diff options
Diffstat (limited to 'net')
71 files changed, 4993 insertions, 2618 deletions
diff --git a/net/atm/common.c b/net/atm/common.c index 49a872d..6dc1230 100644 --- a/net/atm/common.c +++ b/net/atm/common.c @@ -96,7 +96,7 @@ static void vcc_def_wakeup(struct sock *sk) rcu_read_lock(); wq = rcu_dereference(sk->sk_wq); - if (wq_has_sleeper(wq)) + if (skwq_has_sleeper(wq)) wake_up(&wq->wait); rcu_read_unlock(); } @@ -117,7 +117,7 @@ static void vcc_write_space(struct sock *sk) if (vcc_writable(sk)) { wq = rcu_dereference(sk->sk_wq); - if (wq_has_sleeper(wq)) + if (skwq_has_sleeper(wq)) wake_up_interruptible(&wq->wait); sk_wake_async(sk, SOCK_WAKE_SPACE, POLL_OUT); diff --git a/net/atm/mpc.h b/net/atm/mpc.h index 0919a88..cfc7b74 100644 --- a/net/atm/mpc.h +++ b/net/atm/mpc.h @@ -21,11 +21,11 @@ struct mpoa_client { uint8_t our_ctrl_addr[ATM_ESA_LEN]; /* MPC's control ATM address */ rwlock_t ingress_lock; - struct in_cache_ops *in_ops; /* ingress cache operations */ + const struct in_cache_ops *in_ops; /* ingress cache operations */ in_cache_entry *in_cache; /* the ingress cache of this MPC */ rwlock_t egress_lock; - struct eg_cache_ops *eg_ops; /* egress cache operations */ + const struct eg_cache_ops *eg_ops; /* egress cache operations */ eg_cache_entry *eg_cache; /* the egress cache of this MPC */ uint8_t *mps_macs; /* array of MPS MAC addresses, >=1 */ diff --git a/net/atm/mpoa_caches.c b/net/atm/mpoa_caches.c index d1b2d9a..9e60e74 100644 --- a/net/atm/mpoa_caches.c +++ b/net/atm/mpoa_caches.c @@ -534,7 +534,7 @@ static void eg_destroy_cache(struct mpoa_client *mpc) } -static struct in_cache_ops ingress_ops = { +static const struct in_cache_ops ingress_ops = { in_cache_add_entry, /* add_entry */ in_cache_get, /* get */ in_cache_get_with_mask, /* get_with_mask */ @@ -548,7 +548,7 @@ static struct in_cache_ops ingress_ops = { in_destroy_cache /* destroy_cache */ }; -static struct eg_cache_ops egress_ops = { +static const struct eg_cache_ops egress_ops = { eg_cache_add_entry, /* add_entry */ eg_cache_get_by_cache_id, /* get_by_cache_id */ eg_cache_get_by_tag, /* get_by_tag */ diff --git a/net/batman-adv/hard-interface.c b/net/batman-adv/hard-interface.c index f11345e..aa8867e 100644 --- a/net/batman-adv/hard-interface.c +++ b/net/batman-adv/hard-interface.c @@ -464,7 +464,8 @@ int batadv_hardif_enable_interface(struct batadv_hard_iface *hard_iface, hard_iface->soft_iface = soft_iface; bat_priv = netdev_priv(hard_iface->soft_iface); - ret = netdev_master_upper_dev_link(hard_iface->net_dev, soft_iface); + ret = netdev_master_upper_dev_link(hard_iface->net_dev, + soft_iface, NULL, NULL); if (ret) goto err_dev; diff --git a/net/bluetooth/af_bluetooth.c b/net/bluetooth/af_bluetooth.c index 70306cc..5785e8e 100644 --- a/net/bluetooth/af_bluetooth.c +++ b/net/bluetooth/af_bluetooth.c @@ -33,8 +33,6 @@ #include "selftest.h" -#define VERSION "2.21" - /* Bluetooth sockets */ #define BT_MAX_PROTO 8 static const struct net_proto_family *bt_proto[BT_MAX_PROTO]; @@ -671,7 +669,7 @@ static const struct file_operations bt_fops = { }; int bt_procfs_init(struct net *net, const char *name, - struct bt_sock_list* sk_list, + struct bt_sock_list *sk_list, int (* seq_show)(struct seq_file *, void *)) { sk_list->custom_seq_show = seq_show; @@ -687,7 +685,7 @@ void bt_procfs_cleanup(struct net *net, const char *name) } #else int bt_procfs_init(struct net *net, const char *name, - struct bt_sock_list* sk_list, + struct bt_sock_list *sk_list, int (* seq_show)(struct seq_file *, void *)) { return 0; @@ -715,7 +713,7 @@ static int __init bt_init(void) sock_skb_cb_check_size(sizeof(struct bt_skb_cb)); - BT_INFO("Core ver %s", VERSION); + BT_INFO("Core ver %s", BT_SUBSYS_VERSION); err = bt_selftest(); if (err < 0) @@ -789,7 +787,7 @@ subsys_initcall(bt_init); module_exit(bt_exit); MODULE_AUTHOR("Marcel Holtmann <marcel@holtmann.org>"); -MODULE_DESCRIPTION("Bluetooth Core ver " VERSION); -MODULE_VERSION(VERSION); +MODULE_DESCRIPTION("Bluetooth Core ver " BT_SUBSYS_VERSION); +MODULE_VERSION(BT_SUBSYS_VERSION); MODULE_LICENSE("GPL"); MODULE_ALIAS_NETPROTO(PF_BLUETOOTH); diff --git a/net/bluetooth/cmtp/core.c b/net/bluetooth/cmtp/core.c index 298ed37..9e59b66 100644 --- a/net/bluetooth/cmtp/core.c +++ b/net/bluetooth/cmtp/core.c @@ -178,8 +178,7 @@ static inline int cmtp_recv_frame(struct cmtp_session *session, struct sk_buff * cmtp_add_msgpart(session, id, skb->data + hdrlen, len); break; default: - if (session->reassembly[id] != NULL) - kfree_skb(session->reassembly[id]); + kfree_skb(session->reassembly[id]); session->reassembly[id] = NULL; break; } diff --git a/net/bluetooth/hci_conn.c b/net/bluetooth/hci_conn.c index 85b82f7..2d334e0 100644 --- a/net/bluetooth/hci_conn.c +++ b/net/bluetooth/hci_conn.c @@ -178,6 +178,10 @@ static void hci_connect_le_scan_remove(struct hci_conn *conn) hci_dev_hold(conn->hdev); hci_conn_get(conn); + /* Even though we hold a reference to the hdev, many other + * things might get cleaned up meanwhile, including the hdev's + * own workqueue, so we can't use that for scheduling. + */ schedule_work(&conn->le_scan_cleanup); } @@ -781,7 +785,7 @@ struct hci_conn *hci_connect_le(struct hci_dev *hdev, bdaddr_t *dst, u8 role) { struct hci_conn_params *params; - struct hci_conn *conn, *conn_unfinished; + struct hci_conn *conn; struct smp_irk *irk; struct hci_request req; int err; @@ -794,35 +798,22 @@ struct hci_conn *hci_connect_le(struct hci_dev *hdev, bdaddr_t *dst, return ERR_PTR(-EOPNOTSUPP); } - /* Some devices send ATT messages as soon as the physical link is - * established. To be able to handle these ATT messages, the user- - * space first establishes the connection and then starts the pairing - * process. - * - * So if a hci_conn object already exists for the following connection - * attempt, we simply update pending_sec_level and auth_type fields - * and return the object found. - */ - conn = hci_conn_hash_lookup_le(hdev, dst, dst_type); - conn_unfinished = NULL; - if (conn) { - if (conn->state == BT_CONNECT && - test_bit(HCI_CONN_SCANNING, &conn->flags)) { - BT_DBG("will continue unfinished conn %pMR", dst); - conn_unfinished = conn; - } else { - if (conn->pending_sec_level < sec_level) - conn->pending_sec_level = sec_level; - goto done; - } - } - /* Since the controller supports only one LE connection attempt at a * time, we return -EBUSY if there is any connection attempt running. */ if (hci_lookup_le_connect(hdev)) return ERR_PTR(-EBUSY); + /* If there's already a connection object but it's not in + * scanning state it means it must already be established, in + * which case we can't do anything else except report a failure + * to connect. + */ + conn = hci_conn_hash_lookup_le(hdev, dst, dst_type); + if (conn && !test_bit(HCI_CONN_SCANNING, &conn->flags)) { + return ERR_PTR(-EBUSY); + } + /* When given an identity address with existing identity * resolving key, the connection needs to be established * to a resolvable random address. @@ -838,23 +829,20 @@ struct hci_conn *hci_connect_le(struct hci_dev *hdev, bdaddr_t *dst, dst_type = ADDR_LE_DEV_RANDOM; } - if (conn_unfinished) { - conn = conn_unfinished; + if (conn) { bacpy(&conn->dst, dst); } else { conn = hci_conn_add(hdev, LE_LINK, dst, role); + if (!conn) + return ERR_PTR(-ENOMEM); + hci_conn_hold(conn); + conn->pending_sec_level = sec_level; } - if (!conn) - return ERR_PTR(-ENOMEM); - conn->dst_type = dst_type; conn->sec_level = BT_SECURITY_LOW; conn->conn_timeout = conn_timeout; - if (!conn_unfinished) - conn->pending_sec_level = sec_level; - hci_req_init(&req, hdev); /* Disable advertising if we're active. For master role @@ -918,37 +906,9 @@ create_conn: return ERR_PTR(err); } -done: - /* If this is continuation of connect started by hci_connect_le_scan, - * it already called hci_conn_hold and calling it again would mess the - * counter. - */ - if (!conn_unfinished) - hci_conn_hold(conn); - return conn; } -static void hci_connect_le_scan_complete(struct hci_dev *hdev, u8 status, - u16 opcode) -{ - struct hci_conn *conn; - - if (!status) - return; - - BT_ERR("Failed to add device to auto conn whitelist: status 0x%2.2x", - status); - - hci_dev_lock(hdev); - - conn = hci_conn_hash_lookup_state(hdev, LE_LINK, BT_CONNECT); - if (conn) - hci_le_conn_failed(conn, status); - - hci_dev_unlock(hdev); -} - static bool is_connected(struct hci_dev *hdev, bdaddr_t *addr, u8 type) { struct hci_conn *conn; @@ -964,10 +924,9 @@ static bool is_connected(struct hci_dev *hdev, bdaddr_t *addr, u8 type) } /* This function requires the caller holds hdev->lock */ -static int hci_explicit_conn_params_set(struct hci_request *req, +static int hci_explicit_conn_params_set(struct hci_dev *hdev, bdaddr_t *addr, u8 addr_type) { - struct hci_dev *hdev = req->hdev; struct hci_conn_params *params; if (is_connected(hdev, addr, addr_type)) @@ -995,7 +954,6 @@ static int hci_explicit_conn_params_set(struct hci_request *req, } params->explicit_connect = true; - __hci_update_background_scan(req); BT_DBG("addr %pMR (type %u) auto_connect %u", addr, addr_type, params->auto_connect); @@ -1006,11 +964,9 @@ static int hci_explicit_conn_params_set(struct hci_request *req, /* This function requires the caller holds hdev->lock */ struct hci_conn *hci_connect_le_scan(struct hci_dev *hdev, bdaddr_t *dst, u8 dst_type, u8 sec_level, - u16 conn_timeout, u8 role) + u16 conn_timeout) { struct hci_conn *conn; - struct hci_request req; - int err; /* Let's make sure that le is enabled.*/ if (!hci_dev_test_flag(hdev, HCI_LE_ENABLED)) { @@ -1038,29 +994,22 @@ struct hci_conn *hci_connect_le_scan(struct hci_dev *hdev, bdaddr_t *dst, BT_DBG("requesting refresh of dst_addr"); - conn = hci_conn_add(hdev, LE_LINK, dst, role); + conn = hci_conn_add(hdev, LE_LINK, dst, HCI_ROLE_MASTER); if (!conn) return ERR_PTR(-ENOMEM); - hci_req_init(&req, hdev); - - if (hci_explicit_conn_params_set(&req, dst, dst_type) < 0) + if (hci_explicit_conn_params_set(hdev, dst, dst_type) < 0) return ERR_PTR(-EBUSY); conn->state = BT_CONNECT; set_bit(HCI_CONN_SCANNING, &conn->flags); - - err = hci_req_run(&req, hci_connect_le_scan_complete); - if (err && err != -ENODATA) { - hci_conn_del(conn); - return ERR_PTR(err); - } - conn->dst_type = dst_type; conn->sec_level = BT_SECURITY_LOW; conn->pending_sec_level = sec_level; conn->conn_timeout = conn_timeout; + hci_update_background_scan(hdev); + done: hci_conn_hold(conn); return conn; diff --git a/net/bluetooth/hci_core.c b/net/bluetooth/hci_core.c index 62edbf1..89af7e4 100644 --- a/net/bluetooth/hci_core.c +++ b/net/bluetooth/hci_core.c @@ -56,15 +56,6 @@ DEFINE_MUTEX(hci_cb_list_lock); /* HCI ID Numbering */ static DEFINE_IDA(hci_index_ida); -/* ----- HCI requests ----- */ - -#define HCI_REQ_DONE 0 -#define HCI_REQ_PEND 1 -#define HCI_REQ_CANCELED 2 - -#define hci_req_lock(d) mutex_lock(&d->req_lock) -#define hci_req_unlock(d) mutex_unlock(&d->req_lock) - /* ---- HCI debugfs entries ---- */ static ssize_t dut_mode_read(struct file *file, char __user *user_buf, @@ -73,7 +64,7 @@ static ssize_t dut_mode_read(struct file *file, char __user *user_buf, struct hci_dev *hdev = file->private_data; char buf[3]; - buf[0] = hci_dev_test_flag(hdev, HCI_DUT_MODE) ? 'Y': 'N'; + buf[0] = hci_dev_test_flag(hdev, HCI_DUT_MODE) ? 