *device; struct file *filep; - int ret = -ENODEV; + int ret; if (0 == atomic_read(&group->container_users) || !group->container->iommu_driver || !vfio_group_viable(group)) return -EINVAL; - mutex_lock(&group->device_lock); - list_for_each_entry(device, &group->device_list, group_next) { - if (strcmp(dev_name(device->dev), buf)) - continue; + device = vfio_device_get_from_name(group, buf); + if (!device) + return -ENODEV; - ret = device->ops->open(device->device_data); - if (ret) - break; - /* - * We can't use anon_inode_getfd() because we need to modify - * the f_mode flags directly to allow more than just ioctls - */ - ret = get_unused_fd_flags(O_CLOEXEC); - if (ret < 0) { - device->ops->release(device->device_data); - break; - } + ret = device->ops->open(device->device_data); + if (ret) { + vfio_device_put(device); + return ret; + } - filep = anon_inode_getfile("[vfio-device]", &vfio_device_fops, - device, O_RDWR); - if (IS_ERR(filep)) { - put_unused_fd(ret); - ret = PTR_ERR(filep); - device->ops->release(device->device_data); - break; - } + /* + * We can't use anon_inode_getfd() because we need to modify + * the f_mode flags directly to allow more than just ioctls + */ + ret = get_unused_fd_flags(O_CLOEXEC); + if (ret < 0) { + device->ops->release(device->device_data); + vfio_device_put(device); + return ret; + } - /* - * TODO: add an anon_inode interface to do this. - * Appears to be missing by lack of need rather than - * explicitly prevented. Now there's need. - */ - filep->f_mode |= (FMODE_LSEEK | FMODE_PREAD | FMODE_PWRITE); + filep = anon_inode_getfile("[vfio-device]", &vfio_device_fops, + device, O_RDWR); + if (IS_ERR(filep)) { + put_unused_fd(ret); + ret = PTR_ERR(filep); + device->ops->release(device->device_data); + vfio_device_put(device); + return ret; + } + + /* + * TODO: add an anon_inode interface to do this. + * Appears to be missing by lack of need rather than + * explicitly prevented. Now there's need. + */ + filep->f_mode |= (FMODE_LSEEK | FMODE_PREAD | FMODE_PWRITE); - vfio_device_get(device); - atomic_inc(&group->container_users); + atomic_inc(&group->container_users); - fd_install(ret, filep); - break; - } - mutex_unlock(&group->device_lock); + fd_install(ret, filep); return ret; } -- cgit v0.10.2 From ba9f6f64a0ff6b7ecaed72144c179061f8eca378 Mon Sep 17 00:00:00 2001 From: Saurav Kashyap Date: Wed, 10 Jun 2015 11:05:17 -0400 Subject: qla2xxx: Fix hardware lock/unlock issue causing kernel panic. [ Upstream commit ef86cb2059a14b4024c7320999ee58e938873032 ] This patch fixes a kernel panic for qla2xxx Target core Module driver introduced by a fix in the qla2xxx initiator code. Commit ef86cb2 ("qla2xxx: Mark port lost when we receive an RSCN for it.") introduced the regression for qla2xxx Target driver. Stack trace will have following signature --- --- [ffff88081faa3cc8] _raw_spin_lock_irqsave at ffffffff815b1f03 [ffff88081faa3cd0] qlt_fc_port_deleted at ffffffffa096ccd0 [qla2xxx] [ffff88081faa3d20] qla2x00_schedule_rport_del at ffffffffa0913831[qla2xxx] [ffff88081faa3d50] qla2x00_mark_device_lost at ffffffffa09159c5[qla2xxx] [ffff88081faa3db0] qla2x00_async_event at ffffffffa0938d59 [qla2xxx] [ffff88081faa3e30] qla24xx_msix_default at ffffffffa093a326 [qla2xxx] [ffff88081faa3e90] handle_irq_event_percpu at ffffffff810a7b8d [ffff88081faa3ee0] handle_irq_event at ffffffff810a7d32 [ffff88081faa3f10] handle_edge_irq at ffffffff810ab6b9 [ffff88081faa3f30] handle_irq at ffffffff8100619c [ffff88081faa3f70] do_IRQ at ffffffff815b4b1c --- --- Cc: # v3.18+ Signed-off-by: Saurav Kashyap Signed-off-by: Himanshu Madhani Reviewed-by: Nicholas Bellinger Signed-off-by: Nicholas Bellinger diff --git a/drivers/scsi/qla2xxx/qla_init.c b/drivers/scsi/qla2xxx/qla_init.c index 6640131..6c302c4 100644 --- a/drivers/scsi/qla2xxx/qla_init.c +++ b/drivers/scsi/qla2xxx/qla_init.c @@ -2924,6 +2924,7 @@ qla2x00_rport_del(void *data) struct fc_rport *rport; scsi_qla_host_t *vha = fcport->vha; unsigned long flags; + unsigned long vha_flags; spin_lock_irqsave(fcport->vha->host->host_lock, flags); rport = fcport->drport ? fcport->drport: fcport->rport; @@ -2935,7 +2936,9 @@ qla2x00_rport_del(void *data) * Release the target mode FC NEXUS in qla_target.c code * if target mod is enabled. */ + spin_lock_irqsave(&vha->hw->hardware_lock, vha_flags); qlt_fc_port_deleted(vha, fcport); + spin_unlock_irqrestore(&vha->hw->hardware_lock, vha_flags); } } @@ -3303,6 +3306,7 @@ qla2x00_reg_remote_port(scsi_qla_host_t *vha, fc_port_t *fcport) * Create target mode FC NEXUS in qla_target.c if target mode is * enabled.. */ + qlt_fc_port_added(vha, fcport); spin_lock_irqsave(fcport->vha->host->host_lock, flags); diff --git a/drivers/scsi/qla2xxx/qla_target.c b/drivers/scsi/qla2xxx/qla_target.c index b749026..bce2bea0 100644 --- a/drivers/scsi/qla2xxx/qla_target.c +++ b/drivers/scsi/qla2xxx/qla_target.c @@ -782,10 +782,8 @@ void qlt_fc_port_added(struct scsi_qla_host *vha, fc_port_t *fcport) void qlt_fc_port_deleted(struct scsi_qla_host *vha, fc_port_t *fcport) { - struct qla_hw_data *ha = vha->hw; struct qla_tgt *tgt = vha->vha_tgt.qla_tgt; struct qla_tgt_sess *sess; - unsigned long flags; if (!vha->hw->tgt.tgt_ops) return; @@ -793,14 +791,11 @@ void qlt_fc_port_deleted(struct scsi_qla_host *vha, fc_port_t *fcport) if (!tgt || (fcport->port_type != FCT_INITIATOR)) return; - spin_lock_irqsave(&ha->hardware_lock, flags); if (tgt->tgt_stop) { - spin_unlock_irqrestore(&ha->hardware_lock, flags); return; } sess = qlt_find_sess_by_port_name(tgt, fcport->port_name); if (!sess) { - spin_unlock_irqrestore(&ha->hardware_lock, flags); return; } @@ -808,7 +803,6 @@ void qlt_fc_port_deleted(struct scsi_qla_host *vha, fc_port_t *fcport) sess->local = 1; qlt_schedule_sess_for_deletion(sess, false); - spin_unlock_irqrestore(&ha->hardware_lock, flags); } static inline int test_tgt_sess_count(struct qla_tgt *tgt) -- cgit v0.10.2 From b20f02e1418d8a5617b81464c612de09aa55e552 Mon Sep 17 00:00:00 2001 From: Himanshu Madhani Date: Wed, 10 Jun 2015 11:05:18 -0400 Subject: qla2xxx: Enable target mode for ISP27XX Signed-off-by: Himanshu Madhani Signed-off-by: Giridhar Malavali Reviewed-by: Nicholas Bellinger Signed-off-by: Nicholas Bellinger diff --git a/drivers/scsi/qla2xxx/qla_attr.c b/drivers/scsi/qla2xxx/qla_attr.c index 82b92c4..437254e 100644 --- a/drivers/scsi/qla2xxx/qla_attr.c +++ b/drivers/scsi/qla2xxx/qla_attr.c @@ -738,7 +738,7 @@ qla2x00_sysfs_write_reset(struct file *filp, struct kobject *kobj, ql_log(ql_log_info, vha, 0x706f, "Issuing MPI reset.\n"); - if (IS_QLA83XX(ha)) { + if (IS_QLA83XX(ha) || IS_QLA27XX(ha)) { uint32_t idc_control; qla83xx_idc_lock(vha, 0); diff --git a/drivers/scsi/qla2xxx/qla_def.h b/drivers/scsi/qla2xxx/qla_def.h index e86201d..787f901 100644 --- a/drivers/scsi/qla2xxx/qla_def.h +++ b/drivers/scsi/qla2xxx/qla_def.h @@ -3154,13 +3154,13 @@ struct qla_hw_data { /* Bit 21 of fw_attributes decides the MCTP capabilities */ #define IS_MCTP_CAPABLE(ha) (IS_QLA2031(ha) && \ ((ha)->fw_attributes_ext[0] & BIT_0)) -#define IS_PI_UNINIT_CAPABLE(ha) (IS_QLA83XX(ha)) -#define IS_PI_IPGUARD_CAPABLE(ha) (IS_QLA83XX(ha)) +#define IS_PI_UNINIT_CAPABLE(ha) (IS_QLA83XX(ha) || IS_QLA27XX(ha)) +#define IS_PI_IPGUARD_CAPABLE(ha) (IS_QLA83XX(ha) || IS_QLA27XX(ha)) #define IS_PI_DIFB_DIX0_CAPABLE(ha) (0) -#define IS_PI_SPLIT_DET_CAPABLE_HBA(ha) (IS_QLA83XX(ha)) +#define IS_PI_SPLIT_DET_CAPABLE_HBA(ha) (IS_QLA83XX(ha) || IS_QLA27XX(ha)) #define IS_PI_SPLIT_DET_CAPABLE(ha) (IS_PI_SPLIT_DET_CAPABLE_HBA(ha) && \ (((ha)->fw_attributes_h << 16 | (ha)->fw_attributes) & BIT_22)) -#define IS_ATIO_MSIX_CAPABLE(ha) (IS_QLA83XX(ha)) +#define IS_ATIO_MSIX_CAPABLE(ha) (IS_QLA83XX(ha) || IS_QLA27XX(ha)) #define IS_TGT_MODE_CAPABLE(ha) (ha->tgt.atio_q_length) #define IS_SHADOW_REG_CAPABLE(ha) (IS_QLA27XX(ha)) #define IS_DPORT_CAPABLE(ha) (IS_QLA83XX(ha) || IS_QLA27XX(ha)) diff --git a/drivers/scsi/qla2xxx/qla_init.c b/drivers/scsi/qla2xxx/qla_init.c index 6c302c4..17c3bd9 100644 --- a/drivers/scsi/qla2xxx/qla_init.c +++ b/drivers/scsi/qla2xxx/qla_init.c @@ -1538,7 +1538,7 @@ qla2x00_alloc_fw_dump(scsi_qla_host_t *vha) mem_size = (ha->fw_memory_size - 0x11000 + 1) * sizeof(uint16_t); } else if (IS_FWI2_CAPABLE(ha)) { - if (IS_QLA83XX(ha)) + if (IS_QLA83XX(ha) || IS_QLA27XX(ha)) fixed_size = offsetof(struct qla83xx_fw_dump, ext_mem); else if (IS_QLA81XX(ha)) fixed_size = offsetof(struct qla81xx_fw_dump, ext_mem); @@ -1550,7 +1550,7 @@ qla2x00_alloc_fw_dump(scsi_qla_host_t *vha) mem_size = (ha->fw_memory_size - 0x100000 + 1) * sizeof(uint32_t); if (ha->mqenable) { - if (!IS_QLA83XX(ha)) + if (!IS_QLA83XX(ha) && !IS_QLA27XX(ha)) mq_size = sizeof(struct qla2xxx_mq_chain); /* * Allocate maximum buffer size for all queues. diff --git a/drivers/scsi/qla2xxx/qla_mbx.c b/drivers/scsi/qla2xxx/qla_mbx.c index 02b1c1c..b2f713a 100644 --- a/drivers/scsi/qla2xxx/qla_mbx.c +++ b/drivers/scsi/qla2xxx/qla_mbx.c @@ -2415,7 +2415,8 @@ qla2x00_get_resource_cnts(scsi_qla_host_t *vha, uint16_t *cur_xchg_cnt, *orig_iocb_cnt = mcp->mb[10]; if (vha->hw->flags.npiv_supported && max_npiv_vports) *max_npiv_vports = mcp->mb[11]; - if ((IS_QLA81XX(vha->hw) || IS_QLA83XX(vha->hw)) && max_fcfs) + if ((IS_QLA81XX(vha->hw) || IS_QLA83XX(vha->hw) || + IS_QLA27XX(vha->hw)) && max_fcfs) *max_fcfs = mcp->mb[12]; } @@ -3898,7 +3899,7 @@ qla25xx_init_rsp_que(struct scsi_qla_host *vha, struct rsp_que *rsp) spin_lock_irqsave(&ha->hardware_lock, flags); if (!(rsp->options & BIT_0)) { WRT_REG_DWORD(rsp->rsp_q_out, 0); - if (!IS_QLA83XX(ha)) + if (!IS_QLA83XX(ha) && !IS_QLA27XX(ha)) WRT_REG_DWORD(rsp->rsp_q_in, 0); } @@ -5345,7 +5346,7 @@ qla83xx_restart_nic_firmware(scsi_qla_host_t *vha) mbx_cmd_t *mcp = &mc; struct qla_hw_data *ha = vha->hw; - if (!IS_QLA83XX(ha)) + if (!IS_QLA83XX(ha) && !IS_QLA27XX(ha)) return QLA_FUNCTION_FAILED; ql_dbg(ql_dbg_mbx, vha, 0x1143, "Entered %s.\n", __func__); diff --git a/drivers/scsi/qla2xxx/qla_os.c b/drivers/scsi/qla2xxx/qla_os.c index a28815b..7c7528b 100644 --- a/drivers/scsi/qla2xxx/qla_os.c +++ b/drivers/scsi/qla2xxx/qla_os.c @@ -2504,6 +2504,7 @@ qla2x00_probe_one(struct pci_dev *pdev, const struct pci_device_id *id) ha->mbx_count = MAILBOX_REGISTER_COUNT; req_length = REQUEST_ENTRY_CNT_24XX; rsp_length = RESPONSE_ENTRY_CNT_2300; + ha->tgt.atio_q_length = ATIO_ENTRY_CNT_24XX; ha->max_loop_id = SNS_LAST_LOOP_ID_2300; ha->init_cb_size = sizeof(struct mid_init_cb_81xx); ha->gid_list_info_size = 8; diff --git a/drivers/scsi/qla2xxx/qla_sup.c b/drivers/scsi/qla2xxx/qla_sup.c index 028e8c8..2feb5f3 100644 --- a/drivers/scsi/qla2xxx/qla_sup.c +++ b/drivers/scsi/qla2xxx/qla_sup.c @@ -1697,7 +1697,7 @@ qla83xx_select_led_port(struct qla_hw_data *ha) { uint32_t led_select_value = 0; - if (!IS_QLA83XX(ha)) + if (!IS_QLA83XX(ha) && !IS_QLA27XX(ha)) goto out; if (ha->port_no == 0) diff --git a/drivers/scsi/qla2xxx/qla_target.c b/drivers/scsi/qla2xxx/qla_target.c index bce2bea0..9c14d10 100644 --- a/drivers/scsi/qla2xxx/qla_target.c +++ b/drivers/scsi/qla2xxx/qla_target.c @@ -5787,7 +5787,7 @@ qlt_probe_one_stage1(struct scsi_qla_host *base_vha, struct qla_hw_data *ha) if (!QLA_TGT_MODE_ENABLED()) return; - if (ha->mqenable || IS_QLA83XX(ha)) { + if (ha->mqenable || IS_QLA83XX(ha) || IS_QLA27XX(ha)) { ISP_ATIO_Q_IN(base_vha) = &ha->mqiobase->isp25mq.atio_q_in; ISP_ATIO_Q_OUT(base_vha) = &ha->mqiobase->isp25mq.atio_q_out; } else { -- cgit v0.10.2 From 3761f3e8704878b1d3db9751f5c6db6fc918032f Mon Sep 17 00:00:00 2001 From: Quinn Tran Date: Wed, 10 Jun 2015 11:05:19 -0400 Subject: qla2xxx: Add flush after updating ATIOQ consumer index. After updating the consumer index of ATIO Q, a read is required to flush the write to the adapter register. Signed-off-by: Quinn Tran Signed-off-by: Himanshu Madhani Reviewed-by: Nicholas Bellinger Signed-off-by: Nicholas Bellinger diff --git a/drivers/scsi/qla2xxx/qla_target.c b/drivers/scsi/qla2xxx/qla_target.c index 9c14d10..5047c85 100644 --- a/drivers/scsi/qla2xxx/qla_target.c +++ b/drivers/scsi/qla2xxx/qla_target.c @@ -5546,6 +5546,7 @@ qlt_24xx_process_atio_queue(struct scsi_qla_host *vha) /* Adjust ring index */ WRT_REG_DWORD(ISP_ATIO_Q_OUT(vha), ha->tgt.atio_ring_index); + RD_REG_DWORD_RELAXED(ISP_ATIO_Q_OUT(vha)); } void -- cgit v0.10.2 From 810e30bc4658e9c069577bde52394a5af872803c Mon Sep 17 00:00:00 2001 From: Quinn Tran Date: Wed, 10 Jun 2015 11:05:20 -0400 Subject: qla2xxx: release request queue reservation. Request IOCB queue element(s) is reserved during good path IO. Under error condition such as unable to allocate IOCB handle condition, the IOCB count that was reserved is not released. Cc: # v3.18+ Signed-off-by: Quinn Tran Signed-off-by: Himanshu Madhani Reviewed-by: Nicholas Bellinger Signed-off-by: Nicholas Bellinger diff --git a/drivers/scsi/qla2xxx/qla_target.c b/drivers/scsi/qla2xxx/qla_target.c index 5047c85..49a184d 100644 --- a/drivers/scsi/qla2xxx/qla_target.c +++ b/drivers/scsi/qla2xxx/qla_target.c @@ -2339,9 +2339,10 @@ int qlt_xmit_response(struct qla_tgt_cmd *cmd, int xmit_type, res = qlt_build_ctio_crc2_pkt(&prm, vha); else res = qlt_24xx_build_ctio_pkt(&prm, vha); - if (unlikely(res != 0)) + if (unlikely(res != 0)) { + vha->req->cnt += full_req_cnt; goto out_unmap_unlock; - + } pkt = (struct ctio7_to_24xx *)prm.pkt; @@ -2479,8 +2480,11 @@ int qlt_rdy_to_xfer(struct qla_tgt_cmd *cmd) else res = qlt_24xx_build_ctio_pkt(&prm, vha); - if (unlikely(res != 0)) + if (unlikely(res != 0)) { + vha->req->cnt += prm.req_cnt; goto out_unlock_free_unmap; + } + pkt = (struct ctio7_to_24xx *)prm.pkt; pkt->u.status0.flags |= __constant_cpu_to_le16(CTIO7_FLAGS_DATA_OUT | CTIO7_FLAGS_STATUS_MODE_0); -- cgit v0.10.2 From e5fdee875f7918b37f7c3fce3025197d8f214a27 Mon Sep 17 00:00:00 2001 From: Quinn Tran Date: Wed, 10 Jun 2015 11:05:21 -0400 Subject: qla2xxx: adjust debug flags Adjust debug flag to match debug comment. Signed-off-by: Quinn Tran Signed-off-by: Himanshu Madhani Reviewed-by: Nicholas Bellinger Signed-off-by: Nicholas Bellinger diff --git a/drivers/scsi/qla2xxx/qla_target.c b/drivers/scsi/qla2xxx/qla_target.c index 49a184d..cda4434 100644 --- a/drivers/scsi/qla2xxx/qla_target.c +++ b/drivers/scsi/qla2xxx/qla_target.c @@ -3013,7 +3013,7 @@ qlt_abort_cmd_on_host_reset(struct scsi_qla_host *vha, struct qla_tgt_cmd *cmd) dump_stack(); } - cmd->cmd_flags |= BIT_12; + cmd->cmd_flags |= BIT_17; ha->tgt.tgt_ops->free_cmd(cmd); } @@ -3175,7 +3175,7 @@ static void qlt_do_ctio_completion(struct scsi_qla_host *vha, uint32_t handle, skip_term: if (cmd->state == QLA_TGT_STATE_PROCESSED) { - ; + cmd->cmd_flags |= BIT_12; } else if (cmd->state == QLA_TGT_STATE_NEED_DATA) { int rx_status = 0; @@ -3189,9 +3189,11 @@ skip_term: ha->tgt.tgt_ops->handle_data(cmd); return; } else if (cmd->state == QLA_TGT_STATE_ABORTED) { + cmd->cmd_flags |= BIT_18; ql_dbg(ql_dbg_tgt_mgt, vha, 0xf01e, "Aborted command %p (tag %lld) finished\n", cmd, se_cmd->tag); } else { + cmd->cmd_flags |= BIT_19; ql_dbg(ql_dbg_tgt_mgt, vha, 0xf05c, "qla_target(%d): A command in state (%d) should " "not return a CTIO complete\n", vha->vp_idx, cmd->state); @@ -3203,7 +3205,6 @@ skip_term: dump_stack(); } - ha->tgt.tgt_ops->free_cmd(cmd); } diff --git a/drivers/scsi/qla2xxx/qla_target.h b/drivers/scsi/qla2xxx/qla_target.h index 985d76d..dfeeadf 100644 --- a/drivers/scsi/qla2xxx/qla_target.h +++ b/drivers/scsi/qla2xxx/qla_target.h @@ -961,6 +961,9 @@ struct qla_tgt_cmd { * BIT_14 - Back end data received/sent. * BIT_15 - SRR prepare ctio * BIT_16 - complete free + * BIT_17 - flush - qlt_abort_cmd_on_host_reset + * BIT_18 - completion w/abort status + * BIT_19 - completion w/unknown status */ uint32_t cmd_flags; }; diff --git a/drivers/scsi/qla2xxx/tcm_qla2xxx.c b/drivers/scsi/qla2xxx/tcm_qla2xxx.c index 33364b6..32ff9d1 100644 --- a/drivers/scsi/qla2xxx/tcm_qla2xxx.c +++ b/drivers/scsi/qla2xxx/tcm_qla2xxx.c @@ -374,7 +374,7 @@ static int tcm_qla2xxx_write_pending(struct se_cmd *se_cmd) { struct qla_tgt_cmd *cmd = container_of(se_cmd, struct qla_tgt_cmd, se_cmd); - + cmd->cmd_flags |= BIT_3; cmd->bufflen = se_cmd->data_length; cmd->dma_data_direction = target_reverse_dma_direction(se_cmd); @@ -546,7 +546,6 @@ static int tcm_qla2xxx_queue_data_in(struct se_cmd *se_cmd) cmd->sg_cnt = se_cmd->t_data_nents; cmd->sg = se_cmd->t_data_sg; cmd->offset = 0; - cmd->cmd_flags |= BIT_3; cmd->prot_sg_cnt = se_cmd->t_prot_nents; cmd->prot_sg = se_cmd->t_prot_sg; -- cgit v0.10.2 From 6bc85dd595a5438b50ec085668e53ef26058bb90 Mon Sep 17 00:00:00 2001 From: Himanshu Madhani Date: Wed, 10 Jun 2015 11:05:22 -0400 Subject: qla2xxx: Remove msleep in qlt_send_term_exchange Remove unnecessary msleep from qlt_send_term_exchange as it adds latency of 250 msec while sending terminate exchange to an aborted task. Cc: # v3.18+ Signed-off-by: Himanshu Madhani Signed-off-by: Giridhar Malavali Reviewed-by: Nicholas Bellinger Signed-off-by: Nicholas Bellinger diff --git a/drivers/scsi/qla2xxx/qla_target.c b/drivers/scsi/qla2xxx/qla_target.c index cda4434..2491bc9 100644 --- a/drivers/scsi/qla2xxx/qla_target.c +++ b/drivers/scsi/qla2xxx/qla_target.c @@ -2713,7 +2713,7 @@ static int __qlt_send_term_exchange(struct scsi_qla_host *vha, static void qlt_send_term_exchange(struct scsi_qla_host *vha, struct qla_tgt_cmd *cmd, struct atio_from_isp *atio, int ha_locked) { - unsigned long flags; + unsigned long flags = 0; int rc; if (qlt_issue_marker(vha, ha_locked) < 0) @@ -2729,17 +2729,18 @@ static void qlt_send_term_exchange(struct scsi_qla_host *vha, rc = __qlt_send_term_exchange(vha, cmd, atio); if (rc == -ENOMEM) qlt_alloc_qfull_cmd(vha, atio, 0, 0); - spin_unlock_irqrestore(&vha->hw->hardware_lock, flags); done: if (cmd && ((cmd->state != QLA_TGT_STATE_ABORTED) || !cmd->cmd_sent_to_fw)) { - if (!ha_locked && !in_interrupt()) - msleep(250); /* just in case */ - - qlt_unmap_sg(vha, cmd); + if (cmd->sg_mapped) + qlt_unmap_sg(vha, cmd); vha->hw->tgt.tgt_ops->free_cmd(cmd); } + + if (!ha_locked) + spin_unlock_irqrestore(&vha->hw->hardware_lock, flags); + return; } -- cgit v0.10.2 From 9fce12540cb9f91e7f1f539a80b70f0b388bdae0 Mon Sep 17 00:00:00 2001 From: Kanoj Sarcar Date: Wed, 10 Jun 2015 11:05:23 -0400 Subject: qla2xxx: fix command initialization in target mode. Cc: # v3.18+ Signed-off-by: Kanoj Sarcar Signed-off-by: Himanshu Madhani Reviewed-by: Nicholas Bellinger Signed-off-by: Nicholas Bellinger diff --git a/drivers/scsi/qla2xxx/qla_target.c b/drivers/scsi/qla2xxx/qla_target.c index 2491bc9..17d656b 100644 --- a/drivers/scsi/qla2xxx/qla_target.c +++ b/drivers/scsi/qla2xxx/qla_target.c @@ -3345,6 +3345,11 @@ static struct qla_tgt_cmd *qlt_get_tag(scsi_qla_host_t *vha, cmd->loop_id = sess->loop_id; cmd->conf_compl_supported = sess->conf_compl_supported; + cmd->cmd_flags = 0; + cmd->jiffies_at_alloc = get_jiffies_64(); + + cmd->reset_count = vha->hw->chip_reset; + return cmd; } @@ -3451,11 +3456,6 @@ static int qlt_handle_cmd_for_atio(struct scsi_qla_host *vha, return -ENOMEM; } - cmd->cmd_flags = 0; - cmd->jiffies_at_alloc = get_jiffies_64(); - - cmd->reset_count = vha->hw->chip_reset; - cmd->cmd_in_wq = 1; cmd->cmd_flags |= BIT_0; INIT_WORK(&cmd->work, qlt_do_work); -- cgit v0.10.2 From b2032fd567326ad0b2d443bb6d96d2580ec670a5 Mon Sep 17 00:00:00 2001 From: Roland Dreier Date: Tue, 14 Jul 2015 16:00:42 -0400 Subject: qla2xxx: kill sessions/log out initiator on RSCN and port down events To fix some issues talking to ESX, this patch modifies the qla2xxx driver so that it never logs into remote ports. This has the side effect of getting rid of the "rports" entirely, which means we never log out of initiators and never tear down sessions when an initiator goes away. This is mostly OK, except that we can run into trouble if we have initiator A assigned FC address X:Y:Z by the fabric talking to us, and then initiator A goes away. Some time (could be a long time) later, initiator B comes along and also gets FC address X:Y:Z (which is available again, because initiator A is gone). If initiator B starts talking to us, then we'll still have the session for initiator A, and since we look up incoming IO based on the FC address X:Y:Z, initiator B will end up using ACLs for initiator A. Fix this by: 1. Handling RSCN events somewhat differently; instead of completely skipping the processing of fcports, we look through the list, and if an fcport disappears, we tell the target code the tear down the session and tell the HBA FW to release the N_Port handle. 2. Handling "port down" events by flushing all of our sessions. The firmware was already releasing the N_Port handle but we want the target code to drop all the sessions too. Cc: # v3.18+ Signed-off-by: Roland Dreier Signed-off-by: Alexei Potashnik Acked-by: Quinn Tran Signed-off-by: Himanshu Madhani Signed-off-by: Nicholas Bellinger diff --git a/drivers/scsi/qla2xxx/qla_dbg.c b/drivers/scsi/qla2xxx/qla_dbg.c index 0e6ee3c..e9ae6b9 100644 --- a/drivers/scsi/qla2xxx/qla_dbg.c +++ b/drivers/scsi/qla2xxx/qla_dbg.c @@ -68,7 +68,7 @@ * | | | 0xd101-0xd1fe | * | | | 0xd214-0xd2fe | * | Target Mode | 0xe079 | | - * | Target Mode Management | 0xf072 | 0xf002 | + * | Target Mode Management | 0xf080 | 0xf002 | * | | | 0xf046-0xf049 | * | Target Mode Task Management | 0x1000b | | * ---------------------------------------------------------------------- diff --git a/drivers/scsi/qla2xxx/qla_init.c b/drivers/scsi/qla2xxx/qla_init.c index 17c3bd9..e166c46 100644 --- a/drivers/scsi/qla2xxx/qla_init.c +++ b/drivers/scsi/qla2xxx/qla_init.c @@ -3464,20 +3464,43 @@ qla2x00_configure_fabric(scsi_qla_host_t *vha) if ((fcport->flags & FCF_FABRIC_DEVICE) == 0) continue; - if (fcport->scan_state == QLA_FCPORT_SCAN && - atomic_read(&fcport->state) == FCS_ONLINE) { - qla2x00_mark_device_lost(vha, fcport, - ql2xplogiabsentdevice, 0); - if (fcport->loop_id != FC_NO_LOOP_ID && - (fcport->flags & FCF_FCP2_DEVICE) == 0 && - fcport->port_type != FCT_INITIATOR && - fcport->port_type != FCT_BROADCAST) { - ha->isp_ops->fabric_logout(vha, - fcport->loop_id, - fcport->d_id.b.domain, - fcport->d_id.b.area, - fcport->d_id.b.al_pa); - qla2x00_clear_loop_id(fcport); + if (fcport->scan_state == QLA_FCPORT_SCAN) { + if (qla_ini_mode_enabled(base_vha) && + atomic_read(&fcport->state) == FCS_ONLINE) { + qla2x00_mark_device_lost(vha, fcport, + ql2xplogiabsentdevice, 0); + if (fcport->loop_id != FC_NO_LOOP_ID && + (fcport->flags & FCF_FCP2_DEVICE) == 0 && + fcport->port_type != FCT_INITIATOR && + fcport->port_type != FCT_BROADCAST) { + ha->isp_ops->fabric_logout(vha, + fcport->loop_id, + fcport->d_id.b.domain, + fcport->d_id.b.area, + fcport->d_id.b.al_pa); + qla2x00_clear_loop_id(fcport); + } + } else if (!qla_ini_mode_enabled(base_vha)) { + /* + * In target mode, explicitly kill + * sessions and log out of devices + * that are gone, so that we don't + * end up with an initiator using the + * wrong ACL (if the fabric recycles + * an FC address and we have a stale + * session around) and so that we don't + * report initiators that are no longer + * on the fabric. + */ + ql_dbg(ql_dbg_tgt_mgt, vha, 0xf077, + "port gone, logging out/killing session: " + "%8phC state 0x%x flags 0x%x fc4_type 0x%x " + "scan_state %d\n", + fcport->port_name, + atomic_read(&fcport->state), + fcport->flags, fcport->fc4_type, + fcport->scan_state); + qlt_fc_port_deleted(vha, fcport); } } } @@ -3498,6 +3521,28 @@ qla2x00_configure_fabric(scsi_qla_host_t *vha) (fcport->flags & FCF_LOGIN_NEEDED) == 0) continue; + /* + * If we're not an initiator, skip looking for devices + * and logging in. There's no reason for us to do it, + * and it seems to actively cause problems in target + * mode if we race with the initiator logging into us + * (we might get the "port ID used" status back from + * our login command and log out the initiator, which + * seems to cause havoc). + */ + if (!qla_ini_mode_enabled(base_vha)) { + if (fcport->scan_state == QLA_FCPORT_FOUND) { + ql_dbg(ql_dbg_tgt_mgt, vha, 0xf078, + "port %8phC state 0x%x flags 0x%x fc4_type 0x%x " + "scan_state %d (initiator mode disabled; skipping " + "login)\n", fcport->port_name, + atomic_read(&fcport->state), + fcport->flags, fcport->fc4_type, + fcport->scan_state); + } + continue; + } + if (fcport->loop_id == FC_NO_LOOP_ID) { fcport->loop_id = next_loopid; rval = qla2x00_find_new_loop_id( @@ -3524,16 +3569,38 @@ qla2x00_configure_fabric(scsi_qla_host_t *vha) test_bit(LOOP_RESYNC_NEEDED, &vha->dpc_flags)) break; - /* Find a new loop ID to use. */ - fcport->loop_id = next_loopid; - rval = qla2x00_find_new_loop_id(base_vha, fcport); - if (rval != QLA_SUCCESS) { - /* Ran out of IDs to use */ - break; - } + /* + * If we're not an initiator, skip looking for devices + * and logging in. There's no reason for us to do it, + * and it seems to actively cause problems in target + * mode if we race with the initiator logging into us + * (we might get the "port ID used" status back from + * our login command and log out the initiator, which + * seems to cause havoc). + */ + if (qla_ini_mode_enabled(base_vha)) { + /* Find a new loop ID to use. */ + fcport->loop_id = next_loopid; + rval = qla2x00_find_new_loop_id(base_vha, + fcport); + if (rval != QLA_SUCCESS) { + /* Ran out of IDs to use */ + break; + } - /* Login and update database */ - qla2x00_fabric_dev_login(vha, fcport, &next_loopid); + /* Login and update database */ + qla2x00_fabric_dev_login(vha, fcport, + &next_loopid); + } else { + ql_dbg(ql_dbg_tgt_mgt, vha, 0xf079, + "new port %8phC state 0x%x flags 0x%x fc4_type " + "0x%x scan_state %d (initiator mode disabled; " + "skipping login)\n", + fcport->port_name, + atomic_read(&fcport->state), + fcport->flags, fcport->fc4_type, + fcport->scan_state); + } list_move_tail(&fcport->list, &vha->vp_fcports); } @@ -3729,11 +3796,12 @@ qla2x00_find_all_fabric_devs(scsi_qla_host_t *vha, fcport->fp_speed = new_fcport->fp_speed; /* - * If address the same and state FCS_ONLINE, nothing - * changed. + * If address the same and state FCS_ONLINE + * (or in target mode), nothing changed. */ if (fcport->d_id.b24 == new_fcport->d_id.b24 && - atomic_read(&fcport->state) == FCS_ONLINE) { + (atomic_read(&fcport->state) == FCS_ONLINE || + !qla_ini_mode_enabled(base_vha))) { break; } @@ -3753,6 +3821,22 @@ qla2x00_find_all_fabric_devs(scsi_qla_host_t *vha, * Log it out if still logged in and mark it for * relogin later. */ + if (!qla_ini_mode_enabled(base_vha)) { + ql_dbg(ql_dbg_tgt_mgt, vha, 0xf080, + "port changed FC ID, %8phC" + " old %x:%x:%x (loop_id 0x%04x)-> new %x:%x:%x\n", + fcport->port_name, + fcport->d_id.b.domain, + fcport->d_id.b.area, + fcport->d_id.b.al_pa, + fcport->loop_id, + new_fcport->d_id.b.domain, + new_fcport->d_id.b.area, + new_fcport->d_id.b.al_pa); + fcport->d_id.b24 = new_fcport->d_id.b24; + break; + } + fcport->d_id.b24 = new_fcport->d_id.b24; fcport->flags |= FCF_LOGIN_NEEDED; if (fcport->loop_id != FC_NO_LOOP_ID && @@ -3772,6 +3856,7 @@ qla2x00_find_all_fabric_devs(scsi_qla_host_t *vha, if (found) continue; /* If device was not in our fcports list, then add it. */ + new_fcport->scan_state = QLA_FCPORT_FOUND; list_add_tail(&new_fcport->list, new_fcports); /* Allocate a new replacement fcport. */ diff --git a/drivers/scsi/qla2xxx/qla_target.c b/drivers/scsi/qla2xxx/qla_target.c index 17d656b..d9dd354 100644 --- a/drivers/scsi/qla2xxx/qla_target.c +++ b/drivers/scsi/qla2xxx/qla_target.c @@ -113,6 +113,7 @@ static void qlt_abort_cmd_on_host_reset(struct scsi_qla_host *vha, static void qlt_alloc_qfull_cmd(struct scsi_qla_host *vha, struct atio_from_isp *atio, uint16_t status, int qfull); static void qlt_disable_vha(struct scsi_qla_host *vha); +static void qlt_clear_tgt_db(struct qla_tgt *tgt); /* * Global Variables */ @@ -431,10 +432,10 @@ static int qlt_reset(struct scsi_qla_host *vha, void *iocb, int mcmd) loop_id = le16_to_cpu(n->u.isp24.nport_handle); if (loop_id == 0xFFFF) { -#if 0 /* FIXME: Re-enable Global event handling.. */ /* Global event */ - atomic_inc(&ha->tgt.qla_tgt->tgt_global_resets_count); - qlt_clear_tgt_db(ha->tgt.qla_tgt); + atomic_inc(&vha->vha_tgt.qla_tgt->tgt_global_resets_count); + qlt_clear_tgt_db(vha->vha_tgt.qla_tgt); +#if 0 /* FIXME: do we need to choose a session here? */ if (!list_empty(&ha->tgt.qla_tgt->sess_list)) { sess = list_entry(ha->tgt.qla_tgt->sess_list.next, typeof(*sess), sess_list_entry); @@ -788,7 +789,7 @@ void qlt_fc_port_deleted(struct scsi_qla_host *vha, fc_port_t *fcport) if (!vha->hw->tgt.tgt_ops) return; - if (!tgt || (fcport->port_type != FCT_INITIATOR)) + if (!tgt) return; if (tgt->tgt_stop) { -- cgit v0.10.2 From 8b2f5ff3d05c2c48b722c3cc67b8226f1601042b Mon Sep 17 00:00:00 2001 From: Swapnil Nagle Date: Tue, 14 Jul 2015 16:00:43 -0400 Subject: qla2xxx: cleanup cmd in qla workqueue before processing TMR Since cmds go into qla_tgt_wq and TMRs don't, it's possible that TMR like TASK_ABORT can be queued over the cmd for which it was meant. To avoid this race, use a per-port list to keep track of cmds that are enqueued to qla_tgt_wq but not yet processed. When a TMR arrives, iterate through this list and remove any cmds that match the TMR. This patch supports TASK_ABORT and LUN_RESET. Cc: # v3.18+ Signed-off-by: Swapnil Nagle Signed-off-by: Alexei Potashnik Acked-by: Quinn Tran Signed-off-by: Himanshu Madhani Signed-off-by: Nicholas Bellinger diff --git a/drivers/scsi/qla2xxx/qla_dbg.c b/drivers/scsi/qla2xxx/qla_dbg.c index e9ae6b9..e63aa07 100644 --- a/drivers/scsi/qla2xxx/qla_dbg.c +++ b/drivers/scsi/qla2xxx/qla_dbg.c @@ -68,7 +68,7 @@ * | | | 0xd101-0xd1fe | * | | | 0xd214-0xd2fe | * | Target Mode | 0xe079 | | - * | Target Mode Management | 0xf080 | 0xf002 | + * | Target Mode Management | 0xf083 | 0xf002 | * | | | 0xf046-0xf049 | * | Target Mode Task Management | 0x1000b | | * ---------------------------------------------------------------------- diff --git a/drivers/scsi/qla2xxx/qla_def.h b/drivers/scsi/qla2xxx/qla_def.h index 787f901..d9c4ed2 100644 --- a/drivers/scsi/qla2xxx/qla_def.h +++ b/drivers/scsi/qla2xxx/qla_def.h @@ -3579,6 +3579,11 @@ typedef struct scsi_qla_host { uint16_t fcoe_fcf_idx; uint8_t fcoe_vn_port_mac[6]; + /* list of commands waiting on workqueue */ + struct list_head qla_cmd_list; + struct list_head qla_sess_op_cmd_list; + spinlock_t cmd_list_lock; + uint32_t vp_abort_cnt; struct fc_vport *fc_vport; /* holds fc_vport * for each vport */ diff --git a/drivers/scsi/qla2xxx/qla_os.c b/drivers/scsi/qla2xxx/qla_os.c index 7c7528b..7ce395a 100644 --- a/drivers/scsi/qla2xxx/qla_os.c +++ b/drivers/scsi/qla2xxx/qla_os.c @@ -3764,8 +3764,11 @@ struct scsi_qla_host *qla2x00_create_host(struct scsi_host_template *sht, INIT_LIST_HEAD(&vha->vp_fcports); INIT_LIST_HEAD(&vha->work_list); INIT_LIST_HEAD(&vha->list); + INIT_LIST_HEAD(&vha->qla_cmd_list); + INIT_LIST_HEAD(&vha->qla_sess_op_cmd_list); spin_lock_init(&vha->work_lock); + spin_lock_init(&vha->cmd_list_lock); sprintf(vha->host_str, "%s_%ld", QLA2XXX_DRIVER_NAME, vha->host_no); ql_dbg(ql_dbg_init, vha, 0x0041, diff --git a/drivers/scsi/qla2xxx/qla_target.c b/drivers/scsi/qla2xxx/qla_target.c index d9dd354..eca4448 100644 --- a/drivers/scsi/qla2xxx/qla_target.c +++ b/drivers/scsi/qla2xxx/qla_target.c @@ -1170,6 +1170,70 @@ static void qlt_24xx_retry_term_exchange(struct scsi_qla_host *vha, FCP_TMF_CMPL, true); } +static int abort_cmd_for_tag(struct scsi_qla_host *vha, uint32_t tag) +{ + struct qla_tgt_sess_op *op; + struct qla_tgt_cmd *cmd; + + spin_lock(&vha->cmd_list_lock); + + list_for_each_entry(op, &vha->qla_sess_op_cmd_list, cmd_list) { + if (tag == op->atio.u.isp24.exchange_addr) { + op->aborted = true; + spin_unlock(&vha->cmd_list_lock); + return 1; + } + } + + list_for_each_entry(cmd, &vha->qla_cmd_list, cmd_list) { + if (tag == cmd->atio.u.isp24.exchange_addr) { + cmd->state = QLA_TGT_STATE_ABORTED; + spin_unlock(&vha->cmd_list_lock); + return 1; + } + } + + spin_unlock(&vha->cmd_list_lock); + return 0; +} + +/* drop cmds for the given lun + * XXX only looks for cmds on the port through which lun reset was recieved + * XXX does not go through the list of other port (which may have cmds + * for the same lun) + */ +static void abort_cmds_for_lun(struct scsi_qla_host *vha, + uint32_t lun, uint8_t *s_id) +{ + struct qla_tgt_sess_op *op; + struct qla_tgt_cmd *cmd; + uint32_t key; + + key = sid_to_key(s_id); + spin_lock(&vha->cmd_list_lock); + list_for_each_entry(op, &vha->qla_sess_op_cmd_list, cmd_list) { + uint32_t op_key; + uint32_t op_lun; + + op_key = sid_to_key(op->atio.u.isp24.fcp_hdr.s_id); + op_lun = scsilun_to_int( + (struct scsi_lun *)&op->atio.u.isp24.fcp_cmnd.lun); + if (op_key == key && op_lun == lun) + op->aborted = true; + } + list_for_each_entry(cmd, &vha->qla_cmd_list, cmd_list) { + uint32_t cmd_key; + uint32_t cmd_lun; + + cmd_key = sid_to_key(cmd->atio.u.isp24.fcp_hdr.s_id); + cmd_lun = scsilun_to_int( + (struct scsi_lun *)&cmd->atio.u.isp24.fcp_cmnd.lun); + if (cmd_key == key && cmd_lun == lun) + cmd->state = QLA_TGT_STATE_ABORTED; + } + spin_unlock(&vha->cmd_list_lock); +} + /* ha->hardware_lock supposed to be held on entry */ static int __qlt_24xx_handle_abts(struct scsi_qla_host *vha, struct abts_recv_from_24xx *abts, struct qla_tgt_sess *sess) @@ -1194,8 +1258,19 @@ static int __qlt_24xx_handle_abts(struct scsi_qla_host *vha, } spin_unlock(&se_sess->sess_cmd_lock); - if (!found_lun) - return -ENOENT; + /* cmd not in LIO lists, look in qla list */ + if (!found_lun) { + if (abort_cmd_for_tag(vha, abts->exchange_addr_to_abort)) { + /* send TASK_ABORT response immediately */ + qlt_24xx_send_abts_resp(vha, abts, FCP_TMF_CMPL, false); + return 0; + } else { + ql_dbg(ql_dbg_tgt_mgt, vha, 0xf081, + "unable to find cmd in driver or LIO for tag 0x%x\n", + abts->exchange_addr_to_abort); + return -ENOENT; + } + } ql_dbg(ql_dbg_tgt_mgt, vha, 0xf00f, "qla_target(%d): task abort (tag=%d)\n", @@ -3264,6 +3339,13 @@ static void __qlt_do_work(struct qla_tgt_cmd *cmd) if (tgt->tgt_stop) goto out_term; + if (cmd->state == QLA_TGT_STATE_ABORTED) { + ql_dbg(ql_dbg_tgt_mgt, vha, 0xf082, + "cmd with tag %u is aborted\n", + cmd->atio.u.isp24.exchange_addr); + goto out_term; + } + cdb = &atio->u.isp24.fcp_cmnd.cdb[0]; cmd->se_cmd.tag = atio->u.isp24.exchange_addr; cmd->unpacked_lun = scsilun_to_int( @@ -3317,6 +3399,12 @@ out_term: static void qlt_do_work(struct work_struct *work) { struct qla_tgt_cmd *cmd = container_of(work, struct qla_tgt_cmd, work); + scsi_qla_host_t *vha = cmd->vha; + unsigned long flags; + + spin_lock_irqsave(&vha->cmd_list_lock, flags); + list_del(&cmd->cmd_list); + spin_unlock_irqrestore(&vha->cmd_list_lock, flags); __qlt_do_work(cmd); } @@ -3368,14 +3456,25 @@ static void qlt_create_sess_from_atio(struct work_struct *work) unsigned long flags; uint8_t *s_id = op->atio.u.isp24.fcp_hdr.s_id; + spin_lock_irqsave(&vha->cmd_list_lock, flags); + list_del(&op->cmd_list); + spin_unlock_irqrestore(&vha->cmd_list_lock, flags); + + if (op->aborted) { + ql_dbg(ql_dbg_tgt_mgt, vha, 0xf083, + "sess_op with tag %u is aborted\n", + op->atio.u.isp24.exchange_addr); + goto out_term; + } + ql_dbg(ql_dbg_tgt_mgt, vha, 0xf022, - "qla_target(%d): Unable to find wwn login" - " (s_id %x:%x:%x), trying to create it manually\n", - vha->vp_idx, s_id[0], s_id[1], s_id[2]); + "qla_target(%d): Unable to find wwn login" + " (s_id %x:%x:%x), trying to create it manually\n", + vha->vp_idx, s_id[0], s_id[1], s_id[2]); if (op->atio.u.raw.entry_count > 1) { ql_dbg(ql_dbg_tgt_mgt, vha, 0xf023, - "Dropping multy entry atio %p\n", &op->atio); + "Dropping multy entry atio %p\n", &op->atio); goto out_term; } @@ -3440,6 +3539,11 @@ static int qlt_handle_cmd_for_atio(struct scsi_qla_host *vha, memcpy(&op->atio, atio, sizeof(*atio)); op->vha = vha; + + spin_lock(&vha->cmd_list_lock); + list_add_tail(&op->cmd_list, &vha->qla_sess_op_cmd_list); + spin_unlock(&vha->cmd_list_lock); + INIT_WORK(&op->work, qlt_create_sess_from_atio); queue_work(qla_tgt_wq, &op->work); return 0; @@ -3459,6 +3563,11 @@ static int qlt_handle_cmd_for_atio(struct scsi_qla_host *vha, cmd->cmd_in_wq = 1; cmd->cmd_flags |= BIT_0; + + spin_lock(&vha->cmd_list_lock); + list_add_tail(&cmd->cmd_list, &vha->qla_cmd_list); + spin_unlock(&vha->cmd_list_lock); + INIT_WORK(&cmd->work, qlt_do_work); queue_work(qla_tgt_wq, &cmd->work); return 0; @@ -3472,6 +3581,7 @@ static int qlt_issue_task_mgmt(struct qla_tgt_sess *sess, uint32_t lun, struct scsi_qla_host *vha = sess->vha; struct qla_hw_data *ha = vha->hw; struct qla_tgt_mgmt_cmd *mcmd; + struct atio_from_isp *a = (struct atio_from_isp *)iocb; int res; uint8_t tmr_func; @@ -3512,6 +3622,7 @@ static int qlt_issue_task_mgmt(struct qla_tgt_sess *sess, uint32_t lun, ql_dbg(ql_dbg_tgt_tmr, vha, 0x10002, "qla_target(%d): LUN_RESET received\n", sess->vha->vp_idx); tmr_func = TMR_LUN_RESET; + abort_cmds_for_lun(vha, lun, a->u.isp24.fcp_hdr.s_id); break; case QLA_TGT_CLEAR_TS: diff --git a/drivers/scsi/qla2xxx/qla_target.h b/drivers/scsi/qla2xxx/qla_target.h index dfeeadf..3104335 100644 --- a/drivers/scsi/qla2xxx/qla_target.h +++ b/drivers/scsi/qla2xxx/qla_target.h @@ -874,6 +874,8 @@ struct qla_tgt_sess_op { struct scsi_qla_host *vha; struct atio_from_isp atio; struct work_struct work; + struct list_head cmd_list; + bool aborted; }; /* @@ -1076,6 +1078,16 @@ static inline void qla_reverse_ini_mode(struct scsi_qla_host *ha) ha->host->active_mode |= MODE_INITIATOR; } +static inline uint32_t sid_to_key(const uint8_t *s_id) +{ + uint32_t key; + + key = (((unsigned long)s_id[0] << 16) | + ((unsigned long)s_id[1] << 8) | + (unsigned long)s_id[2]); + return key; +} + /* * Exported symbols from qla_target.c LLD logic used by qla2xxx code.. */ diff --git a/drivers/scsi/qla2xxx/tcm_qla2xxx.c b/drivers/scsi/qla2xxx/tcm_qla2xxx.c index 32ff9d1..9ad9b69 100644 --- a/drivers/scsi/qla2xxx/tcm_qla2xxx.c +++ b/drivers/scsi/qla2xxx/tcm_qla2xxx.c @@ -1148,9 +1148,7 @@ static struct qla_tgt_sess *tcm_qla2xxx_find_sess_by_s_id( return NULL; } - key = (((unsigned long)s_id[0] << 16) | - ((unsigned long)s_id[1] << 8) | - (unsigned long)s_id[2]); + key = sid_to_key(s_id); pr_debug("find_sess_by_s_id: 0x%06x\n", key); se_nacl = btree_lookup32(&lport->lport_fcport_map, key); @@ -1185,9 +1183,7 @@ static void tcm_qla2xxx_set_sess_by_s_id( void *slot; int rc; - key = (((unsigned long)s_id[0] << 16) | - ((unsigned long)s_id[1] << 8) | - (unsigned long)s_id[2]); + key = sid_to_key(s_id); pr_debug("set_sess_by_s_id: %06x\n", key); slot = btree_lookup32(&lport->lport_fcport_map, key); -- cgit v0.10.2 From a6ca88780dd66b0700d89419abd17b6b4bb49483 Mon Sep 17 00:00:00 2001 From: Alexei Potashnik Date: Tue, 14 Jul 2015 16:00:44 -0400 Subject: qla2xxx: delay plogi/prli ack until existing sessions are deleted - keep qla_tgt_sess object on the session list until it's freed - modify use of sess->deleted flag to differentiate delayed session deletion that can be cancelled from irreversible one: QLA_SESS_DELETION_PENDING vs QLA_SESS_DELETION_IN_PROGRESS - during IN_PROGRESS deletion all newly arrived commands and TMRs will be rejected, existing commands and TMRs will be terminated when given by the core to the fabric or simply dropped if session logout has already happened (logout terminates all existing exchanges) - new PLOGI will initiate deletion of the following sessions (unless deletion is already IN_PROGRESS): - with the same port_name (with logout) - different port_name, different loop_id but the same port_id (with logout) - different port_name, different port_id, but the same loop_id (without logout) - additionally each new PLOGI will store imm notify iocb in the same port_name session being deleted. When deletion process completes this iocb will be acked. Only the most recent PLOGI iocb is stored. The older ones will be terminated when replaced. - new PRLI will initiate deletion of the following sessions (unless deletion is already IN_PROGRESS): - different port_name, different port_id, but the same loop_id (without logout) Cc: # v3.18+ Signed-off-by: Alexei Potashnik Acked-by: Quinn Tran Signed-off-by: Himanshu Madhani Signed-off-by: Nicholas Bellinger diff --git a/drivers/scsi/qla2xxx/qla_dbg.c b/drivers/scsi/qla2xxx/qla_dbg.c index e63aa07..05793b7 100644 --- a/drivers/scsi/qla2xxx/qla_dbg.c +++ b/drivers/scsi/qla2xxx/qla_dbg.c @@ -67,10 +67,10 @@ * | | | 0xd031-0xd0ff | * | | | 0xd101-0xd1fe | * | | | 0xd214-0xd2fe | - * | Target Mode | 0xe079 | | - * | Target Mode Management | 0xf083 | 0xf002 | + * | Target Mode | 0xe080 | | + * | Target Mode Management | 0xf091 | 0xf002 | * | | | 0xf046-0xf049 | - * | Target Mode Task Management | 0x1000b | | + * | Target Mode Task Management | 0x1000d | | * ---------------------------------------------------------------------- */ diff --git a/drivers/scsi/qla2xxx/qla_def.h b/drivers/scsi/qla2xxx/qla_def.h index d9c4ed2..08a869f 100644 --- a/drivers/scsi/qla2xxx/qla_def.h +++ b/drivers/scsi/qla2xxx/qla_def.h @@ -274,6 +274,7 @@ #define RESPONSE_ENTRY_CNT_FX00 256 /* Number of response entries.*/ struct req_que; +struct qla_tgt_sess; /* * (sd.h is not exported, hence local inclusion) @@ -2026,6 +2027,7 @@ typedef struct fc_port { uint16_t port_id; unsigned long retry_delay_timestamp; + struct qla_tgt_sess *tgt_session; } fc_port_t; #include "qla_mr.h" diff --git a/drivers/scsi/qla2xxx/qla_init.c b/drivers/scsi/qla2xxx/qla_init.c index e166c46..bda7a0d 100644 --- a/drivers/scsi/qla2xxx/qla_init.c +++ b/drivers/scsi/qla2xxx/qla_init.c @@ -115,6 +115,8 @@ qla2x00_async_iocb_timeout(void *data) QLA_LOGIO_LOGIN_RETRIED : 0; qla2x00_post_async_login_done_work(fcport->vha, fcport, lio->u.logio.data); + } else if (sp->type == SRB_LOGOUT_CMD) { + qlt_logo_completion_handler(fcport, QLA_FUNCTION_TIMEOUT); } } @@ -497,7 +499,10 @@ void qla2x00_async_logout_done(struct scsi_qla_host *vha, fc_port_t *fcport, uint16_t *data) { - qla2x00_mark_device_lost(vha, fcport, 1, 0); + /* Don't re-login in target mode */ + if (!fcport->tgt_session) + qla2x00_mark_device_lost(vha, fcport, 1, 0); + qlt_logo_completion_handler(fcport, data[0]); return; } diff --git a/drivers/scsi/qla2xxx/qla_iocb.c b/drivers/scsi/qla2xxx/qla_iocb.c index 36fbd4c..6f02b26 100644 --- a/drivers/scsi/qla2xxx/qla_iocb.c +++ b/drivers/scsi/qla2xxx/qla_iocb.c @@ -1943,6 +1943,9 @@ qla24xx_logout_iocb(srb_t *sp, struct logio_entry_24xx *logio) logio->entry_type = LOGINOUT_PORT_IOCB_TYPE; logio->control_flags = cpu_to_le16(LCF_COMMAND_LOGO|LCF_IMPL_LOGO); + if (!sp->fcport->tgt_session || + !sp->fcport->tgt_session->keep_nport_handle) + logio->control_flags |= cpu_to_le16(LCF_FREE_NPORT); logio->nport_handle = cpu_to_le16(sp->fcport->loop_id); logio->port_id[0] = sp->fcport->d_id.b.al_pa; logio->port_id[1] = sp->fcport->d_id.b.area; diff --git a/drivers/scsi/qla2xxx/qla_target.c b/drivers/scsi/qla2xxx/qla_target.c index eca4448..acb2a50 100644 --- a/drivers/scsi/qla2xxx/qla_target.c +++ b/drivers/scsi/qla2xxx/qla_target.c @@ -114,6 +114,10 @@ static void qlt_alloc_qfull_cmd(struct scsi_qla_host *vha, struct atio_from_isp *atio, uint16_t status, int qfull); static void qlt_disable_vha(struct scsi_qla_host *vha); static void qlt_clear_tgt_db(struct qla_tgt *tgt); +static void qlt_send_notify_ack(struct scsi_qla_host *vha, + struct imm_ntfy_from_isp *ntfy, + uint32_t add_flags, uint16_t resp_code, int resp_code_valid, + uint16_t srr_flags, uint16_t srr_reject_code, uint8_t srr_explan); /* * Global Variables */ @@ -382,14 +386,73 @@ static void qlt_free_session_done(struct work_struct *work) struct qla_tgt *tgt = sess->tgt; struct scsi_qla_host *vha = sess->vha; struct qla_hw_data *ha = vha->hw; + unsigned long flags; + bool logout_started = false; + fc_port_t fcport; + + ql_dbg(ql_dbg_tgt_mgt, vha, 0xf084, + "%s: se_sess %p / sess %p from port %8phC loop_id %#04x" + " s_id %02x:%02x:%02x logout %d keep %d plogi %d\n", + __func__, sess->se_sess, sess, sess->port_name, sess->loop_id, + sess->s_id.b.domain, sess->s_id.b.area, sess->s_id.b.al_pa, + sess->logout_on_delete, sess->keep_nport_handle, + sess->plogi_ack_needed); BUG_ON(!tgt); + + if (sess->logout_on_delete) { + int rc; + + memset(&fcport, 0, sizeof(fcport)); + fcport.loop_id = sess->loop_id; + fcport.d_id = sess->s_id; + memcpy(fcport.port_name, sess->port_name, WWN_SIZE); + fcport.vha = vha; + fcport.tgt_session = sess; + + rc = qla2x00_post_async_logout_work(vha, &fcport, NULL); + if (rc != QLA_SUCCESS) + ql_log(ql_log_warn, vha, 0xf085, + "Schedule logo failed sess %p rc %d\n", + sess, rc); + else + logout_started = true; + } + /* * Release the target session for FC Nexus from fabric module code. */ if (sess->se_sess != NULL) ha->tgt.tgt_ops->free_session(sess); + if (logout_started) { + bool traced = false; + + while (!ACCESS_ONCE(sess->logout_completed)) { + if (!traced) { + ql_dbg(ql_dbg_tgt_mgt, vha, 0xf086, + "%s: waiting for sess %p logout\n", + __func__, sess); + traced = true; + } + msleep(100); + } + + ql_dbg(ql_dbg_tgt_mgt, vha, 0xf087, + "%s: sess %p logout completed\n", + __func__, sess); + } + + spin_lock_irqsave(&ha->hardware_lock, flags); + + if (sess->plogi_ack_needed) + qlt_send_notify_ack(vha, &sess->tm_iocb, + 0, 0, 0, 0, 0, 0); + + list_del(&sess->sess_list_entry); + + spin_unlock_irqrestore(&ha->hardware_lock, flags); + ql_dbg(ql_dbg_tgt_mgt, vha, 0xf001, "Unregistration of sess %p finished\n", sess); @@ -410,9 +473,9 @@ void qlt_unreg_sess(struct qla_tgt_sess *sess) vha->hw->tgt.tgt_ops->clear_nacl_from_fcport_map(sess); - list_del(&sess->sess_list_entry); - if (sess->deleted) - list_del(&sess->del_list_entry); + if (!list_empty(&sess->del_list_entry)) + list_del_init(&sess->del_list_entry); + sess->deleted = QLA_SESS_DELETION_IN_PROGRESS; INIT_WORK(&sess->free_work, qlt_free_session_done); schedule_work(&sess->free_work); @@ -490,27 +553,36 @@ static void qlt_schedule_sess_for_deletion(struct qla_tgt_sess *sess, struct qla_tgt *tgt = sess->tgt; uint32_t dev_loss_tmo = tgt->ha->port_down_retry_count + 5; - if (sess->deleted) - return; + if (sess->deleted) { + /* Upgrade to unconditional deletion in case it was temporary */ + if (immediate && sess->deleted == QLA_SESS_DELETION_PENDING) + list_del(&sess->del_list_entry); + else + return; + } ql_dbg(ql_dbg_tgt, sess->vha, 0xe001, "Scheduling sess %p for deletion\n", sess); - list_add_tail(&sess->del_list_entry, &tgt->del_sess_list); - sess->deleted = 1; - if (immediate) + if (immediate) { dev_loss_tmo = 0; + sess->deleted = QLA_SESS_DELETION_IN_PROGRESS; + list_add(&sess->del_list_entry, &tgt->del_sess_list); + } else { + sess->deleted = QLA_SESS_DELETION_PENDING; + list_add_tail(&sess->del_list_entry, &tgt->del_sess_list); + } sess->expires = jiffies + dev_loss_tmo * HZ; ql_dbg(ql_dbg_tgt, sess->vha, 0xe048, "qla_target(%d): session for port %8phC (loop ID %d) scheduled for " - "deletion in %u secs (expires: %lu) immed: %d\n", + "deletion in %u secs (expires: %lu) immed: %d, logout: %d\n", sess->vha->vp_idx, sess->port_name, sess->loop_id, dev_loss_tmo, - sess->expires, immediate); + sess->expires, immediate, sess->logout_on_delete); if (immediate) - schedule_delayed_work(&tgt->sess_del_work, 0); + mod_delayed_work(system_wq, &tgt->sess_del_work, 0); else schedule_delayed_work(&tgt->sess_del_work, sess->expires - jiffies); @@ -579,9 +651,9 @@ out_free_id_list: /* ha->hardware_lock supposed to be held on entry */ static void qlt_undelete_sess(struct qla_tgt_sess *sess) { - BUG_ON(!sess->deleted); + BUG_ON(sess->deleted != QLA_SESS_DELETION_PENDING); - list_del(&sess->del_list_entry); + list_del_init(&sess->del_list_entry); sess->deleted = 0; } @@ -600,7 +672,9 @@ static void qlt_del_sess_work_fn(struct delayed_work *work) del_list_entry); elapsed = jiffies; if (time_after_eq(elapsed, sess->expires)) { - qlt_undelete_sess(sess); + /* No turning back */ + list_del_init(&sess->del_list_entry); + sess->deleted = QLA_SESS_DELETION_IN_PROGRESS; ql_dbg(ql_dbg_tgt_mgt, vha, 0xf004, "Timeout: sess %p about to be deleted\n", @@ -644,6 +718,13 @@ static struct qla_tgt_sess *qlt_create_sess( fcport->d_id.b.al_pa, fcport->d_id.b.area, fcport->loop_id); + /* Cannot undelete at this point */ + if (sess->deleted == QLA_SESS_DELETION_IN_PROGRESS) { + spin_unlock_irqrestore(&ha->hardware_lock, + flags); + return NULL; + } + if (sess->deleted) qlt_undelete_sess(sess); @@ -674,6 +755,14 @@ static struct qla_tgt_sess *qlt_create_sess( sess->s_id = fcport->d_id; sess->loop_id = fcport->loop_id; sess->local = local; + INIT_LIST_HEAD(&sess->del_list_entry); + + /* Under normal circumstances we want to logout from firmware when + * session eventually ends and release corresponding nport handle. + * In the exception cases (e.g. when new PLOGI is waiting) corresponding + * code will adjust these flags as necessary. */ + sess->logout_on_delete = 1; + sess->keep_nport_handle = 0; ql_dbg(ql_dbg_tgt_mgt, vha, 0xf006, "Adding sess %p to tgt %p via ->check_initiator_node_acl()\n", @@ -751,6 +840,10 @@ void qlt_fc_port_added(struct scsi_qla_host *vha, fc_port_t *fcport) mutex_unlock(&vha->vha_tgt.tgt_mutex); spin_lock_irqsave(&ha->hardware_lock, flags); + } else if (sess->deleted == QLA_SESS_DELETION_IN_PROGRESS) { + /* Point of no return */ + spin_unlock_irqrestore(&ha->hardware_lock, flags); + return; } else { kref_get(&sess->se_sess->sess_kref); @@ -2371,6 +2464,19 @@ int qlt_xmit_response(struct qla_tgt_cmd *cmd, int xmit_type, unsigned long flags = 0; int res; + spin_lock_irqsave(&ha->hardware_lock, flags); + if (cmd->sess && cmd->sess->deleted == QLA_SESS_DELETION_IN_PROGRESS) { + cmd->state = QLA_TGT_STATE_PROCESSED; + if (cmd->sess->logout_completed) + /* no need to terminate. FW already freed exchange. */ + qlt_abort_cmd_on_host_reset(cmd->vha, cmd); + else + qlt_send_term_exchange(vha, cmd, &cmd->atio, 1); + spin_unlock_irqrestore(&ha->hardware_lock, flags); + return 0; + } + spin_unlock_irqrestore(&ha->hardware_lock, flags); + memset(&prm, 0, sizeof(prm)); qlt_check_srr_debug(cmd, &xmit_type); @@ -2532,7 +2638,8 @@ int qlt_rdy_to_xfer(struct qla_tgt_cmd *cmd) spin_lock_irqsave(&ha->hardware_lock, flags); - if (qla2x00_reset_active(vha) || cmd->reset_count != ha->chip_reset) { + if (qla2x00_reset_active(vha) || (cmd->reset_count != ha->chip_reset) || + (cmd->sess && cmd->sess->deleted == QLA_SESS_DELETION_IN_PROGRESS)) { /* * Either a chip reset is active or this request was from * previous life, just abort the processing. @@ -2725,6 +2832,89 @@ out: /* If hardware_lock held on entry, might drop it, then reaquire */ /* This function sends the appropriate CTIO to ISP 2xxx or 24xx */ +static int __qlt_send_term_imm_notif(struct scsi_qla_host *vha, + struct imm_ntfy_from_isp *ntfy) +{ + struct nack_to_isp *nack; + struct qla_hw_data *ha = vha->hw; + request_t *pkt; + int ret = 0; + + ql_dbg(ql_dbg_tgt_tmr, vha, 0xe01c, + "Sending TERM ELS CTIO (ha=%p)\n", ha); + + pkt = (request_t *)qla2x00_alloc_iocbs_ready(vha, NULL); + if (pkt == NULL) { + ql_dbg(ql_dbg_tgt, vha, 0xe080, + "qla_target(%d): %s failed: unable to allocate " + "request packet\n", vha->vp_idx, __func__); + return -ENOMEM; + } + + pkt->entry_type = NOTIFY_ACK_TYPE; + pkt->entry_count = 1; + pkt->handle = QLA_TGT_SKIP_HANDLE | CTIO_COMPLETION_HANDLE_MARK; + + nack = (struct nack_to_isp *)pkt; + nack->ox_id = ntfy->ox_id; + + nack->u.isp24.nport_handle = ntfy->u.isp24.nport_handle; + if (le16_to_cpu(ntfy->u.isp24.status) == IMM_NTFY_ELS) { + nack->u.isp24.flags = ntfy->u.isp24.flags & + __constant_cpu_to_le32(NOTIFY24XX_FLAGS_PUREX_IOCB); + } + + /* terminate */ + nack->u.isp24.flags |= + __constant_cpu_to_le16(NOTIFY_ACK_FLAGS_TERMINATE); + + nack->u.isp24.srr_rx_id = ntfy->u.isp24.srr_rx_id; + nack->u.isp24.status = ntfy->u.isp24.status; + nack->u.isp24.status_subcode = ntfy->u.isp24.status_subcode; + nack->u.isp24.fw_handle = ntfy->u.isp24.fw_handle; + nack->u.isp24.exchange_address = ntfy->u.isp24.exchange_address; + nack->u.isp24.srr_rel_offs = ntfy->u.isp24.srr_rel_offs; + nack->u.isp24.srr_ui = ntfy->u.isp24.srr_ui; + nack->u.isp24.vp_index = ntfy->u.isp24.vp_index; + + qla2x00_start_iocbs(vha, vha->req); + return ret; +} + +static void qlt_send_term_imm_notif(struct scsi_qla_host *vha, + struct imm_ntfy_from_isp *imm, int ha_locked) +{ + unsigned long flags = 0; + int rc; + + if (qlt_issue_marker(vha, ha_locked) < 0) + return; + + if (ha_locked) { + rc = __qlt_send_term_imm_notif(vha, imm); + +#if 0 /* Todo */ + if (rc == -ENOMEM) + qlt_alloc_qfull_cmd(vha, imm, 0, 0); +#endif + goto done; + } + + spin_lock_irqsave(&vha->hw->hardware_lock, flags); + rc = __qlt_send_term_imm_notif(vha, imm); + +#if 0 /* Todo */ + if (rc == -ENOMEM) + qlt_alloc_qfull_cmd(vha, imm, 0, 0); +#endif + +done: + if (!ha_locked) + spin_unlock_irqrestore(&vha->hw->hardware_lock, flags); +} + +/* If hardware_lock held on entry, might drop it, then reaquire */ +/* This function sends the appropriate CTIO to ISP 2xxx or 24xx */ static int __qlt_send_term_exchange(struct scsi_qla_host *vha, struct qla_tgt_cmd *cmd, struct atio_from_isp *atio) @@ -3776,22 +3966,237 @@ static int qlt_abort_task(struct scsi_qla_host *vha, return __qlt_abort_task(vha, iocb, sess); } +void qlt_logo_completion_handler(fc_port_t *fcport, int rc) +{ + if (fcport->tgt_session) { + if (rc != MBS_COMMAND_COMPLETE) { + ql_dbg(ql_dbg_tgt_mgt, fcport->vha, 0xf088, + "%s: se_sess %p / sess %p from" + " port %8phC loop_id %#04x s_id %02x:%02x:%02x" + " LOGO failed: %#x\n", + __func__, + fcport->tgt_session->se_sess, + fcport->tgt_session, + fcport->port_name, fcport->loop_id, + fcport->d_id.b.domain, fcport->d_id.b.area, + fcport->d_id.b.al_pa, rc); + } + + fcport->tgt_session->logout_completed = 1; + } +} + +static void qlt_swap_imm_ntfy_iocb(struct imm_ntfy_from_isp *a, + struct imm_ntfy_from_isp *b) +{ + struct imm_ntfy_from_isp tmp; + memcpy(&tmp, a, sizeof(struct imm_ntfy_from_isp)); + memcpy(a, b, sizeof(struct imm_ntfy_from_isp)); + memcpy(b, &tmp, sizeof(struct imm_ntfy_from_isp)); +} + +/* +* ha->hardware_lock supposed to be held on entry (to protect tgt->sess_list) +* +* Schedules sessions with matching port_id/loop_id but different wwn for +* deletion. Returns existing session with matching wwn if present. +* Null otherwise. +*/ +static struct qla_tgt_sess * +qlt_find_sess_invalidate_other(struct qla_tgt *tgt, uint64_t wwn, + port_id_t port_id, uint16_t loop_id) +{ + struct qla_tgt_sess *sess = NULL, *other_sess; + uint64_t other_wwn; + + list_for_each_entry(other_sess, &tgt->sess_list, sess_list_entry) { + + other_wwn = wwn_to_u64(other_sess->port_name); + + if (wwn == other_wwn) { + WARN_ON(sess); + sess = other_sess; + continue; + } + + /* find other sess with nport_id collision */ + if (port_id.b24 == other_sess->s_id.b24) { + if (loop_id != other_sess->loop_id) { + ql_dbg(ql_dbg_tgt_tmr, tgt->vha, 0x1000c, + "Invalidating sess %p loop_id %d wwn %llx.\n", + other_sess, other_sess->loop_id, other_wwn); + + /* + * logout_on_delete is set by default, but another + * session that has the same s_id/loop_id combo + * might have cleared it when requested this session + * deletion, so don't touch it + */ + qlt_schedule_sess_for_deletion(other_sess, true); + } else { + /* + * Another wwn used to have our s_id/loop_id + * combo - kill the session, but don't log out + */ + sess->logout_on_delete = 0; + qlt_schedule_sess_for_deletion(other_sess, + true); + } + continue; + } + + /* find other sess with nport handle collision */ + if (loop_id == other_sess->loop_id) { + ql_dbg(ql_dbg_tgt_tmr, tgt->vha, 0x1000d, + "Invalidating sess %p loop_id %d wwn %llx.\n", + other_sess, other_sess->loop_id, other_wwn); + + /* Same loop_id but different s_id + * Ok to kill and logout */ + qlt_schedule_sess_for_deletion(other_sess, true); + } + } + + return sess; +} + /* * ha->hardware_lock supposed to be held on entry. Might drop it, then reaquire */ static int qlt_24xx_handle_els(struct scsi_qla_host *vha, struct imm_ntfy_from_isp *iocb) { + struct qla_tgt *tgt = vha->vha_tgt.qla_tgt; + struct qla_tgt_sess *sess = NULL; + uint64_t wwn; + port_id_t port_id; + uint16_t loop_id; + uint16_t wd3_lo; int res = 0; + wwn = wwn_to_u64(iocb->u.isp24.port_name); + + port_id.b.domain = iocb->u.isp24.port_id[2]; + port_id.b.area = iocb->u.isp24.port_id[1]; + port_id.b.al_pa = iocb->u.isp24.port_id[0]; + port_id.b.rsvd_1 = 0; + + loop_id = le16_to_cpu(iocb->u.isp24.nport_handle); + ql_dbg(ql_dbg_tgt_mgt, vha, 0xf026, "qla_target(%d): Port ID: 0x%3phC ELS opcode: 0x%02x\n", vha->vp_idx, iocb->u.isp24.port_id, iocb->u.isp24.status_subcode); + /* res = 1 means ack at the end of thread + * res = 0 means ack async/later. + */ switch (iocb->u.isp24.status_subcode) { case ELS_PLOGI: - case ELS_FLOGI: + + if (wwn) + sess = qlt_find_sess_invalidate_other(tgt, wwn, + port_id, loop_id); + + if (!sess || IS_SW_RESV_ADDR(sess->s_id)) { + res = 1; + break; + } + + if (sess->plogi_ack_needed) { + /* + * Initiator sent another PLOGI before last PLOGI could + * finish. Swap plogi iocbs and terminate old one + * without acking, new one will get acked when session + * deletion completes. + */ + ql_log(ql_log_warn, sess->vha, 0xf089, + "sess %p received double plogi.\n", sess); + + qlt_swap_imm_ntfy_iocb(iocb, &sess->tm_iocb); + + qlt_send_term_imm_notif(vha, iocb, 1); + + res = 0; + break; + } + + res = 0; + + /* + * Save immediate Notif IOCB for Ack when sess is done + * and being deleted. + */ + memcpy(&sess->tm_iocb, iocb, sizeof(sess->tm_iocb)); + sess->plogi_ack_needed = 1; + + /* + * Under normal circumstances we want to release nport handle + * during LOGO process to avoid nport handle leaks inside FW. + * The exception is when LOGO is done while another PLOGI with + * the same nport handle is waiting as might be the case here. + * Note: there is always a possibily of a race where session + * deletion has already started for other reasons (e.g. ACL + * removal) and now PLOGI arrives: + * 1. if PLOGI arrived in FW after nport handle has been freed, + * FW must have assigned this PLOGI a new/same handle and we + * can proceed ACK'ing it as usual when session deletion + * completes. + * 2. if PLOGI arrived in FW before LOGO with LCF_FREE_NPORT + * bit reached it, the handle has now been released. We'll + * get an error when we ACK this PLOGI. Nothing will be sent + * back to initiator. Initiator should eventually retry + * PLOGI and situation will correct itself. + */ + sess->keep_nport_handle = ((sess->loop_id == loop_id) && + (sess->s_id.b24 == port_id.b24)); + qlt_schedule_sess_for_deletion(sess, true); + break; + case ELS_PRLI: + wd3_lo = le16_to_cpu(iocb->u.isp24.u.prli.wd3_lo); + + if (wwn) + sess = qlt_find_sess_invalidate_other(tgt, wwn, port_id, + loop_id); + + if (sess != NULL) { + if (sess->deleted) { + /* + * Impatient initiator sent PRLI before last + * PLOGI could finish. Will force him to re-try, + * while last one finishes. + */ + ql_log(ql_log_warn, sess->vha, 0xf090, + "sess %p PRLI received, before plogi ack.\n", + sess); + qlt_send_term_imm_notif(vha, iocb, 1); + res = 0; + break; + } + + /* + * This shouldn't happen under normal circumstances, + * since we have deleted the old session during PLOGI + */ + ql_dbg(ql_dbg_tgt_mgt, vha, 0xf091, + "PRLI (loop_id %#04x) for existing sess %p (loop_id %#04x)\n", + sess->loop_id, sess, iocb->u.isp24.nport_handle); + + sess->local = 0; + sess->loop_id = loop_id; + sess->s_id = port_id; + + if (wd3_lo & BIT_7) + sess->conf_compl_supported = 1; + + res = 1; + } else { + /* todo: else - create sess here. */ + res = 1; /* send notify ack */ + } + + break; + case ELS_LOGO: case ELS_PRLO: res = qlt_reset(vha, iocb, QLA_TGT_NEXUS_LOSS_SESS); @@ -3809,6 +4214,7 @@ static int qlt_24xx_handle_els(struct scsi_qla_host *vha, break; } + case ELS_FLOGI: /* should never happen */ default: ql_dbg(ql_dbg_tgt_mgt, vha, 0xf061, "qla_target(%d): Unsupported ELS command %x " diff --git a/drivers/scsi/qla2xxx/qla_target.h b/drivers/scsi/qla2xxx/qla_target.h index 3104335..165efb5 100644 --- a/drivers/scsi/qla2xxx/qla_target.h +++ b/drivers/scsi/qla2xxx/qla_target.h @@ -167,7 +167,24 @@ struct imm_ntfy_from_isp { uint32_t srr_rel_offs; uint16_t srr_ui; uint16_t srr_ox_id; - uint8_t reserved_4[19]; + union { + struct { + uint8_t node_name[8]; + } plogi; /* PLOGI/ADISC/PDISC */ + struct { + /* PRLI word 3 bit 0-15 */ + uint16_t wd3_lo; + uint8_t resv0[6]; + } prli; + struct { + uint8_t port_id[3]; + uint8_t resv1; + uint16_t nport_handle; + uint16_t resv2; + } req_els; + } u; + uint8_t port_name[8]; + uint8_t resv3[3]; uint8_t vp_index; uint32_t reserved_5; uint8_t port_id[3]; @@ -234,6 +251,7 @@ struct nack_to_isp { uint8_t reserved[2]; uint16_t ox_id; } __packed; +#define NOTIFY_ACK_FLAGS_TERMINATE BIT_3 #define NOTIFY_ACK_SRR_FLAGS_ACCEPT 0 #define NOTIFY_ACK_SRR_FLAGS_REJECT 1 @@ -878,6 +896,13 @@ struct qla_tgt_sess_op { bool aborted; }; +enum qla_sess_deletion { + QLA_SESS_DELETION_NONE = 0, + QLA_SESS_DELETION_PENDING = 1, /* hopefully we can get rid of + * this one */ + QLA_SESS_DELETION_IN_PROGRESS = 2, +}; + /* * Equivilant to IT Nexus (Initiator-Target) */ @@ -886,8 +911,13 @@ struct qla_tgt_sess { port_id_t s_id; unsigned int conf_compl_supported:1; - unsigned int deleted:1; + unsigned int deleted:2; unsigned int local:1; + unsigned int logout_on_delete:1; + unsigned int plogi_ack_needed:1; + unsigned int keep_nport_handle:1; + + unsigned char logout_completed; struct se_session *se_sess; struct scsi_qla_host *vha; @@ -899,6 +929,10 @@ struct qla_tgt_sess { uint8_t port_name[WWN_SIZE]; struct work_struct free_work; + + union { + struct imm_ntfy_from_isp tm_iocb; + }; }; struct qla_tgt_cmd { @@ -1031,6 +1065,10 @@ struct qla_tgt_srr_ctio { struct qla_tgt_cmd *cmd; }; +/* Check for Switch reserved address */ +#define IS_SW_RESV_ADDR(_s_id) \ + ((_s_id.b.domain == 0xff) && (_s_id.b.area == 0xfc)) + #define QLA_TGT_XMIT_DATA 1 #define QLA_TGT_XMIT_STATUS 2 #define QLA_TGT_XMIT_ALL (QLA_TGT_XMIT_STATUS|QLA_TGT_XMIT_DATA) @@ -1124,5 +1162,6 @@ extern void qlt_stop_phase2(struct qla_tgt *); extern irqreturn_t qla83xx_msix_atio_q(int, void *); extern void qlt_83xx_iospace_config(struct qla_hw_data *); extern int qlt_free_qfull_cmds(struct scsi_qla_host *); +extern void qlt_logo_completion_handler(fc_port_t *, int); #endif /* __QLA_TARGET_H */ diff --git a/drivers/scsi/qla2xxx/tcm_qla2xxx.c b/drivers/scsi/qla2xxx/tcm_qla2xxx.c index 9ad9b69..4e242c7 100644 --- a/drivers/scsi/qla2xxx/tcm_qla2xxx.c +++ b/drivers/scsi/qla2xxx/tcm_qla2xxx.c @@ -1539,6 +1539,10 @@ static void tcm_qla2xxx_update_sess(struct qla_tgt_sess *sess, port_id_t s_id, } sess->conf_compl_supported = conf_compl_supported; + + /* Reset logout parameters to default */ + sess->logout_on_delete = 1; + sess->keep_nport_handle = 0; } /* -- cgit v0.10.2 From daddf5cf9b5c68b81b2bb7133f1dd0fda4552d0b Mon Sep 17 00:00:00 2001 From: Alexei Potashnik Date: Tue, 14 Jul 2015 16:00:45 -0400 Subject: qla2xxx: Abort stale cmds on qla_tgt_wq when plogi arrives cancel any commands from initiator's s_id that are still waiting on qla_tgt_wq when PLOGI arrives. Cc: # v3.18+ Signed-off-by: Alexei Potashnik Acked-by: Quinn Tran Signed-off-by: Himanshu Madhani Signed-off-by: Nicholas Bellinger diff --git a/drivers/scsi/qla2xxx/qla_target.c b/drivers/scsi/qla2xxx/qla_target.c index acb2a50..0b08beb 100644 --- a/drivers/scsi/qla2xxx/qla_target.c +++ b/drivers/scsi/qla2xxx/qla_target.c @@ -4060,6 +4060,38 @@ qlt_find_sess_invalidate_other(struct qla_tgt *tgt, uint64_t wwn, return sess; } +/* Abort any commands for this s_id waiting on qla_tgt_wq workqueue */ +static int abort_cmds_for_s_id(struct scsi_qla_host *vha, port_id_t *s_id) +{ + struct qla_tgt_sess_op *op; + struct qla_tgt_cmd *cmd; + uint32_t key; + int count = 0; + + key = (((u32)s_id->b.domain << 16) | + ((u32)s_id->b.area << 8) | + ((u32)s_id->b.al_pa)); + + spin_lock(&vha->cmd_list_lock); + list_for_each_entry(op, &vha->qla_sess_op_cmd_list, cmd_list) { + uint32_t op_key = sid_to_key(op->atio.u.isp24.fcp_hdr.s_id); + if (op_key == key) { + op->aborted = true; + count++; + } + } + list_for_each_entry(cmd, &vha->qla_cmd_list, cmd_list) { + uint32_t cmd_key = sid_to_key(cmd->atio.u.isp24.fcp_hdr.s_id); + if (cmd_key == key) { + cmd->state = QLA_TGT_STATE_ABORTED; + count++; + } + } + spin_unlock(&vha->cmd_list_lock); + + return count; +} + /* * ha->hardware_lock supposed to be held on entry. Might drop it, then reaquire */ @@ -4093,6 +4125,9 @@ static int qlt_24xx_handle_els(struct scsi_qla_host *vha, switch (iocb->u.isp24.status_subcode) { case ELS_PLOGI: + /* Mark all stale commands in qla_tgt_wq for deletion */ + abort_cmds_for_s_id(vha, &port_id); + if (wwn) sess = qlt_find_sess_invalidate_other(tgt, wwn, port_id, loop_id); -- cgit v0.10.2 From df673274fa4896f25f0bf348d2a3535d74b4cbec Mon Sep 17 00:00:00 2001 From: Alexei Potashnik Date: Tue, 14 Jul 2015 16:00:46 -0400 Subject: qla2xxx: added sess generations to detect RSCN update races RSCN processing in qla2xxx driver can run in parallel with ELS/IO processing. As such the decision to remove disappeared fc port's session could be stale, because a new login sequence has occurred since and created a brand new session. Previous mechanism of dealing with this by delaying deletion request was prone to erroneous deletions if the event that was supposed to cancel the deletion never arrived or has been delayed in processing. New mechanism relies on a time-like generation counter to serialize RSCN updates relative to ELS/IO updates. Cc: # v3.18+ Signed-off-by: Alexei Potashnik Signed-off-by: Himanshu Madhani Signed-off-by: Nicholas Bellinger diff --git a/drivers/scsi/qla2xxx/qla_dbg.c b/drivers/scsi/qla2xxx/qla_dbg.c index 05793b7..8b011ae 100644 --- a/drivers/scsi/qla2xxx/qla_dbg.c +++ b/drivers/scsi/qla2xxx/qla_dbg.c @@ -68,7 +68,7 @@ * | | | 0xd101-0xd1fe | * | | | 0xd214-0xd2fe | * | Target Mode | 0xe080 | | - * | Target Mode Management | 0xf091 | 0xf002 | + * | Target Mode Management | 0xf096 | 0xf002 | * | | | 0xf046-0xf049 | * | Target Mode Task Management | 0x1000d | | * ---------------------------------------------------------------------- diff --git a/drivers/scsi/qla2xxx/qla_def.h b/drivers/scsi/qla2xxx/qla_def.h index 08a869f..9ad819e 100644 --- a/drivers/scsi/qla2xxx/qla_def.h +++ b/drivers/scsi/qla2xxx/qla_def.h @@ -3586,6 +3586,11 @@ typedef struct scsi_qla_host { struct list_head qla_sess_op_cmd_list; spinlock_t cmd_list_lock; + /* Counter to detect races between ELS and RSCN events */ + atomic_t generation_tick; + /* Time when global fcport update has been scheduled */ + int total_fcport_update_gen; + uint32_t vp_abort_cnt; struct fc_vport *fc_vport; /* holds fc_vport * for each vport */ diff --git a/drivers/scsi/qla2xxx/qla_init.c b/drivers/scsi/qla2xxx/qla_init.c index bda7a0d..506621d 100644 --- a/drivers/scsi/qla2xxx/qla_init.c +++ b/drivers/scsi/qla2xxx/qla_init.c @@ -2927,24 +2927,14 @@ qla2x00_rport_del(void *data) { fc_port_t *fcport = data; struct fc_rport *rport; - scsi_qla_host_t *vha = fcport->vha; unsigned long flags; - unsigned long vha_flags; spin_lock_irqsave(fcport->vha->host->host_lock, flags); rport = fcport->drport ? fcport->drport: fcport->rport; fcport->drport = NULL; spin_unlock_irqrestore(fcport->vha->host->host_lock, flags); - if (rport) { + if (rport) fc_remote_port_delete(rport); - /* - * Release the target mode FC NEXUS in qla_target.c code - * if target mod is enabled. - */ - spin_lock_irqsave(&vha->hw->hardware_lock, vha_flags); - qlt_fc_port_deleted(vha, fcport); - spin_unlock_irqrestore(&vha->hw->hardware_lock, vha_flags); - } } /** @@ -3384,6 +3374,7 @@ qla2x00_configure_fabric(scsi_qla_host_t *vha) LIST_HEAD(new_fcports); struct qla_hw_data *ha = vha->hw; struct scsi_qla_host *base_vha = pci_get_drvdata(ha->pdev); + int discovery_gen; /* If FL port exists, then SNS is present */ if (IS_FWI2_CAPABLE(ha)) @@ -3454,6 +3445,14 @@ qla2x00_configure_fabric(scsi_qla_host_t *vha) fcport->scan_state = QLA_FCPORT_SCAN; } + /* Mark the time right before querying FW for connected ports. + * This process is long, asynchronous and by the time it's done, + * collected information might not be accurate anymore. E.g. + * disconnected port might have re-connected and a brand new + * session has been created. In this case session's generation + * will be newer than discovery_gen. */ + qlt_do_generation_tick(vha, &discovery_gen); + rval = qla2x00_find_all_fabric_devs(vha, &new_fcports); if (rval != QLA_SUCCESS) break; @@ -3505,7 +3504,8 @@ qla2x00_configure_fabric(scsi_qla_host_t *vha) atomic_read(&fcport->state), fcport->flags, fcport->fc4_type, fcport->scan_state); - qlt_fc_port_deleted(vha, fcport); + qlt_fc_port_deleted(vha, fcport, + discovery_gen); } } } @@ -4282,6 +4282,14 @@ qla2x00_update_fcports(scsi_qla_host_t *base_vha) atomic_read(&fcport->state) != FCS_UNCONFIGURED) { spin_unlock_irqrestore(&ha->vport_slock, flags); qla2x00_rport_del(fcport); + + /* + * Release the target mode FC NEXUS in + * qla_target.c, if target mod is enabled. + */ + qlt_fc_port_deleted(vha, fcport, + base_vha->total_fcport_update_gen); + spin_lock_irqsave(&ha->vport_slock, flags); } } diff --git a/drivers/scsi/qla2xxx/qla_os.c b/drivers/scsi/qla2xxx/qla_os.c index 7ce395a..be6de2a 100644 --- a/drivers/scsi/qla2xxx/qla_os.c +++ b/drivers/scsi/qla2xxx/qla_os.c @@ -3230,11 +3230,14 @@ qla2x00_schedule_rport_del(struct scsi_qla_host *vha, fc_port_t *fcport, spin_lock_irqsave(vha->host->host_lock, flags); fcport->drport = rport; spin_unlock_irqrestore(vha->host->host_lock, flags); + qlt_do_generation_tick(vha, &base_vha->total_fcport_update_gen); set_bit(FCPORT_UPDATE_NEEDED, &base_vha->dpc_flags); qla2xxx_wake_dpc(base_vha); } else { + int now; fc_remote_port_delete(rport); - qlt_fc_port_deleted(vha, fcport); + qlt_do_generation_tick(vha, &now); + qlt_fc_port_deleted(vha, fcport, now); } } diff --git a/drivers/scsi/qla2xxx/qla_target.c b/drivers/scsi/qla2xxx/qla_target.c index 0b08beb..0b5bd9c 100644 --- a/drivers/scsi/qla2xxx/qla_target.c +++ b/drivers/scsi/qla2xxx/qla_target.c @@ -127,6 +127,16 @@ static struct workqueue_struct *qla_tgt_wq; static DEFINE_MUTEX(qla_tgt_mutex); static LIST_HEAD(qla_tgt_glist); +/* This API intentionally takes dest as a parameter, rather than returning + * int value to avoid caller forgetting to issue wmb() after the store */ +void qlt_do_generation_tick(struct scsi_qla_host *vha, int *dest) +{ + scsi_qla_host_t *base_vha = pci_get_drvdata(vha->hw->pdev); + *dest = atomic_inc_return(&base_vha->generation_tick); + /* memory barrier */ + wmb(); +} + /* ha->hardware_lock supposed to be held on entry (to protect tgt->sess_list) */ static struct qla_tgt_sess *qlt_find_sess_by_port_name( struct qla_tgt *tgt, @@ -576,10 +586,12 @@ static void qlt_schedule_sess_for_deletion(struct qla_tgt_sess *sess, sess->expires = jiffies + dev_loss_tmo * HZ; ql_dbg(ql_dbg_tgt, sess->vha, 0xe048, - "qla_target(%d): session for port %8phC (loop ID %d) scheduled for " - "deletion in %u secs (expires: %lu) immed: %d, logout: %d\n", - sess->vha->vp_idx, sess->port_name, sess->loop_id, dev_loss_tmo, - sess->expires, immediate, sess->logout_on_delete); + "qla_target(%d): session for port %8phC (loop ID %d s_id %02x:%02x:%02x)" + " scheduled for deletion in %u secs (expires: %lu) immed: %d, logout: %d, gen: %#x\n", + sess->vha->vp_idx, sess->port_name, sess->loop_id, + sess->s_id.b.domain, sess->s_id.b.area, sess->s_id.b.al_pa, + dev_loss_tmo, sess->expires, immediate, sess->logout_on_delete, + sess->generation); if (immediate) mod_delayed_work(system_wq, &tgt->sess_del_work, 0); @@ -734,6 +746,9 @@ static struct qla_tgt_sess *qlt_create_sess( if (sess->local && !local) sess->local = 0; + + qlt_do_generation_tick(vha, &sess->generation); + spin_unlock_irqrestore(&ha->hardware_lock, flags); return sess; @@ -795,6 +810,7 @@ static struct qla_tgt_sess *qlt_create_sess( spin_lock_irqsave(&ha->hardware_lock, flags); list_add_tail(&sess->sess_list_entry, &vha->vha_tgt.qla_tgt->sess_list); vha->vha_tgt.qla_tgt->sess_count++; + qlt_do_generation_tick(vha, &sess->generation); spin_unlock_irqrestore(&ha->hardware_lock, flags); ql_dbg(ql_dbg_tgt_mgt, vha, 0xf04b, @@ -808,7 +824,7 @@ static struct qla_tgt_sess *qlt_create_sess( } /* - * Called from drivers/scsi/qla2xxx/qla_init.c:qla2x00_reg_remote_port() + * Called from qla2x00_reg_remote_port() */ void qlt_fc_port_added(struct scsi_qla_host *vha, fc_port_t *fcport) { @@ -874,7 +890,12 @@ void qlt_fc_port_added(struct scsi_qla_host *vha, fc_port_t *fcport) spin_unlock_irqrestore(&ha->hardware_lock, flags); } -void qlt_fc_port_deleted(struct scsi_qla_host *vha, fc_port_t *fcport) +/* + * max_gen - specifies maximum session generation + * at which this deletion requestion is still valid + */ +void +qlt_fc_port_deleted(struct scsi_qla_host *vha, fc_port_t *fcport, int max_gen) { struct qla_tgt *tgt = vha->vha_tgt.qla_tgt; struct qla_tgt_sess *sess; @@ -893,6 +914,15 @@ void qlt_fc_port_deleted(struct scsi_qla_host *vha, fc_port_t *fcport) return; } + if (max_gen - sess->generation < 0) { + ql_dbg(ql_dbg_tgt_mgt, vha, 0xf092, + "Ignoring stale deletion request for se_sess %p / sess %p" + " for port %8phC, req_gen %d, sess_gen %d\n", + sess->se_sess, sess, sess->port_name, max_gen, + sess->generation); + return; + } + ql_dbg(ql_dbg_tgt_mgt, vha, 0xf008, "qla_tgt_fc_port_deleted %p", sess); sess->local = 1; @@ -3970,7 +4000,7 @@ void qlt_logo_completion_handler(fc_port_t *fcport, int rc) { if (fcport->tgt_session) { if (rc != MBS_COMMAND_COMPLETE) { - ql_dbg(ql_dbg_tgt_mgt, fcport->vha, 0xf088, + ql_dbg(ql_dbg_tgt_mgt, fcport->vha, 0xf093, "%s: se_sess %p / sess %p from" " port %8phC loop_id %#04x s_id %02x:%02x:%02x" " LOGO failed: %#x\n", @@ -4099,6 +4129,7 @@ static int qlt_24xx_handle_els(struct scsi_qla_host *vha, struct imm_ntfy_from_isp *iocb) { struct qla_tgt *tgt = vha->vha_tgt.qla_tgt; + struct qla_hw_data *ha = vha->hw; struct qla_tgt_sess *sess = NULL; uint64_t wwn; port_id_t port_id; @@ -4144,7 +4175,7 @@ static int qlt_24xx_handle_els(struct scsi_qla_host *vha, * without acking, new one will get acked when session * deletion completes. */ - ql_log(ql_log_warn, sess->vha, 0xf089, + ql_log(ql_log_warn, sess->vha, 0xf094, "sess %p received double plogi.\n", sess); qlt_swap_imm_ntfy_iocb(iocb, &sess->tm_iocb); @@ -4201,7 +4232,7 @@ static int qlt_24xx_handle_els(struct scsi_qla_host *vha, * PLOGI could finish. Will force him to re-try, * while last one finishes. */ - ql_log(ql_log_warn, sess->vha, 0xf090, + ql_log(ql_log_warn, sess->vha, 0xf095, "sess %p PRLI received, before plogi ack.\n", sess); qlt_send_term_imm_notif(vha, iocb, 1); @@ -4213,7 +4244,7 @@ static int qlt_24xx_handle_els(struct scsi_qla_host *vha, * This shouldn't happen under normal circumstances, * since we have deleted the old session during PLOGI */ - ql_dbg(ql_dbg_tgt_mgt, vha, 0xf091, + ql_dbg(ql_dbg_tgt_mgt, vha, 0xf096, "PRLI (loop_id %#04x) for existing sess %p (loop_id %#04x)\n", sess->loop_id, sess, iocb->u.isp24.nport_handle); @@ -4224,7 +4255,14 @@ static int qlt_24xx_handle_els(struct scsi_qla_host *vha, if (wd3_lo & BIT_7) sess->conf_compl_supported = 1; - res = 1; + } + res = 1; /* send notify ack */ + + /* Make session global (not used in fabric mode) */ + if (ha->current_topology != ISP_CFG_F) { + set_bit(LOOP_RESYNC_NEEDED, &vha->dpc_flags); + set_bit(LOCAL_LOOP_UPDATE, &vha->dpc_flags); + qla2xxx_wake_dpc(vha); } else { /* todo: else - create sess here. */ res = 1; /* send notify ack */ diff --git a/drivers/scsi/qla2xxx/qla_target.h b/drivers/scsi/qla2xxx/qla_target.h index 165efb5..2ceb60f 100644 --- a/drivers/scsi/qla2xxx/qla_target.h +++ b/drivers/scsi/qla2xxx/qla_target.h @@ -919,6 +919,8 @@ struct qla_tgt_sess { unsigned char logout_completed; + int generation; + struct se_session *se_sess; struct scsi_qla_host *vha; struct qla_tgt *tgt; @@ -1086,7 +1088,7 @@ extern int qlt_lport_register(void *, u64, u64, u64, extern void qlt_lport_deregister(struct scsi_qla_host *); extern void qlt_unreg_sess(struct qla_tgt_sess *); extern void qlt_fc_port_added(struct scsi_qla_host *, fc_port_t *); -extern void qlt_fc_port_deleted(struct scsi_qla_host *, fc_port_t *); +extern void qlt_fc_port_deleted(struct scsi_qla_host *, fc_port_t *, int); extern int __init qlt_init(void); extern void qlt_exit(void); extern void qlt_update_vp_map(struct scsi_qla_host *, int); @@ -1163,5 +1165,6 @@ extern irqreturn_t qla83xx_msix_atio_q(int, void *); extern void qlt_83xx_iospace_config(struct qla_hw_data *); extern int qlt_free_qfull_cmds(struct scsi_qla_host *); extern void qlt_logo_completion_handler(fc_port_t *, int); +extern void qlt_do_generation_tick(struct scsi_qla_host *, int *); #endif /* __QLA_TARGET_H */ -- cgit v0.10.2 From d20ed91bb60ca54c42e3326251287ec51ed225a1 Mon Sep 17 00:00:00 2001 From: Alexei Potashnik Date: Tue, 14 Jul 2015 16:00:47 -0400 Subject: qla2xxx: disable scsi_transport_fc registration in target mode There are multiple reasons for disabling this: 1. It provides no functional benefit. We pretty much only get a few more sysfs entries for each port, but all that information is already available from /sys/kernel/debug/target/qla-session-X 2. It already only works in private-loop mode. By disabling we'll be getting more uniform behavior with fabric mode. 3. It creates complications for the new PLOGI handling mechanism: scsi_transport_fc port deletion timer could race with new session from initiator and cause logout after successful login. Cc: # v3.18+ Signed-off-by: Alexei Potashnik Signed-off-by: Himanshu Madhani Signed-off-by: Nicholas Bellinger diff --git a/drivers/scsi/qla2xxx/qla_init.c b/drivers/scsi/qla2xxx/qla_init.c index 506621d..11f2f32 100644 --- a/drivers/scsi/qla2xxx/qla_init.c +++ b/drivers/scsi/qla2xxx/qla_init.c @@ -3340,8 +3340,7 @@ qla2x00_update_fcport(scsi_qla_host_t *vha, fc_port_t *fcport) if (IS_QLAFX00(vha->hw)) { qla2x00_set_fcport_state(fcport, FCS_ONLINE); - qla2x00_reg_remote_port(vha, fcport); - return; + goto reg_port; } fcport->login_retry = 0; fcport->flags &= ~(FCF_LOGIN_NEEDED | FCF_ASYNC_SENT); @@ -3349,7 +3348,16 @@ qla2x00_update_fcport(scsi_qla_host_t *vha, fc_port_t *fcport) qla2x00_set_fcport_state(fcport, FCS_ONLINE); qla2x00_iidma_fcport(vha, fcport); qla24xx_update_fcport_fcp_prio(vha, fcport); - qla2x00_reg_remote_port(vha, fcport); + +reg_port: + if (qla_ini_mode_enabled(vha)) + qla2x00_reg_remote_port(vha, fcport); + else { + /* + * Create target mode FC NEXUS in qla_target.c + */ + qlt_fc_port_added(vha, fcport); + } } /* diff --git a/drivers/scsi/qla2xxx/qla_os.c b/drivers/scsi/qla2xxx/qla_os.c index be6de2a..8a5cac8 100644 --- a/drivers/scsi/qla2xxx/qla_os.c +++ b/drivers/scsi/qla2xxx/qla_os.c @@ -3235,7 +3235,8 @@ qla2x00_schedule_rport_del(struct scsi_qla_host *vha, fc_port_t *fcport, qla2xxx_wake_dpc(base_vha); } else { int now; - fc_remote_port_delete(rport); + if (rport) + fc_remote_port_delete(rport); qlt_do_generation_tick(vha, &now); qlt_fc_port_deleted(vha, fcport, now); } -- cgit v0.10.2 From e52a8b45b9c782937f74b701f8c656d4e5619eb5 Mon Sep 17 00:00:00 2001 From: Alexei Potashnik Date: Tue, 14 Jul 2015 16:00:48 -0400 Subject: qla2xxx: drop cmds/tmrs arrived while session is being deleted If a new initiator (different WWN) shows up on the same fcport, old initiator's session is scheduled for deletion. But there is a small window between it being marked with QLA_SESS_DELETION_IN_PROGRESS and qlt_unret_sess getting called when new session's commands will keep finding old session in the fcport map. This patch drops cmds/tmrs if they find session in the progress of being deleted. Cc: # v3.18+ Signed-off-by: Alexei Potashnik Acked-by: Quinn Tran Signed-off-by: Himanshu Madhani Signed-off-by: Nicholas Bellinger diff --git a/drivers/scsi/qla2xxx/qla_target.c b/drivers/scsi/qla2xxx/qla_target.c index 0b5bd9c..b60f907 100644 --- a/drivers/scsi/qla2xxx/qla_target.c +++ b/drivers/scsi/qla2xxx/qla_target.c @@ -1477,6 +1477,11 @@ static void qlt_24xx_handle_abts(struct scsi_qla_host *vha, return; } + if (sess->deleted == QLA_SESS_DELETION_IN_PROGRESS) { + qlt_24xx_send_abts_resp(vha, abts, FCP_TMF_REJECTED, false); + return; + } + rc = __qlt_24xx_handle_abts(vha, abts, sess); if (rc != 0) { ql_dbg(ql_dbg_tgt_mgt, vha, 0xf054, @@ -3768,6 +3773,16 @@ static int qlt_handle_cmd_for_atio(struct scsi_qla_host *vha, queue_work(qla_tgt_wq, &op->work); return 0; } + + /* Another WWN used to have our s_id. Our PLOGI scheduled its + * session deletion, but it's still in sess_del_work wq */ + if (sess->deleted == QLA_SESS_DELETION_IN_PROGRESS) { + ql_dbg(ql_dbg_io, vha, 0x3061, + "New command while old session %p is being deleted\n", + sess); + return -EFAULT; + } + /* * Do kref_get() before returning + dropping qla_hw_data->hardware_lock. */ @@ -3931,6 +3946,9 @@ static int qlt_handle_task_mgmt(struct scsi_qla_host *vha, void *iocb) sizeof(struct atio_from_isp)); } + if (sess->deleted == QLA_SESS_DELETION_IN_PROGRESS) + return -EFAULT; + return qlt_issue_task_mgmt(sess, unpacked_lun, fn, iocb, 0); } @@ -5603,6 +5621,11 @@ static void qlt_abort_work(struct qla_tgt *tgt, if (!sess) goto out_term; } else { + if (sess->deleted == QLA_SESS_DELETION_IN_PROGRESS) { + sess = NULL; + goto out_term; + } + kref_get(&sess->se_sess->sess_kref); } @@ -5657,6 +5680,11 @@ static void qlt_tmr_work(struct qla_tgt *tgt, if (!sess) goto out_term; } else { + if (sess->deleted == QLA_SESS_DELETION_IN_PROGRESS) { + sess = NULL; + goto out_term; + } + kref_get(&sess->se_sess->sess_kref); } -- cgit v0.10.2 From 7359df25a53386dd33c223672bbd12cb49d0ce4f Mon Sep 17 00:00:00 2001 From: Alexei Potashnik Date: Tue, 14 Jul 2015 16:00:49 -0400 Subject: qla2xxx: terminate exchange when command is aborted by LIO The newly introduced aborted_task TFO callback has to terminate exchange with QLogic driver, since command is being deleted and no status will be queued to the driver at a later point. This patch also moves the burden of releasing one cmd refcount to the aborted_task handler. Changed iSCSI aborted_task logic to satisfy the above requirement. Cc: # v3.18+ Signed-off-by: Alexei Potashnik Acked-by: Quinn Tran Signed-off-by: Himanshu Madhani Signed-off-by: Nicholas Bellinger diff --git a/drivers/scsi/qla2xxx/qla_target.c b/drivers/scsi/qla2xxx/qla_target.c index b60f907..58651ec 100644 --- a/drivers/scsi/qla2xxx/qla_target.c +++ b/drivers/scsi/qla2xxx/qla_target.c @@ -1924,20 +1924,6 @@ static int qlt_pre_xmit_response(struct qla_tgt_cmd *cmd, struct qla_hw_data *ha = vha->hw; struct se_cmd *se_cmd = &cmd->se_cmd; - if (unlikely(cmd->aborted)) { - ql_dbg(ql_dbg_tgt_mgt, vha, 0xf014, - "qla_target(%d): terminating exchange for aborted cmd=%p (se_cmd=%p, tag=%lld)", - vha->vp_idx, cmd, se_cmd, se_cmd->tag); - - cmd->state = QLA_TGT_STATE_ABORTED; - cmd->cmd_flags |= BIT_6; - - qlt_send_term_exchange(vha, cmd, &cmd->atio, 0); - - /* !! At this point cmd could be already freed !! */ - return QLA_TGT_PRE_XMIT_RESP_CMD_ABORTED; - } - prm->cmd = cmd; prm->tgt = tgt; prm->rq_result = scsi_status; @@ -2524,9 +2510,6 @@ int qlt_xmit_response(struct qla_tgt_cmd *cmd, int xmit_type, res = qlt_pre_xmit_response(cmd, &prm, xmit_type, scsi_status, &full_req_cnt); if (unlikely(res != 0)) { - if (res == QLA_TGT_PRE_XMIT_RESP_CMD_ABORTED) - return 0; - return res; } @@ -3092,6 +3075,24 @@ static void qlt_chk_exch_leak_thresh_hold(struct scsi_qla_host *vha) } +void qlt_abort_cmd(struct qla_tgt_cmd *cmd) +{ + struct qla_tgt *tgt = cmd->tgt; + struct scsi_qla_host *vha = tgt->vha; + struct se_cmd *se_cmd = &cmd->se_cmd; + + ql_dbg(ql_dbg_tgt_mgt, vha, 0xf014, + "qla_target(%d): terminating exchange for aborted cmd=%p " + "(se_cmd=%p, tag=%llu)", vha->vp_idx, cmd, &cmd->se_cmd, + se_cmd->tag); + + cmd->state = QLA_TGT_STATE_ABORTED; + cmd->cmd_flags |= BIT_6; + + qlt_send_term_exchange(vha, cmd, &cmd->atio, 0); +} +EXPORT_SYMBOL(qlt_abort_cmd); + void qlt_free_cmd(struct qla_tgt_cmd *cmd) { struct qla_tgt_sess *sess = cmd->sess; diff --git a/drivers/scsi/qla2xxx/qla_target.h b/drivers/scsi/qla2xxx/qla_target.h index 2ceb60f..bca584a 100644 --- a/drivers/scsi/qla2xxx/qla_target.h +++ b/drivers/scsi/qla2xxx/qla_target.h @@ -808,13 +808,6 @@ int qla2x00_wait_for_hba_online(struct scsi_qla_host *); #define FC_TM_REJECT 4 #define FC_TM_FAILED 5 -/* - * Error code of qlt_pre_xmit_response() meaning that cmd's exchange was - * terminated, so no more actions is needed and success should be returned - * to target. - */ -#define QLA_TGT_PRE_XMIT_RESP_CMD_ABORTED 0x1717 - #if (BITS_PER_LONG > 32) || defined(CONFIG_HIGHMEM64G) #define pci_dma_lo32(a) (a & 0xffffffff) #define pci_dma_hi32(a) ((((a) >> 16)>>16) & 0xffffffff) @@ -950,7 +943,6 @@ struct qla_tgt_cmd { unsigned int conf_compl_supported:1; unsigned int sg_mapped:1; unsigned int free_sg:1; - unsigned int aborted:1; /* Needed in case of SRR */ unsigned int write_data_transferred:1; unsigned int ctx_dsd_alloced:1; unsigned int q_full:1; @@ -1134,6 +1126,7 @@ static inline uint32_t sid_to_key(const uint8_t *s_id) extern void qlt_response_pkt_all_vps(struct scsi_qla_host *, response_t *); extern int qlt_rdy_to_xfer(struct qla_tgt_cmd *); extern int qlt_xmit_response(struct qla_tgt_cmd *, int, uint8_t); +extern void qlt_abort_cmd(struct qla_tgt_cmd *); extern void qlt_xmit_tm_rsp(struct qla_tgt_mgmt_cmd *); extern void qlt_free_mcmd(struct qla_tgt_mgmt_cmd *); extern void qlt_free_cmd(struct qla_tgt_cmd *cmd); diff --git a/drivers/scsi/qla2xxx/tcm_qla2xxx.c b/drivers/scsi/qla2xxx/tcm_qla2xxx.c index 4e242c7..9224a06 100644 --- a/drivers/scsi/qla2xxx/tcm_qla2xxx.c +++ b/drivers/scsi/qla2xxx/tcm_qla2xxx.c @@ -541,7 +541,6 @@ static int tcm_qla2xxx_queue_data_in(struct se_cmd *se_cmd) cmd->cmd_flags |= BIT_4; cmd->bufflen = se_cmd->data_length; cmd->dma_data_direction = target_reverse_dma_direction(se_cmd); - cmd->aborted = (se_cmd->transport_state & CMD_T_ABORTED); cmd->sg_cnt = se_cmd->t_data_nents; cmd->sg = se_cmd->t_data_sg; @@ -570,7 +569,6 @@ static int tcm_qla2xxx_queue_status(struct se_cmd *se_cmd) cmd->sg_cnt = 0; cmd->offset = 0; cmd->dma_data_direction = target_reverse_dma_direction(se_cmd); - cmd->aborted = (se_cmd->transport_state & CMD_T_ABORTED); if (cmd->cmd_flags & BIT_5) { pr_crit("Bit_5 already set for cmd = %p.\n", cmd); dump_stack(); @@ -635,14 +633,7 @@ static void tcm_qla2xxx_aborted_task(struct se_cmd *se_cmd) { struct qla_tgt_cmd *cmd = container_of(se_cmd, struct qla_tgt_cmd, se_cmd); - struct scsi_qla_host *vha = cmd->vha; - struct qla_hw_data *ha = vha->hw; - - if (!cmd->sg_mapped) - return; - - pci_unmap_sg(ha->pdev, cmd->sg, cmd->sg_cnt, cmd->dma_data_direction); - cmd->sg_mapped = 0; + qlt_abort_cmd(cmd); } static void tcm_qla2xxx_clear_sess_lookup(struct tcm_qla2xxx_lport *, -- cgit v0.10.2 From 417c20a9bdd1e876384127cf096d8ae8b559066c Mon Sep 17 00:00:00 2001 From: Nicholas Bellinger Date: Wed, 22 Jul 2015 00:24:09 -0700 Subject: iscsi-target: Fix use-after-free during TPG session shutdown This patch fixes a use-after-free bug in iscsit_release_sessions_for_tpg() where se_portal_group->session_lock was incorrectly released/re-acquired while walking the active se_portal_group->tpg_sess_list. The can result in a NULL pointer dereference when iscsit_close_session() shutdown happens in the normal path asynchronously to this code, causing a bogus dereference of an already freed list entry to occur. To address this bug, walk the session list checking for the same state as before, but move entries to a local list to avoid dropping the lock while walking the active list. As before, signal using iscsi_session->session_restatement=1 for those list entries to be released locally by iscsit_free_session() code. Reported-by: Sunilkumar Nadumuttlu Cc: Sunilkumar Nadumuttlu Cc: # v3.1+ Signed-off-by: Nicholas Bellinger diff --git a/drivers/target/iscsi/iscsi_target.c b/drivers/target/iscsi/iscsi_target.c index 4e68b62..afab323 100644 --- a/drivers/target/iscsi/iscsi_target.c +++ b/drivers/target/iscsi/iscsi_target.c @@ -4765,6 +4765,7 @@ int iscsit_release_sessions_for_tpg(struct iscsi_portal_group *tpg, int force) struct iscsi_session *sess; struct se_portal_group *se_tpg = &tpg->tpg_se_tpg; struct se_session *se_sess, *se_sess_tmp; + LIST_HEAD(free_list); int session_count = 0; spin_lock_bh(&se_tpg->session_lock); @@ -4786,14 +4787,17 @@ int iscsit_release_sessions_for_tpg(struct iscsi_portal_group *tpg, int force) } atomic_set(&sess->session_reinstatement, 1); spin_unlock(&sess->conn_lock); - spin_unlock_bh(&se_tpg->session_lock); - iscsit_free_session(sess); - spin_lock_bh(&se_tpg->session_lock); + list_move_tail(&se_sess->sess_list, &free_list); + } + spin_unlock_bh(&se_tpg->session_lock); + list_for_each_entry_safe(se_sess, se_sess_tmp, &free_list, sess_list) { + sess = (struct iscsi_session *)se_sess->fabric_sess_ptr; + + iscsit_free_session(sess); session_count++; } - spin_unlock_bh(&se_tpg->session_lock); pr_debug("Released %d iSCSI Session(s) from Target Portal" " Group: %hu\n", session_count, tpg->tpgt); -- cgit v0.10.2 From e54198657b65625085834847ab6271087323ffea Mon Sep 17 00:00:00 2001 From: Nicholas Bellinger Date: Wed, 22 Jul 2015 23:14:19 -0700 Subject: iscsi-target: Fix iscsit_start_kthreads failure OOPs This patch fixes a regression introduced with the following commit in v4.0-rc1 code, where a iscsit_start_kthreads() failure triggers a NULL pointer dereference OOPs: commit 88dcd2dab5c23b1c9cfc396246d8f476c872f0ca Author: Nicholas Bellinger Date: Thu Feb 26 22:19:15 2015 -0800 iscsi-target: Convert iscsi_thread_set usage to kthread.h To address this bug, move iscsit_start_kthreads() immediately preceeding the transmit of last login response, before signaling a successful transition into full-feature-phase within existing iscsi_target_do_tx_login_io() logic. This ensures that no target-side resource allocation failures can occur after the final login response has been successfully sent. Also, it adds a iscsi_conn->rx_login_comp to allow the RX thread to sleep to prevent other socket related failures until the final iscsi_post_login_handler() call is able to complete. Cc: Sagi Grimberg Cc: # v3.10+ Signed-off-by: Nicholas Bellinger diff --git a/drivers/target/iscsi/iscsi_target.c b/drivers/target/iscsi/iscsi_target.c index afab323..202a428 100644 --- a/drivers/target/iscsi/iscsi_target.c +++ b/drivers/target/iscsi/iscsi_target.c @@ -3998,7 +3998,13 @@ get_immediate: } transport_err: - iscsit_take_action_for_connection_exit(conn); + /* + * Avoid the normal connection failure code-path if this connection + * is still within LOGIN mode, and iscsi_np process context is + * responsible for cleaning up the early connection failure. + */ + if (conn->conn_state != TARG_CONN_STATE_IN_LOGIN) + iscsit_take_action_for_connection_exit(conn); out: return 0; } @@ -4082,7 +4088,7 @@ reject: int iscsi_target_rx_thread(void *arg) { - int ret; + int ret, rc; u8 buffer[ISCSI_HDR_LEN], opcode; u32 checksum = 0, digest = 0; struct iscsi_conn *conn = arg; @@ -4092,10 +4098,16 @@ int iscsi_target_rx_thread(void *arg) * connection recovery / failure event can be triggered externally. */ allow_signal(SIGINT); + /* + * Wait for iscsi_post_login_handler() to complete before allowing + * incoming iscsi/tcp socket I/O, and/or failing the connection. + */ + rc = wait_for_completion_interruptible(&conn->rx_login_comp); + if (rc < 0) + return 0; if (conn->conn_transport->transport_type == ISCSI_INFINIBAND) { struct completion comp; - int rc; init_completion(&comp); rc = wait_for_completion_interruptible(&comp); diff --git a/drivers/target/iscsi/iscsi_target_login.c b/drivers/target/iscsi/iscsi_target_login.c index 3d0fe4f..7e8f65e 100644 --- a/drivers/target/iscsi/iscsi_target_login.c +++ b/drivers/target/iscsi/iscsi_target_login.c @@ -82,6 +82,7 @@ static struct iscsi_login *iscsi_login_init_conn(struct iscsi_conn *conn) init_completion(&conn->conn_logout_comp); init_completion(&conn->rx_half_close_comp); init_completion(&conn->tx_half_close_comp); + init_completion(&conn->rx_login_comp); spin_lock_init(&conn->cmd_lock); spin_lock_init(&conn->conn_usage_lock); spin_lock_init(&conn->immed_queue_lock); @@ -644,7 +645,7 @@ static void iscsi_post_login_start_timers(struct iscsi_conn *conn) iscsit_start_nopin_timer(conn); } -static int iscsit_start_kthreads(struct iscsi_conn *conn) +int iscsit_start_kthreads(struct iscsi_conn *conn) { int ret = 0; @@ -679,6 +680,7 @@ static int iscsit_start_kthreads(struct iscsi_conn *conn) return 0; out_tx: + send_sig(SIGINT, conn->tx_thread, 1); kthread_stop(conn->tx_thread); conn->tx_thread_active = false; out_bitmap: @@ -689,7 +691,7 @@ out_bitmap: return ret; } -int iscsi_post_login_handler( +void iscsi_post_login_handler( struct iscsi_np *np, struct iscsi_conn *conn, u8 zero_tsih) @@ -699,7 +701,6 @@ int iscsi_post_login_handler( struct se_session *se_sess = sess->se_sess; struct iscsi_portal_group *tpg = sess->tpg; struct se_portal_group *se_tpg = &tpg->tpg_se_tpg; - int rc; iscsit_inc_conn_usage_count(conn); @@ -739,10 +740,6 @@ int iscsi_post_login_handler( sess->sess_ops->InitiatorName); spin_unlock_bh(&sess->conn_lock); - rc = iscsit_start_kthreads(conn); - if (rc) - return rc; - iscsi_post_login_start_timers(conn); /* * Determine CPU mask to ensure connection's RX and TX kthreads @@ -751,15 +748,20 @@ int iscsi_post_login_handler( iscsit_thread_get_cpumask(conn); conn->conn_rx_reset_cpumask = 1; conn->conn_tx_reset_cpumask = 1; - + /* + * Wakeup the sleeping iscsi_target_rx_thread() now that + * iscsi_conn is in TARG_CONN_STATE_LOGGED_IN state. + */ + complete(&conn->rx_login_comp); iscsit_dec_conn_usage_count(conn); + if (stop_timer) { spin_lock_bh(&se_tpg->session_lock); iscsit_stop_time2retain_timer(sess); spin_unlock_bh(&se_tpg->session_lock); } iscsit_dec_session_usage_count(sess); - return 0; + return; } iscsi_set_session_parameters(sess->sess_ops, conn->param_list, 1); @@ -800,10 +802,6 @@ int iscsi_post_login_handler( " iSCSI Target Portal Group: %hu\n", tpg->nsessions, tpg->tpgt); spin_unlock_bh(&se_tpg->session_lock); - rc = iscsit_start_kthreads(conn); - if (rc) - return rc; - iscsi_post_login_start_timers(conn); /* * Determine CPU mask to ensure connection's RX and TX kthreads @@ -812,10 +810,12 @@ int iscsi_post_login_handler( iscsit_thread_get_cpumask(conn); conn->conn_rx_reset_cpumask = 1; conn->conn_tx_reset_cpumask = 1; - + /* + * Wakeup the sleeping iscsi_target_rx_thread() now that + * iscsi_conn is in TARG_CONN_STATE_LOGGED_IN state. + */ + complete(&conn->rx_login_comp); iscsit_dec_conn_usage_count(conn); - - return 0; } static void iscsi_handle_login_thread_timeout(unsigned long data) @@ -1380,23 +1380,12 @@ static int __iscsi_target_login_thread(struct iscsi_np *np) if (ret < 0) goto new_sess_out; - if (!conn->sess) { - pr_err("struct iscsi_conn session pointer is NULL!\n"); - goto new_sess_out; - } - iscsi_stop_login_thread_timer(np); - if (signal_pending(current)) - goto new_sess_out; - if (ret == 1) { tpg_np = conn->tpg_np; - ret = iscsi_post_login_handler(np, conn, zero_tsih); - if (ret < 0) - goto new_sess_out; - + iscsi_post_login_handler(np, conn, zero_tsih); iscsit_deaccess_np(np, tpg, tpg_np); } diff --git a/drivers/target/iscsi/iscsi_target_login.h b/drivers/target/iscsi/iscsi_target_login.h index 1c73580..57aa0d0 100644 --- a/drivers/target/iscsi/iscsi_target_login.h +++ b/drivers/target/iscsi/iscsi_target_login.h @@ -12,7 +12,8 @@ extern int iscsit_accept_np(struct iscsi_np *, struct iscsi_conn *); extern int iscsit_get_login_rx(struct iscsi_conn *, struct iscsi_login *); extern int iscsit_put_login_tx(struct iscsi_conn *, struct iscsi_login *, u32); extern void iscsit_free_conn(struct iscsi_np *, struct iscsi_conn *); -extern int iscsi_post_login_handler(struct iscsi_np *, struct iscsi_conn *, u8); +extern int iscsit_start_kthreads(struct iscsi_conn *); +extern void iscsi_post_login_handler(struct iscsi_np *, struct iscsi_conn *, u8); extern void iscsi_target_login_sess_out(struct iscsi_conn *, struct iscsi_np *, bool, bool); extern int iscsi_target_login_thread(void *); diff --git a/drivers/target/iscsi/iscsi_target_nego.c b/drivers/target/iscsi/iscsi_target_nego.c index 8c02fa3..f9cde91 100644 --- a/drivers/target/iscsi/iscsi_target_nego.c +++ b/drivers/target/iscsi/iscsi_target_nego.c @@ -17,6 +17,7 @@ ******************************************************************************/ #include +#include #include #include #include @@ -361,10 +362,24 @@ static int iscsi_target_do_tx_login_io(struct iscsi_conn *conn, struct iscsi_log ntohl(login_rsp->statsn), login->rsp_length); padding = ((-login->rsp_length) & 3); + /* + * Before sending the last login response containing the transition + * bit for full-feature-phase, go ahead and start up TX/RX threads + * now to avoid potential resource allocation failures after the + * final login response has been sent. + */ + if (login->login_complete) { + int rc = iscsit_start_kthreads(conn); + if (rc) { + iscsit_tx_login_rsp(conn, ISCSI_STATUS_CLS_TARGET_ERR, + ISCSI_LOGIN_STATUS_NO_RESOURCES); + return -1; + } + } if (conn->conn_transport->iscsit_put_login_tx(conn, login, login->rsp_length + padding) < 0) - return -1; + goto err; login->rsp_length = 0; mutex_lock(&sess->cmdsn_mutex); @@ -373,6 +388,23 @@ static int iscsi_target_do_tx_login_io(struct iscsi_conn *conn, struct iscsi_log mutex_unlock(&sess->cmdsn_mutex); return 0; + +err: + if (login->login_complete) { + if (conn->rx_thread && conn->rx_thread_active) { + send_sig(SIGINT, conn->rx_thread, 1); + kthread_stop(conn->rx_thread); + } + if (conn->tx_thread && conn->tx_thread_active) { + send_sig(SIGINT, conn->tx_thread, 1); + kthread_stop(conn->tx_thread); + } + spin_lock(&iscsit_global->ts_bitmap_lock); + bitmap_release_region(iscsit_global->ts_bitmap, conn->bitmap_id, + get_order(1)); + spin_unlock(&iscsit_global->ts_bitmap_lock); + } + return -1; } static void iscsi_target_sk_data_ready(struct sock *sk) diff --git a/include/target/iscsi/iscsi_target_core.h b/include/target/iscsi/iscsi_target_core.h index 34117b8..0aedbb2 100644 --- a/include/target/iscsi/iscsi_target_core.h +++ b/include/target/iscsi/iscsi_target_core.h @@ -595,6 +595,7 @@ struct iscsi_conn { int bitmap_id; int rx_thread_active; struct task_struct *rx_thread; + struct completion rx_login_comp; int tx_thread_active; struct task_struct *tx_thread; /* list_head for session connection list */ -- cgit v0.10.2 From 007d038bdf95ccfe2491d0078be54040d110fd06 Mon Sep 17 00:00:00 2001 From: Nicholas Bellinger Date: Thu, 23 Jul 2015 22:30:31 +0000 Subject: iscsi-target: Fix iser explicit logout TX kthread leak This patch fixes a regression introduced with the following commit in v4.0-rc1 code, where an explicit iser-target logout would result in ->tx_thread_active being incorrectly cleared by the logout post handler, and subsequent TX kthread leak: commit 88dcd2dab5c23b1c9cfc396246d8f476c872f0ca Author: Nicholas Bellinger Date: Thu Feb 26 22:19:15 2015 -0800 iscsi-target: Convert iscsi_thread_set usage to kthread.h To address this bug, change iscsit_logout_post_handler_closesession() and iscsit_logout_post_handler_samecid() to only cmpxchg() on ->tx_thread_active for traditional iscsi/tcp connections. This is required because iscsi/tcp connections are invoking logout post handler logic directly from TX kthread context, while iser connections are invoking logout post handler logic from a seperate workqueue context. Cc: Sagi Grimberg Cc: # v3.10+ Signed-off-by: Nicholas Bellinger diff --git a/drivers/target/iscsi/iscsi_target.c b/drivers/target/iscsi/iscsi_target.c index 202a428..cd77a06 100644 --- a/drivers/target/iscsi/iscsi_target.c +++ b/drivers/target/iscsi/iscsi_target.c @@ -4544,7 +4544,18 @@ static void iscsit_logout_post_handler_closesession( struct iscsi_conn *conn) { struct iscsi_session *sess = conn->sess; - int sleep = cmpxchg(&conn->tx_thread_active, true, false); + int sleep = 1; + /* + * Traditional iscsi/tcp will invoke this logic from TX thread + * context during session logout, so clear tx_thread_active and + * sleep if iscsit_close_connection() has not already occured. + * + * Since iser-target invokes this logic from it's own workqueue, + * always sleep waiting for RX/TX thread shutdown to complete + * within iscsit_close_connection(). + */ + if (conn->conn_transport->transport_type == ISCSI_TCP) + sleep = cmpxchg(&conn->tx_thread_active, true, false); atomic_set(&conn->conn_logout_remove, 0); complete(&conn->conn_logout_comp); @@ -4558,7 +4569,10 @@ static void iscsit_logout_post_handler_closesession( static void iscsit_logout_post_handler_samecid( struct iscsi_conn *conn) { - int sleep = cmpxchg(&conn->tx_thread_active, true, false); + int sleep = 1; + + if (conn->conn_transport->transport_type == ISCSI_TCP) + sleep = cmpxchg(&conn->tx_thread_active, true, false); atomic_set(&conn->conn_logout_remove, 0); complete(&conn->conn_logout_comp); -- cgit v0.10.2 From ce9a9fc20a78ad1e5222fae3a83d105f2d2fb9b9 Mon Sep 17 00:00:00 2001 From: Nicholas Bellinger Date: Fri, 24 Jul 2015 02:29:55 +0000 Subject: iser-target: Fix REJECT CM event use-after-free OOPs This patch fixes a bug in iser-target code where the REJECT CM event handler code currently performs a isert_put_conn() for the final isert_conn->kref put, while iscsi_np process context is still blocked in isert_get_login_rx(). Once isert_get_login_rx() is awoking due to login timeout, iscsi_np process context will attempt to invoke iscsi_target_login_sess_out() to cleanup iscsi_conn as expected, and calls isert_wait_conn() + isert_free_conn() which triggers the use-after-free OOPs. To address this bug, move the kref_get_unless_zero() call from isert_connected_handler() into isert_connect_request() immediately preceeding isert_rdma_accept() to ensure the CM handler cleanup paths and isert_free_conn() are always operating with two refs. Cc: Sagi Grimberg Cc: # v3.10+ Signed-off-by: Nicholas Bellinger diff --git a/drivers/infiniband/ulp/isert/ib_isert.c b/drivers/infiniband/ulp/isert/ib_isert.c index 7717009..d851e18 100644 --- a/drivers/infiniband/ulp/isert/ib_isert.c +++ b/drivers/infiniband/ulp/isert/ib_isert.c @@ -775,6 +775,17 @@ isert_connect_request(struct rdma_cm_id *cma_id, struct rdma_cm_event *event) ret = isert_rdma_post_recvl(isert_conn); if (ret) goto out_conn_dev; + /* + * Obtain the second reference now before isert_rdma_accept() to + * ensure that any initiator generated REJECT CM event that occurs + * asynchronously won't drop the last reference until the error path + * in iscsi_target_login_sess_out() does it's ->iscsit_free_conn() -> + * isert_free_conn() -> isert_put_conn() -> kref_put(). + */ + if (!kref_get_unless_zero(&isert_conn->kref)) { + isert_warn("conn %p connect_release is running\n", isert_conn); + goto out_conn_dev; + } ret = isert_rdma_accept(isert_conn); if (ret) @@ -836,11 +847,6 @@ isert_connected_handler(struct rdma_cm_id *cma_id) isert_info("conn %p\n", isert_conn); - if (!kref_get_unless_zero(&isert_conn->kref)) { - isert_warn("conn %p connect_release is running\n", isert_conn); - return; - } - mutex_lock(&isert_conn->mutex); if (isert_conn->state != ISER_CONN_FULL_FEATURE) isert_conn->state = ISER_CONN_UP; -- cgit v0.10.2 From c4cfdd81c8fde84e2c75bc90533c7e1276937d3a Mon Sep 17 00:00:00 2001 From: Sebastian Herbszt Date: Mon, 20 Jul 2015 00:16:01 +0200 Subject: Documentation/target: Fix tcm_mod_builder.py build breakage Fix build breakage and set the protocol identifier based on the parameter. Fixes: 9ac8928e6a3e ("target: simplify the target template registration API") Fixes: e4aae5af810e ("target: change core_tpg_register prototype") Signed-off-by: Sebastian Herbszt Reviewed-by: Christoph Hellwig Signed-off-by: Nicholas Bellinger diff --git a/Documentation/target/tcm_mod_builder.py b/Documentation/target/tcm_mod_builder.py index 949de19..cda56df 100755 --- a/Documentation/target/tcm_mod_builder.py +++ b/Documentation/target/tcm_mod_builder.py @@ -199,7 +199,8 @@ def tcm_mod_build_configfs(proto_ident, fabric_mod_dir_var, fabric_mod_name): buf += "#include \n" buf += "#include \n" buf += "#include \n" - buf += "#include \n\n" + buf += "#include \n" + buf += "#include \n\n" buf += "#include \n" buf += "#include \n" buf += "#include \n" @@ -230,8 +231,14 @@ def tcm_mod_build_configfs(proto_ident, fabric_mod_dir_var, fabric_mod_name): buf += " }\n" buf += " tpg->" + fabric_mod_port + " = " + fabric_mod_port + ";\n" buf += " tpg->" + fabric_mod_port + "_tpgt = tpgt;\n\n" - buf += " ret = core_tpg_register(&" + fabric_mod_name + "_ops, wwn,\n" - buf += " &tpg->se_tpg, SCSI_PROTOCOL_SAS);\n" + + if proto_ident == "FC": + buf += " ret = core_tpg_register(wwn, &tpg->se_tpg, SCSI_PROTOCOL_FCP);\n" + elif proto_ident == "SAS": + buf += " ret = core_tpg_register(wwn, &tpg->se_tpg, SCSI_PROTOCOL_SAS);\n" + elif proto_ident == "iSCSI": + buf += " ret = core_tpg_register(wwn, &tpg->se_tpg, SCSI_PROTOCOL_ISCSI);\n" + buf += " if (ret < 0) {\n" buf += " kfree(tpg);\n" buf += " return NULL;\n" @@ -292,7 +299,7 @@ def tcm_mod_build_configfs(proto_ident, fabric_mod_dir_var, fabric_mod_name): buf += "static const struct target_core_fabric_ops " + fabric_mod_name + "_ops = {\n" buf += " .module = THIS_MODULE,\n" - buf += " .name = " + fabric_mod_name + ",\n" + buf += " .name = \"" + fabric_mod_name + "\",\n" buf += " .get_fabric_name = " + fabric_mod_name + "_get_fabric_name,\n" buf += " .tpg_get_wwn = " + fabric_mod_name + "_get_fabric_wwn,\n" buf += " .tpg_get_tag = " + fabric_mod_name + "_get_tag,\n" @@ -322,17 +329,17 @@ def tcm_mod_build_configfs(proto_ident, fabric_mod_dir_var, fabric_mod_name): buf += " .fabric_make_tpg = " + fabric_mod_name + "_make_tpg,\n" buf += " .fabric_drop_tpg = " + fabric_mod_name + "_drop_tpg,\n" buf += "\n" - buf += " .tfc_wwn_attrs = " + fabric_mod_name + "_wwn_attrs;\n" + buf += " .tfc_wwn_attrs = " + fabric_mod_name + "_wwn_attrs,\n" buf += "};\n\n" buf += "static int __init " + fabric_mod_name + "_init(void)\n" buf += "{\n" - buf += " return target_register_template(" + fabric_mod_name + "_ops);\n" + buf += " return target_register_template(&" + fabric_mod_name + "_ops);\n" buf += "};\n\n" buf += "static void __exit " + fabric_mod_name + "_exit(void)\n" buf += "{\n" - buf += " target_unregister_template(" + fabric_mod_name + "_ops);\n" + buf += " target_unregister_template(&" + fabric_mod_name + "_ops);\n" buf += "};\n\n" buf += "MODULE_DESCRIPTION(\"" + fabric_mod_name.upper() + " series fabric driver\");\n" -- cgit v0.10.2 From e053f96b1a00022b4e2c7ceb7ac0229646626507 Mon Sep 17 00:00:00 2001 From: Denis Carikli Date: Thu, 23 Jul 2015 10:31:12 +0200 Subject: ARM: dts: i.MX35: Fix can support. Since commit 3d42a379b6fa5b46058e3302b1802b29f64865bb ("can: flexcan: add 2nd clock to support imx53 and newer") the can driver requires a dt nodes to have a second clock. Add them to imx35 to fix probing the flex can driver on the respective platforms. Signed-off-by: Denis Carikli Cc: Signed-off-by: Shawn Guo diff --git a/arch/arm/boot/dts/imx35.dtsi b/arch/arm/boot/dts/imx35.dtsi index b6478e9..e6540b5 100644 --- a/arch/arm/boot/dts/imx35.dtsi +++ b/arch/arm/boot/dts/imx35.dtsi @@ -286,8 +286,8 @@ can1: can@53fe4000 { compatible = "fsl,imx35-flexcan", "fsl,p1010-flexcan"; reg = <0x53fe4000 0x1000>; - clocks = <&clks 33>; - clock-names = "ipg"; + clocks = <&clks 33>, <&clks 33>; + clock-names = "ipg", "per"; interrupts = <43>; status = "disabled"; }; @@ -295,8 +295,8 @@ can2: can@53fe8000 { compatible = "fsl,imx35-flexcan", "fsl,p1010-flexcan"; reg = <0x53fe8000 0x1000>; - clocks = <&clks 34>; - clock-names = "ipg"; + clocks = <&clks 34>, <&clks 34>; + clock-names = "ipg", "per"; interrupts = <44>; status = "disabled"; }; -- cgit v0.10.2 From 18a912e9a832dcfc7db9e055c7e41701ff5f9e69 Mon Sep 17 00:00:00 2001 From: Julian Anastasov Date: Wed, 22 Jul 2015 10:43:22 +0300 Subject: ipv4: fib_select_default should match the prefix fib_trie starting from 4.1 can link fib aliases from different prefixes in same list. Make sure the alternative gateways are in same table and for same prefix (0) by checking tb_id and fa_slen. Fixes: 79e5ad2ceb00 ("fib_trie: Remove leaf_info") Signed-off-by: Julian Anastasov Signed-off-by: David S. Miller diff --git a/net/ipv4/fib_semantics.c b/net/ipv4/fib_semantics.c index c7358ea..e107958 100644 --- a/net/ipv4/fib_semantics.c +++ b/net/ipv4/fib_semantics.c @@ -1207,12 +1207,17 @@ void fib_select_default(struct fib_result *res) struct fib_info *fi = NULL, *last_resort = NULL; struct hlist_head *fa_head = res->fa_head; struct fib_table *tb = res->table; + u8 slen = 32 - res->prefixlen; int order = -1, last_idx = -1; struct fib_alias *fa; hlist_for_each_entry_rcu(fa, fa_head, fa_list) { struct fib_info *next_fi = fa->fa_info; + if (fa->fa_slen != slen) + continue; + if (fa->tb_id != tb->tb_id) + continue; if (next_fi->fib_scope != res->scope || fa->fa_type != RTN_UNICAST) continue; -- cgit v0.10.2 From 2392debc2be721a7d5b907cbcbc0ebb858dead01 Mon Sep 17 00:00:00 2001 From: Julian Anastasov Date: Wed, 22 Jul 2015 10:43:23 +0300 Subject: ipv4: consider TOS in fib_select_default fib_select_default considers alternative routes only when res->fi is for the first alias in res->fa_head. In the common case this can happen only when the initial lookup matches the first alias with highest TOS value. This prevents the alternative routes to require specific TOS. This patch solves the problem as follows: - routes that require specific TOS should be returned by fib_select_default only when TOS matches, as already done in fib_table_lookup. This rule implies that depending on the TOS we can have many different lists of alternative gateways and we have to keep the last used gateway (fa_default) in first alias for the TOS instead of using single tb_default value. - as the aliases are ordered by many keys (TOS desc, fib_priority asc), we restrict the possible results to routes with matching TOS and lowest metric (fib_priority) and routes that match any TOS, again with lowest metric. For example, packet with TOS 8 can not use gw3 (not lowest metric), gw4 (different TOS) and gw6 (not lowest metric), all other gateways can be used: tos 8 via gw1 metric 2 <--- res->fa_head and res->fi tos 8 via gw2 metric 2 tos 8 via gw3 metric 3 tos 4 via gw4 tos 0 via gw5 tos 0 via gw6 metric 1 Reported-by: Hagen Paul Pfeifer Signed-off-by: Julian Anastasov Signed-off-by: David S. Miller diff --git a/include/net/ip_fib.h b/include/net/ip_fib.h index 49c142b..5fa643b 100644 --- a/include/net/ip_fib.h +++ b/include/net/ip_fib.h @@ -183,7 +183,6 @@ __be32 fib_info_update_nh_saddr(struct net *net, struct fib_nh *nh); struct fib_table { struct hlist_node tb_hlist; u32 tb_id; - int tb_default; int tb_num_default; struct rcu_head rcu; unsigned long *tb_data; @@ -290,7 +289,7 @@ __be32 fib_compute_spec_dst(struct sk_buff *skb); int fib_validate_source(struct sk_buff *skb, __be32 src, __be32 dst, u8 tos, int oif, struct net_device *dev, struct in_device *idev, u32 *itag); -void fib_select_default(struct fib_result *res); +void fib_select_default(const struct flowi4 *flp, struct fib_result *res); #ifdef CONFIG_IP_ROUTE_CLASSID static inline int fib_num_tclassid_users(struct net *net) { diff --git a/net/ipv4/fib_lookup.h b/net/ipv4/fib_lookup.h index c6211ed..9c02920 100644 --- a/net/ipv4/fib_lookup.h +++ b/net/ipv4/fib_lookup.h @@ -13,6 +13,7 @@ struct fib_alias { u8 fa_state; u8 fa_slen; u32 tb_id; + s16 fa_default; struct rcu_head rcu; }; diff --git a/net/ipv4/fib_semantics.c b/net/ipv4/fib_semantics.c index e107958..3a06586 100644 --- a/net/ipv4/fib_semantics.c +++ b/net/ipv4/fib_semantics.c @@ -1202,28 +1202,40 @@ int fib_sync_down_dev(struct net_device *dev, unsigned long event) } /* Must be invoked inside of an RCU protected region. */ -void fib_select_default(struct fib_result *res) +void fib_select_default(const struct flowi4 *flp, struct fib_result *res) { struct fib_info *fi = NULL, *last_resort = NULL; struct hlist_head *fa_head = res->fa_head; struct fib_table *tb = res->table; u8 slen = 32 - res->prefixlen; int order = -1, last_idx = -1; - struct fib_alias *fa; + struct fib_alias *fa, *fa1 = NULL; + u32 last_prio = res->fi->fib_priority; + u8 last_tos = 0; hlist_for_each_entry_rcu(fa, fa_head, fa_list) { struct fib_info *next_fi = fa->fa_info; if (fa->fa_slen != slen) continue; + if (fa->fa_tos && fa->fa_tos != flp->flowi4_tos) + continue; if (fa->tb_id != tb->tb_id) continue; + if (next_fi->fib_priority > last_prio && + fa->fa_tos == last_tos) { + if (last_tos) + continue; + break; + } + if (next_fi->fib_flags & RTNH_F_DEAD) + continue; + last_tos = fa->fa_tos; + last_prio = next_fi->fib_priority; + if (next_fi->fib_scope != res->scope || fa->fa_type != RTN_UNICAST) continue; - - if (next_fi->fib_priority > res->fi->fib_priority) - break; if (!next_fi->fib_nh[0].nh_gw || next_fi->fib_nh[0].nh_scope != RT_SCOPE_LINK) continue; @@ -1233,10 +1245,11 @@ void fib_select_default(struct fib_result *res) if (!fi) { if (next_fi != res->fi) break; + fa1 = fa; } else if (!fib_detect_death(fi, order, &last_resort, - &last_idx, tb->tb_default)) { + &last_idx, fa1->fa_default)) { fib_result_assign(res, fi); - tb->tb_default = order; + fa1->fa_default = order; goto out; } fi = next_fi; @@ -1244,20 +1257,21 @@ void fib_select_default(struct fib_result *res) } if (order <= 0 || !fi) { - tb->tb_default = -1; + if (fa1) + fa1->fa_default = -1; goto out; } if (!fib_detect_death(fi, order, &last_resort, &last_idx, - tb->tb_default)) { + fa1->fa_default)) { fib_result_assign(res, fi); - tb->tb_default = order; + fa1->fa_default = order; goto out; } if (last_idx >= 0) fib_result_assign(res, last_resort); - tb->tb_default = last_idx; + fa1->fa_default = last_idx; out: return; } diff --git a/net/ipv4/fib_trie.c b/net/ipv4/fib_trie.c index 15d3261..81797e0 100644 --- a/net/ipv4/fib_trie.c +++ b/net/ipv4/fib_trie.c @@ -1171,6 +1171,7 @@ int fib_table_insert(struct fib_table *tb, struct fib_config *cfg) new_fa->fa_state = state & ~FA_S_ACCESSED; new_fa->fa_slen = fa->fa_slen; new_fa->tb_id = tb->tb_id; + new_fa->fa_default = -1; err = switchdev_fib_ipv4_add(key, plen, fi, new_fa->fa_tos, @@ -1222,6 +1223,7 @@ int fib_table_insert(struct fib_table *tb, struct fib_config *cfg) new_fa->fa_state = 0; new_fa->fa_slen = slen; new_fa->tb_id = tb->tb_id; + new_fa->fa_default = -1; /* (Optionally) offload fib entry to switch hardware. */ err = switchdev_fib_ipv4_add(key, plen, fi, tos, cfg->fc_type, @@ -1990,7 +1992,6 @@ struct fib_table *fib_trie_table(u32 id, struct fib_table *alias) return NULL; tb->tb_id = id; - tb->tb_default = -1; tb->tb_num_default = 0; tb->tb_data = (alias ? alias->__data : tb->__data); diff --git a/net/ipv4/route.c b/net/ipv4/route.c index d0362a2..e681b85 100644 --- a/net/ipv4/route.c +++ b/net/ipv4/route.c @@ -2176,7 +2176,7 @@ struct rtable *__ip_route_output_key(struct net *net, struct flowi4 *fl4) if (!res.prefixlen && res.table->tb_num_default > 1 && res.type == RTN_UNICAST && !fl4->flowi4_oif) - fib_select_default(&res); + fib_select_default(fl4, &res); if (!fl4->saddr) fl4->saddr = FIB_RES_PREFSRC(net, res); -- cgit v0.10.2 From fe6bea7f1f3a09fc06d835446d34d3b3b6a543fb Mon Sep 17 00:00:00 2001 From: WANG Cong Date: Tue, 21 Jul 2015 16:31:53 -0700 Subject: sch_plug: purge buffered packets during reset Otherwise the skbuff related structures are not correctly refcount'ed. Cc: Jamal Hadi Salim Signed-off-by: Cong Wang Signed-off-by: David S. Miller diff --git a/net/sched/sch_plug.c b/net/sched/sch_plug.c index 89f8fcf..ade9445 100644 --- a/net/sched/sch_plug.c +++ b/net/sched/sch_plug.c @@ -216,6 +216,7 @@ static struct Qdisc_ops plug_qdisc_ops __read_mostly = { .peek = qdisc_peek_head, .init = plug_init, .change = plug_change, + .reset = qdisc_reset_queue, .owner = THIS_MODULE, }; -- cgit v0.10.2 From 77e62da6e60c7772971f813f588d372a7f1a4167 Mon Sep 17 00:00:00 2001 From: WANG Cong Date: Tue, 21 Jul 2015 16:52:43 -0700 Subject: sch_choke: drop all packets in queue during reset Signed-off-by: Cong Wang Signed-off-by: David S. Miller diff --git a/net/sched/sch_choke.c b/net/sched/sch_choke.c index 93d5742..6a783af 100644 --- a/net/sched/sch_choke.c +++ b/net/sched/sch_choke.c @@ -385,6 +385,19 @@ static void choke_reset(struct Qdisc *sch) { struct choke_sched_data *q = qdisc_priv(sch); + while (q->head != q->tail) { + struct sk_buff *skb = q->tab[q->head]; + + q->head = (q->head + 1) & q->tab_mask; + if (!skb) + continue; + qdisc_qstats_backlog_dec(sch, skb); + --sch->q.qlen; + qdisc_drop(skb, sch); + } + + memset(q->tab, 0, (q->tab_mask + 1) * sizeof(struct sk_buff *)); + q->head = q->tail = 0; red_restart(&q->vars); } -- cgit v0.10.2 From cc9f4daa638e660f7a910b8094122561470ac331 Mon Sep 17 00:00:00 2001 From: Konstantin Khlebnikov Date: Wed, 22 Jul 2015 12:23:20 +0300 Subject: cgroup: net_cls: fix false-positive "suspicious RCU usage" In dev_queue_xmit() net_cls protected with rcu-bh. [ 270.730026] =============================== [ 270.730029] [ INFO: suspicious RCU usage. ] [ 270.730033] 4.2.0-rc3+ #2 Not tainted [ 270.730036] ------------------------------- [ 270.730040] include/linux/cgroup.h:353 suspicious rcu_dereference_check() usage! [ 270.730041] other info that might help us debug this: [ 270.730043] rcu_scheduler_active = 1, debug_locks = 1 [ 270.730045] 2 locks held by dhclient/748: [ 270.730046] #0: (rcu_read_lock_bh){......}, at: [] __dev_queue_xmit+0x50/0x960 [ 270.730085] #1: (&qdisc_tx_lock){+.....}, at: [] __dev_queue_xmit+0x240/0x960 [ 270.730090] stack backtrace: [ 270.730096] CPU: 0 PID: 748 Comm: dhclient Not tainted 4.2.0-rc3+ #2 [ 270.730098] Hardware name: OpenStack Foundation OpenStack Nova, BIOS Bochs 01/01/2011 [ 270.730100] 0000000000000001 ffff8800bafeba58 ffffffff817ad487 0000000000000007 [ 270.730103] ffff880232a0a780 ffff8800bafeba88 ffffffff810ca4f2 ffff88022fb23e00 [ 270.730105] ffff880232a0a780 ffff8800bafebb68 ffff8800bafebb68 ffff8800bafebaa8 [ 270.730108] Call Trace: [ 270.730121] [] dump_stack+0x4c/0x65 [ 270.730148] [] lockdep_rcu_suspicious+0xe2/0x120 [ 270.730153] [] task_cls_state+0x92/0xa0 [ 270.730158] [] cls_cgroup_classify+0x4f/0x120 [cls_cgroup] [ 270.730164] [] tc_classify_compat+0x74/0xc0 [ 270.730166] [] tc_classify+0x33/0x90 [ 270.730170] [] htb_enqueue+0xaa/0x4a0 [sch_htb] [ 270.730172] [] __dev_queue_xmit+0x306/0x960 [ 270.730174] [] ? __dev_queue_xmit+0x50/0x960 [ 270.730176] [] dev_queue_xmit_sk+0x13/0x20 [ 270.730185] [] dev_queue_xmit+0x10/0x20 [ 270.730187] [] packet_snd.isra.62+0x54c/0x760 [ 270.730190] [] packet_sendmsg+0x2f5/0x3f0 [ 270.730203] [] ? sock_def_readable+0x5/0x190 [ 270.730210] [] ? _raw_spin_unlock+0x2b/0x40 [ 270.730216] [] ? unix_dgram_sendmsg+0x5cc/0x640 [ 270.730219] [] sock_sendmsg+0x47/0x50 [ 270.730221] [] sock_write_iter+0x7f/0xd0 [ 270.730232] [] __vfs_write+0xa7/0xf0 [ 270.730234] [] vfs_write+0xb8/0x190 [ 270.730236] [] SyS_write+0x52/0xb0 [ 270.730239] [] entry_SYSCALL_64_fastpath+0x12/0x76 Signed-off-by: Konstantin Khlebnikov Signed-off-by: David S. Miller diff --git a/net/core/netclassid_cgroup.c b/net/core/netclassid_cgroup.c index 1f2a126..6441f47 100644 --- a/net/core/netclassid_cgroup.c +++ b/net/core/netclassid_cgroup.c @@ -23,7 +23,8 @@ static inline struct cgroup_cls_state *css_cls_state(struct cgroup_subsys_state struct cgroup_cls_state *task_cls_state(struct task_struct *p) { - return css_cls_state(task_css(p, net_cls_cgrp_id)); + return css_cls_state(task_css_check(p, net_cls_cgrp_id, + rcu_read_lock_bh_held())); } EXPORT_SYMBOL_GPL(task_cls_state); -- cgit v0.10.2 From 548aedac5128e4bb256bab5c8e1db88e626697fc Mon Sep 17 00:00:00 2001 From: Jaegeuk Kim Date: Mon, 13 Jul 2015 17:44:14 -0700 Subject: f2fs: handle error cases in move_encrypted_block This patch fixes some missing error handlers. Reviewed-by: Chao Yu Signed-off-by: Jaegeuk Kim diff --git a/fs/f2fs/gc.c b/fs/f2fs/gc.c index e1e7361..883a841 100644 --- a/fs/f2fs/gc.c +++ b/fs/f2fs/gc.c @@ -556,27 +556,34 @@ static void move_encrypted_block(struct inode *inode, block_t bidx) if (!fio.encrypted_page) goto put_out; - f2fs_submit_page_bio(&fio); + err = f2fs_submit_page_bio(&fio); + if (err) + goto put_page_out; + + /* write page */ + lock_page(fio.encrypted_page); + + if (unlikely(!PageUptodate(fio.encrypted_page))) + goto put_page_out; + if (unlikely(fio.encrypted_page->mapping != META_MAPPING(fio.sbi))) + goto put_page_out; + + set_page_writeback(fio.encrypted_page); /* allocate block address */ f2fs_wait_on_page_writeback(dn.node_page, NODE); - allocate_data_block(fio.sbi, NULL, fio.blk_addr, &fio.blk_addr, &sum, CURSEG_COLD_DATA); - dn.data_blkaddr = fio.blk_addr; - - /* write page */ - lock_page(fio.encrypted_page); - set_page_writeback(fio.encrypted_page); fio.rw = WRITE_SYNC; f2fs_submit_page_mbio(&fio); + dn.data_blkaddr = fio.blk_addr; set_data_blkaddr(&dn); f2fs_update_extent_cache(&dn); set_inode_flag(F2FS_I(inode), FI_APPEND_WRITE); if (page->index == 0) set_inode_flag(F2FS_I(inode), FI_FIRST_BLOCK_WRITTEN); - +put_page_out: f2fs_put_page(fio.encrypted_page, 1); put_out: f2fs_put_dnode(&dn); -- cgit v0.10.2 From 6282adbf932c226f76e1b83e074448c79976fe75 Mon Sep 17 00:00:00 2001 From: Jaegeuk Kim Date: Sat, 25 Jul 2015 00:29:17 -0700 Subject: f2fs: call set_page_dirty to attach i_wb for cgroup The cgroup attaches inode->i_wb via mark_inode_dirty and when set_page_writeback is called, __inc_wb_stat() updates i_wb's stat. So, we need to explicitly call set_page_dirty->__mark_inode_dirty in prior to any writebacking pages. This patch should resolve the following kernel panic reported by Andreas Reis. https://bugzilla.kernel.org/show_bug.cgi?id=101801 --- Comment #2 from Andreas Reis --- BUG: unable to handle kernel NULL pointer dereference at 00000000000000a8 IP: [] __percpu_counter_add+0x1a/0x90 PGD 2951ff067 PUD 2df43f067 PMD 0 Oops: 0000 [#1] PREEMPT SMP Modules linked in: CPU: 7 PID: 10356 Comm: gcc Tainted: G W 4.2.0-1-cu #1 Hardware name: Gigabyte Technology Co., Ltd. G1.Sniper M5/G1.Sniper M5, BIOS T01 02/03/2015 task: ffff880295044f80 ti: ffff880295140000 task.ti: ffff880295140000 RIP: 0010:[] [] __percpu_counter_add+0x1a/0x90 RSP: 0018:ffff880295143ac8 EFLAGS: 00010082 RAX: 0000000000000003 RBX: ffffea000a526d40 RCX: 0000000000000001 RDX: 0000000000000020 RSI: 0000000000000001 RDI: 0000000000000088 RBP: ffff880295143ae8 R08: 0000000000000000 R09: ffff88008f69bb30 R10: 00000000fffffffa R11: 0000000000000000 R12: 0000000000000088 R13: 0000000000000001 R14: ffff88041d099000 R15: ffff880084a205d0 FS: 00007f8549374700(0000) GS:ffff88042f3c0000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: 00000000000000a8 CR3: 000000033e1d5000 CR4: 00000000001406e0 DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400 Stack: 0000000000000000 ffffea000a526d40 ffff880084a20738 ffff880084a20750 ffff880295143b48 ffffffff811cc91e ffff880000000000 0000000000000296 0000000000000000 ffff880417090198 0000000000000000 ffffea000a526d40 Call Trace: [] __test_set_page_writeback+0xde/0x1d0 [] do_write_data_page+0xe7/0x3a0 [] gc_data_segment+0x5aa/0x640 [] do_garbage_collect+0x138/0x150 [] f2fs_gc+0x1be/0x3e0 [] f2fs_balance_fs+0x81/0x90 [] f2fs_unlink+0x47/0x1d0 [] vfs_unlink+0x109/0x1b0 [] do_unlinkat+0x287/0x2c0 [] SyS_unlink+0x16/0x20 [] entry_SYSCALL_64_fastpath+0x12/0x71 Code: 41 5e 5d c3 0f 1f 00 66 2e 0f 1f 84 00 00 00 00 00 55 48 89 e5 41 55 49 89 f5 41 54 49 89 fc 53 48 83 ec 08 65 ff 05 e6 d9 b6 7e <48> 8b 47 20 48 63 ca 65 8b 18 48 63 db 48 01 f3 48 39 cb 7d 0a RIP [] __percpu_counter_add+0x1a/0x90 RSP CR2: 00000000000000a8 ---[ end trace 5132449a58ed93a3 ]--- note: gcc[10356] exited with preempt_count 2 Signed-off-by: Jaegeuk Kim diff --git a/fs/f2fs/data.c b/fs/f2fs/data.c index 9bedfa8..f71e19a 100644 --- a/fs/f2fs/data.c +++ b/fs/f2fs/data.c @@ -2072,8 +2072,6 @@ static int f2fs_set_data_page_dirty(struct page *page) return 1; } - mark_inode_dirty(inode); - if (!PageDirty(page)) { __set_page_dirty_nobuffers(page); update_dirty_page(inode, page); diff --git a/fs/f2fs/file.c b/fs/f2fs/file.c index ada2a3d..b0f38c3 100644 --- a/fs/f2fs/file.c +++ b/fs/f2fs/file.c @@ -1331,12 +1331,13 @@ static int f2fs_ioc_commit_atomic_write(struct file *filp) if (ret) return ret; - if (f2fs_is_atomic_file(inode)) + if (f2fs_is_atomic_file(inode)) { + clear_inode_flag(F2FS_I(inode), FI_ATOMIC_FILE); commit_inmem_pages(inode, false); + } ret = f2fs_sync_file(filp, 0, LONG_MAX, 0); mnt_drop_write_file(filp); - clear_inode_flag(F2FS_I(inode), FI_ATOMIC_FILE); return ret; } @@ -1387,8 +1388,8 @@ static int f2fs_ioc_abort_volatile_write(struct file *filp) f2fs_balance_fs(F2FS_I_SB(inode)); if (f2fs_is_atomic_file(inode)) { - commit_inmem_pages(inode, false); clear_inode_flag(F2FS_I(inode), FI_ATOMIC_FILE); + commit_inmem_pages(inode, false); } if (f2fs_is_volatile_file(inode)) diff --git a/fs/f2fs/gc.c b/fs/f2fs/gc.c index 883a841..22fb5ef 100644 --- a/fs/f2fs/gc.c +++ b/fs/f2fs/gc.c @@ -568,6 +568,11 @@ static void move_encrypted_block(struct inode *inode, block_t bidx) if (unlikely(fio.encrypted_page->mapping != META_MAPPING(fio.sbi))) goto put_page_out; + set_page_dirty(fio.encrypted_page); + f2fs_wait_on_page_writeback(fio.encrypted_page, META); + if (clear_page_dirty_for_io(fio.encrypted_page)) + dec_page_count(fio.sbi, F2FS_DIRTY_META); + set_page_writeback(fio.encrypted_page); /* allocate block address */ @@ -612,8 +617,8 @@ static void move_data_page(struct inode *inode, block_t bidx, int gc_type) .page = page, .encrypted_page = NULL, }; + set_page_dirty(page); f2fs_wait_on_page_writeback(page, DATA); - if (clear_page_dirty_for_io(page)) inode_dec_dirty_pages(inode); set_cold_data(page); diff --git a/fs/f2fs/inline.c b/fs/f2fs/inline.c index 38e75fb..a13ffcc 100644 --- a/fs/f2fs/inline.c +++ b/fs/f2fs/inline.c @@ -141,6 +141,8 @@ int f2fs_convert_inline_page(struct dnode_of_data *dn, struct page *page) kunmap_atomic(dst_addr); SetPageUptodate(page); no_update: + set_page_dirty(page); + /* clear dirty state */ dirty = clear_page_dirty_for_io(page); diff --git a/fs/f2fs/segment.c b/fs/f2fs/segment.c index 1eb3437..61b97f9 100644 --- a/fs/f2fs/segment.c +++ b/fs/f2fs/segment.c @@ -257,6 +257,7 @@ void commit_inmem_pages(struct inode *inode, bool abort) if (!abort) { lock_page(cur->page); if (cur->page->mapping == inode->i_mapping) { + set_page_dirty(cur->page); f2fs_wait_on_page_writeback(cur->page, DATA); if (clear_page_dirty_for_io(cur->page)) inode_dec_dirty_pages(inode); -- cgit v0.10.2 From 8ca243536d21ae2d08f61b1c5af4ac3d4bb697e4 Mon Sep 17 00:00:00 2001 From: Dan Williams Date: Fri, 24 Jul 2015 23:42:34 -0400 Subject: libnvdimm: fix namespace seed creation A new BLK namespace "seed" device is created whenever the current seed is successfully probed. However, if that namespace is assigned to a BTT it may never directly experience a successful probe as it is a subordinate device to a BTT configuration. The effect of the current code is that no new namespaces can be instantiated, after the seed namespace, to consume available BLK DPA capacity. Fix this by treating a successful BTT probe event as a successful probe event for the backing namespace. Reported-by: Nicholas Moulin Signed-off-by: Dan Williams diff --git a/drivers/nvdimm/region_devs.c b/drivers/nvdimm/region_devs.c index a523342..7384455 100644 --- a/drivers/nvdimm/region_devs.c +++ b/drivers/nvdimm/region_devs.c @@ -458,10 +458,15 @@ static void nd_region_notify_driver_action(struct nvdimm_bus *nvdimm_bus, nvdimm_bus_unlock(dev); } if (is_nd_btt(dev) && probe) { + struct nd_btt *nd_btt = to_nd_btt(dev); + nd_region = to_nd_region(dev->parent); nvdimm_bus_lock(dev); if (nd_region->btt_seed == dev) nd_region_create_btt_seed(nd_region); + if (nd_region->ns_seed == &nd_btt->ndns->dev && + is_nd_blk(dev->parent)) + nd_region_create_blk_seed(nd_region); nvdimm_bus_unlock(dev); } } -- cgit v0.10.2 From 4e5a74f1db8d6bea48306126fd445e2720f07a95 Mon Sep 17 00:00:00 2001 From: Sudip Mukherjee Date: Sat, 25 Jul 2015 13:19:40 +0530 Subject: parport: Revert "parport: fix memory leak" This reverts commit 23c405912b88 ("parport: fix memory leak") par_dev->state was already being removed in parport_unregister_device(). Reported-by: Ying Huang Signed-off-by: Sudip Mukherjee Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/parport/share.c b/drivers/parport/share.c index c02b5f2..5ce5ef2 100644 --- a/drivers/parport/share.c +++ b/drivers/parport/share.c @@ -816,7 +816,6 @@ static void free_pardevice(struct device *dev) struct pardevice *par_dev = to_pardevice(dev); kfree(par_dev->name); - kfree(par_dev->state); kfree(par_dev); } -- cgit v0.10.2 From 2c534c0da0a68418693e10ce1c4146e085f39518 Mon Sep 17 00:00:00 2001 From: Matt Fleming Date: Tue, 21 Jul 2015 15:55:09 +0100 Subject: perf/x86/intel/cqm: Return cached counter value from IRQ context Peter reported the following potential crash which I was able to reproduce with his test program, [ 148.765788] ------------[ cut here ]------------ [ 148.765796] WARNING: CPU: 34 PID: 2840 at kernel/smp.c:417 smp_call_function_many+0xb6/0x260() [ 148.765797] Modules linked in: [ 148.765800] CPU: 34 PID: 2840 Comm: perf Not tainted 4.2.0-rc1+ #4 [ 148.765803] ffffffff81cdc398 ffff88085f105950 ffffffff818bdfd5 0000000000000007 [ 148.765805] 0000000000000000 ffff88085f105990 ffffffff810e413a 0000000000000000 [ 148.765807] ffffffff82301080 0000000000000022 ffffffff8107f640 ffffffff8107f640 [ 148.765809] Call Trace: [ 148.765810] [] dump_stack+0x45/0x57 [ 148.765818] [] warn_slowpath_common+0x8a/0xc0 [ 148.765822] [] ? intel_cqm_stable+0x60/0x60 [ 148.765824] [] ? intel_cqm_stable+0x60/0x60 [ 148.765825] [] warn_slowpath_null+0x1a/0x20 [ 148.765827] [] smp_call_function_many+0xb6/0x260 [ 148.765829] [] ? intel_cqm_stable+0x60/0x60 [ 148.765831] [] on_each_cpu_mask+0x28/0x60 [ 148.765832] [] intel_cqm_event_count+0x7f/0xe0 [ 148.765836] [] perf_output_read+0x2a5/0x400 [ 148.765839] [] perf_output_sample+0x31a/0x590 [ 148.765840] [] ? perf_prepare_sample+0x26d/0x380 [ 148.765841] [] perf_event_output+0x47/0x60 [ 148.765843] [] __perf_event_overflow+0x215/0x240 [ 148.765844] [] perf_event_overflow+0x14/0x20 [ 148.765847] [] intel_pmu_handle_irq+0x1d4/0x440 [ 148.765849] [] ? __perf_event_task_sched_in+0x36/0xa0 [ 148.765853] [] ? vunmap_page_range+0x19d/0x2f0 [ 148.765854] [] ? unmap_kernel_range_noflush+0x11/0x20 [ 148.765859] [] ? ghes_copy_tofrom_phys+0x11e/0x2a0 [ 148.765863] [] ? native_apic_msr_write+0x2b/0x30 [ 148.765865] [] ? x2apic_send_IPI_self+0x1d/0x20 [ 148.765869] [] ? arch_irq_work_raise+0x35/0x40 [ 148.765872] [] ? irq_work_queue+0x66/0x80 [ 148.765875] [] perf_event_nmi_handler+0x26/0x40 [ 148.765877] [] nmi_handle+0x79/0x100 [ 148.765879] [] default_do_nmi+0x42/0x100 [ 148.765880] [] do_nmi+0x83/0xb0 [ 148.765884] [] end_repeat_nmi+0x1e/0x2e [ 148.765886] [] ? __perf_event_task_sched_in+0x36/0xa0 [ 148.765888] [] ? __perf_event_task_sched_in+0x36/0xa0 [ 148.765890] [] ? __perf_event_task_sched_in+0x36/0xa0 [ 148.765891] <> [] finish_task_switch+0x156/0x210 [ 148.765898] [] __schedule+0x341/0x920 [ 148.765899] [] schedule+0x37/0x80 [ 148.765903] [] ? do_page_fault+0x2f/0x80 [ 148.765905] [] schedule_user+0x1a/0x50 [ 148.765907] [] retint_careful+0x14/0x32 [ 148.765908] ---[ end trace e33ff2be78e14901 ]--- The CQM task events are not safe to be called from within interrupt context because they require performing an IPI to read the counter value on all sockets. And performing IPIs from within IRQ context is a "no-no". Make do with the last read counter value currently event in event->count when we're invoked in this context. Reported-by: Peter Zijlstra Signed-off-by: Matt Fleming Cc: Thomas Gleixner Cc: Vikas Shivappa Cc: Kanaka Juvva Cc: Will Auld Cc: Link: http://lkml.kernel.org/r/1437490509-15373-1-git-send-email-matt@codeblueprint.co.uk Signed-off-by: Thomas Gleixner diff --git a/arch/x86/kernel/cpu/perf_event_intel_cqm.c b/arch/x86/kernel/cpu/perf_event_intel_cqm.c index 1880761..63eb68b 100644 --- a/arch/x86/kernel/cpu/perf_event_intel_cqm.c +++ b/arch/x86/kernel/cpu/perf_event_intel_cqm.c @@ -952,6 +952,14 @@ static u64 intel_cqm_event_count(struct perf_event *event) return 0; /* + * Getting up-to-date values requires an SMP IPI which is not + * possible if we're being called in interrupt context. Return + * the cached values instead. + */ + if (unlikely(in_interrupt())) + goto out; + + /* * Notice that we don't perform the reading of an RMID * atomically, because we can't hold a spin lock across the * IPIs. -- cgit v0.10.2 From 1a4e8795711f474b31ff6eac37f3efd304ed8a93 Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Sun, 26 Jul 2015 10:27:37 +0200 Subject: x86/mm/pat: Revert 'Adjust default caching mode translation tables' Toshi explains: "No, the default values need to be set to the fallback types, i.e. minimal supported mode. For WC and WT, UC is the fallback type. When PAT is disabled, pat_init() does update the tables below to enable WT per the default BIOS setup. However, when PAT is enabled, but CPU has PAT -errata, WT falls back to UC per the default values." Revert: ca1fec58bc6a 'x86/mm/pat: Adjust default caching mode translation tables' Requested-by: Toshi Kani Cc: Jan Beulich Link: http://lkml.kernel.org/r/1437577776.3214.252.camel@hp.com Signed-off-by: Thomas Gleixner diff --git a/arch/x86/mm/init.c b/arch/x86/mm/init.c index 7a45322..8533b46 100644 --- a/arch/x86/mm/init.c +++ b/arch/x86/mm/init.c @@ -43,18 +43,18 @@ uint16_t __cachemode2pte_tbl[_PAGE_CACHE_MODE_NUM] = { [_PAGE_CACHE_MODE_WC ] = 0 | _PAGE_PCD, [_PAGE_CACHE_MODE_UC_MINUS] = 0 | _PAGE_PCD, [_PAGE_CACHE_MODE_UC ] = _PAGE_PWT | _PAGE_PCD, - [_PAGE_CACHE_MODE_WT ] = _PAGE_PWT | 0, + [_PAGE_CACHE_MODE_WT ] = 0 | _PAGE_PCD, [_PAGE_CACHE_MODE_WP ] = 0 | _PAGE_PCD, }; EXPORT_SYMBOL(__cachemode2pte_tbl); uint8_t __pte2cachemode_tbl[8] = { [__pte2cm_idx( 0 | 0 | 0 )] = _PAGE_CACHE_MODE_WB, - [__pte2cm_idx(_PAGE_PWT | 0 | 0 )] = _PAGE_CACHE_MODE_WT, + [__pte2cm_idx(_PAGE_PWT | 0 | 0 )] = _PAGE_CACHE_MODE_UC_MINUS, [__pte2cm_idx( 0 | _PAGE_PCD | 0 )] = _PAGE_CACHE_MODE_UC_MINUS, [__pte2cm_idx(_PAGE_PWT | _PAGE_PCD | 0 )] = _PAGE_CACHE_MODE_UC, [__pte2cm_idx( 0 | 0 | _PAGE_PAT)] = _PAGE_CACHE_MODE_WB, - [__pte2cm_idx(_PAGE_PWT | 0 | _PAGE_PAT)] = _PAGE_CACHE_MODE_WT, + [__pte2cm_idx(_PAGE_PWT | 0 | _PAGE_PAT)] = _PAGE_CACHE_MODE_UC_MINUS, [__pte2cm_idx(0 | _PAGE_PCD | _PAGE_PAT)] = _PAGE_CACHE_MODE_UC_MINUS, [__pte2cm_idx(_PAGE_PWT | _PAGE_PCD | _PAGE_PAT)] = _PAGE_CACHE_MODE_UC, }; -- cgit v0.10.2 From cbfe8fa6cd672011c755c3cd85c9ffd4e2d10a6f Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Sun, 26 Jul 2015 12:26:21 -0700 Subject: Linux 4.2-rc4 diff --git a/Makefile b/Makefile index a9ad490..afabc44 100644 --- a/Makefile +++ b/Makefile @@ -1,7 +1,7 @@ VERSION = 4 PATCHLEVEL = 2 SUBLEVEL = 0 -EXTRAVERSION = -rc3 +EXTRAVERSION = -rc4 NAME = Hurr durr I'ma sheep # *DOCUMENTATION* -- cgit v0.10.2 From a153790a788392e75d37262a9c5960d1d91ddb7e Mon Sep 17 00:00:00 2001 From: Linus Walleij Date: Sat, 25 Jul 2015 11:22:03 +0200 Subject: ARM: nomadik: disable UART0 on Nomadik boards The UART0 is not used on these boards, yet active and blocking other use. Fix this by disabling UART0 and setting port aliases to maintain port enumeration to userspace. Signed-off-by: Linus Walleij Signed-off-by: Olof Johansson diff --git a/arch/arm/boot/dts/ste-nomadik-nhk15.dts b/arch/arm/boot/dts/ste-nomadik-nhk15.dts index 3d0b875..3d25dba 100644 --- a/arch/arm/boot/dts/ste-nomadik-nhk15.dts +++ b/arch/arm/boot/dts/ste-nomadik-nhk15.dts @@ -17,6 +17,7 @@ }; aliases { + serial1 = &uart1; stmpe-i2c0 = &stmpe0; stmpe-i2c1 = &stmpe1; }; diff --git a/arch/arm/boot/dts/ste-nomadik-s8815.dts b/arch/arm/boot/dts/ste-nomadik-s8815.dts index 85d3b95..3c140d0 100644 --- a/arch/arm/boot/dts/ste-nomadik-s8815.dts +++ b/arch/arm/boot/dts/ste-nomadik-s8815.dts @@ -15,6 +15,10 @@ bootargs = "root=/dev/ram0 console=ttyAMA1,115200n8 earlyprintk"; }; + aliases { + serial1 = &uart1; + }; + src@101e0000 { /* These chrystal drivers are not used on this board */ disable-sxtalo; diff --git a/arch/arm/boot/dts/ste-nomadik-stn8815.dtsi b/arch/arm/boot/dts/ste-nomadik-stn8815.dtsi index 9a5f2ba..ef794a3 100644 --- a/arch/arm/boot/dts/ste-nomadik-stn8815.dtsi +++ b/arch/arm/boot/dts/ste-nomadik-stn8815.dtsi @@ -757,6 +757,7 @@ clock-names = "uartclk", "apb_pclk"; pinctrl-names = "default"; pinctrl-0 = <&uart0_default_mux>; + status = "disabled"; }; uart1: uart@101fb000 { -- cgit v0.10.2 From 0a927c2f02a2437a57679527e42ab7cbfa14efb1 Mon Sep 17 00:00:00 2001 From: Mike Snitzer Date: Tue, 21 Jul 2015 13:20:46 -0400 Subject: dm thin: return -ENOSPC when erroring retry list due to out of data space Otherwise -EIO would be returned when -ENOSPC should be used consistently. Signed-off-by: Mike Snitzer diff --git a/drivers/md/dm-thin.c b/drivers/md/dm-thin.c index 1c50c58..d2bbe8c 100644 --- a/drivers/md/dm-thin.c +++ b/drivers/md/dm-thin.c @@ -666,16 +666,21 @@ static void requeue_io(struct thin_c *tc) requeue_deferred_cells(tc); } -static void error_retry_list(struct pool *pool) +static void error_retry_list_with_code(struct pool *pool, int error) { struct thin_c *tc; rcu_read_lock(); list_for_each_entry_rcu(tc, &pool->active_thins, list) - error_thin_bio_list(tc, &tc->retry_on_resume_list, -EIO); + error_thin_bio_list(tc, &tc->retry_on_resume_list, error); rcu_read_unlock(); } +static void error_retry_list(struct pool *pool) +{ + return error_retry_list_with_code(pool, -EIO); +} + /* * This section of code contains the logic for processing a thin device's IO. * Much of the code depends on pool object resources (lists, workqueues, etc) @@ -2297,7 +2302,7 @@ static void do_no_space_timeout(struct work_struct *ws) if (get_pool_mode(pool) == PM_OUT_OF_DATA_SPACE && !pool->pf.error_if_no_space) { pool->pf.error_if_no_space = true; notify_of_pool_mode_change_to_oods(pool); - error_retry_list(pool); + error_retry_list_with_code(pool, -ENOSPC); } } -- cgit v0.10.2 From 963ad94853000ab100f5ff19eea80095660d41b4 Mon Sep 17 00:00:00 2001 From: Nikolay Aleksandrov Date: Wed, 22 Jul 2015 13:03:40 +0200 Subject: bridge: netlink: fix slave_changelink/br_setport race conditions Since slave_changelink support was added there have been a few race conditions when using br_setport() since some of the port functions it uses require the bridge lock. It is very easy to trigger a lockup due to some internal spin_lock() usage without bh disabled, also it's possible to get the bridge into an inconsistent state. Signed-off-by: Nikolay Aleksandrov Fixes: 3ac636b8591c ("bridge: implement rtnl_link_ops->slave_changelink") Reviewed-by: Jiri Pirko Signed-off-by: David S. Miller diff --git a/net/bridge/br_netlink.c b/net/bridge/br_netlink.c index 364bdc9..3da5525 100644 --- a/net/bridge/br_netlink.c +++ b/net/bridge/br_netlink.c @@ -693,9 +693,17 @@ static int br_port_slave_changelink(struct net_device *brdev, struct nlattr *tb[], struct nlattr *data[]) { + struct net_bridge *br = netdev_priv(brdev); + int ret; + if (!data) return 0; - return br_setport(br_port_get_rtnl(dev), data); + + spin_lock_bh(&br->lock); + ret = br_setport(br_port_get_rtnl(dev), data); + spin_unlock_bh(&br->lock); + + return ret; } static int br_port_fill_slave_info(struct sk_buff *skb, -- cgit v0.10.2 From 178d23e3cd4811ebe702d60ac31e8bee389a5847 Mon Sep 17 00:00:00 2001 From: Or Gerlitz Date: Wed, 22 Jul 2015 16:53:46 +0300 Subject: net/mlx4_core: Use sink counter for the VF default as fallback Some old PF drivers don't let VFs allocate counters, in that case, use the sink counter so the VF can load and operate properly. Fixes: 6de5f7f6a1fa ('net/mlx4_core: Allocate default counter per port') Reported-by: Sebastian Ott Signed-off-by: Or Gerlitz Signed-off-by: Eran Ben Elisha Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/mellanox/mlx4/main.c b/drivers/net/ethernet/mellanox/mlx4/main.c index 12fbfcb..29c2a01 100644 --- a/drivers/net/ethernet/mellanox/mlx4/main.c +++ b/drivers/net/ethernet/mellanox/mlx4/main.c @@ -2273,6 +2273,11 @@ static int mlx4_allocate_default_counters(struct mlx4_dev *dev) } else if (err == -ENOENT) { err = 0; continue; + } else if (mlx4_is_slave(dev) && err == -EINVAL) { + priv->def_counter[port] = MLX4_SINK_COUNTER_INDEX(dev); + mlx4_warn(dev, "can't allocate counter from old PF driver, using index %d\n", + MLX4_SINK_COUNTER_INDEX(dev)); + err = 0; } else { mlx4_err(dev, "%s: failed to allocate default counter port %d err %d\n", __func__, port + 1, err); -- cgit v0.10.2 From 1c1bf34951e8d17941bf708d1901c47e81b15d55 Mon Sep 17 00:00:00 2001 From: Jack Morgenstein Date: Wed, 22 Jul 2015 16:53:47 +0300 Subject: net/mlx4_core: Fix wrong index in propagating port change event to VFs The port-change event processing in procedure mlx4_eq_int() uses "slave" as the vf_oper array index. Since the value of "slave" is the PF function index, the result is that the PF link state is used for deciding to propagate the event for all the VFs. The VF link state should be used, so the VF function index should be used here. Fixes: 948e306d7d64 ('net/mlx4: Add VF link state support') Signed-off-by: Jack Morgenstein Signed-off-by: Matan Barak Signed-off-by: Or Gerlitz Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/mellanox/mlx4/eq.c b/drivers/net/ethernet/mellanox/mlx4/eq.c index aae13ad..8e81e53 100644 --- a/drivers/net/ethernet/mellanox/mlx4/eq.c +++ b/drivers/net/ethernet/mellanox/mlx4/eq.c @@ -601,7 +601,7 @@ static int mlx4_eq_int(struct mlx4_dev *dev, struct mlx4_eq *eq) continue; mlx4_dbg(dev, "%s: Sending MLX4_PORT_CHANGE_SUBTYPE_DOWN to slave: %d, port:%d\n", __func__, i, port); - s_info = &priv->mfunc.master.vf_oper[slave].vport[port].state; + s_info = &priv->mfunc.master.vf_oper[i].vport[port].state; if (IFLA_VF_LINK_STATE_AUTO == s_info->link_state) { eqe->event.port_change.port = cpu_to_be32( @@ -640,7 +640,7 @@ static int mlx4_eq_int(struct mlx4_dev *dev, struct mlx4_eq *eq) continue; if (i == mlx4_master_func_num(dev)) continue; - s_info = &priv->mfunc.master.vf_oper[slave].vport[port].state; + s_info = &priv->mfunc.master.vf_oper[i].vport[port].state; if (IFLA_VF_LINK_STATE_AUTO == s_info->link_state) { eqe->event.port_change.port = cpu_to_be32( -- cgit v0.10.2 From 9f5b031770b9108b57881c83dffc02cd90ec3961 Mon Sep 17 00:00:00 2001 From: Jack Morgenstein Date: Wed, 22 Jul 2015 16:53:48 +0300 Subject: net/mlx4_core: Relieve cpu load average on the port sending flow When a port is not attached, the FW requires a longer than usual time to execute the SENSE_PORT command. In the command flow, the wait_for_completion_timeout call used in mlx4_cmd_wait puts the kernel thread into the uninterruptible state during this time. This, in turn, due to the computation method, causes the CPU load average to increase. Fix this by using wait_for_completion_interruptible_timeout() for the SENSE_PORT command, which puts the thread in the interruptible state. In this state, the thread does not contribute to the CPU load average. Treat the interrupted case as if the SENSE_PORT command returned port_type = NONE. Fix suggested by Gideon Naim and Bart Van Assche . Signed-off-by: Jack Morgenstein Signed-off-by: Or Gerlitz Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/mellanox/mlx4/cmd.c b/drivers/net/ethernet/mellanox/mlx4/cmd.c index 8204013..0a32020 100644 --- a/drivers/net/ethernet/mellanox/mlx4/cmd.c +++ b/drivers/net/ethernet/mellanox/mlx4/cmd.c @@ -686,6 +686,7 @@ static int mlx4_cmd_wait(struct mlx4_dev *dev, u64 in_param, u64 *out_param, { struct mlx4_cmd *cmd = &mlx4_priv(dev)->cmd; struct mlx4_cmd_context *context; + long ret_wait; int err = 0; down(&cmd->event_sem); @@ -711,8 +712,20 @@ static int mlx4_cmd_wait(struct mlx4_dev *dev, u64 in_param, u64 *out_param, if (err) goto out_reset; - if (!wait_for_completion_timeout(&context->done, - msecs_to_jiffies(timeout))) { + if (op == MLX4_CMD_SENSE_PORT) { + ret_wait = + wait_for_completion_interruptible_timeout(&context->done, + msecs_to_jiffies(timeout)); + if (ret_wait < 0) { + context->fw_status = 0; + context->out_param = 0; + context->result = 0; + } + } else { + ret_wait = (long)wait_for_completion_timeout(&context->done, + msecs_to_jiffies(timeout)); + } + if (!ret_wait) { mlx4_warn(dev, "command 0x%x timed out (go bit not cleared)\n", op); if (op == MLX4_CMD_NOP) { -- cgit v0.10.2 From 62e4c9b4fd3f8fd7d51951f0f10e3f27c58d4037 Mon Sep 17 00:00:00 2001 From: Ido Shamay Date: Wed, 22 Jul 2015 16:53:49 +0300 Subject: net/mlx4_en: Remove BUG_ON assert when checking if ring is full In mlx4_en_is_ring_empty we check if ring surpassed its size. Since the prod and cons indicators are u32, there might be a state where prod wrapped around and cons, making this assert false, although no actual bug exists (other code segment can cope with this state). Signed-off-by: Ido Shamay Signed-off-by: Eugenia Emantayev Signed-off-by: Or Gerlitz Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/mellanox/mlx4/en_rx.c b/drivers/net/ethernet/mellanox/mlx4/en_rx.c index 7a4f20b..9c145dd 100644 --- a/drivers/net/ethernet/mellanox/mlx4/en_rx.c +++ b/drivers/net/ethernet/mellanox/mlx4/en_rx.c @@ -246,7 +246,6 @@ static int mlx4_en_prepare_rx_desc(struct mlx4_en_priv *priv, static inline bool mlx4_en_is_ring_empty(struct mlx4_en_rx_ring *ring) { - BUG_ON((u32)(ring->prod - ring->cons) > ring->actual_size); return ring->prod == ring->cons; } -- cgit v0.10.2 From 81296fc67319d96ea6f7f43a07494394e1236a19 Mon Sep 17 00:00:00 2001 From: Daniel Borkmann Date: Wed, 22 Jul 2015 16:31:49 +0200 Subject: net: sctp: stop spamming klog with rfc6458, 5.3.2. deprecation warnings Back then when we added support for SCTP_SNDINFO/SCTP_RCVINFO from RFC6458 5.3.4/5.3.5, we decided to add a deprecation warning for the (as per RFC deprecated) SCTP_SNDRCV via commit bbbea41d5e53 ("net: sctp: deprecate rfc6458, 5.3.2. SCTP_SNDRCV support"), see [1]. Imho, it was not a good idea, and we should just revert that message for a couple of reasons: 1) It's uapi and therefore set in stone forever. 