'Y' : 'N'; buf[1] = '\n'; buf[2] = '\0'; return simple_read_from_buffer(user_buf, count, ppos, buf, 2); @@ -101,14 +92,14 @@ static ssize_t dut_mode_write(struct file *file, const char __user *user_buf, if (enable == hci_dev_test_flag(hdev, HCI_DUT_MODE)) return -EALREADY; - hci_req_lock(hdev); + hci_req_sync_lock(hdev); if (enable) skb = __hci_cmd_sync(hdev, HCI_OP_ENABLE_DUT_MODE, 0, NULL, HCI_CMD_TIMEOUT); else skb = __hci_cmd_sync(hdev, HCI_OP_RESET, 0, NULL, HCI_CMD_TIMEOUT); - hci_req_unlock(hdev); + hci_req_sync_unlock(hdev); if (IS_ERR(skb)) return PTR_ERR(skb); @@ -133,7 +124,7 @@ static ssize_t vendor_diag_read(struct file *file, char __user *user_buf, struct hci_dev *hdev = file->private_data; char buf[3]; - buf[0] = hci_dev_test_flag(hdev, HCI_VENDOR_DIAG) ? 'Y': 'N'; + buf[0] = hci_dev_test_flag(hdev, HCI_VENDOR_DIAG) ? 'Y' : 'N'; buf[1] = '\n'; buf[2] = '\0'; return simple_read_from_buffer(user_buf, count, ppos, buf, 2); @@ -165,9 +156,9 @@ static ssize_t vendor_diag_write(struct file *file, const char __user *user_buf, !test_bit(HCI_RUNNING, &hdev->flags)) goto done; - hci_req_lock(hdev); + hci_req_sync_lock(hdev); err = hdev->set_diag(hdev, enable); - hci_req_unlock(hdev); + hci_req_sync_unlock(hdev); if (err < 0) return err; @@ -198,197 +189,14 @@ static void hci_debugfs_create_basic(struct hci_dev *hdev) &vendor_diag_fops); } -/* ---- HCI requests ---- */ - -static void hci_req_sync_complete(struct hci_dev *hdev, u8 result, u16 opcode, - struct sk_buff *skb) -{ - BT_DBG("%s result 0x%2.2x", hdev->name, result); - - if (hdev->req_status == HCI_REQ_PEND) { - hdev->req_result = result; - hdev->req_status = HCI_REQ_DONE; - if (skb) - hdev->req_skb = skb_get(skb); - wake_up_interruptible(&hdev->req_wait_q); - } -} - -static void hci_req_cancel(struct hci_dev *hdev, int err) -{ - BT_DBG("%s err 0x%2.2x", hdev->name, err); - - if (hdev->req_status == HCI_REQ_PEND) { - hdev->req_result = err; - hdev->req_status = HCI_REQ_CANCELED; - wake_up_interruptible(&hdev->req_wait_q); - } -} - -struct sk_buff *__hci_cmd_sync_ev(struct hci_dev *hdev, u16 opcode, u32 plen, - const void *param, u8 event, u32 timeout) -{ - DECLARE_WAITQUEUE(wait, current); - struct hci_request req; - struct sk_buff *skb; - int err = 0; - - BT_DBG("%s", hdev->name); - - hci_req_init(&req, hdev); - - hci_req_add_ev(&req, opcode, plen, param, event); - - hdev->req_status = HCI_REQ_PEND; - - add_wait_queue(&hdev->req_wait_q, &wait); - set_current_state(TASK_INTERRUPTIBLE); - - err = hci_req_run_skb(&req, hci_req_sync_complete); - if (err < 0) { - remove_wait_queue(&hdev->req_wait_q, &wait); - set_current_state(TASK_RUNNING); - return ERR_PTR(err); - } - - schedule_timeout(timeout); - - remove_wait_queue(&hdev->req_wait_q, &wait); - - if (signal_pending(current)) - return ERR_PTR(-EINTR); - - switch (hdev->req_status) { - case HCI_REQ_DONE: - err = -bt_to_errno(hdev->req_result); - break; - - case HCI_REQ_CANCELED: - err = -hdev->req_result; - break; - - default: - err = -ETIMEDOUT; - break; - } - - hdev->req_status = hdev->req_result = 0; - skb = hdev->req_skb; - hdev->req_skb = NULL; - - BT_DBG("%s end: err %d", hdev->name, err); - - if (err < 0) { - kfree_skb(skb); - return ERR_PTR(err); - } - - if (!skb) - return ERR_PTR(-ENODATA); - - return skb; -} -EXPORT_SYMBOL(__hci_cmd_sync_ev); - -struct sk_buff *__hci_cmd_sync(struct hci_dev *hdev, u16 opcode, u32 plen, - const void *param, u32 timeout) -{ - return __hci_cmd_sync_ev(hdev, opcode, plen, param, 0, timeout); -} -EXPORT_SYMBOL(__hci_cmd_sync); - -/* Execute request and wait for completion. */ -static int __hci_req_sync(struct hci_dev *hdev, - void (*func)(struct hci_request *req, - unsigned long opt), - unsigned long opt, __u32 timeout) -{ - struct hci_request req; - DECLARE_WAITQUEUE(wait, current); - int err = 0; - - BT_DBG("%s start", hdev->name); - - hci_req_init(&req, hdev); - - hdev->req_status = HCI_REQ_PEND; - - func(&req, opt); - - add_wait_queue(&hdev->req_wait_q, &wait); - set_current_state(TASK_INTERRUPTIBLE); - - err = hci_req_run_skb(&req, hci_req_sync_complete); - if (err < 0) { - hdev->req_status = 0; - - remove_wait_queue(&hdev->req_wait_q, &wait); - set_current_state(TASK_RUNNING); - - /* ENODATA means the HCI request command queue is empty. - * This can happen when a request with conditionals doesn't - * trigger any commands to be sent. This is normal behavior - * and should not trigger an error return. - */ - if (err == -ENODATA) - return 0; - - return err; - } - - schedule_timeout(timeout); - - remove_wait_queue(&hdev->req_wait_q, &wait); - - if (signal_pending(current)) - return -EINTR; - - switch (hdev->req_status) { - case HCI_REQ_DONE: - err = -bt_to_errno(hdev->req_result); - break; - - case HCI_REQ_CANCELED: - err = -hdev->req_result; - break; - - default: - err = -ETIMEDOUT; - break; - } - - hdev->req_status = hdev->req_result = 0; - - BT_DBG("%s end: err %d", hdev->name, err); - - return err; -} - -static int hci_req_sync(struct hci_dev *hdev, - void (*req)(struct hci_request *req, - unsigned long opt), - unsigned long opt, __u32 timeout) -{ - int ret; - - if (!test_bit(HCI_UP, &hdev->flags)) - return -ENETDOWN; - - /* Serialize all requests */ - hci_req_lock(hdev); - ret = __hci_req_sync(hdev, req, opt, timeout); - hci_req_unlock(hdev); - - return ret; -} - -static void hci_reset_req(struct hci_request *req, unsigned long opt) +static int hci_reset_req(struct hci_request *req, unsigned long opt) { BT_DBG("%s %ld", req->hdev->name, opt); /* Reset device */ set_bit(HCI_RESET, &req->hdev->flags); hci_req_add(req, HCI_OP_RESET, 0, NULL); + return 0; } static void bredr_init(struct hci_request *req) @@ -428,7 +236,7 @@ static void amp_init1(struct hci_request *req) hci_req_add(req, HCI_OP_READ_LOCATION_DATA, 0, NULL); } -static void amp_init2(struct hci_request *req) +static int amp_init2(struct hci_request *req) { /* Read Local Supported Features. Not all AMP controllers * support this so it's placed conditionally in the second @@ -436,9 +244,11 @@ static void amp_init2(struct hci_request *req) */ if (req->hdev->commands[14] & 0x20) hci_req_add(req, HCI_OP_READ_LOCAL_FEATURES, 0, NULL); + + return 0; } -static void hci_init1_req(struct hci_request *req, unsigned long opt) +static int hci_init1_req(struct hci_request *req, unsigned long opt) { struct hci_dev *hdev = req->hdev; @@ -461,6 +271,8 @@ static void hci_init1_req(struct hci_request *req, unsigned long opt) BT_ERR("Unknown device type %d", hdev->dev_type); break; } + + return 0; } static void bredr_setup(struct hci_request *req) @@ -531,20 +343,30 @@ static void hci_setup_event_mask(struct hci_request *req) if (lmp_bredr_capable(hdev)) { events[4] |= 0x01; /* Flow Specification Complete */ - events[4] |= 0x02; /* Inquiry Result with RSSI */ - events[4] |= 0x04; /* Read Remote Extended Features Complete */ - events[5] |= 0x08; /* Synchronous Connection Complete */ - events[5] |= 0x10; /* Synchronous Connection Changed */ } else { /* Use a different default for LE-only devices */ memset(events, 0, sizeof(events)); - events[0] |= 0x10; /* Disconnection Complete */ - events[1] |= 0x08; /* Read Remote Version Information Complete */ events[1] |= 0x20; /* Command Complete */ events[1] |= 0x40; /* Command Status */ events[1] |= 0x80; /* Hardware Error */ - events[2] |= 0x04; /* Number of Completed Packets */ - events[3] |= 0x02; /* Data Buffer Overflow */ + + /* If the controller supports the Disconnect command, enable + * the corresponding event. In addition enable packet flow + * control related events. + */ + if (hdev->commands[0] & 0x20) { + events[0] |= 0x10; /* Disconnection Complete */ + events[2] |= 0x04; /* Number of Completed Packets */ + events[3] |= 0x02; /* Data Buffer Overflow */ + } + + /* If the controller supports the Read Remote Version + * Information command, enable the corresponding event. + */ + if (hdev->commands[2] & 0x80) + events[1] |= 0x08; /* Read Remote Version Information + * Complete + */ if (hdev->le_features[0] & HCI_LE_ENCRYPTION) { events[0] |= 0x80; /* Encryption Change */ @@ -552,9 +374,18 @@ static void hci_setup_event_mask(struct hci_request *req) } } - if (lmp_inq_rssi_capable(hdev)) + if (lmp_inq_rssi_capable(hdev) || + test_bit(HCI_QUIRK_FIXUP_INQUIRY_MODE, &hdev->quirks)) events[4] |= 0x02; /* Inquiry Result with RSSI */ + if (lmp_ext_feat_capable(hdev)) + events[4] |= 0x04; /* Read Remote Extended Features Complete */ + + if (lmp_esco_capable(hdev)) { + events[5] |= 0x08; /* Synchronous Connection Complete */ + events[5] |= 0x10; /* Synchronous Connection Changed */ + } + if (lmp_sniffsubr_capable(hdev)) events[5] |= 0x20; /* Sniff Subrating */ @@ -590,7 +421,7 @@ static void hci_setup_event_mask(struct hci_request *req) hci_req_add(req, HCI_OP_SET_EVENT_MASK, sizeof(events), events); } -static void hci_init2_req(struct hci_request *req, unsigned long opt) +static int hci_init2_req(struct hci_request *req, unsigned long opt) { struct hci_dev *hdev = req->hdev; @@ -670,6 +501,8 @@ static void hci_init2_req(struct hci_request *req, unsigned long opt) hci_req_add(req, HCI_OP_WRITE_AUTH_ENABLE, sizeof(enable), &enable); } + + return 0; } static void hci_setup_link_policy(struct hci_request *req) @@ -744,7 +577,7 @@ static void hci_set_event_mask_page_2(struct hci_request *req) hci_req_add(req, HCI_OP_SET_EVENT_MASK_PAGE_2, sizeof(events), events); } -static void hci_init3_req(struct hci_request *req, unsigned long opt) +static int hci_init3_req(struct hci_request *req, unsigned long opt) { struct hci_dev *hdev = req->hdev; u8 p; @@ -777,7 +610,6 @@ static void hci_init3_req(struct hci_request *req, unsigned long opt) u8 events[8]; memset(events, 0, sizeof(events)); - events[0] = 0x0f; if (hdev->le_features[0] & HCI_LE_ENCRYPTION) events[0] |= 0x10; /* LE Long Term Key Request */ @@ -804,6 +636,34 @@ static void hci_init3_req(struct hci_request *req, unsigned long opt) * Report */ + /* If the controller supports the LE Set Scan Enable command, + * enable the corresponding advertising report event. + */ + if (hdev->commands[26] & 0x08) + events[0] |= 0x02; /* LE Advertising Report */ + + /* If the controller supports the LE Create Connection + * command, enable the corresponding event. + */ + if (hdev->commands[26] & 0x10) + events[0] |= 0x01; /* LE Connection Complete */ + + /* If the controller supports the LE Connection Update + * command, enable the corresponding event. + */ + if (hdev->commands[27] & 0x04) + events[0] |= 0x04; /* LE Connection Update + * Complete + */ + + /* If the controller supports the LE Read Remote Used Features + * command, enable the corresponding event. + */ + if (hdev->commands[27] & 0x20) + events[0] |= 0x08; /* LE Read Remote Used + * Features Complete + */ + /* If the controller supports the LE Read Local P-256 * Public Key command, enable the corresponding event. */ @@ -856,9 +716,11 @@ static void hci_init3_req(struct hci_request *req, unsigned long opt) hci_req_add(req, HCI_OP_READ_LOCAL_EXT_FEATURES, sizeof(cp), &cp); } + + return 0; } -static void hci_init4_req(struct hci_request *req, unsigned long opt) +static int hci_init4_req(struct hci_request *req, unsigned long opt) { struct hci_dev *hdev = req->hdev; @@ -909,20 +771,22 @@ static void hci_init4_req(struct hci_request *req, unsigned long opt) hci_req_add(req, HCI_OP_WRITE_SC_SUPPORT, sizeof(support), &support); } + + return 0; } static int __hci_init(struct hci_dev *hdev) { int err; - err = __hci_req_sync(hdev, hci_init1_req, 0, HCI_INIT_TIMEOUT); + err = __hci_req_sync(hdev, hci_init1_req, 0, HCI_INIT_TIMEOUT, NULL); if (err < 0) return err; if (hci_dev_test_flag(hdev, HCI_SETUP)) hci_debugfs_create_basic(hdev); - err = __hci_req_sync(hdev, hci_init2_req, 0, HCI_INIT_TIMEOUT); + err = __hci_req_sync(hdev, hci_init2_req, 0, HCI_INIT_TIMEOUT, NULL); if (err < 0) return err; @@ -933,11 +797,11 @@ static int __hci_init(struct hci_dev *hdev) if (hdev->dev_type != HCI_BREDR) return 0; - err = __hci_req_sync(hdev, hci_init3_req, 0, HCI_INIT_TIMEOUT); + err = __hci_req_sync(hdev, hci_init3_req, 0, HCI_INIT_TIMEOUT, NULL); if (err < 0) return err; - err = __hci_req_sync(hdev, hci_init4_req, 0, HCI_INIT_TIMEOUT); + err = __hci_req_sync(hdev, hci_init4_req, 0, HCI_INIT_TIMEOUT, NULL); if (err < 0) return err; @@ -968,7 +832,7 @@ static int __hci_init(struct hci_dev *hdev) return 0; } -static void hci_init0_req(struct hci_request *req, unsigned long opt) +static int hci_init0_req(struct hci_request *req, unsigned long opt) { struct hci_dev *hdev = req->hdev; @@ -984,6 +848,8 @@ static void hci_init0_req(struct hci_request *req, unsigned long opt) /* Read BD Address */ if (hdev->set_bdaddr) hci_req_add(req, HCI_OP_READ_BD_ADDR, 0, NULL); + + return 0; } static int __hci_unconf_init(struct hci_dev *hdev) @@ -993,7 +859,7 @@ static int __hci_unconf_init(struct hci_dev *hdev) if (test_bit(HCI_QUIRK_RAW_DEVICE, &hdev->quirks)) return 0; - err = __hci_req_sync(hdev, hci_init0_req, 0, HCI_INIT_TIMEOUT); + err = __hci_req_sync(hdev, hci_init0_req, 0, HCI_INIT_TIMEOUT, NULL); if (err < 0) return err; @@ -1003,7 +869,7 @@ static int __hci_unconf_init(struct hci_dev *hdev) return 0; } -static void hci_scan_req(struct hci_request *req, unsigned long opt) +static int hci_scan_req(struct hci_request *req, unsigned long opt) { __u8 scan = opt; @@ -1011,9 +877,10 @@ static void hci_scan_req(struct hci_request *req, unsigned long opt) /* Inquiry and Page scans */ hci_req_add(req, HCI_OP_WRITE_SCAN_ENABLE, 1, &scan); + return 0; } -static void hci_auth_req(struct hci_request *req, unsigned long opt) +static int hci_auth_req(struct hci_request *req, unsigned long opt) { __u8 auth = opt; @@ -1021,9 +888,10 @@ static void hci_auth_req(struct hci_request *req, unsigned long opt) /* Authentication */ hci_req_add(req, HCI_OP_WRITE_AUTH_ENABLE, 1, &auth); + return 0; } -static void hci_encrypt_req(struct hci_request *req, unsigned long opt) +static int hci_encrypt_req(struct hci_request *req, unsigned long opt) { __u8 encrypt = opt; @@ -1031,9 +899,10 @@ static void hci_encrypt_req(struct hci_request *req, unsigned long opt) /* Encryption */ hci_req_add(req, HCI_OP_WRITE_ENCRYPT_MODE, 1, &encrypt); + return 0; } -static void hci_linkpol_req(struct hci_request *req, unsigned long opt) +static int hci_linkpol_req(struct hci_request *req, unsigned long opt) { __le16 policy = cpu_to_le16(opt); @@ -1041,6 +910,7 @@ static void hci_linkpol_req(struct hci_request *req, unsigned long opt) /* Default link policy */ hci_req_add(req, HCI_OP_WRITE_DEF_LINK_POLICY, 2, &policy); + return 0; } /* Get HCI device by index. @@ -1285,7 +1155,7 @@ static int inquiry_cache_dump(struct hci_dev *hdev, int num, __u8 *buf) return copied; } -static void hci_inq_req(struct hci_request *req, unsigned long opt) +static int hci_inq_req(struct hci_request *req, unsigned long opt) { struct hci_inquiry_req *ir = (struct hci_inquiry_req *) opt; struct hci_dev *hdev = req->hdev; @@ -1294,13 +1164,15 @@ static void hci_inq_req(struct hci_request *req, unsigned long opt) BT_DBG("%s", hdev->name); if (test_bit(HCI_INQUIRY, &hdev->flags)) - return; + return 0; /* Start Inquiry */ memcpy(&cp.lap, &ir->lap, 3); cp.length = ir->length; cp.num_rsp = ir->num_rsp; hci_req_add(req, HCI_OP_INQUIRY, sizeof(cp), &cp); + + return 0; } int hci_inquiry(void __user *arg) @@ -1351,7 +1223,7 @@ int hci_inquiry(void __user *arg) if (do_inquiry) { err = hci_req_sync(hdev, hci_inq_req, (unsigned long) &ir, - timeo); + timeo, NULL); if (err < 0) goto done; @@ -1404,7 +1276,7 @@ static int hci_dev_do_open(struct hci_dev *hdev) BT_DBG("%s %p", hdev->name, hdev); - hci_req_lock(hdev); + hci_req_sync_lock(hdev); if (hci_dev_test_flag(hdev, HCI_UNREGISTER)) { ret = -ENODEV; @@ -1557,7 +1429,7 @@ static int hci_dev_do_open(struct hci_dev *hdev) } done: - hci_req_unlock(hdev); + hci_req_sync_unlock(hdev); return ret; } @@ -1651,12 +1523,12 @@ int hci_dev_do_close(struct hci_dev *hdev) cancel_delayed_work(&hdev->power_off); - hci_req_cancel(hdev, ENODEV); - hci_req_lock(hdev); + hci_request_cancel_all(hdev); + hci_req_sync_lock(hdev); if (!test_and_clear_bit(HCI_UP, &hdev->flags)) { cancel_delayed_work_sync(&hdev->cmd_timer); - hci_req_unlock(hdev); + hci_req_sync_unlock(hdev); return 0; } @@ -1674,9 +1546,6 @@ int hci_dev_do_close(struct hci_dev *hdev) if (hci_dev_test_and_clear_flag(hdev, HCI_SERVICE_CACHE)) cancel_delayed_work(&hdev->service_cache); - cancel_delayed_work_sync(&hdev->le_scan_disable); - cancel_delayed_work_sync(&hdev->le_scan_restart); - if (hci_dev_test_flag(hdev, HCI_MGMT)) cancel_delayed_work_sync(&hdev->rpa_expired); @@ -1717,7 +1586,7 @@ int hci_dev_do_close(struct hci_dev *hdev) if (test_bit(HCI_QUIRK_RESET_ON_CLOSE, &hdev->quirks) && !auto_off && !hci_dev_test_flag(hdev, HCI_UNCONFIGURED)) { set_bit(HCI_INIT, &hdev->flags); - __hci_req_sync(hdev, hci_reset_req, 0, HCI_CMD_TIMEOUT); + __hci_req_sync(hdev, hci_reset_req, 0, HCI_CMD_TIMEOUT, NULL); clear_bit(HCI_INIT, &hdev->flags); } @@ -1754,7 +1623,7 @@ int hci_dev_do_close(struct hci_dev *hdev) memset(hdev->dev_class, 0, sizeof(hdev->dev_class)); bacpy(&hdev->random_addr, BDADDR_ANY); - hci_req_unlock(hdev); + hci_req_sync_unlock(hdev); hci_dev_put(hdev); return 0; @@ -1790,7 +1659,7 @@ static int hci_dev_do_reset(struct hci_dev *hdev) BT_DBG("%s %p", hdev->name, hdev); - hci_req_lock(hdev); + hci_req_sync_lock(hdev); /* Drop queues */ skb_queue_purge(&hdev->rx_q); @@ -1812,9 +1681,9 @@ static int hci_dev_do_reset(struct hci_dev *hdev) atomic_set(&hdev->cmd_cnt, 1); hdev->acl_cnt = 0; hdev->sco_cnt = 0; hdev->le_cnt = 0; - ret = __hci_req_sync(hdev, hci_reset_req, 0, HCI_INIT_TIMEOUT); + ret = __hci_req_sync(hdev, hci_reset_req, 0, HCI_INIT_TIMEOUT, NULL); - hci_req_unlock(hdev); + hci_req_sync_unlock(hdev); return ret; } @@ -1947,7 +1816,7 @@ int hci_dev_cmd(unsigned int cmd, void __user *arg) switch (cmd) { case HCISETAUTH: err = hci_req_sync(hdev, hci_auth_req, dr.dev_opt, - HCI_INIT_TIMEOUT); + HCI_INIT_TIMEOUT, NULL); break; case HCISETENCRYPT: @@ -1959,18 +1828,18 @@ int hci_dev_cmd(unsigned int cmd, void __user *arg) if (!test_bit(HCI_AUTH, &hdev->flags)) { /* Auth must be enabled first */ err = hci_req_sync(hdev, hci_auth_req, dr.dev_opt, - HCI_INIT_TIMEOUT); + HCI_INIT_TIMEOUT, NULL); if (err) break; } err = hci_req_sync(hdev, hci_encrypt_req, dr.dev_opt, - HCI_INIT_TIMEOUT); + HCI_INIT_TIMEOUT, NULL); break; case HCISETSCAN: err = hci_req_sync(hdev, hci_scan_req, dr.dev_opt, - HCI_INIT_TIMEOUT); + HCI_INIT_TIMEOUT, NULL); /* Ensure that the connectable and discoverable states * get correctly modified as this was a non-mgmt change. @@ -1981,7 +1850,7 @@ int hci_dev_cmd(unsigned int cmd, void __user *arg) case HCISETLINKPOL: err = hci_req_sync(hdev, hci_linkpol_req, dr.dev_opt, - HCI_INIT_TIMEOUT); + HCI_INIT_TIMEOUT, NULL); break; case HCISETLINKMODE: @@ -2731,7 +2600,8 @@ struct adv_info *hci_find_adv_instance(struct hci_dev *hdev, u8 instance) } /* This function requires the caller holds hdev->lock */ -struct adv_info *hci_get_next_instance(struct hci_dev *hdev, u8 instance) { +struct adv_info *hci_get_next_instance(struct hci_dev *hdev, u8 instance) +{ struct adv_info *cur_instance; cur_instance = hci_find_adv_instance(hdev, instance); @@ -3024,181 +2894,16 @@ void hci_conn_params_clear_disabled(struct hci_dev *hdev) } /* This function requires the caller holds hdev->lock */ -void hci_conn_params_clear_all(struct hci_dev *hdev) +static void hci_conn_params_clear_all(struct hci_dev *hdev) { struct hci_conn_params *params, *tmp; list_for_each_entry_safe(params, tmp, &hdev->le_conn_params, list) hci_conn_params_free(params); - hci_update_background_scan(hdev); - BT_DBG("All LE connection parameters were removed"); } -static void inquiry_complete(struct hci_dev *hdev, u8 status, u16 opcode) -{ - if (status) { - BT_ERR("Failed to start inquiry: status %d", status); - - hci_dev_lock(hdev); - hci_discovery_set_state(hdev, DISCOVERY_STOPPED); - hci_dev_unlock(hdev); - return; - } -} - -static void le_scan_disable_work_complete(struct hci_dev *hdev, u8 status, - u16 opcode) -{ - /* General inquiry access code (GIAC) */ - u8 lap[3] = { 0x33, 0x8b, 0x9e }; - struct hci_cp_inquiry cp; - int err; - - if (status) { - BT_ERR("Failed to disable LE scanning: status %d", status); - return; - } - - hdev->discovery.scan_start = 0; - - switch (hdev->discovery.type) { - case DISCOV_TYPE_LE: - hci_dev_lock(hdev); - hci_discovery_set_state(hdev, DISCOVERY_STOPPED); - hci_dev_unlock(hdev); - break; - - case DISCOV_TYPE_INTERLEAVED: - hci_dev_lock(hdev); - - if (test_bit(HCI_QUIRK_SIMULTANEOUS_DISCOVERY, - &hdev->quirks)) { - /* If we were running LE only scan, change discovery - * state. If we were running both LE and BR/EDR inquiry - * simultaneously, and BR/EDR inquiry is already - * finished, stop discovery, otherwise BR/EDR inquiry - * will stop discovery when finished. If we will resolve - * remote device name, do not change discovery state. - */ - if (!test_bit(HCI_INQUIRY, &hdev->flags) && - hdev->discovery.state != DISCOVERY_RESOLVING) - hci_discovery_set_state(hdev, - DISCOVERY_STOPPED); - } else { - struct hci_request req; - - hci_inquiry_cache_flush(hdev); - - hci_req_init(&req, hdev); - - memset(&cp, 0, sizeof(cp)); - memcpy(&cp.lap, lap, sizeof(cp.lap)); - cp.length = DISCOV_INTERLEAVED_INQUIRY_LEN; - hci_req_add(&req, HCI_OP_INQUIRY, sizeof(cp), &cp); - - err = hci_req_run(&req, inquiry_complete); - if (err) { - BT_ERR("Inquiry request failed: err %d", err); - hci_discovery_set_state(hdev, - DISCOVERY_STOPPED); - } - } - - hci_dev_unlock(hdev); - break; - } -} - -static void le_scan_disable_work(struct work_struct *work) -{ - struct hci_dev *hdev = container_of(work, struct hci_dev, - le_scan_disable.work); - struct hci_request req; - int err; - - BT_DBG("%s", hdev->name); - - cancel_delayed_work_sync(&hdev->le_scan_restart); - - hci_req_init(&req, hdev); - - hci_req_add_le_scan_disable(&req); - - err = hci_req_run(&req, le_scan_disable_work_complete); - if (err) - BT_ERR("Disable LE scanning request failed: err %d", err); -} - -static void le_scan_restart_work_complete(struct hci_dev *hdev, u8 status, - u16 opcode) -{ - unsigned long timeout, duration, scan_start, now; - - BT_DBG("%s", hdev->name); - - if (status) { - BT_ERR("Failed to restart LE scan: status %d", status); - return; - } - - if (!test_bit(HCI_QUIRK_STRICT_DUPLICATE_FILTER, &hdev->quirks) || - !hdev->discovery.scan_start) - return; - - /* When the scan was started, hdev->le_scan_disable has been queued - * after duration from scan_start. During scan restart this job - * has been canceled, and we need to queue it again after proper - * timeout, to make sure that scan does not run indefinitely. - */ - duration = hdev->discovery.scan_duration; - scan_start = hdev->discovery.scan_start; - now = jiffies; - if (now - scan_start <= duration) { - int elapsed; - - if (now >= scan_start) - elapsed = now - scan_start; - else - elapsed = ULONG_MAX - scan_start + now; - - timeout = duration - elapsed; - } else { - timeout = 0; - } - queue_delayed_work(hdev->workqueue, - &hdev->le_scan_disable, timeout); -} - -static void le_scan_restart_work(struct work_struct *work) -{ - struct hci_dev *hdev = container_of(work, struct hci_dev, - le_scan_restart.work); - struct hci_request req; - struct hci_cp_le_set_scan_enable cp; - int err; - - BT_DBG("%s", hdev->name); - - /* If controller is not scanning we are done. */ - if (!hci_dev_test_flag(hdev, HCI_LE_SCAN)) - return; - - hci_req_init(&req, hdev); - - hci_req_add_le_scan_disable(&req); - - memset(&cp, 0, sizeof(cp)); - cp.enable = LE_SCAN_ENABLE; - cp.filter_dup = LE_SCAN_FILTER_DUP_ENABLE; - hci_req_add(&req, HCI_OP_LE_SET_SCAN_ENABLE, sizeof(cp), &cp); - - err = hci_req_run(&req, le_scan_restart_work_complete); - if (err) - BT_ERR("Restart LE scan request failed: err %d", err); -} - /* Copy the Identity Address of the controller. * * If the controller has a public BD_ADDR, then by default use that one. @@ -3298,8 +3003,6 @@ struct hci_dev *hci_alloc_dev(void) INIT_DELAYED_WORK(&hdev->power_off, hci_power_off); INIT_DELAYED_WORK(&hdev->discov_off, hci_discov_off); - INIT_DELAYED_WORK(&hdev->le_scan_disable, le_scan_disable_work); - INIT_DELAYED_WORK(&hdev->le_scan_restart, le_scan_restart_work); INIT_DELAYED_WORK(&hdev->adv_instance_expire, hci_adv_timeout_expire); skb_queue_head_init(&hdev->rx_q); @@ -3310,6 +3013,8 @@ struct hci_dev *hci_alloc_dev(void) INIT_DELAYED_WORK(&hdev->cmd_timer, hci_cmd_timeout); + hci_request_setup(hdev); + hci_init_sysfs(hdev); discovery_init(hdev); @@ -3520,7 +3225,7 @@ int hci_reset_dev(struct hci_dev *hdev) if (!skb) return -ENOMEM; - bt_cb(skb)->pkt_type = HCI_EVENT_PKT; + hci_skb_pkt_type(skb) = HCI_EVENT_PKT; memcpy(skb_put(skb, 3), hw_err, 3); /* Send Hardware Error to upper stack */ @@ -3537,9 +3242,9 @@ int hci_recv_frame(struct hci_dev *hdev, struct sk_buff *skb) return -ENXIO; } - if (bt_cb(skb)->pkt_type != HCI_EVENT_PKT && - bt_cb(skb)->pkt_type != HCI_ACLDATA_PKT && - bt_cb(skb)->pkt_type != HCI_SCODATA_PKT) { + if (hci_skb_pkt_type(skb) != HCI_EVENT_PKT && + hci_skb_pkt_type(skb) != HCI_ACLDATA_PKT && + hci_skb_pkt_type(skb) != HCI_SCODATA_PKT) { kfree_skb(skb); return -EINVAL; } @@ -3561,7 +3266,7 @@ EXPORT_SYMBOL(hci_recv_frame); int hci_recv_diag(struct hci_dev *hdev, struct sk_buff *skb) { /* Mark as diagnostic packet */ - bt_cb(skb)->pkt_type = HCI_DIAG_PKT; + hci_skb_pkt_type(skb) = HCI_DIAG_PKT; /* Time stamp */ __net_timestamp(skb); @@ -3603,7 +3308,8 @@ static void hci_send_frame(struct hci_dev *hdev, struct sk_buff *skb) { int err; - BT_DBG("%s type %d len %d", hdev->name, bt_cb(skb)->pkt_type, skb->len); + BT_DBG("%s type %d len %d", hdev->name, hci_skb_pkt_type(skb), + skb->len); /* Time stamp */ __net_timestamp(skb); @@ -3648,7 +3354,7 @@ int hci_send_cmd(struct hci_dev *hdev, __u16 opcode, __u32 plen, /* Stand-alone HCI commands must be flagged as * single-command requests. */ - bt_cb(skb)->hci.req_start = true; + bt_cb(skb)->hci.req_flags |= HCI_REQ_START; skb_queue_tail(&hdev->cmd_q, skb); queue_work(hdev->workqueue, &hdev->cmd_work); @@ -3685,9 +3391,9 @@ struct sk_buff *hci_cmd_sync(struct hci_dev *hdev, u16 opcode, u32 plen, bt_dev_dbg(hdev, "opcode 0x%4.4x plen %d", opcode, plen); - hci_req_lock(hdev); + hci_req_sync_lock(hdev); skb = __hci_cmd_sync(hdev, opcode, plen, param, timeout); - hci_req_unlock(hdev); + hci_req_sync_unlock(hdev); return skb; } @@ -3716,7 +3422,7 @@ static void hci_queue_acl(struct hci_chan *chan, struct sk_buff_head *queue, skb->len = skb_headlen(skb); skb->data_len = 0; - bt_cb(skb)->pkt_type = HCI_ACLDATA_PKT; + hci_skb_pkt_type(skb) = HCI_ACLDATA_PKT; switch (hdev->dev_type) { case HCI_BREDR: @@ -3756,7 +3462,7 @@ static void hci_queue_acl(struct hci_chan *chan, struct sk_buff_head *queue, do { skb = list; list = list->next; - bt_cb(skb)->pkt_type = HCI_ACLDATA_PKT; + hci_skb_pkt_type(skb) = HCI_ACLDATA_PKT; hci_add_acl_hdr(skb, conn->handle, flags); BT_DBG("%s frag %p len %d", hdev->name, skb, skb->len); @@ -3794,7 +3500,7 @@ void hci_send_sco(struct hci_conn *conn, struct sk_buff *skb) skb_reset_transport_header(skb); memcpy(skb_transport_header(skb), &hdr, HCI_SCO_HDR_SIZE); - bt_cb(skb)->pkt_type = HCI_SCODATA_PKT; + hci_skb_pkt_type(skb) = HCI_SCODATA_PKT; skb_queue_tail(&conn->data_q, skb); queue_work(hdev->workqueue, &hdev->tx_work); @@ -4345,7 +4051,7 @@ static bool hci_req_is_complete(struct hci_dev *hdev) if (!skb) return true; - return bt_cb(skb)->hci.req_start; + return (bt_cb(skb)->hci.req_flags & HCI_REQ_START); } static void hci_resend_last(struct hci_dev *hdev) @@ -4405,20 +4111,20 @@ void hci_req_cmd_complete(struct hci_dev *hdev, u16 opcode, u8 status, * callback would be found in hdev->sent_cmd instead of the * command queue (hdev->cmd_q). */ - if (bt_cb(hdev->sent_cmd)->hci.req_complete) { - *req_complete = bt_cb(hdev->sent_cmd)->hci.req_complete; + if (bt_cb(hdev->sent_cmd)->hci.req_flags & HCI_REQ_SKB) { + *req_complete_skb = bt_cb(hdev->sent_cmd)->hci.req_complete_skb; return; } - if (bt_cb(hdev->sent_cmd)->hci.req_complete_skb) { - *req_complete_skb = bt_cb(hdev->sent_cmd)->hci.req_complete_skb; + if (bt_cb(hdev->sent_cmd)->hci.req_complete) { + *req_complete = bt_cb(hdev->sent_cmd)->hci.req_complete; return; } /* Remove all pending commands belonging to this request */ spin_lock_irqsave(&hdev->cmd_q.lock, flags); while ((skb = __skb_dequeue(&hdev->cmd_q))) { - if (bt_cb(skb)->hci.req_start) { + if (bt_cb(skb)->hci.req_flags & HCI_REQ_START) { __skb_queue_head(&hdev->cmd_q, skb); break; } @@ -4453,7 +4159,7 @@ static void hci_rx_work(struct work_struct *work) if (test_bit(HCI_INIT, &hdev->flags)) { /* Don't process data packets in this states. */ - switch (bt_cb(skb)->pkt_type) { + switch (hci_skb_pkt_type(skb)) { case HCI_ACLDATA_PKT: case HCI_SCODATA_PKT: kfree_skb(skb); @@ -4462,7 +4168,7 @@ static void hci_rx_work(struct work_struct *work) } /* Process frame */ - switch (bt_cb(skb)->pkt_type) { + switch (hci_skb_pkt_type(skb)) { case HCI_EVENT_PKT: BT_DBG("%s Event packet", hdev->name); hci_event_packet(hdev, skb); diff --git a/net/bluetooth/hci_request.c b/net/bluetooth/hci_request.c index 981f8a2..e639671 100644 --- a/net/bluetooth/hci_request.c +++ b/net/bluetooth/hci_request.c @@ -27,6 +27,10 @@ #include "smp.h" #include "hci_request.h" +#define HCI_REQ_DONE 0 +#define HCI_REQ_PEND 1 +#define HCI_REQ_CANCELED 2 + void hci_req_init(struct hci_request *req, struct hci_dev *hdev) { skb_queue_head_init(&req->cmd_q); @@ -56,8 +60,12 @@ static int req_run(struct hci_request *req, hci_req_complete_t complete, return -ENODATA; skb = skb_peek_tail(&req->cmd_q); - bt_cb(skb)->hci.req_complete = complete; - bt_cb(skb)->hci.req_complete_skb = complete_skb; + if (complete) { + bt_cb(skb)->hci.req_complete = complete; + } else if (complete_skb) { + bt_cb(skb)->hci.req_complete_skb = complete_skb; + bt_cb(skb)->hci.req_flags |= HCI_REQ_SKB; + } spin_lock_irqsave(&hdev->cmd_q.lock, flags); skb_queue_splice_tail(&req->cmd_q, &hdev->cmd_q); @@ -78,6 +86,203 @@ int hci_req_run_skb(struct hci_request *req, hci_req_complete_skb_t complete) return req_run(req, NULL, complete); } +static void hci_req_sync_complete(struct hci_dev *hdev, u8 result, u16 opcode, + struct sk_buff *skb) +{ + BT_DBG("%s result 0x%2.2x", hdev->name, result); + + if (hdev->req_status == HCI_REQ_PEND) { + hdev->req_result = result; + hdev->req_status = HCI_REQ_DONE; + if (skb) + hdev->req_skb = skb_get(skb); + wake_up_interruptible(&hdev->req_wait_q); + } +} + +void hci_req_sync_cancel(struct hci_dev *hdev, int err) +{ + BT_DBG("%s err 0x%2.2x", hdev->name, err); + + if (hdev->req_status == HCI_REQ_PEND) { + hdev->req_result = err; + hdev->req_status = HCI_REQ_CANCELED; + wake_up_interruptible(&hdev->req_wait_q); + } +} + +struct sk_buff *__hci_cmd_sync_ev(struct hci_dev *hdev, u16 opcode, u32 plen, + const void *param, u8 event, u32 timeout) +{ + DECLARE_WAITQUEUE(wait, current); + struct hci_request req; + struct sk_buff *skb; + int err = 0; + + BT_DBG("%s", hdev->name); + + hci_req_init(&req, hdev); + + hci_req_add_ev(&req, opcode, plen, param, event); + + hdev->req_status = HCI_REQ_PEND; + + add_wait_queue(&hdev->req_wait_q, &wait); + set_current_state(TASK_INTERRUPTIBLE); + + err = hci_req_run_skb(&req, hci_req_sync_complete); + if (err < 0) { + remove_wait_queue(&hdev->req_wait_q, &wait); + set_current_state(TASK_RUNNING); + return ERR_PTR(err); + } + + schedule_timeout(timeout); + + remove_wait_queue(&hdev->req_wait_q, &wait); + + if (signal_pending(current)) + return ERR_PTR(-EINTR); + + switch (hdev->req_status) { + case HCI_REQ_DONE: + err = -bt_to_errno(hdev->req_result); + break; + + case HCI_REQ_CANCELED: + err = -hdev->req_result; + break; + + default: + err = -ETIMEDOUT; + break; + } + + hdev->req_status = hdev->req_result = 0; + skb = hdev->req_skb; + hdev->req_skb = NULL; + + BT_DBG("%s end: err %d", hdev->name, err); + + if (err < 0) { + kfree_skb(skb); + return ERR_PTR(err); + } + + if (!skb) + return ERR_PTR(-ENODATA); + + return skb; +} +EXPORT_SYMBOL(__hci_cmd_sync_ev); + +struct sk_buff *__hci_cmd_sync(struct hci_dev *hdev, u16 opcode, u32 plen, + const void *param, u32 timeout) +{ + return __hci_cmd_sync_ev(hdev, opcode, plen, param, 0, timeout); +} +EXPORT_SYMBOL(__hci_cmd_sync); + +/* Execute request and wait for completion. */ +int __hci_req_sync(struct hci_dev *hdev, int (*func)(struct hci_request *req, + unsigned long opt), + unsigned long opt, u32 timeout, u8 *hci_status) +{ + struct hci_request req; + DECLARE_WAITQUEUE(wait, current); + int err = 0; + + BT_DBG("%s start", hdev->name); + + hci_req_init(&req, hdev); + + hdev->req_status = HCI_REQ_PEND; + + err = func(&req, opt); + if (err) { + if (hci_status) + *hci_status = HCI_ERROR_UNSPECIFIED; + return err; + } + + add_wait_queue(&hdev->req_wait_q, &wait); + set_current_state(TASK_INTERRUPTIBLE); + + err = hci_req_run_skb(&req, hci_req_sync_complete); + if (err < 0) { + hdev->req_status = 0; + + remove_wait_queue(&hdev->req_wait_q, &wait); + set_current_state(TASK_RUNNING); + + /* ENODATA means the HCI request command queue is empty. + * This can happen when a request with conditionals doesn't + * trigger any commands to be sent. This is normal behavior + * and should not trigger an error return. + */ + if (err == -ENODATA) { + if (hci_status) + *hci_status = 0; + return 0; + } + + if (hci_status) + *hci_status = HCI_ERROR_UNSPECIFIED; + + return err; + } + + schedule_timeout(timeout); + + remove_wait_queue(&hdev->req_wait_q, &wait); + + if (signal_pending(current)) + return -EINTR; + + switch (hdev->req_status) { + case HCI_REQ_DONE: + err = -bt_to_errno(hdev->req_result); + if (hci_status) + *hci_status = hdev->req_result; + break; + + case HCI_REQ_CANCELED: + err = -hdev->req_result; + if (hci_status) + *hci_status = HCI_ERROR_UNSPECIFIED; + break; + + default: + err = -ETIMEDOUT; + if (hci_status) + *hci_status = HCI_ERROR_UNSPECIFIED; + break; + } + + hdev->req_status = hdev->req_result = 0; + + BT_DBG("%s end: err %d", hdev->name, err); + + return err; +} + +int hci_req_sync(struct hci_dev *hdev, int (*req)(struct hci_request *req, + unsigned long opt), + unsigned long opt, u32 timeout, u8 *hci_status) +{ + int ret; + + if (!test_bit(HCI_UP, &hdev->flags)) + return -ENETDOWN; + + /* Serialize all requests */ + hci_req_sync_lock(hdev); + ret = __hci_req_sync(hdev, req, opt, timeout, hci_status); + hci_req_sync_unlock(hdev); + + return ret; +} + struct sk_buff *hci_prepare_cmd(struct hci_dev *hdev, u16 opcode, u32 plen, const void *param) { @@ -98,8 +303,8 @@ struct sk_buff *hci_prepare_cmd(struct hci_dev *hdev, u16 opcode, u32 plen, BT_DBG("skb len %d", skb->len); - bt_cb(skb)->pkt_type = HCI_COMMAND_PKT; - bt_cb(skb)->hci.opcode = opcode; + hci_skb_pkt_type(skb) = HCI_COMMAND_PKT; + hci_skb_opcode(skb) = opcode; return skb; } @@ -128,7 +333,7 @@ void hci_req_add_ev(struct hci_request *req, u16 opcode, u32 plen, } if (skb_queue_empty(&req->cmd_q)) - bt_cb(skb)->hci.req_start = true; + bt_cb(skb)->hci.req_flags |= HCI_REQ_START; bt_cb(skb)->hci.req_event = event; @@ -476,7 +681,7 @@ void hci_update_page_scan(struct hci_dev *hdev) * * This function requires the caller holds hdev->lock. */ -void __hci_update_background_scan(struct hci_request *req) +static void __hci_update_background_scan(struct hci_request *req) { struct hci_dev *hdev = req->hdev; @@ -543,28 +748,6 @@ void __hci_update_background_scan(struct hci_request *req) } } -static void update_background_scan_complete(struct hci_dev *hdev, u8 status, - u16 opcode) -{ - if (status) - BT_DBG("HCI request failed to update background scanning: " - "status 0x%2.2x", status); -} - -void hci_update_background_scan(struct hci_dev *hdev) -{ - int err; - struct hci_request req; - - hci_req_init(&req, hdev); - - __hci_update_background_scan(&req); - - err = hci_req_run(&req, update_background_scan_complete); - if (err && err != -ENODATA) - BT_ERR("Failed to run HCI request: err %d", err); -} - void __hci_abort_conn(struct hci_request *req, struct hci_conn *conn, u8 reason) { @@ -657,3 +840,446 @@ int hci_abort_conn(struct hci_conn *conn, u8 reason) return 0; } + +static int update_bg_scan(struct hci_request *req, unsigned long opt) +{ + hci_dev_lock(req->hdev); + __hci_update_background_scan(req); + hci_dev_unlock(req->hdev); + return 0; +} + +static void bg_scan_update(struct work_struct *work) +{ + struct hci_dev *hdev = container_of(work, struct hci_dev, + bg_scan_update); + struct hci_conn *conn; + u8 status; + int err; + + err = hci_req_sync(hdev, update_bg_scan, 0, HCI_CMD_TIMEOUT, &status); + if (!err) + return; + + hci_dev_lock(hdev); + + conn = hci_conn_hash_lookup_state(hdev, LE_LINK, BT_CONNECT); + if (conn) + hci_le_conn_failed(conn, status); + + hci_dev_unlock(hdev); +} + +static int le_scan_disable(struct hci_request *req, unsigned long opt) +{ + hci_req_add_le_scan_disable(req); + return 0; +} + +static int bredr_inquiry(struct hci_request *req, unsigned long opt) +{ + u8 length = opt; + /* General inquiry access code (GIAC) */ + u8 lap[3] = { 0x33, 0x8b, 0x9e }; + struct hci_cp_inquiry cp; + + BT_DBG("%s", req->hdev->name); + + hci_dev_lock(req->hdev); + hci_inquiry_cache_flush(req->hdev); + hci_dev_unlock(req->hdev); + + memset(&cp, 0, sizeof(cp)); + memcpy(&cp.lap, lap, sizeof(cp.lap)); + cp.length = length; + + hci_req_add(req, HCI_OP_INQUIRY, sizeof(cp), &cp); + + return 0; +} + +static void le_scan_disable_work(struct work_struct *work) +{ + struct hci_dev *hdev = container_of(work, struct hci_dev, + le_scan_disable.work); + u8 status; + + BT_DBG("%s", hdev->name); + + if (!hci_dev_test_flag(hdev, HCI_LE_SCAN)) + return; + + cancel_delayed_work(&hdev->le_scan_restart); + + hci_req_sync(hdev, le_scan_disable, 0, HCI_CMD_TIMEOUT, &status); + if (status) { + BT_ERR("Failed to disable LE scan: status 0x%02x", status); + return; + } + + hdev->discovery.scan_start = 0; + + /* If we were running LE only scan, change discovery state. If + * we were running both LE and BR/EDR inquiry simultaneously, + * and BR/EDR inquiry is already finished, stop discovery, + * otherwise BR/EDR inquiry will stop discovery when finished. + * If we will resolve remote device name, do not change + * discovery state. + */ + + if (hdev->discovery.type == DISCOV_TYPE_LE) + goto discov_stopped; + + if (hdev->discovery.type != DISCOV_TYPE_INTERLEAVED) + return; + + if (test_bit(HCI_QUIRK_SIMULTANEOUS_DISCOVERY, &hdev->quirks)) { + if (!test_bit(HCI_INQUIRY, &hdev->flags) && + hdev->discovery.state != DISCOVERY_RESOLVING) + goto discov_stopped; + + return; + } + + hci_req_sync(hdev, bredr_inquiry, DISCOV_INTERLEAVED_INQUIRY_LEN, + HCI_CMD_TIMEOUT, &status); + if (status) { + BT_ERR("Inquiry failed: status 0x%02x", status); + goto discov_stopped; + } + + return; + +discov_stopped: + hci_dev_lock(hdev); + hci_discovery_set_state(hdev, DISCOVERY_STOPPED); + hci_dev_unlock(hdev); +} + +static int le_scan_restart(struct hci_request *req, unsigned long opt) +{ + struct hci_dev *hdev = req->hdev; + struct hci_cp_le_set_scan_enable cp; + + /* If controller is not scanning we are done. */ + if (!hci_dev_test_flag(hdev, HCI_LE_SCAN)) + return 0; + + hci_req_add_le_scan_disable(req); + + memset(&cp, 0, sizeof(cp)); + cp.enable = LE_SCAN_ENABLE; + cp.filter_dup = LE_SCAN_FILTER_DUP_ENABLE; + hci_req_add(req, HCI_OP_LE_SET_SCAN_ENABLE, sizeof(cp), &cp); + + return 0; +} + +static void le_scan_restart_work(struct work_struct *work) +{ + struct hci_dev *hdev = container_of(work, struct hci_dev, + le_scan_restart.work); + unsigned long timeout, duration, scan_start, now; + u8 status; + + BT_DBG("%s", hdev->name); + + hci_req_sync(hdev, le_scan_restart, 0, HCI_CMD_TIMEOUT, &status); + if (status) { + BT_ERR("Failed to restart LE scan: status %d", status); + return; + } + + hci_dev_lock(hdev); + + if (!test_bit(HCI_QUIRK_STRICT_DUPLICATE_FILTER, &hdev->quirks) || + !hdev->discovery.scan_start) + goto unlock; + + /* When the scan was started, hdev->le_scan_disable has been queued + * after duration from scan_start. During scan restart this job + * has been canceled, and we need to queue it again after proper + * timeout, to make sure that scan does not run indefinitely. + */ + duration = hdev->discovery.scan_duration; + scan_start = hdev->discovery.scan_start; + now = jiffies; + if (now - scan_start <= duration) { + int elapsed; + + if (now >= scan_start) + elapsed = now - scan_start; + else + elapsed = ULONG_MAX - scan_start + now; + + timeout = duration - elapsed; + } else { + timeout = 0; + } + + queue_delayed_work(hdev->req_workqueue, + &hdev->le_scan_disable, timeout); + +unlock: + hci_dev_unlock(hdev); +} + +static void cancel_adv_timeout(struct hci_dev *hdev) +{ + if (hdev->adv_instance_timeout) { + hdev->adv_instance_timeout = 0; + cancel_delayed_work(&hdev->adv_instance_expire); + } +} + +static void disable_advertising(struct hci_request *req) +{ + u8 enable = 0x00; + + hci_req_add(req, HCI_OP_LE_SET_ADV_ENABLE, sizeof(enable), &enable); +} + +static int active_scan(struct hci_request *req, unsigned long opt) +{ + uint16_t interval = opt; + struct hci_dev *hdev = req->hdev; + struct hci_cp_le_set_scan_param param_cp; + struct hci_cp_le_set_scan_enable enable_cp; + u8 own_addr_type; + int err; + + BT_DBG("%s", hdev->name); + + if (hci_dev_test_flag(hdev, HCI_LE_ADV)) { + hci_dev_lock(hdev); + + /* Don't let discovery abort an outgoing connection attempt + * that's using directed advertising. + */ + if (hci_lookup_le_connect(hdev)) { + hci_dev_unlock(hdev); + return -EBUSY; + } + + cancel_adv_timeout(hdev); + hci_dev_unlock(hdev); + + disable_advertising(req); + } + + /* If controller is scanning, it means the background scanning is + * running. Thus, we should temporarily stop it in order to set the + * discovery scanning parameters. + */ + if (hci_dev_test_flag(hdev, HCI_LE_SCAN)) + hci_req_add_le_scan_disable(req); + + /* All active scans will be done with either a resolvable private + * address (when privacy feature has been enabled) or non-resolvable + * private address. + */ + err = hci_update_random_address(req, true, &own_addr_type); + if (err < 0) + own_addr_type = ADDR_LE_DEV_PUBLIC; + + memset(¶m_cp, 0, sizeof(param_cp)); + param_cp.type = LE_SCAN_ACTIVE; + param_cp.interval = cpu_to_le16(interval); + param_cp.window = cpu_to_le16(DISCOV_LE_SCAN_WIN); + param_cp.own_address_type = own_addr_type; + + hci_req_add(req, HCI_OP_LE_SET_SCAN_PARAM, sizeof(param_cp), + ¶m_cp); + + memset(&enable_cp, 0, sizeof(enable_cp)); + enable_cp.enable = LE_SCAN_ENABLE; + enable_cp.filter_dup = LE_SCAN_FILTER_DUP_ENABLE; + + hci_req_add(req, HCI_OP_LE_SET_SCAN_ENABLE, sizeof(enable_cp), + &enable_cp); + + return 0; +} + +static int interleaved_discov(struct hci_request *req, unsigned long opt) +{ + int err; + + BT_DBG("%s", req->hdev->name); + + err = active_scan(req, opt); + if (err) + return err; + + return bredr_inquiry(req, DISCOV_BREDR_INQUIRY_LEN); +} + +static void start_discovery(struct hci_dev *hdev, u8 *status) +{ + unsigned long timeout; + + BT_DBG("%s type %u", hdev->name, hdev->discovery.type); + + switch (hdev->discovery.type) { + case DISCOV_TYPE_BREDR: + if (!hci_dev_test_flag(hdev, HCI_INQUIRY)) + hci_req_sync(hdev, bredr_inquiry, + DISCOV_BREDR_INQUIRY_LEN, HCI_CMD_TIMEOUT, + status); + return; + case DISCOV_TYPE_INTERLEAVED: + /* When running simultaneous discovery, the LE scanning time + * should occupy the whole discovery time sine BR/EDR inquiry + * and LE scanning are scheduled by the controller. + * + * For interleaving discovery in comparison, BR/EDR inquiry + * and LE scanning are done sequentially with separate + * timeouts. + */ + if (test_bit(HCI_QUIRK_SIMULTANEOUS_DISCOVERY, + &hdev->quirks)) { + timeout = msecs_to_jiffies(DISCOV_LE_TIMEOUT); + /* During simultaneous discovery, we double LE scan + * interval. We must leave some time for the controller + * to do BR/EDR inquiry. + */ + hci_req_sync(hdev, interleaved_discov, + DISCOV_LE_SCAN_INT * 2, HCI_CMD_TIMEOUT, + status); + break; + } + + timeout = msecs_to_jiffies(hdev->discov_interleaved_timeout); + hci_req_sync(hdev, active_scan, DISCOV_LE_SCAN_INT, + HCI_CMD_TIMEOUT, status); + break; + case DISCOV_TYPE_LE: + timeout = msecs_to_jiffies(DISCOV_LE_TIMEOUT); + hci_req_sync(hdev, active_scan, DISCOV_LE_SCAN_INT, + HCI_CMD_TIMEOUT, status); + break; + default: + *status = HCI_ERROR_UNSPECIFIED; + return; + } + + if (*status) + return; + + BT_DBG("%s timeout %u ms", hdev->name, jiffies_to_msecs(timeout)); + + /* When service discovery is used and the controller has a + * strict duplicate filter, it is important to remember the + * start and duration of the scan. This is required for + * restarting scanning during the discovery phase. + */ + if (test_bit(HCI_QUIRK_STRICT_DUPLICATE_FILTER, &hdev->quirks) && + hdev->discovery.result_filtering) { + hdev->discovery.scan_start = jiffies; + hdev->discovery.scan_duration = timeout; + } + + queue_delayed_work(hdev->req_workqueue, &hdev->le_scan_disable, + timeout); +} + +bool hci_req_stop_discovery(struct hci_request *req) +{ + struct hci_dev *hdev = req->hdev; + struct discovery_state *d = &hdev->discovery; + struct hci_cp_remote_name_req_cancel cp; + struct inquiry_entry *e; + bool ret = false; + + BT_DBG("%s state %u", hdev->name, hdev->discovery.state); + + if (d->state == DISCOVERY_FINDING || d->state == DISCOVERY_STOPPING) { + if (test_bit(HCI_INQUIRY, &hdev->flags)) + hci_req_add(req, HCI_OP_INQUIRY_CANCEL, 0, NULL); + + if (hci_dev_test_flag(hdev, HCI_LE_SCAN)) { + cancel_delayed_work(&hdev->le_scan_disable); + hci_req_add_le_scan_disable(req); + } + + ret = true; + } else { + /* Passive scanning */ + if (hci_dev_test_flag(hdev, HCI_LE_SCAN)) { + hci_req_add_le_scan_disable(req); + ret = true; + } + } + + /* No further actions needed for LE-only discovery */ + if (d->type == DISCOV_TYPE_LE) + return