2) To be able to run on older and newer kernels, an SCTP application would need to probe for both, SCTP_SNDRCV, but also SCTP_SNDINFO/ SCTP_RCVINFO support, so that on older kernels, it can make use of SCTP_SNDRCV, and on newer kernels SCTP_SNDINFO/SCTP_RCVINFO. In my (limited) experience, a lot of SCTP appliances are migrating to newer kernels only ve(ee)ry slowly. 3) Some people don't have the chance to change their applications, f.e. due to proprietary legacy stuff. So, they'll hit this warning in fast path and are stuck with older kernels. But i.e. due to point 1) I really fail to see the benefit of a warning. So just revert that for now, the issue was reported up Jamal. [1] http://thread.gmane.org/gmane.linux.network/321960/ Reported-by: Jamal Hadi Salim Signed-off-by: Daniel Borkmann Cc: Michael Tuexen Acked-by: Jamal Hadi Salim Signed-off-by: David S. Miller diff --git a/net/sctp/socket.c b/net/sctp/socket.c index 1425ec2..17bef01 100644 --- a/net/sctp/socket.c +++ b/net/sctp/socket.c @@ -2200,12 +2200,6 @@ static int sctp_setsockopt_events(struct sock *sk, char __user *optval, if (copy_from_user(&sctp_sk(sk)->subscribe, optval, optlen)) return -EFAULT; - if (sctp_sk(sk)->subscribe.sctp_data_io_event) - pr_warn_ratelimited(DEPRECATED "%s (pid %d) " - "Requested SCTP_SNDRCVINFO event.\n" - "Use SCTP_RCVINFO through SCTP_RECVRCVINFO option instead.\n", - current->comm, task_pid_nr(current)); - /* At the time when a user app subscribes to SCTP_SENDER_DRY_EVENT, * if there is no data to be sent or retransmit, the stack will * immediately send up this notification. -- cgit v0.10.2 From 69cefc273f942bd7bb347a02e8b5b738d5f6e6f3 Mon Sep 17 00:00:00 2001 From: Lukasz Anaczkowski Date: Tue, 21 Jul 2015 10:41:13 +0200 Subject: intel_pstate: Add get_scaling cpu_defaults param to Knights Landing Scaling for Knights Landing is same as the default scaling (100000). When Knigts Landing support was added to the pstate driver, this parameter was omitted resulting in a kernel panic during boot. Fixes: b34ef932d79a (intel_pstate: Knights Landing support) Reported-by: Yasuaki Ishimatsu Signed-off-by: Dasaratharaman Chandramouli Signed-off-by: Lukasz Anaczkowski Acked-by: Kristen Carlson Accardi Cc: 4.1+ # 4.1+ Signed-off-by: Rafael J. Wysocki diff --git a/drivers/cpufreq/intel_pstate.c b/drivers/cpufreq/intel_pstate.c index 15ada47..fcb929e 100644 --- a/drivers/cpufreq/intel_pstate.c +++ b/drivers/cpufreq/intel_pstate.c @@ -681,6 +681,7 @@ static struct cpu_defaults knl_params = { .get_max = core_get_max_pstate, .get_min = core_get_min_pstate, .get_turbo = knl_get_turbo_pstate, + .get_scaling = core_get_scaling, .set = core_set_pstate, }, }; -- cgit v0.10.2 From d1fe19444d82e399e38c1594c71b850eca8e9de0 Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Thu, 23 Jul 2015 12:05:37 +0200 Subject: inet: frag: don't re-use chainlist for evictor commit 65ba1f1ec0eff ("inet: frags: fix a race between inet_evict_bucket and inet_frag_kill") describes the bug, but the fix doesn't work reliably. Problem is that ->flags member can be set on other cpu without chainlock being held by that task, i.e. the RMW-Cycle can clear INET_FRAG_EVICTED bit after we put the element on the evictor private list. We can crash when walking the 'private' evictor list since an element can be deleted from list underneath the evictor. Join work with Nikolay Alexandrov. Fixes: b13d3cbfb8e8 ("inet: frag: move eviction of queues to work queue") Reported-by: Johan Schuijt Tested-by: Frank Schreuder Signed-off-by: Nikolay Alexandrov Signed-off-by: Florian Westphal Signed-off-by: David S. Miller diff --git a/include/net/inet_frag.h b/include/net/inet_frag.h index e1300b3..56a3a56 100644 --- a/include/net/inet_frag.h +++ b/include/net/inet_frag.h @@ -45,6 +45,7 @@ enum { * @flags: fragment queue flags * @max_size: maximum received fragment size * @net: namespace that this frag belongs to + * @list_evictor: list of queues to forcefully evict (e.g. due to low memory) */ struct inet_frag_queue { spinlock_t lock; @@ -59,6 +60,7 @@ struct inet_frag_queue { __u8 flags; u16 max_size; struct netns_frags *net; + struct hlist_node list_evictor; }; #define INETFRAGS_HASHSZ 1024 diff --git a/net/ipv4/inet_fragment.c b/net/ipv4/inet_fragment.c index 5e346a0..1722348 100644 --- a/net/ipv4/inet_fragment.c +++ b/net/ipv4/inet_fragment.c @@ -151,14 +151,13 @@ evict_again: } fq->flags |= INET_FRAG_EVICTED; - hlist_del(&fq->list); - hlist_add_head(&fq->list, &expired); + hlist_add_head(&fq->list_evictor, &expired); ++evicted; } spin_unlock(&hb->chain_lock); - hlist_for_each_entry_safe(fq, n, &expired, list) + hlist_for_each_entry_safe(fq, n, &expired, list_evictor) f->frag_expire((unsigned long) fq); return evicted; @@ -284,8 +283,7 @@ static inline void fq_unlink(struct inet_frag_queue *fq, struct inet_frags *f) struct inet_frag_bucket *hb; hb = get_frag_bucket_locked(fq, f); - if (!(fq->flags & INET_FRAG_EVICTED)) - hlist_del(&fq->list); + hlist_del(&fq->list); spin_unlock(&hb->chain_lock); } -- cgit v0.10.2 From 0e60d245a0be7fdbb723607f1d6621007916b252 Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Thu, 23 Jul 2015 12:05:38 +0200 Subject: inet: frag: change *_frag_mem_limit functions to take netns_frags as argument Followup patch will call it after inet_frag_queue was freed, so q->net doesn't work anymore (but netf = q->net; free(q); mem_limit(netf) would). Tested-by: Frank Schreuder Signed-off-by: Florian Westphal Signed-off-by: David S. Miller diff --git a/include/net/inet_frag.h b/include/net/inet_frag.h index 56a3a56..e71ca17 100644 --- a/include/net/inet_frag.h +++ b/include/net/inet_frag.h @@ -141,14 +141,14 @@ static inline int frag_mem_limit(struct netns_frags *nf) return percpu_counter_read(&nf->mem); } -static inline void sub_frag_mem_limit(struct inet_frag_queue *q, int i) +static inline void sub_frag_mem_limit(struct netns_frags *nf, int i) { - __percpu_counter_add(&q->net->mem, -i, frag_percpu_counter_batch); + __percpu_counter_add(&nf->mem, -i, frag_percpu_counter_batch); } -static inline void add_frag_mem_limit(struct inet_frag_queue *q, int i) +static inline void add_frag_mem_limit(struct netns_frags *nf, int i) { - __percpu_counter_add(&q->net->mem, i, frag_percpu_counter_batch); + __percpu_counter_add(&nf->mem, i, frag_percpu_counter_batch); } static inline void init_frag_mem_limit(struct netns_frags *nf) diff --git a/net/ieee802154/6lowpan/reassembly.c b/net/ieee802154/6lowpan/reassembly.c index f46e4d1..214d44a 100644 --- a/net/ieee802154/6lowpan/reassembly.c +++ b/net/ieee802154/6lowpan/reassembly.c @@ -207,7 +207,7 @@ found: } else { fq->q.meat += skb->len; } - add_frag_mem_limit(&fq->q, skb->truesize); + add_frag_mem_limit(fq->q.net, skb->truesize); if (fq->q.flags == (INET_FRAG_FIRST_IN | INET_FRAG_LAST_IN) && fq->q.meat == fq->q.len) { @@ -287,7 +287,7 @@ static int lowpan_frag_reasm(struct lowpan_frag_queue *fq, struct sk_buff *prev, clone->data_len = clone->len; head->data_len -= clone->len; head->len -= clone->len; - add_frag_mem_limit(&fq->q, clone->truesize); + add_frag_mem_limit(fq->q.net, clone->truesize); } WARN_ON(head == NULL); @@ -310,7 +310,7 @@ static int lowpan_frag_reasm(struct lowpan_frag_queue *fq, struct sk_buff *prev, } fp = next; } - sub_frag_mem_limit(&fq->q, sum_truesize); + sub_frag_mem_limit(fq->q.net, sum_truesize); head->next = NULL; head->dev = dev; diff --git a/net/ipv4/inet_fragment.c b/net/ipv4/inet_fragment.c index 1722348..4473232 100644 --- a/net/ipv4/inet_fragment.c +++ b/net/ipv4/inet_fragment.c @@ -328,7 +328,7 @@ void inet_frag_destroy(struct inet_frag_queue *q, struct inet_frags *f) fp = xp; } sum = sum_truesize + f->qsize; - sub_frag_mem_limit(q, sum); + sub_frag_mem_limit(q->net, sum); if (f->destructor) f->destructor(q); @@ -388,7 +388,7 @@ static struct inet_frag_queue *inet_frag_alloc(struct netns_frags *nf, q->net = nf; f->constructor(q, arg); - add_frag_mem_limit(q, f->qsize); + add_frag_mem_limit(nf, f->qsize); setup_timer(&q->timer, f->frag_expire, (unsigned long)q); spin_lock_init(&q->lock); diff --git a/net/ipv4/ip_fragment.c b/net/ipv4/ip_fragment.c index 31f71b1..b4a77d02 100644 --- a/net/ipv4/ip_fragment.c +++ b/net/ipv4/ip_fragment.c @@ -309,7 +309,7 @@ static int ip_frag_reinit(struct ipq *qp) kfree_skb(fp); fp = xp; } while (fp); - sub_frag_mem_limit(&qp->q, sum_truesize); + sub_frag_mem_limit(qp->q.net, sum_truesize); qp->q.flags = 0; qp->q.len = 0; @@ -455,7 +455,7 @@ found: qp->q.fragments = next; qp->q.meat -= free_it->len; - sub_frag_mem_limit(&qp->q, free_it->truesize); + sub_frag_mem_limit(qp->q.net, free_it->truesize); kfree_skb(free_it); } } @@ -479,7 +479,7 @@ found: qp->q.stamp = skb->tstamp; qp->q.meat += skb->len; qp->ecn |= ecn; - add_frag_mem_limit(&qp->q, skb->truesize); + add_frag_mem_limit(qp->q.net, skb->truesize); if (offset == 0) qp->q.flags |= INET_FRAG_FIRST_IN; @@ -587,7 +587,7 @@ static int ip_frag_reasm(struct ipq *qp, struct sk_buff *prev, head->len -= clone->len; clone->csum = 0; clone->ip_summed = head->ip_summed; - add_frag_mem_limit(&qp->q, clone->truesize); + add_frag_mem_limit(qp->q.net, clone->truesize); } skb_push(head, head->data - skb_network_header(head)); @@ -615,7 +615,7 @@ static int ip_frag_reasm(struct ipq *qp, struct sk_buff *prev, } fp = next; } - sub_frag_mem_limit(&qp->q, sum_truesize); + sub_frag_mem_limit(qp->q.net, sum_truesize); head->next = NULL; head->dev = dev; diff --git a/net/ipv6/netfilter/nf_conntrack_reasm.c b/net/ipv6/netfilter/nf_conntrack_reasm.c index 6f187c8..6d02498 100644 --- a/net/ipv6/netfilter/nf_conntrack_reasm.c +++ b/net/ipv6/netfilter/nf_conntrack_reasm.c @@ -348,7 +348,7 @@ found: fq->ecn |= ecn; if (payload_len > fq->q.max_size) fq->q.max_size = payload_len; - add_frag_mem_limit(&fq->q, skb->truesize); + add_frag_mem_limit(fq->q.net, skb->truesize); /* The first fragment. * nhoffset is obtained from the first fragment, of course. @@ -430,7 +430,7 @@ nf_ct_frag6_reasm(struct frag_queue *fq, struct net_device *dev) clone->ip_summed = head->ip_summed; NFCT_FRAG6_CB(clone)->orig = NULL; - add_frag_mem_limit(&fq->q, clone->truesize); + add_frag_mem_limit(fq->q.net, clone->truesize); } /* We have to remove fragment header from datagram and to relocate @@ -454,7 +454,7 @@ nf_ct_frag6_reasm(struct frag_queue *fq, struct net_device *dev) head->csum = csum_add(head->csum, fp->csum); head->truesize += fp->truesize; } - sub_frag_mem_limit(&fq->q, head->truesize); + sub_frag_mem_limit(fq->q.net, head->truesize); head->ignore_df = 1; head->next = NULL; diff --git a/net/ipv6/reassembly.c b/net/ipv6/reassembly.c index 8ffa2c8..5c3bbca 100644 --- a/net/ipv6/reassembly.c +++ b/net/ipv6/reassembly.c @@ -330,7 +330,7 @@ found: fq->q.stamp = skb->tstamp; fq->q.meat += skb->len; fq->ecn |= ecn; - add_frag_mem_limit(&fq->q, skb->truesize); + add_frag_mem_limit(fq->q.net, skb->truesize); /* The first fragment. * nhoffset is obtained from the first fragment, of course. @@ -443,7 +443,7 @@ static int ip6_frag_reasm(struct frag_queue *fq, struct sk_buff *prev, head->len -= clone->len; clone->csum = 0; clone->ip_summed = head->ip_summed; - add_frag_mem_limit(&fq->q, clone->truesize); + add_frag_mem_limit(fq->q.net, clone->truesize); } /* We have to remove fragment header from datagram and to relocate @@ -481,7 +481,7 @@ static int ip6_frag_reasm(struct frag_queue *fq, struct sk_buff *prev, } fp = next; } - sub_frag_mem_limit(&fq->q, sum_truesize); + sub_frag_mem_limit(fq->q.net, sum_truesize); head->next = NULL; head->dev = dev; -- cgit v0.10.2 From 5719b296fb81502d0dbbb4e87b3235e5bdcdfc6b Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Thu, 23 Jul 2015 12:05:39 +0200 Subject: inet: frag: don't wait for timer deletion when evicting Frank reports 'NMI watchdog: BUG: soft lockup' errors when load is high. Instead of (potentially) unbounded restarts of the eviction process, just skip to the next entry. One caveat is that, when a netns is exiting, a timer may still be running by the time inet_evict_bucket returns. We use the frag memory accounting to wait for outstanding timers, so that when we free the percpu counter we can be sure no running timer will trip over it. Reported-and-tested-by: Frank Schreuder Signed-off-by: Florian Westphal Signed-off-by: David S. Miller diff --git a/net/ipv4/inet_fragment.c b/net/ipv4/inet_fragment.c index 4473232..a00ca4c 100644 --- a/net/ipv4/inet_fragment.c +++ b/net/ipv4/inet_fragment.c @@ -131,24 +131,14 @@ inet_evict_bucket(struct inet_frags *f, struct inet_frag_bucket *hb) unsigned int evicted = 0; HLIST_HEAD(expired); -evict_again: spin_lock(&hb->chain_lock); hlist_for_each_entry_safe(fq, n, &hb->chain, list) { if (!inet_fragq_should_evict(fq)) continue; - if (!del_timer(&fq->timer)) { - /* q expiring right now thus increment its refcount so - * it won't be freed under us and wait until the timer - * has finished executing then destroy it - */ - atomic_inc(&fq->refcnt); - spin_unlock(&hb->chain_lock); - del_timer_sync(&fq->timer); - inet_frag_put(fq, f); - goto evict_again; - } + if (!del_timer(&fq->timer)) + continue; fq->flags |= INET_FRAG_EVICTED; hlist_add_head(&fq->list_evictor, &expired); @@ -239,18 +229,20 @@ void inet_frags_exit_net(struct netns_frags *nf, struct inet_frags *f) int i; nf->low_thresh = 0; - local_bh_disable(); evict_again: + local_bh_disable(); seq = read_seqbegin(&f->rnd_seqlock); for (i = 0; i < INETFRAGS_HASHSZ ; i++) inet_evict_bucket(f, &f->hash[i]); - if (read_seqretry(&f->rnd_seqlock, seq)) - goto evict_again; - local_bh_enable(); + cond_resched(); + + if (read_seqretry(&f->rnd_seqlock, seq) || + percpu_counter_sum(&nf->mem)) + goto evict_again; percpu_counter_destroy(&nf->mem); } @@ -284,6 +276,7 @@ static inline void fq_unlink(struct inet_frag_queue *fq, struct inet_frags *f) hb = get_frag_bucket_locked(fq, f); hlist_del(&fq->list); + fq->flags |= INET_FRAG_COMPLETE; spin_unlock(&hb->chain_lock); } @@ -295,7 +288,6 @@ void inet_frag_kill(struct inet_frag_queue *fq, struct inet_frags *f) if (!(fq->flags & INET_FRAG_COMPLETE)) { fq_unlink(fq, f); atomic_dec(&fq->refcnt); - fq->flags |= INET_FRAG_COMPLETE; } } EXPORT_SYMBOL(inet_frag_kill); @@ -328,11 +320,12 @@ void inet_frag_destroy(struct inet_frag_queue *q, struct inet_frags *f) fp = xp; } sum = sum_truesize + f->qsize; - sub_frag_mem_limit(q->net, sum); if (f->destructor) f->destructor(q); kmem_cache_free(f->frags_cachep, q); + + sub_frag_mem_limit(nf, sum); } EXPORT_SYMBOL(inet_frag_destroy); -- cgit v0.10.2 From caaecdd3d3f8ec0ea9906c54b1dd8ec8316d26b9 Mon Sep 17 00:00:00 2001 From: Nikolay Aleksandrov Date: Thu, 23 Jul 2015 12:05:40 +0200 Subject: inet: frags: remove INET_FRAG_EVICTED and use list_evictor for the test We can simply remove the INET_FRAG_EVICTED flag to avoid all the flags race conditions with the evictor and use a participation test for the evictor list, when we're at that point (after inet_frag_kill) in the timer there're 2 possible cases: 1. The evictor added the entry to its evictor list while the timer was waiting for the chainlock or 2. The timer unchained the entry and the evictor won't see it In both cases we should be able to see list_evictor correctly due to the sync on the chainlock. Joint work with Florian Westphal. Tested-by: Frank Schreuder Signed-off-by: Nikolay Aleksandrov Signed-off-by: Florian Westphal Signed-off-by: David S. Miller diff --git a/include/net/inet_frag.h b/include/net/inet_frag.h index e71ca17..53eead2 100644 --- a/include/net/inet_frag.h +++ b/include/net/inet_frag.h @@ -21,13 +21,11 @@ struct netns_frags { * @INET_FRAG_FIRST_IN: first fragment has arrived * @INET_FRAG_LAST_IN: final fragment has arrived * @INET_FRAG_COMPLETE: frag queue has been processed and is due for destruction - * @INET_FRAG_EVICTED: frag queue is being evicted */ enum { INET_FRAG_FIRST_IN = BIT(0), INET_FRAG_LAST_IN = BIT(1), INET_FRAG_COMPLETE = BIT(2), - INET_FRAG_EVICTED = BIT(3) }; /** @@ -127,6 +125,11 @@ static inline void inet_frag_put(struct inet_frag_queue *q, struct inet_frags *f inet_frag_destroy(q, f); } +static inline bool inet_frag_evicting(struct inet_frag_queue *q) +{ + return !hlist_unhashed(&q->list_evictor); +} + /* Memory Tracking Functions. */ /* The default percpu_counter batch size is not big enough to scale to diff --git a/net/ipv4/inet_fragment.c b/net/ipv4/inet_fragment.c index a00ca4c..d0a7c03 100644 --- a/net/ipv4/inet_fragment.c +++ b/net/ipv4/inet_fragment.c @@ -140,7 +140,6 @@ inet_evict_bucket(struct inet_frags *f, struct inet_frag_bucket *hb) if (!del_timer(&fq->timer)) continue; - fq->flags |= INET_FRAG_EVICTED; hlist_add_head(&fq->list_evictor, &expired); ++evicted; } diff --git a/net/ipv4/ip_fragment.c b/net/ipv4/ip_fragment.c index b4a77d02..921138f 100644 --- a/net/ipv4/ip_fragment.c +++ b/net/ipv4/ip_fragment.c @@ -202,7 +202,7 @@ static void ip_expire(unsigned long arg) ipq_kill(qp); IP_INC_STATS_BH(net, IPSTATS_MIB_REASMFAILS); - if (!(qp->q.flags & INET_FRAG_EVICTED)) { + if (!inet_frag_evicting(&qp->q)) { struct sk_buff *head = qp->q.fragments; const struct iphdr *iph; int err; diff --git a/net/ipv6/reassembly.c b/net/ipv6/reassembly.c index 5c3bbca..f1159bb 100644 --- a/net/ipv6/reassembly.c +++ b/net/ipv6/reassembly.c @@ -144,7 +144,7 @@ void ip6_expire_frag_queue(struct net *net, struct frag_queue *fq, IP6_INC_STATS_BH(net, __in6_dev_get(dev), IPSTATS_MIB_REASMFAILS); - if (fq->q.flags & INET_FRAG_EVICTED) + if (inet_frag_evicting(&fq->q)) goto out_rcu_unlock; IP6_INC_STATS_BH(net, __in6_dev_get(dev), IPSTATS_MIB_REASMTIMEOUT); -- cgit v0.10.2 From 42288830494cd51873ca745a7a229023df061226 Mon Sep 17 00:00:00 2001 From: Jiri Pirko Date: Thu, 23 Jul 2015 12:20:37 +0200 Subject: niu: don't count tx error twice in case of headroom realloc fails Fixes: a3138df9 ("[NIU]: Add Sun Neptune ethernet driver.") Signed-off-by: Jiri Pirko Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/sun/niu.c b/drivers/net/ethernet/sun/niu.c index 0c5842a..ab6051a 100644 --- a/drivers/net/ethernet/sun/niu.c +++ b/drivers/net/ethernet/sun/niu.c @@ -6658,10 +6658,8 @@ static netdev_tx_t niu_start_xmit(struct sk_buff *skb, struct sk_buff *skb_new; skb_new = skb_realloc_headroom(skb, len); - if (!skb_new) { - rp->tx_errors++; + if (!skb_new) goto out_drop; - } kfree_skb(skb); skb = skb_new; } else -- cgit v0.10.2 From c0a1a0a698bb0f90ec4719de34f37715abfad8a5 Mon Sep 17 00:00:00 2001 From: Lucas Stach Date: Thu, 23 Jul 2015 16:06:19 +0200 Subject: net: fec: use managed DMA API functions to allocate BD ring So it gets freed when the device is going away. This fixes a DMA memory leak on driver probe() fail and driver remove(). Signed-off-by: Lucas Stach Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/freescale/fec_main.c b/drivers/net/ethernet/freescale/fec_main.c index 1f89c59..2945b62 100644 --- a/drivers/net/ethernet/freescale/fec_main.c +++ b/drivers/net/ethernet/freescale/fec_main.c @@ -3115,8 +3115,8 @@ static int fec_enet_init(struct net_device *ndev) fep->bufdesc_size; /* Allocate memory for buffer descriptors. */ - cbd_base = dma_alloc_coherent(NULL, bd_size, &bd_dma, - GFP_KERNEL); + cbd_base = dmam_alloc_coherent(&fep->pdev->dev, bd_size, &bd_dma, + GFP_KERNEL); if (!cbd_base) { return -ENOMEM; } -- cgit v0.10.2 From 32cba57ba74be58589aeb4cb6496183e46a5e3e5 Mon Sep 17 00:00:00 2001 From: Lucas Stach Date: Thu, 23 Jul 2015 16:06:20 +0200 Subject: net: fec: introduce fec_ptp_stop and use in probe fail path This function frees resources and cancels delayed work item that have been initialized in fec_ptp_init(). Use this to do proper error handling if something goes wrong in probe function after fec_ptp_init has been called. Signed-off-by: Lucas Stach Acked-by: Fugang Duan Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/freescale/fec.h b/drivers/net/ethernet/freescale/fec.h index 1eee73c..99d33e2 100644 --- a/drivers/net/ethernet/freescale/fec.h +++ b/drivers/net/ethernet/freescale/fec.h @@ -562,6 +562,7 @@ struct fec_enet_private { }; void fec_ptp_init(struct platform_device *pdev); +void fec_ptp_stop(struct platform_device *pdev); void fec_ptp_start_cyclecounter(struct net_device *ndev); int fec_ptp_set(struct net_device *ndev, struct ifreq *ifr); int fec_ptp_get(struct net_device *ndev, struct ifreq *ifr); diff --git a/drivers/net/ethernet/freescale/fec_main.c b/drivers/net/ethernet/freescale/fec_main.c index 2945b62..5e8b837 100644 --- a/drivers/net/ethernet/freescale/fec_main.c +++ b/drivers/net/ethernet/freescale/fec_main.c @@ -3454,6 +3454,7 @@ failed_register: failed_mii_init: failed_irq: failed_init: + fec_ptp_stop(pdev); if (fep->reg_phy) regulator_disable(fep->reg_phy); failed_regulator: @@ -3473,14 +3474,12 @@ fec_drv_remove(struct platform_device *pdev) struct net_device *ndev = platform_get_drvdata(pdev); struct fec_enet_private *fep = netdev_priv(ndev); - cancel_delayed_work_sync(&fep->time_keep); cancel_work_sync(&fep->tx_timeout_work); + fec_ptp_stop(pdev); unregister_netdev(ndev); fec_enet_mii_remove(fep); if (fep->reg_phy) regulator_disable(fep->reg_phy); - if (fep->ptp_clock) - ptp_clock_unregister(fep->ptp_clock); of_node_put(fep->phy_node); free_netdev(ndev); diff --git a/drivers/net/ethernet/freescale/fec_ptp.c b/drivers/net/ethernet/freescale/fec_ptp.c index a15663a..f457a23 100644 --- a/drivers/net/ethernet/freescale/fec_ptp.c +++ b/drivers/net/ethernet/freescale/fec_ptp.c @@ -604,6 +604,16 @@ void fec_ptp_init(struct platform_device *pdev) schedule_delayed_work(&fep->time_keep, HZ); } +void fec_ptp_stop(struct platform_device *pdev) +{ + struct net_device *ndev = platform_get_drvdata(pdev); + struct fec_enet_private *fep = netdev_priv(ndev); + + cancel_delayed_work_sync(&fep->time_keep); + if (fep->ptp_clock) + ptp_clock_unregister(fep->ptp_clock); +} + /** * fec_ptp_check_pps_event * @fep: the fec_enet_private structure handle -- cgit v0.10.2 From 30342fe65e511007672437741158d493472f427f Mon Sep 17 00:00:00 2001 From: Michael Holzheu Date: Thu, 23 Jul 2015 19:26:15 +0200 Subject: s390/bpf: clear correct BPF accumulator register Currently we assumed the following BPF to eBPF register mapping: - BPF_REG_A -> BPF_REG_7 - BPF_REG_X -> BPF_REG_8 Unfortunately this mapping is wrong. The correct mapping is: - BPF_REG_A -> BPF_REG_0 - BPF_REG_X -> BPF_REG_7 So clear the correct registers and use the BPF_REG_A and BPF_REG_X macros instead of BPF_REG_0/7. Fixes: 054623105728 ("s390/bpf: Add s390x eBPF JIT compiler backend") Cc: stable@vger.kernel.org # 4.0+ Signed-off-by: Michael Holzheu Signed-off-by: Martin Schwidefsky diff --git a/arch/s390/net/bpf_jit_comp.c b/arch/s390/net/bpf_jit_comp.c index fee782a..8d2e516 100644 --- a/arch/s390/net/bpf_jit_comp.c +++ b/arch/s390/net/bpf_jit_comp.c @@ -448,13 +448,13 @@ static void bpf_jit_prologue(struct bpf_jit *jit) EMIT6_DISP_LH(0xe3000000, 0x0004, REG_SKB_DATA, REG_0, BPF_REG_1, offsetof(struct sk_buff, data)); } - /* BPF compatibility: clear A (%b7) and X (%b8) registers */ - if (REG_SEEN(BPF_REG_7)) - /* lghi %b7,0 */ - EMIT4_IMM(0xa7090000, BPF_REG_7, 0); - if (REG_SEEN(BPF_REG_8)) - /* lghi %b8,0 */ - EMIT4_IMM(0xa7090000, BPF_REG_8, 0); + /* BPF compatibility: clear A (%b0) and X (%b7) registers */ + if (REG_SEEN(BPF_REG_A)) + /* lghi %ba,0 */ + EMIT4_IMM(0xa7090000, BPF_REG_A, 0); + if (REG_SEEN(BPF_REG_X)) + /* lghi %bx,0 */ + EMIT4_IMM(0xa7090000, BPF_REG_X, 0); } /* -- cgit v0.10.2 From 5c02a4206538da12c040b51778d310df84c6bf6c Mon Sep 17 00:00:00 2001 From: Andy Shevchenko Date: Fri, 24 Jul 2015 13:49:48 +0300 Subject: avr32: handle NULL as a valid clock object Since NULL is used as valid clock object on optional clocks we have to handle this case in avr32 implementation as well. Fixes: e1824dfe0d8e (net: macb: Adjust tx_clk when link speed changes) Signed-off-by: Andy Shevchenko Acked-by: Hans-Christian Egtvedt diff --git a/arch/avr32/mach-at32ap/clock.c b/arch/avr32/mach-at32ap/clock.c index 23b1a97..52c179b 100644 --- a/arch/avr32/mach-at32ap/clock.c +++ b/arch/avr32/mach-at32ap/clock.c @@ -80,6 +80,9 @@ int clk_enable(struct clk *clk) { unsigned long flags; + if (!clk) + return 0; + spin_lock_irqsave(&clk_lock, flags); __clk_enable(clk); spin_unlock_irqrestore(&clk_lock, flags); @@ -106,6 +109,9 @@ void clk_disable(struct clk *clk) { unsigned long flags; + if (IS_ERR_OR_NULL(clk)) + return; + spin_lock_irqsave(&clk_lock, flags); __clk_disable(clk); spin_unlock_irqrestore(&clk_lock, flags); @@ -117,6 +123,9 @@ unsigned long clk_get_rate(struct clk *clk) unsigned long flags; unsigned long rate; + if (!clk) + return 0; + spin_lock_irqsave(&clk_lock, flags); rate = clk->get_rate(clk); spin_unlock_irqrestore(&clk_lock, flags); @@ -129,6 +138,9 @@ long clk_round_rate(struct clk *clk, unsigned long rate) { unsigned long flags, actual_rate; + if (!clk) + return 0; + if (!clk->set_rate) return -ENOSYS; @@ -145,6 +157,9 @@ int clk_set_rate(struct clk *clk, unsigned long rate) unsigned long flags; long ret; + if (!clk) + return 0; + if (!clk->set_rate) return -ENOSYS; @@ -161,6 +176,9 @@ int clk_set_parent(struct clk *clk, struct clk *parent) unsigned long flags; int ret; + if (!clk) + return 0; + if (!clk->set_parent) return -ENOSYS; @@ -174,7 +192,7 @@ EXPORT_SYMBOL(clk_set_parent); struct clk *clk_get_parent(struct clk *clk) { - return clk->parent; + return !clk ? NULL : clk->parent; } EXPORT_SYMBOL(clk_get_parent); -- cgit v0.10.2 From b214396fb5890af58ff04b42deba7de5a2686338 Mon Sep 17 00:00:00 2001 From: hayeswang Date: Fri, 24 Jul 2015 13:54:23 +0800 Subject: r8152: fix the issue about U1/U2 - Disable U1/U2 during initialization. - Disable lpm when linking is on, and enable it when linking is off. - Disable U1/U2 when enabling runtime suspend. It is possible to let hw stop working, if the U1/U2 request occurs during some situations. The patch is used to avoid it. Signed-off-by: Hayes Wang Signed-off-by: David S. Miller diff --git a/drivers/net/usb/r8152.c b/drivers/net/usb/r8152.c index 7f6419e..e3a0110 100644 --- a/drivers/net/usb/r8152.c +++ b/drivers/net/usb/r8152.c @@ -2166,6 +2166,7 @@ static int rtl8153_enable(struct r8152 *tp) if (test_bit(RTL8152_UNPLUG, &tp->flags)) return -ENODEV; + usb_disable_lpm(tp->udev); set_tx_qlen(tp); rtl_set_eee_plus(tp); r8153_set_rx_early_timeout(tp); @@ -2337,11 +2338,54 @@ static void __rtl_set_wol(struct r8152 *tp, u32 wolopts) device_set_wakeup_enable(&tp->udev->dev, false); } +static void r8153_u1u2en(struct r8152 *tp, bool enable) +{ + u8 u1u2[8]; + + if (enable) + memset(u1u2, 0xff, sizeof(u1u2)); + else + memset(u1u2, 0x00, sizeof(u1u2)); + + usb_ocp_write(tp, USB_TOLERANCE, BYTE_EN_SIX_BYTES, sizeof(u1u2), u1u2); +} + +static void r8153_u2p3en(struct r8152 *tp, bool enable) +{ + u32 ocp_data; + + ocp_data = ocp_read_word(tp, MCU_TYPE_USB, USB_U2P3_CTRL); + if (enable && tp->version != RTL_VER_03 && tp->version != RTL_VER_04) + ocp_data |= U2P3_ENABLE; + else + ocp_data &= ~U2P3_ENABLE; + ocp_write_word(tp, MCU_TYPE_USB, USB_U2P3_CTRL, ocp_data); +} + +static void r8153_power_cut_en(struct r8152 *tp, bool enable) +{ + u32 ocp_data; + + ocp_data = ocp_read_word(tp, MCU_TYPE_USB, USB_POWER_CUT); + if (enable) + ocp_data |= PWR_EN | PHASE2_EN; + else + ocp_data &= ~(PWR_EN | PHASE2_EN); + ocp_write_word(tp, MCU_TYPE_USB, USB_POWER_CUT, ocp_data); + + ocp_data = ocp_read_word(tp, MCU_TYPE_USB, USB_MISC_0); + ocp_data &= ~PCUT_STATUS; + ocp_write_word(tp, MCU_TYPE_USB, USB_MISC_0, ocp_data); +} + static void rtl_runtime_suspend_enable(struct r8152 *tp, bool enable) { if (enable) { u32 ocp_data; + r8153_u1u2en(tp, false); + r8153_u2p3en(tp, false); + __rtl_set_wol(tp, WAKE_ANY); ocp_write_byte(tp, MCU_TYPE_PLA, PLA_CRWECR, CRWECR_CONFIG); @@ -2353,6 +2397,8 @@ static void rtl_runtime_suspend_enable(struct r8152 *tp, bool enable) ocp_write_byte(tp, MCU_TYPE_PLA, PLA_CRWECR, CRWECR_NORAML); } else { __rtl_set_wol(tp, tp->saved_wolopts); + r8153_u2p3en(tp, true); + r8153_u1u2en(tp, true); } } @@ -2599,46 +2645,6 @@ static void r8153_hw_phy_cfg(struct r8152 *tp) set_bit(PHY_RESET, &tp->flags); } -static void r8153_u1u2en(struct r8152 *tp, bool enable) -{ - u8 u1u2[8]; - - if (enable) - memset(u1u2, 0xff, sizeof(u1u2)); - else - memset(u1u2, 0x00, sizeof(u1u2)); - - usb_ocp_write(tp, USB_TOLERANCE, BYTE_EN_SIX_BYTES, sizeof(u1u2), u1u2); -} - -static void r8153_u2p3en(struct r8152 *tp, bool enable) -{ - u32 ocp_data; - - ocp_data = ocp_read_word(tp, MCU_TYPE_USB, USB_U2P3_CTRL); - if (enable) - ocp_data |= U2P3_ENABLE; - else - ocp_data &= ~U2P3_ENABLE; - ocp_write_word(tp, MCU_TYPE_USB, USB_U2P3_CTRL, ocp_data); -} - -static void r8153_power_cut_en(struct r8152 *tp, bool enable) -{ - u32 ocp_data; - - ocp_data = ocp_read_word(tp, MCU_TYPE_USB, USB_POWER_CUT); - if (enable) - ocp_data |= PWR_EN | PHASE2_EN; - else - ocp_data &= ~(PWR_EN | PHASE2_EN); - ocp_write_word(tp, MCU_TYPE_USB, USB_POWER_CUT, ocp_data); - - ocp_data = ocp_read_word(tp, MCU_TYPE_USB, USB_MISC_0); - ocp_data &= ~PCUT_STATUS; - ocp_write_word(tp, MCU_TYPE_USB, USB_MISC_0, ocp_data); -} - static void r8153_first_init(struct r8152 *tp) { u32 ocp_data; @@ -2781,6 +2787,7 @@ static void rtl8153_disable(struct r8152 *tp) r8153_disable_aldps(tp); rtl_disable(tp); r8153_enable_aldps(tp); + usb_enable_lpm(tp->udev); } static int rtl8152_set_speed(struct r8152 *tp, u8 autoneg, u16 speed, u8 duplex) @@ -2901,9 +2908,13 @@ static void rtl8153_up(struct r8152 *tp) if (test_bit(RTL8152_UNPLUG, &tp->flags)) return; + r8153_u1u2en(tp, false); r8153_disable_aldps(tp); r8153_first_init(tp); r8153_enable_aldps(tp); + r8153_u2p3en(tp, true); + r8153_u1u2en(tp, true); + usb_enable_lpm(tp->udev); } static void rtl8153_down(struct r8152 *tp) @@ -2914,6 +2925,7 @@ static void rtl8153_down(struct r8152 *tp) } r8153_u1u2en(tp, false); + r8153_u2p3en(tp, false); r8153_power_cut_en(tp, false); r8153_disable_aldps(tp); r8153_enter_oob(tp); @@ -3252,6 +3264,7 @@ static void r8153_init(struct r8152 *tp) msleep(20); } + usb_disable_lpm(tp->udev); r8153_u2p3en(tp, false); if (tp->version == RTL_VER_04) { @@ -3319,6 +3332,7 @@ static void r8153_init(struct r8152 *tp) r8153_enable_aldps(tp); r8152b_enable_fc(tp); rtl_tally_reset(tp); + r8153_u2p3en(tp, true); } static int rtl8152_suspend(struct usb_interface *intf, pm_message_t message) -- cgit v0.10.2 From 7daed8dc2a4b48c5a7ea5b3243d01837ec1aed0a Mon Sep 17 00:00:00 2001 From: hayeswang Date: Fri, 24 Jul 2015 13:54:24 +0800 Subject: r8152: fix wakeup settings Avoid the driver to enable WOL if the device doesn't support it. Signed-off-by: Hayes Wang Signed-off-by: David S. Miller diff --git a/drivers/net/usb/r8152.c b/drivers/net/usb/r8152.c index e3a0110..d537c30 100644 --- a/drivers/net/usb/r8152.c +++ b/drivers/net/usb/r8152.c @@ -2378,6 +2378,13 @@ static void r8153_power_cut_en(struct r8152 *tp, bool enable) ocp_write_word(tp, MCU_TYPE_USB, USB_MISC_0, ocp_data); } +static bool rtl_can_wakeup(struct r8152 *tp) +{ + struct usb_device *udev = tp->udev; + + return (udev->actconfig->desc.bmAttributes & USB_CONFIG_ATT_WAKEUP); +} + static void rtl_runtime_suspend_enable(struct r8152 *tp, bool enable) { if (enable) { @@ -3417,12 +3424,15 @@ static void rtl8152_get_wol(struct net_device *dev, struct ethtool_wolinfo *wol) if (usb_autopm_get_interface(tp->intf) < 0) return; - mutex_lock(&tp->control); - - wol->supported = WAKE_ANY; - wol->wolopts = __rtl_get_wol(tp); - - mutex_unlock(&tp->control); + if (!rtl_can_wakeup(tp)) { + wol->supported = 0; + wol->wolopts = 0; + } else { + mutex_lock(&tp->control); + wol->supported = WAKE_ANY; + wol->wolopts = __rtl_get_wol(tp); + mutex_unlock(&tp->control); + } usb_autopm_put_interface(tp->intf); } @@ -3432,6 +3442,9 @@ static int rtl8152_set_wol(struct net_device *dev, struct ethtool_wolinfo *wol) struct r8152 *tp = netdev_priv(dev); int ret; + if (!rtl_can_wakeup(tp)) + return -EOPNOTSUPP; + ret = usb_autopm_get_interface(tp->intf); if (ret < 0) goto out_set_wol; @@ -4073,6 +4086,9 @@ static int rtl8152_probe(struct usb_interface *intf, goto out1; } + if (!rtl_can_wakeup(tp)) + __rtl_set_wol(tp, 0); + tp->saved_wolopts = __rtl_get_wol(tp); if (tp->saved_wolopts) device_set_wakeup_enable(&udev->dev, true); -- cgit v0.10.2 From 41cec84cf2858b59121a1cec5e9b09fc1bf1d882 Mon Sep 17 00:00:00 2001 From: hayeswang Date: Fri, 24 Jul 2015 13:54:25 +0800 Subject: r8152: don't enable napi before rx ready Adjust napi_disable() and napi_enable() to avoid r8152_poll() start working before rx ready. Otherwise, it may have race condition for rx_agg. Signed-off-by: Hayes Wang Signed-off-by: David S. Miller diff --git a/drivers/net/usb/r8152.c b/drivers/net/usb/r8152.c index d537c30..144dc64 100644 --- a/drivers/net/usb/r8152.c +++ b/drivers/net/usb/r8152.c @@ -2075,7 +2075,6 @@ static int rtl_start_rx(struct r8152 *tp) { int i, ret = 0; - napi_disable(&tp->napi); INIT_LIST_HEAD(&tp->rx_done); for (i = 0; i < RTL8152_MAX_RX; i++) { INIT_LIST_HEAD(&tp->rx_info[i].list); @@ -2083,7 +2082,6 @@ static int rtl_start_rx(struct r8152 *tp) if (ret) break; } - napi_enable(&tp->napi); if (ret && ++i < RTL8152_MAX_RX) { struct list_head rx_queue; @@ -2951,8 +2949,10 @@ static void set_carrier(struct r8152 *tp) if (!netif_carrier_ok(netdev)) { tp->rtl_ops.enable(tp); set_bit(RTL8152_SET_RX_MODE, &tp->flags); + napi_disable(&tp->napi); netif_carrier_on(netdev); rtl_start_rx(tp); + napi_enable(&tp->napi); } } else { if (netif_carrier_ok(netdev)) { @@ -3395,9 +3395,11 @@ static int rtl8152_resume(struct usb_interface *intf) if (test_bit(SELECTIVE_SUSPEND, &tp->flags)) { rtl_runtime_suspend_enable(tp, false); clear_bit(SELECTIVE_SUSPEND, &tp->flags); + napi_disable(&tp->napi); set_bit(WORK_ENABLE, &tp->flags); if (netif_carrier_ok(tp->netdev)) rtl_start_rx(tp); + napi_enable(&tp->napi); } else { tp->rtl_ops.up(tp); rtl8152_set_speed(tp, AUTONEG_ENABLE, -- cgit v0.10.2 From dfbafc995304ebb9a9b03f65083e6e9cea143b20 Mon Sep 17 00:00:00 2001 From: Sabrina Dubroca Date: Fri, 24 Jul 2015 18:19:25 +0200 Subject: tcp: fix recv with flags MSG_WAITALL | MSG_PEEK Currently, tcp_recvmsg enters a busy loop in sk_wait_data if called with flags = MSG_WAITALL | MSG_PEEK. sk_wait_data waits for sk_receive_queue not empty, but in this case, the receive queue is not empty, but does not contain any skb that we can use. Add a "last skb seen on receive queue" argument to sk_wait_data, so that it sleeps until the receive queue has new skbs. Link: https://bugzilla.kernel.org/show_bug.cgi?id=99461 Link: https://sourceware.org/bugzilla/show_bug.cgi?id=18493 Link: https://bugzilla.redhat.com/show_bug.cgi?id=1205258 Reported-by: Enrico Scholz Reported-by: Dan Searle Signed-off-by: Sabrina Dubroca Acked-by: Eric Dumazet Signed-off-by: David S. Miller diff --git a/include/net/sock.h b/include/net/sock.h index 05a8c1a..f21f070 100644 --- a/include/net/sock.h +++ b/include/net/sock.h @@ -902,7 +902,7 @@ void sk_stream_kill_queues(struct sock *sk); void sk_set_memalloc(struct sock *sk); void sk_clear_memalloc(struct sock *sk); -int sk_wait_data(struct sock *sk, long *timeo); +int sk_wait_data(struct sock *sk, long *timeo, const struct sk_buff *skb); struct request_sock_ops; struct timewait_sock_ops; diff --git a/net/core/sock.c b/net/core/sock.c index 08f16db..8a14f12 100644 --- a/net/core/sock.c +++ b/net/core/sock.c @@ -1967,20 +1967,21 @@ static void __release_sock(struct sock *sk) * sk_wait_data - wait for data to arrive at sk_receive_queue * @sk: sock to wait on * @timeo: for how long + * @skb: last skb seen on sk_receive_queue * * Now socket state including sk->sk_err is changed only under lock, * hence we may omit checks after joining wait queue. * We check receive queue before schedule() only as optimization; * it is very likely that release_sock() added new data. */ -int sk_wait_data(struct sock *sk, long *timeo) +int sk_wait_data(struct sock *sk, long *timeo, const struct sk_buff *skb) { int rc; DEFINE_WAIT(wait); prepare_to_wait(sk_sleep(sk), &wait, TASK_INTERRUPTIBLE); set_bit(SOCK_ASYNC_WAITDATA, &sk->sk_socket->flags); - rc = sk_wait_event(sk, timeo, !skb_queue_empty(&sk->sk_receive_queue)); + rc = sk_wait_event(sk, timeo, skb_peek_tail(&sk->sk_receive_queue) != skb); clear_bit(SOCK_ASYNC_WAITDATA, &sk->sk_socket->flags); finish_wait(sk_sleep(sk), &wait); return rc; diff --git a/net/dccp/proto.c b/net/dccp/proto.c index 52a9401..b5cf13a 100644 --- a/net/dccp/proto.c +++ b/net/dccp/proto.c @@ -886,7 +886,7 @@ verify_sock_status: break; } - sk_wait_data(sk, &timeo); + sk_wait_data(sk, &timeo, NULL); continue; found_ok_skb: if (len > skb->len) diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c index 7f40567..45534a5 100644 --- a/net/ipv4/tcp.c +++ b/net/ipv4/tcp.c @@ -780,7 +780,7 @@ ssize_t tcp_splice_read(struct socket *sock, loff_t *ppos, ret = -EAGAIN; break; } - sk_wait_data(sk, &timeo); + sk_wait_data(sk, &timeo, NULL); if (signal_pending(current)) { ret = sock_intr_errno(timeo); break; @@ -1575,7 +1575,7 @@ int tcp_recvmsg(struct sock *sk, struct msghdr *msg, size_t len, int nonblock, int target; /* Read at least this many bytes */ long timeo; struct task_struct *user_recv = NULL; - struct sk_buff *skb; + struct sk_buff *skb, *last; u32 urg_hole = 0; if (unlikely(flags & MSG_ERRQUEUE)) @@ -1635,7 +1635,9 @@ int tcp_recvmsg(struct sock *sk, struct msghdr *msg, size_t len, int nonblock, /* Next get a buffer. */ + last = skb_peek_tail(&sk->sk_receive_queue); skb_queue_walk(&sk->sk_receive_queue, skb) { + last = skb; /* Now that we have two receive queues this * shouldn't happen. */ @@ -1754,8 +1756,9 @@ int tcp_recvmsg(struct sock *sk, struct msghdr *msg, size_t len, int nonblock, /* Do not sleep, just process backlog. */ release_sock(sk); lock_sock(sk); - } else - sk_wait_data(sk, &timeo); + } else { + sk_wait_data(sk, &timeo, last); + } if (user_recv) { int chunk; diff --git a/net/llc/af_llc.c b/net/llc/af_llc.c index 8fd9feb..8dab4e5 100644 --- a/net/llc/af_llc.c +++ b/net/llc/af_llc.c @@ -613,7 +613,7 @@ static int llc_wait_data(struct sock *sk, long timeo) if (signal_pending(current)) break; rc = 0; - if (sk_wait_data(sk, &timeo)) + if (sk_wait_data(sk, &timeo, NULL)) break; } return rc; @@ -802,7 +802,7 @@ static int llc_ui_recvmsg(struct socket *sock, struct msghdr *msg, size_t len, release_sock(sk); lock_sock(sk); } else - sk_wait_data(sk, &timeo); + sk_wait_data(sk, &timeo, NULL); if ((flags & MSG_PEEK) && peek_seq != llc->copied_seq) { net_dbg_ratelimited("LLC(%s:%d): Application bug, race in MSG_PEEK\n", -- cgit v0.10.2 From f2ce8a9e48385f444389e75cfe293637c3eb5410 Mon Sep 17 00:00:00 2001 From: Andy Shevchenko Date: Fri, 24 Jul 2015 21:23:59 +0300 Subject: net/macb: improve big endian CPU support The commit a50dad355a53 (net: macb: Add big endian CPU support) converted I/O accessors to readl_relaxed() and writel_relaxed() and consequentially broke MACB driver on AVR32 platforms such as ATNGW100. This patch improves I/O access by checking endiannes first and use the corresponding methods. Fixes: a50dad355a53 (net: macb: Add big endian CPU support) Signed-off-by: Andy Shevchenko Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/cadence/macb.c b/drivers/net/ethernet/cadence/macb.c index caeb395..9d06e3d 100644 --- a/drivers/net/ethernet/cadence/macb.c +++ b/drivers/net/ethernet/cadence/macb.c @@ -104,6 +104,57 @@ static void *macb_rx_buffer(struct macb *bp, unsigned int index) return bp->rx_buffers + bp->rx_buffer_size * macb_rx_ring_wrap(index); } +/* I/O accessors */ +static u32 hw_readl_native(struct macb *bp, int offset) +{ + return __raw_readl(bp->regs + offset); +} + +static void hw_writel_native(struct macb *bp, int offset, u32 value) +{ + __raw_writel(value, bp->regs + offset); +} + +static u32 hw_readl(struct macb *bp, int offset) +{ + return readl_relaxed(bp->regs + offset); +} + +static void hw_writel(struct macb *bp, int offset, u32 value) +{ + writel_relaxed(value, bp->regs + offset); +} + +/* + * Find the CPU endianness by using the loopback bit of NCR register. When the + * CPU is in big endian we need to program swaped mode for management + * descriptor access. + */ +static bool hw_is_native_io(void __iomem *addr) +{ + u32 value = MACB_BIT(LLB); + + __raw_writel(value, addr + MACB_NCR); + value = __raw_readl(addr + MACB_NCR); + + /* Write 0 back to disable everything */ + __raw_writel(0, addr + MACB_NCR); + + return value == MACB_BIT(LLB); +} + +static bool hw_is_gem(void __iomem *addr, bool native_io) +{ + u32 id; + + if (native_io) + id = __raw_readl(addr + MACB_MID); + else + id = readl_relaxed(addr + MACB_MID); + + return MACB_BFEXT(IDNUM, id) >= 0x2; +} + static void macb_set_hwaddr(struct macb *bp) { u32 bottom; @@ -449,14 +500,14 @@ err_out: static void macb_update_stats(struct macb *bp) { - u32 __iomem *reg = bp->regs + MACB_PFR; u32 *p = &bp->hw_stats.macb.rx_pause_frames; u32 *end = &bp->hw_stats.macb.tx_pause_frames + 1; + int offset = MACB_PFR; WARN_ON((unsigned long)(end - p - 1) != (MACB_TPF - MACB_PFR) / 4); - for(; p < end; p++, reg++) - *p += readl_relaxed(reg); + for(; p < end; p++, offset += 4) + *p += bp->readl(bp, offset); } static int macb_halt_tx(struct macb *bp) @@ -1603,7 +1654,6 @@ static u32 macb_dbw(struct macb *bp) static void macb_configure_dma(struct macb *bp) { u32 dmacfg; - u32 tmp, ncr; if (macb_is_gem(bp)) { dmacfg = gem_readl(bp, DMACFG) & ~GEM_BF(RXBS, -1L); @@ -1613,22 +1663,11 @@ static void macb_configure_dma(struct macb *bp) dmacfg |= GEM_BIT(TXPBMS) | GEM_BF(RXBMS, -1L); dmacfg &= ~GEM_BIT(ENDIA_PKT); - /* Find the CPU endianness by using the loopback bit of net_ctrl - * register. save it first. When the CPU is in big endian we - * need to program swaped mode for management descriptor access. - */ - ncr = macb_readl(bp, NCR); - __raw_writel(MACB_BIT(LLB), bp->regs + MACB_NCR); - tmp = __raw_readl(bp->regs + MACB_NCR); - - if (tmp == MACB_BIT(LLB)) + if (bp->native_io) dmacfg &= ~GEM_BIT(ENDIA_DESC); else dmacfg |= GEM_BIT(ENDIA_DESC); /* CPU in big endian */ - /* Restore net_ctrl */ - macb_writel(bp, NCR, ncr); - if (bp->dev->features & NETIF_F_HW_CSUM) dmacfg |= GEM_BIT(TXCOEN); else @@ -1902,14 +1941,14 @@ static void gem_update_stats(struct macb *bp) for (i = 0; i < GEM_STATS_LEN; ++i, ++p) { u32 offset = gem_statistics[i].offset; - u64 val = readl_relaxed(bp->regs + offset); + u64 val = bp->readl(bp, offset); bp->ethtool_stats[i] += val; *p += val; if (offset == GEM_OCTTXL || offset == GEM_OCTRXL) { /* Add GEM_OCTTXH, GEM_OCTRXH */ - val = readl_relaxed(bp->regs + offset + 4); + val = bp->readl(bp, offset + 4); bp->ethtool_stats[i] += ((u64)val) << 32; *(++p) += val; } @@ -2190,7 +2229,7 @@ static void macb_configure_caps(struct macb *bp, const struct macb_config *dt_co if (dt_conf) bp->caps = dt_conf->caps; - if (macb_is_gem_hw(bp->regs)) { + if (hw_is_gem(bp->regs, bp->native_io)) { bp->caps |= MACB_CAPS_MACB_IS_GEM; dcfg = gem_readl(bp, DCFG1); @@ -2205,6 +2244,7 @@ static void macb_configure_caps(struct macb *bp, const struct macb_config *dt_co } static void macb_probe_queues(void __iomem *mem, + bool native_io, unsigned int *queue_mask, unsigned int *num_queues) { @@ -2219,7 +2259,7 @@ static void macb_probe_queues(void __iomem *mem, * we are early in the probe process and don't have the * MACB_CAPS_MACB_IS_GEM flag positioned */ - if (!macb_is_gem_hw(mem)) + if (!hw_is_gem(mem, native_io)) return; /* bit 0 is never set but queue 0 always exists */ @@ -2786,6 +2826,7 @@ static int macb_probe(struct platform_device *pdev) struct clk *pclk, *hclk, *tx_clk; unsigned int queue_mask, num_queues; struct macb_platform_data *pdata; + bool native_io; struct phy_device *phydev; struct net_device *dev; struct resource *regs; @@ -2794,6 +2835,11 @@ static int macb_probe(struct platform_device *pdev) struct macb *bp; int err; + regs = platform_get_resource(pdev, IORESOURCE_MEM, 0); + mem = devm_ioremap_resource(&pdev->dev, regs); + if (IS_ERR(mem)) + return PTR_ERR(mem); + if (np) { const struct of_device_id *match; @@ -2809,14 +2855,9 @@ static int macb_probe(struct platform_device *pdev) if (err) return err; - regs = platform_get_resource(pdev, IORESOURCE_MEM, 0); - mem = devm_ioremap_resource(&pdev->dev, regs); - if (IS_ERR(mem)) { - err = PTR_ERR(mem); - goto err_disable_clocks; - } + native_io = hw_is_native_io(mem); - macb_probe_queues(mem, &queue_mask, &num_queues); + macb_probe_queues(mem, native_io, &queue_mask, &num_queues); dev = alloc_etherdev_mq(sizeof(*bp), num_queues); if (!dev) { err = -ENOMEM; @@ -2831,6 +2872,14 @@ static int macb_probe(struct platform_device *pdev) bp->pdev = pdev; bp->dev = dev; bp->regs = mem; + bp->native_io = native_io; + if (native_io) { + bp->readl = hw_readl_native; + bp->writel = hw_writel_native; + } else { + bp->readl = hw_readl; + bp->writel = hw_writel; + } bp->num_queues = num_queues; bp->queue_mask = queue_mask; if (macb_config) diff --git a/drivers/net/ethernet/cadence/macb.h b/drivers/net/ethernet/cadence/macb.h index d746559..f245340 100644 --- a/drivers/net/ethernet/cadence/macb.h +++ b/drivers/net/ethernet/cadence/macb.h @@ -429,18 +429,12 @@ | GEM_BF(name, value)) /* Register access macros */ -#define macb_readl(port,reg) \ - readl_relaxed((port)->regs + MACB_##reg) -#define macb_writel(port,reg,value) \ - writel_relaxed((value), (port)->regs + MACB_##reg) -#define gem_readl(port, reg) \ - readl_relaxed((port)->regs + GEM_##reg) -#define gem_writel(port, reg, value) \ - writel_relaxed((value), (port)->regs + GEM_##reg) -#define queue_readl(queue, reg) \ - readl_relaxed((queue)->bp->regs + (queue)->reg) -#define queue_writel(queue, reg, value) \ - writel_relaxed((value), (queue)->bp->regs + (queue)->reg) +#define macb_readl(port, reg) (port)->readl((port), MACB_##reg) +#define macb_writel(port, reg, value) (port)->writel((port), MACB_##reg, (value)) +#define gem_readl(port, reg) (port)->readl((port), GEM_##reg) +#define gem_writel(port, reg, value) (port)->writel((port), GEM_##reg, (value)) +#define queue_readl(queue, reg) (queue)->bp->readl((queue)->bp, (queue)->reg) +#define queue_writel(queue, reg, value) (queue)->bp->writel((queue)->bp, (queue)->reg, (value)) /* Conditional GEM/MACB macros. These perform the operation to the correct * register dependent on whether the device is a GEM or a MACB. For registers @@ -785,6 +779,11 @@ struct macb_queue { struct macb { void __iomem *regs; + bool native_io; + + /* hardware IO accessors */ + u32 (*readl)(struct macb *bp, int offset); + void (*writel)(struct macb *bp, int offset, u32 value); unsigned int rx_tail; unsigned int rx_prepared_head; @@ -843,9 +842,4 @@ static inline bool macb_is_gem(struct macb *bp) return !!(bp->caps & MACB_CAPS_MACB_IS_GEM); } -static inline bool macb_is_gem_hw(void __iomem *addr) -{ - return !!(MACB_BFEXT(IDNUM, readl_relaxed(addr + MACB_MID)) >= 0x2); -} - #endif /* _MACB_H */ -- cgit v0.10.2 From f36dbe6a285e0628e67b7bd9d3b08599184d7d2c Mon Sep 17 00:00:00 2001 From: Andy Shevchenko Date: Fri, 24 Jul 2015 21:24:00 +0300 Subject: net/macb: check if macb_config present The commit 98b5a0f4a228 introduces jumbo frame support, but also it assumes that macb_config present which is not always true. The configuration without macb_config fails to boot. Unable to handle kernel NULL pointer dereference at virtual address 00000010 ptbr = 90350000 pgd = 00000000 Oops: Kernel access of bad area, sig: 11 [#1] FRAME_POINTER chip: 0x01f:0x1e82 rev 2 Modules linked in: CPU: 0 PID: 1 Comm: swapper Not tainted 4.2.0-rc3-next-20150723+ #13 task: 91c26000 ti: 91c28000 task.ti: 91c28000 PC is at macb_probe+0x140/0x61c Fixes: 98b5a0f4a228 (net: macb: Add support for jumbo frames) Signed-off-by: Andy Shevchenko Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/cadence/macb.c b/drivers/net/ethernet/cadence/macb.c index 9d06e3d..6433232 100644 --- a/drivers/net/ethernet/cadence/macb.c +++ b/drivers/net/ethernet/cadence/macb.c @@ -2887,9 +2887,8 @@ static int macb_probe(struct platform_device *pdev) bp->pclk = pclk; bp->hclk = hclk; bp->tx_clk = tx_clk; - if (macb_config->jumbo_max_len) { + if (macb_config) bp->jumbo_max_len = macb_config->jumbo_max_len; - } spin_lock_init(&bp->lock); -- cgit v0.10.2 From a35919e174350d3e236ea5f179177396e87b6894 Mon Sep 17 00:00:00 2001 From: Andy Shevchenko Date: Fri, 24 Jul 2015 21:24:01 +0300 Subject: net/macb: use dev_*() when netdev is not yet registered To avoid messages like macb macb.0 (unnamed net_device) (uninitialized): Cadence caps 0x00000000 macb macb.0 (unnamed net_device) (uninitialized): invalid hw address, using random let's use dev_*() macros. Signed-off-by: Andy Shevchenko Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/cadence/macb.c b/drivers/net/ethernet/cadence/macb.c index 6433232..c4ffab9 100644 --- a/drivers/net/ethernet/cadence/macb.c +++ b/drivers/net/ethernet/cadence/macb.c @@ -211,7 +211,7 @@ static void macb_get_hwaddr(struct macb *bp) } } - netdev_info(bp->dev, "invalid hw address, using random\n"); + dev_info(&bp->pdev->dev, "invalid hw address, using random\n"); eth_hw_addr_random(bp->dev); } @@ -2240,7 +2240,7 @@ static void macb_configure_caps(struct macb *bp, const struct macb_config *dt_co bp->caps |= MACB_CAPS_FIFO_MODE; } - netdev_dbg(bp->dev, "Cadence caps 0x%08x\n", bp->caps); + dev_dbg(&bp->pdev->dev, "Cadence caps 0x%08x\n", bp->caps); } static void macb_probe_queues(void __iomem *mem, -- cgit v0.10.2 From 8bcbf82f31a94f8e4f939ac57478d808263c3890 Mon Sep 17 00:00:00 2001 From: Andy Shevchenko Date: Fri, 24 Jul 2015 21:24:02 +0300 Subject: net/macb: suppress compiler warnings MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit This patch fixes the following warnings: drivers/net/ethernet/cadence/macb.c: In function ‘macb_handle_link_change’: drivers/net/ethernet/cadence/macb.c:266: warning: comparison between signed and unsigned drivers/net/ethernet/cadence/macb.c:267: warning: comparison between signed and unsigned drivers/net/ethernet/cadence/macb.c:291: warning: comparison between signed and unsigned drivers/net/ethernet/cadence/macb.c: In function ‘gem_update_stats’: drivers/net/ethernet/cadence/macb.c:1908: warning: comparison between signed and unsigned drivers/net/ethernet/cadence/macb.c: In function ‘gem_get_ethtool_strings’: drivers/net/ethernet/cadence/macb.c:1988: warning: comparison between signed and unsigned Signed-off-by: Andy Shevchenko Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/cadence/macb.c b/drivers/net/ethernet/cadence/macb.c index c4ffab9..77a5270 100644 --- a/drivers/net/ethernet/cadence/macb.c +++ b/drivers/net/ethernet/cadence/macb.c @@ -303,7 +303,6 @@ static void macb_handle_link_change(struct net_device *dev) struct macb *bp = netdev_priv(dev); struct phy_device *phydev = bp->phy_dev; unsigned long flags; - int status_change = 0; spin_lock_irqsave(&bp->lock, flags); @@ -1936,7 +1935,7 @@ static int macb_change_mtu(struct net_device *dev, int new_mtu) static void gem_update_stats(struct macb *bp) { - int i; + unsigned int i; u32 *p = &bp->hw_stats.gem.tx_octets_31_0; for (i = 0; i < GEM_STATS_LEN; ++i, ++p) { @@ -2015,7 +2014,7 @@ static int gem_get_sset_count(struct net_device *dev, int sset) static void gem_get_ethtool_strings(struct net_device *dev, u32 sset, u8 *p) { - int i; + unsigned int i; switch (sset) { case ETH_SS_STATS: diff --git a/drivers/net/ethernet/cadence/macb.h b/drivers/net/ethernet/cadence/macb.h index f245340..2aa102e 100644 --- a/drivers/net/ethernet/cadence/macb.h +++ b/drivers/net/ethernet/cadence/macb.h @@ -816,9 +816,9 @@ struct macb { struct mii_bus *mii_bus; struct phy_device *phy_dev; - unsigned int link; - unsigned int speed; - unsigned int duplex; + int link; + int speed; + int duplex; u32 caps; unsigned int dma_burst_length; -- cgit v0.10.2 From 94b295edc2c678212376a8aca9308cf6ee430b89 Mon Sep 17 00:00:00 2001 From: Andy Shevchenko Date: Fri, 24 Jul 2015 21:24:03 +0300 Subject: net/macb: replace macb_count_tx_descriptors() by DIV_ROUND_UP() macb_count_tx_descriptors() repeats the generic macro DIV_ROUND_UP(). The patch does a replacement. There is no functional change. Signed-off-by: Andy Shevchenko Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/cadence/macb.c b/drivers/net/ethernet/cadence/macb.c index 77a5270..c638757 100644 --- a/drivers/net/ethernet/cadence/macb.c +++ b/drivers/net/ethernet/cadence/macb.c @@ -1157,12 +1157,6 @@ static void macb_poll_controller(struct net_device *dev) } #endif -static inline unsigned int macb_count_tx_descriptors(struct macb *bp, - unsigned int len) -{ - return (len + bp->max_tx_length - 1) / bp->max_tx_length; -} - static unsigned int macb_tx_map(struct macb *bp, struct macb_queue *queue, struct sk_buff *skb) @@ -1313,11 +1307,11 @@ static int macb_start_xmit(struct sk_buff *skb, struct net_device *dev) * socket buffer: skb fragments of jumbo frames may need to be * splitted into many buffer descriptors. */ - count = macb_count_tx_descriptors(bp, skb_headlen(skb)); + count = DIV_ROUND_UP(skb_headlen(skb), bp->max_tx_length); nr_frags = skb_shinfo(skb)->nr_frags; for (f = 0; f < nr_frags; f++) { frag_size = skb_frag_size(&skb_shinfo(skb)->frags[f]); - count += macb_count_tx_descriptors(bp, frag_size); + count += DIV_ROUND_UP(frag_size, bp->max_tx_length); } spin_lock_irqsave(&bp->lock, flags); -- cgit v0.10.2 From e018a0cce3d849bc73e72686c571420adc40bad2 Mon Sep 17 00:00:00 2001 From: Andy Shevchenko Date: Fri, 24 Jul 2015 21:24:04 +0300 Subject: net/macb: convert to kernel doc This patch coverts struct description to the kernel doc format. There is no functional change. Signed-off-by: Andy Shevchenko Signed-off-by: David S. Miller diff --git a/include/linux/platform_data/macb.h b/include/linux/platform_data/macb.h index 044a124..21b15f6 100644 --- a/include/linux/platform_data/macb.h +++ b/include/linux/platform_data/macb.h @@ -8,11 +8,19 @@ #ifndef __MACB_PDATA_H__ #define __MACB_PDATA_H__ +/** + * struct macb_platform_data - platform data for MACB Ethernet + * @phy_mask: phy mask passed when register the MDIO bus + * within the driver + * @phy_irq_pin: PHY IRQ + * @is_rmii: using RMII interface? + * @rev_eth_addr: reverse Ethernet address byte order + */ struct macb_platform_data { u32 phy_mask; - int phy_irq_pin; /* PHY IRQ */ - u8 is_rmii; /* using RMII interface? */ - u8 rev_eth_addr; /* reverse Ethernet address byte order */ + int phy_irq_pin; + u8 is_rmii; + u8 rev_eth_addr; }; #endif /* __MACB_PDATA_H__ */ -- cgit v0.10.2 From 7025e88a79d64aa4ba58fd03d630a78b12cce6ae Mon Sep 17 00:00:00 2001 From: WingMan Kwok Date: Fri, 24 Jul 2015 15:02:29 -0400 Subject: net: netcp: Fixes SGMII reset on network interface shutdown This patch asserts SGMII RTRESET, i.e. resetting the SGMII Tx/Rx logic, during network interface shutdown to avoid having the hardware wedge when shutting down with high incoming traffic rates. This is cleared (brought out of RTRESET) when the interface is brought back up. Signed-off-by: WingMan Kwok Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/ti/netcp.h b/drivers/net/ethernet/ti/netcp.h index bbacf5c..a8a7306 100644 --- a/drivers/net/ethernet/ti/netcp.h +++ b/drivers/net/ethernet/ti/netcp.h @@ -223,6 +223,7 @@ void *netcp_device_find_module(struct netcp_device *netcp_device, /* SGMII functions */ int netcp_sgmii_reset(void __iomem *sgmii_ofs, int port); +bool netcp_sgmii_rtreset(void __iomem *sgmii_ofs, int port, bool set); int netcp_sgmii_get_port_link(void __iomem *sgmii_ofs, int port); int netcp_sgmii_config(void __iomem *sgmii_ofs, int port, u32 interface); diff --git a/drivers/net/ethernet/ti/netcp_ethss.c b/drivers/net/ethernet/ti/netcp_ethss.c index 9b7e0a3..a218812 100644 --- a/drivers/net/ethernet/ti/netcp_ethss.c +++ b/drivers/net/ethernet/ti/netcp_ethss.c @@ -1901,11 +1901,28 @@ static void gbe_port_config(struct gbe_priv *gbe_dev, struct gbe_slave *slave, writel(slave->mac_control, GBE_REG_ADDR(slave, emac_regs, mac_control)); } +static void gbe_sgmii_rtreset(struct gbe_priv *priv, + struct gbe_slave *slave, bool set) +{ + void __iomem *sgmii_port_regs; + + if (SLAVE_LINK_IS_XGMII(slave)) + return; + + if ((priv->ss_version == GBE_SS_VERSION_14) && (slave->slave_num >= 2)) + sgmii_port_regs = priv->sgmii_port34_regs; + else + sgmii_port_regs = priv->sgmii_port_regs; + + netcp_sgmii_rtreset(sgmii_port_regs, slave->slave_num, set); +} + static void gbe_slave_stop(struct gbe_intf *intf) { struct gbe_priv *gbe_dev = intf->gbe_dev; struct gbe_slave *slave = intf->slave; + gbe_sgmii_rtreset(gbe_dev, slave, true); gbe_port_reset(slave); /* Disable forwarding */ cpsw_ale_control_set(gbe_dev->ale, slave->port_num, @@ -1947,6 +1964,7 @@ static int gbe_slave_open(struct gbe_intf *gbe_intf) gbe_sgmii_config(priv, slave); gbe_port_reset(slave); + gbe_sgmii_rtreset(priv, slave, false); gbe_port_config(priv, slave, priv->rx_packet_max); gbe_set_slave_mac(slave, gbe_intf); /* enable forwarding */ diff --git a/drivers/net/ethernet/ti/netcp_sgmii.c b/drivers/net/ethernet/ti/netcp_sgmii.c index dbeb142..5d8419f 100644 --- a/drivers/net/ethernet/ti/netcp_sgmii.c +++ b/drivers/net/ethernet/ti/netcp_sgmii.c @@ -18,6 +18,9 @@ #include "netcp.h" +#define SGMII_SRESET_RESET BIT(0) +#define SGMII_SRESET_RTRESET BIT(1) + #define SGMII_REG_STATUS_LOCK BIT(4) #define SGMII_REG_STATUS_LINK BIT(0) #define SGMII_REG_STATUS_AUTONEG BIT(2) @@ -51,12 +54,35 @@ static void sgmii_write_reg_bit(void __iomem *base, int reg, u32 val) int netcp_sgmii_reset(void __iomem *sgmii_ofs, int port) { /* Soft reset */ - sgmii_write_reg_bit(sgmii_ofs, SGMII_SRESET_REG(port), 0x1); - while (sgmii_read_reg(sgmii_ofs, SGMII_SRESET_REG(port)) != 0x0) + sgmii_write_reg_bit(sgmii_ofs, SGMII_SRESET_REG(port), + SGMII_SRESET_RESET); + + while ((sgmii_read_reg(sgmii_ofs, SGMII_SRESET_REG(port)) & + SGMII_SRESET_RESET) != 0x0) ; + return 0; } +/* port is 0 based */ +bool netcp_sgmii_rtreset(void __iomem *sgmii_ofs, int port, bool set) +{ + u32 reg; + bool oldval; + + /* Initiate a soft reset */ + reg = sgmii_read_reg(sgmii_ofs, SGMII_SRESET_REG(port)); + oldval = (reg & SGMII_SRESET_RTRESET) != 0x0; + if (set) + reg |= SGMII_SRESET_RTRESET; + else + reg &= ~SGMII_SRESET_RTRESET; + sgmii_write_reg(sgmii_ofs, SGMII_SRESET_REG(port), reg); + wmb(); + + return oldval; +} + int netcp_sgmii_get_port_link(void __iomem *sgmii_ofs, int port) { u32 status = 0, link = 0; -- cgit v0.10.2 From 8fff755e9f8d0f70a595e79f248695ce6aef5cc3 Mon Sep 17 00:00:00 2001 From: Andrew Lunn Date: Sat, 25 Jul 2015 22:38:02 +0200 Subject: net: fec: Ensure clocks are enabled while using mdio bus When a switch is attached to the mdio bus, the mdio bus can be used while the interface is not open. If the IPG clock is not enabled, MDIO reads/writes will simply time out. Add support for runtime PM to control this clock. Enable/disable this clock using runtime PM, with open()/close() and mdio read()/write() function triggering runtime PM operations. Since PM is optional, the IPG clock is enabled at probe and is no longer modified by fec_enet_clk_enable(), thus if PM is not enabled in the kernel, it is guaranteed the clock is running when MDIO operations are performed. Signed-off-by: Andrew Lunn Signed-off-by: Lucas Stach Cc: tyler.baker@linaro.org Cc: fabio.estevam@freescale.com Cc: shawn.guo@linaro.org Tested-by: Fabio Estevam Tested-by: Tyler Baker Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/freescale/fec_main.c b/drivers/net/ethernet/freescale/fec_main.c index 5e8b837..32e3807c 100644 --- a/drivers/net/ethernet/freescale/fec_main.c +++ b/drivers/net/ethernet/freescale/fec_main.c @@ -24,6 +24,7 @@ #include #include #include +#include #include #include #include @@ -77,6 +78,7 @@ static void fec_enet_itr_coal_init(struct net_device *ndev); #define FEC_ENET_RAEM_V 0x8 #define FEC_ENET_RAFL_V 0x8 #define FEC_ENET_OPD_V 0xFFF0 +#define FEC_MDIO_PM_TIMEOUT 100 /* ms */ static struct platform_device_id fec_devtype[] = { { @@ -1767,7 +1769,13 @@ static void fec_enet_adjust_link(struct net_device *ndev) static int fec_enet_mdio_read(struct mii_bus *bus, int mii_id, int regnum) { struct fec_enet_private *fep = bus->priv; + struct device *dev = &fep->pdev->dev; unsigned long time_left; + int ret = 0; + + ret = pm_runtime_get_sync(dev); + if (IS_ERR_VALUE(ret)) + return ret; fep->mii_timeout = 0; init_completion(&fep->mdio_done); @@ -1783,18 +1791,30 @@ static int fec_enet_mdio_read(struct mii_bus *bus, int mii_id, int regnum) if (time_left == 0) { fep->mii_timeout = 1; netdev_err(fep->netdev, "MDIO read timeout\n"); - return -ETIMEDOUT; + ret = -ETIMEDOUT; + goto out; } - /* return value */ - return FEC_MMFR_DATA(readl(fep->hwp + FEC_MII_DATA)); + ret = FEC_MMFR_DATA(readl(fep->hwp + FEC_MII_DATA)); + +out: + pm_runtime_mark_last_busy(dev); + pm_runtime_put_autosuspend(dev); + + return ret; } static int fec_enet_mdio_write(struct mii_bus *bus, int mii_id, int regnum, u16 value) { struct fec_enet_private *fep = bus->priv; + struct device *dev = &fep->pdev->dev; unsigned long time_left; + int ret = 0; + + ret = pm_runtime_get_sync(dev); + if (IS_ERR_VALUE(ret)) + return ret; fep->mii_timeout = 0; init_completion(&fep->mdio_done); @@ -1811,10 +1831,13 @@ static int fec_enet_mdio_write(struct mii_bus *bus, int mii_id, int regnum, if (time_left == 0) { fep->mii_timeout = 1; netdev_err(fep->netdev, "MDIO write timeout\n"); - return -ETIMEDOUT; + ret = -ETIMEDOUT; } - return 0; + pm_runtime_mark_last_busy(dev); + pm_runtime_put_autosuspend(dev); + + return ret; } static int fec_enet_clk_enable(struct net_device *ndev, bool enable) @@ -1826,9 +1849,6 @@ static int fec_enet_clk_enable(struct net_device *ndev, bool enable) ret = clk_prepare_enable(fep->clk_ahb); if (ret) return ret; - ret = clk_prepare_enable(fep->clk_ipg); - if (ret) - goto failed_clk_ipg; if (fep->clk_enet_out) { ret = clk_prepare_enable(fep->clk_enet_out); if (ret) @@ -1852,7 +1872,6 @@ static int fec_enet_clk_enable(struct net_device *ndev, bool enable) } } else { clk_disable_unprepare(fep->clk_ahb); - clk_disable_unprepare(fep->clk_ipg); if (fep->clk_enet_out) clk_disable_unprepare(fep->clk_enet_out); if (fep->clk_ptp) { @@ -1874,8 +1893,6 @@ failed_clk_ptp: if (fep->clk_enet_out) clk_disable_unprepare(fep->clk_enet_out); failed_clk_enet_out: - clk_disable_unprepare(fep->clk_ipg); -failed_clk_ipg: clk_disable_unprepare(fep->clk_ahb); return ret; @@ -2847,10 +2864,14 @@ fec_enet_open(struct net_device *ndev) struct fec_enet_private *fep = netdev_priv(ndev); int ret; + ret = pm_runtime_get_sync(&fep->pdev->dev); + if (IS_ERR_VALUE(ret)) + return ret; + pinctrl_pm_select_default_state(&fep->pdev->dev); ret = fec_enet_clk_enable(ndev, true); if (ret) - return ret; + goto clk_enable; /* I should reset the ring buffers here, but I don't yet know * a simple way to do that. @@ -2881,6 +2902,9 @@ err_enet_mii_probe: fec_enet_free_buffers(ndev); err_enet_alloc: fec_enet_clk_enable(ndev, false); +clk_enable: + pm_runtime_mark_last_busy(&fep->pdev->dev); + pm_runtime_put_autosuspend(&fep->pdev->dev); pinctrl_pm_select_sleep_state(&fep->pdev->dev); return ret; } @@ -2903,6 +2927,9 @@ fec_enet_close(struct net_device *ndev) fec_enet_clk_enable(ndev, false); pinctrl_pm_select_sleep_state(&fep->pdev->dev); + pm_runtime_mark_last_busy(&fep->pdev->dev); + pm_runtime_put_autosuspend(&fep->pdev->dev); + fec_enet_free_buffers(ndev); return 0; @@ -3388,6 +3415,10 @@ fec_probe(struct platform_device *pdev) if (ret) goto failed_clk; + ret = clk_prepare_enable(fep->clk_ipg); + if (ret) + goto failed_clk_ipg; + fep->reg_phy = devm_regulator_get(&pdev->dev, "phy"); if (!IS_ERR(fep->reg_phy)) { ret = regulator_enable(fep->reg_phy); @@ -3400,6 +3431,11 @@ fec_probe(struct platform_device *pdev) fep->reg_phy = NULL; } + pm_runtime_set_autosuspend_delay(&pdev->dev, FEC_MDIO_PM_TIMEOUT); + pm_runtime_use_autosuspend(&pdev->dev); + pm_runtime_set_active(&pdev->dev); + pm_runtime_enable(&pdev->dev); + fec_reset_phy(pdev); if (fep->bufdesc_ex) @@ -3447,6 +3483,10 @@ fec_probe(struct platform_device *pdev) fep->rx_copybreak = COPYBREAK_DEFAULT; INIT_WORK(&fep->tx_timeout_work, fec_enet_timeout_work); + + pm_runtime_mark_last_busy(&pdev->dev); + pm_runtime_put_autosuspend(&pdev->dev); + return 0; failed_register: @@ -3458,6 +3498,8 @@ failed_init: if (fep->reg_phy) regulator_disable(fep->reg_phy); failed_regulator: + clk_disable_unprepare(fep->clk_ipg); +failed_clk_ipg: fec_enet_clk_enable(ndev, false); failed_clk: failed_phy: @@ -3567,7 +3609,28 @@ failed_clk: return ret; } -static SIMPLE_DEV_PM_OPS(fec_pm_ops, fec_suspend, fec_resume); +static int __maybe_unused fec_runtime_suspend(struct device *dev) +{ + struct net_device *ndev = dev_get_drvdata(dev); + struct fec_enet_private *fep = netdev_priv(ndev); + + clk_disable_unprepare(fep->clk_ipg); + + return 0; +} + +static int __maybe_unused fec_runtime_resume(struct device *dev) +{ + struct net_device *ndev = dev_get_drvdata(dev); + struct fec_enet_private *fep = netdev_priv(ndev); + + return clk_prepare_enable(fep->clk_ipg); +} + +static const struct dev_pm_ops fec_pm_ops = { + SET_SYSTEM_SLEEP_PM_OPS(fec_suspend, fec_resume) + SET_RUNTIME_PM_OPS(fec_runtime_suspend, fec_runtime_resume, NULL) +}; static struct platform_driver fec_driver = { .driver = { -- cgit v0.10.2 From ac8c79304280da6ef05c348a9da03ab04898b994 Mon Sep 17 00:00:00 2001 From: Kamil Dudka Date: Wed, 15 Jul 2015 17:18:15 +0200 Subject: drm/nouveau: hold mutex when calling nouveau_abi16_fini() This was the only access to cli->abi16 without holding the mutex. Signed-off-by: Kamil Dudka diff --git a/drivers/gpu/drm/nouveau/nouveau_drm.c b/drivers/gpu/drm/nouveau/nouveau_drm.c index 649024d..01c78a4 100644 --- a/drivers/gpu/drm/nouveau/nouveau_drm.c +++ b/drivers/gpu/drm/nouveau/nouveau_drm.c @@ -865,8 +865,10 @@ nouveau_drm_preclose(struct drm_device *dev, struct drm_file *fpriv) pm_runtime_get_sync(dev->dev); + mutex_lock(&cli->mutex); if (cli->abi16) nouveau_abi16_fini(cli->abi16); + mutex_unlock(&cli->mutex); mutex_lock(&drm->client.mutex); list_del(&cli->head); -- cgit v0.10.2 From 7512223b1ece29a5968ed8b67ccb891d21b7834b Mon Sep 17 00:00:00 2001 From: Kamil Dudka Date: Wed, 15 Jul 2015 22:57:43 +0200 Subject: drm/nouveau/drm/nv04-nv40/instmem: protect access to priv->heap by mutex This fixes the list_del corruption reported at . Signed-off-by: Kamil Dudka diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/instmem/nv04.c b/drivers/gpu/drm/nouveau/nvkm/subdev/instmem/nv04.c index 80614f1..282143f 100644 --- a/drivers/gpu/drm/nouveau/nvkm/subdev/instmem/nv04.c +++ b/drivers/gpu/drm/nouveau/nvkm/subdev/instmem/nv04.c @@ -50,7 +50,12 @@ nv04_instobj_dtor(struct nvkm_object *object) { struct nv04_instmem_priv *priv = (void *)nvkm_instmem(object); struct nv04_instobj_priv *node = (void *)object; + struct nvkm_subdev *subdev = (void *)priv; + + mutex_lock(&subdev->mutex); nvkm_mm_free(&priv->heap, &node->mem); + mutex_unlock(&subdev->mutex); + nvkm_instobj_destroy(&node->base); } @@ -62,6 +67,7 @@ nv04_instobj_ctor(struct nvkm_object *parent, struct nvkm_object *engine, struct nv04_instmem_priv *priv = (void *)nvkm_instmem(parent); struct nv04_instobj_priv *node; struct nvkm_instobj_args *args = data; + struct nvkm_subdev *subdev = (void *)priv; int ret; if (!args->align) @@ -72,8 +78,10 @@ nv04_instobj_ctor(struct nvkm_object *parent, struct nvkm_object *engine, if (ret) return ret; + mutex_lock(&subdev->mutex); ret = nvkm_mm_head(&priv->heap, 0, 1, args->size, args->size, args->align, &node->mem); + mutex_unlock(&subdev->mutex); if (ret) return ret; -- cgit v0.10.2 From 9694554691d26226f84fa1d6f500b7e7c9288510 Mon Sep 17 00:00:00 2001 From: Roy Spliet Date: Sat, 23 May 2015 10:37:42 +0200 Subject: drm/nouveau/clk/gt215: u32->s32 for difference in req. and set clock This difference can of course be negative too... Signed-off-by: Roy Spliet Signed-off-by: Ben Skeggs diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/clk/gt215.c b/drivers/gpu/drm/nouveau/nvkm/subdev/clk/gt215.c index 822d32a2..065e9f5 100644 --- a/drivers/gpu/drm/nouveau/nvkm/subdev/clk/gt215.c +++ b/drivers/gpu/drm/nouveau/nvkm/subdev/clk/gt215.c @@ -180,7 +180,8 @@ gt215_clk_info(struct nvkm_clk *clock, int clk, u32 khz, struct gt215_clk_info *info) { struct gt215_clk_priv *priv = (void *)clock; - u32 oclk, sclk, sdiv, diff; + u32 oclk, sclk, sdiv; + s32 diff; info->clk = 0; -- cgit v0.10.2 From f5654d9555b470d05be9d6d26cfb8fb79239ac3e Mon Sep 17 00:00:00 2001 From: Thierry Reding Date: Thu, 16 Oct 2014 11:54:54 +0200 Subject: drm/nouveau: Do not leak client objects The memory allocated for a nouveau_cli object in nouveau_cli_create() is never freed. Free the memory in nouveau_cli_destroy() to plug this leak. kmemleak recorded this after running a couple of nouveau test programs. Note that kmemleak points at drm_open_helper() because for some reason it thinks that skipping the first two stack frames is a good idea. Signed-off-by: Thierry Reding Signed-off-by: Ben Skeggs diff --git a/drivers/gpu/drm/nouveau/nouveau_drm.c b/drivers/gpu/drm/nouveau/nouveau_drm.c index 01c78a4..477cbb1 100644 --- a/drivers/gpu/drm/nouveau/nouveau_drm.c +++ b/drivers/gpu/drm/nouveau/nouveau_drm.c @@ -128,6 +128,7 @@ nouveau_cli_destroy(struct nouveau_cli *cli) nvkm_vm_ref(NULL, &nvxx_client(&cli->base)->vm, NULL); nvif_client_fini(&cli->base); usif_client_fini(cli); + kfree(cli); } static void -- cgit v0.10.2 From a67e14b0b6e35355411a4f2b05f8d087a196dac9 Mon Sep 17 00:00:00 2001 From: Alexandre Courbot Date: Wed, 8 Apr 2015 18:08:13 +0900 Subject: drm/nouveau/platform: fix compile error if !CONFIG_IOMMU The lack of IOMMU API support can make nouveau_platform_probe_iommu() fail to compile because struct iommu_ops is then empty. Fix this by skipping IOMMU probe in that case - lack of IOMMU on platform devices is sub-optimal, but is not an error. Signed-off-by: Alexandre Courbot Signed-off-by: Ben Skeggs diff --git a/drivers/gpu/drm/nouveau/nouveau_platform.c b/drivers/gpu/drm/nouveau/nouveau_platform.c index 775277f..dcfbbfa 100644 --- a/drivers/gpu/drm/nouveau/nouveau_platform.c +++ b/drivers/gpu/drm/nouveau/nouveau_platform.c @@ -92,6 +92,8 @@ static int nouveau_platform_power_down(struct nouveau_platform_gpu *gpu) return 0; } +#if IS_ENABLED(CONFIG_IOMMU_API) + static void nouveau_platform_probe_iommu(struct device *dev, struct nouveau_platform_gpu *gpu) { @@ -158,6 +160,20 @@ static void nouveau_platform_remove_iommu(struct device *dev, } } +#else + +static void nouveau_platform_probe_iommu(struct device *dev, + struct nouveau_platform_gpu *gpu) +{ +} + +static void nouveau_platform_remove_iommu(struct device *dev, + struct nouveau_platform_gpu *gpu) +{ +} + +#endif + static int nouveau_platform_probe(struct platform_device *pdev) { struct nouveau_platform_gpu *gpu; -- cgit v0.10.2 From 9c56be4cf3d6db5a20c3b6483bd4cdc21c15cf4f Mon Sep 17 00:00:00 2001 From: Alexandre Courbot Date: Tue, 28 Apr 2015 16:17:07 +0900 Subject: drm/nouveau/ibus/gk20a: increase SM wait timeout Increase clock timeout for SYS, FPB and GPC in order to avoid operation failure at high gpcclk rate. Signed-off-by: Alexandre Courbot Signed-off-by: Ben Skeggs diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/ibus/gk20a.c b/drivers/gpu/drm/nouveau/nvkm/subdev/ibus/gk20a.c index c0fdb89..24dcdfb 100644 --- a/drivers/gpu/drm/nouveau/nvkm/subdev/ibus/gk20a.c +++ b/drivers/gpu/drm/nouveau/nvkm/subdev/ibus/gk20a.c @@ -38,6 +38,14 @@ gk20a_ibus_init_priv_ring(struct gk20a_ibus_priv *priv) nv_wr32(priv, 0x12004c, 0x4); nv_wr32(priv, 0x122204, 0x2); nv_rd32(priv, 0x122204); + + /* + * Bug: increase clock timeout to avoid operation failure at high + * gpcclk rate. + */ + nv_wr32(priv, 0x122354, 0x800); + nv_wr32(priv, 0x128328, 0x800); + nv_wr32(priv, 0x124320, 0x800); } static void -- cgit v0.10.2 From 1addc12648521d15ef33c1a88d0354850190dfa7 Mon Sep 17 00:00:00 2001 From: Alexandre Courbot Date: Thu, 16 Apr 2015 11:33:32 +0900 Subject: drm/nouveau/fifo/gk104: kick channels when deactivating them Kicking channels is part of their deactivation process. Maxwell chips are particularly sensitive to this, and can start fetching the previous pushbuffer of a recycled channel if this is not done. While we are at it, improve the channel preemption code to only wait for bit 20 of 0x002634 to turn to 0, as it is the bit indicating a preempt is pending. Signed-off-by: Alexandre Courbot Signed-off-by: Ben Skeggs diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gk104.c b/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gk104.c index e10f964..52c22b0 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gk104.