ret; + + if (d->state == DISCOVERY_RESOLVING || d->state == DISCOVERY_STOPPING) { + e = hci_inquiry_cache_lookup_resolve(hdev, BDADDR_ANY, + NAME_PENDING); + if (!e) + return ret; + + bacpy(&cp.bdaddr, &e->data.bdaddr); + hci_req_add(req, HCI_OP_REMOTE_NAME_REQ_CANCEL, sizeof(cp), + &cp); + ret = true; + } + + return ret; +} + +static int stop_discovery(struct hci_request *req, unsigned long opt) +{ + hci_dev_lock(req->hdev); + hci_req_stop_discovery(req); + hci_dev_unlock(req->hdev); + + return 0; +} + +static void discov_update(struct work_struct *work) +{ + struct hci_dev *hdev = container_of(work, struct hci_dev, + discov_update); + u8 status = 0; + + switch (hdev->discovery.state) { + case DISCOVERY_STARTING: + start_discovery(hdev, &status); + mgmt_start_discovery_complete(hdev, status); + if (status) + hci_discovery_set_state(hdev, DISCOVERY_STOPPED); + else + hci_discovery_set_state(hdev, DISCOVERY_FINDING); + break; + case DISCOVERY_STOPPING: + hci_req_sync(hdev, stop_discovery, 0, HCI_CMD_TIMEOUT, &status); + mgmt_stop_discovery_complete(hdev, status); + if (!status) + hci_discovery_set_state(hdev, DISCOVERY_STOPPED); + break; + case DISCOVERY_STOPPED: + default: + return; + } +} + +void hci_request_setup(struct hci_dev *hdev) +{ + INIT_WORK(&hdev->discov_update, discov_update); + INIT_WORK(&hdev->bg_scan_update, bg_scan_update); + INIT_DELAYED_WORK(&hdev->le_scan_disable, le_scan_disable_work); + INIT_DELAYED_WORK(&hdev->le_scan_restart, le_scan_restart_work); +} + +void hci_request_cancel_all(struct hci_dev *hdev) +{ + hci_req_sync_cancel(hdev, ENODEV); + + cancel_work_sync(&hdev->discov_update); + cancel_work_sync(&hdev->bg_scan_update); + cancel_delayed_work_sync(&hdev->le_scan_disable); + cancel_delayed_work_sync(&hdev->le_scan_restart); +} diff --git a/net/bluetooth/hci_request.h b/net/bluetooth/hci_request.h index 25c7f13..6b9e59f 100644 --- a/net/bluetooth/hci_request.h +++ b/net/bluetooth/hci_request.h @@ -20,6 +20,9 @@ SOFTWARE IS DISCLAIMED. */ +#define hci_req_sync_lock(hdev) mutex_lock(&hdev->req_lock) +#define hci_req_sync_unlock(hdev) mutex_unlock(&hdev->req_lock) + struct hci_request { struct hci_dev *hdev; struct sk_buff_head cmd_q; @@ -41,21 +44,37 @@ void hci_req_cmd_complete(struct hci_dev *hdev, u16 opcode, u8 status, hci_req_complete_t *req_complete, hci_req_complete_skb_t *req_complete_skb); +int hci_req_sync(struct hci_dev *hdev, int (*req)(struct hci_request *req, + unsigned long opt), + unsigned long opt, u32 timeout, u8 *hci_status); +int __hci_req_sync(struct hci_dev *hdev, int (*func)(struct hci_request *req, + unsigned long opt), + unsigned long opt, u32 timeout, u8 *hci_status); +void hci_req_sync_cancel(struct hci_dev *hdev, int err); + struct sk_buff *hci_prepare_cmd(struct hci_dev *hdev, u16 opcode, u32 plen, const void *param); void hci_req_add_le_scan_disable(struct hci_request *req); void hci_req_add_le_passive_scan(struct hci_request *req); +/* Returns true if HCI commands were queued */ +bool hci_req_stop_discovery(struct hci_request *req); + void hci_update_page_scan(struct hci_dev *hdev); void __hci_update_page_scan(struct hci_request *req); int hci_update_random_address(struct hci_request *req, bool require_privacy, u8 *own_addr_type); -void hci_update_background_scan(struct hci_dev *hdev); -void __hci_update_background_scan(struct hci_request *req); - int hci_abort_conn(struct hci_conn *conn, u8 reason); void __hci_abort_conn(struct hci_request *req, struct hci_conn *conn, u8 reason); + +static inline void hci_update_background_scan(struct hci_dev *hdev) +{ + queue_work(hdev->req_workqueue, &hdev->bg_scan_update); +} + +void hci_request_setup(struct hci_dev *hdev); +void hci_request_cancel_all(struct hci_dev *hdev); diff --git a/net/bluetooth/hci_sock.c b/net/bluetooth/hci_sock.c index b1eb8c0..41f579b 100644 --- a/net/bluetooth/hci_sock.c +++ b/net/bluetooth/hci_sock.c @@ -26,6 +26,8 @@ #include <linux/export.h> #include <asm/unaligned.h> +#include <generated/compile.h> +#include <generated/utsrelease.h> #include <net/bluetooth/bluetooth.h> #include <net/bluetooth/hci_core.h> @@ -120,13 +122,13 @@ static bool is_filtered_packet(struct sock *sk, struct sk_buff *skb) /* Apply filter */ flt = &hci_pi(sk)->filter; - flt_type = bt_cb(skb)->pkt_type & HCI_FLT_TYPE_BITS; + flt_type = hci_skb_pkt_type(skb) & HCI_FLT_TYPE_BITS; if (!test_bit(flt_type, &flt->type_mask)) return true; /* Extra filter for event packets only */ - if (bt_cb(skb)->pkt_type != HCI_EVENT_PKT) + if (hci_skb_pkt_type(skb) != HCI_EVENT_PKT) return false; flt_event = (*(__u8 *)skb->data & HCI_FLT_EVENT_BITS); @@ -170,19 +172,19 @@ void hci_send_to_sock(struct hci_dev *hdev, struct sk_buff *skb) continue; if (hci_pi(sk)->channel == HCI_CHANNEL_RAW) { - if (bt_cb(skb)->pkt_type != HCI_COMMAND_PKT && - bt_cb(skb)->pkt_type != HCI_EVENT_PKT && - bt_cb(skb)->pkt_type != HCI_ACLDATA_PKT && - bt_cb(skb)->pkt_type != HCI_SCODATA_PKT) + if (hci_skb_pkt_type(skb) != HCI_COMMAND_PKT && + hci_skb_pkt_type(skb) != HCI_EVENT_PKT && + hci_skb_pkt_type(skb) != HCI_ACLDATA_PKT && + hci_skb_pkt_type(skb) != HCI_SCODATA_PKT) continue; if (is_filtered_packet(sk, skb)) continue; } else if (hci_pi(sk)->channel == HCI_CHANNEL_USER) { if (!bt_cb(skb)->incoming) continue; - if (bt_cb(skb)->pkt_type != HCI_EVENT_PKT && - bt_cb(skb)->pkt_type != HCI_ACLDATA_PKT && - bt_cb(skb)->pkt_type != HCI_SCODATA_PKT) + if (hci_skb_pkt_type(skb) != HCI_EVENT_PKT && + hci_skb_pkt_type(skb) != HCI_ACLDATA_PKT && + hci_skb_pkt_type(skb) != HCI_SCODATA_PKT) continue; } else { /* Don't send frame to other channel types */ @@ -196,7 +198,7 @@ void hci_send_to_sock(struct hci_dev *hdev, struct sk_buff *skb) continue; /* Put type byte before the data */ - memcpy(skb_push(skb_copy, 1), &bt_cb(skb)->pkt_type, 1); + memcpy(skb_push(skb_copy, 1), &hci_skb_pkt_type(skb), 1); } nskb = skb_clone(skb_copy, GFP_ATOMIC); @@ -262,7 +264,7 @@ void hci_send_to_monitor(struct hci_dev *hdev, struct sk_buff *skb) BT_DBG("hdev %p len %d", hdev, skb->len); - switch (bt_cb(skb)->pkt_type) { + switch (hci_skb_pkt_type(skb)) { case HCI_COMMAND_PKT: opcode = cpu_to_le16(HCI_MON_COMMAND_PKT); break; @@ -294,7 +296,7 @@ void hci_send_to_monitor(struct hci_dev *hdev, struct sk_buff *skb) return; /* Put header before the data */ - hdr = (void *) skb_push(skb_copy, HCI_MON_HDR_SIZE); + hdr = (void *)skb_push(skb_copy, HCI_MON_HDR_SIZE); hdr->opcode = opcode; hdr->index = cpu_to_le16(hdev->id); hdr->len = cpu_to_le16(skb->len); @@ -375,7 +377,7 @@ static struct sk_buff *create_monitor_event(struct hci_dev *hdev, int event) __net_timestamp(skb); - hdr = (void *) skb_push(skb, HCI_MON_HDR_SIZE); + hdr = (void *)skb_push(skb, HCI_MON_HDR_SIZE); hdr->opcode = opcode; hdr->index = cpu_to_le16(hdev->id); hdr->len = cpu_to_le16(skb->len - HCI_MON_HDR_SIZE); @@ -383,6 +385,29 @@ static struct sk_buff *create_monitor_event(struct hci_dev *hdev, int event) return skb; } +static void send_monitor_note(struct sock *sk, const char *text) +{ + size_t len = strlen(text); + struct hci_mon_hdr *hdr; + struct sk_buff *skb; + + skb = bt_skb_alloc(len + 1, GFP_ATOMIC); + if (!skb) + return; + + strcpy(skb_put(skb, len + 1), text); + + __net_timestamp(skb); + + hdr = (void *)skb_push(skb, HCI_MON_HDR_SIZE); + hdr->opcode = cpu_to_le16(HCI_MON_SYSTEM_NOTE); + hdr->index = cpu_to_le16(HCI_DEV_NONE); + hdr->len = cpu_to_le16(skb->len - HCI_MON_HDR_SIZE); + + if (sock_queue_rcv_skb(sk, skb)) + kfree_skb(skb); +} + static void send_monitor_replay(struct sock *sk) { struct hci_dev *hdev; @@ -436,18 +461,18 @@ static void hci_si_event(struct hci_dev *hdev, int type, int dlen, void *data) if (!skb) return; - hdr = (void *) skb_put(skb, HCI_EVENT_HDR_SIZE); + hdr = (void *)skb_put(skb, HCI_EVENT_HDR_SIZE); hdr->evt = HCI_EV_STACK_INTERNAL; hdr->plen = sizeof(*ev) + dlen; - ev = (void *) skb_put(skb, sizeof(*ev) + dlen); + ev = (void *)skb_put(skb, sizeof(*ev) + dlen); ev->type = type; memcpy(ev->data, data, dlen); bt_cb(skb)->incoming = 1; __net_timestamp(skb); - bt_cb(skb)->pkt_type = HCI_EVENT_PKT; + hci_skb_pkt_type(skb) = HCI_EVENT_PKT; hci_send_to_sock(hdev, skb); kfree_skb(skb); } @@ -653,20 +678,20 @@ static int hci_sock_bound_ioctl(struct sock *sk, unsigned int cmd, return -EOPNOTSUPP; case HCIGETCONNINFO: - return hci_get_conn_info(hdev, (void __user *) arg); + return hci_get_conn_info(hdev, (void __user *)arg); case HCIGETAUTHINFO: - return hci_get_auth_info(hdev, (void __user *) arg); + return hci_get_auth_info(hdev, (void __user *)arg); case HCIBLOCKADDR: if (!capable(CAP_NET_ADMIN)) return -EPERM; - return hci_sock_blacklist_add(hdev, (void __user *) arg); + return hci_sock_blacklist_add(hdev, (void __user *)arg); case HCIUNBLOCKADDR: if (!capable(CAP_NET_ADMIN)) return -EPERM; - return hci_sock_blacklist_del(hdev, (void __user *) arg); + return hci_sock_blacklist_del(hdev, (void __user *)arg); } return -ENOIOCTLCMD; @@ -675,7 +700,7 @@ static int hci_sock_bound_ioctl(struct sock *sk, unsigned int cmd, static int hci_sock_ioctl(struct socket *sock, unsigned int cmd, unsigned long arg) { - void __user *argp = (void __user *) arg; + void __user *argp = (void __user *)arg; struct sock *sk = sock->sk; int err; @@ -872,11 +897,27 @@ static int hci_sock_bind(struct socket *sock, struct sockaddr *addr, */ hci_sock_set_flag(sk, HCI_SOCK_TRUSTED); + send_monitor_note(sk, "Linux version " UTS_RELEASE + " (" UTS_MACHINE ")"); + send_monitor_note(sk, "Bluetooth subsystem version " + BT_SUBSYS_VERSION); send_monitor_replay(sk); atomic_inc(&monitor_promisc); break; + case HCI_CHANNEL_LOGGING: + if (haddr.hci_dev != HCI_DEV_NONE) { + err = -EINVAL; + goto done; + } + + if (!capable(CAP_NET_ADMIN)) { + err = -EPERM; + goto done; + } + break; + default: if (!hci_mgmt_chan_find(haddr.hci_channel)) { err = -EINVAL; @@ -926,7 +967,7 @@ done: static int hci_sock_getname(struct socket *sock, struct sockaddr *addr, int *addr_len, int peer) { - struct sockaddr_hci *haddr = (struct sockaddr_hci *) addr; + struct sockaddr_hci *haddr = (struct sockaddr_hci *)addr; struct sock *sk = sock->sk; struct hci_dev *hdev; int err = 0; @@ -991,8 +1032,8 @@ static void hci_sock_cmsg(struct sock *sk, struct msghdr *msg, } } -static int hci_sock_recvmsg(struct socket *sock, struct msghdr *msg, size_t len, - int flags) +static int hci_sock_recvmsg(struct socket *sock, struct msghdr *msg, + size_t len, int flags) { int noblock = flags & MSG_DONTWAIT; struct sock *sk = sock->sk; @@ -1004,6 +1045,9 @@ static int hci_sock_recvmsg(struct socket *sock, struct msghdr *msg, size_t len, if (flags & MSG_OOB) return -EOPNOTSUPP; + if (hci_pi(sk)->channel == HCI_CHANNEL_LOGGING) + return -EOPNOTSUPP; + if (sk->sk_state == BT_CLOSED) return 0; @@ -1150,6 +1194,90 @@ done: return err; } +static int hci_logging_frame(struct sock *sk, struct msghdr *msg, int len) +{ + struct hci_mon_hdr *hdr; + struct sk_buff *skb; + struct hci_dev *hdev; + u16 index; + int err; + + /* The logging frame consists at minimum of the standard header, + * the priority byte, the ident length byte and at least one string + * terminator NUL byte. Anything shorter are invalid packets. + */ + if (len < sizeof(*hdr) + 3) + return -EINVAL; + + skb = bt_skb_send_alloc(sk, len, msg->msg_flags & MSG_DONTWAIT, &err); + if (!skb) + return err; + + if (memcpy_from_msg(skb_put(skb, len), msg, len)) { + err = -EFAULT; + goto drop; + } + + hdr = (void *)skb->data; + + if (__le16_to_cpu(hdr->len) != len - sizeof(*hdr)) { + err = -EINVAL; + goto drop; + } + + if (__le16_to_cpu(hdr->opcode) == 0x0000) { + __u8 priority = skb->data[sizeof(*hdr)]; + __u8 ident_len = skb->data[sizeof(*hdr) + 1]; + + /* Only the priorities 0-7 are valid and with that any other + * value results in an invalid packet. + * + * The priority byte is followed by an ident length byte and + * the NUL terminated ident string. Check that the ident + * length is not overflowing the packet and also that the + * ident string itself is NUL terminated. In case the ident + * length is zero, the length value actually doubles as NUL + * terminator identifier. + * + * The message follows the ident string (if present) and + * must be NUL terminated. Otherwise it is not a valid packet. + */ + if (priority > 7 || skb->data[len - 1] != 0x00 || + ident_len > len - sizeof(*hdr) - 3 || + skb->data[sizeof(*hdr) + ident_len + 1] != 0x00) { + err = -EINVAL; + goto drop; + } + } else { + err = -EINVAL; + goto drop; + } + + index = __le16_to_cpu(hdr->index); + + if (index != MGMT_INDEX_NONE) { + hdev = hci_dev_get(index); + if (!hdev) { + err = -ENODEV; + goto drop; + } + } else { + hdev = NULL; + } + + hdr->opcode = cpu_to_le16(HCI_MON_USER_LOGGING); + + hci_send_to_channel(HCI_CHANNEL_MONITOR, skb, HCI_SOCK_TRUSTED, NULL); + err = len; + + if (hdev) + hci_dev_put(hdev); + +drop: + kfree_skb(skb); + return err; +} + static int hci_sock_sendmsg(struct socket *sock, struct msghdr *msg, size_t len) { @@ -1179,6 +1307,9 @@ static int hci_sock_sendmsg(struct socket *sock, struct msghdr *msg, case HCI_CHANNEL_MONITOR: err = -EOPNOTSUPP; goto done; + case HCI_CHANNEL_LOGGING: + err = hci_logging_frame(sk, msg, len); + goto done; default: mutex_lock(&mgmt_chan_list_lock); chan = __hci_mgmt_chan_find(hci_pi(sk)->channel); @@ -1211,7 +1342,7 @@ static int hci_sock_sendmsg(struct socket *sock, struct msghdr *msg, goto drop; } - bt_cb(skb)->pkt_type = *((unsigned char *) skb->data); + hci_skb_pkt_type(skb) = skb->data[0]; skb_pull(skb, 1); if (hci_pi(sk)->channel == HCI_CHANNEL_USER) { @@ -1220,16 +1351,16 @@ static int hci_sock_sendmsg(struct socket *sock, struct msghdr *msg, * * However check that the packet type is valid. */ - if (bt_cb(skb)->pkt_type != HCI_COMMAND_PKT && - bt_cb(skb)->pkt_type != HCI_ACLDATA_PKT && - bt_cb(skb)->pkt_type != HCI_SCODATA_PKT) { + if (hci_skb_pkt_type(skb) != HCI_COMMAND_PKT && + hci_skb_pkt_type(skb) != HCI_ACLDATA_PKT && + hci_skb_pkt_type(skb) != HCI_SCODATA_PKT) { err = -EINVAL; goto drop; } skb_queue_tail(&hdev->raw_q, skb); queue_work(hdev->workqueue, &hdev->tx_work); - } else if (bt_cb(skb)->pkt_type == HCI_COMMAND_PKT) { + } else if (hci_skb_pkt_type(skb) == HCI_COMMAND_PKT) { u16 opcode = get_unaligned_le16(skb->data); u16 ogf = hci_opcode_ogf(opcode); u16 ocf = hci_opcode_ocf(opcode); @@ -1242,6 +1373,11 @@ static int hci_sock_sendmsg(struct socket *sock, struct msghdr *msg, goto drop; } + /* Since the opcode has already been extracted here, store + * a copy of the value for later use by the drivers. + */ + hci_skb_opcode(skb) = opcode; + if (ogf == 0x3f) { skb_queue_tail(&hdev->raw_q, skb); queue_work(hdev->workqueue, &hdev->tx_work); @@ -1249,7 +1385,7 @@ static int hci_sock_sendmsg(struct socket *sock, struct msghdr *msg, /* Stand-alone HCI commands must be flagged as * single-command requests. */ - bt_cb(skb)->hci.req_start = true; + bt_cb(skb)->hci.req_flags |= HCI_REQ_START; skb_queue_tail(&hdev->cmd_q, skb); queue_work(hdev->workqueue, &hdev->cmd_work); @@ -1260,8 +1396,8 @@ static int hci_sock_sendmsg(struct socket *sock, struct msghdr *msg, goto drop; } - if (bt_cb(skb)->pkt_type != HCI_ACLDATA_PKT && - bt_cb(skb)->pkt_type != HCI_SCODATA_PKT) { + if (hci_skb_pkt_type(skb) != HCI_ACLDATA_PKT && + hci_skb_pkt_type(skb) != HCI_SCODATA_PKT) { err = -EINVAL; goto drop; } diff --git a/net/bluetooth/l2cap_core.c b/net/bluetooth/l2cap_core.c index 66e8b6e..39a5149 100644 --- a/net/bluetooth/l2cap_core.c +++ b/net/bluetooth/l2cap_core.c @@ -6538,8 +6538,6 @@ static int l2cap_rx(struct l2cap_chan *chan, struct l2cap_ctrl *control, static int l2cap_stream_rx(struct l2cap_chan *chan, struct l2cap_ctrl *control, struct sk_buff *skb) { - int err = 0; - BT_DBG("chan %p, control %p, skb %p, state %d", chan, control, skb, chan->rx_state); @@ -6570,7 +6568,7 @@ static int l2cap_stream_rx(struct l2cap_chan *chan, struct l2cap_ctrl *control, chan->last_acked_seq = control->txseq; chan->expected_tx_seq = __next_seq(chan, control->txseq); - return err; + return 0; } static int l2cap_data_rcv(struct l2cap_chan *chan, struct sk_buff *skb) @@ -7113,8 +7111,6 @@ int l2cap_chan_connect(struct l2cap_chan *chan, __le16 psm, u16 cid, chan->dcid = cid; if (bdaddr_type_is_le(dst_type)) { - u8 role; - /* Convert from L2CAP channel address type to HCI address type */ if (dst_type == BDADDR_LE_PUBLIC) @@ -7123,14 +7119,15 @@ int l2cap_chan_connect(struct l2cap_chan *chan, __le16 psm, u16 cid, dst_type = ADDR_LE_DEV_RANDOM; if (hci_dev_test_flag(hdev, HCI_ADVERTISING)) - role = HCI_ROLE_SLAVE; + hcon = hci_connect_le(hdev, dst, dst_type, + chan->sec_level, + HCI_LE_CONN_TIMEOUT, + HCI_ROLE_SLAVE); else - role = HCI_ROLE_MASTER; + hcon = hci_connect_le_scan(hdev, dst, dst_type, + chan->sec_level, + HCI_LE_CONN_TIMEOUT); - hcon = hci_connect_le_scan(hdev, dst, dst_type, - chan->sec_level, - HCI_LE_CONN_TIMEOUT, - role); } else { u8 auth_type = l2cap_get_auth_type(chan); hcon = hci_connect_acl(hdev, dst, chan->sec_level, auth_type); diff --git a/net/bluetooth/mgmt.c b/net/bluetooth/mgmt.c index 7f22119..3d9d2e4 100644 --- a/net/bluetooth/mgmt.c +++ b/net/bluetooth/mgmt.c @@ -38,7 +38,7 @@ #include "mgmt_util.h" #define MGMT_VERSION 1 -#define MGMT_REVISION 10 +#define MGMT_REVISION 11 static const u16 mgmt_commands[] = { MGMT_OP_READ_INDEX_LIST, @@ -102,6 +102,7 @@ static const u16 mgmt_commands[] = { MGMT_OP_READ_ADV_FEATURES, MGMT_OP_ADD_ADVERTISING, MGMT_OP_REMOVE_ADVERTISING, + MGMT_OP_GET_ADV_SIZE_INFO, }; static const u16 mgmt_events[] = { @@ -1416,49 +1417,6 @@ static void clean_up_hci_complete(struct hci_dev *hdev, u8 status, u16 opcode) } } -static bool hci_stop_discovery(struct hci_request *req) -{ - struct hci_dev *hdev = req->hdev; - struct hci_cp_remote_name_req_cancel cp; - struct inquiry_entry *e; - - switch (hdev->discovery.state) { - case DISCOVERY_FINDING: - if (test_bit(HCI_INQUIRY, &hdev->flags)) - hci_req_add(req, HCI_OP_INQUIRY_CANCEL, 0, NULL); - - if (hci_dev_test_flag(hdev, HCI_LE_SCAN)) { - cancel_delayed_work(&hdev->le_scan_disable); - hci_req_add_le_scan_disable(req); - } - - return true; - - case DISCOVERY_RESOLVING: - e = hci_inquiry_cache_lookup_resolve(hdev, BDADDR_ANY, - NAME_PENDING); - if (!e) - break; - - bacpy(&cp.bdaddr, &e->data.bdaddr); - hci_req_add(req, HCI_OP_REMOTE_NAME_REQ_CANCEL, sizeof(cp), - &cp); - - return true; - - default: - /* Passive scanning */ - if (hci_dev_test_flag(hdev, HCI_LE_SCAN)) { - hci_req_add_le_scan_disable(req); - return true; - } - - break; - } - - return false; -} - static void advertising_added(struct sock *sk, struct hci_dev *hdev, u8 instance) { @@ -1636,7 +1594,7 @@ static int clean_up_hci_state(struct hci_dev *hdev) if (hci_dev_test_flag(hdev, HCI_LE_ADV)) disable_advertising(&req); - discov_stopped = hci_stop_discovery(&req); + discov_stopped = hci_req_stop_discovery(&req); list_for_each_entry(conn, &hdev->conn_hash.list, list) { /* 0x15 == Terminated due to Power Off */ @@ -2510,8 +2468,8 @@ static void le_enable_complete(struct hci_dev *hdev, u8 status, u16 opcode) hci_req_init(&req, hdev); update_adv_data(&req); update_scan_rsp_data(&req); - __hci_update_background_scan(&req); hci_req_run(&req, NULL); + hci_update_background_scan(hdev); } unlock: @@ -3561,8 +3519,7 @@ static int pair_device(struct sock *sk, struct hci_dev *hdev, void *data, conn = hci_connect_le_scan(hdev, &cp->addr.bdaddr, addr_type, sec_level, - HCI_LE_CONN_TIMEOUT, - HCI_ROLE_MASTER); + HCI_LE_CONN_TIMEOUT); } if (IS_ERR(conn)) { @@ -4164,145 +4121,9 @@ done: return err; } -static bool trigger_bredr_inquiry(struct hci_request *req, u8 *status) -{ - struct hci_dev *hdev = req->hdev; - struct hci_cp_inquiry cp; - /* General inquiry access code (GIAC) */ - u8 lap[3] = { 0x33, 0x8b, 0x9e }; - - *status = mgmt_bredr_support(hdev); - if (*status) - return false; - - if (hci_dev_test_flag(hdev, HCI_INQUIRY)) { - *status = MGMT_STATUS_BUSY; - return false; - } - - hci_inquiry_cache_flush(hdev); - - memset(&cp, 0, sizeof(cp)); - memcpy(&cp.lap, lap, sizeof(cp.lap)); - cp.length = DISCOV_BREDR_INQUIRY_LEN; - - hci_req_add(req, HCI_OP_INQUIRY, sizeof(cp), &cp); - - return true; -} - -static bool trigger_le_scan(struct hci_request *req, u16 interval, u8 *status) -{ - struct hci_dev *hdev = req->hdev; - struct hci_cp_le_set_scan_param param_cp; - struct hci_cp_le_set_scan_enable enable_cp; - u8 own_addr_type; - int err; - - *status = mgmt_le_support(hdev); - if (*status) - return false; - - if (hci_dev_test_flag(hdev, HCI_LE_ADV)) { - /* Don't let discovery abort an outgoing connection attempt - * that's using directed advertising. - */ - if (hci_lookup_le_connect(hdev)) { - *status = MGMT_STATUS_REJECTED; - return false; - } - - cancel_adv_timeout(hdev); - disable_advertising(req); - } - - /* If controller is scanning, it means the background scanning is - * running. Thus, we should temporarily stop it in order to set the - * discovery scanning parameters. - */ - if (hci_dev_test_flag(hdev, HCI_LE_SCAN)) - hci_req_add_le_scan_disable(req); - - /* All active scans will be done with either a resolvable private - * address (when privacy feature has been enabled) or non-resolvable - * private address. - */ - err = hci_update_random_address(req, true, &own_addr_type); - if (err < 0) { - *status = MGMT_STATUS_FAILED; - return false; - } - - memset(¶m_cp, 0, sizeof(param_cp)); - param_cp.type = LE_SCAN_ACTIVE; - param_cp.interval = cpu_to_le16(interval); - param_cp.window = cpu_to_le16(DISCOV_LE_SCAN_WIN); - param_cp.own_address_type = own_addr_type; - - hci_req_add(req, HCI_OP_LE_SET_SCAN_PARAM, sizeof(param_cp), - ¶m_cp); - - memset(&enable_cp, 0, sizeof(enable_cp)); - enable_cp.enable = LE_SCAN_ENABLE; - enable_cp.filter_dup = LE_SCAN_FILTER_DUP_ENABLE; - - hci_req_add(req, HCI_OP_LE_SET_SCAN_ENABLE, sizeof(enable_cp), - &enable_cp); - - return true; -} - -static bool trigger_discovery(struct hci_request *req, u8 *status) -{ - struct hci_dev *hdev = req->hdev; - - switch (hdev->discovery.type) { - case DISCOV_TYPE_BREDR: - if (!trigger_bredr_inquiry(req, status)) - return false; - break; - - case DISCOV_TYPE_INTERLEAVED: - if (test_bit(HCI_QUIRK_SIMULTANEOUS_DISCOVERY, - &hdev->quirks)) { - /* During simultaneous discovery, we double LE scan - * interval. We must leave some time for the controller - * to do BR/EDR inquiry. - */ - if (!trigger_le_scan(req, DISCOV_LE_SCAN_INT * 2, - status)) - return false; - - if (!trigger_bredr_inquiry(req, status)) - return false; - - return true; - } - - if (!hci_dev_test_flag(hdev, HCI_BREDR_ENABLED)) { - *status = MGMT_STATUS_NOT_SUPPORTED; - return false; - } - /* fall through */ - - case DISCOV_TYPE_LE: - if (!trigger_le_scan(req, DISCOV_LE_SCAN_INT, status)) - return false; - break; - - default: - *status = MGMT_STATUS_INVALID_PARAMS; - return false; - } - - return true; -} - -static void start_discovery_complete(struct hci_dev *hdev, u8 status, - u16 