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gk104.c @@ -166,14 +166,30 @@ gk104_fifo_context_attach(struct nvkm_object *parent, } static int +gk104_fifo_chan_kick(struct gk104_fifo_chan *chan) +{ + struct nvkm_object *obj = (void *)chan; + struct gk104_fifo_priv *priv = (void *)obj->engine; + + nv_wr32(priv, 0x002634, chan->base.chid); + if (!nv_wait(priv, 0x002634, 0x100000, 0x000000)) { + nv_error(priv, "channel %d [%s] kick timeout\n", + chan->base.chid, nvkm_client_name(chan)); + return -EBUSY; + } + + return 0; +} + +static int gk104_fifo_context_detach(struct nvkm_object *parent, bool suspend, struct nvkm_object *object) { struct nvkm_bar *bar = nvkm_bar(parent); - struct gk104_fifo_priv *priv = (void *)parent->engine; struct gk104_fifo_base *base = (void *)parent->parent; struct gk104_fifo_chan *chan = (void *)parent; u32 addr; + int ret; switch (nv_engidx(object->engine)) { case NVDEV_ENGINE_SW : return 0; @@ -188,13 +204,9 @@ gk104_fifo_context_detach(struct nvkm_object *parent, bool suspend, return -EINVAL; } - nv_wr32(priv, 0x002634, chan->base.chid); - if (!nv_wait(priv, 0x002634, 0xffffffff, chan->base.chid)) { - nv_error(priv, "channel %d [%s] kick timeout\n", - chan->base.chid, nvkm_client_name(chan)); - if (suspend) - return -EBUSY; - } + ret = gk104_fifo_chan_kick(chan); + if (ret && suspend) + return ret; if (addr) { nv_wo32(base, addr + 0x00, 0x00000000); @@ -319,6 +331,7 @@ gk104_fifo_chan_fini(struct nvkm_object *object, bool suspend) gk104_fifo_runlist_update(priv, chan->engine); } + gk104_fifo_chan_kick(chan); nv_wr32(priv, 0x800000 + (chid * 8), 0x00000000); return nvkm_fifo_channel_fini(&chan->base, suspend); } -- cgit v0.10.2 From 19bf09cecfec1891069f1d5353a0298debd98713 Mon Sep 17 00:00:00 2001 From: Alexandre Courbot Date: Wed, 29 Apr 2015 23:04:23 +0900 Subject: drm/nouveau/gr/gf100: wait on bottom half of FE's pipeline When emitting the ICMD bundle, wait on the bottom half (bit 3 of the GR_STATUS register) instead of upper half (bit 2) to make sure methods are effectively emitted. Signed-off-by: Alexandre Courbot Signed-off-by: Ben Skeggs diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/gr/gf100.c b/drivers/gpu/drm/nouveau/nvkm/engine/gr/gf100.c index 5606c25..01efc2c 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/gr/gf100.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/gr/gf100.c @@ -699,7 +699,7 @@ gf100_gr_icmd(struct gf100_gr_priv *priv, const struct gf100_gr_pack *p) while (addr < next) { nv_wr32(priv, 0x400200, addr); - nv_wait(priv, 0x400700, 0x00000002, 0x00000000); + nv_wait(priv, 0x400700, 0x00000004, 0x00000000); addr += init->pitch; } } -- cgit v0.10.2 From 4a8cf4513dd4279d342eb41abdb9db65285732b3 Mon Sep 17 00:00:00 2001 From: Alexandre Courbot Date: Mon, 27 Apr 2015 17:25:11 +0900 Subject: drm/nouveau/gr/gf100: wait for GR idle after GO_IDLE bundle After submitting a GO_IDLE bundle, one must wait for GR to effectively be idle before submitting the next bundle. Failure to do so may result in undefined behavior in some rare cases. Signed-off-by: Alexandre Courbot Reported-by: Kary Jin Signed-off-by: Ben Skeggs diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/gr/gf100.c b/drivers/gpu/drm/nouveau/nvkm/engine/gr/gf100.c index 01efc2c..ca11ddb 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/gr/gf100.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/gr/gf100.c @@ -663,6 +663,37 @@ gf100_gr_zbc_init(struct gf100_gr_priv *priv) gf100_gr_zbc_clear_depth(priv, index); } +/** + * Wait until GR goes idle. GR is considered idle if it is disabled by the + * MC (0x200) register, or GR is not busy and a context switch is not in + * progress. + */ +int +gf100_gr_wait_idle(struct gf100_gr_priv *priv) +{ + unsigned long end_jiffies = jiffies + msecs_to_jiffies(2000); + bool gr_enabled, ctxsw_active, gr_busy; + + do { + /* + * required to make sure FIFO_ENGINE_STATUS (0x2640) is + * up-to-date + */ + nv_rd32(priv, 0x400700); + + gr_enabled = nv_rd32(priv, 0x200) & 0x1000; + ctxsw_active = nv_rd32(priv, 0x2640) & 0x8000; + gr_busy = nv_rd32(priv, 0x40060c) & 0x1; + + if (!gr_enabled || (!gr_busy && !ctxsw_active)) + return 0; + } while (time_before(jiffies, end_jiffies)); + + nv_error(priv, "wait for idle timeout (en: %d, ctxsw: %d, busy: %d)\n", + gr_enabled, ctxsw_active, gr_busy); + return -EAGAIN; +} + void gf100_gr_mmio(struct gf100_gr_priv *priv, const struct gf100_gr_pack *p) { @@ -699,6 +730,12 @@ gf100_gr_icmd(struct gf100_gr_priv *priv, const struct gf100_gr_pack *p) while (addr < next) { nv_wr32(priv, 0x400200, addr); + /** + * Wait for GR to go idle after submitting a + * GO_IDLE bundle + */ + if ((addr & 0xffff) == 0xe100) + gf100_gr_wait_idle(priv); nv_wait(priv, 0x400700, 0x00000004, 0x00000000); addr += init->pitch; } diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/gr/gf100.h b/drivers/gpu/drm/nouveau/nvkm/engine/gr/gf100.h index 8af1a89..c9533fd 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/gr/gf100.h +++ b/drivers/gpu/drm/nouveau/nvkm/engine/gr/gf100.h @@ -181,6 +181,7 @@ struct gf100_gr_oclass { int ppc_nr; }; +int gf100_gr_wait_idle(struct gf100_gr_priv *); void gf100_gr_mmio(struct gf100_gr_priv *, const struct gf100_gr_pack *); void gf100_gr_icmd(struct gf100_gr_priv *, const struct gf100_gr_pack *); void gf100_gr_mthd(struct gf100_gr_priv *, const struct gf100_gr_pack *); -- cgit v0.10.2 From 3693d544056b822f7f2454da5acb7eda9471b902 Mon Sep 17 00:00:00 2001 From: Samuel Pitoiset Date: Sun, 7 Jun 2015 22:40:11 +0200 Subject: drm/nouveau/pm: prevent freeing the wrong engine context This fixes a crash when multiple PM engine contexts are created. Signed-off-by: Samuel Pitoiset Reviewed-by: Martin Peres Signed-off-by: Ben Skeggs diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/pm/base.c b/drivers/gpu/drm/nouveau/nvkm/engine/pm/base.c index 2006c44..274457c 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/pm/base.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/pm/base.c @@ -332,9 +332,12 @@ static void nvkm_perfctx_dtor(struct nvkm_object *object) { struct nvkm_pm *ppm = (void *)object->engine; + struct nvkm_perfctx *ctx = (void *)object; + mutex_lock(&nv_subdev(ppm)->mutex); - nvkm_engctx_destroy(&ppm->context->base); - ppm->context = NULL; + nvkm_engctx_destroy(&ctx->base); + if (ppm->context == ctx) + ppm->context = NULL; mutex_unlock(&nv_subdev(ppm)->mutex); } -- cgit v0.10.2 From 305c1959ea60ddcae5142a20c50db849a40c2a35 Mon Sep 17 00:00:00 2001 From: Samuel Pitoiset Date: Sun, 7 Jun 2015 22:40:12 +0200 Subject: drm/nouveau/pm: fix a potential race condition when creating an engine context There is always the possiblity that the ppm->context pointer would get partially updated and accidentally would equal ctx. This would allow two contexts to co-exist, which is not acceptable. Moving the test to the critical section takes care of this problem. Signed-off-by: Samuel Pitoiset Signed-off-by: Martin Peres Signed-off-by: Ben Skeggs diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/pm/base.c b/drivers/gpu/drm/nouveau/nvkm/engine/pm/base.c index 274457c..4cf36a3 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/pm/base.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/pm/base.c @@ -358,12 +358,11 @@ nvkm_perfctx_ctor(struct nvkm_object *parent, struct nvkm_object *engine, mutex_lock(&nv_subdev(ppm)->mutex); if (ppm->context == NULL) ppm->context = ctx; - mutex_unlock(&nv_subdev(ppm)->mutex); - if (ctx != ppm->context) - return -EBUSY; + ret = -EBUSY; + mutex_unlock(&nv_subdev(ppm)->mutex); - return 0; + return ret; } struct nvkm_oclass -- cgit v0.10.2 From 1196bcf92142ad6f022df01ab8120aefa908eaea Mon Sep 17 00:00:00 2001 From: Thierry Reding Date: Mon, 21 Jul 2014 14:02:58 +0200 Subject: drm/nouveau/disp: Use NULL for pointers The return type of exec_lookup() is struct nvkm_output *, so it should return NULL rather than 0. Signed-off-by: Thierry Reding Reviewed-by: Emil Velikov Signed-off-by: Ben Skeggs diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/disp/gf110.c b/drivers/gpu/drm/nouveau/nvkm/engine/disp/gf110.c index 9ef6728..7f2f05f 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/disp/gf110.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/disp/gf110.c @@ -809,7 +809,7 @@ exec_lookup(struct nv50_disp_priv *priv, int head, int or, u32 ctrl, case 0x00000900: type = DCB_OUTPUT_DP; mask = 2; break; default: nv_error(priv, "unknown SOR mc 0x%08x\n", ctrl); - return 0x0000; + return NULL; } } -- cgit v0.10.2 From 360ccb8436ce83b9c25f969cee7b1b607899063b Mon Sep 17 00:00:00 2001 From: Ilia Mirkin Date: Thu, 18 Jun 2015 23:59:06 -0400 Subject: drm/nouveau/bios: add 0x59 and 0x5a opcodes Opcode 0x5a is a register write for data looked up from another part of the VBIOS image. 0x59 is a more complex opcode, but we may as well recognize it. These occur on a single known instance of Riva TNT2 hardware. Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=91025 Signed-off-by: Ilia Mirkin Signed-off-by: Ben Skeggs diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/bios/init.c b/drivers/gpu/drm/nouveau/nvkm/subdev/bios/init.c index f67cdae..1f590f8 100644 --- a/drivers/gpu/drm/nouveau/nvkm/subdev/bios/init.c +++ b/drivers/gpu/drm/nouveau/nvkm/subdev/bios/init.c @@ -577,6 +577,9 @@ init_reserved(struct nvbios_init *init) u8 length, i; switch (opcode) { + case 0x59: + length = 7; + break; case 0xaa: length = 4; break; @@ -1285,6 +1288,25 @@ init_zm_reg_sequence(struct nvbios_init *init) } /** + * INIT_ZM_REG_INDIRECT - opcode 0x5a + * + */ +static void +init_zm_reg_indirect(struct nvbios_init *init) +{ + struct nvkm_bios *bios = init->bios; + u32 reg = nv_ro32(bios, init->offset + 1); + u16 addr = nv_ro16(bios, init->offset + 5); + u32 data = nv_ro32(bios, addr); + + trace("ZM_REG_INDIRECT\tR[0x%06x] = VBIOS[0x%04x] = 0x%08x\n", + reg, addr, data); + init->offset += 7; + + init_wr32(init, addr, data); +} + +/** * INIT_SUB_DIRECT - opcode 0x5b * */ @@ -2145,6 +2167,8 @@ static struct nvbios_init_opcode { [0x56] = { init_condition_time }, [0x57] = { init_ltime }, [0x58] = { init_zm_reg_sequence }, + [0x59] = { init_reserved }, + [0x5a] = { init_zm_reg_indirect }, [0x5b] = { init_sub_direct }, [0x5c] = { init_jump }, [0x5e] = { init_i2c_if }, -- cgit v0.10.2 From d31b11d85814ff669cfeb93c0f85b9d659a85ef8 Mon Sep 17 00:00:00 2001 From: Ilia Mirkin Date: Fri, 19 Jun 2015 01:19:40 -0400 Subject: drm/nouveau/bios: add proper support for opcode 0x59 More analysis shows that this is identical to 0x79 except that it loads the frequency indirectly from elsewhere in the VBIOS. Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=91025 Signed-off-by: Ilia Mirkin Signed-off-by: Ben Skeggs diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/bios/init.c b/drivers/gpu/drm/nouveau/nvkm/subdev/bios/init.c index 1f590f8..f4611e3 100644 --- a/drivers/gpu/drm/nouveau/nvkm/subdev/bios/init.c +++ b/drivers/gpu/drm/nouveau/nvkm/subdev/bios/init.c @@ -577,9 +577,6 @@ init_reserved(struct nvbios_init *init) u8 length, i; switch (opcode) { - case 0x59: - length = 7; - break; case 0xaa: length = 4; break; @@ -1288,6 +1285,25 @@ init_zm_reg_sequence(struct nvbios_init *init) } /** + * INIT_PLL_INDIRECT - opcode 0x59 + * + */ +static void +init_pll_indirect(struct nvbios_init *init) +{ + struct nvkm_bios *bios = init->bios; + u32 reg = nv_ro32(bios, init->offset + 1); + u16 addr = nv_ro16(bios, init->offset + 5); + u32 freq = (u32)nv_ro16(bios, addr) * 1000; + + trace("PLL_INDIRECT\tR[0x%06x] =PLL= VBIOS[%04x] = %dkHz\n", + reg, addr, freq); + init->offset += 7; + + init_prog_pll(init, reg, freq); +} + +/** * INIT_ZM_REG_INDIRECT - opcode 0x5a * */ @@ -2167,7 +2183,7 @@ static struct nvbios_init_opcode { [0x56] = { init_condition_time }, [0x57] = { init_ltime }, [0x58] = { init_zm_reg_sequence }, - [0x59] = { init_reserved }, + [0x59] = { init_pll_indirect }, [0x5a] = { init_zm_reg_indirect }, [0x5b] = { init_sub_direct }, [0x5c] = { init_jump }, -- cgit v0.10.2 From d108142c0840ce389cd9898aa76943b3fb430b83 Mon Sep 17 00:00:00 2001 From: Ilia Mirkin Date: Mon, 29 Jun 2015 04:07:20 -0400 Subject: drm/nouveau/fbcon/nv11-: correctly account for ring space usage The RING_SPACE macro accounts how much space is used up so it's important to ask it for the right amount. Incorrect accounting of this can cause page faults down the line as writes are attempted outside of the ring. Signed-off-by: Ilia Mirkin Cc: stable@vger.kernel.org Signed-off-by: Ben Skeggs diff --git a/drivers/gpu/drm/nouveau/nv04_fbcon.c b/drivers/gpu/drm/nouveau/nv04_fbcon.c index 4ef602c..495c576 100644 --- a/drivers/gpu/drm/nouveau/nv04_fbcon.c +++ b/drivers/gpu/drm/nouveau/nv04_fbcon.c @@ -203,7 +203,7 @@ nv04_fbcon_accel_init(struct fb_info *info) if (ret) return ret; - if (RING_SPACE(chan, 49)) { + if (RING_SPACE(chan, 49 + (device->info.chipset >= 0x11 ? 4 : 0))) { nouveau_fbcon_gpu_lockup(info); return 0; } -- cgit v0.10.2 From 4fd26cb1e4049c7a630d86d52864a5722c7453ac Mon Sep 17 00:00:00 2001 From: Ilia Mirkin Date: Mon, 29 Jun 2015 20:43:46 -0400 Subject: drm/nouveau/fbcon/gf100-: reduce RING_SPACE allocation We only emit 58 words to the ring, not 60. Signed-off-by: Ilia Mirkin Signed-off-by: Ben Skeggs diff --git a/drivers/gpu/drm/nouveau/nvc0_fbcon.c b/drivers/gpu/drm/nouveau/nvc0_fbcon.c index 6124667..fcd2e5f 100644 --- a/drivers/gpu/drm/nouveau/nvc0_fbcon.c +++ b/drivers/gpu/drm/nouveau/nvc0_fbcon.c @@ -188,7 +188,7 @@ nvc0_fbcon_accel_init(struct fb_info *info) return -EINVAL; } - ret = RING_SPACE(chan, 60); + ret = RING_SPACE(chan, 58); if (ret) { WARN_ON(1); nouveau_fbcon_gpu_lockup(info); -- cgit v0.10.2 From b7eea2d7e67cd4b65f42c00624277dae780f7009 Mon Sep 17 00:00:00 2001 From: Ilia Mirkin Date: Mon, 29 Jun 2015 20:49:48 -0400 Subject: drm/nouveau/fbcon/g80: reduce PUSH_SPACE alloc, fire ring on accel init Only 58 words get written to the ring, not 59. Also, normalize the accel init wrt nvc0 and nv04 fbcon impls by firing the ring at accel init time rather than waiting until "later". Signed-off-by: Ilia Mirkin Signed-off-by: Ben Skeggs diff --git a/drivers/gpu/drm/nouveau/nv50_fbcon.c b/drivers/gpu/drm/nouveau/nv50_fbcon.c index 394c89a..901130b 100644 --- a/drivers/gpu/drm/nouveau/nv50_fbcon.c +++ b/drivers/gpu/drm/nouveau/nv50_fbcon.c @@ -188,7 +188,7 @@ nv50_fbcon_accel_init(struct fb_info *info) if (ret) return ret; - ret = RING_SPACE(chan, 59); + ret = RING_SPACE(chan, 58); if (ret) { nouveau_fbcon_gpu_lockup(info); return ret; @@ -252,6 +252,7 @@ nv50_fbcon_accel_init(struct fb_info *info) OUT_RING(chan, info->var.yres_virtual); OUT_RING(chan, upper_32_bits(fb->vma.offset)); OUT_RING(chan, lower_32_bits(fb->vma.offset)); + FIRE_RING(chan); return 0; } -- cgit v0.10.2 From 134bf30c06f057d6b8d90132e8f8b3cd2be79572 Mon Sep 17 00:00:00 2001 From: Colin Ian King Date: Thu, 23 Jul 2015 16:47:59 +0100 Subject: dm cache policy smq: fix alloc_bitset check that always evaluates as false static analysis by cppcheck has found a check on alloc_bitset that always evaluates as false and hence never finds an allocation failure: [drivers/md/dm-cache-policy-smq.c:1689]: (warning) Logical conjunction always evaluates to false: !EXPR && EXPR. Fix this by removing the incorrect mq->cache_hit_bits check Signed-off-by: Colin Ian King Signed-off-by: Mike Snitzer diff --git a/drivers/md/dm-cache-policy-smq.c b/drivers/md/dm-cache-policy-smq.c index b6f2265..48a4a82 100644 --- a/drivers/md/dm-cache-policy-smq.c +++ b/drivers/md/dm-cache-policy-smq.c @@ -1686,7 +1686,7 @@ static struct dm_cache_policy *smq_create(dm_cblock_t cache_size, if (from_cblock(cache_size)) { mq->cache_hit_bits = alloc_bitset(from_cblock(cache_size)); - if (!mq->cache_hit_bits && mq->cache_hit_bits) { + if (!mq->cache_hit_bits) { DMERR("couldn't allocate cache hit bitset"); goto bad_cache_hit_bits; } -- cgit v0.10.2 From 6ed443c07f1e7f819983422b16598facb152250d Mon Sep 17 00:00:00 2001 From: Baruch Siach Date: Sun, 5 Jul 2015 09:55:44 +0300 Subject: dm crypt: update wiki page URL Cryptsetup moved to gitlab. This is a leftover from commit e44f23b32dc7 (dm crypt: update URLs to new cryptsetup project page, 2015-04-05). Signed-off-by: Baruch Siach Signed-off-by: Mike Snitzer diff --git a/drivers/md/Kconfig b/drivers/md/Kconfig index b597273..bfec3bd 100644 --- a/drivers/md/Kconfig +++ b/drivers/md/Kconfig @@ -259,7 +259,7 @@ config DM_CRYPT the ciphers you're going to use in the cryptoapi configuration. For further information on dm-crypt and userspace tools see: - + To compile this code as a module, choose M here: the module will be called dm-crypt. -- cgit v0.10.2 From 39da809e86274abbd9010365906045e02da3944a Mon Sep 17 00:00:00 2001 From: Rob Herring Date: Sun, 5 Jul 2015 10:18:41 -0500 Subject: of: add HAS_IOMEM depends to OF_ADDRESS MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit On UML builds, of_address.c fails to compile: ../drivers/of/address.c:873:2: error: implicit declaration of function ‘ioremap’ [-Werror=implicit-function-declaration] This is due to CONFIG_OF now being user selectable. Add a dependency on HAS_IOMEM to OF_ADDRESS in order to fix this. Signed-off-by: Rob Herring Cc: Grant Likely diff --git a/drivers/of/Kconfig b/drivers/of/Kconfig index 8df1b17..59bb855 100644 --- a/drivers/of/Kconfig +++ b/drivers/of/Kconfig @@ -47,7 +47,7 @@ config OF_DYNAMIC config OF_ADDRESS def_bool y - depends on !SPARC + depends on !SPARC && HAS_IOMEM select OF_ADDRESS_PCI if PCI config OF_ADDRESS_PCI -- cgit v0.10.2 From 35068ce8cbf1749ef1a4b9b1493af83b8488c37b Mon Sep 17 00:00:00 2001 From: Tomeu Vizoso Date: Wed, 1 Jul 2015 09:10:43 +0200 Subject: of: constify drv arg of of_driver_match_device stub With this change the stub has the same signature as the actual function, preventing this compiler warning when building without CONFIG_OF: drivers/base/property.c: In function 'fwnode_driver_match_device': >> drivers/base/property.c:608:38: warning: passing argument 2 of 'of_driver_match_device' discards 'const' qualifier from pointer target type return of_driver_match_device(dev, drv); ^ In file included from drivers/base/property.c:18:0: include/linux/of_device.h:61:19: note: expected 'struct device_driver *' but argument is of type 'const struct device_driver *' static inline int of_driver_match_device(struct device *dev, ^ Signed-off-by: Tomeu Vizoso Signed-off-by: Rob Herring diff --git a/include/linux/of_device.h b/include/linux/of_device.h index 4c50854..cc7dd687 100644 --- a/include/linux/of_device.h +++ b/include/linux/of_device.h @@ -59,7 +59,7 @@ void of_dma_configure(struct device *dev, struct device_node *np); #else /* CONFIG_OF */ static inline int of_driver_match_device(struct device *dev, - struct device_driver *drv) + const struct device_driver *drv) { return 0; } -- cgit v0.10.2 From 3b6e644e6540405fd3582e0a90412fccbddd0e45 Mon Sep 17 00:00:00 2001 From: Alban Bedel Date: Fri, 3 Jul 2015 10:14:57 +0200 Subject: DEVICETREE: Misc fix for the AR7100 SPI controller binding Fix the clocks property documentation and use lower case for hex values in the example. Signed-off-by: Alban Bedel Signed-off-by: Rob Herring diff --git a/Documentation/devicetree/bindings/spi/spi-ath79.txt b/Documentation/devicetree/bindings/spi/spi-ath79.txt index f1ad9c3..9c696fa 100644 --- a/Documentation/devicetree/bindings/spi/spi-ath79.txt +++ b/Documentation/devicetree/bindings/spi/spi-ath79.txt @@ -3,7 +3,7 @@ Binding for Qualcomm Atheros AR7xxx/AR9xxx SPI controller Required properties: - compatible: has to be "qca,-spi", "qca,ar7100-spi" as fallback. - reg: Base address and size of the controllers memory area -- clocks: phandle to the AHB clock. +- clocks: phandle of the AHB clock. - clock-names: has to be "ahb". - #address-cells: <1>, as required by generic SPI binding. - #size-cells: <0>, also as required by generic SPI binding. @@ -12,9 +12,9 @@ Child nodes as per the generic SPI binding. Example: - spi@1F000000 { + spi@1f000000 { compatible = "qca,ar9132-spi", "qca,ar7100-spi"; - reg = <0x1F000000 0x10>; + reg = <0x1f000000 0x10>; clocks = <&pll 2>; clock-names = "ahb"; -- cgit v0.10.2 From 599ad5ac9a59587a46f6540a8e2ab79dff811f7d Mon Sep 17 00:00:00 2001 From: Krzysztof Kozlowski Date: Fri, 10 Jul 2015 15:26:12 +0900 Subject: of: Drop owner assignment from platform and i2c driver platform_driver and i2c_driver do not need to set an owner because core will set it. Signed-off-by: Krzysztof Kozlowski Signed-off-by: Rob Herring diff --git a/drivers/of/unittest.c b/drivers/of/unittest.c index 1801634..9f71770b6 100644 --- a/drivers/of/unittest.c +++ b/drivers/of/unittest.c @@ -979,7 +979,6 @@ static struct platform_driver unittest_driver = { .remove = unittest_remove, .driver = { .name = "unittest", - .owner = THIS_MODULE, .of_match_table = of_match_ptr(unittest_match), }, }; @@ -1666,7 +1665,6 @@ static const struct i2c_device_id unittest_i2c_dev_id[] = { static struct i2c_driver unittest_i2c_dev_driver = { .driver = { .name = "unittest-i2c-dev", - .owner = THIS_MODULE, }, .probe = unittest_i2c_dev_probe, .remove = unittest_i2c_dev_remove, @@ -1761,7 +1759,6 @@ static const struct i2c_device_id unittest_i2c_mux_id[] = { static struct i2c_driver unittest_i2c_mux_driver = { .driver = { .name = "unittest-i2c-mux", - .owner = THIS_MODULE, }, .probe = unittest_i2c_mux_probe, .remove = unittest_i2c_mux_remove, -- cgit v0.10.2 From 03d5eb65b53889fe98a5ecddfe205c16e3093190 Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Mon, 27 Jul 2015 10:23:19 -0400 Subject: NFS: Fix a memory leak in nfs_do_recoalesce If the function exits early, then we must put those requests that were not processed back onto the &mirror->pg_list so they can be cleaned up by nfs_pgio_error(). Fixes: a7d42ddb30997 ("nfs: add mirroring support to pgio layer") Cc: stable@vger.kernel.org # v4.0+ Signed-off-by: Trond Myklebust diff --git a/fs/nfs/pagelist.c b/fs/nfs/pagelist.c index 1da68d3..8ea5920 100644 --- a/fs/nfs/pagelist.c +++ b/fs/nfs/pagelist.c @@ -1109,8 +1109,11 @@ static int nfs_do_recoalesce(struct nfs_pageio_descriptor *desc) nfs_list_remove_request(req); if (__nfs_pageio_add_request(desc, req)) continue; - if (desc->pg_error < 0) + if (desc->pg_error < 0) { + list_splice_tail(&head, &mirror->pg_list); + mirror->pg_recoalesce = 1; return 0; + } break; } } while (mirror->pg_recoalesce); -- cgit v0.10.2 From d4c30454db732d5a377abfc309c7a629f3bbaeb0 Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Fri, 24 Jul 2015 12:31:33 -0400 Subject: NFS: Don't clear desc->pg_moreio in nfs_do_recoalesce() Recoalescing does not affect whether or not we've already sent off I/O, and doing so means that we end up sending a bunch of synchronous for cases where we actually need to be using unstable writes. Signed-off-by: Trond Myklebust diff --git a/fs/nfs/pagelist.c b/fs/nfs/pagelist.c index 8ea5920..4984bbe 100644 --- a/fs/nfs/pagelist.c +++ b/fs/nfs/pagelist.c @@ -1100,8 +1100,6 @@ static int nfs_do_recoalesce(struct nfs_pageio_descriptor *desc) mirror->pg_base = 0; mirror->pg_recoalesce = 0; - desc->pg_moreio = 0; - while (!list_empty(&head)) { struct nfs_page *req; -- cgit v0.10.2 From 7932c0bd7740f4cd2aa168d3ce0199e7af7d72d5 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Marc-Andr=C3=A9=20Lureau?= Date: Fri, 17 Jul 2015 15:32:03 +0200 Subject: vhost: actually track log eventfd file MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit While reviewing vhost log code, I found out that log_file is never set. Note: I haven't tested the change (QEMU doesn't use LOG_FD yet). Cc: stable@vger.kernel.org Signed-off-by: Marc-André Lureau Signed-off-by: Michael S. Tsirkin diff --git a/drivers/vhost/vhost.c b/drivers/vhost/vhost.c index a9fe859..95bdb90 100644 --- a/drivers/vhost/vhost.c +++ b/drivers/vhost/vhost.c @@ -995,6 +995,7 @@ long vhost_dev_ioctl(struct vhost_dev *d, unsigned int ioctl, void __user *argp) } if (eventfp != d->log_file) { filep = d->log_file; + d->log_file = eventfp; ctx = d->log_ctx; d->log_ctx = eventfp ? eventfd_ctx_fileget(eventfp) : NULL; -- cgit v0.10.2 From 1e0994730f772580ff98754eb5595190cdf371ef Mon Sep 17 00:00:00 2001 From: Igor Mammedov Date: Wed, 15 Jul 2015 16:48:50 +0200 Subject: vhost: fix error handling for memory region alloc callers of vhost_kvzalloc() expect the same behaviour on allocation error as from kmalloc/vmalloc i.e. NULL return value. So just return vzmalloc() returned value instead of returning ERR_PTR(-ENOMEM) Fixes: 4de7255f7d2be5 ("vhost: extend memory regions allocation to vmalloc") Spotted-by: Dan Carpenter Suggested-by: Julia Lawall Signed-off-by: Igor Mammedov Signed-off-by: Michael S. Tsirkin diff --git a/drivers/vhost/vhost.c b/drivers/vhost/vhost.c index 95bdb90..eec2f11 100644 --- a/drivers/vhost/vhost.c +++ b/drivers/vhost/vhost.c @@ -683,11 +683,8 @@ static void *vhost_kvzalloc(unsigned long size) { void *n = kzalloc(size, GFP_KERNEL | __GFP_NOWARN | __GFP_REPEAT); - if (!n) { + if (!n) n = vzalloc(size); - if (!n) - return ERR_PTR(-ENOMEM); - } return n; } -- cgit v0.10.2 From bdcc2cd14e4e86f1fb19c78e191a1b8da3d6dea0 Mon Sep 17 00:00:00 2001 From: "J. Bruce Fields" Date: Thu, 23 Jul 2015 11:08:43 -0400 Subject: NFSv4.2: handle NFS-specific llseek errors Handle NFS-specific llseek errors instead of letting them leak out to userspace. Reported-by: Benjamin Coddington Signed-off-by: J. Bruce Fields Signed-off-by: Trond Myklebust diff --git a/fs/nfs/nfs42proc.c b/fs/nfs/nfs42proc.c index f486b80..d731bbf 100644 --- a/fs/nfs/nfs42proc.c +++ b/fs/nfs/nfs42proc.c @@ -135,7 +135,7 @@ int nfs42_proc_deallocate(struct file *filep, loff_t offset, loff_t len) return err; } -loff_t nfs42_proc_llseek(struct file *filep, loff_t offset, int whence) +static loff_t _nfs42_proc_llseek(struct file *filep, loff_t offset, int whence) { struct inode *inode = file_inode(filep); struct nfs42_seek_args args = { @@ -171,6 +171,23 @@ loff_t nfs42_proc_llseek(struct file *filep, loff_t offset, int whence) return vfs_setpos(filep, res.sr_offset, inode->i_sb->s_maxbytes); } +loff_t nfs42_proc_llseek(struct file *filep, loff_t offset, int whence) +{ + struct nfs_server *server = NFS_SERVER(file_inode(filep)); + struct nfs4_exception exception = { }; + int err; + + do { + err = _nfs42_proc_llseek(filep, offset, whence); + if (err == -ENOTSUPP) + return -EOPNOTSUPP; + err = nfs4_handle_exception(server, err, &exception); + } while (exception.retry); + + return err; +} + + static void nfs42_layoutstat_prepare(struct rpc_task *task, void *calldata) { -- cgit v0.10.2 From 743c69e7c089ba1bea1b207c5829dd079a4e98f9 Mon Sep 17 00:00:00 2001 From: NeilBrown Date: Mon, 27 Jul 2015 10:55:35 +1000 Subject: sunrpc: translate -EAGAIN to -ENOBUFS when socket is writable. The networking layer does not reliably report the distinction between a non-block write failing because: 1/ the queue is too full already and 2/ a memory allocation attempt failed. The distinction is important because in the first case it is appropriate to retry as soon as the socket reports that it is writable, and in the second case a small delay is required as the socket will most likely report as writable but kmalloc could still fail. sk_stream_wait_memory() exhibits this distinction nicely, setting 'vm_wait' if a small wait is needed. However in the non-blocking case it always returns -EAGAIN no matter the cause of the failure. This -EAGAIN call get all the way to sunrpc. The sunrpc layer expects EAGAIN to indicate the first cause, and ENOBUFS to indicate the second. Various documentation suggests that this is not unreasonable, but does not guarantee the desired error codes. The result of getting -EAGAIN when -ENOBUFS is expected is that the send is tried again in a tight loop and soft lockups are reported. so: add tests after calls to xs_sendpages() to translate -EAGAIN into -ENOBUFS if the socket is writable. This cannot happen inside xs_sendpages() as the test for "is socket writable" is different between TCP and UDP. With this change, the tight loop retrying xs_sendpages() becomes a loop which only retries every 250ms, and so will not trigger a soft-lockup warning. It is possible that the write did fail because the queue was too full and by the time xs_sendpages() completed, the queue was writable again. In this case an extra 250ms delay is inserted that isn't really needed. This circumstance suggests a degree of congestion so a delay is not necessarily a bad thing, and it can only cause a single 250ms delay, not a series of them. Signed-off-by: NeilBrown Signed-off-by: Trond Myklebust diff --git a/net/sunrpc/xprtsock.c b/net/sunrpc/xprtsock.c index 44c1927..4f48b1a 100644 --- a/net/sunrpc/xprtsock.c +++ b/net/sunrpc/xprtsock.c @@ -527,6 +527,10 @@ static int xs_local_send_request(struct rpc_task *task) true, &sent); dprintk("RPC: %s(%u) = %d\n", __func__, xdr->len - req->rq_bytes_sent, status); + + if (status == -EAGAIN && sock_writeable(transport->inet)) + status = -ENOBUFS; + if (likely(sent > 0) || status == 0) { req->rq_bytes_sent += sent; req->rq_xmit_bytes_sent += sent; @@ -590,6 +594,9 @@ static int xs_udp_send_request(struct rpc_task *task) if (status == -EPERM) goto process_status; + if (status == -EAGAIN && sock_writeable(transport->inet)) + status = -ENOBUFS; + if (sent > 0 || status == 0) { req->rq_xmit_bytes_sent += sent; if (sent >= req->rq_slen) @@ -687,6 +694,8 @@ static int xs_tcp_send_request(struct rpc_task *task) status = -EAGAIN; break; } + if (status == -EAGAIN && sk_stream_is_writeable(transport->inet)) + status = -ENOBUFS; switch (status) { case -ENOTSOCK: -- cgit v0.10.2 From f91b1feada0b6f0a4d33648155b3ded2c4e0707e Mon Sep 17 00:00:00 2001 From: Ard Biesheuvel Date: Sun, 26 Jul 2015 14:59:00 +0200 Subject: arm64/efi: map the entire UEFI vendor string before reading it At boot, the UTF-16 UEFI vendor string is copied from the system table into a char array with a size of 100 bytes. However, this size of 100 bytes is also used for memremapping() the source, which may not be sufficient if the vendor string exceeds 50 UTF-16 characters, and the placement of the vendor string inside a 4 KB page happens to leave the end unmapped. So use the correct '100 * sizeof(efi_char16_t)' for the size of the mapping. Signed-off-by: Ard Biesheuvel Fixes: f84d02755f5a ("arm64: add EFI runtime services") Cc: # 3.16+ Signed-off-by: Catalin Marinas diff --git a/arch/arm64/kernel/efi.c b/arch/arm64/kernel/efi.c index 9d4aa18..e8ca6ea 100644 --- a/arch/arm64/kernel/efi.c +++ b/arch/arm64/kernel/efi.c @@ -122,12 +122,12 @@ static int __init uefi_init(void) /* Show what we know for posterity */ c16 = early_memremap(efi_to_phys(efi.systab->fw_vendor), - sizeof(vendor)); + sizeof(vendor) * sizeof(efi_char16_t)); if (c16) { for (i = 0; i < (int) sizeof(vendor) - 1 && *c16; ++i) vendor[i] = c16[i]; vendor[i] = '\0'; - early_memunmap(c16, sizeof(vendor)); + early_memunmap(c16, sizeof(vendor) * sizeof(efi_char16_t)); } pr_info("EFI v%u.%.02u by %s\n", -- cgit v0.10.2 From 7a6e0706c09b121d4656b055d9b0494320246f4f Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Mon, 27 Jul 2015 14:24:48 -0700 Subject: macb: Fix build with macro'ized readl/writel. If an architecture defines readl/writel using CPP macros, we get the following kinds of build failure: > > > drivers/net/ethernet/cadence/macb.c:164:1: error: macro "writel" > > > passed 3 arguments, but takes just 2 > macb_or_gem_writel(bp, SA1B, bottom); > ^ Rename the methods so that this doesn't happen. Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/cadence/macb.c b/drivers/net/ethernet/cadence/macb.c index c638757..bf9eb2e 100644 --- a/drivers/net/ethernet/cadence/macb.c +++ b/drivers/net/ethernet/cadence/macb.c @@ -506,7 +506,7 @@ static void macb_update_stats(struct macb *bp) WARN_ON((unsigned long)(end - p - 1) != (MACB_TPF - MACB_PFR) / 4); for(; p < end; p++, offset += 4) - *p += bp->readl(bp, offset); + *p += bp->macb_reg_readl(bp, offset); } static int macb_halt_tx(struct macb *bp) @@ -1934,14 +1934,14 @@ static void gem_update_stats(struct macb *bp) for (i = 0; i < GEM_STATS_LEN; ++i, ++p) { u32 offset = gem_statistics[i].offset; - u64 val = bp->readl(bp, offset); + u64 val = bp->macb_reg_readl(bp, offset); bp->ethtool_stats[i] += val; *p += val; if (offset == GEM_OCTTXL || offset == GEM_OCTRXL) { /* Add GEM_OCTTXH, GEM_OCTRXH */ - val = bp->readl(bp, offset + 4); + val = bp->macb_reg_readl(bp, offset + 4); bp->ethtool_stats[i] += ((u64)val) << 32; *(++p) += val; } @@ -2867,11 +2867,11 @@ static int macb_probe(struct platform_device *pdev) bp->regs = mem; bp->native_io = native_io; if (native_io) { - bp->readl = hw_readl_native; - bp->writel = hw_writel_native; + bp->macb_reg_readl = hw_readl_native; + bp->macb_reg_writel = hw_writel_native; } else { - bp->readl = hw_readl; - bp->writel = hw_writel; + bp->macb_reg_readl = hw_readl; + bp->macb_reg_writel = hw_writel; } bp->num_queues = num_queues; bp->queue_mask = queue_mask; diff --git a/drivers/net/ethernet/cadence/macb.h b/drivers/net/ethernet/cadence/macb.h index 2aa102e..1895b6b 100644 --- a/drivers/net/ethernet/cadence/macb.h +++ b/drivers/net/ethernet/cadence/macb.h @@ -429,12 +429,12 @@ | GEM_BF(name, value)) /* Register access macros */ -#define macb_readl(port, reg) (port)->readl((port), MACB_##reg) -#define macb_writel(port, reg, value) (port)->writel((port), MACB_##reg, (value)) -#define gem_readl(port, reg) (port)->readl((port), GEM_##reg) -#define gem_writel(port, reg, value) (port)->writel((port), GEM_##reg, (value)) -#define queue_readl(queue, reg) (queue)->bp->readl((queue)->bp, (queue)->reg) -#define queue_writel(queue, reg, value) (queue)->bp->writel((queue)->bp, (queue)->reg, (value)) +#define macb_readl(port, reg) (port)->macb_reg_readl((port), MACB_##reg) +#define macb_writel(port, reg, value) (port)->macb_reg_writel((port), MACB_##reg, (value)) +#define gem_readl(port, reg) (port)->macb_reg_readl((port), GEM_##reg) +#define gem_writel(port, reg, value) (port)->macb_reg_writel((port), GEM_##reg, (value)) +#define queue_readl(queue, reg) (queue)->bp->macb_reg_readl((queue)->bp, (queue)->reg) +#define queue_writel(queue, reg, value) (queue)->bp->macb_reg_writel((queue)->bp, (queue)->reg, (value)) /* Conditional GEM/MACB macros. These perform the operation to the correct * register dependent on whether the device is a GEM or a MACB. For registers @@ -782,8 +782,8 @@ struct macb { bool native_io; /* hardware IO accessors */ - u32 (*readl)(struct macb *bp, int offset); - void (*writel)(struct macb *bp, int offset, u32 value); + u32 (*macb_reg_readl)(struct macb *bp, int offset); + void (*macb_reg_writel)(struct macb *bp, int offset, u32 value); unsigned int rx_tail; unsigned int rx_prepared_head; -- cgit v0.10.2 From 1513069edcf8dd86cfd8d5daef482b97d6b93df6 Mon Sep 17 00:00:00 2001 From: Alexander Duyck Date: Mon, 27 Jul 2015 13:08:06 -0700 Subject: fib_trie: Drop unnecessary calls to leaf_pull_suffix It was reported that update_suffix was taking a long time on systems where a large number of leaves were attached to a single node. As it turns out fib_table_flush was calling update_suffix for each leaf that didn't have all of the aliases stripped from it. As a result, on this large node removing one leaf would result in us calling update_suffix for every other leaf on the node. The fix is to just remove the calls to leaf_pull_suffix since they are redundant as we already have a call in resize that will go through and update the suffix length for the node before we exit out of fib_table_flush or fib_table_flush_external. Reported-by: David Ahern Signed-off-by: Alexander Duyck Tested-by: David Ahern Signed-off-by: David S. Miller diff --git a/net/ipv4/fib_trie.c b/net/ipv4/fib_trie.c index 81797e0..37c4bb8 100644 --- a/net/ipv4/fib_trie.c +++ b/net/ipv4/fib_trie.c @@ -1793,8 +1793,6 @@ void fib_table_flush_external(struct fib_table *tb) if (hlist_empty(&n->leaf)) { put_child_root(pn, n->key, NULL); node_free(n); - } else { - leaf_pull_suffix(pn, n); } } } @@ -1864,8 +1862,6 @@ int fib_table_flush(struct fib_table *tb) if (hlist_empty(&n->leaf)) { put_child_root(pn, n->key, NULL); node_free(n); - } else { - leaf_pull_suffix(pn, n); } } -- cgit v0.10.2 From c5c62f1bb0e1fc94ab77ec01e92ccab5cb249742 Mon Sep 17 00:00:00 2001 From: Ivan Vecera Date: Thu, 23 Jul 2015 16:37:43 +0200 Subject: macvtap: fix network header pointer for VLAN tagged pkts Network header is set with offset ETH_HLEN but it is not true for VLAN (multiple-)tagged and results in checksum issues in lower devices. v2: leave skb->protocol untouched (thx Vlad), comment added v3: moved after skb_probe_transport_header() call (thx Toshiaki) Signed-off-by: Ivan Vecera Signed-off-by: David S. Miller diff --git a/drivers/net/macvtap.c b/drivers/net/macvtap.c index 3b933bb..edd7734 100644 --- a/drivers/net/macvtap.c +++ b/drivers/net/macvtap.c @@ -719,6 +719,7 @@ static ssize_t macvtap_get_user(struct macvtap_queue *q, struct msghdr *m, struct virtio_net_hdr vnet_hdr = { 0 }; int vnet_hdr_len = 0; int copylen = 0; + int depth; bool zerocopy = false; size_t linear; ssize_t n; @@ -804,6 +805,12 @@ static ssize_t macvtap_get_user(struct macvtap_queue *q, struct msghdr *m, skb_probe_transport_header(skb, ETH_HLEN); + /* Move network header to the right position for VLAN tagged packets */ + if ((skb->protocol == htons(ETH_P_8021Q) || + skb->protocol == htons(ETH_P_8021AD)) && + __vlan_get_protocol(skb, skb->protocol, &depth) != 0) + skb_set_network_header(skb, depth); + rcu_read_lock(); vlan = rcu_dereference(q->vlan); /* copy skb_ubuf_info for callback when skb has no error */ -- cgit v0.10.2 From f580dd042823294b5b548e0f8bf1ba7a4b114fa5 Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Sat, 11 Jul 2015 17:48:52 +0200 Subject: SUNRPC: Report TCP errors to the caller Signed-off-by: Trond Myklebust diff --git a/net/sunrpc/xprtsock.c b/net/sunrpc/xprtsock.c index 4f48b1a..6a21368 100644 --- a/net/sunrpc/xprtsock.c +++ b/net/sunrpc/xprtsock.c @@ -677,9 +677,6 @@ static int xs_tcp_send_request(struct rpc_task *task) dprintk("RPC: xs_tcp_send_request(%u) = %d\n", xdr->len - req->rq_bytes_sent, status); - if (unlikely(sent == 0 && status < 0)) - break; - /* If we've sent the entire packet, immediately * reset the count of bytes sent. */ req->rq_bytes_sent += sent; @@ -689,10 +686,12 @@ static int xs_tcp_send_request(struct rpc_task *task) return 0; } - if (sent != 0) - continue; - status = -EAGAIN; - break; + if (status < 0) + break; + if (sent == 0) { + status = -EAGAIN; + break; + } } if (status == -EAGAIN && sk_stream_is_writeable(transport->inet)) status = -ENOBUFS; -- cgit v0.10.2 From 158cd4af8dedbda0d612d448c724c715d0dda649 Mon Sep 17 00:00:00 2001 From: Lars Westerhoff Date: Tue, 28 Jul 2015 01:32:21 +0300 Subject: packet: missing dev_put() in packet_do_bind() When binding a PF_PACKET socket, the use count of the bound interface is always increased with dev_hold in dev_get_by_{index,name}. However, when rebound with the same protocol and device as in the previous bind the use count of the interface was not decreased. Ultimately, this caused the deletion of the interface to fail with the following message: unregister_netdevice: waiting for dummy0 to become free. Usage count = 1 This patch moves the dev_put out of the conditional part that was only executed when either the protocol or device changed on a bind. Fixes: 902fefb82ef7 ('packet: improve socket create/bind latency in some cases') Signed-off-by: Lars Westerhoff Signed-off-by: Dan Carpenter Reviewed-by: Daniel Borkmann Signed-off-by: David S. Miller diff --git a/net/packet/af_packet.c b/net/packet/af_packet.c index c9e8741..c7c42eb 100644 --- a/net/packet/af_packet.c +++ b/net/packet/af_packet.c @@ -2784,7 +2784,7 @@ static int packet_release(struct socket *sock) static int packet_do_bind(struct sock *sk, struct net_device *dev, __be16 proto) { struct packet_sock *po = pkt_sk(sk); - const struct net_device *dev_curr; + struct net_device *dev_curr; __be16 proto_curr; bool need_rehook; @@ -2808,15 +2808,13 @@ static int packet_do_bind(struct sock *sk, struct net_device *dev, __be16 proto) po->num = proto; po->prot_hook.type = proto; - - if (po->prot_hook.dev) - dev_put(po->prot_hook.dev); - po->prot_hook.dev = dev; po->ifindex = dev ? dev->ifindex : 0; packet_cached_dev_assign(po, dev); } + if (dev_curr) + dev_put(dev_curr); if (proto == 0 || !need_rehook) goto out_unlock; -- cgit v0.10.2 From ab80ee3895a76f1beba35847911d89833fbb68ab Mon Sep 17 00:00:00 2001 From: Jiri Kosina Date: Fri, 24 Jul 2015 14:33:21 -0700 Subject: Input: synaptics - dump ext10 capabilities as well Make extended capabilities obtained through $10 query also available in touchpad identification. Signed-off-by: Jiri Kosina Reviewed-by: Benjamin Tissoires Signed-off-by: Dmitry Torokhov diff --git a/drivers/input/mouse/synaptics.c b/drivers/input/mouse/synaptics.c index 3a32caf..6025eb4 100644 --- a/drivers/input/mouse/synaptics.c +++ b/drivers/input/mouse/synaptics.c @@ -1484,12 +1484,12 @@ static int __synaptics_init(struct psmouse *psmouse, bool absolute_mode) priv->pkt_type = SYN_MODEL_NEWABS(priv->model_id) ? SYN_NEWABS : SYN_OLDABS; psmouse_info(psmouse, - "Touchpad model: %ld, fw: %ld.%ld, id: %#lx, caps: %#lx/%#lx/%#lx, board id: %lu, fw id: %lu\n", + "Touchpad model: %ld, fw: %ld.%ld, id: %#lx, caps: %#lx/%#lx/%#lx/%#lx, board id: %lu, fw id: %lu\n", SYN_ID_MODEL(priv->identity), SYN_ID_MAJOR(priv->identity), SYN_ID_MINOR(priv->identity), priv->model_id, priv->capabilities, priv->ext_cap, priv->ext_cap_0c, - priv->board_id, priv->firmware_id); + priv->ext_cap_10, priv->board_id, priv->firmware_id); set_input_params(psmouse, priv); -- cgit v0.10.2 From efbd34702fb1cb66e08606a623a8a76f86b300f9 Mon Sep 17 00:00:00 2001 From: Henrik Rydberg Date: Fri, 24 Jul 2015 14:44:37 -0700 Subject: Input: bcm5974 - prepare for a new trackpad generation With the advent of the Macbook Pro 12, we see a new generation of trackpads, capable of force sensoring and haptic feedback. This patch prepares for the new device by adding configuration data for the code paths that would otherwise look different. Tested-by: John Horan Tested-by: Jochen Radmacher Tested-by: Yang Hongyang Tested-by: Yen-Chin, Lee Tested-by: George Hilios Tested-by: Janez Urevc Signed-off-by: Henrik Rydberg Signed-off-by: Dmitry Torokhov diff --git a/drivers/input/mouse/bcm5974.c b/drivers/input/mouse/bcm5974.c index b10709f..a596b9b 100644 --- a/drivers/input/mouse/bcm5974.c +++ b/drivers/input/mouse/bcm5974.c @@ -184,17 +184,37 @@ enum tp_type { }; /* trackpad finger data offsets, le16-aligned */ -#define FINGER_TYPE1 (13 * sizeof(__le16)) -#define FINGER_TYPE2 (15 * sizeof(__le16)) -#define FINGER_TYPE3 (19 * sizeof(__le16)) +#define HEADER_TYPE1 (13 * sizeof(__le16)) +#define HEADER_TYPE2 (15 * sizeof(__le16)) +#define HEADER_TYPE3 (19 * sizeof(__le16)) /* trackpad button data offsets */ +#define BUTTON_TYPE1 0 #define BUTTON_TYPE2 15 #define BUTTON_TYPE3 23 /* list of device capability bits */ #define HAS_INTEGRATED_BUTTON 1 +/* trackpad finger data block size */ +#define FSIZE_TYPE1 (14 * sizeof(__le16)) +#define FSIZE_TYPE2 (14 * sizeof(__le16)) +#define FSIZE_TYPE3 (14 * sizeof(__le16)) + +/* offset from header to finger struct */ +#define DELTA_TYPE1 (0 * sizeof(__le16)) +#define DELTA_TYPE2 (0 * sizeof(__le16)) +#define DELTA_TYPE3 (0 * sizeof(__le16)) + +/* usb control message mode switch data */ +#define USBMSG_TYPE1 8, 0x300, 0, 0, 0x1, 0x8 +#define USBMSG_TYPE2 8, 0x300, 0, 0, 0x1, 0x8 +#define USBMSG_TYPE3 8, 0x300, 0, 0, 0x1, 0x8 + +/* Wellspring initialization constants */ +#define BCM5974_WELLSPRING_MODE_READ_REQUEST_ID 1 +#define BCM5974_WELLSPRING_MODE_WRITE_REQUEST_ID 9 + /* trackpad finger structure, le16-aligned */ struct tp_finger { __le16 origin; /* zero when switching track finger */ @@ -213,8 +233,6 @@ struct tp_finger { /* trackpad finger data size, empirically at least ten fingers */ #define MAX_FINGERS 16 -#define SIZEOF_FINGER sizeof(struct tp_finger) -#define SIZEOF_ALL_FINGERS (MAX_FINGERS * SIZEOF_FINGER) #define MAX_FINGER_ORIENTATION 16384 /* device-specific parameters */ @@ -232,8 +250,17 @@ struct bcm5974_config { int bt_datalen; /* data length of the button interface */ int tp_ep; /* the endpoint of the trackpad interface */ enum tp_type tp_type; /* type of trackpad interface */ - int tp_offset; /* offset to trackpad finger data */ + int tp_header; /* bytes in header block */ int tp_datalen; /* data length of the trackpad interface */ + int tp_button; /* offset to button data */ + int tp_fsize; /* bytes in single finger block */ + int tp_delta; /* offset from header to finger struct */ + int um_size; /* usb control message length */ + int um_req_val; /* usb control message value */ + int um_req_idx; /* usb control message index */ + int um_switch_idx; /* usb control message mode switch index */ + int um_switch_on; /* usb control message mode switch on */ + int um_switch_off; /* usb control message mode switch off */ struct bcm5974_param p; /* finger pressure limits */ struct bcm5974_param w; /* finger width limits */ struct bcm5974_param x; /* horizontal limits */ @@ -259,6 +286,24 @@ struct bcm5974 { int slots[MAX_FINGERS]; /* slot assignments */ }; +/* trackpad finger block data, le16-aligned */ +static const struct tp_finger *get_tp_finger(const struct bcm5974 *dev, int i) +{ + const struct bcm5974_config *c = &dev->cfg; + u8 *f_base = dev->tp_data + c->tp_header + c->tp_delta; + + return (const struct tp_finger *)(f_base + i * c->tp_fsize); +} + +#define DATAFORMAT(type) \ + type, \ + HEADER_##type, \ + HEADER_##type + (MAX_FINGERS) * (FSIZE_##type), \ + BUTTON_##type, \ + FSIZE_##type, \ + DELTA_##type, \ + USBMSG_##type + /* logical signal quality */ #define SN_PRESSURE 45 /* pressure signal-to-noise ratio */ #define SN_WIDTH 25 /* width signal-to-noise ratio */ @@ -273,7 +318,7 @@ static const struct bcm5974_config bcm5974_config_table[] = { USB_DEVICE_ID_APPLE_WELLSPRING_JIS, 0, 0x84, sizeof(struct bt_data), - 0x81, TYPE1, FINGER_TYPE1, FINGER_TYPE1 + SIZEOF_ALL_FINGERS, + 0x81, DATAFORMAT(TYPE1), { SN_PRESSURE, 0, 256 }, { SN_WIDTH, 0, 2048 }, { SN_COORD, -4824, 5342 }, @@ -286,7 +331,7 @@ static const struct bcm5974_config bcm5974_config_table[] = { USB_DEVICE_ID_APPLE_WELLSPRING2_JIS, 0, 0x84, sizeof(struct bt_data), - 0x81, TYPE1, FINGER_TYPE1, FINGER_TYPE1 + SIZEOF_ALL_FINGERS, + 0x81, DATAFORMAT(TYPE1), { SN_PRESSURE, 0, 256 }, { SN_WIDTH, 0, 2048 }, { SN_COORD, -4824, 4824 }, @@ -299,7 +344,7 @@ static const struct bcm5974_config bcm5974_config_table[] = { USB_DEVICE_ID_APPLE_WELLSPRING3_JIS, HAS_INTEGRATED_BUTTON, 0x84, sizeof(struct bt_data), - 0x81, TYPE2, FINGER_TYPE2, FINGER_TYPE2 + SIZEOF_ALL_FINGERS, + 0x81, DATAFORMAT(TYPE2), { SN_PRESSURE, 0, 300 }, { SN_WIDTH, 0, 2048 }, { SN_COORD, -4460, 5166 }, @@ -312,7 +357,7 @@ static const struct bcm5974_config bcm5974_config_table[] = { USB_DEVICE_ID_APPLE_WELLSPRING4_JIS, HAS_INTEGRATED_BUTTON, 0x84, sizeof(struct bt_data), - 0x81, TYPE2, FINGER_TYPE2, FINGER_TYPE2 + SIZEOF_ALL_FINGERS, + 0x81, DATAFORMAT(TYPE2), { SN_PRESSURE, 0, 300 }, { SN_WIDTH, 0, 2048 }, { SN_COORD, -4620, 5140 }, @@ -325,7 +370,7 @@ static const struct bcm5974_config bcm5974_config_table[] = { USB_DEVICE_ID_APPLE_WELLSPRING4A_JIS, HAS_INTEGRATED_BUTTON, 0x84, sizeof(struct bt_data), - 0x81, TYPE2, FINGER_TYPE2, FINGER_TYPE2 + SIZEOF_ALL_FINGERS, + 0x81, DATAFORMAT(TYPE2), { SN_PRESSURE, 0, 300 }, { SN_WIDTH, 0, 2048 }, { SN_COORD, -4616, 5112 }, @@ -338,7 +383,7 @@ static const struct bcm5974_config bcm5974_config_table[] = { USB_DEVICE_ID_APPLE_WELLSPRING5_JIS, HAS_INTEGRATED_BUTTON, 0x84, sizeof(struct bt_data), - 0x81, TYPE2, FINGER_TYPE2, FINGER_TYPE2 + SIZEOF_ALL_FINGERS, + 0x81, DATAFORMAT(TYPE2), { SN_PRESSURE, 0, 300 }, { SN_WIDTH, 0, 2048 }, { SN_COORD, -4415, 5050 }, @@ -351,7 +396,7 @@ static const struct bcm5974_config bcm5974_config_table[] = { USB_DEVICE_ID_APPLE_WELLSPRING6_JIS, HAS_INTEGRATED_BUTTON, 0x84, sizeof(struct bt_data), - 0x81, TYPE2, FINGER_TYPE2, FINGER_TYPE2 + SIZEOF_ALL_FINGERS, + 0x81, DATAFORMAT(TYPE2), { SN_PRESSURE, 0, 300 }, { SN_WIDTH, 0, 2048 }, { SN_COORD, -4620, 5140 }, @@ -364,7 +409,7 @@ static const struct bcm5974_config bcm5974_config_table[] = { USB_DEVICE_ID_APPLE_WELLSPRING5A_JIS, HAS_INTEGRATED_BUTTON, 0x84, sizeof(struct bt_data), - 0x81, TYPE2, FINGER_TYPE2, FINGER_TYPE2 + SIZEOF_ALL_FINGERS, + 0x81, DATAFORMAT(TYPE2), { SN_PRESSURE, 0, 300 }, { SN_WIDTH, 0, 2048 }, { SN_COORD, -4750, 5280 }, @@ -377,7 +422,7 @@ static const struct bcm5974_config bcm5974_config_table[] = { USB_DEVICE_ID_APPLE_WELLSPRING6A_JIS, HAS_INTEGRATED_BUTTON, 0x84, sizeof(struct bt_data), - 0x81, TYPE2, FINGER_TYPE2, FINGER_TYPE2 + SIZEOF_ALL_FINGERS, + 0x81, DATAFORMAT(TYPE2), { SN_PRESSURE, 0, 300 }, { SN_WIDTH, 0, 2048 }, { SN_COORD, -4620, 5140 }, @@ -390,7 +435,7 @@ static const struct bcm5974_config bcm5974_config_table[] = { USB_DEVICE_ID_APPLE_WELLSPRING7_JIS, HAS_INTEGRATED_BUTTON, 0x84, sizeof(struct bt_data), - 0x81, TYPE2, FINGER_TYPE2, FINGER_TYPE2 + SIZEOF_ALL_FINGERS, + 0x81, DATAFORMAT(TYPE2), { SN_PRESSURE, 0, 300 }, { SN_WIDTH, 0, 2048 }, { SN_COORD, -4750, 5280 }, @@ -403,7 +448,7 @@ static const struct bcm5974_config bcm5974_config_table[] = { USB_DEVICE_ID_APPLE_WELLSPRING7A_JIS, HAS_INTEGRATED_BUTTON, 0x84, sizeof(struct bt_data), - 0x81, TYPE2, FINGER_TYPE2, FINGER_TYPE2 + SIZEOF_ALL_FINGERS, + 0x81, DATAFORMAT(TYPE2), { SN_PRESSURE, 0, 300 }, { SN_WIDTH, 0, 2048 }, { SN_COORD, -4750, 5280 }, @@ -416,7 +461,7 @@ static const struct bcm5974_config bcm5974_config_table[] = { USB_DEVICE_ID_APPLE_WELLSPRING8_JIS, HAS_INTEGRATED_BUTTON, 0, sizeof(struct bt_data), - 0x83, TYPE3, FINGER_TYPE3, FINGER_TYPE3 + SIZEOF_ALL_FINGERS, + 0x83, DATAFORMAT(TYPE3), { SN_PRESSURE, 0, 300 }, { SN_WIDTH, 0, 2048 }, { SN_COORD, -4620, 5140 }, @@ -549,19 +594,18 @@ static int report_tp_state(struct bcm5974 *dev, int size) struct input_dev *input = dev->input; int raw_n, i, n = 0; - if (size < c->tp_offset || (size - c->tp_offset) % SIZEOF_FINGER != 0) + if (size < c->tp_header || (size - c->tp_header) % c->tp_fsize != 0) return -EIO; - /* finger data, le16-aligned */ - f = (const struct tp_finger *)(dev->tp_data + c->tp_offset); - raw_n = (size - c->tp_offset) / SIZEOF_FINGER; + raw_n = (size - c->tp_header) / c->tp_fsize; for (i = 0; i < raw_n; i++) { - if (raw2int(f[i].touch_major) == 0) + f = get_tp_finger(dev, i); + if (raw2int(f->touch_major) == 0) continue; - dev->pos[n].x = raw2int(f[i].abs_x); - dev->pos[n].y = c->y.min + c->y.max - raw2int(f[i].abs_y); - dev->index[n++] = &f[i]; + dev->pos[n].x = raw2int(f->abs_x); + dev->pos[n].y = c->y.min + c->y.max - raw2int(f->abs_y); + dev->index[n++] = f; } input_mt_assign_slots(input, dev->slots, dev->pos, n, 0); @@ -572,32 +616,22 @@ static int report_tp_state(struct bcm5974 *dev, int size) input_mt_sync_frame(input); - report_synaptics_data(input, c, f, raw_n); + report_synaptics_data(input, c, get_tp_finger(dev, 0), raw_n); - /* type 2 reports button events via ibt only */ - if (c->tp_type == TYPE2) { - int ibt = raw2int(dev->tp_data[BUTTON_TYPE2]); + /* later types report button events via integrated button only */ + if (c->caps & HAS_INTEGRATED_BUTTON) { + int ibt = raw2int(dev->tp_data[c->tp_button]); input_report_key(input, BTN_LEFT, ibt); } - if (c->tp_type == TYPE3) - input_report_key(input, BTN_LEFT, dev->tp_data[BUTTON_TYPE3]); - input_sync(input); return 0; } -/* Wellspring initialization constants */ -#define BCM5974_WELLSPRING_MODE_READ_REQUEST_ID 1 -#define BCM5974_WELLSPRING_MODE_WRITE_REQUEST_ID 9 -#define BCM5974_WELLSPRING_MODE_REQUEST_VALUE 0x300 -#define BCM5974_WELLSPRING_MODE_REQUEST_INDEX 0 -#define BCM5974_WELLSPRING_MODE_VENDOR_VALUE 0x01 -#define BCM5974_WELLSPRING_MODE_NORMAL_VALUE 0x08 - static int bcm5974_wellspring_mode(struct bcm5974 *dev, bool on) { + const struct bcm5974_config *c = &dev->cfg; int retval = 0, size; char *data; @@ -605,7 +639,7 @@ static int bcm5974_wellspring_mode(struct bcm5974 *dev, bool on) if (dev->cfg.tp_type == TYPE3) return 0; - data = kmalloc(8, GFP_KERNEL); + data = kmalloc(c->um_size, GFP_KERNEL); if (!data) { dev_err(&dev->intf->dev, "out of memory\n"); retval = -ENOMEM; @@ -616,28 +650,24 @@ static int bcm5974_wellspring_mode(struct bcm5974 *dev, bool on) size = usb_control_msg(dev->udev, usb_rcvctrlpipe(dev->udev, 0), BCM5974_WELLSPRING_MODE_READ_REQUEST_ID, USB_DIR_IN | USB_TYPE_CLASS | USB_RECIP_INTERFACE, - BCM5974_WELLSPRING_MODE_REQUEST_VALUE, - BCM5974_WELLSPRING_MODE_REQUEST_INDEX, data, 8, 5000); + c->um_req_val, c->um_req_idx, data, c->um_size, 5000); - if (size != 8) { + if (size != c->um_size) { dev_err(&dev->intf->dev, "could not read from device\n"); retval = -EIO; goto out; } /* apply the mode switch */ - data[0] = on ? - BCM5974_WELLSPRING_MODE_VENDOR_VALUE : - BCM5974_WELLSPRING_MODE_NORMAL_VALUE; + data[c->um_switch_idx] = on ? c->um_switch_on : c->um_switch_off; /* write configuration */ size = usb_control_msg(dev->udev, usb_sndctrlpipe(dev->udev, 0), BCM5974_WELLSPRING_MODE_WRITE_REQUEST_ID, USB_DIR_OUT | USB_TYPE_CLASS | USB_RECIP_INTERFACE, - BCM5974_WELLSPRING_MODE_REQUEST_VALUE, - BCM5974_WELLSPRING_MODE_REQUEST_INDEX, data, 8, 5000); + c->um_req_val, c->um_req_idx, data, c->um_size, 5000); - if (size != 8) { + if (size != c->um_size) { dev_err(&dev->intf->dev, "could not write to device\n"); retval = -EIO; goto out; -- cgit v0.10.2 From a4a2c54560f2c57b88ba0283f141b44f594c2337 Mon Sep 17 00:00:00 2001 From: Henrik Rydberg Date: Fri, 24 Jul 2015 14:45:05 -0700 Subject: HID: apple: Add support for the 2015 Macbook Pro This patch adds keyboard support for MacbookPro12,1 as WELLSPRING9 (0x0272, 0x0273, 0x0274). The touchpad is handled in a separate bcm5974 patch, as usual. Tested-by: John Horan Tested-by: Jochen Radmacher Tested-by: Yang Hongyang Tested-by: Yen-Chin, Lee Tested-by: George Hilios Tested-by: Janez Urevc Signed-off-by: Henrik Rydberg Acked-by: Jiri Kosina Signed-off-by: Dmitry Torokhov diff --git a/drivers/hid/hid-apple.c b/drivers/hid/hid-apple.c index f822fd2..884d82f 100644 --- a/drivers/hid/hid-apple.c +++ b/drivers/hid/hid-apple.c @@ -546,6 +546,12 @@ static const struct hid_device_id apple_devices[] = { .driver_data = APPLE_HAS_FN | APPLE_ISO_KEYBOARD }, { HID_USB_DEVICE(USB_VENDOR_ID_APPLE, USB_DEVICE_ID_APPLE_WELLSPRING8_JIS), .driver_data = APPLE_HAS_FN | APPLE_RDESC_JIS }, + { HID_USB_DEVICE(USB_VENDOR_ID_APPLE, USB_DEVICE_ID_APPLE_WELLSPRING9_ANSI), + .driver_data = APPLE_HAS_FN }, + { HID_USB_DEVICE(USB_VENDOR_ID_APPLE, USB_DEVICE_ID_APPLE_WELLSPRING9_ISO), + .driver_data = APPLE_HAS_FN | APPLE_ISO_KEYBOARD }, + { HID_USB_DEVICE(USB_VENDOR_ID_APPLE, USB_DEVICE_ID_APPLE_WELLSPRING9_JIS), + .driver_data = APPLE_HAS_FN | APPLE_RDESC_JIS }, { HID_BLUETOOTH_DEVICE(USB_VENDOR_ID_APPLE, USB_DEVICE_ID_APPLE_ALU_WIRELESS_2009_ANSI), .driver_data = APPLE_NUMLOCK_EMULATION | APPLE_HAS_FN }, { HID_BLUETOOTH_DEVICE(USB_VENDOR_ID_APPLE, USB_DEVICE_ID_APPLE_ALU_WIRELESS_2009_ISO), diff --git a/drivers/hid/hid-core.c b/drivers/hid/hid-core.c index 56ce8c2..c57c007 100644 --- a/drivers/hid/hid-core.c +++ b/drivers/hid/hid-core.c @@ -1760,6 +1760,9 @@ static const struct hid_device_id hid_have_special_driver[] = { { HID_USB_DEVICE(USB_VENDOR_ID_APPLE, USB_DEVICE_ID_APPLE_WELLSPRING8_ANSI) }, { HID_USB_DEVICE(USB_VENDOR_ID_APPLE, USB_DEVICE_ID_APPLE_WELLSPRING8_ISO) }, { HID_USB_DEVICE(USB_VENDOR_ID_APPLE, USB_DEVICE_ID_APPLE_WELLSPRING8_JIS) }, + { HID_USB_DEVICE(USB_VENDOR_ID_APPLE, USB_DEVICE_ID_APPLE_WELLSPRING9_ANSI) }, + { HID_USB_DEVICE(USB_VENDOR_ID_APPLE, USB_DEVICE_ID_APPLE_WELLSPRING9_ISO) }, + { HID_USB_DEVICE(USB_VENDOR_ID_APPLE, USB_DEVICE_ID_APPLE_WELLSPRING9_JIS) }, { HID_BLUETOOTH_DEVICE(USB_VENDOR_ID_APPLE, USB_DEVICE_ID_APPLE_ALU_WIRELESS_2009_ANSI) }, { HID_BLUETOOTH_DEVICE(USB_VENDOR_ID_APPLE, USB_DEVICE_ID_APPLE_ALU_WIRELESS_2009_ISO) }, { HID_BLUETOOTH_DEVICE(USB_VENDOR_ID_APPLE, USB_DEVICE_ID_APPLE_ALU_WIRELESS_2009_JIS) }, @@ -2450,6 +2453,9 @@ static const struct hid_device_id hid_mouse_ignore_list[] = { { HID_USB_DEVICE(USB_VENDOR_ID_APPLE, USB_DEVICE_ID_APPLE_WELLSPRING8_ANSI) }, { HID_USB_DEVICE(USB_VENDOR_ID_APPLE, USB_DEVICE_ID_APPLE_WELLSPRING8_ISO) }, { HID_USB_DEVICE(USB_VENDOR_ID_APPLE, USB_DEVICE_ID_APPLE_WELLSPRING8_JIS) }, + { HID_USB_DEVICE(USB_VENDOR_ID_APPLE, USB_DEVICE_ID_APPLE_WELLSPRING9_ANSI) }, + { HID_USB_DEVICE(USB_VENDOR_ID_APPLE, USB_DEVICE_ID_APPLE_WELLSPRING9_ISO) }, + { HID_USB_DEVICE(USB_VENDOR_ID_APPLE, USB_DEVICE_ID_APPLE_WELLSPRING9_JIS) }, { HID_USB_DEVICE(USB_VENDOR_ID_APPLE, USB_DEVICE_ID_APPLE_FOUNTAIN_TP_ONLY) }, { HID_USB_DEVICE(USB_VENDOR_ID_APPLE, USB_DEVICE_ID_APPLE_GEYSER1_TP_ONLY) }, { } diff --git a/drivers/hid/hid-ids.h b/drivers/hid/hid-ids.h index 9c47867..5274e9d 100644 --- a/drivers/hid/hid-ids.h +++ b/drivers/hid/hid-ids.h @@ -142,6 +142,9 @@ #define USB_DEVICE_ID_APPLE_WELLSPRING8_ANSI 0x0290 #define USB_DEVICE_ID_APPLE_WELLSPRING8_ISO 0x0291 #define USB_DEVICE_ID_APPLE_WELLSPRING8_JIS 0x0292 +#define USB_DEVICE_ID_APPLE_WELLSPRING9_ANSI 0x0272 +#define USB_DEVICE_ID_APPLE_WELLSPRING9_ISO 0x0273 +#define USB_DEVICE_ID_APPLE_WELLSPRING9_JIS 0x0274 #define USB_DEVICE_ID_APPLE_FOUNTAIN_TP_ONLY 0x030a #define USB_DEVICE_ID_APPLE_GEYSER1_TP_ONLY 0x030b #define USB_DEVICE_ID_APPLE_IRCONTROL 0x8240 -- cgit v0.10.2 From d58069265c9d15c04c9e3832cd1d9dffe9d4d5f6 Mon Sep 17 00:00:00 2001 From: John Horan Date: Fri, 24 Jul 2015 14:45:33 -0700 Subject: Input: bcm5974 - add support for the 2015 Macbook Pro Add support for the MacBookPro12,1 model. This patch needs to be applied together with the accompanied HID patch, as usual. Tested-by: John Horan Tested-by: Jochen Radmacher Tested-by: Yang Hongyang Tested-by: Yen-Chin, Lee Tested-by: George Hilios Tested-by: Janez Urevc Signed-off-by: Henrik Rydberg Signed-off-by: Dmitry Torokhov diff --git a/drivers/input/mouse/bcm5974.c b/drivers/input/mouse/bcm5974.c index a596b9b..30e3442 100644 --- a/drivers/input/mouse/bcm5974.c +++ b/drivers/input/mouse/bcm5974.c @@ -2,6 +2,7 @@ * Apple USB BCM5974 (Macbook Air and Penryn Macbook Pro) multitouch driver * * Copyright (C) 2008 Henrik Rydberg (rydberg@euromail.se) + * Copyright (C) 2015 John Horan (knasher@gmail.com) * * The USB initialization and package decoding was made by * Scott Shawcroft as part of the touchd user-space driver project: @@ -91,6 +92,10 @@ #define USB_DEVICE_ID_APPLE_WELLSPRING8_ANSI 0x0290 #define USB_DEVICE_ID_APPLE_WELLSPRING8_ISO 0x0291 #define USB_DEVICE_ID_APPLE_WELLSPRING8_JIS 0x0292 +/* MacbookPro12,1 (2015) */ +#define USB_DEVICE_ID_APPLE_WELLSPRING9_ANSI 0x0272 +#define USB_DEVICE_ID_APPLE_WELLSPRING9_ISO 0x0273 +#define USB_DEVICE_ID_APPLE_WELLSPRING9_JIS 0x0274 #define BCM5974_DEVICE(prod) { \ .match_flags = (USB_DEVICE_ID_MATCH_DEVICE | \ @@ -152,6 +157,10 @@ static const struct usb_device_id bcm5974_table[] = { BCM5974_DEVICE(USB_DEVICE_ID_APPLE_WELLSPRING8_ANSI), BCM5974_DEVICE(USB_DEVICE_ID_APPLE_WELLSPRING8_ISO), BCM5974_DEVICE(USB_DEVICE_ID_APPLE_WELLSPRING8_JIS), + /* MacbookPro12,1 */ + BCM5974_DEVICE(USB_DEVICE_ID_APPLE_WELLSPRING9_ANSI), + BCM5974_DEVICE(USB_DEVICE_ID_APPLE_WELLSPRING9_ISO), + BCM5974_DEVICE(USB_DEVICE_ID_APPLE_WELLSPRING9_JIS), /* Terminating entry */ {} }; @@ -180,18 +189,21 @@ struct bt_data { enum tp_type { TYPE1, /* plain trackpad */ TYPE2, /* button integrated in trackpad */ - TYPE3 /* additional header fields since June 2013 */ + TYPE3, /* additional header fields since June 2013 */ + TYPE4 /* additional header field for pressure data */ }; /* trackpad finger data offsets, le16-aligned */ #define HEADER_TYPE1 (13 * sizeof(__le16)) #define HEADER_TYPE2 (15 * sizeof(__le16)) #define HEADER_TYPE3 (19 * sizeof(__le16)) +#define HEADER_TYPE4 (23 * sizeof(__le16)) /* trackpad button data offsets */ #define BUTTON_TYPE1 0 #define BUTTON_TYPE2 15 #define BUTTON_TYPE3 23 +#define BUTTON_TYPE4 31 /* list of device capability bits */ #define HAS_INTEGRATED_BUTTON 1 @@ -200,16 +212,19 @@ enum tp_type { #define FSIZE_TYPE1 (14 * sizeof(__le16)) #define FSIZE_TYPE2 (14 * sizeof(__le16)) #define FSIZE_TYPE3 (14 * sizeof(__le16)) +#define FSIZE_TYPE4 (15 * sizeof(__le16)) /* offset from header to finger struct */ #define DELTA_TYPE1 (0 * sizeof(__le16)) #define DELTA_TYPE2 (0 * sizeof(__le16)) #define DELTA_TYPE3 (0 * sizeof(__le16)) +#define DELTA_TYPE4 (1 * sizeof(__le16)) /* usb control message mode switch data */ #define USBMSG_TYPE1 8, 0x300, 0, 0, 0x1, 0x8 #define USBMSG_TYPE2 8, 0x300, 0, 0, 0x1, 0x8 #define USBMSG_TYPE3 8, 0x300, 0, 0, 0x1, 0x8 +#define USBMSG_TYPE4 2, 0x302, 2, 1, 0x1, 0x0 /* Wellspring initialization constants */ #define BCM5974_WELLSPRING_MODE_READ_REQUEST_ID 1 @@ -227,7 +242,8 @@ struct tp_finger { __le16 orientation; /* 16384 when point, else 15 bit angle */ __le16 touch_major; /* touch area, major axis */ __le16 touch_minor; /* touch area, minor axis */ - __le16 unused[3]; /* zeros */ + __le16 unused[2]; /* zeros */ + __le16 pressure; /* pressure on forcetouch touchpad */ __le16 multi; /* one finger: varies, more fingers: constant */ } __attribute__((packed,aligned(2))); @@ -468,6 +484,19 @@ static const struct bcm5974_config bcm5974_config_table[] = { { SN_COORD, -150, 6600 }, { SN_ORIENT, -MAX_FINGER_ORIENTATION, MAX_FINGER_ORIENTATION } }, + { + USB_DEVICE_ID_APPLE_WELLSPRING9_ANSI, + USB_DEVICE_ID_APPLE_WELLSPRING9_ISO, + USB_DEVICE_ID_APPLE_WELLSPRING9_JIS, + HAS_INTEGRATED_BUTTON, + 0, sizeof(struct bt_data), + 0x83, DATAFORMAT(TYPE4), + { SN_PRESSURE, 0, 300 }, + { SN_WIDTH, 0, 2048 }, + { SN_COORD, -4828, 5345 }, + { SN_COORD, -203, 6803 }, + { SN_ORIENT, -MAX_FINGER_ORIENTATION, MAX_FINGER_ORIENTATION } + }, {} }; -- cgit v0.10.2 From ca4da5dd1f99fe9c59f1709fb43e818b18ad20e0 Mon Sep 17 00:00:00 2001 From: Colin Ian King Date: Mon, 27 Jul 2015 15:23:43 +0100 Subject: KEYS: ensure we free the assoc array edit if edit is valid __key_link_end is not freeing the associated array edit structure and this leads to a 512 byte memory leak each time an identical existing key is added with add_key(). The reason the add_key() system call returns okay is that key_create_or_update() calls __key_link_begin() before checking to see whether it can update a key directly rather than adding/replacing - which it turns out it can. Thus __key_link() is not called through __key_instantiate_and_link() and __key_link_end() must cancel the edit. CVE-2015-1333 Signed-off-by: Colin Ian King Signed-off-by: David Howells Signed-off-by: James Morris diff --git a/security/keys/keyring.c b/security/keys/keyring.c index e72548b..d334370 100644 --- a/security/keys/keyring.c +++ b/security/keys/keyring.c @@ -1181,9 +1181,11 @@ void __key_link_end(struct key *keyring, if (index_key->type == &key_type_keyring) up_write(&keyring_serialise_link_sem); - if (edit && !edit->dead_leaf) { - key_payload_reserve(keyring, - keyring->datalen - KEYQUOTA_LINK_BYTES); + if (edit) { + if (!edit->dead_leaf) { + key_payload_reserve(keyring, + keyring->datalen - KEYQUOTA_LINK_BYTES); + } assoc_array_cancel_edit(edit); } up_write(&keyring->sem); -- cgit v0.10.2 From 0b991f5cdcd6201e5401f83ca3a672343c3bfc49 Mon Sep 17 00:00:00 2001 From: Heiko Carstens Date: Mon, 27 Jul 2015 09:53:49 +0200 Subject: s390/cachinfo: add missing facility check to init_cache_level() Stephen Powell reported the following crash on a z890 machine: Kernel BUG at 00000000001219d0 [verbose debug info unavailable] illegal operation: 0001 ilc:3 [#1] SMP Krnl PSW : 0704e00180000000 00000000001219d0 (init_cache_level+0x38/0xe0) R:0 T:1 IO:1 EX:1 Key:0 M:1 W:0 P:0 AS:3 CC:2 PM:0 EA:3 Krnl Code: 00000000001219c2: a7840056 brc 8,121a6e 00000000001219c6: a7190000 lghi %r1,0 #00000000001219ca: eb101000004c ecag %r1,%r0,0(%r1) >00000000001219d0: a7390000 lghi %r3,0 00000000001219d4: e310f0a00024 stg %r1,160(%r15) 00000000001219da: a7080000 lhi %r0,0 00000000001219de: a7b9f000 lghi %r11,-4096 00000000001219e2: c0a0002899d9 larl %r10,634d94 Call Trace: [<0000000000478ee2>] detect_cache_attributes+0x2a/0x2b8 [<000000000097c9b0>] cacheinfo_sysfs_init+0x60/0xc8 [<00000000001001c0>] do_one_initcall+0x98/0x1c8 [<000000000094fdc2>] kernel_init_freeable+0x212/0x2d8 [<000000000062352e>] kernel_init+0x26/0x118 [<000000000062fd2e>] kernel_thread_starter+0x6/0xc The illegal operation was executed because of a missing facility check, which should have made sure that the ECAG execution would only be executed on machines which have the general-instructions-extension facility installed. Reported-and-tested-by: Stephen Powell Cc: stable@vger.kernel.org # v4.0+ Signed-off-by: Heiko Carstens Signed-off-by: Martin Schwidefsky diff --git a/arch/s390/kernel/cache.c b/arch/s390/kernel/cache.c index bff5e3b..8ba3243 100644 --- a/arch/s390/kernel/cache.c +++ b/arch/s390/kernel/cache.c @@ -138,6 +138,8 @@ int init_cache_level(unsigned int cpu) union cache_topology ct; enum cache_type ctype; + if (!test_facility(34)) + return -EOPNOTSUPP; if (!this_cpu_ci) return -EINVAL; ct.raw = ecag(EXTRACT_TOPOLOGY, 0, 0); -- cgit v0.10.2 From 697bb728d9e2367020aa0c5af7363809d7658e43 Mon Sep 17 00:00:00 2001 From: Ben Skeggs Date: Tue, 28 Jul 2015 17:20:57 +1000 Subject: drm/nouveau/kms/nv50-: guard against enabling cursor on disabled heads Userspace has started doing this, which upsets the display class hw error checking in various unpleasant ways. Signed-off-by: Ben Skeggs diff --git a/drivers/gpu/drm/nouveau/nv50_display.c b/drivers/gpu/drm/nouveau/nv50_display.c index 7da7958..981342d 100644 --- a/drivers/gpu/drm/nouveau/nv50_display.c +++ b/drivers/gpu/drm/nouveau/nv50_display.c @@ -979,7 +979,7 @@ nv50_crtc_cursor_show_hide(struct nouveau_crtc *nv_crtc, bool show, bool update) { struct nv50_mast *mast = nv50_mast(nv_crtc->base.dev); - if (show && nv_crtc->cursor.nvbo) + if (show && nv_crtc->cursor.nvbo && nv_crtc->base.enabled) nv50_crtc_cursor_show(nv_crtc); else nv50_crtc_cursor_hide(nv_crtc); -- cgit v0.10.2 From eb48b12ee5e05dd3ea473d008873f5228a63eb7a Mon Sep 17 00:00:00 2001 From: Alexandre Courbot Date: Thu, 9 Jul 2015 17:15:14 +0900 Subject: drm/nouveau/nouveau/ttm: fix tiled system memory with Maxwell Add Maxwell to the switch statement that sets node->memtype, otherwise all tiling information is ignored for buffers in system memory. While we are at it, make that switch statement explicitly complain the next time we meet a non-handled card family. Signed-off-by: Alexandre Courbot Signed-off-by: Ben Skeggs diff --git a/drivers/gpu/drm/nouveau/nouveau_ttm.c b/drivers/gpu/drm/nouveau/nouveau_ttm.c index 18f4497..7464aef3 100644 --- a/drivers/gpu/drm/nouveau/nouveau_ttm.c +++ b/drivers/gpu/drm/nouveau/nouveau_ttm.c @@ -175,15 +175,24 @@ nouveau_gart_manager_new(struct ttm_mem_type_manager *man, node->page_shift = 12; switch (drm->device.info.family) { + case NV_DEVICE_INFO_V0_TNT: + case NV_DEVICE_INFO_V0_CELSIUS: + case NV_DEVICE_INFO_V0_KELVIN: + case NV_DEVICE_INFO_V0_RANKINE: + case NV_DEVICE_INFO_V0_CURIE: + break; case NV_DEVICE_INFO_V0_TESLA: if (drm->device.info.chipset != 0x50) node->memtype = (nvbo->tile_flags & 0x7f00) >> 8; break; case NV_DEVICE_INFO_V0_FERMI: case NV_DEVICE_INFO_V0_KEPLER: + case NV_DEVICE_INFO_V0_MAXWELL: node->memtype = (nvbo->tile_flags & 0xff00) >> 8; break; default: + NV_WARN(drm, "%s: unhandled family type %x\n", __func__, + drm->device.info.family); break; } -- cgit v0.10.2 From 3471648a7569512e10f154cdfe5076c341a5c099 Mon Sep 17 00:00:00 2001 From: Jeff Layton Date: Fri, 10 Jul 2015 15:58:42 -0400 Subject: nfs: plug memory leak when ->prepare_layoutcommit fails "data" is currently leaked when the prepare_layoutcommit operation returns an error. Put the cred before taking the spinlock in that case, take the lock and then goto out_unlock which will drop the lock and then free "data". Signed-off-by: Jeff Layton Signed-off-by: Trond Myklebust diff --git a/fs/nfs/pnfs.c b/fs/nfs/pnfs.c index 18aa3b7..70bf706 100644 --- a/fs/nfs/pnfs.c +++ b/fs/nfs/pnfs.c @@ -2221,13 +2221,12 @@ pnfs_layoutcommit_inode(struct inode *inode, bool sync) if (ld->prepare_layoutcommit) { status = ld->prepare_layoutcommit(&data->args); if (status) { + put_rpccred(data->cred); spin_lock(&inode->i_lock); set_bit(NFS_INO_LAYOUTCOMMIT, &nfsi->flags); if (end_pos > nfsi->layout->plh_lwb) nfsi->layout->plh_lwb = end_pos; - spin_unlock(&inode->i_lock); - put_rpccred(data->cred); - goto clear_layoutcommitting; + goto out_unlock; } } -- cgit v0.10.2 From a49c269111a5b3c1fd2a98f36fa27423b94549f8 Mon Sep 17 00:00:00 2001 From: Kinglong Mee Date: Mon, 27 Jul 2015 15:31:38 +0800 Subject: nfs: Fix an oops caused by using other thread's stack space in ASYNC mode An oops caused by using other thread's stack space in sunrpc ASYNC sending thread. [ 9839.007187] ------------[ cut here ]------------ [ 9839.007923] kernel BUG at fs/nfs/nfs4xdr.c:910! [ 9839.008069] invalid opcode: 0000 [#1] SMP [ 9839.008069] Modules linked in: blocklayoutdriver rpcsec_gss_krb5 nfsv4 dns_resolver nfs fscache snd_hda_codec_generic snd_hda_intel snd_hda_controller snd_hda_codec snd_hwdep snd_seq snd_seq_device snd_pcm joydev iosf_mbi crct10dif_pclmul snd_timer crc32_pclmul crc32c_intel ghash_clmulni_intel snd soundcore ppdev pvpanic parport_pc i2c_piix4 serio_raw virtio_balloon parport acpi_cpufreq nfsd nfs_acl lockd grace auth_rpcgss sunrpc qxl drm_kms_helper virtio_net virtio_console virtio_blk ttm drm virtio_pci virtio_ring virtio ata_generic pata_acpi [ 9839.008069] CPU: 0 PID: 308 Comm: kworker/0:1H Not tainted 4.0.0-0.rc4.git1.3.fc23.x86_64 #1 [ 9839.008069] Hardware name: Bochs Bochs, BIOS Bochs 01/01/2011 [ 9839.008069] Workqueue: rpciod rpc_async_schedule [sunrpc] [ 9839.008069] task: ffff8800d8b4d8e0 ti: ffff880036678000 task.ti: ffff880036678000 [ 9839.008069] RIP: 0010:[] [] reserve_space.part.73+0x9/0x10 [nfsv4] [ 9839.008069] RSP: 0018:ffff88003667ba58 EFLAGS: 00010246 [ 9839.008069] RAX: 0000000000000000 RBX: 000000001fc15e18 RCX: ffff8800c0193800 [ 9839.008069] RDX: ffff8800e4ae3f24 RSI: 000000001fc15e2c RDI: ffff88003667bcd0 [ 9839.008069] RBP: ffff88003667ba58 R08: ffff8800d9173008 R09: 0000000000000003 [ 9839.008069] R10: ffff88003667bcd0 R11: 000000000000000c R12: 0000000000010000 [ 9839.008069] R13: ffff8800d9173350 R14: 0000000000000000 R15: ffff8800c0067b98 [ 9839.008069] FS: 0000000000000000(0000) GS:ffff88011fc00000(0000) knlGS:0000000000000000 [ 9839.008069] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 [ 9839.008069] CR2: 00007f988c9c8bb0 CR3: 00000000d99b6000 CR4: 00000000000407f0 [ 9839.008069] Stack: [ 9839.008069] ffff88003667bbc8 ffffffffa03412c5 00000000c6c55680 ffff880000000003 [ 9839.008069] 0000000000000088 00000010c6c55680 0001000000000002 ffffffff816e87e9 [ 9839.008069] 0000000000000000 00000000477290e2 ffff88003667bab8 ffffffff81327ba3 [ 9839.008069] Call Trace: [ 9839.008069] [] encode_attrs+0x435/0x530 [nfsv4] [ 9839.008069] [] ? inet_sendmsg+0x69/0xb0 [ 9839.008069] [] ? selinux_socket_sendmsg+0x23/0x30 [ 9839.008069] [] ? do_sock_sendmsg+0x9f/0xc0 [ 9839.008069] [] ? kernel_sendmsg+0x58/0x70 [ 9839.008069] [] ? xdr_reserve_space+0x20/0x170 [sunrpc] [ 9839.008069] [] ? xdr_reserve_space+0x20/0x170 [sunrpc] [ 9839.008069] [] ? nfs4_xdr_enc_open_noattr+0x130/0x130 [nfsv4] [ 9839.008069] [] encode_open+0x2d5/0x340 [nfsv4] [ 9839.008069] [] ? nfs4_xdr_enc_open_noattr+0x130/0x130 [nfsv4] [ 9839.008069] [] ? xdr_encode_opaque+0x19/0x20 [sunrpc] [ 9839.008069] [] ? encode_string+0x2b/0x40 [nfsv4] [ 9839.008069] [] nfs4_xdr_enc_open+0xb3/0x140 [nfsv4] [ 9839.008069] [] rpcauth_wrap_req+0xac/0xf0 [sunrpc] [ 9839.008069] [] call_transmit+0x18b/0x2d0 [sunrpc] [ 9839.008069] [] ? call_decode+0x860/0x860 [sunrpc] [ 9839.008069] [] ? call_decode+0x860/0x860 [sunrpc] [ 9839.008069] [] __rpc_execute+0x90/0x460 [sunrpc] [ 9839.008069] [] rpc_async_schedule+0x15/0x20 [sunrpc] [ 9839.008069] [] process_one_work+0x1bb/0x410 [ 9839.008069] [] worker_thread+0x53/0x470 [ 9839.008069] [] ? process_one_work+0x410/0x410 [ 9839.008069] [] ? process_one_work+0x410/0x410 [ 9839.008069] [] kthread+0xd8/0xf0 [ 9839.008069] [] ? kthread_worker_fn+0x180/0x180 [ 9839.008069] [] ret_from_fork+0x58/0x90 [ 9839.008069] [] ? kthread_worker_fn+0x180/0x180 [ 9839.008069] Code: 00 00 48 c7 c7 21 fa 37 a0 e8 94 1c d6 e0 c6 05 d2 17 05 00 01 8b 03 eb d7 66 0f 1f 84 00 00 00 00 00 66 66 66 66 90 55 48 89 e5 <0f> 0b 0f 1f 44 00 00 66 66 66 66 90 55 48 89 e5 41 54 53 89 f3 [ 9839.008069] RIP [] reserve_space.part.73+0x9/0x10 [nfsv4] [ 9839.008069] RSP [ 9839.071114] ---[ end trace cc14c03adb522e94 ]--- Signed-off-by: Kinglong Mee Signed-off-by: Trond Myklebust diff --git a/fs/nfs/internal.h b/fs/nfs/internal.h index 9e6475b..7970138 100644 --- a/fs/nfs/internal.h +++ b/fs/nfs/internal.h @@ -296,6 +296,22 @@ extern struct rpc_procinfo nfs4_procedures[]; #ifdef CONFIG_NFS_V4_SECURITY_LABEL extern struct nfs4_label *nfs4_label_alloc(struct nfs_server *server, gfp_t flags); +static inline struct nfs4_label * +nfs4_label_copy(struct nfs4_label *dst, struct nfs4_label *src) +{ + if (!dst || !src) + return NULL; + + if (src->len > NFS4_MAXLABELLEN) + return NULL; + + dst->lfs = src->lfs; + dst->pi = src->pi; + dst->len = src->len; + memcpy(dst->label, src->label, src->len); + + return dst; +} static inline void nfs4_label_free(struct nfs4_label *label) { if (label) { @@ -316,6 +332,11 @@ static inline void nfs4_label_free(void *label) {} static inline void nfs_zap_label_cache_locked(struct nfs_inode *nfsi) { } +static inline struct nfs4_label * +nfs4_label_copy(struct nfs4_label *dst, struct nfs4_label *src) +{ + return NULL; +} #endif /* CONFIG_NFS_V4_SECURITY_LABEL */ /* proc.c */ diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c index c85ffe6..e94a964 100644 --- a/fs/nfs/nfs4proc.c +++ b/fs/nfs/nfs4proc.c @@ -912,6 +912,7 @@ struct nfs4_opendata { struct nfs_open_confirmres c_res; struct nfs4_string owner_name; struct nfs4_string group_name; + struct nfs4_label *a_label; struct nfs_fattr f_attr; struct nfs4_label *f_label; struct dentry *dir; @@ -1015,6 +1016,10 @@ static struct nfs4_opendata *nfs4_opendata_alloc(struct dentry *dentry, if (IS_ERR(p->f_label)) goto err_free_p; + p->a_label = nfs4_label_alloc(server, gfp_mask); + if (IS_ERR(p->a_label)) + goto err_free_f; + alloc_seqid = server->nfs_client->cl_mvops->alloc_seqid; p->o_arg.seqid = alloc_seqid(&sp->so_seqid, gfp_mask); if (IS_ERR(p->o_arg.seqid)) @@ -1043,7 +1048,7 @@ static struct nfs4_opendata *nfs4_opendata_alloc(struct dentry *dentry, p->o_arg.server = server; p->o_arg.bitmask = nfs4_bitmask(server, label); p->o_arg.open_bitmap = &nfs4_fattr_bitmap[0]; - p->o_arg.label = label; + p->o_arg.label = nfs4_label_copy(p->a_label, label); p->o_arg.claim = nfs4_map_atomic_open_claim(server, claim); switch (p->o_arg.claim) { case NFS4_OPEN_CLAIM_NULL: @@ -1076,6 +1081,8 @@ static struct nfs4_opendata *nfs4_opendata_alloc(struct dentry *dentry, return p; err_free_label: + nfs4_label_free(p->a_label); +err_free_f: nfs4_label_free(p->f_label); err_free_p: kfree(p); @@ -1095,6 +1102,7 @@ static void nfs4_opendata_free(struct kref *kref) nfs4_put_open_state(p->state); nfs4_put_state_owner(p->owner); + nfs4_label_free(p->a_label); nfs4_label_free(p->f_label); dput(p->dir); -- cgit v0.10.2 From 71b65445f0ed04c2afe3660f829779fddb2890c1 Mon Sep 17 00:00:00 2001 From: Mika Westerberg Date: Tue, 28 Jul 2015 13:51:21 +0300 Subject: ACPI / PM: Use target_state to set the device power state Commit 20dacb71ad28 ("ACPI / PM: Rework device power management to follow ACPI 6") changed the device power management to use D3hot if the device in question does not have _PR3 method even if D3cold was requested by the caller. However, if the device has _PR3 device->power.state is also set to D3hot instead of D3Cold after power resources have been turned off because device->power.state will be assigned from "state" instead of "target_state". Next time the device is transitioned to D0, acpi_power_transition() will find that the current power state of the device is D3hot instead of D3cold which causes it to power down all resources required for the current (wrong) state D3hot. Below is a simplified ASL example of a real touch panel device which triggers the problem: Scope (TPL1) { Name (_PR0, Package (1) { \_SB.PCI0.I2C1.PXTC }) Name (_PR3, Package (1) { \_SB.PCI0.I2C1.PXTC }) ... } In both D0 and D3hot the same power resource is required. However, when acpi_power_transition() turns off power resources required for D3hot (as the device is transitioned to D0) it powers down PXTC which then makes the device to lose its power. Fix this by assigning "target_state" to the device power state instead of "state" that is always D3hot even for devices with valid _PR3. Fixes: 20dacb71ad28 (ACPI / PM: Rework device power management to follow ACPI 6) Signed-off-by: Mika Westerberg Signed-off-by: Rafael J. Wysocki diff --git a/drivers/acpi/device_pm.c b/drivers/acpi/device_pm.c index 717afcd..88dbbb1 100644 --- a/drivers/acpi/device_pm.c +++ b/drivers/acpi/device_pm.c @@ -231,7 +231,7 @@ int acpi_device_set_power(struct acpi_device *device, int state) dev_warn(&device->dev, "Failed to change power state to %s\n", acpi_power_state_string(state)); } else { - device->power.state = state; + device->power.state = target_state; ACPI_DEBUG_PRINT((ACPI_DB_INFO, "Device [%s] transitioned to %s\n", device->pnp.bus_id, -- cgit v0.10.2 From 559ed40752dc63e68f9b9ad301b20e6a3fe5cf21 Mon Sep 17 00:00:00 2001 From: "Rafael J. Wysocki" Date: Sun, 26 Jul 2015 02:07:47 +0200 Subject: cpufreq: Avoid attempts to create duplicate symbolic links After commit 87549141d516 (cpufreq: Stop migrating sysfs files on hotplug) there is a problem with CPUs that share cpufreq policy objects with other CPUs and are initially offline. Say CPU1 shares a policy with CPU0 which is online and is registered first. As part of the registration process, cpufreq_add_dev() is called for it. It creates the policy object and a symbolic link to it from the CPU1's sysfs directory. If CPU1 is registered subsequently and it is offline at that time, cpufreq_add_dev() will attempt to create a symbolic link to the policy object for it, but that link is present already, so a warning about that will be triggered. To avoid that warning, make cpufreq use an additional CPU mask containing related CPUs that are actually present for each policy object. That mask is initialized when the policy object is populated after its creation (for the first online CPU using it) and it includes CPUs from the "policy CPUs" mask returned by the cpufreq driver's ->init() callback that are physically present at that time. Symbolic links to the policy are created only for the CPUs in that mask. If cpufreq_add_dev() is invoked for an offline CPU, it checks the new mask and only creates the symlink if the CPU was not in it (the CPU is added to the mask at the same time). In turn, cpufreq_remove_dev() drops the given CPU from the new mask, removes its symlink to the policy object and returns, unless it is the CPU owning the policy object. In that case, the policy object is moved to a new CPU's sysfs directory or deleted if the CPU being removed was the last user of the policy. While at it, notice that cpufreq_remove_dev() can't fail, because its return value is ignored, so make it ignore return values from __cpufreq_remove_dev_prepare() and __cpufreq_remove_dev_finish() and prevent these functions from aborting on errors returned by __cpufreq_governor(). Also drop the now unused sif argument from them. Fixes: 87549141d516 (cpufreq: Stop migrating sysfs files on hotplug) Signed-off-by: Rafael J. Wysocki Reported-and-tested-by: Russell King Acked-by: Viresh Kumar diff --git a/drivers/cpufreq/cpufreq.c b/drivers/cpufreq/cpufreq.c index 26063af..7a3c30c 100644 --- a/drivers/cpufreq/cpufreq.c +++ b/drivers/cpufreq/cpufreq.c @@ -1002,7 +1002,7 @@ static int cpufreq_add_dev_symlink(struct cpufreq_policy *policy) int ret = 0; /* Some related CPUs might not be present (physically hotplugged) */ - for_each_cpu_and(j, policy->related_cpus, cpu_present_mask) { + for_each_cpu(j, policy->real_cpus) { if (j == policy->kobj_cpu) continue; @@ -1019,7 +1019,7 @@ static void cpufreq_remove_dev_symlink(struct cpufreq_policy *policy) unsigned int j; /* Some related CPUs might not be present (physically hotplugged) */ - for_each_cpu_and(j, policy->related_cpus, cpu_present_mask) { + for_each_cpu(j, policy->real_cpus) { if (j == policy->kobj_cpu) continue; @@ -1163,11 +1163,14 @@ static struct cpufreq_policy *cpufreq_policy_alloc(struct device *dev) if (!zalloc_cpumask_var(&policy->related_cpus, GFP_KERNEL)) goto err_free_cpumask; + if (!zalloc_cpumask_var(&policy->real_cpus, GFP_KERNEL)) + goto err_free_rcpumask; + ret = kobject_init_and_add(&policy->kobj, &ktype_cpufreq, &dev->kobj, "cpufreq"); if (ret) { pr_err("%s: failed to init policy->kobj: %d\n", __func__, ret); - goto err_free_rcpumask; + goto err_free_real_cpus; } INIT_LIST_HEAD(&policy->policy_list); @@ -1184,6 +1187,8 @@ static struct cpufreq_policy *cpufreq_policy_alloc(struct device *dev) return policy; +err_free_real_cpus: + free_cpumask_var(policy->real_cpus); err_free_rcpumask: free_cpumask_var(policy->related_cpus); err_free_cpumask: @@ -1234,6 +1239,7 @@ static void cpufreq_policy_free(struct cpufreq_policy *policy, bool notify) write_unlock_irqrestore(&cpufreq_driver_lock, flags); cpufreq_policy_put_kobj(policy, notify); + free_cpumask_var(policy->real_cpus); free_cpumask_var(policy->related_cpus); free_cpumask_var(policy->cpus); kfree(policy); @@ -1258,14 +1264,17 @@ static int cpufreq_add_dev(struct device *dev, struct subsys_interface *sif) pr_debug("adding CPU %u\n", cpu); - /* - * Only possible if 'cpu' wasn't physically present earlier and we are - * here from subsys_interface add callback. A hotplug notifier will - * follow and we will handle it like logical CPU hotplug then. For now, - * just create the sysfs link. - */ - if (cpu_is_offline(cpu)) - return add_cpu_dev_symlink(per_cpu(cpufreq_cpu_data, cpu), cpu); + if (cpu_is_offline(cpu)) { + /* + * Only possible if we are here from the subsys_interface add + * callback. A hotplug notifier will follow and we will handle + * it as CPU online then. For now, just create the sysfs link, + * unless there is no policy or the link is already present. + */ + policy = per_cpu(cpufreq_cpu_data, cpu); + return policy && !cpumask_test_and_set_cpu(cpu, policy->real_cpus) + ? add_cpu_dev_symlink(policy, cpu) : 0; + } if (!down_read_trylock(&cpufreq_rwsem)) return 0; @@ -1307,6 +1316,10 @@ static int cpufreq_add_dev(struct device *dev, struct subsys_interface *sif) /* related cpus should atleast have policy->cpus */ cpumask_or(policy->related_cpus, policy->related_cpus, policy->cpus); + /* Remember which CPUs have been present at the policy creation time. */ + if (!recover_policy) + cpumask_and(policy->real_cpus, policy->cpus, cpu_present_mask); + /* * affected cpus must always be the one, which are online. We aren't * managing offline cpus here. @@ -1420,8 +1433,7 @@ nomem_out: return ret; } -static int __cpufreq_remove_dev_prepare(struct device *dev, - struct subsys_interface *sif) +static int __cpufreq_remove_dev_prepare(struct device *dev) { unsigned int cpu = dev->id; int ret = 0; @@ -1437,10 +1449,8 @@ static int __cpufreq_remove_dev_prepare(struct device *dev, if (has_target()) { ret = __cpufreq_governor(policy, CPUFREQ_GOV_STOP); - if (ret) { + if (ret) pr_err("%s: Failed to stop governor\n", __func__); - return ret; - } } down_write(&policy->rwsem); @@ -1473,8 +1483,7 @@ static int __cpufreq_remove_dev_prepare(struct device *dev, return ret; } -static int __cpufreq_remove_dev_finish(struct device *dev, - struct subsys_interface *sif) +static int __cpufreq_remove_dev_finish(struct device *dev) { unsigned int cpu = dev->id; int ret; @@ -1492,10 +1501,8 @@ static int __cpufreq_remove_dev_finish(struct device *dev, /* If cpu is last user of policy, free policy */ if (has_target()) { ret = __cpufreq_governor(policy, CPUFREQ_GOV_POLICY_EXIT); - if (ret) { + if (ret) pr_err("%s: Failed to exit governor\n", __func__); - return ret; - } } /* @@ -1506,10 +1513,6 @@ static int __cpufreq_remove_dev_finish(struct device *dev, if (cpufreq_driver->exit) cpufreq_driver->exit(policy); - /* Free the policy only if the driver is getting removed. */ - if (sif) - cpufreq_policy_free(policy, true); - return 0; } @@ -1521,42 +1524,41 @@ static int __cpufreq_remove_dev_finish(struct device *dev, static int cpufreq_remove_dev(struct device *dev, struct subsys_interface *sif) { unsigned int cpu = dev->id; - int ret; - - /* - * Only possible if 'cpu' is getting physically removed now. A hotplug - * notifier should have already been called and we just need to remove - * link or free policy here. - */ - if (cpu_is_offline(cpu)) { - struct cpufreq_policy *policy = per_cpu(cpufreq_cpu_data, cpu); - struct cpumask mask; + struct cpufreq_policy *policy = per_cpu(cpufreq_cpu_data, cpu); - if (!policy) - return 0; + if (!policy) + return 0; - cpumask_copy(&mask, policy->related_cpus); - cpumask_clear_cpu(cpu, &mask); + if (cpu_online(cpu)) { + __cpufreq_remove_dev_prepare(dev); + __cpufreq_remove_dev_finish(dev); + } - /* - * Free policy only if all policy->related_cpus are removed - * physically. - */ - if (cpumask_intersects(&mask, cpu_present_mask)) { - remove_cpu_dev_symlink(policy, cpu); - return 0; - } + cpumask_clear_cpu(cpu, policy->real_cpus); + if (cpumask_empty(policy->real_cpus)) { cpufreq_policy_free(policy, true); return 0; } - ret = __cpufreq_remove_dev_prepare(dev, sif); + if (cpu != policy->kobj_cpu) { + remove_cpu_dev_symlink(policy, cpu); + } else { + /* + * The CPU owning the policy object is going away. Move it to + * another suitable CPU. + */ + unsigned int new_cpu = cpumask_first(policy->real_cpus); + struct device *new_dev = get_cpu_device(new_cpu); + + dev_dbg(dev, "%s: Moving policy object to CPU%u\n", __func__, new_cpu); - if (!ret) - ret = __cpufreq_remove_dev_finish(dev, sif); + sysfs_remove_link(&new_dev->kobj, "cpufreq"); + policy->kobj_cpu = new_cpu; + WARN_ON(kobject_move(&policy->kobj, &new_dev->kobj)); + } - return ret; + return 0; } static void handle_update(struct work_struct *work) @@ -2395,11 +2397,11 @@ static int cpufreq_cpu_callback(struct notifier_block *nfb, break; case CPU_DOWN_PREPARE: - __cpufreq_remove_dev_prepare(dev, NULL); + __cpufreq_remove_dev_prepare(dev); break; case CPU_POST_DEAD: - __cpufreq_remove_dev_finish(dev, NULL); + __cpufreq_remove_dev_finish(dev); break; case CPU_DOWN_FAILED: diff --git a/include/linux/cpufreq.h b/include/linux/cpufreq.h index 29ad97c..bde1e56 100644 --- a/include/linux/cpufreq.h +++ b/include/linux/cpufreq.h @@ -62,6 +62,7 @@ struct cpufreq_policy { /* CPUs sharing clock, require sw coordination */ cpumask_var_t cpus; /* Online CPUs only */ cpumask_var_t related_cpus; /* Online + Offline CPUs */ + cpumask_var_t real_cpus; /* Related and present */ unsigned int shared_type; /* ACPI: ANY or ALL affected CPUs should set cpufreq */ -- cgit v0.10.2 From 3213afb8bf5cf8d8c68a2c2376bf1dda52afae5d Mon Sep 17 00:00:00 2001 From: Dmitry Torokhov Date: Tue, 28 Jul 2015 10:25:03 -0700 Subject: Revert "Input: zforce - don't overwrite the stack" This reverts commit 7d01cd261c76f95913c81554a751968a1d282d3a because with given FRAME_MAXSIZE of 257 the check will never trigger and it causes warnings from GCC (with -Wtype-limits). Also the check was incorrect as it was not accounting for the already read 2 bytes of data stored in the buffer. diff --git a/drivers/input/touchscreen/zforce_ts.c b/drivers/input/touchscreen/zforce_ts.c index 80285c7..f58a196 100644 --- a/drivers/input/touchscreen/zforce_ts.c +++ b/drivers/input/touchscreen/zforce_ts.c @@ -429,7 +429,7 @@ static int zforce_read_packet(struct zforce_ts *ts, u8 *buf) goto unlock; } - if (buf[PAYLOAD_LENGTH] == 0 || buf[PAYLOAD_LENGTH] > FRAME_MAXSIZE) { + if (buf[PAYLOAD_LENGTH] == 0) { dev_err(&client->dev, "invalid payload length: %d\n", buf[PAYLOAD_LENGTH]); ret = -EIO; -- cgit v0.10.2 From b2442c5a7fe92cca08437070c8a45a7aa0d1703e Mon Sep 17 00:00:00 2001 From: Dave Chinner Date: Wed, 29 Jul 2015 11:48:00 +1000 Subject: xfs: call dax_fault on read page faults for DAX When modifying the patch series to handle the XFS MMAP_LOCK nesting of page faults, I botched the conversion of the read page fault path, and so it is only every calling through the page cache. Re-add the necessary __dax_fault() call for such files. Because the get_blocks callback on read faults may not set up the mapping buffer correctly to allow unwritten extent completion to be run, we need to allow callers of __dax_fault() to pass a null complete_unwritten() callback. The DAX code always zeros the unwritten page when it is read faulted so there are no stale data exposure issues with not doing the conversion. The only downside will be the potential for increased CPU overhead on repeated read faults of the same page. If this proves to be a problem, then the filesystem needs to fix it's get_block callback and provide a convert_unwritten() callback to the read fault path. Signed-off-by: Dave Chinner Reviewed-by: Matthew Wilcox Reviewed-by: Brian Foster Signed-off-by: Dave Chinner diff --git a/fs/dax.c b/fs/dax.c index c3e21cc..a7f77e1 100644 --- a/fs/dax.c +++ b/fs/dax.c @@ -319,6 +319,12 @@ static int dax_insert_mapping(struct inode *inode, struct buffer_head *bh, * @vma: The virtual memory area where the fault occurred * @vmf: The description of the fault * @get_block: The filesystem method used to translate file offsets to blocks + * @complete_unwritten: The filesystem method used to convert unwritten blocks + * to written so the data written to them is exposed. This is required for + * required by write faults for filesystems that will return unwritten + * extent mappings from @get_block, but it is optional for reads as + * dax_insert_mapping() will always zero unwritten blocks. If the fs does + * not support unwritten extents, the it should pass NULL. * * When a page fault occurs, filesystems may call this helper in their * fault handler for DAX files. __dax_fault() assumes the caller has done all @@ -437,8 +443,12 @@ int __dax_fault(struct vm_area_struct *vma, struct vm_fault *vmf, * as for normal BH based IO completions. */ error = dax_insert_mapping(inode, &bh, vma, vmf); - if (buffer_unwritten(&bh)) - complete_unwritten(&bh, !error); + if (buffer_unwritten(&bh)) { + if (complete_unwritten) + complete_unwritten(&bh, !error); + else + WARN_ON_ONCE(!(vmf->flags & FAULT_FLAG_WRITE)); + } out: if (error == -ENOMEM) diff --git a/fs/xfs/xfs_file.c b/fs/xfs/xfs_file.c index f0e8249..db4acc1 100644 --- a/fs/xfs/xfs_file.c +++ b/fs/xfs/xfs_file.c @@ -1514,18 +1514,27 @@ xfs_filemap_fault( struct vm_area_struct *vma, struct vm_fault *vmf) { - struct xfs_inode *ip = XFS_I(file_inode(vma->vm_file)); + struct inode *inode = file_inode(vma->vm_file); int ret; - trace_xfs_filemap_fault(ip); + trace_xfs_filemap_fault(XFS_I(inode)); /* DAX can shortcut the normal fault path on write faults! */ - if ((vmf->flags & FAULT_FLAG_WRITE) && IS_DAX(VFS_I(ip))) + if ((vmf->flags & FAULT_FLAG_WRITE) && IS_DAX(inode)) return xfs_filemap_page_mkwrite(vma, vmf); - xfs_ilock(ip, XFS_MMAPLOCK_SHARED); - ret = filemap_fault(vma, vmf); - xfs_iunlock(ip, XFS_MMAPLOCK_SHARED); + xfs_ilock(XFS_I(inode), XFS_MMAPLOCK_SHARED); + if (IS_DAX(inode)) { + /* + * we do not want to trigger unwritten extent conversion on read + * faults - that is unnecessary overhead and would also require + * changes to xfs_get_blocks_direct() to map unwritten extent + * ioend for conversion on read-only mappings. + */ + ret = __dax_fault(vma, vmf, xfs_get_blocks_direct, NULL); + } else + ret = filemap_fault(vma, vmf); + xfs_iunlock(XFS_I(inode), XFS_MMAPLOCK_SHARED); return ret; } -- cgit v0.10.2 From e3c32ee9e3e747fec01eb38e6610a9157d44c3ea Mon Sep 17 00:00:00 2001 From: Dave Chinner Date: Wed, 29 Jul 2015 11:48:01 +1000 Subject: xfs: remote attribute headers contain an invalid LSN In recent testing, a system that crashed failed log recovery on restart with a bad symlink buffer magic number: XFS (vda): Starting recovery (logdev: internal) XFS (vda): Bad symlink block magic! XFS: Assertion failed: 0, file: fs/xfs/xfs_log_recover.c, line: 2060 On examination of the log via xfs_logprint, none of the symlink buffers in the log had a bad magic number, nor were any other types of buffer log format headers mis-identified as symlink buffers. Tracing was used to find the buffer the kernel was tripping over, and xfs_db identified it's contents as: 000: 5841524d 00000000 00000346 64d82b48 8983e692 d71e4680 a5f49e2c b317576e 020: 00000000 00602038 00000000 006034ce d0020000 00000000 4d4d4d4d 4d4d4d4d 040: 4d4d4d4d 4d4d4d4d 4d4d4d4d 4d4d4d4d 4d4d4d4d 4d4d4d4d 4d4d4d4d 4d4d4d4d 060: 4d4d4d4d 4d4d4d4d 4d4d4d4d 4d4d4d4d 4d4d4d4d 4d4d4d4d 4d4d4d4d 4d4d4d4d ..... This is a remote attribute buffer, which are notable in that they are not logged but are instead written synchronously by the remote attribute code so that they exist on disk before the attribute transactions are committed to the journal. The above remote attribute block has an invalid LSN in it - cycle 0xd002000, block 0 - which means when log recovery comes along to determine if the transaction that writes to the underlying block should be replayed, it sees a block that has a future LSN and so does not replay the buffer data in the transaction. Instead, it validates the buffer magic number and attaches the buffer verifier to it. It is this buffer magic number check that is failing in the above assert, indicating that we skipped replay due to the LSN of the underlying buffer. The problem here is that the remote attribute buffers cannot have a valid LSN placed into them, because the transaction that contains the attribute tree pointer changes and the block allocation that the attribute data is being written to hasn't yet been committed. Hence the LSN field in the attribute block is completely unwritten, thereby leaving the underlying contents of the block in the LSN field. It could have any value, and hence a future overwrite of the block by log recovery may or may not work correctly. Fix this by always writing an invalid LSN to the remote attribute block, as any buffer in log recovery that needs to write over the remote attribute should occur. We are protected from having old data written over the attribute by the fact that freeing the block before the remote attribute is written will result in the buffer being marked stale in the log and so all changes prior to the buffer stale transaction will be cancelled by log recovery. Hence it is safe to ignore the LSN in the case or synchronously written, unlogged metadata such as remote attribute blocks, and to ensure we do that correctly, we need to write an invalid LSN to all remote attribute blocks to trigger immediate recovery of metadata that is written over the top. As a further protection for filesystems that may already have remote attribute blocks with bad LSNs on disk, change the log recovery code to always trigger immediate recovery of metadata over remote attribute blocks. cc: Signed-off-by: Dave Chinner Reviewed-by: Brian Foster Signed-off-by: Dave Chinner diff --git a/fs/xfs/libxfs/xfs_attr_remote.c b/fs/xfs/libxfs/xfs_attr_remote.c index 20de88d..2faec26 100644 --- a/fs/xfs/libxfs/xfs_attr_remote.c +++ b/fs/xfs/libxfs/xfs_attr_remote.c @@ -159,11 +159,10 @@ xfs_attr3_rmt_write_verify( struct xfs_buf *bp) { struct xfs_mount *mp = bp->b_target->bt_mount; - struct xfs_buf_log_item *bip = bp->b_fspriv; + int blksize = mp->m_attr_geo->blksize; char *ptr; int len; xfs_daddr_t bno; - int blksize = mp->m_attr_geo->blksize; /* no verification of non-crc buffers */ if (!xfs_sb_version_hascrc(&mp->m_sb)) @@ -175,16 +174,22 @@ xfs_attr3_rmt_write_verify( ASSERT(len >= blksize); while (len > 0) { + struct xfs_attr3_rmt_hdr *rmt = (struct xfs_attr3_rmt_hdr *)ptr; + if (!xfs_attr3_rmt_verify(mp, ptr, blksize, bno)) { xfs_buf_ioerror(bp, -EFSCORRUPTED); xfs_verifier_error(bp); return; } - if (bip) { - struct xfs_attr3_rmt_hdr *rmt; - rmt = (struct xfs_attr3_rmt_hdr *)ptr; - rmt->rm_lsn = cpu_to_be64(bip->bli_item.li_lsn); + /* + * Ensure we aren't writing bogus LSNs to disk. See + * xfs_attr3_rmt_hdr_set() for the explanation. + */ + if (rmt->rm_lsn != cpu_to_be64(NULLCOMMITLSN)) { + xfs_buf_ioerror(bp, -EFSCORRUPTED); + xfs_verifier_error(bp); + return; } xfs_update_cksum(ptr, blksize, XFS_ATTR3_RMT_CRC_OFF); @@ -221,6 +226,18 @@ xfs_attr3_rmt_hdr_set( rmt->rm_owner = cpu_to_be64(ino); rmt->rm_blkno = cpu_to_be64(bno); + /* + * Remote attribute blocks are written synchronously, so we don't + * have an LSN that we can stamp in them that makes any sense to log + * recovery. To ensure that log recovery handles overwrites of these + * blocks sanely (i.e. once they've been freed and reallocated as some + * other type of metadata) we need to ensure that the LSN has a value + * that tells log recovery to ignore the LSN and overwrite the buffer + * with whatever is in it's log. To do this, we use the magic + * NULLCOMMITLSN to indicate that the LSN is invalid. + */ + rmt->rm_lsn = cpu_to_be64(NULLCOMMITLSN); + return sizeof(struct xfs_attr3_rmt_hdr); } diff --git a/fs/xfs/xfs_log_recover.c b/fs/xfs/xfs_log_recover.c index 01dd228..480ebba 100644 --- a/fs/xfs/xfs_log_recover.c +++ b/fs/xfs/xfs_log_recover.c @@ -1886,9 +1886,14 @@ xlog_recover_get_buf_lsn( uuid = &((struct xfs_dir3_blk_hdr *)blk)->uuid; break; case XFS_ATTR3_RMT_MAGIC: - lsn = be64_to_cpu(((struct xfs_attr3_rmt_hdr *)blk)->rm_lsn); - uuid = &((struct xfs_attr3_rmt_hdr *)blk)->rm_uuid; - break; + /* + * Remote attr blocks are written synchronously, rather than + * being logged. That means they do not contain a valid LSN + * (i.e. transactionally ordered) in them, and hence any time we + * see a buffer to replay over the top of a remote attribute + * block we should simply do so. + */ + goto recover_immediately; case XFS_SB_MAGIC: lsn = be64_to_cpu(((struct xfs_dsb *)blk)->sb_lsn); uuid = &((struct xfs_dsb *)blk)->sb_uuid; -- cgit v0.10.2 From df150ed102baa0e78c06e08e975dfb47147dd677 Mon Sep 17 00:00:00 2001 From: Dave Chinner Date: Wed, 29 Jul 2015 11:48:02 +1000 Subject: xfs: remote attributes need to be considered data We don't log remote attribute contents, and instead write them synchronously before we commit the block allocation and attribute tree update transaction. As a result we are writing to the allocated space before the allcoation has been made permanent. As a result, we cannot consider this allocation to be a metadata allocation. Metadata allocation can take blocks from the free list and so reuse them before the transaction that freed the block is committed to disk. This behaviour is perfectly fine for journalled metadata changes as log recovery will ensure the free operation is replayed before the overwrite, but for remote attribute writes this is not the case. Hence we have to consider the remote attribute blocks to contain data and allocate accordingly. We do this by dropping the XFS_BMAPI_METADATA flag from the block allocation. This means the allocation will not use blocks that are on the busy list without first ensuring that the freeing transaction has been committed to disk and the blocks removed from the busy list. This ensures we will never overwrite a freed block without first ensuring that it is really free. cc: Signed-off-by: Dave Chinner Reviewed-by: Brian Foster Signed-off-by: Dave Chinner diff --git a/fs/xfs/libxfs/xfs_attr_remote.c b/fs/xfs/libxfs/xfs_attr_remote.c index 2faec26..dd71403 100644 --- a/fs/xfs/libxfs/xfs_attr_remote.c +++ b/fs/xfs/libxfs/xfs_attr_remote.c @@ -451,14 +451,21 @@ xfs_attr_rmtval_set( /* * Allocate a single extent, up to the size of the value. + * + * Note that we have to consider this a data allocation as we + * write the remote attribute without logging the contents. + * Hence we must ensure that we aren't using blocks that are on + * the busy list so that we don't overwrite blocks which have + * recently been freed but their transactions are not yet + * committed to disk. If we overwrite the contents of a busy + * extent and then crash then the block may not contain the + * correct metadata after log recovery occurs. */ xfs_bmap_init(args->flist, args->firstblock); nmap = 1; error = xfs_bmapi_write(args->trans, dp, (xfs_fileoff_t)lblkno, - blkcnt, - XFS_BMAPI_ATTRFORK | XFS_BMAPI_METADATA, - args->firstblock, args->total, &map, &nmap, - args->flist); + blkcnt, XFS_BMAPI_ATTRFORK, args->firstblock, + args->total, &map, &nmap, args->flist); if (!error) { error = xfs_bmap_finish(&args->trans, args->flist, &committed); -- cgit v0.10.2 From 76b91c32dd86f624b5df038dcb68cddd5a18d355 Mon Sep 17 00:00:00 2001 From: Nikolay Aleksandrov Date: Thu, 23 Jul 2015 11:01:05 -0700 Subject: bridge: stp: when using userspace stp stop kernel hello and hold timers These should be handled only by the respective STP which is in control. They become problematic for devices with limited resources with many ports because the hold_timer is per port and fires each second and the hello timer fires each 2 seconds even though it's global. While in user-space STP mode these timers are completely unnecessary so it's better to keep them off. Also ensure that when the bridge is up these timers are started only when running with kernel STP. Signed-off-by: Satish Ashok Signed-off-by: Nikolay Aleksandrov Signed-off-by: David S. Miller diff --git a/net/bridge/br_stp.c b/net/bridge/br_stp.c index b4b6dab..ed74ffa 100644 --- a/net/bridge/br_stp.c +++ b/net/bridge/br_stp.c @@ -209,8 +209,9 @@ void br_transmit_config(struct net_bridge_port *p) br_send_config_bpdu(p, &bpdu); p->topology_change_ack = 0; p->config_pending = 0; - mod_timer(&p->hold_timer, - round_jiffies(jiffies + BR_HOLD_TIME)); + if (p->br->stp_enabled == BR_KERNEL_STP) + mod_timer(&p->hold_timer, + round_jiffies(jiffies + BR_HOLD_TIME)); } } diff --git a/net/bridge/br_stp_if.c b/net/bridge/br_stp_if.c index a2730e7..4ca449a 100644 --- a/net/bridge/br_stp_if.c +++ b/net/bridge/br_stp_if.c @@ -48,7 +48,8 @@ void br_stp_enable_bridge(struct net_bridge *br) struct net_bridge_port *p; spin_lock_bh(&br->lock); - mod_timer(&br->hello_timer, jiffies + br->hello_time); + if (br->stp_enabled == BR_KERNEL_STP) + mod_timer(&br->hello_timer, jiffies + br->hello_time); mod_timer(&br->gc_timer, jiffies + HZ/10); br_config_bpdu_generation(br); @@ -127,6 +128,7 @@ static void br_stp_start(struct net_bridge *br) int r; char *argv[] = { BR_STP_PROG, br->dev->name, "start", NULL }; char *envp[] = { NULL }; + struct net_bridge_port *p; r = call_usermodehelper(BR_STP_PROG, argv, envp, UMH_WAIT_PROC); @@ -140,6 +142,10 @@ static void br_stp_start(struct net_bridge *br) if (r == 0) { br->stp_enabled = BR_USER_STP; br_debug(br, "userspace STP started\n"); + /* Stop hello and hold timers */ + del_timer(&br->hello_timer); + list_for_each_entry(p, &br->port_list, list) + del_timer(&p->hold_timer); } else { br->stp_enabled = BR_KERNEL_STP; br_debug(br, "using kernel STP\n"); @@ -156,12 +162,17 @@ static void br_stp_stop(struct net_bridge *br) int r; char *argv[] = { BR_STP_PROG, br->dev->name, "stop", NULL }; char *envp[] = { NULL }; + struct net_bridge_port *p; if (br->stp_enabled == BR_USER_STP) { r = call_usermodehelper(BR_STP_PROG, argv, envp, UMH_WAIT_PROC); br_info(br, "userspace STP stopped, return code %d\n", r); /* To start timers on any ports left in blocking */ + mod_timer(&br->hello_timer, jiffies + br->hello_time); + list_for_each_entry(p, &br->port_list, list) + mod_timer(&p->hold_timer, + round_jiffies(jiffies + BR_HOLD_TIME)); spin_lock_bh(&br->lock); br_port_state_selection(br); spin_unlock_bh(&br->lock); diff --git a/net/bridge/br_stp_timer.c b/net/bridge/br_stp_timer.c index 7caf7fa..5f0f5af 100644 --- a/net/bridge/br_stp_timer.c +++ b/net/bridge/br_stp_timer.c @@ -40,7 +40,9 @@ static void br_hello_timer_expired(unsigned long arg) if (br->dev->flags & IFF_UP) { br_config_bpdu_generation(br); - mod_timer(&br->hello_timer, round_jiffies(jiffies + br->hello_time)); + if (br->stp_enabled != BR_USER_STP) + mod_timer(&br->hello_timer, + round_jiffies(jiffies + br->hello_time)); } spin_unlock(&br->lock); } -- cgit v0.10.2 From 865b804244f228e80fb62abe464296399253cce8 Mon Sep 17 00:00:00 2001 From: David Ward Date: Sun, 26 Jul 2015 12:18:58 -0400 Subject: net/ipv4: suppress NETDEV_UP notification on address lifetime update This notification causes the FIB to be updated, which is not needed because the address already exists, and more importantly it may undo intentional changes that were made to the FIB after the address was originally added. (As a point of comparison, when an address becomes deprecated because its preferred lifetime expired, a notification on this chain is not generated.) The motivation for this commit is fixing an incompatibility between DHCP clients which set and update the address lifetime according to the lease, and a commercial VPN client which replaces kernel routes in a way that outbound traffic is sent only through the tunnel (and disconnects if any further route changes are detected via netlink). Signed-off-by: David Ward Signed-off-by: David S. Miller diff --git a/net/ipv4/devinet.c b/net/ipv4/devinet.c index e813196..2d9cb17 100644 --- a/net/ipv4/devinet.c +++ b/net/ipv4/devinet.c @@ -882,7 +882,6 @@ static int inet_rtm_newaddr(struct sk_buff *skb, struct nlmsghdr *nlh) queue_delayed_work(system_power_efficient_wq, &check_lifetime_work, 0); rtmsg_ifa(RTM_NEWADDR, ifa, nlh, NETLINK_CB(skb).portid); - blocking_notifier_call_chain(&inetaddr_chain, NETDEV_UP, ifa); } return 0; } -- cgit v0.10.2 From 11c91ef98f37cd743098de26160fffd7f9bd40e1 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Mon, 27 Jul 2015 11:33:50 +0200 Subject: arp: filter NOARP neighbours for SIOCGARP When arp is off on a device, and ioctl(SIOCGARP) is queried, a buggy answer is given with MAC address of the device, instead of the mac address of the destination/gateway. We filter out NUD_NOARP neighbours for /proc/net/arp, we must do the same for SIOCGARP ioctl. Tested: lpaa23:~# ./arp 10.246.7.190 MAC=00:01:e8:22:cb:1d // correct answer lpaa23:~# ip link set dev eth0 arp off lpaa23:~# cat /proc/net/arp # check arp table is now 'empty' IP address HW type Flags HW address Mask Device lpaa23:~# ./arp 10.246.7.190 MAC=00:1a:11:c3:0d:7f // buggy answer before patch (this is eth0 mac) After patch : lpaa23:~# ip link set dev eth0 arp off lpaa23:~# ./arp 10.246.7.190 ioctl(SIOCGARP) failed: No such device or address Signed-off-by: Eric Dumazet Reported-by: Vytautas Valancius Cc: Willem de Bruijn Signed-off-by: David S. Miller diff --git a/net/ipv4/arp.c b/net/ipv4/arp.c index 933a928..6c8b1fb 100644 --- a/net/ipv4/arp.c +++ b/net/ipv4/arp.c @@ -1017,14 +1017,16 @@ static int arp_req_get(struct arpreq *r, struct net_device *dev) neigh = neigh_lookup(&arp_tbl, &ip, dev); if (neigh) { - read_lock_bh(&neigh->lock); - memcpy(r->arp_ha.sa_data, neigh->ha, dev->addr_len); - r->arp_flags = arp_state_to_flags(neigh); - read_unlock_bh(&neigh->lock); - r->arp_ha.sa_family = dev->type; - strlcpy(r->arp_dev, dev->name, sizeof(r->arp_dev)); + if (!(neigh->nud_state & NUD_NOARP)) { + read_lock_bh(&neigh->lock); + memcpy(r->arp_ha.sa_data, neigh->ha, dev->addr_len); + r->arp_flags = arp_state_to_flags(neigh); + read_unlock_bh(&neigh->lock); + r->arp_ha.sa_family = dev->type; + strlcpy(r->arp_dev, dev->name, sizeof(r->arp_dev)); + err = 0; + } neigh_release(neigh); - err = 0; } return err; } -- cgit v0.10.2 From 3a05d12f46cb95a6a685114819363a56e6170996 Mon Sep 17 00:00:00 2001 From: Takashi Iwai Date: Wed, 29 Jul 2015 09:04:52 +0200 Subject: ALSA: hda - Apply a fixup to Dell Vostro 5480 Dell Vostro 5480 (1028:069a) needs the very same quirk used for Vostro 5470 model to make bass speakers properly working. Reported-and-tested-by: Paulo Roberto de Oliveira Castro Cc: Signed-off-by: Takashi Iwai diff --git a/sound/pci/hda/patch_realtek.c b/sound/pci/hda/patch_realtek.c index 4ae877c..18ae17e 100644 --- a/sound/pci/hda/patch_realtek.c +++ b/sound/pci/hda/patch_realtek.c @@ -5185,6 +5185,7 @@ static const struct snd_pci_quirk alc269_fixup_tbl[] = { SND_PCI_QUIRK(0x1028, 0x064a, "Dell", ALC293_FIXUP_DELL1_MIC_NO_PRESENCE), SND_PCI_QUIRK(0x1028, 0x064b, "Dell", ALC293_FIXUP_DELL1_MIC_NO_PRESENCE), SND_PCI_QUIRK(0x1028, 0x0665, "Dell XPS 13", ALC288_FIXUP_DELL_XPS_13), + SND_PCI_QUIRK(0x1028, 0x069a, "Dell Vostro 5480", ALC290_FIXUP_SUBWOOFER_HSJACK), SND_PCI_QUIRK(0x1028, 0x06c7, "Dell", ALC255_FIXUP_DELL1_MIC_NO_PRESENCE), SND_PCI_QUIRK(0x1028, 0x06d9, "Dell", ALC293_FIXUP_DELL1_MIC_NO_PRESENCE), SND_PCI_QUIRK(0x1028, 0x06da, "Dell", ALC293_FIXUP_DELL1_MIC_NO_PRESENCE), -- cgit v0.10.2 From dbd46ab412b8fb395f2b0ff6f6a7eec9df311550 Mon Sep 17 00:00:00 2001 From: Alexander Drozdov Date: Tue, 28 Jul 2015 13:57:01 +0300 Subject: packet: tpacket_snd(): fix signed/unsigned comparison tpacket_fill_skb() can return a negative value (-errno) which is stored in tp_len variable. In that case the following condition will be (but shouldn't be) true: tp_len > dev->mtu + dev->hard_header_len as dev->mtu and dev->hard_header_len are both unsigned. That may lead to just returning an incorrect EMSGSIZE errno to the user. Fixes: 52f1454f629fa ("packet: allow to transmit +4 byte in TX_RING slot for VLAN case") Signed-off-by: Alexander Drozdov Acked-by: Daniel Borkmann Signed-off-by: David S. Miller diff --git a/net/packet/af_packet.c b/net/packet/af_packet.c index c7c42eb..ed458b3 100644 --- a/net/packet/af_packet.c +++ b/net/packet/af_packet.c @@ -2403,7 +2403,8 @@ static int tpacket_snd(struct packet_sock *po, struct msghdr *msg) } tp_len = tpacket_fill_skb(po, skb, ph, dev, size_max, proto, addr, hlen); - if (tp_len > dev->mtu + dev->hard_header_len) { + if (likely(tp_len >= 0) && + tp_len > dev->mtu + dev->hard_header_len) { struct ethhdr *ehdr; /* Earlier code assumed this would be a VLAN pkt, * double-check this now that we have the actual -- cgit v0.10.2 From 2d1cb7f658fb9c3ba8f9dab8aca297d4dfdec835 Mon Sep 17 00:00:00 2001 From: Yao-Wen Mao Date: Wed, 29 Jul 2015 15:13:54 +0800 Subject: ALSA: usb-audio: add dB range mapping for some devices Add the correct dB ranges of Bose Companion 5 and Drangonfly DAC 1.2. Signed-off-by: Yao-Wen Mao Cc: Signed-off-by: Takashi Iwai diff --git a/sound/usb/mixer_maps.c b/sound/usb/mixer_maps.c index e5000da..6a803ef 100644 --- a/sound/usb/mixer_maps.c +++ b/sound/usb/mixer_maps.c @@ -341,6 +341,20 @@ static const struct usbmix_name_map scms_usb3318_map[] = { { 0 } }; +/* Bose companion 5, the dB conversion factor is 16 instead of 256 */ +static struct usbmix_dB_map bose_companion5_dB = {-5006, -6}; +static struct usbmix_name_map bose_companion5_map[] = { + { 3, NULL, .dB = &bose_companion5_dB }, + { 0 } /* terminator */ +}; + +/* Dragonfly DAC 1.2, the dB conversion factor is 1 instead of 256 */ +static struct usbmix_dB_map dragonfly_1_2_dB = {0, 5000}; +static struct usbmix_name_map dragonfly_1_2_map[] = { + { 7, NULL, .dB = &dragonfly_1_2_dB }, + { 0 } /* terminator */ +}; + /* * Control map entries */ @@ -451,6 +465,16 @@ static struct usbmix_ctl_map usbmix_ctl_maps[] = { .id = USB_ID(0x25c4, 0x0003), .map = scms_usb3318_map, }, + { + /* Bose Companion 5 */ + .id = USB_ID(0x05a7, 0x1020), + .map = bose_companion5_map, + }, + { + /* Dragonfly DAC 1.2 */ + .id = USB_ID(0x21b4, 0x0081), + .map = dragonfly_1_2_map, + }, { 0 } /* terminator */ }; -- cgit v0.10.2 From ee0a227b7ac6e75f28e10269f81c7ec6eb600952 Mon Sep 17 00:00:00 2001 From: Chris Wilson Date: Wed, 15 Jul 2015 09:50:42 +0100 Subject: drm/i915: Replace WARN inside I915_READ64_2x32 with retry loop MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Since we may conceivably encounter situations where the upper part of the 64bit register changes between reads, for example when a timestamp counter overflows, change the WARN into a retry loop. Signed-off-by: Chris Wilson Cc: MichaÅ‚ Winiarski Cc: stable@vger.kernel.org Signed-off-by: Daniel Vetter diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h index 5f27290..fd1de45 100644 --- a/drivers/gpu/drm/i915/i915_drv.h +++ b/drivers/gpu/drm/i915/i915_drv.h @@ -3303,15 +3303,14 @@ int intel_freq_opcode(struct drm_i915_private *dev_priv, int val); #define I915_READ64(reg) dev_priv->uncore.funcs.mmio_readq(dev_priv, (reg), true) #define I915_READ64_2x32(lower_reg, upper_reg) ({ \ - u32 upper = I915_READ(upper_reg); \ - u32 lower = I915_READ(lower_reg); \ - u32 tmp = I915_READ(upper_reg); \ - if (upper != tmp) { \ - upper = tmp; \ - lower = I915_READ(lower_reg); \ - WARN_ON(I915_READ(upper_reg) != upper); \ - } \ - (u64)upper << 32 | lower; }) + u32 upper, lower, tmp; \ + tmp = I915_READ(upper_reg); \ + do { \ + upper = tmp; \ + lower = I915_READ(lower_reg); \ + tmp = I915_READ(upper_reg); \ + } while (upper != tmp); \ + (u64)upper << 32 | lower; }) #define POSTING_READ(reg) (void)I915_READ_NOTRACE(reg) #define POSTING_READ16(reg) (void)I915_READ16_NOTRACE(reg) -- cgit v0.10.2 From 4b563ea317c2262987f0675abf15066614f536a1 Mon Sep 17 00:00:00 2001 From: Jie Yang Date: Tue, 28 Jul 2015 16:01:05 +0800 Subject: ASoC: Intel: haswell: fix initialize 'NULL device *' issue Add initialization for sst_hsw.dev at init stage, which fix the 'NULL device *' warning issues. Signed-off-by: Jie Yang Signed-off-by: Mark Brown diff --git a/sound/soc/intel/haswell/sst-haswell-ipc.c b/sound/soc/intel/haswell/sst-haswell-ipc.c index f95f271..f6efa9d 100644 --- a/sound/soc/intel/haswell/sst-haswell-ipc.c +++ b/sound/soc/intel/haswell/sst-haswell-ipc.c @@ -2119,6 +2119,8 @@ int sst_hsw_dsp_init(struct device *dev, struct sst_pdata *pdata) if (hsw == NULL) return -ENOMEM; + hsw->dev = dev; + ipc = &hsw->ipc; ipc->dev = dev; ipc->ops.tx_msg = hsw_tx_msg; -- cgit v0.10.2 From 45f503df1ba445359b94e1758c5e4f2c3460c8e4 Mon Sep 17 00:00:00 2001 From: Jie Yang Date: Tue, 28 Jul 2015 16:01:06 +0800 Subject: ASoC: Intel: sst_byt: fix initialize 'NULL device *' issue Add initialization for sst_byt.dev at init stage, which fix the 'NULL device *' warning issues. Signed-off-by: Jie Yang Signed-off-by: Mark Brown diff --git a/sound/soc/intel/baytrail/sst-baytrail-ipc.c b/sound/soc/intel/baytrail/sst-baytrail-ipc.c index 4c01bb4..5bbaa66 100644 --- a/sound/soc/intel/baytrail/sst-baytrail-ipc.c +++ b/sound/soc/intel/baytrail/sst-baytrail-ipc.c @@ -701,6 +701,8 @@ int sst_byt_dsp_init(struct device *dev, struct sst_pdata *pdata) if (byt == NULL) return -ENOMEM; + byt->dev = dev; + ipc = &byt->ipc; ipc->dev = dev; ipc->ops.tx_msg = byt_tx_msg; -- cgit v0.10.2 From 342e84490574cbb2a9c5b1d0886a112ad2bcf4d7 Mon Sep 17 00:00:00 2001 From: "U. Artie Eoff" Date: Tue, 28 Jul 2015 13:29:56 -0700 Subject: ALSA: hda - Fix race between PM ops and HDA init/probe PM ops could be triggered before HDA is done initializing and cause PM to set HDA controller to D3Hot. This can result in "CORB reset timeout#2, CORBRP = 65535" and "no codecs initialized". Additionally, PM ops can be triggered before azx_probe_continue finishes (async probe). This can result in a NULL deref kernel crash. To fix this, avoid PM ops if !chip->running. Signed-off-by: U. Artie Eoff Signed-off-by: Takashi Iwai diff --git a/sound/pci/hda/hda_intel.c b/sound/pci/hda/hda_intel.c index 735bdcb..c38c68f 100644 --- a/sound/pci/hda/hda_intel.c +++ b/sound/pci/hda/hda_intel.c @@ -867,7 +867,7 @@ static int azx_suspend(struct device *dev) chip = card->private_data; hda = container_of(chip, struct hda_intel, chip); - if (chip->disabled || hda->init_failed) + if (chip->disabled || hda->init_failed || !chip->running) return 0; bus = azx_bus(chip); @@ -902,7 +902,7 @@ static int azx_resume(struct device *dev) chip = card->private_data; hda = container_of(chip, struct hda_intel, chip); - if (chip->disabled || hda->init_failed) + if (chip->disabled || hda->init_failed || !chip->running) return 0; if (chip->driver_caps & AZX_DCAPS_I915_POWERWELL @@ -1027,7 +1027,7 @@ static int azx_runtime_idle(struct device *dev) return 0; if (!power_save_controller || !azx_has_pm_runtime(chip) || - azx_bus(chip)->codec_powered) + azx_bus(chip)->codec_powered || !chip->running) return -EBUSY; return 0; -- cgit v0.10.2 From 3508e6590d4729ac07f01f7ae2256c2f9dc738b8 Mon Sep 17 00:00:00 2001 From: Mike Snitzer Date: Wed, 29 Jul 2015 14:11:28 -0400 Subject: Revert "dm cache: do not wake_worker() in free_migration()" This reverts commit 386cb7cdeeef97e0bf082a8d6bbfc07a2ccce07b. Taking the wake_worker() out of free_migration() will slow writeback dramatically, and hence adaptability. Say we have 10k blocks that need writing back, but are only able to issue 5 concurrently due to the migration bandwidth: it's imperative that we wake_worker() immediately after migration completion; waiting for the next 1 second wake up (via do_waker) means it'll take a long time to write that all back. Reported-by: Joe Thornber Signed-off-by: Mike Snitzer diff --git a/drivers/md/dm-cache-target.c b/drivers/md/dm-cache-target.c index b680da5..64e96a2 100644 --- a/drivers/md/dm-cache-target.c +++ b/drivers/md/dm-cache-target.c @@ -424,6 +424,7 @@ static void free_migration(struct dm_cache_migration *mg) wake_up(&cache->migration_wait); mempool_free(mg, cache->migration_pool); + wake_worker(cache); } static int prealloc_data_structs(struct cache *cache, struct prealloc *p) -- cgit v0.10.2 From 795e633a2dc6cbbeac68bc7f6006082150d38bb7 Mon Sep 17 00:00:00 2001 From: Mike Snitzer Date: Wed, 29 Jul 2015 13:48:23 -0400 Subject: dm cache: fix device destroy hang due to improper prealloc_used accounting Commit 665022d72f9 ("dm cache: avoid calls to prealloc_free_structs() if possible") introduced a regression that caused the removal of a DM cache device to hang in cache_postsuspend()'s call to wait_for_migrations() with the following stack trace: [] schedule+0x37/0x80 [] cache_postsuspend+0xbb/0x470 [dm_cache] [] ? prepare_to_wait_event+0xf0/0xf0 [] dm_table_postsuspend_targets+0x47/0x60 [dm_mod] [] __dm_destroy+0x215/0x250 [dm_mod] [] dm_destroy+0x13/0x20 [dm_mod] [] dev_remove+0x10d/0x170 [dm_mod] [] ? dev_suspend+0x240/0x240 [dm_mod] [] ctl_ioctl+0x255/0x4d0 [dm_mod] [] ? SYSC_semtimedop+0x280/0xe10 [] dm_ctl_ioctl+0x13/0x20 [dm_mod] [] do_vfs_ioctl+0x2d2/0x4b0 [] ? __audit_syscall_entry+0xaf/0x100 [] ? do_audit_syscall_entry+0x66/0x70 [] SyS_ioctl+0x79/0x90 [] ? syscall_trace_leave+0xb8/0x110 [] entry_SYSCALL_64_fastpath+0x12/0x71 Fix this by accounting for the call to prealloc_data_structs() immediately _before_ the call as opposed to after. This is needed because it is possible to break out of the control loop after the call to prealloc_data_structs() but before prealloc_used was set to true. Signed-off-by: Mike Snitzer diff --git a/drivers/md/dm-cache-target.c b/drivers/md/dm-cache-target.c index 64e96a2..1fe93cf 100644 --- a/drivers/md/dm-cache-target.c +++ b/drivers/md/dm-cache-target.c @@ -1967,6 +1967,7 @@ static void process_deferred_bios(struct cache *cache) * this bio might require one, we pause until there are some * prepared mappings to process. */ + prealloc_used = true; if (prealloc_data_structs(cache, &structs)) { spin_lock_irqsave(&cache->lock, flags); bio_list_merge(&cache->deferred_bios, &bios); @@ -1982,7 +1983,6 @@ static void process_deferred_bios(struct cache *cache) process_discard_bio(cache, &structs, bio); else process_bio(cache, &structs, bio); - prealloc_used = true; } if (prealloc_used) @@ -2011,6 +2011,7 @@ static void process_deferred_cells(struct cache *cache) * this bio might require one, we pause until there are some * prepared mappings to process. */ + prealloc_used = true; if (prealloc_data_structs(cache, &structs)) { spin_lock_irqsave(&cache->lock, flags); list_splice(&cells, &cache->deferred_cells); @@ -2019,7 +2020,6 @@ static void process_deferred_cells(struct cache *cache) } process_cell(cache, &structs, cell); - prealloc_used = true; } if (prealloc_used) @@ -2081,6 +2081,7 @@ static void writeback_some_dirty_blocks(struct cache *cache) if (policy_writeback_work(cache->policy, &oblock, &cblock, busy)) break; /* no work to do */ + prealloc_used = true; if (prealloc_data_structs(cache, &structs) || get_cell(cache, oblock, &structs, &old_ocell)) { policy_set_dirty(cache->policy, oblock); @@ -2088,7 +2089,6 @@ static void writeback_some_dirty_blocks(struct cache *cache) } writeback(cache, &structs, oblock, cblock, old_ocell); - prealloc_used = true; } if (prealloc_used) -- cgit v0.10.2 From df356d5e81b04dd51dd9f23f2bce7d73dd929899 Mon Sep 17 00:00:00 2001 From: Toshiaki Makita Date: Tue, 28 Jul 2015 19:05:37 +0900 Subject: bridge: Fix network header pointer for vlan tagged packets There are several devices that can receive vlan tagged packets with CHECKSUM_PARTIAL like tap, possibly veth and xennet. When (multiple) vlan tagged packets with CHECKSUM_PARTIAL are forwarded by bridge to a device with the IP_CSUM feature, they end up with checksum error because before entering bridge, the network header is set to ETH_HLEN (not including vlan header length) in __netif_receive_skb_core(), get_rps_cpu(), or drivers' rx functions, and nobody fixes the pointer later. Since the network header is exepected to be ETH_HLEN in flow-dissection and hash-calculation in RPS in rx path, and since the header pointer fix is needed only in tx path, set the appropriate network header on forwarding packets. Signed-off-by: Toshiaki Makita Signed-off-by: David S. Miller diff --git a/net/bridge/br_forward.c b/net/bridge/br_forward.c index 0ff6e1b..fa7bfce 100644 --- a/net/bridge/br_forward.c +++ b/net/bridge/br_forward.c @@ -37,15 +37,30 @@ static inline int should_deliver(const struct net_bridge_port *p, int br_dev_queue_push_xmit(struct sock *sk, struct sk_buff *skb) { - if (!is_skb_forwardable(skb->dev, skb)) { - kfree_skb(skb); - } else { - skb_push(skb, ETH_HLEN); - br_drop_fake_rtable(skb); - skb_sender_cpu_clear(skb); - dev_queue_xmit(skb); + if (!is_skb_forwardable(skb->dev, skb)) + goto drop; + + skb_push(skb, ETH_HLEN); + br_drop_fake_rtable(skb); + skb_sender_cpu_clear(skb); + + if (skb->ip_summed == CHECKSUM_PARTIAL && + (skb->protocol == htons(ETH_P_8021Q) || + skb->protocol == htons(ETH_P_8021AD))) { + int depth; + + if (!__vlan_get_protocol(skb, skb->protocol, &depth)) + goto drop; + + skb_set_network_header(skb, depth); } + dev_queue_xmit(skb); + + return 0; + +drop: + kfree_skb(skb); return 0; } EXPORT_SYMBOL_GPL(br_dev_queue_push_xmit); -- cgit v0.10.2 From e13af53e7d5a8cea8992d9b61fac69bd0ed8d845 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Michel=20D=C3=A4nzer?= Date: Wed, 22 Jul 2015 17:29:00 +0900 Subject: drm/radeon: Drop drm/ prefix for including drm.h in radeon_drm.h MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit This allows radeon_drm.h to be reused verbatim in libdrm. Reviewed-by: Christian König Signed-off-by: Michel Dänzer diff --git a/include/uapi/drm/radeon_drm.h b/include/uapi/drm/radeon_drm.h index 1ef7666..01aa2a8 100644 --- a/include/uapi/drm/radeon_drm.h +++ b/include/uapi/drm/radeon_drm.h @@ -33,7 +33,7 @@ #ifndef __RADEON_DRM_H__ #define __RADEON_DRM_H__ -#include +#include "drm.h" /* WARNING: If you change any of these defines, make sure to change the * defines in the X server file (radeon_sarea.h) -- cgit v0.10.2 From b3fcf36aeef3aeb890d9413c2066048ec7fda7e5 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Michel=20D=C3=A4nzer?= Date: Wed, 22 Jul 2015 17:29:01 +0900 Subject: drm/amdgpu: Drop drm/ prefix for including drm.h in amdgpu_drm.h MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit This allows amdgpu_drm.h to be reused verbatim in libdrm. Reviewed-by: Alex Deucher Reviewed-by: Christian König Signed-off-by: Michel Dänzer diff --git a/include/uapi/drm/amdgpu_drm.h b/include/uapi/drm/amdgpu_drm.h index d708a53..fbdd118 100644 --- a/include/uapi/drm/amdgpu_drm.h +++ b/include/uapi/drm/amdgpu_drm.h @@ -32,7 +32,7 @@ #ifndef __AMDGPU_DRM_H__ #define __AMDGPU_DRM_H__ -#include +#include "drm.h" #define DRM_AMDGPU_GEM_CREATE 0x00 #define DRM_AMDGPU_GEM_MMAP 0x01 -- cgit v0.10.2 From d0ea397e22f9ad0113c1dbdaab14eded050472eb Mon Sep 17 00:00:00 2001 From: Alex Deucher Date: Thu, 23 Jul 2015 10:01:09 -0400 Subject: drm/radeon: rework audio detect (v4) 1. Always assign audio function pointers even if the display does not support audio. We need to properly disable the audio stream when when using a non-audio capable monitor. Fixes purple line on some hdmi monitors. 2. Check if a pin is in use by another encoder before disabling it. v2: make sure we've fetched the edid before checking audio and look up the encoder before calling audio_detect since connector->encoder may not be assigned yet. Separate pin and afmt. They are allocated at different times and have no dependency on eachother. v3: fix connector fetching in encoder functions v4: fix missed dig->pin check in dce6_afmt_write_latency_fields bugs: https://bugzilla.kernel.org/show_bug.cgi?id=93701 https://bugzilla.redhat.com/show_bug.cgi?id=1236337 https://bugs.freedesktop.org/show_bug.cgi?id=91041 Signed-off-by: Alex Deucher Cc: stable@vger.kernel.org diff --git a/drivers/gpu/drm/radeon/dce6_afmt.c b/drivers/gpu/drm/radeon/dce6_afmt.c index 68fd9fc..44480c1 100644 --- a/drivers/gpu/drm/radeon/dce6_afmt.c +++ b/drivers/gpu/drm/radeon/dce6_afmt.c @@ -93,30 +93,26 @@ void dce6_afmt_select_pin(struct drm_encoder *encoder) struct radeon_device *rdev = encoder->dev->dev_private; struct radeon_encoder *radeon_encoder = to_radeon_encoder(encoder); struct radeon_encoder_atom_dig *dig = radeon_encoder->enc_priv; - u32 offset; - if (!dig || !dig->afmt || !dig->afmt->pin) + if (!dig || !dig->afmt || !dig->pin) return; - offset = dig->afmt->offset; - - WREG32(AFMT_AUDIO_SRC_CONTROL + offset, - AFMT_AUDIO_SRC_SELECT(dig->afmt->pin->id)); + WREG32(AFMT_AUDIO_SRC_CONTROL + dig->afmt->offset, + AFMT_AUDIO_SRC_SELECT(dig->pin->id)); } void dce6_afmt_write_latency_fields(struct drm_encoder *encoder, - struct drm_connector *connector, struct drm_display_mode *mode) + struct drm_connector *connector, + struct drm_display_mode *mode) { struct radeon_device *rdev = encoder->dev->dev_private; struct radeon_encoder *radeon_encoder = to_radeon_encoder(encoder); struct radeon_encoder_atom_dig *dig = radeon_encoder->enc_priv; - u32 tmp = 0, offset; + u32 tmp = 0; - if (!dig || !dig->afmt || !dig->afmt->pin) + if (!dig || !dig->afmt || !dig->pin) return; - offset = dig->afmt->pin->offset; - if (mode->flags & DRM_MODE_FLAG_INTERLACE) { if (connector->latency_present[1]) tmp = VIDEO_LIPSYNC(connector->video_latency[1]) | @@ -130,24 +126,24 @@ void dce6_afmt_write_latency_fields(struct drm_encoder *encoder, else tmp = VIDEO_LIPSYNC(0) | AUDIO_LIPSYNC(0); } - WREG32_ENDPOINT(offset, AZ_F0_CODEC_PIN_CONTROL_RESPONSE_LIPSYNC, tmp); + WREG32_ENDPOINT(dig->pin->offset, + AZ_F0_CODEC_PIN_CONTROL_RESPONSE_LIPSYNC, tmp); } void dce6_afmt_hdmi_write_speaker_allocation(struct drm_encoder *encoder, - u8 *sadb, int sad_count) + u8 *sadb, int sad_count) { struct radeon_device *rdev = encoder->dev->dev_private; struct radeon_encoder *radeon_encoder = to_radeon_encoder(encoder); struct radeon_encoder_atom_dig *dig = radeon_encoder->enc_priv; - u32 offset, tmp; + u32 tmp; - if (!dig || !dig->afmt || !dig->afmt->pin) + if (!dig || !dig->afmt || !dig->pin) return; - offset = dig->afmt->pin->offset; - /* program the speaker allocation */ - tmp = RREG32_ENDPOINT(offset, AZ_F0_CODEC_PIN_CONTROL_CHANNEL_SPEAKER); + tmp = RREG32_ENDPOINT(dig->pin->offset, + AZ_F0_CODEC_PIN_CONTROL_CHANNEL_SPEAKER); tmp &= ~(DP_CONNECTION | SPEAKER_ALLOCATION_MASK); /* set HDMI mode */ tmp |= HDMI_CONNECTION; @@ -155,24 +151,24 @@ void dce6_afmt_hdmi_write_speaker_allocation(struct drm_encoder *encoder, tmp |= SPEAKER_ALLOCATION(sadb[0]); else tmp |= SPEAKER_ALLOCATION(5); /* stereo */ - WREG32_ENDPOINT(offset, AZ_F0_CODEC_PIN_CONTROL_CHANNEL_SPEAKER, tmp); + WREG32_ENDPOINT(dig->pin->offset, + AZ_F0_CODEC_PIN_CONTROL_CHANNEL_SPEAKER, tmp); } void dce6_afmt_dp_write_speaker_allocation(struct drm_encoder *encoder, - u8 *sadb, int sad_count) + u8 *sadb, int sad_count) { struct radeon_device *rdev = encoder->dev->dev_private; struct radeon_encoder *radeon_encoder = to_radeon_encoder(encoder); struct radeon_encoder_atom_dig *dig = radeon_encoder->enc_priv; - u32 offset, tmp; + u32 tmp; - if (!dig || !dig->afmt || !dig->afmt->pin) + if (!dig || !dig->afmt || !dig->pin) return; - offset = dig->afmt->pin->offset; - /* program the speaker allocation */ - tmp = RREG32_ENDPOINT(offset, AZ_F0_CODEC_PIN_CONTROL_CHANNEL_SPEAKER); + tmp = RREG32_ENDPOINT(dig->pin->offset, + AZ_F0_CODEC_PIN_CONTROL_CHANNEL_SPEAKER); tmp &= ~(HDMI_CONNECTION | SPEAKER_ALLOCATION_MASK); /* set DP mode */ tmp |= DP_CONNECTION; @@ -180,13 +176,13 @@ void dce6_afmt_dp_write_speaker_allocation(struct drm_encoder *encoder, tmp |= SPEAKER_ALLOCATION(sadb[0]); else tmp |= SPEAKER_ALLOCATION(5); /* stereo */ - WREG32_ENDPOINT(offset, AZ_F0_CODEC_PIN_CONTROL_CHANNEL_SPEAKER, tmp); + WREG32_ENDPOINT(dig->pin->offset, + AZ_F0_CODEC_PIN_CONTROL_CHANNEL_SPEAKER, tmp); } void dce6_afmt_write_sad_regs(struct drm_encoder *encoder, - struct cea_sad *sads, int sad_count) + struct cea_sad *sads, int sad_count) { - u32 offset; int i; struct radeon_encoder *radeon_encoder = to_radeon_encoder(encoder); struct radeon_encoder_atom_dig *dig = radeon_encoder->enc_priv; @@ -206,11 +202,9 @@ void dce6_afmt_write_sad_regs(struct drm_encoder *encoder, { AZ_F0_CODEC_PIN_CONTROL_AUDIO_DESCRIPTOR13, HDMI_AUDIO_CODING_TYPE_WMA_PRO }, }; - if (!dig || !dig->afmt || !dig->afmt->pin) + if (!dig || !dig->afmt || !dig->pin) return; - offset = dig->afmt->pin->offset; - for (i = 0; i < ARRAY_SIZE(eld_reg_to_type); i++) { u32 value = 0; u8 stereo_freqs = 0; @@ -237,7 +231,7 @@ void dce6_afmt_write_sad_regs(struct drm_encoder *encoder, value |= SUPPORTED_FREQUENCIES_STEREO(stereo_freqs); - WREG32_ENDPOINT(offset, eld_reg_to_type[i][0], value); + WREG32_ENDPOINT(dig->pin->offset, eld_reg_to_type[i][0], value); } } @@ -253,7 +247,7 @@ void dce6_audio_enable(struct radeon_device *rdev, } void dce6_hdmi_audio_set_dto(struct radeon_device *rdev, - struct radeon_crtc *crtc, unsigned int clock) + struct radeon_crtc *crtc, unsigned int clock) { /* Two dtos; generally use dto0 for HDMI */ u32 value = 0; @@ -272,7 +266,7 @@ void dce6_hdmi_audio_set_dto(struct radeon_device *rdev, } void dce6_dp_audio_set_dto(struct radeon_device *rdev, - struct radeon_crtc *crtc, unsigned int clock) + struct radeon_crtc *crtc, unsigned int clock) { /* Two dtos; generally use dto1 for DP */ u32 value = 0; diff --git a/drivers/gpu/drm/radeon/radeon_audio.c b/drivers/gpu/drm/radeon/radeon_audio.c index fa719c5..59b3d32 100644 --- a/drivers/gpu/drm/radeon/radeon_audio.c +++ b/drivers/gpu/drm/radeon/radeon_audio.c @@ -245,6 +245,28 @@ static struct radeon_audio_funcs dce6_dp_funcs = { static void radeon_audio_enable(struct radeon_device *rdev, struct r600_audio_pin *pin, u8 enable_mask) { + struct drm_encoder *encoder; + struct radeon_encoder *radeon_encoder; + struct radeon_encoder_atom_dig *dig; + int pin_count = 0; + + if (!pin) + return; + + if (rdev->mode_info.mode_config_initialized) { + list_for_each_entry(encoder, &rdev->ddev->mode_config.encoder_list, head) { + if (radeon_encoder_is_digital(encoder)) { + radeon_encoder = to_radeon_encoder(encoder); + dig = radeon_encoder->enc_priv; + if (dig->pin == pin) + pin_count++; + } + } + + if ((pin_count > 1) && (enable_mask == 0)) + return; + } + if (rdev->audio.funcs->enable) rdev->audio.funcs->enable(rdev, pin, enable_mask); } @@ -336,24 +358,13 @@ void radeon_audio_endpoint_wreg(struct radeon_device *rdev, u32 offset, static void radeon_audio_write_sad_regs(struct drm_encoder *encoder) { - struct radeon_encoder *radeon_encoder; - struct drm_connector *connector; - struct radeon_connector *radeon_connector = NULL; + struct drm_connector *connector = radeon_get_connector_for_encoder(encoder); + struct radeon_encoder *radeon_encoder = to_radeon_encoder(encoder); struct cea_sad *sads; int sad_count; - list_for_each_entry(connector, - &encoder->dev->mode_config.connector_list, head) { - if (connector->encoder == encoder) { - radeon_connector = to_radeon_connector(connector); - break; - } - } - - if (!radeon_connector) { - DRM_ERROR("Couldn't find encoder's connector\n"); + if (!connector) return; - } sad_count = drm_edid_to_sad(radeon_connector_edid(connector), &sads); if (sad_count <= 0) { @@ -362,8 +373,6 @@ static void radeon_audio_write_sad_regs(struct drm_encoder *encoder) } BUG_ON(!sads); - radeon_encoder = to_radeon_encoder(encoder); - if (radeon_encoder->audio && radeon_encoder->audio->write_sad_regs) radeon_encoder->audio->write_sad_regs(encoder, sads, sad_count); @@ -372,27 +381,16 @@ static void radeon_audio_write_sad_regs(struct drm_encoder *encoder) static void radeon_audio_write_speaker_allocation(struct drm_encoder *encoder) { + struct drm_connector *connector = radeon_get_connector_for_encoder(encoder); struct radeon_encoder *radeon_encoder = to_radeon_encoder(encoder); - struct drm_connector *connector; - struct radeon_connector *radeon_connector = NULL; u8 *sadb = NULL; int sad_count; - list_for_each_entry(connector, - &encoder->dev->mode_config.connector_list, head) { - if (connector->encoder == encoder) { - radeon_connector = to_radeon_connector(connector); - break; - } - } - - if (!radeon_connector) { - DRM_ERROR("Couldn't find encoder's connector\n"); + if (!connector) return; - } - sad_count = drm_edid_to_speaker_allocation( - radeon_connector_edid(connector), &sadb); + sad_count = drm_edid_to_speaker_allocation(radeon_connector_edid(connector), + &sadb); if (sad_count < 0) { DRM_DEBUG("Couldn't read Speaker Allocation Data Block: %d\n", sad_count); @@ -406,26 +404,13 @@ static void radeon_audio_write_speaker_allocation(struct drm_encoder *encoder) } static void radeon_audio_write_latency_fields(struct drm_encoder *encoder, - struct drm_display_mode *mode) + struct drm_display_mode *mode) { - struct radeon_encoder *radeon_encoder; - struct drm_connector *connector; - struct radeon_connector *radeon_connector = 0; - - list_for_each_entry(connector, - &encoder->dev->mode_config.connector_list, head) { - if (connector->encoder == encoder) { - radeon_connector = to_radeon_connector(connector); - break; - } - } + struct drm_connector *connector = radeon_get_connector_for_encoder(encoder); + struct radeon_encoder *radeon_encoder = to_radeon_encoder(encoder); - if (!radeon_connector) { - DRM_ERROR("Couldn't find encoder's connector\n"); + if (!connector) return; - } - - radeon_encoder = to_radeon_encoder(encoder); if (radeon_encoder->audio && radeon_encoder->audio->write_latency_fields) radeon_encoder->audio->write_latency_fields(encoder, connector, mode); @@ -451,29 +436,23 @@ static void radeon_audio_select_pin(struct drm_encoder *encoder) } void radeon_audio_detect(struct drm_connector *connector, + struct drm_encoder *encoder, enum drm_connector_status status) { - struct radeon_device *rdev; - struct radeon_encoder *radeon_encoder; + struct drm_device *dev = connector->dev; + struct radeon_device *rdev = dev->dev_private; + struct radeon_encoder *radeon_encoder = to_radeon_encoder(encoder); struct radeon_encoder_atom_dig *dig; - if (!connector || !connector->encoder) + if (!radeon_audio_chipset_supported(rdev)) return; - rdev = connector->encoder->dev->dev_private; - - if (!radeon_audio_chipset_supported(rdev)) + if (!radeon_encoder_is_digital(encoder)) return; - radeon_encoder = to_radeon_encoder(connector->encoder); dig = radeon_encoder->enc_priv; if (status == connector_status_connected) { - if (!drm_detect_monitor_audio(radeon_connector_edid(connector))) { - radeon_encoder->audio = NULL; - return; - } - if (connector->connector_type == DRM_MODE_CONNECTOR_DisplayPort) { struct radeon_connector *radeon_connector = to_radeon_connector(connector); @@ -486,11 +465,17 @@ void radeon_audio_detect(struct drm_connector *connector, radeon_encoder->audio = rdev->audio.hdmi_funcs; } - dig->afmt->pin = radeon_audio_get_pin(connector->encoder); - radeon_audio_enable(rdev, dig->afmt->pin, 0xf); + if (drm_detect_monitor_audio(radeon_connector_edid(connector))) { + if (!dig->pin) + dig->pin = radeon_audio_get_pin(encoder); + radeon_audio_enable(rdev, dig->pin, 0xf); + } else { + radeon_audio_enable(rdev, dig->pin, 0); + dig->pin = NULL; + } } else { - radeon_audio_enable(rdev, dig->afmt->pin, 0); - dig->afmt->pin = NULL; + radeon_audio_enable(rdev, dig->pin, 0); + dig->pin = NULL; } } @@ -518,29 +503,18 @@ static void radeon_audio_set_dto(struct drm_encoder *encoder, unsigned int clock } static int radeon_audio_set_avi_packet(struct drm_encoder *encoder, - struct drm_display_mode *mode) + struct drm_display_mode *mode) { struct radeon_device *rdev = encoder->dev->dev_private; struct radeon_encoder *radeon_encoder = to_radeon_encoder(encoder); struct radeon_encoder_atom_dig *dig = radeon_encoder->enc_priv; - struct drm_connector *connector; - struct radeon_connector *radeon_connector = NULL; + struct drm_connector *connector = radeon_get_connector_for_encoder(encoder); u8 buffer[HDMI_INFOFRAME_HEADER_SIZE + HDMI_AVI_INFOFRAME_SIZE]; struct hdmi_avi_infoframe frame; int err; - list_for_each_entry(connector, - &encoder->dev->mode_config.connector_list, head) { - if (connector->encoder == encoder) { - radeon_connector = to_radeon_connector(connector); - break; - } - } - - if (!radeon_connector) { - DRM_ERROR("Couldn't find encoder's connector\n"); - return -ENOENT; - } + if (!connector) + return -EINVAL; err = drm_hdmi_avi_infoframe_from_display_mode(&frame, mode); if (err < 0) { @@ -563,8 +537,8 @@ static int radeon_audio_set_avi_packet(struct drm_encoder *encoder, return err; } - if (dig && dig->afmt && - radeon_encoder->audio && radeon_encoder->audio->set_avi_packet) + if (dig && dig->afmt && radeon_encoder->audio && + radeon_encoder->audio->set_avi_packet) radeon_encoder->audio->set_avi_packet(rdev, dig->afmt->offset, buffer, sizeof(buffer)); @@ -745,7 +719,7 @@ static void radeon_audio_hdmi_mode_set(struct drm_encoder *encoder, } static void radeon_audio_dp_mode_set(struct drm_encoder *encoder, - struct drm_display_mode *mode) + struct drm_display_mode *mode) { struct drm_device *dev = encoder->dev; struct radeon_device *rdev = dev->dev_private; @@ -756,6 +730,9 @@ static void radeon_audio_dp_mode_set(struct drm_encoder *encoder, struct radeon_connector_atom_dig *dig_connector = radeon_connector->con_priv; + if (!connector) + return; + if (!dig || !dig->afmt) return; @@ -774,7 +751,7 @@ static void radeon_audio_dp_mode_set(struct drm_encoder *encoder, } void radeon_audio_mode_set(struct drm_encoder *encoder, - struct drm_display_mode *mode) + struct drm_display_mode *mode) { struct radeon_encoder *radeon_encoder = to_radeon_encoder(encoder); diff --git a/drivers/gpu/drm/radeon/radeon_audio.h b/drivers/gpu/drm/radeon/radeon_audio.h index 8438304..059cc30 100644 --- a/drivers/gpu/drm/radeon/radeon_audio.h +++ b/drivers/gpu/drm/radeon/radeon_audio.h @@ -68,7 +68,8 @@ struct radeon_audio_funcs int radeon_audio_init(struct radeon_device *rdev); void radeon_audio_detect(struct drm_connector *connector, - enum drm_connector_status status); + struct drm_encoder *encoder, + enum drm_connector_status status); u32 radeon_audio_endpoint_rreg(struct radeon_device *rdev, u32 offset, u32 reg); void radeon_audio_endpoint_wreg(struct radeon_device *rdev, diff --git a/drivers/gpu/drm/radeon/radeon_connectors.c b/drivers/gpu/drm/radeon/radeon_connectors.c index cebb65e..94b21ae 100644 --- a/drivers/gpu/drm/radeon/radeon_connectors.c +++ b/drivers/gpu/drm/radeon/radeon_connectors.c @@ -1379,8 +1379,16 @@ out: /* updated in get modes as well since we need to know if it's analog or digital */ radeon_connector_update_scratch_regs(connector, ret); - if (radeon_audio != 0) - radeon_audio_detect(connector, ret); + if ((radeon_audio != 0) && radeon_connector->use_digital) { + const struct drm_connector_helper_funcs *connector_funcs = + connector->helper_private; + + encoder = connector_funcs->best_encoder(connector); + if (encoder && (encoder->encoder_type == DRM_MODE_ENCODER_TMDS)) { + radeon_connector_get_edid(connector); + radeon_audio_detect(connector, encoder, ret); + } + } exit: pm_runtime_mark_last_busy(connector->dev->dev); @@ -1717,8 +1725,10 @@ radeon_dp_detect(struct drm_connector *connector, bool force) radeon_connector_update_scratch_regs(connector, ret); - if (radeon_audio != 0) - radeon_audio_detect(connector, ret); + if ((radeon_audio != 0) && encoder) { + radeon_connector_get_edid(connector); + radeon_audio_detect(connector, encoder, ret); + } out: pm_runtime_mark_last_busy(connector->dev->dev); diff --git a/drivers/gpu/drm/radeon/radeon_mode.h b/drivers/gpu/drm/radeon/radeon_mode.h index 07909d8..aecc3e3 100644 --- a/drivers/gpu/drm/radeon/radeon_mode.h +++ b/drivers/gpu/drm/radeon/radeon_mode.h @@ -237,7 +237,6 @@ struct radeon_afmt { int offset; bool last_buffer_filled_status; int id; - struct r600_audio_pin *pin; }; struct radeon_mode_info { @@ -439,6 +438,7 @@ struct radeon_encoder_atom_dig { uint8_t backlight_level; int panel_mode; struct radeon_afmt *afmt; + struct r600_audio_pin *pin; int active_mst_links; }; -- cgit v0.10.2 From 7726e72b3d6879ee5fc743a230eb6f5afa12844b Mon Sep 17 00:00:00 2001 From: Alex Deucher Date: Fri, 24 Jul 2015 00:42:02 -0400 Subject: drm/radeon: rework audio modeset to handle non-audio hdmi features Need to setup the deep color and avi packets regardless of audio setup. Signed-off-by: Alex Deucher diff --git a/drivers/gpu/drm/radeon/atombios_encoders.c b/drivers/gpu/drm/radeon/atombios_encoders.c index dd39f43..c387259 100644 --- a/drivers/gpu/drm/radeon/atombios_encoders.c +++ b/drivers/gpu/drm/radeon/atombios_encoders.c @@ -2299,8 +2299,7 @@ radeon_atom_encoder_mode_set(struct drm_encoder *encoder, encoder_mode = atombios_get_encoder_mode(encoder); if (connector && (radeon_audio != 0) && ((encoder_mode == ATOM_ENCODER_MODE_HDMI) || - (ENCODER_MODE_IS_DP(encoder_mode) && - drm_detect_monitor_audio(radeon_connector_edid(connector))))) + ENCODER_MODE_IS_DP(encoder_mode))) radeon_audio_mode_set(encoder, adjusted_mode); } diff --git a/drivers/gpu/drm/radeon/radeon_audio.c b/drivers/gpu/drm/radeon/radeon_audio.c index 59b3d32..fbc8d88 100644 --- a/drivers/gpu/drm/radeon/radeon_audio.c +++ b/drivers/gpu/drm/radeon/radeon_audio.c @@ -696,26 +696,37 @@ static void radeon_audio_hdmi_mode_set(struct drm_encoder *encoder, { struct radeon_encoder *radeon_encoder = to_radeon_encoder(encoder); struct radeon_encoder_atom_dig *dig = radeon_encoder->enc_priv; + struct drm_connector *connector = radeon_get_connector_for_encoder(encoder); if (!dig || !dig->afmt) return; - radeon_audio_set_mute(encoder, true); + if (!connector) + return; - radeon_audio_write_speaker_allocation(encoder); - radeon_audio_write_sad_regs(encoder); - radeon_audio_write_latency_fields(encoder, mode); - radeon_audio_set_dto(encoder, mode->clock); - radeon_audio_set_vbi_packet(encoder); - radeon_hdmi_set_color_depth(encoder); - radeon_audio_update_acr(encoder, mode->clock); - radeon_audio_set_audio_packet(encoder); - radeon_audio_select_pin(encoder); + if (drm_detect_monitor_audio(radeon_connector_edid(connector))) { + radeon_audio_set_mute(encoder, true); - if (radeon_audio_set_avi_packet(encoder, mode) < 0) - return; + radeon_audio_write_speaker_allocation(encoder); + radeon_audio_write_sad_regs(encoder); + radeon_audio_write_latency_fields(encoder, mode); + radeon_audio_set_dto(encoder, mode->clock); + radeon_audio_set_vbi_packet(encoder); + radeon_hdmi_set_color_depth(encoder); + radeon_audio_update_acr(encoder, mode->clock); + radeon_audio_set_audio_packet(encoder); + radeon_audio_select_pin(encoder); + + if (radeon_audio_set_avi_packet(encoder, mode) < 0) + return; + + radeon_audio_set_mute(encoder, false); + } else { + radeon_hdmi_set_color_depth(encoder); - radeon_audio_set_mute(encoder, false); + if (radeon_audio_set_avi_packet(encoder, mode) < 0) + return; + } } static void radeon_audio_dp_mode_set(struct drm_encoder *encoder, @@ -730,24 +741,26 @@ static void radeon_audio_dp_mode_set(struct drm_encoder *encoder, struct radeon_connector_atom_dig *dig_connector = radeon_connector->con_priv; - if (!connector) + if (!dig || !dig->afmt) return; - if (!dig || !dig->afmt) + if (!connector) return; - radeon_audio_write_speaker_allocation(encoder); - radeon_audio_write_sad_regs(encoder); - radeon_audio_write_latency_fields(encoder, mode); - if (rdev->clock.dp_extclk || ASIC_IS_DCE5(rdev)) - radeon_audio_set_dto(encoder, rdev->clock.default_dispclk * 10); - else - radeon_audio_set_dto(encoder, dig_connector->dp_clock); - radeon_audio_set_audio_packet(encoder); - radeon_audio_select_pin(encoder); + if (drm_detect_monitor_audio(radeon_connector_edid(connector))) { + radeon_audio_write_speaker_allocation(encoder); + radeon_audio_write_sad_regs(encoder); + radeon_audio_write_latency_fields(encoder, mode); + if (rdev->clock.dp_extclk || ASIC_IS_DCE5(rdev)) + radeon_audio_set_dto(encoder, rdev->clock.default_dispclk * 10); + else + radeon_audio_set_dto(encoder, dig_connector->dp_clock); + radeon_audio_set_audio_packet(encoder); + radeon_audio_select_pin(encoder); - if (radeon_audio_set_avi_packet(encoder, mode) < 0) - return; + if (radeon_audio_set_avi_packet(encoder, mode) < 0) + return; + } } void radeon_audio_mode_set(struct drm_encoder *encoder, -- cgit v0.10.2 From 0a90a0cff9f429f886f423967ae053150dce9259 Mon Sep 17 00:00:00 2001 From: Alex Deucher Date: Mon, 27 Jul 2015 19:24:31 -0400 Subject: drm/radeon/combios: add some validation of lvds values Fixes a broken hsync start value uncovered by: abc0b1447d4974963548777a5ba4a4457c82c426 (drm: Perform basic sanity checks on probed modes) The driver handled the bad hsync start elsewhere, but the above commit prevented it from getting added. bug: https://bugs.freedesktop.org/show_bug.cgi?id=91401 Signed-off-by: Alex Deucher Cc: stable@vger.kernel.org diff --git a/drivers/gpu/drm/radeon/radeon_combios.c b/drivers/gpu/drm/radeon/radeon_combios.c index 3e5f6b7..c097d3a 100644 --- a/drivers/gpu/drm/radeon/radeon_combios.c +++ b/drivers/gpu/drm/radeon/radeon_combios.c @@ -1255,10 +1255,15 @@ struct radeon_encoder_lvds *radeon_combios_get_lvds_info(struct radeon_encoder if ((RBIOS16(tmp) == lvds->native_mode.hdisplay) && (RBIOS16(tmp + 2) == lvds->native_mode.vdisplay)) { + u32 hss = (RBIOS16(tmp + 21) - RBIOS16(tmp + 19) - 1) * 8; + + if (hss > lvds->native_mode.hdisplay) + hss = (10 - 1) * 8; + lvds->native_mode.htotal = lvds->native_mode.hdisplay + (RBIOS16(tmp + 17) - RBIOS16(tmp + 19)) * 8; lvds->native_mode.hsync_start = lvds->native_mode.hdisplay + - (RBIOS16(tmp + 21) - RBIOS16(tmp + 19) - 1) * 8; + hss; lvds->native_mode.hsync_end = lvds->native_mode.hsync_start + (RBIOS8(tmp + 23) * 8); -- cgit v0.10.2 From 8faf0e08d5a78ae5f1752b1d69f97ed70afa625f Mon Sep 17 00:00:00 2001 From: Alex Deucher Date: Tue, 28 Jul 2015 11:50:31 -0400 Subject: drm/amdgpu: clean up init sequence for failures MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit If we fail during device init, record what state each block is in so that we can tear down clearly. Fixes various problems on device init failure. Reviewed-by: Christian König Signed-off-by: Alex Deucher diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu.h b/drivers/gpu/drm/amd/amdgpu/amdgpu.h index e9fde72..31b00f9 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu.h @@ -1866,6 +1866,12 @@ typedef void (*amdgpu_wreg_t)(struct amdgpu_device*, uint32_t, uint32_t); typedef uint32_t (*amdgpu_block_rreg_t)(struct amdgpu_device*, uint32_t, uint32_t); typedef void (*amdgpu_block_wreg_t)(struct amdgpu_device*, uint32_t, uint32_t, uint32_t); +struct amdgpu_ip_block_status { + bool valid; + bool sw; + bool hw; +}; + struct amdgpu_device { struct device *dev; struct drm_device *ddev; @@ -2008,7 +2014,7 @@ struct amdgpu_device { const struct amdgpu_ip_block_version *ip_blocks; int num_ip_blocks; - bool *ip_block_enabled; + struct amdgpu_ip_block_status *ip_block_status; struct mutex mn_lock; DECLARE_HASHTABLE(mn_hash, 7); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c index d79009b..99f158e 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c @@ -1191,8 +1191,9 @@ static int amdgpu_early_init(struct amdgpu_device *adev) return -EINVAL; } - adev->ip_block_enabled = kcalloc(adev->num_ip_blocks, sizeof(bool), GFP_KERNEL); - if (adev->ip_block_enabled == NULL) + adev->ip_block_status = kcalloc(adev->num_ip_blocks, + sizeof(struct amdgpu_ip_block_status), GFP_KERNEL); + if (adev->ip_block_status == NULL) return -ENOMEM; if (adev->ip_blocks == NULL) { @@ -1203,18 +1204,18 @@ static int amdgpu_early_init(struct amdgpu_device *adev) for (i = 0; i < adev->num_ip_blocks; i++) { if ((amdgpu_ip_block_mask & (1 << i)) == 0) { DRM_ERROR("disabled ip block: %d\n", i); - adev->ip_block_enabled[i] = false; + adev->ip_block_status[i].valid = false; } else { if (adev->ip_blocks[i].funcs->early_init) { r = adev->ip_blocks[i].funcs->early_init((void *)adev); if (r == -ENOENT) - adev->ip_block_enabled[i] = false; + adev->ip_block_status[i].valid = false; else if (r) return r; else - adev->ip_block_enabled[i] = true; + adev->ip_block_status[i].valid = true; } else { - adev->ip_block_enabled[i] = true; + adev->ip_block_status[i].valid = true; } } } @@ -1227,11 +1228,12 @@ static int amdgpu_init(struct amdgpu_device *adev) int i, r; for (i = 0; i < adev->num_ip_blocks; i++) { - if (!adev->ip_block_enabled[i]) + if (!adev->ip_block_status[i].valid) continue; r = adev->ip_blocks[i].funcs->sw_init((void *)adev); if (r) return r; + adev->ip_block_status[i].sw = true; /* need to do gmc hw init early so we can allocate gpu mem */ if (adev->ip_blocks[i].type == AMD_IP_BLOCK_TYPE_GMC) { r = amdgpu_vram_scratch_init(adev); @@ -1243,11 +1245,12 @@ static int amdgpu_init(struct amdgpu_device *adev) r = amdgpu_wb_init(adev); if (r) return r; + adev->ip_block_status[i].hw = true; } } for (i = 0; i < adev->num_ip_blocks; i++) { - if (!adev->ip_block_enabled[i]) + if (!adev->ip_block_status[i].sw) continue; /* gmc hw init is done early */ if (adev->ip_blocks[i].type == AMD_IP_BLOCK_TYPE_GMC) @@ -1255,6 +1258,7 @@ static int amdgpu_init(struct amdgpu_device *adev) r = adev->ip_blocks[i].funcs->hw_init((void *)adev); if (r) return r; + adev->ip_block_status[i].hw = true; } return 0; @@ -1265,7 +1269,7 @@ static int amdgpu_late_init(struct amdgpu_device *adev) int i = 0, r; for (i = 0; i < adev->num_ip_blocks; i++) { - if (!adev->ip_block_enabled[i]) + if (!adev->ip_block_status[i].valid) continue; /* enable clockgating to save power */ r = adev->ip_blocks[i].funcs->set_clockgating_state((void *)adev, @@ -1287,7 +1291,7 @@ static int amdgpu_fini(struct amdgpu_device *adev) int i, r; for (i = adev->num_ip_blocks - 1; i >= 0; i--) { - if (!adev->ip_block_enabled[i]) + if (!adev->ip_block_status[i].hw) continue; if (adev->ip_blocks[i].type == AMD_IP_BLOCK_TYPE_GMC) { amdgpu_wb_fini(adev); @@ -1300,14 +1304,16 @@ static int amdgpu_fini(struct amdgpu_device *adev) return r; r = adev->ip_blocks[i].funcs->hw_fini((void *)adev); /* XXX handle errors */ + adev->ip_block_status[i].hw = false; } for (i = adev->num_ip_blocks - 1; i >= 0; i--) { - if (!adev->ip_block_enabled[i]) + if (!adev->ip_block_status[i].sw) continue; r = adev->ip_blocks[i].funcs->sw_fini((void *)adev); /* XXX handle errors */ - adev->ip_block_enabled[i] = false; + adev->ip_block_status[i].sw = false; + adev->ip_block_status[i].valid = false; } return 0; @@ -1318,7 +1324,7 @@ static int amdgpu_suspend(struct amdgpu_device *adev) int i, r; for (i = adev->num_ip_blocks - 1; i >= 0; i--) { - if (!adev->ip_block_enabled[i]) + if (!adev->ip_block_status[i].valid) continue; /* ungate blocks so that suspend can properly shut them down */ r = adev->ip_blocks[i].funcs->set_clockgating_state((void *)adev, @@ -1336,7 +1342,7 @@ static int amdgpu_resume(struct amdgpu_device *adev) int i, r; for (i = 0; i < adev->num_ip_blocks; i++) { - if (!adev->ip_block_enabled[i]) + if (!adev->ip_block_status[i].valid) continue; r = adev->ip_blocks[i].funcs->resume(adev); if (r) @@ -1582,8 +1588,8 @@ void amdgpu_device_fini(struct amdgpu_device *adev) amdgpu_fence_driver_fini(adev); amdgpu_fbdev_fini(adev); r = amdgpu_fini(adev); - kfree(adev->ip_block_enabled); - adev->ip_block_enabled = NULL; + kfree(adev->ip_block_status); + adev->ip_block_status = NULL; adev->accel_working = false; /* free i2c buses */ amdgpu_i2c_fini(adev); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c index 31ad444..d316bda 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c @@ -235,7 +235,7 @@ static int amdgpu_info_ioctl(struct drm_device *dev, void *data, struct drm_file for (i = 0; i < adev->num_ip_blocks; i++) { if (adev->ip_blocks[i].type == type && - adev->ip_block_enabled[i]) { + adev->ip_block_status[i].valid) { ip.hw_ip_version_major = adev->ip_blocks[i].major; ip.hw_ip_version_minor = adev->ip_blocks[i].minor; ip.capabilities_flags = 0; @@ -274,7 +274,7 @@ static int amdgpu_info_ioctl(struct drm_device *dev, void *data, struct drm_file for (i = 0; i < adev->num_ip_blocks; i++) if (adev->ip_blocks[i].type == type && - adev->ip_block_enabled[i] && + adev->ip_block_status[i].valid && count < AMDGPU_HW_IP_INSTANCE_MAX_COUNT) count++; -- cgit v0.10.2 From c193fa91b9182465a4a01665ad4096a6cdb4db2d Mon Sep 17 00:00:00 2001 From: Dan Carpenter Date: Tue, 28 Jul 2015 18:51:29 +0300 Subject: drm/amdgpu: information leak in amdgpu_info_ioctl() We recently changed the drm_amdgpu_info_device struct so now there is a 4 byte hole at the end. We need to initialize it so we don't disclose secret information from the stack. Fixes: fa92754e9c47 ('drm/amdgpu: add VCE harvesting instance query') Signed-off-by: Dan Carpenter Signed-off-by: Alex Deucher diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c index d316bda..9736892 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c @@ -416,7 +416,7 @@ static int amdgpu_info_ioctl(struct drm_device *dev, void *data, struct drm_file return n ? -EFAULT : 0; } case AMDGPU_INFO_DEV_INFO: { - struct drm_amdgpu_info_device dev_info; + struct drm_amdgpu_info_device dev_info = {}; struct amdgpu_cu_info cu_info; dev_info.device_id = dev->pdev->device; -- cgit v0.10.2 From 93323131d66db68802e646204c0562cddc81a651 Mon Sep 17 00:00:00 2001 From: "monk.liu" Date: Wed, 15 Jul 2015 17:21:45 +0800 Subject: drm/amdgpu: different emit_ib for gfx and compute MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit compute ring didn't use const engine byfar, so ignore CE things in compute routine Signed-off-by: monk.liu Reviewed-by: Christian König diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c index 2c188fb..2db6ab0 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c @@ -2561,7 +2561,7 @@ static bool gfx_v7_0_ring_emit_semaphore(struct amdgpu_ring *ring, * sheduling on the ring. This function schedules the IB * on the gfx ring for execution by the GPU. */ -static void gfx_v7_0_ring_emit_ib(struct amdgpu_ring *ring, +static void gfx_v7_0_ring_emit_ib_gfx(struct amdgpu_ring *ring, struct amdgpu_ib *ib) { bool need_ctx_switch = ring->current_ctx != ib->ctx; @@ -2569,15 +2569,10 @@ static void gfx_v7_0_ring_emit_ib(struct amdgpu_ring *ring, u32 next_rptr = ring->wptr + 5; /* drop the CE preamble IB for the same context */ - if ((ring->type == AMDGPU_RING_TYPE_GFX) && - (ib->flags & AMDGPU_IB_FLAG_PREAMBLE) && - !need_ctx_switch) + if ((ib->flags & AMDGPU_IB_FLAG_PREAMBLE) && !need_ctx_switch) return; - if (ring->type == AMDGPU_RING_TYPE_COMPUTE) - control |= INDIRECT_BUFFER_VALID; - - if (need_ctx_switch && ring->type == AMDGPU_RING_TYPE_GFX) + if (need_ctx_switch) next_rptr += 2; next_rptr += 4; @@ -2588,7 +2583,7 @@ static void gfx_v7_0_ring_emit_ib(struct amdgpu_ring *ring, amdgpu_ring_write(ring, next_rptr); /* insert SWITCH_BUFFER packet before first IB in the ring frame */ - if (need_ctx_switch && ring->type == AMDGPU_RING_TYPE_GFX) { + if (need_ctx_switch) { amdgpu_ring_write(ring, PACKET3(PACKET3_SWITCH_BUFFER, 0)); amdgpu_ring_write(ring, 0); } @@ -2611,6 +2606,35 @@ static void gfx_v7_0_ring_emit_ib(struct amdgpu_ring *ring, amdgpu_ring_write(ring, control); } +static void gfx_v7_0_ring_emit_ib_compute(struct amdgpu_ring *ring, + struct amdgpu_ib *ib) +{ + u32 header, control = 0; + u32 next_rptr = ring->wptr + 5; + + control |= INDIRECT_BUFFER_VALID; + next_rptr += 4; + amdgpu_ring_write(ring, PACKET3(PACKET3_WRITE_DATA, 3)); + amdgpu_ring_write(ring, WRITE_DATA_DST_SEL(5) | WR_CONFIRM); + amdgpu_ring_write(ring, ring->next_rptr_gpu_addr & 0xfffffffc); + amdgpu_ring_write(ring, upper_32_bits(ring->next_rptr_gpu_addr) & 0xffffffff); + amdgpu_ring_write(ring, next_rptr); + + header = PACKET3(PACKET3_INDIRECT_BUFFER, 2); + + control |= ib->length_dw | + (ib->vm ? (ib->vm->ids[ring->idx].id << 24) : 0); + + amdgpu_ring_write(ring, header); + amdgpu_ring_write(ring, +#ifdef __BIG_ENDIAN + (2 << 0) | +#endif + (ib->gpu_addr & 0xFFFFFFFC)); + amdgpu_ring_write(ring, upper_32_bits(ib->gpu_addr) & 0xFFFF); + amdgpu_ring_write(ring, control); +} + /** * gfx_v7_0_ring_test_ib - basic ring IB test * @@ -5555,7 +5579,7 @@ static const struct amdgpu_ring_funcs gfx_v7_0_ring_funcs_gfx = { .get_wptr = gfx_v7_0_ring_get_wptr_gfx, .set_wptr = gfx_v7_0_ring_set_wptr_gfx, .parse_cs = NULL, - .emit_ib = gfx_v7_0_ring_emit_ib, + .emit_ib = gfx_v7_0_ring_emit_ib_gfx, .emit_fence = gfx_v7_0_ring_emit_fence_gfx, .emit_semaphore = gfx_v7_0_ring_emit_semaphore, .emit_vm_flush = gfx_v7_0_ring_emit_vm_flush, @@ -5571,7 +5595,7 @@ static const struct amdgpu_ring_funcs gfx_v7_0_ring_funcs_compute = { .get_wptr = gfx_v7_0_ring_get_wptr_compute, .set_wptr = gfx_v7_0_ring_set_wptr_compute, .parse_cs = NULL, - .emit_ib = gfx_v7_0_ring_emit_ib, + .emit_ib = gfx_v7_0_ring_emit_ib_compute, .emit_fence = gfx_v7_0_ring_emit_fence_compute, .emit_semaphore = gfx_v7_0_ring_emit_semaphore, .emit_vm_flush = gfx_v7_0_ring_emit_vm_flush, diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c index 1c7c992..9e1d4dd 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c @@ -3753,7 +3753,7 @@ static void gfx_v8_0_ring_emit_hdp_flush(struct amdgpu_ring *ring) amdgpu_ring_write(ring, 0x20); /* poll interval */ } -static void gfx_v8_0_ring_emit_ib(struct amdgpu_ring *ring, +static void gfx_v8_0_ring_emit_ib_gfx(struct amdgpu_ring *ring, struct amdgpu_ib *ib) { bool need_ctx_switch = ring->current_ctx != ib->ctx; @@ -3761,15 +3761,10 @@ static void gfx_v8_0_ring_emit_ib(struct amdgpu_ring *ring, u32 next_rptr = ring->wptr + 5; /* drop the CE preamble IB for the same context */ - if ((ring->type == AMDGPU_RING_TYPE_GFX) && - (ib->flags & AMDGPU_IB_FLAG_PREAMBLE) && - !need_ctx_switch) + if ((ib->flags & AMDGPU_IB_FLAG_PREAMBLE) && !need_ctx_switch) return; - if (ring->type == AMDGPU_RING_TYPE_COMPUTE) - control |= INDIRECT_BUFFER_VALID; - - if (need_ctx_switch && ring->type == AMDGPU_RING_TYPE_GFX) + if (need_ctx_switch) next_rptr += 2; next_rptr += 4; @@ -3780,7 +3775,7 @@ static void gfx_v8_0_ring_emit_ib(struct amdgpu_ring *ring, amdgpu_ring_write(ring, next_rptr); /* insert SWITCH_BUFFER packet before first IB in the ring frame */ - if (need_ctx_switch && ring->type == AMDGPU_RING_TYPE_GFX) { + if (need_ctx_switch) { amdgpu_ring_write(ring, PACKET3(PACKET3_SWITCH_BUFFER, 0)); amdgpu_ring_write(ring, 0); } @@ -3803,6 +3798,36 @@ static void gfx_v8_0_ring_emit_ib(struct amdgpu_ring *ring, amdgpu_ring_write(ring, control); } +static void gfx_v8_0_ring_emit_ib_compute(struct amdgpu_ring *ring, + struct amdgpu_ib *ib) +{ + u32 header, control = 0; + u32 next_rptr = ring->wptr + 5; + + control |= INDIRECT_BUFFER_VALID; + + next_rptr += 4; + amdgpu_ring_write(ring, PACKET3(PACKET3_WRITE_DATA, 3)); + amdgpu_ring_write(ring, WRITE_DATA_DST_SEL(5) | WR_CONFIRM); + amdgpu_ring_write(ring, ring->next_rptr_gpu_addr & 0xfffffffc); + amdgpu_ring_write(ring, upper_32_bits(ring->next_rptr_gpu_addr) & 0xffffffff); + amdgpu_ring_write(ring, next_rptr); + + header = PACKET3(PACKET3_INDIRECT_BUFFER, 2); + + control |= ib->length_dw | + (ib->vm ? (ib->vm->ids[ring->idx].id << 24) : 0); + + amdgpu_ring_write(ring, header); + amdgpu_ring_write(ring, +#ifdef __BIG_ENDIAN + (2 << 0) | +#endif + (ib->gpu_addr & 0xFFFFFFFC)); + amdgpu_ring_write(ring, upper_32_bits(ib->gpu_addr) & 0xFFFF); + amdgpu_ring_write(ring, control); +} + static void gfx_v8_0_ring_emit_fence_gfx(struct amdgpu_ring *ring, u64 addr, u64 seq, unsigned flags) { @@ -4224,7 +4249,7 @@ static const struct amdgpu_ring_funcs gfx_v8_0_ring_funcs_gfx = { .get_wptr = gfx_v8_0_ring_get_wptr_gfx, .set_wptr = gfx_v8_0_ring_set_wptr_gfx, .parse_cs = NULL, - .emit_ib = gfx_v8_0_ring_emit_ib, + .emit_ib = gfx_v8_0_ring_emit_ib_gfx, .emit_fence = gfx_v8_0_ring_emit_fence_gfx, .emit_semaphore = gfx_v8_0_ring_emit_semaphore, .emit_vm_flush = gfx_v8_0_ring_emit_vm_flush, @@ -4240,7 +4265,7 @@ static const struct amdgpu_ring_funcs gfx_v8_0_ring_funcs_compute = { .get_wptr = gfx_v8_0_ring_get_wptr_compute, .set_wptr = gfx_v8_0_ring_set_wptr_compute, .parse_cs = NULL, - .emit_ib = gfx_v8_0_ring_emit_ib, + .emit_ib = gfx_v8_0_ring_emit_ib_compute, .emit_fence = gfx_v8_0_ring_emit_fence_compute, .emit_semaphore = gfx_v8_0_ring_emit_semaphore, .emit_vm_flush = gfx_v8_0_ring_emit_vm_flush, -- cgit v0.10.2 From e722b71a540362eebdbae060430dc5b06b990c38 Mon Sep 17 00:00:00 2001 From: "monk.liu" Date: Fri, 17 Jul 2015 17:10:09 +0800 Subject: drm/amdgpu: hdp_flush is not needed for inside IB MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit hdp flush is not needed for IBs that dispatched from kernel inside because there is no video memory host access Signed-off-by: monk.liu Reviewed-by: Christian König diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ib.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ib.c index 52dff75..bc0fac6 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ib.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ib.c @@ -180,16 +180,16 @@ int amdgpu_ib_schedule(struct amdgpu_device *adev, unsigned num_ibs, if (vm) { /* do context switch */ amdgpu_vm_flush(ring, vm, ib->sync.last_vm_update); - } - if (vm && ring->funcs->emit_gds_switch) - amdgpu_ring_emit_gds_switch(ring, ib->vm->ids[ring->idx].id, - ib->gds_base, ib->gds_size, - ib->gws_base, ib->gws_size, - ib->oa_base, ib->oa_size); + if (ring->funcs->emit_gds_switch) + amdgpu_ring_emit_gds_switch(ring, ib->vm->ids[ring->idx].id, + ib->gds_base, ib->gds_size, + ib->gws_base, ib->gws_size, + ib->oa_base, ib->oa_size); - if (ring->funcs->emit_hdp_flush) - amdgpu_ring_emit_hdp_flush(ring); + if (ring->funcs->emit_hdp_flush) + amdgpu_ring_emit_hdp_flush(ring); + } old_ctx = ring->current_ctx; for (i = 0; i < num_ibs; ++i) { -- cgit v0.10.2 From 194a33643b1161fe7a054fa9bf43875ae0f6e1e8 Mon Sep 17 00:00:00 2001 From: "monk.liu" Date: Wed, 22 Jul 2015 13:29:28 +0800 Subject: drm/amdgpu: add new parameter to seperate map and unmap MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Signed-off-by: monk.liu Reviewed-by: Christian König diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c index ae43b58..4afc507 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c @@ -449,7 +449,7 @@ out: * vital here, so they are not reported back to userspace. */ static void amdgpu_gem_va_update_vm(struct amdgpu_device *adev, - struct amdgpu_bo_va *bo_va) + struct amdgpu_bo_va *bo_va, uint32_t operation) { struct ttm_validate_buffer tv, *entry; struct amdgpu_bo_list_entry *vm_bos; @@ -485,7 +485,9 @@ static void amdgpu_gem_va_update_vm(struct amdgpu_device *adev, if (r) goto error_unlock; - r = amdgpu_vm_bo_update(adev, bo_va, &bo_va->bo->tbo.mem); + + if (operation == AMDGPU_VA_OP_MAP) + r = amdgpu_vm_bo_update(adev, bo_va, &bo_va->bo->tbo.mem); error_unlock: mutex_unlock(&bo_va->vm->mutex); @@ -580,7 +582,7 @@ int amdgpu_gem_va_ioctl(struct drm_device *dev, void *data, } if (!r && !(args->flags & AMDGPU_VM_DELAY_UPDATE)) - amdgpu_gem_va_update_vm(adev, bo_va); + amdgpu_gem_va_update_vm(adev, bo_va, args->operation); drm_gem_object_unreference_unlocked(gobj); return r; -- cgit v0.10.2 From ac45146733b03a1c8e3a6a33720bdc42804cc09b Mon Sep 17 00:00:00 2001 From: Rob Clark Date: Thu, 23 Jul 2015 13:44:56 -0400 Subject: drm/msm: fix msm_gem_prime_get_sg_table() We need to return a new sgt, since the caller takes ownership of it. Reported-by: Stanimir Varbanov Signed-off-by: Rob Clark diff --git a/drivers/gpu/drm/msm/msm_gem_prime.c b/drivers/gpu/drm/msm/msm_gem_prime.c index dd7a7ab..831461b 100644 --- a/drivers/gpu/drm/msm/msm_gem_prime.c +++ b/drivers/gpu/drm/msm/msm_gem_prime.c @@ -23,8 +23,12 @@ struct sg_table *msm_gem_prime_get_sg_table(struct drm_gem_object *obj) { struct msm_gem_object *msm_obj = to_msm_bo(obj); - BUG_ON(!msm_obj->sgt); /* should have already pinned! */ - return msm_obj->sgt; + int npages = obj->size >> PAGE_SHIFT; + + if (WARN_ON(!msm_obj->pages)) /* should have already pinned! */ + return NULL; + + return drm_prime_pages_to_sg(msm_obj->pages, npages); } void *msm_gem_prime_vmap(struct drm_gem_object *obj) -- cgit v0.10.2 From a1c3e3e01ee301de6a13e696ef8775f40ca339ac Mon Sep 17 00:00:00 2001 From: Archit Taneja Date: Fri, 26 Jun 2015 15:49:43 +0530 Subject: drm/msm: mdp4: Fix drm_framebuffer dereference crash mdp4_get_frame_format() can dereference a drm_framebuffer when it's NULL. Call it in mdp4_plane_mode_set only when we know fb is non-NULL. Signed-off-by: Archit Taneja Signed-off-by: Rob Clark diff --git a/drivers/gpu/drm/msm/mdp/mdp4/mdp4_plane.c b/drivers/gpu/drm/msm/mdp/mdp4/mdp4_plane.c index 0d1dbb7..247a424 100644 --- a/drivers/gpu/drm/msm/mdp/mdp4/mdp4_plane.c +++ b/drivers/gpu/drm/msm/mdp/mdp4/mdp4_plane.c @@ -220,13 +220,15 @@ static int mdp4_plane_mode_set(struct drm_plane *plane, uint32_t op_mode = 0; uint32_t phasex_step = MDP4_VG_PHASE_STEP_DEFAULT; uint32_t phasey_step = MDP4_VG_PHASE_STEP_DEFAULT; - enum mdp4_frame_format frame_type = mdp4_get_frame_format(fb); + enum mdp4_frame_format frame_type; if (!(crtc && fb)) { DBG("%s: disabled!", mdp4_plane->name); return 0; } + frame_type = mdp4_get_frame_format(fb); + /* src values are in Q16 fixed point, convert to integer: */ src_x = src_x >> 16; src_y = src_y >> 16; -- cgit v0.10.2 From 99fc1bc48f352185f1711795f0829bbf503c0712 Mon Sep 17 00:00:00 2001 From: Wentao Xu Date: Mon, 22 Jun 2015 11:53:42 -0400 Subject: drm/msm: change to uninterruptible wait in atomic commit The atomic commit cannot easily undo and return an error once the state is swapped. Change to uninterruptible wait, and ignore the timeout error. Signed-off-by: Wentao Xu Signed-off-by: Rob Clark diff --git a/drivers/gpu/drm/msm/msm_atomic.c b/drivers/gpu/drm/msm/msm_atomic.c index 1b22d8b..1ceb4f2 100644 --- a/drivers/gpu/drm/msm/msm_atomic.c +++ b/drivers/gpu/drm/msm/msm_atomic.c @@ -283,12 +283,8 @@ int msm_atomic_commit(struct drm_device *dev, timeout = ktime_add_ms(ktime_get(), 1000); - ret = msm_wait_fence_interruptable(dev, c->fence, &timeout); - if (ret) { - WARN_ON(ret); // TODO unswap state back? or?? - commit_destroy(c); - return ret; - } + /* uninterruptible wait */ + msm_wait_fence(dev, c->fence, &timeout, false); complete_commit(c); diff --git a/drivers/gpu/drm/msm/msm_drv.c b/drivers/gpu/drm/msm/msm_drv.c index b7ef56e..d3467b1 100644 --- a/drivers/gpu/drm/msm/msm_drv.c +++ b/drivers/gpu/drm/msm/msm_drv.c @@ -637,8 +637,8 @@ static void msm_debugfs_cleanup(struct drm_minor *minor) * Fences: */ -int msm_wait_fence_interruptable(struct drm_device *dev, uint32_t fence, - ktime_t *timeout) +int msm_wait_fence(struct drm_device *dev, uint32_t fence, + ktime_t *timeout , bool interruptible) { struct msm_drm_private *priv = dev->dev_private; int ret; @@ -667,7 +667,12 @@ int msm_wait_fence_interruptable(struct drm_device *dev, uint32_t fence, remaining_jiffies = timespec_to_jiffies(&ts); } - ret = wait_event_interruptible_timeout(priv->fence_event, + if (interruptible) + ret = wait_event_interruptible_timeout(priv->fence_event, + fence_completed(dev, fence), + remaining_jiffies); + else + ret = wait_event_timeout(priv->fence_event, fence_completed(dev, fence), remaining_jiffies); @@ -853,7 +858,7 @@ static int msm_ioctl_wait_fence(struct drm_device *dev, void *data, return -EINVAL; } - return msm_wait_fence_interruptable(dev, args->fence, &timeout); + return msm_wait_fence(dev, args->fence, &timeout, true); } static const struct drm_ioctl_desc msm_ioctls[] = { diff --git a/drivers/gpu/drm/msm/msm_drv.h b/drivers/gpu/drm/msm/msm_drv.h index e7c5ea1..4ff0ec9 100644 --- a/drivers/gpu/drm/msm/msm_drv.h +++ b/drivers/gpu/drm/msm/msm_drv.h @@ -164,8 +164,8 @@ int msm_atomic_commit(struct drm_device *dev, int msm_register_mmu(struct drm_device *dev, struct msm_mmu *mmu); -int msm_wait_fence_interruptable(struct drm_device *dev, uint32_t fence, - ktime_t *timeout); +int msm_wait_fence(struct drm_device *dev, uint32_t fence, + ktime_t *timeout, bool interruptible); int msm_queue_fence_cb(struct drm_device *dev, struct msm_fence_cb *cb, uint32_t fence); void msm_update_fence(struct drm_device *dev, uint32_t fence); diff --git a/drivers/gpu/drm/msm/msm_gem.c b/drivers/gpu/drm/msm/msm_gem.c index f211b80..c76cc85 100644 --- a/drivers/gpu/drm/msm/msm_gem.c +++ b/drivers/gpu/drm/msm/msm_gem.c @@ -460,7 +460,7 @@ int msm_gem_cpu_prep(struct drm_gem_object *obj, uint32_t op, ktime_t *timeout) if (op & MSM_PREP_NOSYNC) timeout = NULL; - ret = msm_wait_fence_interruptable(dev, fence, timeout); + ret = msm_wait_fence(dev, fence, timeout, true); } /* TODO cache maintenance */ -- cgit v0.10.2 From b4cba04f05ed6b9b2278547295ecc5c40180e612 Mon Sep 17 00:00:00 2001 From: Wentao Xu Date: Fri, 19 Jun 2015 14:03:42 -0400 Subject: drm/msm/mdp5: release SMB (shared memory blocks) in various cases Release all blocks after the pipe is disabled, even when vsync didn't happen in some error cases. Allow requesting SMB multiple times before configuring to hardware, by releasing blocks not programmed to hardware yet for shrinking case. This fixes a potential leak of shared memory pool blocks. Signed-off-by: Wentao Xu Tested-by: Archit Taneja Signed-off-by: Rob Clark diff --git a/drivers/gpu/drm/msm/mdp/mdp5/mdp5_kms.c b/drivers/gpu/drm/msm/mdp/mdp5/mdp5_kms.c index 206f758..e253db5 100644 --- a/drivers/gpu/drm/msm/mdp/mdp5/mdp5_kms.c +++ b/drivers/gpu/drm/msm/mdp/mdp5/mdp5_kms.c @@ -76,7 +76,20 @@ static void mdp5_prepare_commit(struct msm_kms *kms, struct drm_atomic_state *st static void mdp5_complete_commit(struct msm_kms *kms, struct drm_atomic_state *state) { + int i; struct mdp5_kms *mdp5_kms = to_mdp5_kms(to_mdp_kms(kms)); + int nplanes = mdp5_kms->dev->mode_config.num_total_plane; + + for (i = 0; i < nplanes; i++) { + struct drm_plane *plane = state->planes[i]; + struct drm_plane_state *plane_state = state->plane_states[i]; + + if (!plane) + continue; + + mdp5_plane_complete_commit(plane, plane_state); + } + mdp5_disable(mdp5_kms); } diff --git a/drivers/gpu/drm/msm/mdp/mdp5/mdp5_kms.h b/drivers/gpu/drm/msm/mdp/mdp5/mdp5_kms.h index e0eb245..e79ac09 100644 --- a/drivers/gpu/drm/msm/mdp/mdp5/mdp5_kms.h +++ b/drivers/gpu/drm/msm/mdp/mdp5/mdp5_kms.h @@ -227,6 +227,8 @@ void mdp5_plane_install_properties(struct drm_plane *plane, struct drm_mode_object *obj); uint32_t mdp5_plane_get_flush(struct drm_plane *plane); void mdp5_plane_complete_flip(struct drm_plane *plane); +void mdp5_plane_complete_commit(struct drm_plane *plane, + struct drm_plane_state *state); enum mdp5_pipe mdp5_plane_pipe(struct drm_plane *plane); struct drm_plane *mdp5_plane_init(struct drm_device *dev, enum mdp5_pipe pipe, bool private_plane, uint32_t reg_offset); diff --git a/drivers/gpu/drm/msm/mdp/mdp5/mdp5_plane.c b/drivers/gpu/drm/msm/mdp/mdp5/mdp5_plane.c index 57b8f56..2227556 100644 --- a/drivers/gpu/drm/msm/mdp/mdp5/mdp5_plane.c +++ b/drivers/gpu/drm/msm/mdp/mdp5/mdp5_plane.c @@ -31,8 +31,6 @@ struct mdp5_plane { uint32_t nformats; uint32_t formats[32]; - - bool enabled; }; #define to_mdp5_plane(x) container_of(x, struct mdp5_plane, base) @@ -56,22 +54,6 @@ static bool plane_enabled(struct drm_plane_state *state) return state->fb && state->crtc; } -static int mdp5_plane_disable(struct drm_plane *plane) -{ - struct mdp5_plane *mdp5_plane = to_mdp5_plane(plane); - struct mdp5_kms *mdp5_kms = get_kms(plane); - enum mdp5_pipe pipe = mdp5_plane->pipe; - - DBG("%s: disable", mdp5_plane->name); - - if (mdp5_kms) { - /* Release the memory we requested earlier from the SMP: */ - mdp5_smp_release(mdp5_kms->smp, pipe); - } - - return 0; -} - static void mdp5_plane_destroy(struct drm_plane *plane) { struct mdp5_plane *mdp5_plane = to_mdp5_plane(plane); @@ -224,7 +206,6 @@ static void mdp5_plane_atomic_update(struct drm_plane *plane, if (!plane_enabled(state)) { to_mdp5_plane_state(state)->pending = true; - mdp5_plane_disable(plane); } else if (to_mdp5_plane_state(state)->mode_changed) { int ret; to_mdp5_plane_state(state)->pending = true; @@ -602,6 +583,20 @@ uint32_t mdp5_plane_get_flush(struct drm_plane *plane) return mdp5_plane->flush_mask; } +/* called after vsync in thread context */ +void mdp5_plane_complete_commit(struct drm_plane *plane, + struct drm_plane_state *state) +{ + struct mdp5_kms *mdp5_kms = get_kms(plane); + struct mdp5_plane *mdp5_plane = to_mdp5_plane(plane); + enum mdp5_pipe pipe = mdp5_plane->pipe; + + if (!plane_enabled(plane->state)) { + DBG("%s: free SMP", mdp5_plane->name); + mdp5_smp_release(mdp5_kms->smp, pipe); + } +} + /* initialize plane */ struct drm_plane *mdp5_plane_init(struct drm_device *dev, enum mdp5_pipe pipe, bool private_plane, uint32_t reg_offset) diff --git a/drivers/gpu/drm/msm/mdp/mdp5/mdp5_smp.c b/drivers/gpu/drm/msm/mdp/mdp5/mdp5_smp.c index 16702ae..64a27d8 100644 --- a/drivers/gpu/drm/msm/mdp/mdp5/mdp5_smp.c +++ b/drivers/gpu/drm/msm/mdp/mdp5/mdp5_smp.c @@ -34,22 +34,44 @@ * and CANNOT be re-allocated (eg: MMB0 and MMB1 both tied to RGB0). * * For each block that can be dynamically allocated, it can be either - * free, or pending/in-use by a client. The updates happen in three steps: + * free: + * The block is free. + * + * pending: + * The block is allocated to some client and not free. + * + * configured: + * The block is allocated to some client, and assigned to that + * client in MDP5_MDP_SMP_ALLOC registers. + * + * inuse: + * The block is being actively used by a client. + * + * The updates happen in the following steps: * * 1) mdp5_smp_request(): * When plane scanout is setup, calculate required number of - * blocks needed per client, and request. Blocks not inuse or - * pending by any other client are added to client's pending - * set. + * blocks needed per client, and request. Blocks neither inuse nor + * configured nor pending by any other client are added to client's + * pending set. + * For shrinking, blocks in pending but not in configured can be freed + * directly, but those already in configured will be freed later by + * mdp5_smp_commit. * * 2) mdp5_smp_configure(): * As hw is programmed, before FLUSH, MDP5_MDP_SMP_ALLOC registers * are configured for the union(pending, inuse) + * Current pending is copied to configured. + * It is assumed that mdp5_smp_request and mdp5_smp_configure not run + * concurrently for the same pipe. * * 3) mdp5_smp_commit(): - * After next vblank, copy pending -> inuse. Optionally update + * After next vblank, copy configured -> inuse. Optionally update * MDP5_SMP_ALLOC registers if there are newly unused blocks * + * 4) mdp5_smp_release(): + * Must be called after the pipe is disabled and no longer uses any SMB + * * On the next vblank after changes have been committed to hw, the * client's pending blocks become it's in-use blocks (and no-longer * in-use blocks become available to other clients). @@ -77,6 +99,9 @@ struct mdp5_smp { struct mdp5_client_smp_state client_state[MAX_CLIENTS]; }; +static void update_smp_state(struct mdp5_smp *smp, + u32 cid, mdp5_smp_state_t *assigned); + static inline struct mdp5_kms *get_kms(struct mdp5_smp *smp) { @@ -149,7 +174,12 @@ static int smp_request_block(struct mdp5_smp *smp, for (i = cur_nblks; i > nblks; i--) { int blk = find_first_bit(ps->pending, cnt); clear_bit(blk, ps->pending); - /* don't clear in global smp_state until _commit() */ + + /* clear in global smp_state if not in configured + * otherwise until _commit() + */ + if (!test_bit(blk, ps->configured)) + clear_bit(blk, smp->state); } } @@ -223,10 +253,33 @@ int mdp5_smp_request(struct mdp5_smp *smp, enum mdp5_pipe pipe, u32 fmt, u32 wid /* Release SMP blocks for all clients of the pipe */ void mdp5_smp_release(struct mdp5_smp *smp, enum mdp5_pipe pipe) { - int i, nblks; + int i; + unsigned long flags; + int cnt = smp->blk_cnt; + + for (i = 0; i < pipe2nclients(pipe); i++) { + mdp5_smp_state_t assigned; + u32 cid = pipe2client(pipe, i); + struct mdp5_client_smp_state *ps = &smp->client_state[cid]; + + spin_lock_irqsave(&smp->state_lock, flags); + + /* clear hw assignment */ + bitmap_or(assigned, ps->inuse, ps->configured, cnt); + update_smp_state(smp, CID_UNUSED, &assigned); + + /* free to global pool */ + bitmap_andnot(smp->state, smp->state, ps->pending, cnt); + bitmap_andnot(smp->state, smp->state, assigned, cnt); + + /* clear client's infor */ + bitmap_zero(ps->pending, cnt); + bitmap_zero(ps->configured, cnt); + bitmap_zero(ps->inuse, cnt); + + spin_unlock_irqrestore(&smp->state_lock, flags); + } - for (i = 0, nblks = 0; i < pipe2nclients(pipe); i++) - smp_request_block(smp, pipe2client(pipe, i), 0); set_fifo_thresholds(smp, pipe, 0); } @@ -274,12 +327,20 @@ void mdp5_smp_configure(struct mdp5_smp *smp, enum mdp5_pipe pipe) u32 cid = pipe2client(pipe, i); struct mdp5_client_smp_state *ps = &smp->client_state[cid]; - bitmap_or(assigned, ps->inuse, ps->pending, cnt); + /* + * if vblank has not happened since last smp_configure + * skip the configure for now + */ + if (!bitmap_equal(ps->inuse, ps->configured, cnt)) + continue; + + bitmap_copy(ps->configured, ps->pending, cnt); + bitmap_or(assigned, ps->inuse, ps->configured, cnt); update_smp_state(smp, cid, &assigned); } } -/* step #3: after vblank, copy pending -> inuse: */ +/* step #3: after vblank, copy configured -> inuse: */ void mdp5_smp_commit(struct mdp5_smp *smp, enum mdp5_pipe pipe) { int cnt = smp->blk_cnt; @@ -295,7 +356,7 @@ void mdp5_smp_commit(struct mdp5_smp *smp, enum mdp5_pipe pipe) * using, which can be released and made available to other * clients: */ - if (bitmap_andnot(released, ps->inuse, ps->pending, cnt)) { + if (bitmap_andnot(released, ps->inuse, ps->configured, cnt)) { unsigned long flags; spin_lock_irqsave(&smp->state_lock, flags); @@ -306,7 +367,7 @@ void mdp5_smp_commit(struct mdp5_smp *smp, enum mdp5_pipe pipe) update_smp_state(smp, CID_UNUSED, &released); } - bitmap_copy(ps->inuse, ps->pending, cnt); + bitmap_copy(ps->inuse, ps->configured, cnt); } } diff --git a/drivers/gpu/drm/msm/mdp/mdp5/mdp5_smp.h b/drivers/gpu/drm/msm/mdp/mdp5/mdp5_smp.h index e47179f..5b6c236 100644 --- a/drivers/gpu/drm/msm/mdp/mdp5/mdp5_smp.h +++ b/drivers/gpu/drm/msm/mdp/mdp5/mdp5_smp.h @@ -23,6 +23,7 @@ struct mdp5_client_smp_state { mdp5_smp_state_t inuse; + mdp5_smp_state_t configured; mdp5_smp_state_t pending; }; -- cgit v0.10.2 From 544586f742b43c6fd5fcb74c794d33b2ef189e64 Mon Sep 17 00:00:00 2001 From: Satish Ashok Date: Tue, 28 Jul 2015 03:28:27 -0700 Subject: bridge: mcast: give fast leave precedence over multicast router and querier When fast leave is configured on a bridge port and an IGMP leave is received for a group, the group is not deleted immediately if there is a router detected or if multicast querier is configured. Ideally the group should be deleted immediately when fast leave is configured. Signed-off-by: Satish Ashok Signed-off-by: David S. Miller diff --git a/net/bridge/br_multicast.c b/net/bridge/br_multicast.c index 79db489..0b39dcc 100644 --- a/net/bridge/br_multicast.c +++ b/net/bridge/br_multicast.c @@ -1416,8 +1416,7 @@ br_multicast_leave_group(struct net_bridge *br, spin_lock(&br->multicast_lock); if (!netif_running(br->dev) || - (port && port->state == BR_STATE_DISABLED) || - timer_pending(&other_query->timer)) + (port && port->state == BR_STATE_DISABLED)) goto out; mdb = mlock_dereference(br->mdb, br); @@ -1425,6 +1424,31 @@ br_multicast_leave_group(struct net_bridge *br, if (!mp) goto out; + if (port && (port->flags & BR_MULTICAST_FAST_LEAVE)) { + struct net_bridge_port_group __rcu **pp; + + for (pp = &mp->ports; + (p = mlock_dereference(*pp, br)) != NULL; + pp = &p->next) { + if (p->port != port) + continue; + + rcu_assign_pointer(*pp, p->next); + hlist_del_init(&p->mglist); + del_timer(&p->timer); + call_rcu_bh(&p->rcu, br_multicast_free_pg); + br_mdb_notify(br->dev, port, group, RTM_DELMDB); + + if (!mp->ports && !mp->mglist && + netif_running(br->dev)) + mod_timer(&mp->timer, jiffies); + } + goto out; + } + + if (timer_pending(&other_query->timer)) + goto out; + if (br->multicast_querier) { __br_multicast_send_query(br, port, &mp->addr); @@ -1450,28 +1474,6 @@ br_multicast_leave_group(struct net_bridge *br, } } - if (port && (port->flags & BR_MULTICAST_FAST_LEAVE)) { - struct net_bridge_port_group __rcu **pp; - - for (pp = &mp->ports; - (p = mlock_dereference(*pp, br)) != NULL; - pp = &p->next) { - if (p->port != port) - continue; - - rcu_assign_pointer(*pp, p->next); - hlist_del_init(&p->mglist); - del_timer(&p->timer); - call_rcu_bh(&p->rcu, br_multicast_free_pg); - br_mdb_notify(br->dev, port, group, RTM_DELMDB); - - if (!mp->ports && !mp->mglist && - netif_running(br->dev)) - mod_timer(&mp->timer, jiffies); - } - goto out; - } - now = jiffies; time = now + br->multicast_last_member_count * br->multicast_last_member_interval; -- cgit v0.10.2 From 7ae90a4f96486e3e20274afa1b8329802f5e1981 Mon Sep 17 00:00:00 2001 From: Nikolay Aleksandrov Date: Tue, 28 Jul 2015 13:10:44 +0200 Subject: bridge: mdb: fix delmdb state in the notification Since mdb states were introduced when deleting an entry the state was left as it was set in the delete request from the user which leads to the following output when doing a monitor (for example): $ bridge mdb add dev br0 port eth3 grp 239.0.0.1 permanent (monitor) dev br0 port eth3 grp 239.0.0.1 permanent $ bridge mdb del dev br0 port eth3 grp 239.0.0.1 permanent (monitor) dev br0 port eth3 grp 239.0.0.1 temp ^^^ Note the "temp" state in the delete notification which is wrong since the entry was permanent, the state in a delete is always reported as "temp" regardless of the real state of the entry. After this patch: $ bridge mdb add dev br0 port eth3 grp 239.0.0.1 permanent (monitor) dev br0 port eth3 grp 239.0.0.1 permanent $ bridge mdb del dev br0 port eth3 grp 239.0.0.1 permanent (monitor) dev br0 port eth3 grp 239.0.0.1 permanent There's one important note to make here that the state is actually not matched when doing a delete, so one can delete a permanent entry by stating "temp" in the end of the command, I've chosen this fix in order not to break user-space tools which rely on this (incorrect) behaviour. So to give an example after this patch and using the wrong state: $ bridge mdb add dev br0 port eth3 grp 239.0.0.1 permanent (monitor) dev br0 port eth3 grp 239.0.0.1 permanent $ bridge mdb del dev br0 port eth3 grp 239.0.0.1 temp (monitor) dev br0 port eth3 grp 239.0.0.1 permanent Note the state of the entry that got deleted is correct in the notification. Signed-off-by: Nikolay Aleksandrov Fixes: ccb1c31a7a87 ("bridge: add flags to distinguish permanent mdb entires") Signed-off-by: David S. Miller diff --git a/net/bridge/br_mdb.c b/net/bridge/br_mdb.c index 1198a3d..c943219 100644 --- a/net/bridge/br_mdb.c +++ b/net/bridge/br_mdb.c @@ -445,6 +445,7 @@ static int __br_mdb_del(struct net_bridge *br, struct br_mdb_entry *entry) if (p->port->state == BR_STATE_DISABLED) goto unlock; + entry->state = p->state; rcu_assign_pointer(*pp, p->next); hlist_del_init(&p->mglist); del_timer(&p->timer); -- cgit v0.10.2 From 2482abb93ebf7bfbf85965ca907f0058ff968c59 Mon Sep 17 00:00:00 2001 From: Daniel Borkmann Date: Tue, 28 Jul 2015 15:26:36 +0200 Subject: ebpf, x86: fix general protection fault when tail call is invoked With eBPF JIT compiler enabled on x86_64, I was able to reliably trigger the following general protection fault out of an eBPF program with a simple tail call, f.e. tracex5 (or a stripped down version of it): [ 927.097918] general protection fault: 0000 [#1] SMP DEBUG_PAGEALLOC [...] [ 927.100870] task: ffff8801f228b780 ti: ffff880016a64000 task.ti: ffff880016a64000 [ 927.102096] RIP: 0010:[] [] 0xffffffffa002440d [ 927.103390] RSP: 0018:ffff880016a67a68 EFLAGS: 00010006 [ 927.104683] RAX: 5a5a5a5a5a5a5a5a RBX: 0000000000000000 RCX: 0000000000000001 [ 927.105921] RDX: 0000000000000000 RSI: ffff88014e438000 RDI: ffff880016a67e00 [ 927.107137] RBP: ffff880016a67c90 R08: 0000000000000000 R09: 0000000000000001 [ 927.108351] R10: 0000000000000000 R11: 0000000000000000 R12: ffff880016a67e00 [ 927.109567] R13: 0000000000000000 R14: ffff88026500e460 R15: ffff880220a81520 [ 927.110787] FS: 00007fe7d5c1f740(0000) GS:ffff880265000000(0000) knlGS:0000000000000000 [ 927.112021] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 [ 927.113255] CR2: 0000003e7bbb91a0 CR3: 000000006e04b000 CR4: 00000000001407e0 [ 927.114500] Stack: [ 927.115737] ffffc90008cdb000 ffff880016a67e00 ffff88026500e460 ffff880220a81520 [ 927.117005] 0000000100000000 000000000000001b ffff880016a67aa8 ffffffff8106c548 [ 927.118276] 00007ffcdaf22e58 0000000000000000 0000000000000000 ffff880016a67ff0 [ 927.119543] Call Trace: [ 927.120797] [] ? lookup_address+0x28/0x30 [ 927.122058] [] ? __module_text_address+0x16/0x70 [ 927.123314] [] ? is_ftrace_trampoline+0x3e/0x70 [ 927.124562] [] ? __kernel_text_address+0x5f/0x80 [ 927.125806] [] ? print_context_stack+0x7f/0xf0 [ 927.127033] [] ? __lock_acquire+0x572/0x2050 [ 927.128254] [] ? __lock_acquire+0x572/0x2050 [ 927.129461] [] ? trace_call_bpf+0x3a/0x140 [ 927.130654] [] trace_call_bpf+0x8a/0x140 [ 927.131837] [] ? trace_call_bpf+0x3a/0x140 [ 927.133015] [] kprobe_perf_func+0x28/0x220 [ 927.134195] [] kprobe_dispatcher+0x38/0x60 [ 927.135367] [] ? seccomp_phase1+0x1/0x230 [ 927.136523] [] kprobe_ftrace_handler+0xf0/0x150 [ 927.137666] [] ? seccomp_phase1+0x5/0x230 [ 927.138802] [] ftrace_ops_recurs_func+0x5c/0xb0 [ 927.139934] [] 0xffffffffa022b0d5 [ 927.141066] [] ? seccomp_phase1+0x1/0x230 [ 927.142199] [] seccomp_phase1+0x5/0x230 [ 927.143323] [] syscall_trace_enter_phase1+0xc4/0x150 [ 927.144450] [] ? seccomp_phase1+0x5/0x230 [ 927.145572] [] ? syscall_trace_enter_phase1+0xc4/0x150 [ 927.146666] [] tracesys+0xd/0x44 [ 927.147723] Code: 48 8b 46 10 48 39 d0 76 2c 8b 85 fc fd ff ff 83 f8 20 77 21 83 c0 01 89 85 fc fd ff ff 48 8d 44 d6 80 48 8b 00 48 83 f8 00 74 0a <48> 8b 40 20 48 83 c0 33 ff e0 48 89 d8 48 8b 9d d8 fd ff ff 4c [ 927.150046] RIP [] 0xffffffffa002440d The code section with the instructions that traps points into the eBPF JIT image of the root program (the one invoking the tail call instruction). Using bpf_jit_disasm -o on the eBPF root program image: [...] 4e: mov -0x204(%rbp),%eax 8b 85 fc fd ff ff 54: cmp $0x20,%eax <--- if (tail_call_cnt > MAX_TAIL_CALL_CNT) 83 f8 20 57: ja 0x000000000000007a 77 21 59: add $0x1,%eax <--- tail_call_cnt++ 83 c0 01 5c: mov %eax,-0x204(%rbp) 89 85 fc fd ff ff 62: lea -0x80(%rsi,%rdx,8),%rax <--- prog = array->prog[index] 48 8d 44 d6 80 67: mov (%rax),%rax 48 8b 00 6a: cmp $0x0,%rax <--- check for NULL 48 83 f8 00 6e: je 0x000000000000007a 74 0a 70: mov 0x20(%rax),%rax <--- GPF triggered here! fetch of bpf_func 48 8b 40 20 [ matches <48> 8b 40 20 ... from above ] 74: add $0x33,%rax <--- prologue skip of new prog 48 83 c0 33 78: jmpq *%rax <--- jump to new prog insns ff e0 [...] The problem is that rax has 5a5a5a5a5a5a5a5a, which suggests a tail call jump to map slot 0 is pointing to a poisoned page. The issue is the following: lea instruction has a wrong offset, i.e. it should be ... lea 0x80(%rsi,%rdx,8),%rax ... but it actually seems to be ... lea -0x80(%rsi,%rdx,8),%rax ... where 0x80 is offsetof(struct bpf_array, prog), thus the offset needs to be positive instead of negative. Disassembling the interpreter, we btw similarly do: [...] c88: lea 0x80(%rax,%rdx,8),%rax <--- prog = array->prog[index] 48 8d 84 d0 80 00 00 00 c90: add $0x1,%r13d 41 83 c5 01 c94: mov (%rax),%rax 48 8b 00 [...] Now the other interesting fact is that this panic triggers only when things like CONFIG_LOCKDEP are being used. In that case offsetof(struct bpf_array, prog) starts at offset 0x80 and in non-CONFIG_LOCKDEP case at offset 0x50. Reason is that the work_struct inside struct bpf_map grows by 48 bytes in my case due to the lockdep_map member (which also has CONFIG_LOCK_STAT enabled members). Changing the emitter to always use the 4 byte displacement in the lea instruction fixes the panic on my side. It increases the tail call instruction emission by 3 more byte, but it should cover us from various combinations (and perhaps other future increases on related structures). After patch, disassembly: [...] 9e: lea 0x80(%rsi,%rdx,8),%rax <--- CONFIG_LOCKDEP/CONFIG_LOCK_STAT 48 8d 84 d6 80 00 00 00 a6: mov (%rax),%rax 48 8b 00 [...] [...] 9e: lea 0x50(%rsi,%rdx,8),%rax <--- No CONFIG_LOCKDEP 48 8d 84 d6 50 00 00 00 a6: mov (%rax),%rax 48 8b 00 [...] Fixes: b52f00e6a715 ("x86: bpf_jit: implement bpf_tail_call() helper") Signed-off-by: Daniel Borkmann Acked-by: Alexei Starovoitov Signed-off-by: David S. Miller diff --git a/arch/x86/net/bpf_jit_comp.c b/arch/x86/net/bpf_jit_comp.c index 579a8fd..be2e7a2 100644 --- a/arch/x86/net/bpf_jit_comp.c +++ b/arch/x86/net/bpf_jit_comp.c @@ -269,7 +269,7 @@ static void emit_bpf_tail_call(u8 **pprog) EMIT4(0x48, 0x8B, 0x46, /* mov rax, qword ptr [rsi + 16] */ offsetof(struct bpf_array, map.max_entries)); EMIT3(0x48, 0x39, 0xD0); /* cmp rax, rdx */ -#define OFFSET1 44 /* number of bytes to jump */ +#define OFFSET1 47 /* number of bytes to jump */ EMIT2(X86_JBE, OFFSET1); /* jbe out */ label1 = cnt; @@ -278,15 +278,15 @@ static void emit_bpf_tail_call(u8 **pprog) */ EMIT2_off32(0x8B, 0x85, -STACKSIZE + 36); /* mov eax, dword ptr [rbp - 516] */ EMIT3(0x83, 0xF8, MAX_TAIL_CALL_CNT); /* cmp eax, MAX_TAIL_CALL_CNT */ -#define OFFSET2 33 +#define OFFSET2 36 EMIT2(X86_JA, OFFSET2); /* ja out */ label2 = cnt; EMIT3(0x83, 0xC0, 0x01); /* add eax, 1 */ EMIT2_off32(0x89, 0x85, -STACKSIZE + 36); /* mov dword ptr [rbp - 516], eax */ /* prog = array->prog[index]; */ - EMIT4(0x48, 0x8D, 0x44, 0xD6); /* lea rax, [rsi + rdx * 8 + 0x50] */ - EMIT1(offsetof(struct bpf_array, prog)); + EMIT4_off32(0x48, 0x8D, 0x84, 0xD6, /* lea rax, [rsi + rdx * 8 + offsetof(...)] */ + offsetof(struct bpf_array, prog)); EMIT3(0x48, 0x8B, 0x00); /* mov rax, qword ptr [rax] */ /* if (prog == NULL) -- cgit v0.10.2 From 01a030996e1e2c268a4a484e2cbd3722b705414d Mon Sep 17 00:00:00 2001 From: "Karicheri, Muralidharan" Date: Tue, 28 Jul 2015 18:20:12 -0400 Subject: net: netcp: fix cleanup interface list in netcp_remove() Currently if user do rmmod keystone_netcp.ko following warning is seen :- [ 59.035891] ------------[ cut here ]------------ [ 59.040535] WARNING: CPU: 2 PID: 1619 at drivers/net/ethernet/ti/ netcp_core.c:2127 netcp_remove) This is because the interface list is not cleaned up in netcp_remove. This patch fixes this. Also fix some checkpatch related warnings. Signed-off-by: Murali Karicheri Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/ti/netcp_core.c b/drivers/net/ethernet/ti/netcp_core.c index ec8ed30..a1c6961 100644 --- a/drivers/net/ethernet/ti/netcp_core.c +++ b/drivers/net/ethernet/ti/netcp_core.c @@ -2112,6 +2112,7 @@ probe_quit: static int netcp_remove(struct platform_device *pdev) { struct netcp_device *netcp_device = platform_get_drvdata(pdev); + struct netcp_intf *netcp_intf, *netcp_tmp; struct netcp_inst_modpriv *inst_modpriv, *tmp; struct netcp_module *module; @@ -2123,8 +2124,16 @@ static int netcp_remove(struct platform_device *pdev) list_del(&inst_modpriv->inst_list); kfree(inst_modpriv); } - WARN(!list_empty(&netcp_device->interface_head), "%s interface list not empty!\n", - pdev->name); + + /* now that all modules are removed, clean up the interfaces */ + list_for_each_entry_safe(netcp_intf, netcp_tmp, + &netcp_device->interface_head, + interface_list) { + netcp_delete_interface(netcp_device, netcp_intf->ndev); + } + + WARN(!list_empty(&netcp_device->interface_head), + "%s interface list not empty!\n", pdev->name); devm_kfree(&pdev->dev, netcp_device); pm_runtime_put_sync(&pdev->dev); -- cgit v0.10.2 From c20afae75c32ec69eccc4ee432fa27d426e85a83 Mon Sep 17 00:00:00 2001 From: "Karicheri, Muralidharan" Date: Tue, 28 Jul 2015 18:20:13 -0400 Subject: net: netcp: ethss: fix up incorrect use of list api The code seems to assume a null is returned when the list is empty from first_sec_slave() to break the loop which is incorrect. Fix the code by using list_empty(). Signed-off-by: Murali Karicheri Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/ti/netcp_ethss.c b/drivers/net/ethernet/ti/netcp_ethss.c index a218812..d11d617 100644 --- a/drivers/net/ethernet/ti/netcp_ethss.c +++ b/drivers/net/ethernet/ti/netcp_ethss.c @@ -2508,10 +2508,9 @@ static void free_secondary_ports(struct gbe_priv *gbe_dev) { struct gbe_slave *slave; - for (;;) { + while (!list_empty(&gbe_dev->secondary_slaves)) { slave = first_sec_slave(gbe_dev); - if (!slave) - break; + if (slave->phy) phy_disconnect(slave->phy); list_del(&slave->slave_list); -- cgit v0.10.2 From 31a184b7acbc06d894c562ef884a94d6d78d0236 Mon Sep 17 00:00:00 2001 From: "Karicheri, Muralidharan" Date: Tue, 28 Jul 2015 18:20:14 -0400 Subject: net: netcp: ethss: cleanup gbe_probe() and gbe_remove() functions This patch clean up error handle code to use goto label properly. In some cases, the code unnecessarily use goto instead of just returning the error code. Code also make explicit calls to devm_* APIs on error which is not necessary. In the gbe_remove() also it makes similar calls which is also unnecessary. Also fix few checkpatch warnings Signed-off-by: Murali Karicheri Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/ti/netcp_core.c b/drivers/net/ethernet/ti/netcp_core.c index a1c6961..9749dfd 100644 --- a/drivers/net/ethernet/ti/netcp_core.c +++ b/drivers/net/ethernet/ti/netcp_core.c @@ -2135,7 +2135,6 @@ static int netcp_remove(struct platform_device *pdev) WARN(!list_empty(&netcp_device->interface_head), "%s interface list not empty!\n", pdev->name); - devm_kfree(&pdev->dev, netcp_device); pm_runtime_put_sync(&pdev->dev); pm_runtime_disable(&pdev->dev); platform_set_drvdata(pdev, NULL); diff --git a/drivers/net/ethernet/ti/netcp_ethss.c b/drivers/net/ethernet/ti/netcp_ethss.c index d11d617..1974a8a 100644 --- a/drivers/net/ethernet/ti/netcp_ethss.c +++ b/drivers/net/ethernet/ti/netcp_ethss.c @@ -2856,14 +2856,13 @@ static int gbe_probe(struct netcp_device *netcp_device, struct device *dev, &gbe_dev->dma_chan_name); if (ret < 0) { dev_err(dev, "missing \"tx-channel\" parameter\n"); - ret = -ENODEV; - goto quit; + return -EINVAL; } if (!strcmp(node->name, "gbe")) { ret = get_gbe_resource_version(gbe_dev, node); if (ret) - goto quit; + return ret; dev_dbg(dev, "ss_version: 0x%08x\n", gbe_dev->ss_version); @@ -2874,22 +2873,20 @@ static int gbe_probe(struct netcp_device *netcp_device, struct device *dev, else ret = -ENODEV; - if (ret) - goto quit; } else if (!strcmp(node->name, "xgbe")) { ret = set_xgbe_ethss10_priv(gbe_dev, node); if (ret) - goto quit; + return ret; ret = netcp_xgbe_serdes_init(gbe_dev->xgbe_serdes_regs, gbe_dev->ss_regs); - if (ret) - goto quit; } else { dev_err(dev, "unknown GBE node(%s)\n", node->name); ret = -ENODEV; - goto quit; } + if (ret) + return ret; + interfaces = of_get_child_by_name(node, "interfaces"); if (!interfaces) dev_err(dev, "could not find interfaces\n"); @@ -2897,11 +2894,11 @@ static int gbe_probe(struct netcp_device *netcp_device, struct device *dev, ret = netcp_txpipe_init(&gbe_dev->tx_pipe, netcp_device, gbe_dev->dma_chan_name, gbe_dev->tx_queue_id); if (ret) - goto quit; + return ret; ret = netcp_txpipe_open(&gbe_dev->tx_pipe); if (ret) - goto quit; + return ret; /* Create network interfaces */ INIT_LIST_HEAD(&gbe_dev->gbe_intf_head); @@ -2916,6 +2913,7 @@ static int gbe_probe(struct netcp_device *netcp_device, struct device *dev, if (gbe_dev->num_slaves >= gbe_dev->max_num_slaves) break; } + of_node_put(interfaces); if (!gbe_dev->num_slaves) dev_warn(dev, "No network interface configured\n"); @@ -2928,9 +2926,10 @@ static int gbe_probe(struct netcp_device *netcp_device, struct device *dev, of_node_put(secondary_ports); if (!gbe_dev->num_slaves) { - dev_err(dev, "No network interface or secondary ports configured\n"); + dev_err(dev, + "No network interface or secondary ports configured\n"); ret = -ENODEV; - goto quit; + goto free_sec_ports; } memset(&ale_params, 0, sizeof(ale_params)); @@ -2944,7 +2943,7 @@ static int gbe_probe(struct netcp_device *netcp_device, struct device *dev, if (!gbe_dev->ale) { dev_err(gbe_dev->dev, "error initializing ale engine\n"); ret = -ENODEV; - goto quit; + goto free_sec_ports; } else { dev_dbg(gbe_dev->dev, "Created a gbe ale engine\n"); } @@ -2960,14 +2959,8 @@ static int gbe_probe(struct netcp_device *netcp_device, struct device *dev, *inst_priv = gbe_dev; return 0; -quit: - if (gbe_dev->hw_stats) - devm_kfree(dev, gbe_dev->hw_stats); - cpsw_ale_destroy(gbe_dev->ale); - if (gbe_dev->ss_regs) - devm_iounmap(dev, gbe_dev->ss_regs); - of_node_put(interfaces); - devm_kfree(dev, gbe_dev); +free_sec_ports: + free_secondary_ports(gbe_dev); return ret; } @@ -3040,12 +3033,9 @@ static int gbe_remove(struct netcp_device *netcp_device, void *inst_priv) free_secondary_ports(gbe_dev); if (!list_empty(&gbe_dev->gbe_intf_head)) - dev_alert(gbe_dev->dev, "unreleased ethss interfaces present\n"); + dev_alert(gbe_dev->dev, + "unreleased ethss interfaces present\n"); - devm_kfree(gbe_dev->dev, gbe_dev->hw_stats); - devm_iounmap(gbe_dev->dev, gbe_dev->ss_regs); - memset(gbe_dev, 0x00, sizeof(*gbe_dev)); - devm_kfree(gbe_dev->dev, gbe_dev); return 0; } -- cgit v0.10.2 From 0d6aaffc3a6db642e0a165ba4d17d6d7bbaf5201 Mon Sep 17 00:00:00 2001 From: Guenter Roeck Date: Mon, 27 Jul 2015 10:21:46 -0700 Subject: hwmon: (nct7904) Rename pwm attributes to match hwmon ABI pwm attributes have well defined names, which should be used. Cc: Vadim V. Vlasov Cc: stable@vger.kernel.org #v4.1+ Signed-off-by: Guenter Roeck diff --git a/Documentation/hwmon/nct7904 b/Documentation/hwmon/nct7904 index 014f112..57fffe3 100644 --- a/Documentation/hwmon/nct7904 +++ b/Documentation/hwmon/nct7904 @@ -35,11 +35,11 @@ temp1_input Local temperature (1/1000 degree, temp[2-9]_input CPU temperatures (1/1000 degree, 0.125 degree resolution) -fan[1-4]_mode R/W, 0/1 for manual or SmartFan mode +pwm[1-4]_enable R/W, 1/2 for manual or SmartFan mode Setting SmartFan mode is supported only if it has been previously configured by BIOS (or configuration EEPROM) -fan[1-4]_pwm R/O in SmartFan mode, R/W in manual control mode +pwm[1-4] R/O in SmartFan mode, R/W in manual control mode The driver checks sensor control registers and does not export the sensors that are not enabled. Anyway, a sensor that is enabled may actually be not diff --git a/drivers/hwmon/nct7904.c b/drivers/hwmon/nct7904.c index b77b82f..6153df73 100644 --- a/drivers/hwmon/nct7904.c +++ b/drivers/hwmon/nct7904.c @@ -412,8 +412,9 @@ static ssize_t show_pwm(struct device *dev, return sprintf(buf, "%d\n", val); } -static ssize_t store_mode(struct device *dev, struct device_attribute *devattr, - const char *buf, size_t count) +static ssize_t store_enable(struct device *dev, + struct device_attribute *devattr, + const char *buf, size_t count) { int index = to_sensor_dev_attr(devattr)->index; struct nct7904_data *data = dev_get_drvdata(dev); @@ -422,18 +423,18 @@ static ssize_t store_mode(struct device *dev, struct device_attribute *devattr, if (kstrtoul(buf, 10, &val) < 0) return -EINVAL; - if (val > 1 || (val && !data->fan_mode[index])) + if (val < 1 || val > 2 || (val == 2 && !data->fan_mode[index])) return -EINVAL; ret = nct7904_write_reg(data, BANK_3, FANCTL1_FMR_REG + index, - val ? data->fan_mode[index] : 0); + val == 2 ? data->fan_mode[index] : 0); return ret ? ret : count; } -/* Return 0 for manual mode or 1 for SmartFan mode */ -static ssize_t show_mode(struct device *dev, - struct device_attribute *devattr, char *buf) +/* Return 1 for manual mode or 2 for SmartFan mode */ +static ssize_t show_enable(struct device *dev, + struct device_attribute *devattr, char *buf) { int index = to_sensor_dev_attr(devattr)->index; struct nct7904_data *data = dev_get_drvdata(dev); @@ -443,36 +444,36 @@ static ssize_t show_mode(struct device *dev, if (val < 0) return val; - return sprintf(buf, "%d\n", val ? 1 : 0); + return sprintf(buf, "%d\n", val ? 2 : 1); } /* 2 attributes per channel: pwm and mode */ -static SENSOR_DEVICE_ATTR(fan1_pwm, S_IRUGO | S_IWUSR, +static SENSOR_DEVICE_ATTR(pwm1, S_IRUGO | S_IWUSR, show_pwm, store_pwm, 0); -static SENSOR_DEVICE_ATTR(fan1_mode, S_IRUGO | S_IWUSR, - show_mode, store_mode, 0); -static SENSOR_DEVICE_ATTR(fan2_pwm, S_IRUGO | S_IWUSR, +static SENSOR_DEVICE_ATTR(pwm1_enable, S_IRUGO | S_IWUSR, + show_enable, store_enable, 0); +static SENSOR_DEVICE_ATTR(pwm2, S_IRUGO | S_IWUSR, show_pwm, store_pwm, 1); -static SENSOR_DEVICE_ATTR(fan2_mode, S_IRUGO | S_IWUSR, - show_mode, store_mode, 1); -static SENSOR_DEVICE_ATTR(fan3_pwm, S_IRUGO | S_IWUSR, +static SENSOR_DEVICE_ATTR(pwm2_enable, S_IRUGO | S_IWUSR, + show_enable, store_enable, 1); +static SENSOR_DEVICE_ATTR(pwm3, S_IRUGO | S_IWUSR, show_pwm, store_pwm, 2); -static SENSOR_DEVICE_ATTR(fan3_mode, S_IRUGO | S_IWUSR, - show_mode, store_mode, 2); -static SENSOR_DEVICE_ATTR(fan4_pwm, S_IRUGO | S_IWUSR, +static SENSOR_DEVICE_ATTR(pwm3_enable, S_IRUGO | S_IWUSR, + show_enable, store_enable, 2); +static SENSOR_DEVICE_ATTR(pwm4, S_IRUGO | S_IWUSR, show_pwm, store_pwm, 3); -static SENSOR_DEVICE_ATTR(fan4_mode, S_IRUGO | S_IWUSR, - show_mode, store_mode, 3); +static SENSOR_DEVICE_ATTR(pwm4_enable, S_IRUGO | S_IWUSR, + show_enable, store_enable, 3); static struct attribute *nct7904_fanctl_attrs[] = { - &sensor_dev_attr_fan1_pwm.dev_attr.attr, - &sensor_dev_attr_fan1_mode.dev_attr.attr, - &sensor_dev_attr_fan2_pwm.dev_attr.attr, - &sensor_dev_attr_fan2_mode.dev_attr.attr, - &sensor_dev_attr_fan3_pwm.dev_attr.attr, - &sensor_dev_attr_fan3_mode.dev_attr.attr, - &sensor_dev_attr_fan4_pwm.dev_attr.attr, - &sensor_dev_attr_fan4_mode.dev_attr.attr, + &sensor_dev_attr_pwm1.dev_attr.attr, + &sensor_dev_attr_pwm1_enable.dev_attr.attr, + &sensor_dev_attr_pwm2.dev_attr.attr, + &sensor_dev_attr_pwm2_enable.dev_attr.attr, + &sensor_dev_attr_pwm3.dev_attr.attr, + &sensor_dev_attr_pwm3_enable.dev_attr.attr, + &sensor_dev_attr_pwm4.dev_attr.attr, + &sensor_dev_attr_pwm4_enable.dev_attr.attr, NULL }; -- cgit v0.10.2 From 9200bc4c28cd8992eb5379345abd6b4f0c93df16 Mon Sep 17 00:00:00 2001 From: Guenter Roeck Date: Sat, 4 Jul 2015 13:23:42 -0700 Subject: hwmon: (nct7802) Fix integer overflow seen when writing voltage limits Writing a large value into a voltage limit attribute can result in an overflow due to an auto-conversion from unsigned long to unsigned int. Cc: Constantine Shulyupin Reviewed-by: Jean Delvare Cc: stable@vger.kernel.org # v4.1+ Signed-off-by: Guenter Roeck diff --git a/drivers/hwmon/nct7802.c b/drivers/hwmon/nct7802.c index 28fcb2e..fbfc02b 100644 --- a/drivers/hwmon/nct7802.c +++ b/drivers/hwmon/nct7802.c @@ -195,7 +195,7 @@ abort: } static int nct7802_write_voltage(struct nct7802_data *data, int nr, int index, - unsigned int voltage) + unsigned long voltage) { int shift = 8 - REG_VOLTAGE_LIMIT_MSB_SHIFT[index - 1][nr]; int err; -- cgit v0.10.2 From c8507fb235bea3314a02a67ddda0d4e6cf01fa78 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Wed, 29 Jul 2015 12:01:41 +0200 Subject: ipv6: flush nd cache on IFF_NOARP change This patch is the IPv6 equivalent of commit 6c8b4e3ff81b ("arp: flush arp cache on IFF_NOARP change") Without it, we keep buggy neighbours in the cache, with destination MAC address equal to our own MAC address. Tested: tcpdump -i eth0 -s 0 ip6 -n -e & ip link set dev eth0 arp off ping6 remote // sends buggy frames ip link set dev eth0 arp on ping6 remote // should work once kernel is patched Signed-off-by: Eric Dumazet Reported-by: Mario Fanelli Signed-off-by: David S. Miller diff --git a/net/ipv6/ndisc.c b/net/ipv6/ndisc.c index 0a05b35..c53331c 100644 --- a/net/ipv6/ndisc.c +++ b/net/ipv6/ndisc.c @@ -1650,6 +1650,7 @@ int ndisc_rcv(struct sk_buff *skb) static int ndisc_netdev_event(struct notifier_block *this, unsigned long event, void *ptr) { struct net_device *dev = netdev_notifier_info_to_dev(ptr); + struct netdev_notifier_change_info *change_info; struct net *net = dev_net(dev); struct inet6_dev *idev; @@ -1664,6 +1665,11 @@ static int ndisc_netdev_event(struct notifier_block *this, unsigned long event, ndisc_send_unsol_na(dev); in6_dev_put(idev); break; + case NETDEV_CHANGE: + change_info = ptr; + if (change_info->flags_changed & IFF_NOARP) + neigh_changeaddr(&nd_tbl, dev); + break; case NETDEV_DOWN: neigh_ifdown(&nd_tbl, dev); fib6_run_gc(0, net, false); -- cgit v0.10.2 From 4b561c17d91e9311639dd856255de0987f7ed9b2 Mon Sep 17 00:00:00 2001 From: Sunil Goutham Date: Wed, 29 Jul 2015 16:49:36 +0300 Subject: net: thunderx: Fix data integrity issues with LDWB Switching back to LDD transactions from LDWB. While transmitting packets out with LDWB transactions data integrity issues are seen very frequently. hence switching back to LDD. Signed-off-by: Sunil Goutham Signed-off-by: Robert Richter Signed-off-by: Aleksey Makarov Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/cavium/thunder/nicvf_queues.c b/drivers/net/ethernet/cavium/thunder/nicvf_queues.c index d69d228d..4dae6aa 100644 --- a/drivers/net/ethernet/cavium/thunder/nicvf_queues.c +++ b/drivers/net/ethernet/cavium/thunder/nicvf_queues.c @@ -992,7 +992,7 @@ static inline void nicvf_sq_add_gather_subdesc(struct snd_queue *sq, int qentry, memset(gather, 0, SND_QUEUE_DESC_SIZE); gather->subdesc_type = SQ_DESC_TYPE_GATHER; - gather->ld_type = NIC_SEND_LD_TYPE_E_LDWB; + gather->ld_type = NIC_SEND_LD_TYPE_E_LDD; gather->size = size; gather->addr = data; } -- cgit v0.10.2 From 143ceb0b8a1570d1f899900bf05df19ecd673a18 Mon Sep 17 00:00:00 2001 From: Sunil Goutham Date: Wed, 29 Jul 2015 16:49:37 +0300 Subject: net: thunderx: Fix memory leak while tearing down interface Fixed 'tso_hdrs' memory not being freed properly. Also fixed SQ skbuff maintenance issues. Signed-off-by: Sunil Goutham Signed-off-by: Aleksey Makarov Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/cavium/thunder/nicvf_main.c b/drivers/net/ethernet/cavium/thunder/nicvf_main.c index 8b119a0..2890cd6 100644 --- a/drivers/net/ethernet/cavium/thunder/nicvf_main.c +++ b/drivers/net/ethernet/cavium/thunder/nicvf_main.c @@ -425,6 +425,7 @@ static void nicvf_snd_pkt_handler(struct net_device *netdev, if (skb) { prefetch(skb); dev_consume_skb_any(skb); + sq->skbuff[cqe_tx->sqe_ptr] = (u64)NULL; } } diff --git a/drivers/net/ethernet/cavium/thunder/nicvf_queues.c b/drivers/net/ethernet/cavium/thunder/nicvf_queues.c index 4dae6aa..4c91959 100644 --- a/drivers/net/ethernet/cavium/thunder/nicvf_queues.c +++ b/drivers/net/ethernet/cavium/thunder/nicvf_queues.c @@ -382,7 +382,8 @@ static void nicvf_free_snd_queue(struct nicvf *nic, struct snd_queue *sq) return; if (sq->tso_hdrs) - dma_free_coherent(&nic->pdev->dev, sq->dmem.q_len, + dma_free_coherent(&nic->pdev->dev, + sq->dmem.q_len * TSO_HEADER_SIZE, sq->tso_hdrs, sq->tso_hdrs_phys); kfree(sq->skbuff); @@ -863,10 +864,11 @@ void nicvf_sq_free_used_descs(struct net_device *netdev, struct snd_queue *sq, continue; } skb = (struct sk_buff *)sq->skbuff[sq->head]; + if (skb) + dev_kfree_skb_any(skb); atomic64_add(1, (atomic64_t *)&netdev->stats.tx_packets); atomic64_add(hdr->tot_len, (atomic64_t *)&netdev->stats.tx_bytes); - dev_kfree_skb_any(skb); nicvf_put_sq_desc(sq, hdr->subdesc_cnt + 1); } } @@ -1048,7 +1050,7 @@ static int nicvf_sq_append_tso(struct nicvf *nic, struct snd_queue *sq, } nicvf_sq_add_hdr_subdesc(sq, hdr_qentry, seg_subdescs - 1, skb, seg_len); - sq->skbuff[hdr_qentry] = 0; + sq->skbuff[hdr_qentry] = (u64)NULL; qentry = nicvf_get_nxt_sqentry(sq, qentry); desc_cnt += seg_subdescs; -- cgit v0.10.2 From 32c1b965f41639081af84cd3f320a81d9c30a06e Mon Sep 17 00:00:00 2001 From: Sunil Goutham Date: Wed, 29 Jul 2015 16:49:38 +0300 Subject: net: thunderx: Fix RQ_DROP miscalculation With earlier configured value sufficient number of CQEs are not being reserved for transmitted packets. Hence under heavy incoming traffic load, receive notifications will take away most of the CQ thus transmit notifications will be lost resulting in tx skbs not being freed. Finally SQ will be full and it will be stopped, watchdog timer will kick in. After this fix receive notifications will not take morethan half of CQ reserving the rest for transmit notifications. Also changed CQ & SQ sizes from 16k to 4k. This is also due to the receive notifications taking first half of CQ under heavy load and time taken by NAPI to clear transmit notifications will increase with higher queue sizes. Again results in SQ being stopped. Signed-off-by: Sunil Goutham Signed-off-by: Aleksey Makarov Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/cavium/thunder/nicvf_queues.h b/drivers/net/ethernet/cavium/thunder/nicvf_queues.h index 8341bdf..f0937b7 100644 --- a/drivers/net/ethernet/cavium/thunder/nicvf_queues.h +++ b/drivers/net/ethernet/cavium/thunder/nicvf_queues.h @@ -62,7 +62,7 @@ #define SND_QUEUE_CNT 8 #define CMP_QUEUE_CNT 8 /* Max of RCV and SND qcount */ -#define SND_QSIZE SND_QUEUE_SIZE4 +#define SND_QSIZE SND_QUEUE_SIZE2 #define SND_QUEUE_LEN (1ULL << (SND_QSIZE + 10)) #define MAX_SND_QUEUE_LEN (1ULL << (SND_QUEUE_SIZE6 + 10)) #define SND_QUEUE_THRESH 2ULL @@ -70,7 +70,10 @@ /* Since timestamp not enabled, otherwise 2 */ #define MAX_CQE_PER_PKT_XMIT 1 -#define CMP_QSIZE CMP_QUEUE_SIZE4 +/* Keep CQ and SQ sizes same, if timestamping + * is enabled this equation will change. + */ +#define CMP_QSIZE CMP_QUEUE_SIZE2 #define CMP_QUEUE_LEN (1ULL << (CMP_QSIZE + 10)) #define CMP_QUEUE_CQE_THRESH 0 #define CMP_QUEUE_TIMER_THRESH 220 /* 10usec */ @@ -87,7 +90,12 @@ #define MAX_CQES_FOR_TX ((SND_QUEUE_LEN / MIN_SQ_DESC_PER_PKT_XMIT) * \ MAX_CQE_PER_PKT_XMIT) -#define RQ_CQ_DROP ((CMP_QUEUE_LEN - MAX_CQES_FOR_TX) / 256) +/* Calculate number of CQEs to reserve for all SQEs. + * Its 1/256th level of CQ size. + * '+ 1' to account for pipelining + */ +#define RQ_CQ_DROP ((256 / (CMP_QUEUE_LEN / \ + (CMP_QUEUE_LEN - MAX_CQES_FOR_TX))) + 1) /* Descriptor size in bytes */ #define SND_QUEUE_DESC_SIZE 16 -- cgit v0.10.2 From c62cd3c45198c212f85f53bfc254dd83f01b0fbc Mon Sep 17 00:00:00 2001 From: Sunil Goutham Date: Wed, 29 Jul 2015 16:49:39 +0300 Subject: net: thunderx: Fix memory leak when changing queue count Fix for memory leak when changing queue/channel count via ethtool Signed-off-by: Sunil Goutham Signed-off-by: Aleksey Makarov Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/cavium/thunder/nicvf_ethtool.c b/drivers/net/ethernet/cavium/thunder/nicvf_ethtool.c index 16bd2d7..b22dee6 100644 --- a/drivers/net/ethernet/cavium/thunder/nicvf_ethtool.c +++ b/drivers/net/ethernet/cavium/thunder/nicvf_ethtool.c @@ -126,6 +126,7 @@ static void nicvf_set_msglevel(struct net_device *netdev, u32 lvl) static void nicvf_get_strings(struct net_device *netdev, u32 sset, u8 *data) { + struct nicvf *nic = netdev_priv(netdev); int stats, qidx; if (sset != ETH_SS_STATS) @@ -141,7 +142,7 @@ static void nicvf_get_strings(struct net_device *netdev, u32 sset, u8 *data) data += ETH_GSTRING_LEN; } - for (qidx = 0; qidx < MAX_RCV_QUEUES_PER_QS; qidx++) { + for (qidx = 0; qidx < nic->qs->rq_cnt; qidx++) { for (stats = 0; stats < nicvf_n_queue_stats; stats++) { sprintf(data, "rxq%d: %s", qidx, nicvf_queue_stats[stats].name); @@ -149,7 +150,7 @@ static void nicvf_get_strings(struct net_device *netdev, u32 sset, u8 *data) } } - for (qidx = 0; qidx < MAX_SND_QUEUES_PER_QS; qidx++) { + for (qidx = 0; qidx < nic->qs->sq_cnt; qidx++) { for (stats = 0; stats < nicvf_n_queue_stats; stats++) { sprintf(data, "txq%d: %s", qidx, nicvf_queue_stats[stats].name); @@ -170,12 +171,14 @@ static void nicvf_get_strings(struct net_device *netdev, u32 sset, u8 *data) static int nicvf_get_sset_count(struct net_device *netdev, int sset) { + struct nicvf *nic = netdev_priv(netdev); + if (sset != ETH_SS_STATS) return -EINVAL; return nicvf_n_hw_stats + nicvf_n_drv_stats + (nicvf_n_queue_stats * - (MAX_RCV_QUEUES_PER_QS + MAX_SND_QUEUES_PER_QS)) + + (nic->qs->rq_cnt + nic->qs->sq_cnt)) + BGX_RX_STATS_COUNT + BGX_TX_STATS_COUNT; } @@ -197,13 +200,13 @@ static void nicvf_get_ethtool_stats(struct net_device *netdev, *(data++) = ((u64 *)&nic->drv_stats) [nicvf_drv_stats[stat].index]; - for (qidx = 0; qidx < MAX_RCV_QUEUES_PER_QS; qidx++) { + for (qidx = 0; qidx < nic->qs->rq_cnt; qidx++) { for (stat = 0; stat < nicvf_n_queue_stats; stat++) *(data++) = ((u64 *)&nic->qs->rq[qidx].stats) [nicvf_queue_stats[stat].index]; } - for (qidx = 0; qidx < MAX_SND_QUEUES_PER_QS; qidx++) { + for (qidx = 0; qidx < nic->qs->sq_cnt; qidx++) { for (stat = 0; stat < nicvf_n_queue_stats; stat++) *(data++) = ((u64 *)&nic->qs->sq[qidx].stats) [nicvf_queue_stats[stat].index]; @@ -543,6 +546,7 @@ static int nicvf_set_channels(struct net_device *dev, { struct nicvf *nic = netdev_priv(dev); int err = 0; + bool if_up = netif_running(dev); if (!channel->rx_count || !channel->tx_count) return -EINVAL; @@ -551,6 +555,9 @@ static int nicvf_set_channels(struct net_device *dev, if (channel->tx_count > MAX_SND_QUEUES_PER_QS) return -EINVAL; + if (if_up) + nicvf_stop(dev); + nic->qs->rq_cnt = channel->rx_count; nic->qs->sq_cnt = channel->tx_count; nic->qs->cq_cnt = max(nic->qs->rq_cnt, nic->qs->sq_cnt); @@ -559,11 +566,9 @@ static int nicvf_set_channels(struct net_device *dev, if (err) return err; - if (!netif_running(dev)) - return err; + if (if_up) + nicvf_open(dev); - nicvf_stop(dev); - nicvf_open(dev); netdev_info(dev, "Setting num Tx rings to %d, Rx rings to %d success\n", nic->qs->sq_cnt, nic->qs->rq_cnt); -- cgit v0.10.2 From 2cb468e01e6fdad3d8a00da6cda71094e0e05676 Mon Sep 17 00:00:00 2001 From: Sunil Goutham Date: Wed, 29 Jul 2015 16:49:40 +0300 Subject: net: thunderx: Fix TSO packet statistic Fixing TSO packages not being counted. Signed-off-by: Sunil Goutham Signed-off-by: Aleksey Makarov Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/cavium/thunder/nicvf_queues.c b/drivers/net/ethernet/cavium/thunder/nicvf_queues.c index 4c91959..500fdbe 100644 --- a/drivers/net/ethernet/cavium/thunder/nicvf_queues.c +++ b/drivers/net/ethernet/cavium/thunder/nicvf_queues.c @@ -1064,6 +1064,7 @@ static int nicvf_sq_append_tso(struct nicvf *nic, struct snd_queue *sq, /* Inform HW to xmit all TSO segments */ nicvf_queue_reg_write(nic, NIC_QSET_SQ_0_7_DOOR, skb_get_queue_mapping(skb), desc_cnt); + nic->drv_stats.tx_tso++; return 1; } -- cgit v0.10.2 From f8ce9666fa63da7e5afd9ff6e3221f86555621f3 Mon Sep 17 00:00:00 2001 From: Sunil Goutham Date: Wed, 29 Jul 2015 16:49:41 +0300 Subject: net: thunderx: Suppress alloc_pages() failure warnings Suppressing standard alloc_pages() warnings. Some kernel configs limit alloc size and the network driver may fail. Do not drop a kernel warning in this case, instead just drop a oneliner that the network driver could not be loaded since the buffer could not be allocated. Signed-off-by: Sunil Goutham Signed-off-by: Aleksey Makarov Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/cavium/thunder/nicvf_queues.c b/drivers/net/ethernet/cavium/thunder/nicvf_queues.c index 500fdbe..ca4240a 100644 --- a/drivers/net/ethernet/cavium/thunder/nicvf_queues.c +++ b/drivers/net/ethernet/cavium/thunder/nicvf_queues.c @@ -103,9 +103,11 @@ static inline int nicvf_alloc_rcv_buffer(struct nicvf *nic, gfp_t gfp, /* Allocate a new page */ if (!nic->rb_page) { - nic->rb_page = alloc_pages(gfp | __GFP_COMP, order); + nic->rb_page = alloc_pages(gfp | __GFP_COMP | __GFP_NOWARN, + order); if (!nic->rb_page) { - netdev_err(nic->netdev, "Failed to allocate new rcv buffer\n"); + netdev_err(nic->netdev, + "Failed to allocate new rcv buffer\n"); return -ENOMEM; } nic->rb_page_offset = 0; -- cgit v0.10.2 From 74840b83bd59ee51c591714470c6629ac18f424a Mon Sep 17 00:00:00 2001 From: Sunil Goutham Date: Wed, 29 Jul 2015 16:49:42 +0300 Subject: net: thunderx: Wakeup TXQ only if CQE_TX are processed Previously TXQ is wakedup whenever napi is executed and irrespective of if any CQE_TX are processed or not. Added 'txq_stop' and 'txq_wake' counters to aid in debugging if there are any future issues. Signed-off-by: Sunil Goutham Signed-off-by: Aleksey Makarov Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/cavium/thunder/nic.h b/drivers/net/ethernet/cavium/thunder/nic.h index dda8a02..a0be076 100644 --- a/drivers/net/ethernet/cavium/thunder/nic.h +++ b/drivers/net/ethernet/cavium/thunder/nic.h @@ -216,8 +216,9 @@ struct nicvf_drv_stats { /* Tx */ u64 tx_frames_ok; u64 tx_drops; - u64 tx_busy; u64 tx_tso; + u64 txq_stop; + u64 txq_wake; }; struct nicvf { diff --git a/drivers/net/ethernet/cavium/thunder/nicvf_ethtool.c b/drivers/net/ethernet/cavium/thunder/nicvf_ethtool.c index b22dee6..a4228e6 100644 --- a/drivers/net/ethernet/cavium/thunder/nicvf_ethtool.c +++ b/drivers/net/ethernet/cavium/thunder/nicvf_ethtool.c @@ -66,9 +66,10 @@ static const struct nicvf_stat nicvf_drv_stats[] = { NICVF_DRV_STAT(rx_frames_jumbo), NICVF_DRV_STAT(rx_drops), NICVF_DRV_STAT(tx_frames_ok), - NICVF_DRV_STAT(tx_busy), NICVF_DRV_STAT(tx_tso), NICVF_DRV_STAT(tx_drops), + NICVF_DRV_STAT(txq_stop), + NICVF_DRV_STAT(txq_wake), }; static const struct nicvf_stat nicvf_queue_stats[] = { diff --git a/drivers/net/ethernet/cavium/thunder/nicvf_main.c b/drivers/net/ethernet/cavium/thunder/nicvf_main.c index 2890cd6..422d46e 100644 --- a/drivers/net/ethernet/cavium/thunder/nicvf_main.c +++ b/drivers/net/ethernet/cavium/thunder/nicvf_main.c @@ -477,12 +477,13 @@ static void nicvf_rcv_pkt_handler(struct net_device *netdev, static int nicvf_cq_intr_handler(struct net_device *netdev, u8 cq_idx, struct napi_struct *napi, int budget) { - int processed_cqe, work_done = 0; + int processed_cqe, work_done = 0, tx_done = 0; int cqe_count, cqe_head; struct nicvf *nic = netdev_priv(netdev); struct queue_set *qs = nic->qs; struct cmp_queue *cq = &qs->cq[cq_idx]; struct cqe_rx_t *cq_desc; + struct netdev_queue *txq; spin_lock_bh(&cq->lock); loop: @@ -497,8 +498,8 @@ loop: cqe_head = nicvf_queue_reg_read(nic, NIC_QSET_CQ_0_7_HEAD, cq_idx) >> 9; cqe_head &= 0xFFFF; - netdev_dbg(nic->netdev, "%s cqe_count %d cqe_head %d\n", - __func__, cqe_count, cqe_head); + netdev_dbg(nic->netdev, "%s CQ%d cqe_count %d cqe_head %d\n", + __func__, cq_idx, cqe_count, cqe_head); while (processed_cqe < cqe_count) { /* Get the CQ descriptor */ cq_desc = (struct cqe_rx_t *)GET_CQ_DESC(cq, cqe_head); @@ -512,8 +513,8 @@ loop: break; } - netdev_dbg(nic->netdev, "cq_desc->cqe_type %d\n", - cq_desc->cqe_type); + netdev_dbg(nic->netdev, "CQ%d cq_desc->cqe_type %d\n", + cq_idx, cq_desc->cqe_type); switch (cq_desc->cqe_type) { case CQE_TYPE_RX: nicvf_rcv_pkt_handler(netdev, napi, cq, @@ -523,6 +524,7 @@ loop: case CQE_TYPE_SEND: nicvf_snd_pkt_handler(netdev, cq, (void *)cq_desc, CQE_TYPE_SEND); + tx_done++; break; case CQE_TYPE_INVALID: case CQE_TYPE_RX_SPLIT: @@ -533,8 +535,9 @@ loop: } processed_cqe++; } - netdev_dbg(nic->netdev, "%s processed_cqe %d work_done %d budget %d\n", - __func__, processed_cqe, work_done, budget); + netdev_dbg(nic->netdev, + "%s CQ%d processed_cqe %d work_done %d budget %d\n", + __func__, cq_idx, processed_cqe, work_done, budget); /* Ring doorbell to inform H/W to reuse processed CQEs */ nicvf_queue_reg_write(nic, NIC_QSET_CQ_0_7_DOOR, @@ -544,6 +547,19 @@ loop: goto loop; done: + /* Wakeup TXQ if its stopped earlier due to SQ full */ + if (tx_done) { + txq = netdev_get_tx_queue(netdev, cq_idx); + if (netif_tx_queue_stopped(txq)) { + netif_tx_wake_queue(txq); + nic->drv_stats.txq_wake++; + if (netif_msg_tx_err(nic)) + netdev_warn(netdev, + "%s: Transmit queue wakeup SQ%d\n", + netdev->name, cq_idx); + } + } + spin_unlock_bh(&cq->lock); return work_done; } @@ -555,15 +571,10 @@ static int nicvf_poll(struct napi_struct *napi, int budget) struct net_device *netdev = napi->dev; struct nicvf *nic = netdev_priv(netdev); struct nicvf_cq_poll *cq; - struct netdev_queue *txq; cq = container_of(napi, struct nicvf_cq_poll, napi); work_done = nicvf_cq_intr_handler(netdev, cq->cq_idx, napi, budget); - txq = netdev_get_tx_queue(netdev, cq->cq_idx); - if (netif_tx_queue_stopped(txq)) - netif_tx_wake_queue(txq); - if (work_done < budget) { /* Slow packet rate, exit polling */ napi_complete(napi); @@ -836,7 +847,7 @@ static netdev_tx_t nicvf_xmit(struct sk_buff *skb, struct net_device *netdev) if (!nicvf_sq_append_skb(nic, skb) && !netif_tx_queue_stopped(txq)) { netif_tx_stop_queue(txq); - nic->drv_stats.tx_busy++; + nic->drv_stats.txq_stop++; if (netif_msg_tx_err(nic)) netdev_warn(netdev, "%s: Transmit ring full, stopping SQ%d\n", @@ -989,6 +1000,9 @@ int nicvf_open(struct net_device *netdev) for (qidx = 0; qidx < qs->rbdr_cnt; qidx++) nicvf_enable_intr(nic, NICVF_INTR_RBDR, qidx); + nic->drv_stats.txq_stop = 0; + nic->drv_stats.txq_wake = 0; + netif_carrier_on(netdev); netif_tx_start_all_queues(netdev); -- cgit v0.10.2 From 3d7a8aaad8720edb301d40d4a9d7fa906c76ba34 Mon Sep 17 00:00:00 2001 From: Sunil Goutham Date: Wed, 29 Jul 2015 16:49:43 +0300 Subject: net: thunderx: Set watchdog timeout value If a txq (SQ) remains in stopped state after this timeout its considered as stuck and interface is reinited. Signed-off-by: Sunil Goutham Signed-off-by: Aleksey Makarov Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/cavium/thunder/nic.h b/drivers/net/ethernet/cavium/thunder/nic.h index a0be076..8aee250 100644 --- a/drivers/net/ethernet/cavium/thunder/nic.h +++ b/drivers/net/ethernet/cavium/thunder/nic.h @@ -125,6 +125,15 @@ */ #define NICPF_CLK_PER_INT_TICK 2 +/* Time to wait before we decide that a SQ is stuck. + * + * Since both pkt rx and tx notifications are done with same CQ, + * when packets are being received at very high rate (eg: L2 forwarding) + * then freeing transmitted skbs will be delayed and watchdog + * will kick in, resetting interface. Hence keeping this value high. + */ +#define NICVF_TX_TIMEOUT (50 * HZ) + struct nicvf_cq_poll { u8 cq_idx; /* Completion queue index */ struct napi_struct napi; diff --git a/drivers/net/ethernet/cavium/thunder/nicvf_main.c b/drivers/net/ethernet/cavium/thunder/nicvf_main.c index 422d46e..abee2d7 100644 --- a/drivers/net/ethernet/cavium/thunder/nicvf_main.c +++ b/drivers/net/ethernet/cavium/thunder/nicvf_main.c @@ -1293,6 +1293,7 @@ static int nicvf_probe(struct pci_dev *pdev, const struct pci_device_id *ent) netdev->hw_features = netdev->features; netdev->netdev_ops = &nicvf_netdev_ops; + netdev->watchdog_timeo = NICVF_TX_TIMEOUT; INIT_WORK(&nic->reset_task, nicvf_reset_task); -- cgit v0.10.2 From b49087dd0fa27d61b55f7c77d0b857e3b5055161 Mon Sep 17 00:00:00 2001 From: Sunil Goutham Date: Wed, 29 Jul 2015 16:49:44 +0300 Subject: net: thunderx: Fix crash when changing rss with mutliple traffic flows This fixes a crash when changing rss with multiple traffic flows. While interface teardown, disable tx queues after all NAPI threads are done. If done otherwise tx queues might be woken up inside NAPI if any CQE_TX are processed. Signed-off-by: Sunil Goutham Signed-off-by: Aleksey Makarov Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/cavium/thunder/nicvf_main.c b/drivers/net/ethernet/cavium/thunder/nicvf_main.c index abee2d7..c7a29a3 100644 --- a/drivers/net/ethernet/cavium/thunder/nicvf_main.c +++ b/drivers/net/ethernet/cavium/thunder/nicvf_main.c @@ -234,7 +234,7 @@ static void nicvf_handle_mbx_intr(struct nicvf *nic) nic->duplex == DUPLEX_FULL ? "Full duplex" : "Half duplex"); netif_carrier_on(nic->netdev); - netif_tx_wake_all_queues(nic->netdev); + netif_tx_start_all_queues(nic->netdev); } else { netdev_info(nic->netdev, "%s: Link is Down\n", nic->netdev->name); @@ -551,7 +551,7 @@ done: if (tx_done) { txq = netdev_get_tx_queue(netdev, cq_idx); if (netif_tx_queue_stopped(txq)) { - netif_tx_wake_queue(txq); + netif_tx_start_queue(txq); nic->drv_stats.txq_wake++; if (netif_msg_tx_err(nic)) netdev_warn(netdev, @@ -845,7 +845,7 @@ static netdev_tx_t nicvf_xmit(struct sk_buff *skb, struct net_device *netdev) return NETDEV_TX_OK; } - if (!nicvf_sq_append_skb(nic, skb) && !netif_tx_queue_stopped(txq)) { + if (!netif_tx_queue_stopped(txq) && !nicvf_sq_append_skb(nic, skb)) { netif_tx_stop_queue(txq); nic->drv_stats.txq_stop++; if (netif_msg_tx_err(nic)) @@ -871,7 +871,6 @@ int nicvf_stop(struct net_device *netdev) nicvf_send_msg_to_pf(nic, &mbx); netif_carrier_off(netdev); - netif_tx_disable(netdev); /* Disable RBDR & QS error interrupts */ for (qidx = 0; qidx < qs->rbdr_cnt; qidx++) { @@ -906,6 +905,8 @@ int nicvf_stop(struct net_device *netdev) kfree(cq_poll); } + netif_tx_disable(netdev); + /* Free resources */ nicvf_config_data_transfer(nic, false); -- cgit v0.10.2 From 4adf4351145e65dd98407299496ef00d82430b48 Mon Sep 17 00:00:00 2001 From: Sunil Goutham Date: Wed, 29 Jul 2015 16:49:45 +0300 Subject: net: thunderx: Add PCI driver shutdown routine Signed-off-by: Sunil Goutham Signed-off-by: Aleksey Makarov Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/cavium/thunder/nicvf_main.c b/drivers/net/ethernet/cavium/thunder/nicvf_main.c index c7a29a3..3b90afb 100644 --- a/drivers/net/ethernet/cavium/thunder/nicvf_main.c +++ b/drivers/net/ethernet/cavium/thunder/nicvf_main.c @@ -1335,11 +1335,17 @@ static void nicvf_remove(struct pci_dev *pdev) pci_disable_device(pdev); } +static void nicvf_shutdown(struct pci_dev *pdev) +{ + nicvf_remove(pdev); +} + static struct pci_driver nicvf_driver = { .name = DRV_NAME, .id_table = nicvf_id_table, .probe = nicvf_probe, .remove = nicvf_remove, + .shutdown = nicvf_shutdown, }; static int __init nicvf_init_module(void) -- cgit v0.10.2 From 60f83c898776c363ef8ebb66f14cbc748c1df1e8 Mon Sep 17 00:00:00 2001 From: Thanneeru Srinivasulu Date: Wed, 29 Jul 2015 16:49:46 +0300 Subject: net: thunderx: Fix for crash while BGX teardown Cortina phy does not have kernel driver and we don't attach device with phy layer for intefaces like XFI, XLAUI etc, Hence check for interface type before calling disconnect. Signed-off-by: Thanneeru Srinivasulu Signed-off-by: Aleksey Makarov Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/cavium/thunder/thunder_bgx.c b/drivers/net/ethernet/cavium/thunder/thunder_bgx.c index 633ec05..b961a89 100644 --- a/drivers/net/ethernet/cavium/thunder/thunder_bgx.c +++ b/drivers/net/ethernet/cavium/thunder/thunder_bgx.c @@ -673,7 +673,10 @@ static void bgx_lmac_disable(struct bgx *bgx, u8 lmacid) bgx_reg_write(bgx, lmacid, BGX_CMRX_CFG, cmrx_cfg); bgx_flush_dmac_addrs(bgx, lmacid); - if (lmac->phydev) + if ((bgx->lmac_type != BGX_MODE_XFI) && + (bgx->lmac_type != BGX_MODE_XLAUI) && + (bgx->lmac_type != BGX_MODE_40G_KR) && + (bgx->lmac_type != BGX_MODE_10G_KR) && lmac->phydev) phy_disconnect(lmac->phydev); lmac->phydev = NULL; -- cgit v0.10.2 From f4eaed28c7834fc049c754f63e6988bbd73778d9 Mon Sep 17 00:00:00 2001 From: Daniel Borkmann Date: Wed, 29 Jul 2015 18:40:56 +0200 Subject: act_bpf: fix memory leaks when replacing bpf programs We currently trigger multiple memory leaks when replacing bpf actions, besides others: comm "tc", pid 1909, jiffies 4294851310 (age 1602.796s) hex dump (first 32 bytes): 01 00 00 00 03 00 00 00 00 00 00 00 00 00 00 00 ................ 18 b0 98 6d 00 88 ff ff 00 00 00 00 00 00 00 00 ...m............ backtrace: [] kmemleak_alloc+0x4e/0xb0 [] __vmalloc_node_range+0x1bd/0x2c0 [] __vmalloc+0x4a/0x50 [] bpf_prog_alloc+0x3a/0xa0 [] bpf_prog_create+0x44/0xa0 [] tcf_bpf_init+0x28b/0x3c0 [act_bpf] [] tcf_action_init_1+0x191/0x1b0 [] tcf_action_init+0x82/0xf0 [] tcf_exts_validate+0xb2/0xc0 [] cls_bpf_modify_existing+0x98/0x340 [cls_bpf] [] cls_bpf_change+0x1a6/0x274 [cls_bpf] [] tc_ctl_tfilter+0x335/0x910 [] rtnetlink_rcv_msg+0x95/0x240 [] netlink_rcv_skb+0xaf/0xc0 [] rtnetlink_rcv+0x2e/0x40 [] netlink_unicast+0xef/0x1b0 Issue is that the old content from tcf_bpf is allocated and needs to be released when we replace it. We seem to do that since the beginning of act_bpf on the filter and insns, later on the name as well. Example test case, after patch: # FOO="1,6 0 0 4294967295," # BAR="1,6 0 0 4294967294," # tc actions add action bpf bytecode "$FOO" index 2 # tc actions show action bpf action order 0: bpf bytecode '1,6 0 0 4294967295' default-action pipe index 2 ref 1 bind 0 # tc actions replace action bpf bytecode "$BAR" index 2 # tc actions show action bpf action order 0: bpf bytecode '1,6 0 0 4294967294' default-action pipe index 2 ref 1 bind 0 # tc actions replace action bpf bytecode "$FOO" index 2 # tc actions show action bpf action order 0: bpf bytecode '1,6 0 0 4294967295' default-action pipe index 2 ref 1 bind 0 # tc actions del action bpf index 2 [...] # echo "scan" > /sys/kernel/debug/kmemleak # cat /sys/kernel/debug/kmemleak | grep "comm \"tc\"" | wc -l 0 Fixes: d23b8ad8ab23 ("tc: add BPF based action") Signed-off-by: Daniel Borkmann Signed-off-by: David S. Miller diff --git a/net/sched/act_bpf.c b/net/sched/act_bpf.c index 1df7828..d0edeb7 100644 --- a/net/sched/act_bpf.c +++ b/net/sched/act_bpf.c @@ -27,9 +27,10 @@ struct tcf_bpf_cfg { struct bpf_prog *filter; struct sock_filter *bpf_ops; - char *bpf_name; + const char *bpf_name; u32 bpf_fd; u16 bpf_num_ops; + bool is_ebpf; }; static int tcf_bpf(struct sk_buff *skb, const struct tc_action *act, @@ -207,6 +208,7 @@ static int tcf_bpf_init_from_ops(struct nlattr **tb, struct tcf_bpf_cfg *cfg) cfg->bpf_ops = bpf_ops; cfg->bpf_num_ops = bpf_num_ops; cfg->filter = fp; + cfg->is_ebpf = false; return 0; } @@ -241,18 +243,40 @@ static int tcf_bpf_init_from_efd(struct nlattr **tb, struct tcf_bpf_cfg *cfg) cfg->bpf_fd = bpf_fd; cfg->bpf_name = name; cfg->filter = fp; + cfg->is_ebpf = true; return 0; } +static void tcf_bpf_cfg_cleanup(const struct tcf_bpf_cfg *cfg) +{ + if (cfg->is_ebpf) + bpf_prog_put(cfg->filter); + else + bpf_prog_destroy(cfg->filter); + + kfree(cfg->bpf_ops); + kfree(cfg->bpf_name); +} + +static void tcf_bpf_prog_fill_cfg(const struct tcf_bpf *prog, + struct tcf_bpf_cfg *cfg) +{ + cfg->is_ebpf = tcf_bpf_is_ebpf(prog); + cfg->filter = prog->filter; + + cfg->bpf_ops = prog->bpf_ops; + cfg->bpf_name = prog->bpf_name; +} + static int tcf_bpf_init(struct net *net, struct nlattr *nla, struct nlattr *est, struct tc_action *act, int replace, int bind) { struct nlattr *tb[TCA_ACT_BPF_MAX + 1]; + struct tcf_bpf_cfg cfg, old; struct tc_act_bpf *parm; struct tcf_bpf *prog; - struct tcf_bpf_cfg cfg; bool is_bpf, is_ebpf; int ret; @@ -301,6 +325,9 @@ static int tcf_bpf_init(struct net *net, struct nlattr *nla, prog = to_bpf(act); spin_lock_bh(&prog->tcf_lock); + if (ret != ACT_P_CREATED) + tcf_bpf_prog_fill_cfg(prog, &old); + prog->bpf_ops = cfg.bpf_ops; prog->bpf_name = cfg.bpf_name; @@ -316,32 +343,22 @@ static int tcf_bpf_init(struct net *net, struct nlattr *nla, if (ret == ACT_P_CREATED) tcf_hash_insert(act); + else + tcf_bpf_cfg_cleanup(&old); return ret; destroy_fp: - if (is_ebpf) - bpf_prog_put(cfg.filter); - else - bpf_prog_destroy(cfg.filter); - - kfree(cfg.bpf_ops); - kfree(cfg.bpf_name); - + tcf_bpf_cfg_cleanup(&cfg); return ret; } static void tcf_bpf_cleanup(struct tc_action *act, int bind) { - const struct tcf_bpf *prog = act->priv; - - if (tcf_bpf_is_ebpf(prog)) - bpf_prog_put(prog->filter); - else - bpf_prog_destroy(prog->filter); + struct tcf_bpf_cfg tmp; - kfree(prog->bpf_ops); - kfree(prog->bpf_name); + tcf_bpf_prog_fill_cfg(act->priv, &tmp); + tcf_bpf_cfg_cleanup(&tmp); } static struct tc_action_ops act_bpf_ops __read_mostly = { -- cgit v0.10.2 From 15f1bb1f1e067be7088ed43ef23d59629bd24348 Mon Sep 17 00:00:00 2001 From: Shahed Shaikh Date: Wed, 29 Jul 2015 07:55:35 -0400 Subject: qlcnic: Fix corruption while copying Use proper typecasting while performing byte-by-byte copy Signed-off-by: Shahed Shaikh Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/qlogic/qlcnic/qlcnic_83xx_init.c b/drivers/net/ethernet/qlogic/qlcnic/qlcnic_83xx_init.c index 33669c2..753ea8b 100644 --- a/drivers/net/ethernet/qlogic/qlcnic/qlcnic_83xx_init.c +++ b/drivers/net/ethernet/qlogic/qlcnic/qlcnic_83xx_init.c @@ -1415,7 +1415,7 @@ static int qlcnic_83xx_copy_fw_file(struct qlcnic_adapter *adapter) if (fw->size & 0xF) { addr = dest + size; for (i = 0; i < (fw->size & 0xF); i++) - data[i] = temp[size + i]; + data[i] = ((u8 *)temp)[size + i]; for (; i < 16; i++) data[i] = 0; ret = qlcnic_ms_mem_write128(adapter, addr, -- cgit v0.10.2 From 55c99a4dc50fb749076592e8c66c235ca0124360 Mon Sep 17 00:00:00 2001 From: Joerg Roedel Date: Tue, 28 Jul 2015 16:58:47 +0200 Subject: iommu/amd: Use iommu_attach_group() Since the conversion to default domains the iommu_attach_device function only works for devices with their own group. But this isn't always true for current IOMMUv2 capable devices, so use iommu_attach_group instead. Signed-off-by: Joerg Roedel diff --git a/drivers/iommu/amd_iommu_v2.c b/drivers/iommu/amd_iommu_v2.c index 3465faf..f7b875b 100644 --- a/drivers/iommu/amd_iommu_v2.c +++ b/drivers/iommu/amd_iommu_v2.c @@ -132,11 +132,19 @@ static struct device_state *get_device_state(u16 devid) static void free_device_state(struct device_state *dev_state) { + struct iommu_group *group; + /* * First detach device from domain - No more PRI requests will arrive * from that device after it is unbound from the IOMMUv2 domain. */ - iommu_detach_device(dev_state->domain, &dev_state->pdev->dev); + group = iommu_group_get(&dev_state->pdev->dev); + if (WARN_ON(!group)) + return; + + iommu_detach_group(dev_state->domain, group); + + iommu_group_put(group); /* Everything is down now, free the IOMMUv2 domain */ iommu_domain_free(dev_state->domain); @@ -731,6 +739,7 @@ EXPORT_SYMBOL(amd_iommu_unbind_pasid); int amd_iommu_init_device(struct pci_dev *pdev, int pasids) { struct device_state *dev_state; + struct iommu_group *group; unsigned long flags; int ret, tmp; u16 devid; @@ -776,10 +785,16 @@ int amd_iommu_init_device(struct pci_dev *pdev, int pasids) if (ret) goto out_free_domain; - ret = iommu_attach_device(dev_state->domain, &pdev->dev); - if (ret != 0) + group = iommu_group_get(&pdev->dev); + if (!group) goto out_free_domain; + ret = iommu_attach_group(dev_state->domain, group); + if (ret != 0) + goto out_drop_group; + + iommu_group_put(group); + spin_lock_irqsave(&state_lock, flags); if (__get_device_state(devid) != NULL) { @@ -794,6 +809,9 @@ int amd_iommu_init_device(struct pci_dev *pdev, int pasids) return 0; +out_drop_group: + iommu_group_put(group); + out_free_domain: iommu_domain_free(dev_state->domain); -- cgit v0.10.2 From 1e6a7b04c033fe76ec7fe746ef6a3b22ab9502b2 Mon Sep 17 00:00:00 2001 From: Joerg Roedel Date: Tue, 28 Jul 2015 16:58:48 +0200 Subject: iommu/amd: Use iommu core for passthrough mode Remove the AMD IOMMU driver implementation for passthrough mode and rely on the new iommu core features for that. Signed-off-by: Joerg Roedel diff --git a/drivers/iommu/amd_iommu.c b/drivers/iommu/amd_iommu.c index a57e9b7..89e6d4b 100644 --- a/drivers/iommu/amd_iommu.c +++ b/drivers/iommu/amd_iommu.c @@ -76,8 +76,6 @@ LIST_HEAD(hpet_map); * Domain for untranslated devices - only allocated * if iommu=pt passed on kernel cmd line. */ -static struct protection_domain *pt_domain; - static const struct iommu_ops amd_iommu_ops; static ATOMIC_NOTIFIER_HEAD(ppr_notifier); @@ -96,7 +94,7 @@ struct iommu_dev_data { struct protection_domain *domain; /* Domain the device is bound to */ u16 devid; /* PCI Device ID */ bool iommu_v2; /* Device can make use of IOMMUv2 */ - bool passthrough; /* Default for device is pt_domain */ + bool passthrough; /* Device is identity mapped */ struct { bool enabled; int qdep; @@ -116,7 +114,6 @@ struct iommu_cmd { struct kmem_cache *amd_iommu_irq_cache; static void update_domain(struct protection_domain *domain); -static int alloc_passthrough_domain(void); static int protection_domain_init(struct protection_domain *domain); /**************************************************************************** @@ -2221,15 +2218,6 @@ static void __detach_device(struct iommu_dev_data *dev_data) do_detach(head); spin_unlock_irqrestore(&domain->lock, flags); - - /* - * If we run in passthrough mode the device must be assigned to the - * passthrough domain if it is detached from any other domain. - * Make sure we can deassign from the pt_domain itself. - */ - if (dev_data->passthrough && - (dev_data->domain == NULL && domain != pt_domain)) - __attach_device(dev_data, pt_domain); } /* @@ -2287,7 +2275,7 @@ static int amd_iommu_add_device(struct device *dev) BUG_ON(!dev_data); - if (dev_data->iommu_v2) + if (iommu_pass_through || dev_data->iommu_v2) iommu_request_dm_for_dev(dev); /* Domains are initialized for this device - have a look what we ended up with */ @@ -2947,21 +2935,6 @@ out_err: return NULL; } -static int alloc_passthrough_domain(void) -{ - if (pt_domain != NULL) - return 0; - - /* allocate passthrough domain */ - pt_domain = protection_domain_alloc(); - if (!pt_domain) - return -ENOMEM; - - pt_domain->mode = PAGE_MODE_NONE; - - return 0; -} - static struct iommu_domain *amd_iommu_domain_alloc(unsigned type) { struct protection_domain *pdomain; @@ -3222,33 +3195,6 @@ static const struct iommu_ops amd_iommu_ops = { * *****************************************************************************/ -int __init amd_iommu_init_passthrough(void) -{ - struct iommu_dev_data *dev_data; - struct pci_dev *dev = NULL; - int ret; - - ret = alloc_passthrough_domain(); - if (ret) - return ret; - - for_each_pci_dev(dev) { - if (!check_device(&dev->dev)) - continue; - - dev_data = get_dev_data(&dev->dev); - dev_data->passthrough = true; - - attach_device(&dev->dev, pt_domain); - } - - amd_iommu_stats_init(); - - pr_info("AMD-Vi: Initialized for Passthrough Mode\n"); - - return 0; -} - /* IOMMUv2 specific functions */ int amd_iommu_register_ppr_notifier(struct notifier_block *nb) { diff --git a/drivers/iommu/amd_iommu_init.c b/drivers/iommu/amd_iommu_init.c index dbda9ae..a24495e 100644 --- a/drivers/iommu/amd_iommu_init.c +++ b/drivers/iommu/amd_iommu_init.c @@ -2026,14 +2026,6 @@ static bool detect_ivrs(void) return true; } -static int amd_iommu_init_dma(void) -{ - if (iommu_pass_through) - return amd_iommu_init_passthrough(); - else - return amd_iommu_init_dma_ops(); -} - /**************************************************************************** * * AMD IOMMU Initialization State Machine @@ -2073,7 +2065,7 @@ static int __init state_next(void) init_state = ret ? IOMMU_INIT_ERROR : IOMMU_INTERRUPTS_EN; break; case IOMMU_INTERRUPTS_EN: - ret = amd_iommu_init_dma(); + ret = amd_iommu_init_dma_ops(); init_state = ret ? IOMMU_INIT_ERROR : IOMMU_DMA_OPS; break; case IOMMU_DMA_OPS: -- cgit v0.10.2 From 02ca20212f0dde5c90be8de19cc159726b5561aa Mon Sep 17 00:00:00 2001 From: Joerg Roedel Date: Tue, 28 Jul 2015 16:58:49 +0200 Subject: iommu/amd: Allow non-IOMMUv2 devices in IOMMUv2 domains Since devices with IOMMUv2 functionality might be in the same group as devices without it, allow those devices in IOMMUv2 domains too. Otherwise attaching the group with the IOMMUv2 device to the domain will fail. Signed-off-by: Joerg Roedel diff --git a/drivers/iommu/amd_iommu.c b/drivers/iommu/amd_iommu.c index 89e6d4b..6d3dae9 100644 --- a/drivers/iommu/amd_iommu.c +++ b/drivers/iommu/amd_iommu.c @@ -2164,15 +2164,17 @@ static int attach_device(struct device *dev, dev_data = get_dev_data(dev); if (domain->flags & PD_IOMMUV2_MASK) { - if (!dev_data->iommu_v2 || !dev_data->passthrough) + if (!dev_data->passthrough) return -EINVAL; - if (pdev_iommuv2_enable(pdev) != 0) - return -EINVAL; + if (dev_data->iommu_v2) { + if (pdev_iommuv2_enable(pdev) != 0) + return -EINVAL; - dev_data->ats.enabled = true; - dev_data->ats.qdep = pci_ats_queue_depth(pdev); - dev_data->pri_tlp = pci_pri_tlp_required(pdev); + dev_data->ats.enabled = true; + dev_data->ats.qdep = pci_ats_queue_depth(pdev); + dev_data->pri_tlp = pci_pri_tlp_required(pdev); + } } else if (amd_iommu_iotlb_sup && pci_enable_ats(pdev, PAGE_SHIFT) == 0) { dev_data->ats.enabled = true; @@ -2237,7 +2239,7 @@ static void detach_device(struct device *dev) __detach_device(dev_data); write_unlock_irqrestore(&amd_iommu_devtable_lock, flags); - if (domain->flags & PD_IOMMUV2_MASK) + if (domain->flags & PD_IOMMUV2_MASK && dev_data->iommu_v2) pdev_iommuv2_disable(to_pci_dev(dev)); else if (dev_data->ats.enabled) pci_disable_ats(to_pci_dev(dev)); -- cgit v0.10.2 From 323023245771589c53869396e3297c703d347852 Mon Sep 17 00:00:00 2001 From: Joerg Roedel Date: Tue, 28 Jul 2015 16:58:50 +0200 Subject: iommu/amd: Use swiotlb in passthrough mode In passthrough mode (iommu=pt) all devices are identity mapped. If a device does not support 64bit DMA it might still need remapping. Make sure swiotlb is initialized to provide this remapping. Signed-off-by: Joerg Roedel diff --git a/drivers/iommu/amd_iommu.c b/drivers/iommu/amd_iommu.c index 6d3dae9..e29baa6 100644 --- a/drivers/iommu/amd_iommu.c +++ b/drivers/iommu/amd_iommu.c @@ -2282,12 +2282,10 @@ static int amd_iommu_add_device(struct device *dev) /* Domains are initialized for this device - have a look what we ended up with */ domain = iommu_get_domain_for_dev(dev); - if (domain->type == IOMMU_DOMAIN_IDENTITY) { + if (domain->type == IOMMU_DOMAIN_IDENTITY) dev_data->passthrough = true; - dev->archdata.dma_ops = &nommu_dma_ops; - } else { + else dev->archdata.dma_ops = &amd_iommu_dma_ops; - } out: iommu_completion_wait(iommu); @@ -2852,8 +2850,8 @@ int __init amd_iommu_init_api(void) int __init amd_iommu_init_dma_ops(void) { + swiotlb = iommu_pass_through ? 1 : 0; iommu_detected = 1; - swiotlb = 0; amd_iommu_stats_init(); -- cgit v0.10.2 From 52717828356b643a1650fca845b4af488a954cca Mon Sep 17 00:00:00 2001 From: Joerg Roedel Date: Tue, 28 Jul 2015 16:58:51 +0200 Subject: iommu/amd: Set global dma_ops if swiotlb is disabled Some AMD systems also have non-PCI devices which can do DMA. Those can't be handled by the AMD IOMMU, as the hardware can only handle PCI. These devices would end up with no dma_ops, as neither the per-device nor the global dma_ops will get set. SWIOTLB provides global dma_ops when it is active, so make sure there are global dma_ops too when swiotlb is disabled. Signed-off-by: Joerg Roedel diff --git a/drivers/iommu/amd_iommu.c b/drivers/iommu/amd_iommu.c index e29baa6..fa9508b 100644 --- a/drivers/iommu/amd_iommu.c +++ b/drivers/iommu/amd_iommu.c @@ -2853,6 +2853,15 @@ int __init amd_iommu_init_dma_ops(void) swiotlb = iommu_pass_through ? 1 : 0; iommu_detected = 1; + /* + * In case we don't initialize SWIOTLB (actually the common case + * when AMD IOMMU is enabled), make sure there are global + * dma_ops set as a fall-back for devices not handled by this + * driver (for example non-PCI devices). + */ + if (!swiotlb) + dma_ops = &nommu_dma_ops; + amd_iommu_stats_init(); if (amd_iommu_unmap_flush) -- cgit v0.10.2 From b8d65e9662b1ffb3b52a65fd11b0b968022dc6a1 Mon Sep 17 00:00:00 2001 From: Alistair Popple Date: Thu, 30 Jul 2015 16:53:54 +1000 Subject: powerpc/eeh-powernv: Fix unbalanced IRQ warning pnv_eeh_next_error() re-enables the eeh opal event interrupt but it gets called from a loop if there are more outstanding events to process, resulting in a warning due to enabling an already enabled interrupt. Instead the interrupt should only be re-enabled once the last outstanding event has been processed. Tested-by: Daniel Axtens Reported-by: Daniel Axtens Signed-off-by: Alistair Popple Acked-by: Gavin Shan Signed-off-by: Michael Ellerman diff --git a/arch/powerpc/platforms/powernv/eeh-powernv.c b/arch/powerpc/platforms/powernv/eeh-powernv.c index 5cf5e6e..7cf0df8 100644 --- a/arch/powerpc/platforms/powernv/eeh-powernv.c +++ b/arch/powerpc/platforms/powernv/eeh-powernv.c @@ -1478,7 +1478,7 @@ static int pnv_eeh_next_error(struct eeh_pe **pe) } /* Unmask the event */ - if (eeh_enabled()) + if (ret == EEH_NEXT_ERR_NONE && eeh_enabled()) enable_irq(eeh_event_irq); return ret; -- cgit v0.10.2 From d0e30adc42d979e4adc36b6c112b57337423b70c Mon Sep 17 00:00:00 2001 From: Chris Wilson Date: Wed, 29 Jul 2015 20:02:48 +0100 Subject: drm/i915: Mark PIN_USER binding as GLOBAL_BIND without the aliasing ppgtt If the device does not support the aliasing ppgtt, we must translate user bind requests (PIN_USER) from LOCAL_BIND to a GLOBAL_BIND. However, since this is device specific we cannot do this conveniently in the upper layers and so must manage the vma->bound flags in the backend. Partial revert of commit 75d04a3773ecee617847de963ae4195d6aa74c28 [4.2-rc1] Author: Mika Kuoppala Date: Tue Apr 28 17:56:17 2015 +0300 drm/i915/gtt: Allocate va range only if vma is not bound Note this was spotted by Daniel originally, but we dropped the ball in getting the fix in before the bug going wild. Sorry all. Reported-by: Vincent Legoll vincent.legoll@gmail.com Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=91133 References: https://bugs.freedesktop.org/show_bug.cgi?id=90224 Signed-off-by: Chris Wilson Cc: Michel Thierry Cc: Daniel Vetter Cc: Mika Kuoppala Cc: Jani Nikula Signed-off-by: Daniel Vetter diff --git a/drivers/gpu/drm/i915/i915_gem_gtt.c b/drivers/gpu/drm/i915/i915_gem_gtt.c index 56b52a4..31e8269 100644 --- a/drivers/gpu/drm/i915/i915_gem_gtt.c +++ b/drivers/gpu/drm/i915/i915_gem_gtt.c @@ -1923,6 +1923,17 @@ static int ggtt_bind_vma(struct i915_vma *vma, vma->vm->insert_entries(vma->vm, pages, vma->node.start, cache_level, pte_flags); + + /* Note the inconsistency here is due to absence of the + * aliasing ppgtt on gen4 and earlier. Though we always + * request PIN_USER for execbuffer (translated to LOCAL_BIND), + * without the appgtt, we cannot honour that request and so + * must substitute it with a global binding. Since we do this + * behind the upper layers back, we need to explicitly set + * the bound flag ourselves. + */ + vma->bound |= GLOBAL_BIND; + } if (dev_priv->mm.aliasing_ppgtt && flags & LOCAL_BIND) { -- cgit v0.10.2 From 5eb3e5a5e11d14f9deb2a4b83555443b69ab9940 Mon Sep 17 00:00:00 2001 From: Chris Wilson Date: Sun, 28 Jun 2015 09:19:26 +0100 Subject: drm/i915: Declare the swizzling unknown for L-shaped configurations The old style of memory interleaving swizzled upto the end of the first even bank of memory, and then used the remainder as unswizzled on the unpaired bank - i.e. swizzling is not constant for all memory. This causes problems when we try to migrate memory and so the kernel prevents migration at all when we detect L-shaped inconsistent swizzling. However, this issue also extends to userspace who try to manually detile into memory as the swizzling for an individual page is unknown (it depends on its physical address only known to the kernel), userspace cannot correctly swizzle. Note that this is a new attempt for the previously merged one, reverted in commit d82c0ba6e306f079407f07003e53c262d683397b Author: Daniel Vetter Date: Tue Jul 14 12:29:27 2015 +0200 Revert "drm/i915: Declare the swizzling unknown for L-shaped configurations" This is cc: stable since we need it to fix up troubles with wc cpu mmaps that userspace recently started to use widely. Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=91105 Signed-off-by: Chris Wilson Cc: Daniel Vetter Cc: stable@vger.kernel.org [danvet: Add note about previous (failed attempt).] Signed-off-by: Daniel Vetter diff --git a/drivers/gpu/drm/i915/i915_gem_tiling.c b/drivers/gpu/drm/i915/i915_gem_tiling.c index 633bd1f..d19c9db 100644 --- a/drivers/gpu/drm/i915/i915_gem_tiling.c +++ b/drivers/gpu/drm/i915/i915_gem_tiling.c @@ -464,7 +464,10 @@ i915_gem_get_tiling(struct drm_device *dev, void *data, } /* Hide bit 17 from the user -- see comment in i915_gem_set_tiling */ - args->phys_swizzle_mode = args->swizzle_mode; + if (dev_priv->quirks & QUIRK_PIN_SWIZZLED_PAGES) + args->phys_swizzle_mode = I915_BIT_6_SWIZZLE_UNKNOWN; + else + args->phys_swizzle_mode = args->swizzle_mode; if (args->swizzle_mode == I915_BIT_6_SWIZZLE_9_17) args->swizzle_mode = I915_BIT_6_SWIZZLE_9; if (args->swizzle_mode == I915_BIT_6_SWIZZLE_9_10_17) -- cgit v0.10.2 From 7cc03e48965453b5df1cce5062c826189b04b960 Mon Sep 17 00:00:00 2001 From: Dmitry Skorodumov Date: Tue, 28 Jul 2015 18:38:32 +0400 Subject: x86/efi: Use all 64 bit of efi_memmap in setup_e820() The efi_info structure stores low 32 bits of memory map in efi_memmap and high 32 bits in efi_memmap_hi. While constructing pointer in the setup_e820(), need to take into account all 64 bit of the pointer. It is because on 64bit machine the function efi_get_memory_map() may return full 64bit pointer and before the patch that pointer was truncated. The issue is triggered on Parallles virtual machine and fixed with this patch. Signed-off-by: Dmitry Skorodumov Cc: Denis V. Lunev Cc: Signed-off-by: Matt Fleming diff --git a/arch/x86/boot/compressed/eboot.c b/arch/x86/boot/compressed/eboot.c index 2c82bd1..7d69afd 100644 --- a/arch/x86/boot/compressed/eboot.c +++ b/arch/x86/boot/compressed/eboot.c @@ -1193,6 +1193,10 @@ static efi_status_t setup_e820(struct boot_params *params, unsigned int e820_type = 0; unsigned long m = efi->efi_memmap; +#ifdef CONFIG_X86_64 + m |= (u64)efi->efi_memmap_hi << 32; +#endif + d = (efi_memory_desc_t *)(m + (i * efi->efi_memdesc_size)); switch (d->type) { case EFI_RESERVED_TYPE: -- cgit v0.10.2 From 9115c7589b11349a1c3099758b4bded579ff69e0 Mon Sep 17 00:00:00 2001 From: Ricardo Neri Date: Wed, 15 Jul 2015 19:36:03 -0700 Subject: efi: Check for NULL efi kernel parameters Even though it is documented how to specifiy efi parameters, it is possible to cause a kernel panic due to a dereference of a NULL pointer when parsing such parameters if "efi" alone is given: PANIC: early exception 0e rip 10:ffffffff812fb361 error 0 cr2 0 [ 0.000000] CPU: 0 PID: 0 Comm: swapper Not tainted 4.2.0-rc1+ #450 [ 0.000000] ffffffff81fe20a9 ffffffff81e03d50 ffffffff8184bb0f 00000000000003f8 [ 0.000000] 0000000000000000 ffffffff81e03e08 ffffffff81f371a1 64656c62616e6520 [ 0.000000] 0000000000000069 000000000000005f 0000000000000000 0000000000000000 [ 0.000000] Call Trace: [ 0.000000] [] dump_stack+0x45/0x57 [ 0.000000] [] early_idt_handler_common+0x81/0xae [ 0.000000] [] ? parse_option_str+0x11/0x90 [ 0.000000] [] arch_parse_efi_cmdline+0x15/0x42 [ 0.000000] [] do_early_param+0x50/0x8a [ 0.000000] [] parse_args+0x1e3/0x400 [ 0.000000] [] parse_early_options+0x24/0x28 [ 0.000000] [] ? loglevel+0x31/0x31 [ 0.000000] [] parse_early_param+0x31/0x3d [ 0.000000] [] setup_arch+0x2de/0xc08 [ 0.000000] [] ? vprintk_default+0x1a/0x20 [ 0.000000] [] start_kernel+0x90/0x423 [ 0.000000] [] x86_64_start_reservations+0x2a/0x2c [ 0.000000] [] x86_64_start_kernel+0xeb/0xef [ 0.000000] RIP 0xffffffff81ba2efc This panic is not reproducible with "efi=" as this will result in a non-NULL zero-length string. Thus, verify that the pointer to the parameter string is not NULL. This is consistent with other parameter-parsing functions which check for NULL pointers. Signed-off-by: Ricardo Neri Cc: Dave Young Cc: Signed-off-by: Matt Fleming diff --git a/arch/x86/platform/efi/efi.c b/arch/x86/platform/efi/efi.c index cfba30f..e4308fe 100644 --- a/arch/x86/platform/efi/efi.c +++ b/arch/x86/platform/efi/efi.c @@ -972,6 +972,11 @@ u64 efi_mem_attributes(unsigned long phys_addr) static int __init arch_parse_efi_cmdline(char *str) { + if (!str) { + pr_warn("need at least one option\n"); + return -EINVAL; + } + if (parse_option_str(str, "old_map")) set_bit(EFI_OLD_MEMMAP, &efi.flags); if (parse_option_str(str, "debug")) diff --git a/drivers/firmware/efi/efi.c b/drivers/firmware/efi/efi.c index 9fa8084..d6144e3 100644 --- a/drivers/firmware/efi/efi.c +++ b/drivers/firmware/efi/efi.c @@ -58,6 +58,11 @@ bool efi_runtime_disabled(void) static int __init parse_efi_cmdline(char *str) { + if (!str) { + pr_warn("need at least one option\n"); + return -EINVAL; + } + if (parse_option_str(str, "noruntime")) disable_runtime = true; -- cgit v0.10.2 From 36b8e180e1e929e00b351c3b72aab3147fc14116 Mon Sep 17 00:00:00 2001 From: Brian King Date: Tue, 14 Jul 2015 11:41:29 -0500 Subject: ipr: Fix locking for unit attention handling Make sure we have the host lock held when calling scsi_report_bus_reset. Fixes a crash seen as the __devices list in the scsi host was changing as we were iterating through it. Cc: Reviewed-by: Wen Xiong Reviewed-by: Gabriel Krisman Bertazi Signed-off-by: Brian King Reviewed-by: Martin K. Petersen Signed-off-by: James Bottomley diff --git a/drivers/scsi/ipr.c b/drivers/scsi/ipr.c index 8827448..486ffaf 100644 --- a/drivers/scsi/ipr.c +++ b/drivers/scsi/ipr.c @@ -6263,21 +6263,23 @@ static void ipr_scsi_done(struct ipr_cmnd *ipr_cmd) struct ipr_ioa_cfg *ioa_cfg = ipr_cmd->ioa_cfg; struct scsi_cmnd *scsi_cmd = ipr_cmd->scsi_cmd; u32 ioasc = be32_to_cpu(ipr_cmd->s.ioasa.hdr.ioasc); - unsigned long hrrq_flags; + unsigned long lock_flags; scsi_set_resid(scsi_cmd, be32_to_cpu(ipr_cmd->s.ioasa.hdr.residual_data_len)); if (likely(IPR_IOASC_SENSE_KEY(ioasc) == 0)) { scsi_dma_unmap(scsi_cmd); - spin_lock_irqsave(ipr_cmd->hrrq->lock, hrrq_flags); + spin_lock_irqsave(ipr_cmd->hrrq->lock, lock_flags); list_add_tail(&ipr_cmd->queue, &ipr_cmd->hrrq->hrrq_free_q); scsi_cmd->scsi_done(scsi_cmd); - spin_unlock_irqrestore(ipr_cmd->hrrq->lock, hrrq_flags); + spin_unlock_irqrestore(ipr_cmd->hrrq->lock, lock_flags); } else { - spin_lock_irqsave(ipr_cmd->hrrq->lock, hrrq_flags); + spin_lock_irqsave(ioa_cfg->host->host_lock, lock_flags); + spin_lock(&ipr_cmd->hrrq->_lock); ipr_erp_start(ioa_cfg, ipr_cmd); - spin_unlock_irqrestore(ipr_cmd->hrrq->lock, hrrq_flags); + spin_unlock(&ipr_cmd->hrrq->_lock); + spin_unlock_irqrestore(ioa_cfg->host->host_lock, lock_flags); } } -- cgit v0.10.2 From bb7c54339e6a10ecce5c4961adf5e75b3cf0af30 Mon Sep 17 00:00:00 2001 From: Brian King Date: Tue, 14 Jul 2015 11:41:31 -0500 Subject: ipr: Fix incorrect trace indexing When ipr's internal driver trace was changed to an atomic, a signed/unsigned bug slipped in which results in us indexing backwards in our memory buffer writing on memory that does not belong to us. This patch fixes this by removing the modulo and instead just mask off the low bits. Cc: Tested-by: Wen Xiong Reviewed-by: Wen Xiong Reviewed-by: Gabriel Krisman Bertazi Signed-off-by: Brian King Reviewed-by: Martin K. Petersen Signed-off-by: James Bottomley diff --git a/drivers/scsi/ipr.c b/drivers/scsi/ipr.c index 486ffaf..7ae9cd7 100644 --- a/drivers/scsi/ipr.c +++ b/drivers/scsi/ipr.c @@ -599,9 +599,10 @@ static void ipr_trc_hook(struct ipr_cmnd *ipr_cmd, { struct ipr_trace_entry *trace_entry; struct ipr_ioa_cfg *ioa_cfg = ipr_cmd->ioa_cfg; + unsigned int trace_index; - trace_entry = &ioa_cfg->trace[atomic_add_return - (1, &ioa_cfg->trace_index)%IPR_NUM_TRACE_ENTRIES]; + trace_index = atomic_add_return(1, &ioa_cfg->trace_index) & IPR_TRACE_INDEX_MASK; + trace_entry = &ioa_cfg->trace[trace_index]; trace_entry->time = jiffies; trace_entry->op_code = ipr_cmd->ioarcb.cmd_pkt.cdb[0]; trace_entry->type = type; diff --git a/drivers/scsi/ipr.h b/drivers/scsi/ipr.h index 73790a1..6b97ee4 100644 --- a/drivers/scsi/ipr.h +++ b/drivers/scsi/ipr.h @@ -1486,6 +1486,7 @@ struct ipr_ioa_cfg { #define IPR_NUM_TRACE_INDEX_BITS 8 #define IPR_NUM_TRACE_ENTRIES (1 << IPR_NUM_TRACE_INDEX_BITS) +#define IPR_TRACE_INDEX_MASK (IPR_NUM_TRACE_ENTRIES - 1) #define IPR_TRACE_SIZE (sizeof(struct ipr_trace_entry) * IPR_NUM_TRACE_ENTRIES) char trace_start[8]; #define IPR_TRACE_START_LABEL "trace" -- cgit v0.10.2 From 3f1c0581310d5d94bd72740231507e763a6252a4 Mon Sep 17 00:00:00 2001 From: Brian King Date: Tue, 14 Jul 2015 11:41:33 -0500 Subject: ipr: Fix invalid array indexing for HRRQ Fixes another signed / unsigned array indexing bug in the ipr driver. Currently, when hrrq_index wraps, it becomes a negative number. We do the modulo, but still have a negative number, so we end up indexing backwards in the array. Given where the hrrq array is located in memory, we probably won't actually reference memory we don't own, but nonetheless ipr is still looking at data within struct ipr_ioa_cfg and interpreting it as struct ipr_hrr_queue data, so bad things could certainly happen. Each ipr adapter has anywhere from 1 to 16 HRRQs. By default, we use 2 on new adapters. Let's take an example: Assume ioa_cfg->hrrq_index=0x7fffffffe and ioa_cfg->hrrq_num=4: The atomic_add_return will then return -1. We mod this with 3 and get -2, add one and get -1 for an array index. On adapters which support more than a single HRRQ, we dedicate HRRQ to adapter initialization and error interrupts so that we can optimize the other queues for fast path I/O. So all normal I/O uses HRRQ 1-15. So we want to spread the I/O requests across those HRRQs. With the default module parameter settings, this bug won't hit, only when someone sets the ipr.number_of_msix parameter to a value larger than 3 is when bad things start to happen. Cc: Tested-by: Wen Xiong Reviewed-by: Wen Xiong Reviewed-by: Gabriel Krisman Bertazi Signed-off-by: Brian King Reviewed-by: Martin K. Petersen Signed-off-by: James Bottomley diff --git a/drivers/scsi/ipr.c b/drivers/scsi/ipr.c index 7ae9cd7..a9aa389 100644 --- a/drivers/scsi/ipr.c +++ b/drivers/scsi/ipr.c @@ -1052,10 +1052,15 @@ static void ipr_send_blocking_cmd(struct ipr_cmnd *ipr_cmd, static int ipr_get_hrrq_index(struct ipr_ioa_cfg *ioa_cfg) { + unsigned int hrrq; + if (ioa_cfg->hrrq_num == 1) - return 0; - else - return (atomic_add_return(1, &ioa_cfg->hrrq_index) % (ioa_cfg->hrrq_num - 1)) + 1; + hrrq = 0; + else { + hrrq = atomic_add_return(1, &ioa_cfg->hrrq_index); + hrrq = (hrrq % (ioa_cfg->hrrq_num - 1)) + 1; + } + return hrrq; } /** -- cgit v0.10.2 From 0c958ecc69c277b25f38f72bc6d18ab145e8167c Mon Sep 17 00:00:00 2001 From: Tony Battersby Date: Thu, 16 Jul 2015 11:40:41 -0400 Subject: scsi: fix memory leak with scsi-mq Fix a memory leak with scsi-mq triggered by commands with large data transfer length. __sg_alloc_table() sets both table->nents and table->orig_nents to the same value. When the scatterlist is DMA-mapped, table->nents is overwritten with the (possibly smaller) size of the DMA-mapped scatterlist, while table->orig_nents retains the original size of the allocated scatterlist. scsi_free_sgtable() should therefore check orig_nents instead of nents, and all code that initializes sdb->table without calling __sg_alloc_table() should set both nents and orig_nents. Fixes: d285203cf647 ("scsi: add support for a blk-mq based I/O path.") Cc: # 3.17+ Signed-off-by: Tony Battersby Reviewed-by: Christoph Hellwig Reviewed-by: Ewan D. Milne Signed-off-by: James Bottomley diff --git a/drivers/scsi/scsi_error.c b/drivers/scsi/scsi_error.c index 106884a..cfadcce 100644 --- a/drivers/scsi/scsi_error.c +++ b/drivers/scsi/scsi_error.c @@ -944,7 +944,7 @@ void scsi_eh_prep_cmnd(struct scsi_cmnd *scmd, struct scsi_eh_save *ses, scmd->sdb.length); scmd->sdb.table.sgl = &ses->sense_sgl; scmd->sc_data_direction = DMA_FROM_DEVICE; - scmd->sdb.table.nents = 1; + scmd->sdb.table.nents = scmd->sdb.table.orig_nents = 1; scmd->cmnd[0] = REQUEST_SENSE; scmd->cmnd[4] = scmd->sdb.length; scmd->cmd_len = COMMAND_SIZE(scmd->cmnd[0]); diff --git a/drivers/scsi/scsi_lib.c b/drivers/scsi/scsi_lib.c index b1a2631..448ebda 100644 --- a/drivers/scsi/scsi_lib.c +++ b/drivers/scsi/scsi_lib.c @@ -583,7 +583,7 @@ static struct scatterlist *scsi_sg_alloc(unsigned int nents, gfp_t gfp_mask) static void scsi_free_sgtable(struct scsi_data_buffer *sdb, bool mq) { - if (mq && sdb->table.nents <= SCSI_MAX_SG_SEGMENTS) + if (mq && sdb->table.orig_nents <= SCSI_MAX_SG_SEGMENTS) return; __sg_free_table(&sdb->table, SCSI_MAX_SG_SEGMENTS, mq, scsi_sg_free); } @@ -597,8 +597,8 @@ static int scsi_alloc_sgtable(struct scsi_data_buffer *sdb, int nents, bool mq) if (mq) { if (nents <= SCSI_MAX_SG_SEGMENTS) { - sdb->table.nents = nents; - sg_init_table(sdb->table.sgl, sdb->table.nents); + sdb->table.nents = sdb->table.orig_nents = nents; + sg_init_table(sdb->table.sgl, nents); return 0; } first_chunk = sdb->table.sgl; -- cgit v0.10.2 From 646c4b75494747887f936513b669bb8a2d794459 Mon Sep 17 00:00:00 2001 From: Jiang Liu Date: Thu, 30 Jul 2015 15:51:32 +0800 Subject: x86/irq: Use the caller provided polarity setting in mp_check_pin_attr() Commit d32932d02e18 ("x86/irq: Convert IOAPIC to use hierarchical irqdomain interfaces") introduced a regression which causes malfunction of interrupt lines. The reason is that the conversion of mp_check_pin_attr() missed to update the polarity selection of the interrupt pin with the caller provided setting and instead uses a stale attribute value. That in turn results in chosing the wrong interrupt flow handler. Use the caller supplied setting to configure the pin correctly which also choses the correct interrupt flow handler. This restores the original behaviour and on the affected machine/driver (Surface Pro 3, i2c controller) all IOAPIC IRQ configuration are identical to v4.1. Fixes: d32932d02e18 ("x86/irq: Convert IOAPIC to use hierarchical irqdomain interfaces") Reported-and-tested-by: Matt Fleming Reported-and-tested-by: Chen Yu Signed-off-by: Jiang Liu Cc: Tony Luck Cc: Chen Yu Cc: Yinghai Lu Link: http://lkml.kernel.org/r/1438242695-23531-1-git-send-email-jiang.liu@linux.intel.com Signed-off-by: Thomas Gleixner diff --git a/arch/x86/kernel/apic/io_apic.c b/arch/x86/kernel/apic/io_apic.c index 845dc0d..206052e 100644 --- a/arch/x86/kernel/apic/io_apic.c +++ b/arch/x86/kernel/apic/io_apic.c @@ -943,7 +943,7 @@ static bool mp_check_pin_attr(int irq, struct irq_alloc_info *info) */ if (irq < nr_legacy_irqs() && data->count == 1) { if (info->ioapic_trigger != data->trigger) - mp_register_handler(irq, data->trigger); + mp_register_handler(irq, info->ioapic_trigger); data->entry.trigger = data->trigger = info->ioapic_trigger; data->entry.polarity = data->polarity = info->ioapic_polarity; } -- cgit v0.10.2 From 54d9ffc416022cf82649fe5665875d8962dfac41 Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Thu, 30 Jul 2015 12:38:06 +0200 Subject: MAINTAINERS: Appoint Marc Zyngier as irqchips co-maintainer Signed-off-by: Thomas Gleixner Acked-by: Jason Cooper Cc: Marc Zyngier diff --git a/MAINTAINERS b/MAINTAINERS index 9289ecb..577be68 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -5600,6 +5600,7 @@ F: kernel/irq/ IRQCHIP DRIVERS M: Thomas Gleixner M: Jason Cooper +M: Marc Zyngier L: linux-kernel@vger.kernel.org S: Maintained T: git git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip.git irq/core -- cgit v0.10.2 From b09dec2c77dc0ef0bb43ade1628ab4b390e74c6d Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Thu, 30 Jul 2015 12:40:55 +0200 Subject: MAINTAINERS: Appoint Jiang and Marc as irqdomain maintainers Ben was pretty surprised that he is still listed as the maintainer and he has no objections against transferring the duty to those who rumaged in and revamped that code in the recent past. Add kernel/irq/msi.c to the affected files as it's part of the shiny new hierarchical irqdomain machinery. Signed-off-by: Thomas Gleixner Acked-by: Benjamin Herrenschmidt Cc: Jiang Liu Cc: Marc Zyngier Cc: Grant Likely diff --git a/MAINTAINERS b/MAINTAINERS index 577be68..a9ae6c1 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -5609,11 +5609,14 @@ F: Documentation/devicetree/bindings/interrupt-controller/ F: drivers/irqchip/ IRQ DOMAINS (IRQ NUMBER MAPPING LIBRARY) -M: Benjamin Herrenschmidt +M: Jiang Liu +M: Marc Zyngier S: Maintained +T: git git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip.git irq/core F: Documentation/IRQ-domain.txt F: include/linux/irqdomain.h F: kernel/irq/irqdomain.c +F: kernel/irq/msi.c ISAPNP M: Jaroslav Kysela -- cgit v0.10.2 From 649ccd08534ee26deb2e5b08509800d0e95167f5 Mon Sep 17 00:00:00 2001 From: Takashi Iwai Date: Thu, 30 Jul 2015 22:30:29 +0200 Subject: ALSA: hda - Fix MacBook Pro 5,2 quirk MacBook Pro 5,2 with ALC889 codec had already a fixup entry, but this seems not working correctly, a fix for pin NID 0x15 is needed in addition. It's equivalent with the fixup for MacBook Air 1,1, so use this instead. Bugzilla: https://bugzilla.kernel.org/show_bug.cgi?id=102131 Reported-and-tested-by: Jeffery Miller Cc: Signed-off-by: Takashi Iwai diff --git a/sound/pci/hda/patch_realtek.c b/sound/pci/hda/patch_realtek.c index 18ae17e..c456c04 100644 --- a/sound/pci/hda/patch_realtek.c +++ b/sound/pci/hda/patch_realtek.c @@ -2222,7 +2222,7 @@ static const struct snd_pci_quirk alc882_fixup_tbl[] = { SND_PCI_QUIRK(0x106b, 0x4300, "iMac 9,1", ALC889_FIXUP_IMAC91_VREF), SND_PCI_QUIRK(0x106b, 0x4600, "MacbookPro 5,2", ALC889_FIXUP_IMAC91_VREF), SND_PCI_QUIRK(0x106b, 0x4900, "iMac 9,1 Aluminum", ALC889_FIXUP_IMAC91_VREF), - SND_PCI_QUIRK(0x106b, 0x4a00, "Macbook 5,2", ALC889_FIXUP_IMAC91_VREF), + SND_PCI_QUIRK(0x106b, 0x4a00, "Macbook 5,2", ALC889_FIXUP_MBA11_VREF), SND_PCI_QUIRK(0x1071, 0x8258, "Evesham Voyaeger", ALC882_FIXUP_EAPD), SND_PCI_QUIRK(0x1462, 0x7350, "MSI-7350", ALC889_FIXUP_CD), -- cgit v0.10.2 From e501139a515018f913761d3bf4239313c87e721c Mon Sep 17 00:00:00 2001 From: hayeswang Date: Wed, 29 Jul 2015 20:39:08 +0800 Subject: r8152: add pre_reset and post_reset Add rtl8152_pre_reset() and rtl8152_post_reset() which are used when calling usb_reset_device(). The two functions could reduce the time of reset when calling usb_reset_device() after probe(). Signed-off-by: Hayes Wang Signed-off-by: David S. Miller diff --git a/drivers/net/usb/r8152.c b/drivers/net/usb/r8152.c index 144dc64..e1b6d6d 100644 --- a/drivers/net/usb/r8152.c +++ b/drivers/net/usb/r8152.c @@ -3342,6 +3342,58 @@ static void r8153_init(struct r8152 *tp) r8153_u2p3en(tp, true); } +static int rtl8152_pre_reset(struct usb_interface *intf) +{ + struct r8152 *tp = usb_get_intfdata(intf); + struct net_device *netdev; + + if (!tp) + return 0; + + netdev = tp->netdev; + if (!netif_running(netdev)) + return 0; + + napi_disable(&tp->napi); + clear_bit(WORK_ENABLE, &tp->flags); + usb_kill_urb(tp->intr_urb); + cancel_delayed_work_sync(&tp->schedule); + if (netif_carrier_ok(netdev)) { + netif_stop_queue(netdev); + mutex_lock(&tp->control); + tp->rtl_ops.disable(tp); + mutex_unlock(&tp->control); + } + + return 0; +} + +static int rtl8152_post_reset(struct usb_interface *intf) +{ + struct r8152 *tp = usb_get_intfdata(intf); + struct net_device *netdev; + + if (!tp) + return 0; + + netdev = tp->netdev; + if (!netif_running(netdev)) + return 0; + + set_bit(WORK_ENABLE, &tp->flags); + if (netif_carrier_ok(netdev)) { + mutex_lock(&tp->control); + tp->rtl_ops.enable(tp); + rtl8152_set_rx_mode(netdev); + mutex_unlock(&tp->control); + netif_wake_queue(netdev); + } + + napi_enable(&tp->napi); + + return 0; +} + static int rtl8152_suspend(struct usb_interface *intf, pm_message_t message) { struct r8152 *tp = usb_get_intfdata(intf); @@ -4164,6 +4216,8 @@ static struct usb_driver rtl8152_driver = { .suspend = rtl8152_suspend, .resume = rtl8152_resume, .reset_resume = rtl8152_resume, + .pre_reset = rtl8152_pre_reset, + .post_reset = rtl8152_post_reset, .supports_autosuspend = 1, .disable_hub_initiated_lpm = 1, }; -- cgit v0.10.2 From 37608f3e57dda037a230afb7dc8da9a63f100e06 Mon Sep 17 00:00:00 2001 From: hayeswang Date: Wed, 29 Jul 2015 20:39:09 +0800 Subject: r8152: reset device when tx timeout The device reset is necessary if the hw becomes abnormal and stops transmitting packets. Signed-off-by: Hayes Wang Signed-off-by: David S. Miller diff --git a/drivers/net/usb/r8152.c b/drivers/net/usb/r8152.c index e1b6d6d..ad8cbc6 100644 --- a/drivers/net/usb/r8152.c +++ b/drivers/net/usb/r8152.c @@ -27,7 +27,7 @@ #include /* Version Information */ -#define DRIVER_VERSION "v1.08.0 (2015/01/13)" +#define DRIVER_VERSION "v1.08.1 (2015/07/28)" #define DRIVER_AUTHOR "Realtek linux nic maintainers " #define DRIVER_DESC "Realtek RTL8152/RTL8153 Based USB Ethernet Adapters" #define MODULENAME "r8152" @@ -1902,11 +1902,10 @@ static void rtl_drop_queued_tx(struct r8152 *tp) static void rtl8152_tx_timeout(struct net_device *netdev) { struct r8152 *tp = netdev_priv(netdev); - int i; netif_warn(tp, tx_err, netdev, "Tx timeout\n"); - for (i = 0; i < RTL8152_MAX_TX; i++) - usb_unlink_urb(tp->tx_info[i].urb); + + usb_queue_reset_device(tp->intf); } static void rtl8152_set_rx_mode(struct net_device *netdev) -- cgit v0.10.2 From 28e6b67f0b292f557468c139085303b15f1a678f Mon Sep 17 00:00:00 2001 From: Daniel Borkmann Date: Wed, 29 Jul 2015 23:35:25 +0200 Subject: net: sched: fix refcount imbalance in actions Since commit 55334a5db5cd ("net_sched: act: refuse to remove bound action outside"), we end up with a wrong reference count for a tc action. Test case 1: FOO="1,6 0 0 4294967295," BAR="1,6 0 0 4294967294," tc filter add dev foo parent 1: bpf bytecode "$FOO" flowid 1:1 \ action bpf bytecode "$FOO" tc actions show action bpf action order 0: bpf bytecode '1,6 0 0 4294967295' default-action pipe index 1 ref 1 bind 1 tc actions replace action bpf bytecode "$BAR" index 1 tc actions show action bpf action order 0: bpf bytecode '1,6 0 0 4294967294' default-action pipe index 1 ref 2 bind 1 tc actions replace action bpf bytecode "$FOO" index 1 tc actions show action bpf action order 0: bpf bytecode '1,6 0 0 4294967295' default-action pipe index 1 ref 3 bind 1 Test case 2: FOO="1,6 0 0 4294967295," tc filter add dev foo parent 1: bpf bytecode "$FOO" flowid 1:1 action ok tc actions show action gact action order 0: gact action pass random type none pass val 0 index 1 ref 1 bind 1 tc actions add action drop index 1 RTNETLINK answers: File exists [...] tc actions show action gact action order 0: gact action pass random type none pass val 0 index 1 ref 2 bind 1 tc actions add action drop index 1 RTNETLINK answers: File exists [...] tc actions show action gact action order 0: gact action pass random type none pass val 0 index 1 ref 3 bind 1 What happens is that in tcf_hash_check(), we check tcf_common for a given index and increase tcfc_refcnt and conditionally tcfc_bindcnt when we've found an existing action. Now there are the following cases: 1) We do a late binding of an action. In that case, we leave the tcfc_refcnt/tcfc_bindcnt increased and are done with the ->init() handler. This is correctly handeled. 2) We replace the given action, or we try to add one without replacing and find out that the action at a specific index already exists (thus, we go out with error in that case). In case of 2), we have to undo the reference count increase from tcf_hash_check() in the tcf_hash_check() function. Currently, we fail to do so because of the 'tcfc_bindcnt > 0' check which bails out early with an -EPERM error. Now, while commit 55334a5db5cd prevents 'tc actions del action ...' on an already classifier-bound action to drop the reference count (which could then become negative, wrap around etc), this restriction only accounts for invocations outside a specific action's ->init() handler. One possible solution would be to add a flag thus we possibly trigger the -EPERM ony in situations where it is indeed relevant. After the patch, above test cases have correct reference count again. Fixes: 55334a5db5cd ("net_sched: act: refuse to remove bound action outside") Signed-off-by: Daniel Borkmann Reviewed-by: Cong Wang Signed-off-by: David S. Miller diff --git a/include/net/act_api.h b/include/net/act_api.h index 3ee4c92..931738b 100644 --- a/include/net/act_api.h +++ b/include/net/act_api.h @@ -99,7 +99,6 @@ struct tc_action_ops { int tcf_hash_search(struct tc_action *a, u32 index); void tcf_hash_destroy(struct tc_action *a); -int tcf_hash_release(struct tc_action *a, int bind); u32 tcf_hash_new_index(struct tcf_hashinfo *hinfo); int tcf_hash_check(u32 index, struct tc_action *a, int bind); int tcf_hash_create(u32 index, struct nlattr *est, struct tc_action *a, @@ -107,6 +106,13 @@ int tcf_hash_create(u32 index, struct nlattr *est, struct tc_action *a, void tcf_hash_cleanup(struct tc_action *a, struct nlattr *est); void tcf_hash_insert(struct tc_action *a); +int __tcf_hash_release(struct tc_action *a, bool bind, bool strict); + +static inline int tcf_hash_release(struct tc_action *a, bool bind) +{ + return __tcf_hash_release(a, bind, false); +} + int tcf_register_action(struct tc_action_ops *a, unsigned int mask); int tcf_unregister_action(struct tc_action_ops *a); int tcf_action_destroy(struct list_head *actions, int bind); diff --git a/net/sched/act_api.c b/net/sched/act_api.c index af427a3..43ec926 100644 --- a/net/sched/act_api.c +++ b/net/sched/act_api.c @@ -45,7 +45,7 @@ void tcf_hash_destroy(struct tc_action *a) } EXPORT_SYMBOL(tcf_hash_destroy); -int tcf_hash_release(struct tc_action *a, int bind) +int __tcf_hash_release(struct tc_action *a, bool bind, bool strict) { struct tcf_common *p = a->priv; int ret = 0; @@ -53,7 +53,7 @@ int tcf_hash_release(struct tc_action *a, int bind) if (p) { if (bind) p->tcfc_bindcnt--; - else if (p->tcfc_bindcnt > 0) + else if (strict && p->tcfc_bindcnt > 0) return -EPERM; p->tcfc_refcnt--; @@ -64,9 +64,10 @@ int tcf_hash_release(struct tc_action *a, int bind) ret = 1; } } + return ret; } -EXPORT_SYMBOL(tcf_hash_release); +EXPORT_SYMBOL(__tcf_hash_release); static int tcf_dump_walker(struct sk_buff *skb, struct netlink_callback *cb, struct tc_action *a) @@ -136,7 +137,7 @@ static int tcf_del_walker(struct sk_buff *skb, struct tc_action *a) head = &hinfo->htab[tcf_hash(i, hinfo->hmask)]; hlist_for_each_entry_safe(p, n, head, tcfc_head) { a->priv = p; - ret = tcf_hash_release(a, 0); + ret = __tcf_hash_release(a, false, true); if (ret == ACT_P_DELETED) { module_put(a->ops->owner); n_i++; @@ -408,7 +409,7 @@ int tcf_action_destroy(struct list_head *actions, int bind) int ret = 0; list_for_each_entry_safe(a, tmp, actions, list) { - ret = tcf_hash_release(a, bind); + ret = __tcf_hash_release(a, bind, true); if (ret == ACT_P_DELETED) module_put(a->ops->owner); else if (ret < 0) -- cgit v0.10.2 From 8a68173691f036613e3d4e6bf8dc129d4a7bf383 Mon Sep 17 00:00:00 2001 From: Sowmini Varadhan Date: Thu, 30 Jul 2015 15:50:36 +0200 Subject: net: sk_clone_lock() should only do get_net() if the parent is not a kernel socket The newsk returned by sk_clone_lock should hold a get_net() reference if, and only if, the parent is not a kernel socket (making this similar to sk_alloc()). E.g,. for the SYN_RECV path, tcp_v4_syn_recv_sock->..inet_csk_clone_lock sets up the syn_recv newsk from sk_clone_lock. When the parent (listen) socket is a kernel socket (defined in sk_alloc() as having sk_net_refcnt == 0), then the newsk should also have a 0 sk_net_refcnt and should not hold a get_net() reference. Fixes: 26abe14379f8 ("net: Modify sk_alloc to not reference count the netns of kernel sockets.") Acked-by: Eric Dumazet Cc: Eric W. Biederman Signed-off-by: Sowmini Varadhan Signed-off-by: David S. Miller diff --git a/net/core/sock.c b/net/core/sock.c index 8a14f12..193901d 100644 --- a/net/core/sock.c +++ b/net/core/sock.c @@ -1497,7 +1497,8 @@ struct sock *sk_clone_lock(const struct sock *sk, const gfp_t priority) sock_copy(newsk, sk); /* SANITY */ - get_net(sock_net(newsk)); + if (likely(newsk->sk_net_refcnt)) + get_net(sock_net(newsk)); sk_node_init(&newsk->sk_node); sock_lock_init(newsk); bh_lock_sock(newsk); -- cgit v0.10.2 From aa1acff356bbedfd03b544051f5b371746735d89 Mon Sep 17 00:00:00 2001 From: Andy Lutomirski Date: Thu, 30 Jul 2015 14:31:31 -0700 Subject: x86/xen: Probe target addresses in set_aliased_prot() before the hypercall The update_va_mapping hypercall can fail if the VA isn't present in the guest's page tables. Under certain loads, this can result in an OOPS when the target address is in unpopulated vmap space. While we're at it, add comments to help explain what's going on. This isn't a great long-term fix. This code should probably be changed to use something like set_memory_ro. Signed-off-by: Andy Lutomirski Cc: Andrew Cooper Cc: Andy Lutomirski Cc: Boris Ostrovsky Cc: Borislav Petkov Cc: Brian Gerst Cc: David Vrabel Cc: Denys Vlasenko Cc: H. Peter Anvin Cc: Jan Beulich Cc: Konrad Rzeszutek Wilk Cc: Linus Torvalds Cc: Peter Zijlstra Cc: Sasha Levin Cc: Steven Rostedt Cc: Thomas Gleixner Cc: security@kernel.org Cc: Cc: xen-devel Link: http://lkml.kernel.org/r/0b0e55b995cda11e7829f140b833ef932fcabe3a.1438291540.git.luto@kernel.org Signed-off-by: Ingo Molnar diff --git a/arch/x86/xen/enlighten.c b/arch/x86/xen/enlighten.c index 0b95c9b..11d6fb4 100644 --- a/arch/x86/xen/enlighten.c +++ b/arch/x86/xen/enlighten.c @@ -483,6 +483,7 @@ static void set_aliased_prot(void *v, pgprot_t prot) pte_t pte; unsigned long pfn; struct page *page; + unsigned char dummy; ptep = lookup_address((unsigned long)v, &level); BUG_ON(ptep == NULL); @@ -492,6 +493,32 @@ static void set_aliased_prot(void *v, pgprot_t prot) pte = pfn_pte(pfn, prot); + /* + * Careful: update_va_mapping() will fail if the virtual address + * we're poking isn't populated in the page tables. We don't + * need to worry about the direct map (that's always in the page + * tables), but we need to be careful about vmap space. In + * particular, the top level page table can lazily propagate + * entries between processes, so if we've switched mms since we + * vmapped the target in the first place, we might not have the + * top-level page table entry populated. + * + * We disable preemption because we want the same mm active when + * we probe the target and when we issue the hypercall. We'll + * have the same nominal mm, but if we're a kernel thread, lazy + * mm dropping could change our pgd. + * + * Out of an abundance of caution, this uses __get_user() to fault + * in the target address just in case there's some obscure case + * in which the target address isn't readable. + */ + + preempt_disable(); + + pagefault_disable(); /* Avoid warnings due to being atomic. */ + __get_user(dummy, (unsigned char __user __force *)v); + pagefault_enable(); + if (HYPERVISOR_update_va_mapping((unsigned long)v, pte, 0)) BUG(); @@ -503,6 +530,8 @@ static void set_aliased_prot(void *v, pgprot_t prot) BUG(); } else kmap_flush_unused(); + + preempt_enable(); } static void xen_alloc_ldt(struct desc_struct *ldt, unsigned entries) @@ -510,6 +539,17 @@ static void xen_alloc_ldt(struct desc_struct *ldt, unsigned entries) const unsigned entries_per_page = PAGE_SIZE / LDT_ENTRY_SIZE; int i; + /* + * We need to mark the all aliases of the LDT pages RO. We + * don't need to call vm_flush_aliases(), though, since that's + * only responsible for flushing aliases out the TLBs, not the + * page tables, and Xen will flush the TLB for us if needed. + * + * To avoid confusing future readers: none of this is necessary + * to load the LDT. The hypervisor only checks this when the + * LDT is faulted in due to subsequent descriptor access. + */ + for(i = 0; i < entries; i += entries_per_page) set_aliased_prot(ldt + i, PAGE_KERNEL_RO); } -- cgit v0.10.2 From 37868fe113ff2ba814b3b4eb12df214df555f8dc Mon Sep 17 00:00:00 2001 From: Andy Lutomirski Date: Thu, 30 Jul 2015 14:31:32 -0700 Subject: x86/ldt: Make modify_ldt synchronous modify_ldt() has questionable locking and does not synchronize threads. Improve it: redesign the locking and synchronize all threads' LDTs using an IPI on all modifications. This will dramatically slow down modify_ldt in multithreaded programs, but there shouldn't be any multithreaded programs that care about modify_ldt's performance in the first place. This fixes some fallout from the CVE-2015-5157 fixes. Signed-off-by: Andy Lutomirski Reviewed-by: Borislav Petkov Cc: Andrew Cooper Cc: Andy Lutomirski Cc: Boris Ostrovsky Cc: Borislav Petkov Cc: Brian Gerst Cc: Denys Vlasenko Cc: H. Peter Anvin Cc: Jan Beulich Cc: Konrad Rzeszutek Wilk Cc: Linus Torvalds Cc: Peter Zijlstra Cc: Sasha Levin Cc: Steven Rostedt Cc: Thomas Gleixner Cc: security@kernel.org Cc: Cc: xen-devel Link: http://lkml.kernel.org/r/4c6978476782160600471bd865b318db34c7b628.1438291540.git.luto@kernel.org Signed-off-by: Ingo Molnar diff --git a/arch/x86/include/asm/desc.h b/arch/x86/include/asm/desc.h index a0bf89f..4e10d73 100644 --- a/arch/x86/include/asm/desc.h +++ b/arch/x86/include/asm/desc.h @@ -280,21 +280,6 @@ static inline void clear_LDT(void) set_ldt(NULL, 0); } -/* - * load one particular LDT into the current CPU - */ -static inline void load_LDT_nolock(mm_context_t *pc) -{ - set_ldt(pc->ldt, pc->size); -} - -static inline void load_LDT(mm_context_t *pc) -{ - preempt_disable(); - load_LDT_nolock(pc); - preempt_enable(); -} - static inline unsigned long get_desc_base(const struct desc_struct *desc) { return (unsigned)(desc->base0 | ((desc->base1) << 16) | ((desc->base2) << 24)); diff --git a/arch/x86/include/asm/mmu.h b/arch/x86/include/asm/mmu.h index 09b9620..364d274 100644 --- a/arch/x86/include/asm/mmu.h +++ b/arch/x86/include/asm/mmu.h @@ -9,8 +9,7 @@ * we put the segment information here. */ typedef struct { - void *ldt; - int size; + struct ldt_struct *ldt; #ifdef CONFIG_X86_64 /* True if mm supports a task running in 32 bit compatibility mode. */ diff --git a/arch/x86/include/asm/mmu_context.h b/arch/x86/include/asm/mmu_context.h index 804a3a6..984abfe 100644 --- a/arch/x86/include/asm/mmu_context.h +++ b/arch/x86/include/asm/mmu_context.h @@ -34,6 +34,50 @@ static inline void load_mm_cr4(struct mm_struct *mm) {} #endif /* + * ldt_structs can be allocated, used, and freed, but they are never + * modified while live. + */ +struct ldt_struct { + /* + * Xen requires page-aligned LDTs with special permissions. This is + * needed to prevent us from installing evil descriptors such as + * call gates. On native, we could merge the ldt_struct and LDT + * allocations, but it's not worth trying to optimize. + */ + struct desc_struct *entries; + int size; +}; + +static inline void load_mm_ldt(struct mm_struct *mm) +{ + struct ldt_struct *ldt; + + /* lockless_dereference synchronizes with smp_store_release */ + ldt = lockless_dereference(mm->context.ldt); + + /* + * Any change to mm->context.ldt is followed by an IPI to all + * CPUs with the mm active. The LDT will not be freed until + * after the IPI is handled by all such CPUs. This means that, + * if the ldt_struct changes before we return, the values we see + * will be safe, and the new values will be loaded before we run + * any user code. + * + * NB: don't try to convert this to use RCU without extreme care. + * We would still need IRQs off, because we don't want to change + * the local LDT after an IPI loaded a newer value than the one + * that we can see. + */ + + if (unlikely(ldt)) + set_ldt(ldt->entries, ldt->size); + else + clear_LDT(); + + DEBUG_LOCKS_WARN_ON(preemptible()); +} + +/* * Used for LDT copy/destruction. */ int init_new_context(struct task_struct *tsk, struct mm_struct *mm); @@ -78,12 +122,12 @@ static inline void switch_mm(struct mm_struct *prev, struct mm_struct *next, * was called and then modify_ldt changed * prev->context.ldt but suppressed an IPI to this CPU. * In this case, prev->context.ldt != NULL, because we - * never free an LDT while the mm still exists. That - * means that next->context.ldt != prev->context.ldt, - * because mms never share an LDT. + * never set context.ldt to NULL while the mm still + * exists. That means that next->context.ldt != + * prev->context.ldt, because mms never share an LDT. */ if (unlikely(prev->context.ldt != next->context.ldt)) - load_LDT_nolock(&next->context); + load_mm_ldt(next); } #ifdef CONFIG_SMP else { @@ -106,7 +150,7 @@ static inline void switch_mm(struct mm_struct *prev, struct mm_struct *next, load_cr3(next->pgd); trace_tlb_flush(TLB_FLUSH_ON_TASK_SWITCH, TLB_FLUSH_ALL); load_mm_cr4(next); - load_LDT_nolock(&next->context); + load_mm_ldt(next); } } #endif diff --git a/arch/x86/kernel/cpu/common.c b/arch/x86/kernel/cpu/common.c index 922c5e0..cb9e5df 100644 --- a/arch/x86/kernel/cpu/common.c +++ b/arch/x86/kernel/cpu/common.c @@ -1410,7 +1410,7 @@ void cpu_init(void) load_sp0(t, ¤t->thread); set_tss_desc(cpu, t); load_TR_desc(); - load_LDT(&init_mm.context); + load_mm_ldt(&init_mm); clear_all_debug_regs(); dbg_restore_debug_regs(); @@ -1459,7 +1459,7 @@ void cpu_init(void) load_sp0(t, thread); set_tss_desc(cpu, t); load_TR_desc(); - load_LDT(&init_mm.context); + load_mm_ldt(&init_mm); t->x86_tss.io_bitmap_base = offsetof(struct tss_struct, io_bitmap); diff --git a/arch/x86/kernel/cpu/perf_event.c b/arch/x86/kernel/cpu/perf_event.c index 3658de4..9469dfa 100644 --- a/arch/x86/kernel/cpu/perf_event.c +++ b/arch/x86/kernel/cpu/perf_event.c @@ -2179,21 +2179,25 @@ static unsigned long get_segment_base(unsigned int segment) int idx = segment >> 3; if ((segment & SEGMENT_TI_MASK) == SEGMENT_LDT) { + struct ldt_struct *ldt; + if (idx > LDT_ENTRIES) return 0; - if (idx > current->active_mm->context.size) + /* IRQs are off, so this synchronizes with smp_store_release */ + ldt = lockless_dereference(current->active_mm->context.ldt); + if (!ldt || idx > ldt->size) return 0; - desc = current->active_mm->context.ldt; + desc = &ldt->entries[idx]; } else { if (idx > GDT_ENTRIES) return 0; - desc = raw_cpu_ptr(gdt_page.gdt); + desc = raw_cpu_ptr(gdt_page.gdt) + idx; } - return get_desc_base(desc + idx); + return get_desc_base(desc); } #ifdef CONFIG_COMPAT diff --git a/arch/x86/kernel/ldt.c b/arch/x86/kernel/ldt.c index c37886d..2bcc052 100644 --- a/arch/x86/kernel/ldt.c +++ b/arch/x86/kernel/ldt.c @@ -12,6 +12,7 @@ #include #include #include +#include #include #include @@ -20,82 +21,82 @@ #include #include -#ifdef CONFIG_SMP +/* context.lock is held for us, so we don't need any locking. */ static void flush_ldt(void *current_mm) { - if (current->active_mm == current_mm) - load_LDT(¤t->active_mm->context); + mm_context_t *pc; + + if (current->active_mm != current_mm) + return; + + pc = ¤t->active_mm->context; + set_ldt(pc->ldt->entries, pc->ldt->size); } -#endif -static int alloc_ldt(mm_context_t *pc, int mincount, int reload) +/* The caller must call finalize_ldt_struct on the result. LDT starts zeroed. */ +static struct ldt_struct *alloc_ldt_struct(int size) { - void *oldldt, *newldt; - int oldsize; - - if (mincount <= pc->size) - return 0; - oldsize = pc->size; - mincount = (mincount + (PAGE_SIZE / LDT_ENTRY_SIZE - 1)) & - (~(PAGE_SIZE / LDT_ENTRY_SIZE - 1)); - if (mincount * LDT_ENTRY_SIZE > PAGE_SIZE) - newldt = vmalloc(mincount * LDT_ENTRY_SIZE); + struct ldt_struct *new_ldt; + int alloc_size; + + if (size > LDT_ENTRIES) + return NULL; + + new_ldt = kmalloc(sizeof(struct ldt_struct), GFP_KERNEL); + if (!new_ldt) + return NULL; + + BUILD_BUG_ON(LDT_ENTRY_SIZE != sizeof(struct desc_struct)); + alloc_size = size * LDT_ENTRY_SIZE; + + /* + * Xen is very picky: it requires a page-aligned LDT that has no + * trailing nonzero bytes in any page that contains LDT descriptors. + * Keep it simple: zero the whole allocation and never allocate less + * than PAGE_SIZE. + */ + if (alloc_size > PAGE_SIZE) + new_ldt->entries = vzalloc(alloc_size); else - newldt = (void *)__get_free_page(GFP_KERNEL); - - if (!newldt) - return -ENOMEM; + new_ldt->entries = kzalloc(PAGE_SIZE, GFP_KERNEL); - if (oldsize) - memcpy(newldt, pc->ldt, oldsize * LDT_ENTRY_SIZE); - oldldt = pc->ldt; - memset(newldt + oldsize * LDT_ENTRY_SIZE, 0, - (mincount - oldsize) * LDT_ENTRY_SIZE); + if (!new_ldt->entries) { + kfree(new_ldt); + return NULL; + } - paravirt_alloc_ldt(newldt, mincount); + new_ldt->size = size; + return new_ldt; +} -#ifdef CONFIG_X86_64 - /* CHECKME: Do we really need this ? */ - wmb(); -#endif - pc->ldt = newldt; - wmb(); - pc->size = mincount; - wmb(); - - if (reload) { -#ifdef CONFIG_SMP - preempt_disable(); - load_LDT(pc); - if (!cpumask_equal(mm_cpumask(current->mm), - cpumask_of(smp_processor_id()))) - smp_call_function(flush_ldt, current->mm, 1); - preempt_enable(); -#else - load_LDT(pc); -#endif - } - if (oldsize) { - paravirt_free_ldt(oldldt, oldsize); - if (oldsize * LDT_ENTRY_SIZE > PAGE_SIZE) - vfree(oldldt); - else - put_page(virt_to_page(oldldt)); - } - return 0; +/* After calling this, the LDT is immutable. */ +static void finalize_ldt_struct(struct ldt_struct *ldt) +{ + paravirt_alloc_ldt(ldt->entries, ldt->size); } -static inline int copy_ldt(mm_context_t *new, mm_context_t *old) +/* context.lock is held */ +static void install_ldt(struct mm_struct *current_mm, + struct ldt_struct *ldt) { - int err = alloc_ldt(new, old->size, 0); - int i; + /* Synchronizes with lockless_dereference in load_mm_ldt. */ + smp_store_release(¤t_mm->context.ldt, ldt); + + /* Activate the LDT for all CPUs using current_mm. */ + on_each_cpu_mask(mm_cpumask(current_mm), flush_ldt, current_mm, true); +} - if (err < 0) - return err; +static void free_ldt_struct(struct ldt_struct *ldt) +{ + if (likely(!ldt)) + return; - for (i = 0; i < old->size; i++) - write_ldt_entry(new->ldt, i, old->ldt + i * LDT_ENTRY_SIZE); - return 0; + paravirt_free_ldt(ldt->entries, ldt->size); + if (ldt->size * LDT_ENTRY_SIZE > PAGE_SIZE) + vfree(ldt->entries); + else + kfree(ldt->entries); + kfree(ldt); } /* @@ -104,17 +105,37 @@ static inline int copy_ldt(mm_context_t *new, mm_context_t *old) */ int init_new_context(struct task_struct *tsk, struct mm_struct *mm) { + struct ldt_struct *new_ldt; struct mm_struct *old_mm; int retval = 0; mutex_init(&mm->context.lock); - mm->context.size = 0; old_mm = current->mm; - if (old_mm && old_mm->context.size > 0) { - mutex_lock(&old_mm->context.lock); - retval = copy_ldt(&mm->context, &old_mm->context); - mutex_unlock(&old_mm->context.lock); + if (!old_mm) { + mm->context.ldt = NULL; + return 0; } + + mutex_lock(&old_mm->context.lock); + if (!old_mm->context.ldt) { + mm->context.ldt = NULL; + goto out_unlock; + } + + new_ldt = alloc_ldt_struct(old_mm->context.ldt->size); + if (!new_ldt) { + retval = -ENOMEM; + goto out_unlock; + } + + memcpy(new_ldt->entries, old_mm->context.ldt->entries, + new_ldt->size * LDT_ENTRY_SIZE); + finalize_ldt_struct(new_ldt); + + mm->context.ldt = new_ldt; + +out_unlock: + mutex_unlock(&old_mm->context.lock); return retval; } @@ -125,53 +146,47 @@ int init_new_context(struct task_struct *tsk, struct mm_struct *mm) */ void destroy_context(struct mm_struct *mm) { - if (mm->context.size) { -#ifdef CONFIG_X86_32 - /* CHECKME: Can this ever happen ? */ - if (mm == current->active_mm) - clear_LDT(); -#endif - paravirt_free_ldt(mm->context.ldt, mm->context.size); - if (mm->context.size * LDT_ENTRY_SIZE > PAGE_SIZE) - vfree(mm->context.ldt); - else - put_page(virt_to_page(mm->context.ldt)); - mm->context.size = 0; - } + free_ldt_struct(mm->context.ldt); + mm->context.ldt = NULL; } static int read_ldt(void __user *ptr, unsigned long bytecount) { - int err; + int retval; unsigned long size; struct mm_struct *mm = current->mm; - if (!mm->context.size) - return 0; + mutex_lock(&mm->context.lock); + + if (!mm->context.ldt) { + retval = 0; + goto out_unlock; + } + if (bytecount > LDT_ENTRY_SIZE * LDT_ENTRIES) bytecount = LDT_ENTRY_SIZE * LDT_ENTRIES; - mutex_lock(&mm->context.lock); - size = mm->context.size * LDT_ENTRY_SIZE; + size = mm->context.ldt->size * LDT_ENTRY_SIZE; if (size > bytecount) size = bytecount; - err = 0; - if (copy_to_user(ptr, mm->context.ldt, size)) - err = -EFAULT; - mutex_unlock(&mm->context.lock); - if (err < 0) - goto error_return; + if (copy_to_user(ptr, mm->context.ldt->entries, size)) { + retval = -EFAULT; + goto out_unlock; + } + if (size != bytecount) { - /* zero-fill the rest */ - if (clear_user(ptr + size, bytecount - size) != 0) { - err = -EFAULT; - goto error_return; + /* Zero-fill the rest and pretend we read bytecount bytes. */ + if (clear_user(ptr + size, bytecount - size)) { + retval = -EFAULT; + goto out_unlock; } } - return bytecount; -error_return: - return err; + retval = bytecount; + +out_unlock: + mutex_unlock(&mm->context.lock); + return retval; } static int read_default_ldt(void __user *ptr, unsigned long bytecount) @@ -195,6 +210,8 @@ static int write_ldt(void __user *ptr, unsigned long bytecount, int oldmode) struct desc_struct ldt; int error; struct user_desc ldt_info; + int oldsize, newsize; + struct ldt_struct *new_ldt, *old_ldt; error = -EINVAL; if (bytecount != sizeof(ldt_info)) @@ -213,34 +230,39 @@ static int write_ldt(void __user *ptr, unsigned long bytecount, int oldmode) goto out; } - mutex_lock(&mm->context.lock); - if (ldt_info.entry_number >= mm->context.size) { - error = alloc_ldt(¤t->mm->context, - ldt_info.entry_number + 1, 1); - if (error < 0) - goto out_unlock; - } - - /* Allow LDTs to be cleared by the user. */ - if (ldt_info.base_addr == 0 && ldt_info.limit == 0) { - if (oldmode || LDT_empty(&ldt_info)) { - memset(&ldt, 0, sizeof(ldt)); - goto install; + if ((oldmode && !ldt_info.base_addr && !ldt_info.limit) || + LDT_empty(&ldt_info)) { + /* The user wants to clear the entry. */ + memset(&ldt, 0, sizeof(ldt)); + } else { + if (!IS_ENABLED(CONFIG_X86_16BIT) && !ldt_info.seg_32bit) { + error = -EINVAL; + goto out; } + + fill_ldt(&ldt, &ldt_info); + if (oldmode) + ldt.avl = 0; } - if (!IS_ENABLED(CONFIG_X86_16BIT) && !ldt_info.seg_32bit) { - error = -EINVAL; + mutex_lock(&mm->context.lock); + + old_ldt = mm->context.ldt; + oldsize = old_ldt ? old_ldt->size : 0; + newsize = max((int)(ldt_info.entry_number + 1), oldsize); + + error = -ENOMEM; + new_ldt = alloc_ldt_struct(newsize); + if (!new_ldt) goto out_unlock; - } - fill_ldt(&ldt, &ldt_info); - if (oldmode) - ldt.avl = 0; + if (old_ldt) + memcpy(new_ldt->entries, old_ldt->entries, oldsize * LDT_ENTRY_SIZE); + new_ldt->entries[ldt_info.entry_number] = ldt; + finalize_ldt_struct(new_ldt); - /* Install the new entry ... */ -install: - write_ldt_entry(mm->context.ldt, ldt_info.entry_number, &ldt); + install_ldt(mm, new_ldt); + free_ldt_struct(old_ldt); error = 0; out_unlock: diff --git a/arch/x86/kernel/process_64.c b/arch/x86/kernel/process_64.c index 71d7849..f6b9163 100644 --- a/arch/x86/kernel/process_64.c +++ b/arch/x86/kernel/process_64.c @@ -121,11 +121,11 @@ void __show_regs(struct pt_regs *regs, int all) void release_thread(struct task_struct *dead_task) { if (dead_task->mm) { - if (dead_task->mm->context.size) { + if (dead_task->mm->context.ldt) { pr_warn("WARNING: dead process %s still has LDT? <%p/%d>\n", dead_task->comm, dead_task->mm->context.ldt, - dead_task->mm->context.size); + dead_task->mm->context.ldt->size); BUG(); } } diff --git a/arch/x86/kernel/step.c b/arch/x86/kernel/step.c index 9b4d51d..6273324 100644 --- a/arch/x86/kernel/step.c +++ b/arch/x86/kernel/step.c @@ -5,6 +5,7 @@ #include #include #include +#include unsigned long convert_ip_to_linear(struct task_struct *child, struct pt_regs *regs) { @@ -30,10 +31,11 @@ unsigned long convert_ip_to_linear(struct task_struct *child, struct pt_regs *re seg &= ~7UL; mutex_lock(&child->mm->context.lock); - if (unlikely((seg >> 3) >= child->mm->context.size)) + if (unlikely(!child->mm->context.ldt || + (seg >> 3) >= child->mm->context.ldt->size)) addr = -1L; /* bogus selector, access would fault */ else { - desc = child->mm->context.ldt + seg; + desc = &child->mm->context.ldt->entries[seg]; base = get_desc_base(desc); /* 16-bit code segment? */ diff --git a/arch/x86/power/cpu.c b/arch/x86/power/cpu.c index 0d7dd1f..9ab5279 100644 --- a/arch/x86/power/cpu.c +++ b/arch/x86/power/cpu.c @@ -22,6 +22,7 @@ #include #include #include +#include #ifdef CONFIG_X86_32 __visible unsigned long saved_context_ebx; @@ -153,7 +154,7 @@ static void fix_processor_context(void) syscall_init(); /* This sets MSR_*STAR and related */ #endif load_TR_desc(); /* This does ltr */ - load_LDT(¤t->active_mm->context); /* This does lldt */ + load_mm_ldt(current->active_mm); /* This does lldt */ fpu__resume_cpu(); } -- cgit v0.10.2 From 1c1cc454aa694a89572689515fdaaf27b8c9f42a Mon Sep 17 00:00:00 2001 From: Joerg Roedel Date: Thu, 30 Jul 2015 11:24:45 +0200 Subject: iommu/amd: Allow non-ATS devices in IOMMUv2 domains With the grouping of multi-function devices a non-ATS capable device might also end up in the same domain as an IOMMUv2 capable device. So handle this situation gracefully and don't consider it a bug anymore. Tested-by: Oded Gabbay Signed-off-by: Joerg Roedel diff --git a/drivers/iommu/amd_iommu.c b/drivers/iommu/amd_iommu.c index fa9508b..658ee39 100644 --- a/drivers/iommu/amd_iommu.c +++ b/drivers/iommu/amd_iommu.c @@ -3318,7 +3318,12 @@ static int __flush_pasid(struct protection_domain *domain, int pasid, struct amd_iommu *iommu; int qdep; - BUG_ON(!dev_data->ats.enabled); + /* + There might be non-IOMMUv2 capable devices in an IOMMUv2 + * domain. + */ + if (!dev_data->ats.enabled) + continue; qdep = dev_data->ats.qdep; iommu = amd_iommu_rlookup_table[dev_data->devid]; -- cgit v0.10.2 From 20cadcb4df3eebd82410eab4aa91d5b083e16cd1 Mon Sep 17 00:00:00 2001 From: Ludovic Desroches Date: Wed, 17 Jun 2015 16:22:26 +0200 Subject: dmaengine: at_xdmac: fix bug about channel configuration When using descriptor view 2 or higher, we don't write the configuration into AT_XDMAC_CC register because this configuration will be fetch from the descriptor. Unfortunately, the PROT bit is not updated with this method, we have to do it manually before enabling the channel. Signed-off-by: Ludovic Desroches Signed-off-by: Vinod Koul diff --git a/drivers/dma/at_xdmac.c b/drivers/dma/at_xdmac.c index cf1213d..52ca1cc 100644 --- a/drivers/dma/at_xdmac.c +++ b/drivers/dma/at_xdmac.c @@ -359,18 +359,19 @@ static void at_xdmac_start_xfer(struct at_xdmac_chan *atchan, * descriptor view 2 since some fields of the configuration register * depend on transfer size and src/dest addresses. */ - if (at_xdmac_chan_is_cyclic(atchan)) { + if (at_xdmac_chan_is_cyclic(atchan)) reg = AT_XDMAC_CNDC_NDVIEW_NDV1; - at_xdmac_chan_write(atchan, AT_XDMAC_CC, first->lld.mbr_cfg); - } else if (first->lld.mbr_ubc & AT_XDMAC_MBR_UBC_NDV3) { + else if (first->lld.mbr_ubc & AT_XDMAC_MBR_UBC_NDV3) reg = AT_XDMAC_CNDC_NDVIEW_NDV3; - } else { - /* - * No need to write AT_XDMAC_CC reg, it will be done when the - * descriptor is fecthed. - */ + else reg = AT_XDMAC_CNDC_NDVIEW_NDV2; - } + /* + * Even if the register will be updated from the configuration in the + * descriptor when using view 2 or higher, the PROT bit won't be set + * properly. This bit can be modified only by using the channel + * configuration register. + */ + at_xdmac_chan_write(atchan, AT_XDMAC_CC, first->lld.mbr_cfg); reg |= AT_XDMAC_CNDC_NDDUP | AT_XDMAC_CNDC_NDSUP -- cgit v0.10.2 From 93dce3a6434f632dbd684ec1d9c3dc66a14e27e1 Mon Sep 17 00:00:00 2001 From: Cyrille Pitchen Date: Thu, 18 Jun 2015 13:25:41 +0200 Subject: dmaengine: at_hdmac: fix residue computation MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit As claimed by the programmer datasheet and confirmed by the IP designer, the Block Transfer Size (BTSIZE) bitfield of the Channel x Control A Register (CTRLAx) always refers to a number of Source Width (SRC_WIDTH) transfers. Both the SRC_WIDTH and BTSIZE bitfields can be extacted from the CTRLAx register to compute the DMA residue. So the 'tx_width' field is useless and can be removed from the struct at_desc. Before this patch, atc_prep_slave_sg() was not consistent: BTSIZE was correctly initialized according to the SRC_WIDTH but 'tx_width' was always set to reg_width, which was incorrect for MEM_TO_DEV transfers. It led to bad DMA residue when 'tx_width' != SRC_WIDTH. Also the 'tx_width' field was mostly set only in the first and last descriptors. Depending on the kind of DMA transfer, this field remained uninitialized for intermediate descriptors. The accurate DMA residue was computed only when the currently processed descriptor was the first or the last of the chain. This algorithm was a little bit odd. An accurate DMA residue can always be computed using the SRC_WIDTH and BTSIZE bitfields in the CTRLAx register. Finally, the test to check whether the currently processed descriptor is the last of the chain was wrong: for cyclic transfer, last_desc->lli.dscr is NOT equal to zero, since set_desc_eol() is never called, but logically equal to first_desc->txd.phys. This bug has a side effect on the drivers/tty/serial/atmel_serial.c driver, which uses cyclic DMA transfer to receive data. Since the DMA residue was wrong each time the DMA transfer reaches the second (and last) period of the transfer, no more data were received by the USART driver till the cyclic DMA transfer loops back to the first period. Signed-off-by: Cyrille Pitchen Acked-by: Torsten Fleischer Tested-by: Jirí Prchal Acked-by: Nicolas Ferre Signed-off-by: Vinod Koul diff --git a/drivers/dma/at_hdmac.c b/drivers/dma/at_hdmac.c index 59892126..d3629b7 100644 --- a/drivers/dma/at_hdmac.c +++ b/drivers/dma/at_hdmac.c @@ -48,6 +48,8 @@ BIT(DMA_SLAVE_BUSWIDTH_2_BYTES) |\ BIT(DMA_SLAVE_BUSWIDTH_4_BYTES)) +#define ATC_MAX_DSCR_TRIALS 10 + /* * Initial number of descriptors to allocate for each channel. This could * be increased during dma usage. @@ -285,28 +287,19 @@ static struct at_desc *atc_get_desc_by_cookie(struct at_dma_chan *atchan, * * @current_len: the number of bytes left before reading CTRLA * @ctrla: the value of CTRLA - * @desc: the descriptor containing the transfer width */ -static inline int atc_calc_bytes_left(int current_len, u32 ctrla, - struct at_desc *desc) +static inline int atc_calc_bytes_left(int current_len, u32 ctrla) { - return current_len - ((ctrla & ATC_BTSIZE_MAX) << desc->tx_width); -} + u32 btsize = (ctrla & ATC_BTSIZE_MAX); + u32 src_width = ATC_REG_TO_SRC_WIDTH(ctrla); -/** - * atc_calc_bytes_left_from_reg - calculates the number of bytes left according - * to the current value of CTRLA. - * - * @current_len: the number of bytes left before reading CTRLA - * @atchan: the channel to read CTRLA for - * @desc: the descriptor containing the transfer width - */ -static inline int atc_calc_bytes_left_from_reg(int current_len, - struct at_dma_chan *atchan, struct at_desc *desc) -{ - u32 ctrla = channel_readl(atchan, CTRLA); - - return atc_calc_bytes_left(current_len, ctrla, desc); + /* + * According to the datasheet, when reading the Control A Register + * (ctrla), the Buffer Transfer Size (btsize) bitfield refers to the + * number of transfers completed on the Source Interface. + * So btsize is always a number of source width transfers. + */ + return current_len - (btsize << src_width); } /** @@ -320,7 +313,7 @@ static int atc_get_bytes_left(struct dma_chan *chan, dma_cookie_t cookie) struct at_desc *desc_first = atc_first_active(atchan); struct at_desc *desc; int ret; - u32 ctrla, dscr; + u32 ctrla, dscr, trials; /* * If the cookie doesn't match to the currently running transfer then @@ -346,15 +339,82 @@ static int atc_get_bytes_left(struct dma_chan *chan, dma_cookie_t cookie) * the channel's DSCR register and compare it against the value * of the hardware linked list structure of each child * descriptor. + * + * The CTRLA register provides us with the amount of data + * already read from the source for the current child + * descriptor. So we can compute a more accurate residue by also + * removing the number of bytes corresponding to this amount of + * data. + * + * However, the DSCR and CTRLA registers cannot be read both + * atomically. Hence a race condition may occur: the first read + * register may refer to one child descriptor whereas the second + * read may refer to a later child descriptor in the list + * because of the DMA transfer progression inbetween the two + * reads. + * + * One solution could have been to pause the DMA transfer, read + * the DSCR and CTRLA then resume the DMA transfer. Nonetheless, + * this approach presents some drawbacks: + * - If the DMA transfer is paused, RX overruns or TX underruns + * are more likey to occur depending on the system latency. + * Taking the USART driver as an example, it uses a cyclic DMA + * transfer to read data from the Receive Holding Register + * (RHR) to avoid RX overruns since the RHR is not protected + * by any FIFO on most Atmel SoCs. So pausing the DMA transfer + * to compute the residue would break the USART driver design. + * - The atc_pause() function masks interrupts but we'd rather + * avoid to do so for system latency purpose. + * + * Then we'd rather use another solution: the DSCR is read a + * first time, the CTRLA is read in turn, next the DSCR is read + * a second time. If the two consecutive read values of the DSCR + * are the same then we assume both refers to the very same + * child descriptor as well as the CTRLA value read inbetween + * does. For cyclic tranfers, the assumption is that a full loop + * is "not so fast". + * If the two DSCR values are different, we read again the CTRLA + * then the DSCR till two consecutive read values from DSCR are + * equal or till the maxium trials is reach. + * This algorithm is very unlikely not to find a stable value for + * DSCR. */ - ctrla = channel_readl(atchan, CTRLA); - rmb(); /* ensure CTRLA is read before DSCR */ dscr = channel_readl(atchan, DSCR); + rmb(); /* ensure DSCR is read before CTRLA */ + ctrla = channel_readl(atchan, CTRLA); + for (trials = 0; trials < ATC_MAX_DSCR_TRIALS; ++trials) { + u32 new_dscr; + + rmb(); /* ensure DSCR is read after CTRLA */ + new_dscr = channel_readl(atchan, DSCR); + + /* + * If the DSCR register value has not changed inside the + * DMA controller since the previous read, we assume + * that both the dscr and ctrla values refers to the + * very same descriptor. + */ + if (likely(new_dscr == dscr)) + break; + + /* + * DSCR has changed inside the DMA controller, so the + * previouly read value of CTRLA may refer to an already + * processed descriptor hence could be outdated. + * We need to update ctrla to match the current + * descriptor. + */ + dscr = new_dscr; + rmb(); /* ensure DSCR is read before CTRLA */ + ctrla = channel_readl(atchan, CTRLA); + } + if (unlikely(trials >= ATC_MAX_DSCR_TRIALS)) + return -ETIMEDOUT; /* for the first descriptor we can be more accurate */ if (desc_first->lli.dscr == dscr) - return atc_calc_bytes_left(ret, ctrla, desc_first); + return atc_calc_bytes_left(ret, ctrla); ret -= desc_first->len; list_for_each_entry(desc, &desc_first->tx_list, desc_node) { @@ -365,16 +425,14 @@ static int atc_get_bytes_left(struct dma_chan *chan, dma_cookie_t cookie) } /* - * For the last descriptor in the chain we can calculate + * For the current descriptor in the chain we can calculate * the remaining bytes using the channel's register. - * Note that the transfer width of the first and last - * descriptor may differ. */ - if (!desc->lli.dscr) - ret = atc_calc_bytes_left_from_reg(ret, atchan, desc); + ret = atc_calc_bytes_left(ret, ctrla); } else { /* single transfer */ - ret = atc_calc_bytes_left_from_reg(ret, atchan, desc_first); + ctrla = channel_readl(atchan, CTRLA); + ret = atc_calc_bytes_left(ret, ctrla); } return ret; @@ -726,7 +784,6 @@ atc_prep_dma_interleaved(struct dma_chan *chan, desc->txd.cookie = -EBUSY; desc->total_len = desc->len = len; - desc->tx_width = dwidth; /* set end-of-link to the last link descriptor of list*/ set_desc_eol(desc); @@ -804,10 +861,6 @@ atc_prep_dma_memcpy(struct dma_chan *chan, dma_addr_t dest, dma_addr_t src, first->txd.cookie = -EBUSY; first->total_len = len; - /* set transfer width for the calculation of the residue */ - first->tx_width = src_width; - prev->tx_width = src_width; - /* set end-of-link to the last link descriptor of list*/ set_desc_eol(desc); @@ -956,10 +1009,6 @@ atc_prep_slave_sg(struct dma_chan *chan, struct scatterlist *sgl, first->txd.cookie = -EBUSY; first->total_len = total_len; - /* set transfer width for the calculation of the residue */ - first->tx_width = reg_width; - prev->tx_width = reg_width; - /* first link descriptor of list is responsible of flags */ first->txd.flags = flags; /* client is in control of this ack */ @@ -1077,12 +1126,6 @@ atc_prep_dma_sg(struct dma_chan *chan, desc->txd.cookie = 0; desc->len = len; - /* - * Although we only need the transfer width for the first and - * the last descriptor, its easier to set it to all descriptors. - */ - desc->tx_width = src_width; - atc_desc_chain(&first, &prev, desc); /* update the lengths and addresses for the next loop cycle */ @@ -1256,7 +1299,6 @@ atc_prep_dma_cyclic(struct dma_chan *chan, dma_addr_t buf_addr, size_t buf_len, /* First descriptor of the chain embedds additional information */ first->txd.cookie = -EBUSY; first->total_len = buf_len; - first->tx_width = reg_width; return &first->txd; diff --git a/drivers/dma/at_hdmac_regs.h b/drivers/dma/at_hdmac_regs.h index bc8d5eb..7f5a082 100644 --- a/drivers/dma/at_hdmac_regs.h +++ b/drivers/dma/at_hdmac_regs.h @@ -112,6 +112,7 @@ #define ATC_SRC_WIDTH_BYTE (0x0 << 24) #define ATC_SRC_WIDTH_HALFWORD (0x1 << 24) #define ATC_SRC_WIDTH_WORD (0x2 << 24) +#define ATC_REG_TO_SRC_WIDTH(r) (((r) >> 24) & 0x3) #define ATC_DST_WIDTH_MASK (0x3 << 28) /* Destination Single Transfer Size */ #define ATC_DST_WIDTH(x) ((x) << 28) #define ATC_DST_WIDTH_BYTE (0x0 << 28) @@ -182,7 +183,6 @@ struct at_lli { * @txd: support for the async_tx api * @desc_node: node on the channed descriptors list * @len: descriptor byte count - * @tx_width: transfer width * @total_len: total transaction byte count */ struct at_desc { @@ -194,7 +194,6 @@ struct at_desc { struct dma_async_tx_descriptor txd; struct list_head desc_node; size_t len; - u32 tx_width; size_t total_len; /* Interleaved data */ -- cgit v0.10.2 From 1c8a38b1268aebc1a903b21b11575077e02d2cf7 Mon Sep 17 00:00:00 2001 From: Cyrille Pitchen Date: Tue, 30 Jun 2015 14:36:57 +0200 Subject: dmaengine: at_xdmac: fix transfer data width in at_xdmac_prep_slave_sg() This patch adds the missing update of the transfer data width in at_xdmac_prep_slave_sg(). Indeed, for each item in the scatter-gather list, we check whether the transfer length is aligned with the data width provided by dmaengine_slave_config(). If so, we directly use this data width for the current part of the transfer we are preparing. Otherwise, the data width is reduced to 8 bits (1 byte). Of course, the actual number of register accesses must also be updated to match the new data width. So one chunk was missing in the original patch (see Fixes tag below): the number of register accesses was correctly set to (len >> fixed_dwidth) in mbr_ubc but the real data width was not updated in mbr_cfg. Since mbr_cfg may change for each part of the scatter-gather transfer this also explains why the original patch used the Descriptor View 2 instead of the Descriptor View 1. Let's take the example of a DMA transfer to write 8bit data into an Atmel USART with FIFOs. When FIFOs are enabled in the USART, its Transmit Holding Register (THR) works in multidata mode, that is to say that up to 4 8bit data can be written into the THR in a single 32bit access and it is still possible to write only one data with a 8bit access. To take advantage of this new feature, the DMA driver was modified to allow multiple dwidths when doing slave transfers. For instance, when the total length is 22 bytes, the USART driver splits the transfer into 2 parts: First part: 20 bytes transferred through 5 32bit writes into THR Second part: 2 bytes transferred though 2 8bit writes into THR For the second part, the data width was first set to 4_BYTES by the USART driver thanks to dmaengine_slave_config() then at_xdmac_prep_slave_sg() reduces this data width to 1_BYTE because the 2 byte length is not aligned with the original 4_BYTES data width. Since the data width is modified, the actual number of writes into THR must be set accordingly. Signed-off-by: Cyrille Pitchen Fixes: 6d3a7d9e3ada ("dmaengine: at_xdmac: allow muliple dwidths when doing slave transfers") Cc: stable@vger.kernel.org #4.0 and later Acked-by: Nicolas Ferre Acked-by: Ludovic Desroches Signed-off-by: Vinod Koul diff --git a/drivers/dma/at_xdmac.c b/drivers/dma/at_xdmac.c index 52ca1cc..40afa2a 100644 --- a/drivers/dma/at_xdmac.c +++ b/drivers/dma/at_xdmac.c @@ -682,15 +682,16 @@ at_xdmac_prep_slave_sg(struct dma_chan *chan, struct scatterlist *sgl, desc->lld.mbr_sa = mem; desc->lld.mbr_da = atchan->sconfig.dst_addr; } - desc->lld.mbr_cfg = atchan->cfg; - dwidth = at_xdmac_get_dwidth(desc->lld.mbr_cfg); + dwidth = at_xdmac_get_dwidth(atchan->cfg); fixed_dwidth = IS_ALIGNED(len, 1 << dwidth) - ? at_xdmac_get_dwidth(desc->lld.mbr_cfg) + ? dwidth : AT_XDMAC_CC_DWIDTH_BYTE; desc->lld.mbr_ubc = AT_XDMAC_MBR_UBC_NDV2 /* next descriptor view */ | AT_XDMAC_MBR_UBC_NDEN /* next descriptor dst parameter update */ | AT_XDMAC_MBR_UBC_NSEN /* next descriptor src parameter update */ | (len >> fixed_dwidth); /* microblock length */ + desc->lld.mbr_cfg = (atchan->cfg & ~AT_XDMAC_CC_DWIDTH_MASK) | + AT_XDMAC_CC_DWIDTH(fixed_dwidth); dev_dbg(chan2dev(chan), "%s: lld: mbr_sa=%pad, mbr_da=%pad, mbr_ubc=0x%08x\n", __func__, &desc->lld.mbr_sa, &desc->lld.mbr_da, desc->lld.mbr_ubc); -- cgit v0.10.2 From cda8e937191c100025168ba3e22ab316c7298007 Mon Sep 17 00:00:00 2001 From: Rameshwar Prasad Sahu Date: Tue, 7 Jul 2015 15:34:25 +0530 Subject: dmaengine: xgene-dma: Fix the resource map to handle overlapping There is an overlap in dma ring cmd csr region due to sharing of ethernet ring cmd csr region. This patch fix the resource overlapping by mapping the entire dma ring cmd csr region. Signed-off-by: Rameshwar Prasad Sahu Signed-off-by: Vinod Koul diff --git a/Documentation/devicetree/bindings/dma/apm-xgene-dma.txt b/Documentation/devicetree/bindings/dma/apm-xgene-dma.txt index d305876..c53e0b0 100644 --- a/Documentation/devicetree/bindings/dma/apm-xgene-dma.txt +++ b/Documentation/devicetree/bindings/dma/apm-xgene-dma.txt @@ -35,7 +35,7 @@ Example: device_type = "dma"; reg = <0x0 0x1f270000 0x0 0x10000>, <0x0 0x1f200000 0x0 0x10000>, - <0x0 0x1b008000 0x0 0x2000>, + <0x0 0x1b000000 0x0 0x400000>, <0x0 0x1054a000 0x0 0x100>; interrupts = <0x0 0x82 0x4>, <0x0 0xb8 0x4>, diff --git a/arch/arm64/boot/dts/apm/apm-storm.dtsi b/arch/arm64/boot/dts/apm/apm-storm.dtsi index 0689c3f..58093ed 100644 --- a/arch/arm64/boot/dts/apm/apm-storm.dtsi +++ b/arch/arm64/boot/dts/apm/apm-storm.dtsi @@ -823,7 +823,7 @@ device_type = "dma"; reg = <0x0 0x1f270000 0x0 0x10000>, <0x0 0x1f200000 0x0 0x10000>, - <0x0 0x1b008000 0x0 0x2000>, + <0x0 0x1b000000 0x0 0x400000>, <0x0 0x1054a000 0x0 0x100>; interrupts = <0x0 0x82 0x4>, <0x0 0xb8 0x4>, diff --git a/drivers/dma/xgene-dma.c b/drivers/dma/xgene-dma.c index 620fd55ec..dff22ab 100644 --- a/drivers/dma/xgene-dma.c +++ b/drivers/dma/xgene-dma.c @@ -111,6 +111,7 @@ #define XGENE_DMA_MEM_RAM_SHUTDOWN 0xD070 #define XGENE_DMA_BLK_MEM_RDY 0xD074 #define XGENE_DMA_BLK_MEM_RDY_VAL 0xFFFFFFFF +#define XGENE_DMA_RING_CMD_SM_OFFSET 0x8000 /* X-Gene SoC EFUSE csr register and bit defination */ #define XGENE_SOC_JTAG1_SHADOW 0x18 @@ -1887,6 +1888,8 @@ static int xgene_dma_get_resources(struct platform_device *pdev, return -ENOMEM; } + pdma->csr_ring_cmd += XGENE_DMA_RING_CMD_SM_OFFSET; + /* Get efuse csr region */ res = platform_get_resource(pdev, IORESOURCE_MEM, 3); if (!res) { -- cgit v0.10.2 From 0ec9ebc706fbd394bc233d87ac7aaad1c4f3ab54 Mon Sep 17 00:00:00 2001 From: Thomas Petazzoni Date: Wed, 8 Jul 2015 16:28:14 +0200 Subject: dmaengine: mv_xor: fix big endian operation in register mode Commit 6f166312c6ea2 ("dmaengine: mv_xor: add support for a38x command in descriptor mode") introduced the support for a feature that appeared in Armada 38x: specifying the operation to be performed in a per-descriptor basis rather than globally per channel. However, when doing so, it changed the function mv_chan_set_mode() to use: if (IS_ENABLED(__BIG_ENDIAN)) instead of: #if defined(__BIG_ENDIAN) While IS_ENABLED() is perfectly fine for CONFIG_* symbols, it is not for other symbols such as __BIG_ENDIAN that is provided directly by the compiler. Consequently, the commit broke support for big-endian, as the XOR_DESCRIPTOR_SWAP flag was not set in the XOR channel configuration register. The primarily visible effect was some nasty warnings and failures appearing during the self-test of the XOR unit: [ 1.197368] mv_xor d0060900.xor: error on chan 0. intr cause 0x00000082 [ 1.197393] mv_xor d0060900.xor: config 0x00008440 [ 1.197410] mv_xor d0060900.xor: activation 0x00000000 [ 1.197427] mv_xor d0060900.xor: intr cause 0x00000082 [ 1.197443] mv_xor d0060900.xor: intr mask 0x000003f7 [ 1.197460] mv_xor d0060900.xor: error cause 0x00000000 [ 1.197477] mv_xor d0060900.xor: error addr 0x00000000 [ 1.197491] ------------[ cut here ]------------ [ 1.197513] WARNING: CPU: 0 PID: 1 at ../drivers/dma/mv_xor.c:664 mv_xor_interrupt_handler+0x14c/0x170() See also: http://storage.kernelci.org/next/next-20150617/arm-mvebu_v7_defconfig+CONFIG_CPU_BIG_ENDIAN=y/lab-khilman/boot-armada-xp-openblocks-ax3-4.txt Signed-off-by: Thomas Petazzoni Fixes: 6f166312c6ea2 ("dmaengine: mv_xor: add support for a38x command in descriptor mode") Reviewed-by: Maxime Ripard Signed-off-by: Vinod Koul diff --git a/drivers/dma/mv_xor.c b/drivers/dma/mv_xor.c index fbaf1ea..f1325f6 100644 --- a/drivers/dma/mv_xor.c +++ b/drivers/dma/mv_xor.c @@ -162,10 +162,11 @@ static void mv_chan_set_mode(struct mv_xor_chan *chan, config &= ~0x7; config |= op_mode; - if (IS_ENABLED(__BIG_ENDIAN)) - config |= XOR_DESCRIPTOR_SWAP; - else - config &= ~XOR_DESCRIPTOR_SWAP; +#if defined(__BIG_ENDIAN) + config |= XOR_DESCRIPTOR_SWAP; +#else + config &= ~XOR_DESCRIPTOR_SWAP; +#endif writel_relaxed(config, XOR_CONFIG(chan)); chan->current_type = type; -- cgit v0.10.2 From 8c8fe97b2b8a216523e2faf1ccca66ddab634e3e Mon Sep 17 00:00:00 2001 From: Jun Nie Date: Fri, 10 Jul 2015 20:02:49 +0800 Subject: Revert "dmaengine: virt-dma: don't always free descriptor upon completion" This reverts commit b9855f03d560d351e95301b9de0bc3cad3b31fe9. The patch break existing DMA usage case. For example, audio SOC dmaengine never release channel and cause virt-dma to cache too much memory in descriptor to exhaust system memory. Signed-off-by: Vinod Koul diff --git a/drivers/dma/virt-dma.c b/drivers/dma/virt-dma.c index 7d2c17d..6f80432 100644 --- a/drivers/dma/virt-dma.c +++ b/drivers/dma/virt-dma.c @@ -29,7 +29,7 @@ dma_cookie_t vchan_tx_submit(struct dma_async_tx_descriptor *tx) spin_lock_irqsave(&vc->lock, flags); cookie = dma_cookie_assign(tx); - list_move_tail(&vd->node, &vc->desc_submitted); + list_add_tail(&vd->node, &vc->desc_submitted); spin_unlock_irqrestore(&vc->lock, flags); dev_dbg(vc->chan.device->dev, "vchan %p: txd %p[%x]: submitted\n", @@ -83,10 +83,8 @@ static void vchan_complete(unsigned long arg) cb_data = vd->tx.callback_param; list_del(&vd->node); - if (async_tx_test_ack(&vd->tx)) - list_add(&vd->node, &vc->desc_allocated); - else - vc->desc_free(vd); + + vc->desc_free(vd); if (cb) cb(cb_data); @@ -98,13 +96,9 @@ void vchan_dma_desc_free_list(struct virt_dma_chan *vc, struct list_head *head) while (!list_empty(head)) { struct virt_dma_desc *vd = list_first_entry(head, struct virt_dma_desc, node); - if (async_tx_test_ack(&vd->tx)) { - list_move_tail(&vd->node, &vc->desc_allocated); - } else { - dev_dbg(vc->chan.device->dev, "txd %p: freeing\n", vd); - list_del(&vd->node); - vc->desc_free(vd); - } + list_del(&vd->node); + dev_dbg(vc->chan.device->dev, "txd %p: freeing\n", vd); + vc->desc_free(vd); } } EXPORT_SYMBOL_GPL(vchan_dma_desc_free_list); @@ -114,7 +108,6 @@ void vchan_init(struct virt_dma_chan *vc, struct dma_device *dmadev) dma_cookie_init(&vc->chan); spin_lock_init(&vc->lock); - INIT_LIST_HEAD(&vc->desc_allocated); INIT_LIST_HEAD(&vc->desc_submitted); INIT_LIST_HEAD(&vc->desc_issued); INIT_LIST_HEAD(&vc->desc_completed); diff --git a/drivers/dma/virt-dma.h b/drivers/dma/virt-dma.h index 189e75d..181b952 100644 --- a/drivers/dma/virt-dma.h +++ b/drivers/dma/virt-dma.h @@ -29,7 +29,6 @@ struct virt_dma_chan { spinlock_t lock; /* protected by vc.lock */ - struct list_head desc_allocated; struct list_head desc_submitted; struct list_head desc_issued; struct list_head desc_completed; @@ -56,16 +55,11 @@ static inline struct dma_async_tx_descriptor *vchan_tx_prep(struct virt_dma_chan struct virt_dma_desc *vd, unsigned long tx_flags) { extern dma_cookie_t vchan_tx_submit(struct dma_async_tx_descriptor *); - unsigned long flags; dma_async_tx_descriptor_init(&vd->tx, &vc->chan); vd->tx.flags = tx_flags; vd->tx.tx_submit = vchan_tx_submit; - spin_lock_irqsave(&vc->lock, flags); - list_add_tail(&vd->node, &vc->desc_allocated); - spin_unlock_irqrestore(&vc->lock, flags); - return &vd->tx; } @@ -128,8 +122,7 @@ static inline struct virt_dma_desc *vchan_next_desc(struct virt_dma_chan *vc) } /** - * vchan_get_all_descriptors - obtain all allocated, submitted and issued - * descriptors + * vchan_get_all_descriptors - obtain all submitted and issued descriptors * vc: virtual channel to get descriptors from * head: list of descriptors found * @@ -141,7 +134,6 @@ static inline struct virt_dma_desc *vchan_next_desc(struct virt_dma_chan *vc) static inline void vchan_get_all_descriptors(struct virt_dma_chan *vc, struct list_head *head) { - list_splice_tail_init(&vc->desc_allocated, head); list_splice_tail_init(&vc->desc_submitted, head); list_splice_tail_init(&vc->desc_issued, head); list_splice_tail_init(&vc->desc_completed, head); @@ -149,14 +141,11 @@ static inline void vchan_get_all_descriptors(struct virt_dma_chan *vc, static inline void vchan_free_chan_resources(struct virt_dma_chan *vc) { - struct virt_dma_desc *vd; unsigned long flags; LIST_HEAD(head); spin_lock_irqsave(&vc->lock, flags); vchan_get_all_descriptors(vc, &head); - list_for_each_entry(vd, &head, node) - async_tx_clear_ack(&vd->tx); spin_unlock_irqrestore(&vc->lock, flags); vchan_dma_desc_free_list(vc, &head); -- cgit v0.10.2 From c1bfa985ded82cacdfc6403e78f329c44e35534a Mon Sep 17 00:00:00 2001 From: Murali Karicheri Date: Fri, 29 May 2015 12:04:13 -0400 Subject: ARM: dts: keystone: fix dt bindings to use post div register for mainpll All of the keystone devices have a separate register to hold post divider value for main pll clock. Currently the fixed-postdiv value used for k2hk/l/e SoCs works by sheer luck as u-boot happens to use a value of 2 for this. Now that we have fixed this in the pll clock driver change the dt bindings for the same. Signed-off-by: Murali Karicheri Acked-by: Santosh Shilimkar Signed-off-by: Olof Johansson diff --git a/arch/arm/boot/dts/k2e-clocks.dtsi b/arch/arm/boot/dts/k2e-clocks.dtsi index 4773d6a..d56d68f 100644 --- a/arch/arm/boot/dts/k2e-clocks.dtsi +++ b/arch/arm/boot/dts/k2e-clocks.dtsi @@ -13,9 +13,8 @@ clocks { #clock-cells = <0>; compatible = "ti,keystone,main-pll-clock"; clocks = <&refclksys>; - reg = <0x02620350 4>, <0x02310110 4>; - reg-names = "control", "multiplier"; - fixed-postdiv = <2>; + reg = <0x02620350 4>, <0x02310110 4>, <0x02310108 4>; + reg-names = "control", "multiplier", "post-divider"; }; papllclk: papllclk@2620358 { diff --git a/arch/arm/boot/dts/k2hk-clocks.dtsi b/arch/arm/boot/dts/k2hk-clocks.dtsi index d5adee3..af9b719 100644 --- a/arch/arm/boot/dts/k2hk-clocks.dtsi +++ b/arch/arm/boot/dts/k2hk-clocks.dtsi @@ -22,9 +22,8 @@ clocks { #clock-cells = <0>; compatible = "ti,keystone,main-pll-clock"; clocks = <&refclksys>; - reg = <0x02620350 4>, <0x02310110 4>; - reg-names = "control", "multiplier"; - fixed-postdiv = <2>; + reg = <0x02620350 4>, <0x02310110 4>, <0x02310108 4>; + reg-names = "control", "multiplier", "post-divider"; }; papllclk: papllclk@2620358 { diff --git a/arch/arm/boot/dts/k2l-clocks.dtsi b/arch/arm/boot/dts/k2l-clocks.dtsi index eb1e3e2..ef8464b 100644 --- a/arch/arm/boot/dts/k2l-clocks.dtsi +++ b/arch/arm/boot/dts/k2l-clocks.dtsi @@ -22,9 +22,8 @@ clocks { #clock-cells = <0>; compatible = "ti,keystone,main-pll-clock"; clocks = <&refclksys>; - reg = <0x02620350 4>, <0x02310110 4>; - reg-names = "control", "multiplier"; - fixed-postdiv = <2>; + reg = <0x02620350 4>, <0x02310110 4>, <0x02310108 4>; + reg-names = "control", "multiplier", "post-divider"; }; papllclk: papllclk@2620358 { -- cgit v0.10.2 From 5175f7106cc55a1bcf97bf7d5ba0900017ebcef8 Mon Sep 17 00:00:00 2001 From: WANG Cong Date: Thu, 30 Jul 2015 17:12:21 -0700 Subject: act_pedit: check binding before calling tcf_hash_release() When we share an action within a filter, the bind refcnt should increase, therefore we should not call tcf_hash_release(). Fixes: 1a29321ed045 ("net_sched: act: Dont increment refcnt on replace") Cc: Jamal Hadi Salim Cc: Daniel Borkmann Signed-off-by: Cong Wang Signed-off-by: Cong Wang Acked-by: Daniel Borkmann Signed-off-by: David S. Miller diff --git a/net/sched/act_pedit.c b/net/sched/act_pedit.c index 17e6d66..ff8b466 100644 --- a/net/sched/act_pedit.c +++ b/net/sched/act_pedit.c @@ -68,13 +68,12 @@ static int tcf_pedit_init(struct net *net, struct nlattr *nla, } ret = ACT_P_CREATED; } else { - p = to_pedit(a); - tcf_hash_release(a, bind); if (bind) return 0; + tcf_hash_release(a, bind); if (!ovr) return -EEXIST; - + p = to_pedit(a); if (p->tcfp_nkeys && p->tcfp_nkeys != parm->nkeys) { keys = kmalloc(ksize, GFP_KERNEL); if (keys == NULL) -- cgit v0.10.2 From 8486830549189b109bfd40e8d38c6dd8b8f476db Mon Sep 17 00:00:00 2001 From: Claudiu Manoil Date: Fri, 31 Jul 2015 18:38:31 +0300 Subject: gianfar: Fix warning when CONFIG_PM off CC drivers/net/ethernet/freescale/gianfar.o drivers/net/ethernet/freescale/gianfar.c:568:13: warning: 'lock_tx_qs' defined but not used [-Wunused-function] static void lock_tx_qs(struct gfar_private *priv) ^ drivers/net/ethernet/freescale/gianfar.c:576:13: warning: 'unlock_tx_qs' defined but not used [-Wunused-function] static void unlock_tx_qs(struct gfar_private *priv) ^ Reported-by: Scott Wood Signed-off-by: Claudiu Manoil Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/freescale/gianfar.c b/drivers/net/ethernet/freescale/gianfar.c index ff87502..caad327 100644 --- a/drivers/net/ethernet/freescale/gianfar.c +++ b/drivers/net/ethernet/freescale/gianfar.c @@ -565,6 +565,7 @@ static void gfar_ints_enable(struct gfar_private *priv) } } +#ifdef CONFIG_PM static void lock_tx_qs(struct gfar_private *priv) { int i; @@ -580,6 +581,7 @@ static void unlock_tx_qs(struct gfar_private *priv) for (i = 0; i < priv->num_tx_queues; i++) spin_unlock(&priv->tx_queue[i]->txlock); } +#endif static int gfar_alloc_tx_queues(struct gfar_private *priv) { -- cgit v0.10.2 From 614b42426cc3483e8d5bc68a158c2dd47dc831d0 Mon Sep 17 00:00:00 2001 From: Claudiu Manoil Date: Fri, 31 Jul 2015 18:38:32 +0300 Subject: gianfar: Fix suspend/resume for wol magic packet If we disable NAPI in the first place we can mask the device's interrupts (and halt it) without fearing that imask may be concurrently accessed from interrupt context, so there's no need to do local_irq_save() around gfar_halt_nodisable(). lock_rx_qs()/unlock_tx_qs() are just obsoleted and potentially buggy routines. The txlock is currently used in the driver only to manage TX congestion, it has nothing to do with halting the device. With these changes, the TX processing is stopped before gfar_halt(). Compact gfar_halt() is used instead of gfar_halt_nodisable(), as it disables Rx/TX DMA h/w blocks and the Rx/TX h/w queues. gfar_start() re-enables all these blocks on resume. Enabling the magic-packet mode remains the same, note that the RX block is re-enabled just before entering sleep mode. Add IRQF_NO_SUSPEND flag for the error interrupt line, to signal that the interrupt line must remain active during sleep in order to wake the system by magic packet (MAG) reception interrupt. (On some systems the MAG interrupt did trigger w/o this flag as well, but on others it didn't.) Without these fixes, when suspended during fair Tx traffic the interface occasionally failed to be woken up by magic packet. Signed-off-by: Claudiu Manoil Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/freescale/gianfar.c b/drivers/net/ethernet/freescale/gianfar.c index caad327..084c25d 100644 --- a/drivers/net/ethernet/freescale/gianfar.c +++ b/drivers/net/ethernet/freescale/gianfar.c @@ -565,24 +565,6 @@ static void gfar_ints_enable(struct gfar_private *priv) } } -#ifdef CONFIG_PM -static void lock_tx_qs(struct gfar_private *priv) -{ - int i; - - for (i = 0; i < priv->num_tx_queues; i++) - spin_lock(&priv->tx_queue[i]->txlock); -} - -static void unlock_tx_qs(struct gfar_private *priv) -{ - int i; - - for (i = 0; i < priv->num_tx_queues; i++) - spin_unlock(&priv->tx_queue[i]->txlock); -} -#endif - static int gfar_alloc_tx_queues(struct gfar_private *priv) { int i; @@ -1542,48 +1524,37 @@ static int gfar_suspend(struct device *dev) struct gfar_private *priv = dev_get_drvdata(dev); struct net_device *ndev = priv->ndev; struct gfar __iomem *regs = priv->gfargrp[0].regs; - unsigned long flags; u32 tempval; - int magic_packet = priv->wol_en && (priv->device_flags & FSL_GIANFAR_DEV_HAS_MAGIC_PACKET); + if (!netif_running(ndev)) + return 0; + + disable_napi(priv); + netif_tx_lock(ndev); netif_device_detach(ndev); + netif_tx_unlock(ndev); - if (netif_running(ndev)) { + gfar_halt(priv); - local_irq_save(flags); - lock_tx_qs(priv); + if (magic_packet) { + /* Enable interrupt on Magic Packet */ + gfar_write(®s->imask, IMASK_MAG); - gfar_halt_nodisable(priv); + /* Enable Magic Packet mode */ + tempval = gfar_read(®s->maccfg2); + tempval |= MACCFG2_MPEN; + gfar_write(®s->maccfg2, tempval); - /* Disable Tx, and Rx if wake-on-LAN is disabled. */ + /* re-enable the Rx block */ tempval = gfar_read(®s->maccfg1); - - tempval &= ~MACCFG1_TX_EN; - - if (!magic_packet) - tempval &= ~MACCFG1_RX_EN; - + tempval |= MACCFG1_RX_EN; gfar_write(®s->maccfg1, tempval); - unlock_tx_qs(priv); - local_irq_restore(flags); - - disable_napi(priv); - - if (magic_packet) { - /* Enable interrupt on Magic Packet */ - gfar_write(®s->imask, IMASK_MAG); - - /* Enable Magic Packet mode */ - tempval = gfar_read(®s->maccfg2); - tempval |= MACCFG2_MPEN; - gfar_write(®s->maccfg2, tempval); - } else { - phy_stop(priv->phydev); - } + } else { + phy_stop(priv->phydev); } return 0; @@ -1594,37 +1565,26 @@ static int gfar_resume(struct device *dev) struct gfar_private *priv = dev_get_drvdata(dev); struct net_device *ndev = priv->ndev; struct gfar __iomem *regs = priv->gfargrp[0].regs; - unsigned long flags; u32 tempval; int magic_packet = priv->wol_en && (priv->device_flags & FSL_GIANFAR_DEV_HAS_MAGIC_PACKET); - if (!netif_running(ndev)) { - netif_device_attach(ndev); + if (!netif_running(ndev)) return 0; - } - if (!magic_packet && priv->phydev) + if (magic_packet) { + /* Disable Magic Packet mode */ + tempval = gfar_read(®s->maccfg2); + tempval &= ~MACCFG2_MPEN; + gfar_write(®s->maccfg2, tempval); + } else { phy_start(priv->phydev); - - /* Disable Magic Packet mode, in case something - * else woke us up. - */ - local_irq_save(flags); - lock_tx_qs(priv); - - tempval = gfar_read(®s->maccfg2); - tempval &= ~MACCFG2_MPEN; - gfar_write(®s->maccfg2, tempval); + } gfar_start(priv); - unlock_tx_qs(priv); - local_irq_restore(flags); - netif_device_attach(ndev); - enable_napi(priv); return 0; @@ -2047,7 +2007,8 @@ static int register_grp_irqs(struct gfar_priv_grp *grp) /* Install our interrupt handlers for Error, * Transmit, and Receive */ - err = request_irq(gfar_irq(grp, ER)->irq, gfar_error, 0, + err = request_irq(gfar_irq(grp, ER)->irq, gfar_error, + IRQF_NO_SUSPEND, gfar_irq(grp, ER)->name, grp); if (err < 0) { netif_err(priv, intr, dev, "Can't get IRQ %d\n", @@ -2070,7 +2031,8 @@ static int register_grp_irqs(struct gfar_priv_grp *grp) goto rx_irq_fail; } } else { - err = request_irq(gfar_irq(grp, TX)->irq, gfar_interrupt, 0, + err = request_irq(gfar_irq(grp, TX)->irq, gfar_interrupt, + IRQF_NO_SUSPEND, gfar_irq(grp, TX)->name, grp); if (err < 0) { netif_err(priv, intr, dev, "Can't get IRQ %d\n", -- cgit v0.10.2 From b0734b6dc895258b74c6e7a441cb47b6b0ba3465 Mon Sep 17 00:00:00 2001 From: Claudiu Manoil Date: Fri, 31 Jul 2015 18:38:33 +0300 Subject: gianfar: Enable device wakeup when appropriate The wol_en flag is 0 by default anyway, and we have the following inconsistency: a MAGIC packet wol capable eth interface is registered as a wake-up source but unable to wake-up the system as wol_en is 0 (wake-on flag set to 'd'). Calling set_wakeup_enable() at netdev open is just redundant because wol_en is 0 by default. Let only ethtool call set_wakeup_enable() for now. The bflock is obviously obsoleted, its utility has been corroded over time. The bitfield flags used today in gianfar are accessed only on the init/ config path, with no real possibility of concurrency - nothing that would justify smth. like bflock. Signed-off-by: Claudiu Manoil Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/freescale/gianfar.c b/drivers/net/ethernet/freescale/gianfar.c index 084c25d..2b7610f 100644 --- a/drivers/net/ethernet/freescale/gianfar.c +++ b/drivers/net/ethernet/freescale/gianfar.c @@ -1360,7 +1360,6 @@ static int gfar_probe(struct platform_device *ofdev) priv->dev = &ofdev->dev; SET_NETDEV_DEV(dev, &ofdev->dev); - spin_lock_init(&priv->bflock); INIT_WORK(&priv->reset_task, gfar_reset_task); platform_set_drvdata(ofdev, priv); @@ -1454,9 +1453,8 @@ static int gfar_probe(struct platform_device *ofdev) goto register_fail; } - device_init_wakeup(&dev->dev, - priv->device_flags & - FSL_GIANFAR_DEV_HAS_MAGIC_PACKET); + device_set_wakeup_capable(&dev->dev, priv->device_flags & + FSL_GIANFAR_DEV_HAS_MAGIC_PACKET); /* fill out IRQ number and name fields */ for (i = 0; i < priv->num_grps; i++) { @@ -2133,8 +2131,6 @@ static int gfar_enet_open(struct net_device *dev) if (err) return err; - device_set_wakeup_enable(&dev->dev, priv->wol_en); - return err; } diff --git a/drivers/net/ethernet/freescale/gianfar.h b/drivers/net/ethernet/freescale/gianfar.h index daa1d37..5545e41 100644 --- a/drivers/net/ethernet/freescale/gianfar.h +++ b/drivers/net/ethernet/freescale/gianfar.h @@ -1145,9 +1145,6 @@ struct gfar_private { int oldduplex; int oldlink; - /* Bitfield update lock */ - spinlock_t bflock; - uint32_t msg_enable; struct work_struct reset_task; diff --git a/drivers/net/ethernet/freescale/gianfar_ethtool.c b/drivers/net/ethernet/freescale/gianfar_ethtool.c index fda12fb..3c0a8f8 100644 --- a/drivers/net/ethernet/freescale/gianfar_ethtool.c +++ b/drivers/net/ethernet/freescale/gianfar_ethtool.c @@ -653,7 +653,6 @@ static void gfar_get_wol(struct net_device *dev, struct ethtool_wolinfo *wol) static int gfar_set_wol(struct net_device *dev, struct ethtool_wolinfo *wol) { struct gfar_private *priv = netdev_priv(dev); - unsigned long flags; if (!(priv->device_flags & FSL_GIANFAR_DEV_HAS_MAGIC_PACKET) && wol->wolopts != 0) @@ -664,9 +663,7 @@ static int gfar_set_wol(struct net_device *dev, struct ethtool_wolinfo *wol) device_set_wakeup_enable(&dev->dev, wol->wolopts & WAKE_MAGIC); - spin_lock_irqsave(&priv->bflock, flags); - priv->wol_en = !!device_may_wakeup(&dev->dev); - spin_unlock_irqrestore(&priv->bflock, flags); + priv->wol_en = !!device_may_wakeup(&dev->dev); return 0; } -- cgit v0.10.2 From ea111545843e657504193e4f726e1116b96b8141 Mon Sep 17 00:00:00 2001 From: Joachim Eastwood Date: Fri, 31 Jul 2015 19:13:22 +0200 Subject: stmmac: fix missing MODULE_LICENSE in stmmac_platform Commit 50649ab14982 ("stmmac: drop driver from stmmac platform code") was a bit overzealous in removing code and dropped the MODULE_* macro's that are still needed since stmmac_platform can be a module. Fix this by putting the macro's remvoed in 50649ab14982 back. This fixes the following errors when used as a module: stmmac_platform: module license 'unspecified' taints kernel. Disabling lock debugging due to kernel taint stmmac_platform: Unknown symbol devm_kmalloc (err 0) stmmac_platform: Unknown symbol stmmac_suspend (err 0) stmmac_platform: Unknown symbol platform_get_irq_byname (err 0) stmmac_platform: Unknown symbol stmmac_dvr_remove (err 0) stmmac_platform: Unknown symbol platform_get_resource (err 0) stmmac_platform: Unknown symbol of_get_phy_mode (err 0) stmmac_platform: Unknown symbol of_property_read_u32_array (err 0) stmmac_platform: Unknown symbol of_alias_get_id (err 0) stmmac_platform: Unknown symbol stmmac_resume (err 0) stmmac_platform: Unknown symbol stmmac_dvr_probe (err 0) Fixes: 50649ab14982 ("stmmac: drop driver from stmmac platform code") Reported-by: Igor Gnatenko Signed-off-by: Joachim Eastwood Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c index f3918c7..bcdc895 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c @@ -413,3 +413,7 @@ static int stmmac_pltfr_resume(struct device *dev) SIMPLE_DEV_PM_OPS(stmmac_pltfr_pm_ops, stmmac_pltfr_suspend, stmmac_pltfr_resume); EXPORT_SYMBOL_GPL(stmmac_pltfr_pm_ops); + +MODULE_DESCRIPTION("STMMAC 10/100/1000 Ethernet platform support"); +MODULE_AUTHOR("Giuseppe Cavallaro "); +MODULE_LICENSE("GPL"); -- cgit v0.10.2 From 97242f99a013950af63effa0732f8ef7db4e31ec Mon Sep 17 00:00:00 2001 From: Al Viro Date: Sat, 1 Aug 2015 19:59:28 -0400 Subject: link_path_walk(): be careful when failing with ENOTDIR In RCU mode we might end up with dentry evicted just we check that it's a directory. In such case we should return ECHILD rather than ENOTDIR, so that pathwalk would be retries in non-RCU mode. Breakage had been introduced in commit b18825a - prior to that we were looking at nd->inode, which had been fetched before verifying that ->d_seq was still valid. That form of check would only be satisfied if at some point the pathname prefix would indeed have resolved to a non-directory. The fix consists of checking ->d_seq after we'd run into a non-directory dentry, and failing with ECHILD in case of mismatch. Note that all branches since 3.12 have that problem... Signed-off-by: Al Viro diff --git a/fs/namei.c b/fs/namei.c index ae4e4c1..fbbcf09 100644 --- a/fs/namei.c +++ b/fs/namei.c @@ -1954,8 +1954,13 @@ OK: continue; } } - if (unlikely(!d_can_lookup(nd->path.dentry))) + if (unlikely(!d_can_lookup(nd->path.dentry))) { + if (nd->flags & LOOKUP_RCU) { + if (unlazy_walk(nd, NULL, 0)) + return -ECHILD; + } return -ENOTDIR; + } } } -- cgit v0.10.2 From 27667f4744fc5a0f3e50910e78740bac5670d18b Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Wed, 29 Jul 2015 22:18:16 -0700 Subject: i915: temporary fix for DP MST docking station NULL pointer dereference Ted Ts'o reports that his Lenovo T540p ThinkPad crashes at boot if attached to the docking station. This is a regression that he was able to bisect to commit 8c7b5ccb7298: "drm/i915: Use atomic helpers for computing changed flags:" The reason seems to be the new call to drm_atomic_helper_check_modeset() added to intel_modeset_compute_config(), which in turn calls update_connector_routing(), and somehow ends up picking a NULL crtc for the connector state, causing the subsequent drm_crtc_index() to OOPS. Daniel Vetter says that the fundamental issue seems to be confusion in the encoder selection, and this isn't the right fix, but while he chases down the proper fix, this at least avoids the NULL pointer dereference and makes Ted's docking station work again. Reported-bisected-and-tested-by: Theodore Ts'o Cc: Daniel Vetter Cc: Mani Nikula Cc: Dave Airlie Signed-off-by: Linus Torvalds diff --git a/drivers/gpu/drm/drm_atomic_helper.c b/drivers/gpu/drm/drm_atomic_helper.c index 5b59d5ad..aac2122 100644 --- a/drivers/gpu/drm/drm_atomic_helper.c +++ b/drivers/gpu/drm/drm_atomic_helper.c @@ -230,10 +230,12 @@ update_connector_routing(struct drm_atomic_state *state, int conn_idx) } connector_state->best_encoder = new_encoder; - idx = drm_crtc_index(connector_state->crtc); + if (connector_state->crtc) { + idx = drm_crtc_index(connector_state->crtc); - crtc_state = state->crtc_states[idx]; - crtc_state->mode_changed = true; + crtc_state = state->crtc_states[idx]; + crtc_state->mode_changed = true; + } DRM_DEBUG_ATOMIC("[CONNECTOR:%d:%s] using [ENCODER:%d:%s] on [CRTC:%d]\n", connector->base.id, -- cgit v0.10.2 From 74d33293e467df61de1b1d8b2fbe29e550dec33b Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Sun, 2 Aug 2015 18:34:55 -0700 Subject: Linux 4.2-rc5 diff --git a/Makefile b/Makefile index afabc44..e79448d 100644 --- a/Makefile +++ b/Makefile @@ -1,7 +1,7 @@ VERSION = 4 PATCHLEVEL = 2 SUBLEVEL = 0 -EXTRAVERSION = -rc4 +EXTRAVERSION = -rc5 NAME = Hurr durr I'ma sheep # *DOCUMENTATION* -- cgit v0.10.2 From 8ec7cfce3762299ae289c384e281b2f4010ae231 Mon Sep 17 00:00:00 2001 From: Tomer Barletz Date: Sun, 2 Aug 2015 02:08:57 -0700 Subject: ALSA: oxygen: Fix logical-not-parentheses warning This fixes the following warning, that is seen with gcc 5.1: warning: logical not is only applied to the left hand side of comparison [-Wlogical-not-parentheses]. Signed-off-by: Tomer Barletz Signed-off-by: Takashi Iwai diff --git a/sound/pci/oxygen/oxygen_mixer.c b/sound/pci/oxygen/oxygen_mixer.c index 6492bca..4ca1266 100644 --- a/sound/pci/oxygen/oxygen_mixer.c +++ b/sound/pci/oxygen/oxygen_mixer.c @@ -88,7 +88,7 @@ static int dac_mute_put(struct snd_kcontrol *ctl, int changed; mutex_lock(&chip->mutex); - changed = !value->value.integer.value[0] != chip->dac_mute; + changed = (!value->value.integer.value[0]) != chip->dac_mute; if (changed) { chip->dac_mute = !value->value.integer.value[0]; chip->model.update_dac_mute(chip); -- cgit v0.10.2 From a094935e4ebdf5c22c45b8aeeb2d88e9e8c53dbf Mon Sep 17 00:00:00 2001 From: Bard Liao Date: Mon, 3 Aug 2015 12:17:39 +0800 Subject: ASoC: rt5645: Fix lost pin setting for DMIC1 I2S2_DAC pin can be used for I2S or GPIO. We should set it as GPIO if we use GPIO5 as DMIC1 data pin. Signed-off-by: Bard Liao Signed-off-by: Mark Brown diff --git a/sound/soc/codecs/rt5645.c b/sound/soc/codecs/rt5645.c index e9cc3aa..961bd7e 100644 --- a/sound/soc/codecs/rt5645.c +++ b/sound/soc/codecs/rt5645.c @@ -3341,6 +3341,8 @@ static int rt5645_i2c_probe(struct i2c_client *i2c, break; case RT5645_DMIC_DATA_GPIO5: + regmap_update_bits(rt5645->regmap, RT5645_GPIO_CTRL1, + RT5645_I2S2_DAC_PIN_MASK, RT5645_I2S2_DAC_PIN_GPIO); regmap_update_bits(rt5645->regmap, RT5645_DMIC_CTRL1, RT5645_DMIC_1_DP_MASK, RT5645_DMIC_1_DP_GPIO5); regmap_update_bits(rt5645->regmap, RT5645_GPIO_CTRL1, diff --git a/sound/soc/codecs/rt5645.h b/sound/soc/codecs/rt5645.h index 0353a6a..278bb9f 100644 --- a/sound/soc/codecs/rt5645.h +++ b/sound/soc/codecs/rt5645.h @@ -1693,6 +1693,10 @@ #define RT5645_GP6_PIN_SFT 6 #define RT5645_GP6_PIN_GPIO6 (0x0 << 6) #define RT5645_GP6_PIN_DMIC2_SDA (0x1 << 6) +#define RT5645_I2S2_DAC_PIN_MASK (0x1 << 4) +#define RT5645_I2S2_DAC_PIN_SFT 4 +#define RT5645_I2S2_DAC_PIN_I2S (0x0 << 4) +#define RT5645_I2S2_DAC_PIN_GPIO (0x1 << 4) #define RT5645_GP8_PIN_MASK (0x1 << 3) #define RT5645_GP8_PIN_SFT 3 #define RT5645_GP8_PIN_GPIO8 (0x0 << 3) -- cgit v0.10.2 From 9b06dc939489152b583131f49929ed1c6ae83740 Mon Sep 17 00:00:00 2001 From: Jeeja KP Date: Tue, 4 Aug 2015 09:28:38 +0530 Subject: ALSA: HDA: Fix stream assignment for host in decoupled mode This fixes issue in assigning host stream in case of decoupled mode. The check to verify if the stream is already in use was wrong so fix that Signed-off-by: Jeeja KP Signed-off-by: Vinod Koul Signed-off-by: Takashi Iwai diff --git a/sound/hda/ext/hdac_ext_stream.c b/sound/hda/ext/hdac_ext_stream.c index f8ffbdb..3de47dd 100644 --- a/sound/hda/ext/hdac_ext_stream.c +++ b/sound/hda/ext/hdac_ext_stream.c @@ -299,7 +299,7 @@ hdac_ext_host_stream_assign(struct hdac_ext_bus *ebus, if (stream->direction != substream->stream) continue; - if (stream->opened) { + if (!stream->opened) { if (!hstream->decoupled) snd_hdac_ext_stream_decouple(ebus, hstream, true); res = hstream; -- cgit v0.10.2 From 5d942ce63c8fd98794a8ba9af559925c8432a052 Mon Sep 17 00:00:00 2001 From: Jeeja KP Date: Tue, 4 Aug 2015 09:28:39 +0530 Subject: ALSA: HDA: Dont check return for snd_hdac_chip_readl The snd_hdac_chip_readl return can never be less than zeros, so no point in checking for the return value This fixes following static checker warnings in snd_hdac_ext_bus_parse_capabilities sound/hda/ext/hdac_ext_controller.c:47 snd_hdac_ext_bus_parse_capabilities() warn: unsigned 'offset' is never less than zero. sound/hda/ext/hdac_ext_controller.c:54 snd_hdac_ext_bus_parse_capabilities() warn: unsigned 'cur_cap' is never less than zero. Signed-off-by: Jeeja KP Reported-by: Dan Carpenter Signed-off-by: Vinod Koul Signed-off-by: Takashi Iwai diff --git a/sound/hda/ext/hdac_ext_controller.c b/sound/hda/ext/hdac_ext_controller.c index b2da19b..358f161 100644 --- a/sound/hda/ext/hdac_ext_controller.c +++ b/sound/hda/ext/hdac_ext_controller.c @@ -44,16 +44,10 @@ int snd_hdac_ext_bus_parse_capabilities(struct hdac_ext_bus *ebus) offset = snd_hdac_chip_readl(bus, LLCH); - if (offset < 0) - return -EIO; - /* Lets walk the linked capabilities list */ do { cur_cap = _snd_hdac_chip_read(l, bus, offset); - if (cur_cap < 0) - return -EIO; - dev_dbg(bus->dev, "Capability version: 0x%x\n", ((cur_cap & AZX_CAP_HDR_VER_MASK) >> AZX_CAP_HDR_VER_OFF)); -- cgit v0.10.2 From 5406898354ebfb11f49b955fb5e49a62786a542f Mon Sep 17 00:00:00 2001 From: Mengdong Lin Date: Tue, 4 Aug 2015 15:47:35 +0100 Subject: ASoC: topology: fix typo in soc_tplg_kcontrol_bind_io() Signed-off-by: Mengdong Lin Signed-off-by: Liam Girdwood Signed-off-by: Mark Brown diff --git a/sound/soc/soc-topology.c b/sound/soc/soc-topology.c index d096068..6a547c6 100644 --- a/sound/soc/soc-topology.c +++ b/sound/soc/soc-topology.c @@ -534,7 +534,7 @@ static int soc_tplg_kcontrol_bind_io(struct snd_soc_tplg_ctl_hdr *hdr, k->put = bops[i].put; if (k->get == NULL && bops[i].id == hdr->ops.get) k->get = bops[i].get; - if (k->info == NULL && ops[i].id == hdr->ops.info) + if (k->info == NULL && bops[i].id == hdr->ops.info) k->info = bops[i].info; } -- cgit v0.10.2 From c85523d1d97cc86aadc388221aa83ae9bc1e7cca Mon Sep 17 00:00:00 2001 From: Takashi Sakamoto Date: Wed, 5 Aug 2015 09:21:04 +0900 Subject: Revert "ALSA: fireworks: add support for AudioFire2 quirk" This reverts commit 9c6893e0be38b6ca9a56a854226e51dee0a16a5a. The fix is superseded by the next commit as a better implementation for supporting AudioFire2/AudioFire4/AudioFirePre8 quirks. Signed-off-by: Takashi Sakamoto Signed-off-by: Takashi Iwai diff --git a/sound/firewire/fireworks/fireworks.c b/sound/firewire/fireworks/fireworks.c index c670db4..2682e7e 100644 --- a/sound/firewire/fireworks/fireworks.c +++ b/sound/firewire/fireworks/fireworks.c @@ -248,8 +248,6 @@ efw_probe(struct fw_unit *unit, err = get_hardware_info(efw); if (err < 0) goto error; - if (entry->model_id == MODEL_ECHO_AUDIOFIRE_2) - efw->is_af2 = true; if (entry->model_id == MODEL_ECHO_AUDIOFIRE_9) efw->is_af9 = true; diff --git a/sound/firewire/fireworks/fireworks.h b/sound/firewire/fireworks/fireworks.h index c33252b..4f0201a 100644 --- a/sound/firewire/fireworks/fireworks.h +++ b/sound/firewire/fireworks/fireworks.h @@ -70,7 +70,6 @@ struct snd_efw { bool resp_addr_changable; /* for quirks */ - bool is_af2; bool is_af9; u32 firmware_version; diff --git a/sound/firewire/fireworks/fireworks_stream.c b/sound/firewire/fireworks/fireworks_stream.c index a0762dd..c55db1b 100644 --- a/sound/firewire/fireworks/fireworks_stream.c +++ b/sound/firewire/fireworks/fireworks_stream.c @@ -172,9 +172,6 @@ int snd_efw_stream_init_duplex(struct snd_efw *efw) efw->tx_stream.flags |= CIP_DBC_IS_END_EVENT; /* Fireworks reset dbc at bus reset. */ efw->tx_stream.flags |= CIP_SKIP_DBC_ZERO_CHECK; - /* AudioFire2 starts packets with non-zero dbc. */ - if (efw->is_af2) - efw->tx_stream.flags |= CIP_SKIP_INIT_DBC_CHECK; /* AudioFire9 always reports wrong dbs. */ if (efw->is_af9) efw->tx_stream.flags |= CIP_WRONG_DBS; -- cgit v0.10.2 From 18f5ed365d3f188a91149d528c853000330a4a58 Mon Sep 17 00:00:00 2001 From: Takashi Sakamoto Date: Wed, 5 Aug 2015 09:21:05 +0900 Subject: ALSA: fireworks/firewire-lib: add support for recent firmware quirk Fireworks uses TSB43CB43(IceLynx-Micro) as its IEC 61883-1/6 interface. This chip includes ARM7 core, and loads and runs program. The firmware is stored in on-board memory and loaded every powering-on from it. Echo Audio ships several versions of firmwares for each model. These firmwares have each quirk and the quirk changes a sequence of packets. As long as I investigated, AudioFire2/AudioFire4/AudioFirePre8 have a quirk to transfer a first packet with 0x02 in its dbc field. This causes ALSA Fireworks driver to detect discontinuity. In this case, firmware version 5.7.0, 5.7.3 and 5.8.0 are used. Payload CIP CIP quadlets header1 header2 02 00050002 90ffffff <- 42 0005000a 90013000 42 00050012 90014400 42 0005001a 90015800 02 0005001a 90ffffff 42 00050022 90019000 42 0005002a 9001a400 42 00050032 9001b800 02 00050032 90ffffff 42 0005003a 9001d000 42 00050042 9001e400 42 0005004a 9001f800 02 0005004a 90ffffff (AudioFire2 with firmware version 5.7.) $ dmesg snd-fireworks fw1.0: Detect discontinuity of CIP: 00 02 These models, AudioFire8 (since Jul 2009 ) and Gibson Robot Interface Pack series uses the same ARM binary as their firmware. Thus, this quirk may be observed among them. This commit adds a new member for AMDTP structure. This member represents the value of dbc field in a first AMDTP packet. Drivers can set it with a preferred value according to model's quirk. Tested-by: Johannes Oertei Signed-off-by: Takashi Sakamoto Cc: Signed-off-by: Takashi Iwai diff --git a/sound/firewire/amdtp.c b/sound/firewire/amdtp.c index 7bb988f..2a153d2 100644 --- a/sound/firewire/amdtp.c +++ b/sound/firewire/amdtp.c @@ -740,8 +740,9 @@ static int handle_in_packet(struct amdtp_stream *s, s->data_block_counter != UINT_MAX) data_block_counter = s->data_block_counter; - if (((s->flags & CIP_SKIP_DBC_ZERO_CHECK) && data_block_counter == 0) || - (s->data_block_counter == UINT_MAX)) { + if (((s->flags & CIP_SKIP_DBC_ZERO_CHECK) && + data_block_counter == s->tx_first_dbc) || + s->data_block_counter == UINT_MAX) { lost = false; } else if (!(s->flags & CIP_DBC_IS_END_EVENT)) { lost = data_block_counter != s->data_block_counter; diff --git a/sound/firewire/amdtp.h b/sound/firewire/amdtp.h index 26b9093..b2cf9e7 100644 --- a/sound/firewire/amdtp.h +++ b/sound/firewire/amdtp.h @@ -157,6 +157,8 @@ struct amdtp_stream { /* quirk: fixed interval of dbc between previos/current packets. */ unsigned int tx_dbc_interval; + /* quirk: indicate the value of dbc field in a first packet. */ + unsigned int tx_first_dbc; bool callbacked; wait_queue_head_t callback_wait; diff --git a/sound/firewire/fireworks/fireworks.c b/sound/firewire/fireworks/fireworks.c index 2682e7e..c94a432 100644 --- a/sound/firewire/fireworks/fireworks.c +++ b/sound/firewire/fireworks/fireworks.c @@ -248,8 +248,16 @@ efw_probe(struct fw_unit *unit, err = get_hardware_info(efw); if (err < 0) goto error; + /* AudioFire8 (since 2009) and AudioFirePre8 */ if (entry->model_id == MODEL_ECHO_AUDIOFIRE_9) efw->is_af9 = true; + /* These models uses the same firmware. */ + if (entry->model_id == MODEL_ECHO_AUDIOFIRE_2 || + entry->model_id == MODEL_ECHO_AUDIOFIRE_4 || + entry->model_id == MODEL_ECHO_AUDIOFIRE_9 || + entry->model_id == MODEL_GIBSON_RIP || + entry->model_id == MODEL_GIBSON_GOLDTOP) + efw->is_fireworks3 = true; snd_efw_proc_init(efw); diff --git a/sound/firewire/fireworks/fireworks.h b/sound/firewire/fireworks/fireworks.h index 4f0201a..084d414 100644 --- a/sound/firewire/fireworks/fireworks.h +++ b/sound/firewire/fireworks/fireworks.h @@ -71,6 +71,7 @@ struct snd_efw { /* for quirks */ bool is_af9; + bool is_fireworks3; u32 firmware_version; unsigned int midi_in_ports; diff --git a/sound/firewire/fireworks/fireworks_stream.c b/sound/firewire/fireworks/fireworks_stream.c index c55db1b..7e353f1 100644 --- a/sound/firewire/fireworks/fireworks_stream.c +++ b/sound/firewire/fireworks/fireworks_stream.c @@ -172,6 +172,15 @@ int snd_efw_stream_init_duplex(struct snd_efw *efw) efw->tx_stream.flags |= CIP_DBC_IS_END_EVENT; /* Fireworks reset dbc at bus reset. */ efw->tx_stream.flags |= CIP_SKIP_DBC_ZERO_CHECK; + /* + * But Recent firmwares starts packets with non-zero dbc. + * Driver version 5.7.6 installs firmware version 5.7.3. + */ + if (efw->is_fireworks3 && + (efw->firmware_version == 0x5070000 || + efw->firmware_version == 0x5070300 || + efw->firmware_version == 0x5080000)) + efw->tx_stream.tx_first_dbc = 0x02; /* AudioFire9 always reports wrong dbs. */ if (efw->is_af9) efw->tx_stream.flags |= CIP_WRONG_DBS; -- cgit v0.10.2 From 73851b36fe73819f8c201971e913324d4846a7ea Mon Sep 17 00:00:00 2001 From: Hui Wang Date: Wed, 5 Aug 2015 18:03:34 +0800 Subject: ALSA: hda - one Dell machine needs the headphone white noise fixup The fixup ALC292_FIXUP_DISABLE_AAMIX can fix the white noise of the headphone on this Dell machine. Cc: Signed-off-by: Hui Wang Signed-off-by: Takashi Iwai diff --git a/sound/pci/hda/patch_realtek.c b/sound/pci/hda/patch_realtek.c index c456c04..0b9847a 100644 --- a/sound/pci/hda/patch_realtek.c +++ b/sound/pci/hda/patch_realtek.c @@ -5189,6 +5189,7 @@ static const struct snd_pci_quirk alc269_fixup_tbl[] = { SND_PCI_QUIRK(0x1028, 0x06c7, "Dell", ALC255_FIXUP_DELL1_MIC_NO_PRESENCE), SND_PCI_QUIRK(0x1028, 0x06d9, "Dell", ALC293_FIXUP_DELL1_MIC_NO_PRESENCE), SND_PCI_QUIRK(0x1028, 0x06da, "Dell", ALC293_FIXUP_DELL1_MIC_NO_PRESENCE), + SND_PCI_QUIRK(0x1028, 0x06de, "Dell", ALC292_FIXUP_DISABLE_AAMIX), SND_PCI_QUIRK(0x1028, 0x164a, "Dell", ALC293_FIXUP_DELL1_MIC_NO_PRESENCE), SND_PCI_QUIRK(0x1028, 0x164b, "Dell", ALC293_FIXUP_DELL1_MIC_NO_PRESENCE), SND_PCI_QUIRK(0x103c, 0x1586, "HP", ALC269_FIXUP_HP_MUTE_LED_MIC2), -- cgit v0.10.2 From 6dc6db790a67d28e46abefc44ca1a3bd438b2920 Mon Sep 17 00:00:00 2001 From: "Subhransu S. Prusty" Date: Mon, 29 Jun 2015 17:36:44 +0100 Subject: ASoC: topology: Add subsequence in topology Some widgets may need sorting within, So add this support in topology. Signed-off-by: Subhransu S. Prusty Signed-off-by: Mark Brown diff --git a/include/uapi/sound/asoc.h b/include/uapi/sound/asoc.h index 1221520..7ae13fb 100644 --- a/include/uapi/sound/asoc.h +++ b/include/uapi/sound/asoc.h @@ -347,6 +347,7 @@ struct snd_soc_tplg_dapm_widget { __le32 reg; /* negative reg = no direct dapm */ __le32 shift; /* bits to shift */ __le32 mask; /* non-shifted mask */ + __le32 subseq; /* sort within widget type */ __u32 invert; /* invert the power bit */ __u32 ignore_suspend; /* kept enabled over suspend */ __u16 event_flags; diff --git a/sound/soc/soc-topology.c b/sound/soc/soc-topology.c index 6a547c6..9f2b048 100644 --- a/sound/soc/soc-topology.c +++ b/sound/soc/soc-topology.c @@ -1351,6 +1351,7 @@ static int soc_tplg_dapm_widget_create(struct soc_tplg *tplg, template.reg = w->reg; template.shift = w->shift; template.mask = w->mask; + template.subseq = w->subseq; template.on_val = w->invert ? 0 : 1; template.off_val = w->invert ? 1 : 0; template.ignore_suspend = w->ignore_suspend; -- cgit v0.10.2 From c3879956957b8de9fd6cbad604e668fd00c6506c Mon Sep 17 00:00:00 2001 From: Vinod Koul Date: Mon, 29 Jun 2015 17:36:46 +0100 Subject: ASoC: topology: add private data to manifest The topology file manifest should include a private data field. This allows vendors to specify vendor data in the manifest, like timestamps, hashes, additional information for removing platform configuration out of drivers and making these configurable per platform Signed-off-by: Vinod Koul Signed-off-by: Mark Brown diff --git a/include/uapi/sound/asoc.h b/include/uapi/sound/asoc.h index 7ae13fb..d550c8d 100644 --- a/include/uapi/sound/asoc.h +++ b/include/uapi/sound/asoc.h @@ -238,6 +238,7 @@ struct snd_soc_tplg_manifest { __le32 graph_elems; /* number of graph elements */ __le32 dai_elems; /* number of DAI elements */ __le32 dai_link_elems; /* number of DAI link elements */ + struct snd_soc_tplg_private priv; } __attribute__((packed)); /* -- cgit v0.10.2 From 28a87eebcad40101b1b273cbd4f2a02c104f9367 Mon Sep 17 00:00:00 2001 From: Mengdong Lin Date: Wed, 5 Aug 2015 14:41:13 +0100 Subject: ASoC: topology: Update TLV support so we can support more TLV types Currently the TLV topology structure is targeted at only supporting the DB scale data. This patch extends support for the other TLV types so they can be easily added at a later stage. TLV structure is moved to common topology control header since it's a common field for controls and can be processed in a general way. Users must set a proper access flag for a control since it's used to decide if the TLV field is valid and if a TLV callback is needed. Removed the following fields from topology TLV struct: - size/count: type can decide the size. - numid: not needed to initialize TLV for kcontrol. - data: replaced by the type specific struct. Added TLV structure to generic control header and removed TLV structure from mixer control. Signed-off-by: Mengdong Lin Signed-off-by: Liam Girdwood Signed-off-by: Mark Brown diff --git a/include/uapi/sound/asoc.h b/include/uapi/sound/asoc.h index 2819fc1..aa3a79b 100644 --- a/include/uapi/sound/asoc.h +++ b/include/uapi/sound/asoc.h @@ -137,11 +137,19 @@ struct snd_soc_tplg_private { /* * Kcontrol TLV data. */ +struct snd_soc_tplg_tlv_dbscale { + __le32 min; + __le32 step; + __le32 mute; +} __attribute__((packed)); + struct snd_soc_tplg_ctl_tlv { - __le32 size; /* in bytes aligned to 4 */ - __le32 numid; /* control element numeric identification */ - __le32 count; /* number of elem in data array */ - __le32 data[SND_SOC_TPLG_TLV_SIZE]; + __le32 size; /* in bytes of this structure */ + __le32 type; /* SNDRV_CTL_TLVT_*, type of TLV */ + union { + __le32 data[SND_SOC_TPLG_TLV_SIZE]; + struct snd_soc_tplg_tlv_dbscale scale; + }; } __attribute__((packed)); /* @@ -172,7 +180,7 @@ struct snd_soc_tplg_ctl_hdr { char name[SNDRV_CTL_ELEM_ID_NAME_MAXLEN]; __le32 access; struct snd_soc_tplg_kcontrol_ops_id ops; - __le32 tlv_size; /* non zero means control has TLV data */ + struct snd_soc_tplg_ctl_tlv tlv; } __attribute__((packed)); /* @@ -260,7 +268,6 @@ struct snd_soc_tplg_mixer_control { __le32 invert; __le32 num_channels; struct snd_soc_tplg_channel channel[SND_SOC_TPLG_MAX_CHAN]; - struct snd_soc_tplg_ctl_tlv tlv; struct snd_soc_tplg_private priv; } __attribute__((packed)); diff --git a/sound/soc/soc-topology.c b/sound/soc/soc-topology.c index 2c70f30..31068b8 100644 --- a/sound/soc/soc-topology.c +++ b/sound/soc/soc-topology.c @@ -33,6 +33,7 @@ #include #include #include +#include /* * We make several passes over the data (since it wont necessarily be ordered) @@ -579,28 +580,51 @@ static int soc_tplg_init_kcontrol(struct soc_tplg *tplg, return 0; } + +static int soc_tplg_create_tlv_db_scale(struct soc_tplg *tplg, + struct snd_kcontrol_new *kc, struct snd_soc_tplg_tlv_dbscale *scale) +{ + unsigned int item_len = 2 * sizeof(unsigned int); + unsigned int *p; + + p = kzalloc(item_len + 2 * sizeof(unsigned int), GFP_KERNEL); + if (!p) + return -ENOMEM; + + p[0] = SNDRV_CTL_TLVT_DB_SCALE; + p[1] = item_len; + p[2] = scale->min; + p[3] = (scale->step & TLV_DB_SCALE_MASK) + | (scale->mute ? TLV_DB_SCALE_MUTE : 0); + + kc->tlv.p = (void *)p; + return 0; +} + static int soc_tplg_create_tlv(struct soc_tplg *tplg, - struct snd_kcontrol_new *kc, struct snd_soc_tplg_ctl_tlv *tplg_tlv) + struct snd_kcontrol_new *kc, struct snd_soc_tplg_ctl_hdr *tc) { - struct snd_ctl_tlv *tlv; - int size; + struct snd_soc_tplg_ctl_tlv *tplg_tlv; - if (tplg_tlv->count == 0) + if (!(tc->access & SNDRV_CTL_ELEM_ACCESS_TLV_READWRITE)) return 0; - size = ((tplg_tlv->count + (sizeof(unsigned int) - 1)) & - ~(sizeof(unsigned int) - 1)); - tlv = kzalloc(sizeof(*tlv) + size, GFP_KERNEL); - if (tlv == NULL) - return -ENOMEM; - - dev_dbg(tplg->dev, " created TLV type %d size %d bytes\n", - tplg_tlv->numid, size); + if (tc->access & SNDRV_CTL_ELEM_ACCESS_TLV_CALLBACK) { + kc->tlv.c = snd_soc_bytes_tlv_callback; + } else { + tplg_tlv = &tc->tlv; + switch (tplg_tlv->type) { + case SNDRV_CTL_TLVT_DB_SCALE: + return soc_tplg_create_tlv_db_scale(tplg, kc, + &tplg_tlv->scale); - tlv->numid = tplg_tlv->numid; - tlv->length = size; - memcpy(&tlv->tlv[0], tplg_tlv->data, size); - kc->tlv.p = (void *)tlv; + /* TODO: add support for other TLV types */ + default: + dev_dbg(tplg->dev, "Unsupported TLV type %d\n", + tplg_tlv->type); + return -EINVAL; + } + } return 0; } @@ -772,7 +796,7 @@ static int soc_tplg_dmixer_create(struct soc_tplg *tplg, unsigned int count, } /* create any TLV data */ - soc_tplg_create_tlv(tplg, &kc, &mc->tlv); + soc_tplg_create_tlv(tplg, &kc, &mc->hdr); /* register control here */ err = soc_tplg_add_kcontrol(tplg, &kc, -- cgit v0.10.2 From cb88498b36ab01cbe3a0d95cd097e4afdff4c6fd Mon Sep 17 00:00:00 2001 From: Mengdong Lin Date: Wed, 5 Aug 2015 14:41:14 +0100 Subject: ASoC: topology: Add ops support to byte controls UAPI Add UAPI support for setting byte control ops. Rename the ops structure to be more generic so it can be sued by other objects too. Signed-off-by: Mengdong Lin Signed-off-by: Liam Girdwood Signed-off-by: Mark Brown diff --git a/include/uapi/sound/asoc.h b/include/uapi/sound/asoc.h index aa3a79b..d5281ac 100644 --- a/include/uapi/sound/asoc.h +++ b/include/uapi/sound/asoc.h @@ -163,9 +163,11 @@ struct snd_soc_tplg_channel { } __attribute__((packed)); /* - * Kcontrol Operations IDs + * Genericl Operations IDs, for binding Kcontrol or Bytes ext ops + * Kcontrol ops need get/put/info. + * Bytes ext ops need get/put. */ -struct snd_soc_tplg_kcontrol_ops_id { +struct snd_soc_tplg_io_ops { __le32 get; __le32 put; __le32 info; @@ -179,7 +181,7 @@ struct snd_soc_tplg_ctl_hdr { __le32 type; char name[SNDRV_CTL_ELEM_ID_NAME_MAXLEN]; __le32 access; - struct snd_soc_tplg_kcontrol_ops_id ops; + struct snd_soc_tplg_io_ops ops; struct snd_soc_tplg_ctl_tlv tlv; } __attribute__((packed)); @@ -311,6 +313,7 @@ struct snd_soc_tplg_bytes_control { __le32 mask; __le32 base; __le32 num_regs; + struct snd_soc_tplg_io_ops ext_ops; struct snd_soc_tplg_private priv; } __attribute__((packed)); -- cgit v0.10.2 From c7bcf8777a539e64dafc7417c00047aee6eb8909 Mon Sep 17 00:00:00 2001 From: Liam Girdwood Date: Wed, 5 Aug 2015 14:41:15 +0100 Subject: ASoC: topology: Add private data type and bump ABI version to 3 Add ID for standalone private data object types and bump ABI version to 3 in order to userpsace features. Signed-off-by: Liam Girdwood Signed-off-by: Mark Brown diff --git a/include/uapi/sound/asoc.h b/include/uapi/sound/asoc.h index d5281ac..51b8066 100644 --- a/include/uapi/sound/asoc.h +++ b/include/uapi/sound/asoc.h @@ -77,7 +77,7 @@ #define SND_SOC_TPLG_NUM_TEXTS 16 /* ABI version */ -#define SND_SOC_TPLG_ABI_VERSION 0x2 +#define SND_SOC_TPLG_ABI_VERSION 0x3 /* Max size of TLV data */ #define SND_SOC_TPLG_TLV_SIZE 32 @@ -97,7 +97,8 @@ #define SND_SOC_TPLG_TYPE_PCM 7 #define SND_SOC_TPLG_TYPE_MANIFEST 8 #define SND_SOC_TPLG_TYPE_CODEC_LINK 9 -#define SND_SOC_TPLG_TYPE_MAX SND_SOC_TPLG_TYPE_CODEC_LINK +#define SND_SOC_TPLG_TYPE_PDATA 10 +#define SND_SOC_TPLG_TYPE_MAX SND_SOC_TPLG_TYPE_PDATA /* vendor block IDs - please add new vendor types to end */ #define SND_SOC_TPLG_TYPE_VENDOR_FW 1000 -- cgit v0.10.2