opcode) +void mgmt_start_discovery_complete(struct hci_dev *hdev, u8 status) { struct mgmt_pending_cmd *cmd; - unsigned long timeout; BT_DBG("status %d", status); @@ -4317,62 +4138,34 @@ static void start_discovery_complete(struct hci_dev *hdev, u8 status, mgmt_pending_remove(cmd); } - if (status) { - hci_discovery_set_state(hdev, DISCOVERY_STOPPED); - goto unlock; - } - - hci_discovery_set_state(hdev, DISCOVERY_FINDING); + hci_dev_unlock(hdev); +} - /* If the scan involves LE scan, pick proper timeout to schedule - * hdev->le_scan_disable that will stop it. - */ - switch (hdev->discovery.type) { +static bool discovery_type_is_valid(struct hci_dev *hdev, uint8_t type, + uint8_t *mgmt_status) +{ + switch (type) { case DISCOV_TYPE_LE: - timeout = msecs_to_jiffies(DISCOV_LE_TIMEOUT); + *mgmt_status = mgmt_le_support(hdev); + if (*mgmt_status) + return false; break; case DISCOV_TYPE_INTERLEAVED: - /* When running simultaneous discovery, the LE scanning time - * should occupy the whole discovery time sine BR/EDR inquiry - * and LE scanning are scheduled by the controller. - * - * For interleaving discovery in comparison, BR/EDR inquiry - * and LE scanning are done sequentially with separate - * timeouts. - */ - if (test_bit(HCI_QUIRK_SIMULTANEOUS_DISCOVERY, &hdev->quirks)) - timeout = msecs_to_jiffies(DISCOV_LE_TIMEOUT); - else - timeout = msecs_to_jiffies(hdev->discov_interleaved_timeout); - break; + *mgmt_status = mgmt_le_support(hdev); + if (*mgmt_status) + return false; + /* Intentional fall-through */ case DISCOV_TYPE_BREDR: - timeout = 0; + *mgmt_status = mgmt_bredr_support(hdev); + if (*mgmt_status) + return false; break; default: - BT_ERR("Invalid discovery type %d", hdev->discovery.type); - timeout = 0; - break; - } - - if (timeout) { - /* When service discovery is used and the controller has - * a strict duplicate filter, it is important to remember - * the start and duration of the scan. This is required - * for restarting scanning during the discovery phase. - */ - if (test_bit(HCI_QUIRK_STRICT_DUPLICATE_FILTER, - &hdev->quirks) && - hdev->discovery.result_filtering) { - hdev->discovery.scan_start = jiffies; - hdev->discovery.scan_duration = timeout; - } - - queue_delayed_work(hdev->workqueue, - &hdev->le_scan_disable, timeout); + *mgmt_status = MGMT_STATUS_INVALID_PARAMS; + return false; } -unlock: - hci_dev_unlock(hdev); + return true; } static int start_discovery(struct sock *sk, struct hci_dev *hdev, @@ -4380,7 +4173,6 @@ static int start_discovery(struct sock *sk, struct hci_dev *hdev, { struct mgmt_cp_start_discovery *cp = data; struct mgmt_pending_cmd *cmd; - struct hci_request req; u8 status; int err; @@ -4403,14 +4195,12 @@ static int start_discovery(struct sock *sk, struct hci_dev *hdev, goto failed; } - cmd = mgmt_pending_add(sk, MGMT_OP_START_DISCOVERY, hdev, data, len); - if (!cmd) { - err = -ENOMEM; + if (!discovery_type_is_valid(hdev, cp->type, &status)) { + err = mgmt_cmd_complete(sk, hdev->id, MGMT_OP_START_DISCOVERY, + status, &cp->type, sizeof(cp->type)); goto failed; } - cmd->cmd_complete = generic_cmd_complete; - /* Clear the discovery filter first to free any previously * allocated memory for the UUID list. */ @@ -4419,22 +4209,17 @@ static int start_discovery(struct sock *sk, struct hci_dev *hdev, hdev->discovery.type = cp->type; hdev->discovery.report_invalid_rssi = false; - hci_req_init(&req, hdev); - - if (!trigger_discovery(&req, &status)) { - err = mgmt_cmd_complete(sk, hdev->id, MGMT_OP_START_DISCOVERY, - status, &cp->type, sizeof(cp->type)); - mgmt_pending_remove(cmd); + cmd = mgmt_pending_add(sk, MGMT_OP_START_DISCOVERY, hdev, data, len); + if (!cmd) { + err = -ENOMEM; goto failed; } - err = hci_req_run(&req, start_discovery_complete); - if (err < 0) { - mgmt_pending_remove(cmd); - goto failed; - } + cmd->cmd_complete = generic_cmd_complete; hci_discovery_set_state(hdev, DISCOVERY_STARTING); + queue_work(hdev->req_workqueue, &hdev->discov_update); + err = 0; failed: hci_dev_unlock(hdev); @@ -4453,7 +4238,6 @@ static int start_service_discovery(struct sock *sk, struct hci_dev *hdev, { struct mgmt_cp_start_service_discovery *cp = data; struct mgmt_pending_cmd *cmd; - struct hci_request req; const u16 max_uuid_count = ((U16_MAX - sizeof(*cp)) / 16); u16 uuid_count, expected_len; u8 status; @@ -4502,6 +4286,13 @@ static int start_service_discovery(struct sock *sk, struct hci_dev *hdev, goto failed; } + if (!discovery_type_is_valid(hdev, cp->type, &status)) { + err = mgmt_cmd_complete(sk, hdev->id, + MGMT_OP_START_SERVICE_DISCOVERY, + status, &cp->type, sizeof(cp->type)); + goto failed; + } + cmd = mgmt_pending_add(sk, MGMT_OP_START_SERVICE_DISCOVERY, hdev, data, len); if (!cmd) { @@ -4534,30 +4325,16 @@ static int start_service_discovery(struct sock *sk, struct hci_dev *hdev, } } - hci_req_init(&req, hdev); - - if (!trigger_discovery(&req, &status)) { - err = mgmt_cmd_complete(sk, hdev->id, - MGMT_OP_START_SERVICE_DISCOVERY, - status, &cp->type, sizeof(cp->type)); - mgmt_pending_remove(cmd); - goto failed; - } - - err = hci_req_run(&req, start_discovery_complete); - if (err < 0) { - mgmt_pending_remove(cmd); - goto failed; - } - hci_discovery_set_state(hdev, DISCOVERY_STARTING); + queue_work(hdev->req_workqueue, &hdev->discov_update); + err = 0; failed: hci_dev_unlock(hdev); return err; } -static void stop_discovery_complete(struct hci_dev *hdev, u8 status, u16 opcode) +void mgmt_stop_discovery_complete(struct hci_dev *hdev, u8 status) { struct mgmt_pending_cmd *cmd; @@ -4571,9 +4348,6 @@ static void stop_discovery_complete(struct hci_dev *hdev, u8 status, u16 opcode) mgmt_pending_remove(cmd); } - if (!status) - hci_discovery_set_state(hdev, DISCOVERY_STOPPED); - hci_dev_unlock(hdev); } @@ -4582,7 +4356,6 @@ static int stop_discovery(struct sock *sk, struct hci_dev *hdev, void *data, { struct mgmt_cp_stop_discovery *mgmt_cp = data; struct mgmt_pending_cmd *cmd; - struct hci_request req; int err; BT_DBG("%s", hdev->name); @@ -4611,24 +4384,9 @@ static int stop_discovery(struct sock *sk, struct hci_dev *hdev, void *data, cmd->cmd_complete = generic_cmd_complete; - hci_req_init(&req, hdev); - - hci_stop_discovery(&req); - - err = hci_req_run(&req, stop_discovery_complete); - if (!err) { - hci_discovery_set_state(hdev, DISCOVERY_STOPPING); - goto unlock; - } - - mgmt_pending_remove(cmd); - - /* If no HCI commands were sent we're done */ - if (err == -ENODATA) { - err = mgmt_cmd_complete(sk, hdev->id, MGMT_OP_STOP_DISCOVERY, 0, - &mgmt_cp->type, sizeof(mgmt_cp->type)); - hci_discovery_set_state(hdev, DISCOVERY_STOPPED); - } + hci_discovery_set_state(hdev, DISCOVERY_STOPPING); + queue_work(hdev->req_workqueue, &hdev->discov_update); + err = 0; unlock: hci_dev_unlock(hdev); @@ -6076,10 +5834,9 @@ static bool is_connected(struct hci_dev *hdev, bdaddr_t *addr, u8 type) } /* This function requires the caller holds hdev->lock */ -static int hci_conn_params_set(struct hci_request *req, bdaddr_t *addr, +static int hci_conn_params_set(struct hci_dev *hdev, bdaddr_t *addr, u8 addr_type, u8 auto_connect) { - struct hci_dev *hdev = req->hdev; struct hci_conn_params *params; params = hci_conn_params_add(hdev, addr, addr_type); @@ -6099,26 +5856,17 @@ static int hci_conn_params_set(struct hci_request *req, bdaddr_t *addr, */ if (params->explicit_connect) list_add(¶ms->action, &hdev->pend_le_conns); - - __hci_update_background_scan(req); break; case HCI_AUTO_CONN_REPORT: if (params->explicit_connect) list_add(¶ms->action, &hdev->pend_le_conns); else list_add(¶ms->action, &hdev->pend_le_reports); - __hci_update_background_scan(req); break; case HCI_AUTO_CONN_DIRECT: case HCI_AUTO_CONN_ALWAYS: - if (!is_connected(hdev, addr, addr_type)) { + if (!is_connected(hdev, addr, addr_type)) list_add(¶ms->action, &hdev->pend_le_conns); - /* If we are in scan phase of connecting, we were - * already added to pend_le_conns and scanning. - */ - if (params->auto_connect != HCI_AUTO_CONN_EXPLICIT) - __hci_update_background_scan(req); - } break; } @@ -6142,31 +5890,10 @@ static void device_added(struct sock *sk, struct hci_dev *hdev, mgmt_event(MGMT_EV_DEVICE_ADDED, hdev, &ev, sizeof(ev), sk); } -static void add_device_complete(struct hci_dev *hdev, u8 status, u16 opcode) -{ - struct mgmt_pending_cmd *cmd; - - BT_DBG("status 0x%02x", status); - - hci_dev_lock(hdev); - - cmd = pending_find(MGMT_OP_ADD_DEVICE, hdev); - if (!cmd) - goto unlock; - - cmd->cmd_complete(cmd, mgmt_status(status)); - mgmt_pending_remove(cmd); - -unlock: - hci_dev_unlock(hdev); -} - static int add_device(struct sock *sk, struct hci_dev *hdev, void *data, u16 len) { struct mgmt_cp_add_device *cp = data; - struct mgmt_pending_cmd *cmd; - struct hci_request req; u8 auto_conn, addr_type; int err; @@ -6183,24 +5910,15 @@ static int add_device(struct sock *sk, struct hci_dev *hdev, MGMT_STATUS_INVALID_PARAMS, &cp->addr, sizeof(cp->addr)); - hci_req_init(&req, hdev); - hci_dev_lock(hdev); - cmd = mgmt_pending_add(sk, MGMT_OP_ADD_DEVICE, hdev, data, len); - if (!cmd) { - err = -ENOMEM; - goto unlock; - } - - cmd->cmd_complete = addr_cmd_complete; - if (cp->addr.type == BDADDR_BREDR) { /* Only incoming connections action is supported for now */ if (cp->action != 0x01) { - err = cmd->cmd_complete(cmd, - MGMT_STATUS_INVALID_PARAMS); - mgmt_pending_remove(cmd); + err = mgmt_cmd_complete(sk, hdev->id, + MGMT_OP_ADD_DEVICE, + MGMT_STATUS_INVALID_PARAMS, + &cp->addr, sizeof(cp->addr)); goto unlock; } @@ -6209,7 +5927,7 @@ static int add_device(struct sock *sk, struct hci_dev *hdev, if (err) goto unlock; - __hci_update_page_scan(&req); + hci_update_page_scan(hdev); goto added; } @@ -6229,33 +5947,31 @@ static int add_device(struct sock *sk, struct hci_dev *hdev, * hci_conn_params_lookup. */ if (!hci_is_identity_address(&cp->addr.bdaddr, addr_type)) { - err = cmd->cmd_complete(cmd, MGMT_STATUS_INVALID_PARAMS); - mgmt_pending_remove(cmd); + err = mgmt_cmd_complete(sk, hdev->id, MGMT_OP_ADD_DEVICE, + MGMT_STATUS_INVALID_PARAMS, + &cp->addr, sizeof(cp->addr)); goto unlock; } /* If the connection parameters don't exist for this device, * they will be created and configured with defaults. */ - if (hci_conn_params_set(&req, &cp->addr.bdaddr, addr_type, + if (hci_conn_params_set(hdev, &cp->addr.bdaddr, addr_type, auto_conn) < 0) { - err = cmd->cmd_complete(cmd, MGMT_STATUS_FAILED); - mgmt_pending_remove(cmd); + err = mgmt_cmd_complete(sk, hdev->id, MGMT_OP_ADD_DEVICE, + MGMT_STATUS_FAILED, &cp->addr, + sizeof(cp->addr)); goto unlock; } + hci_update_background_scan(hdev); + added: device_added(sk, hdev, &cp->addr.bdaddr, cp->addr.type, cp->action); - err = hci_req_run(&req, add_device_complete); - if (err < 0) { - /* ENODATA means no HCI commands were needed (e.g. if - * the adapter is powered off). - */ - if (err == -ENODATA) - err = cmd->cmd_complete(cmd, MGMT_STATUS_SUCCESS); - mgmt_pending_remove(cmd); - } + err = mgmt_cmd_complete(sk, hdev->id, MGMT_OP_ADD_DEVICE, + MGMT_STATUS_SUCCESS, &cp->addr, + sizeof(cp->addr)); unlock: hci_dev_unlock(hdev); @@ -6273,55 +5989,25 @@ static void device_removed(struct sock *sk, struct hci_dev *hdev, mgmt_event(MGMT_EV_DEVICE_REMOVED, hdev, &ev, sizeof(ev), sk); } -static void remove_device_complete(struct hci_dev *hdev, u8 status, u16 opcode) -{ - struct mgmt_pending_cmd *cmd; - - BT_DBG("status 0x%02x", status); - - hci_dev_lock(hdev); - - cmd = pending_find(MGMT_OP_REMOVE_DEVICE, hdev); - if (!cmd) - goto unlock; - - cmd->cmd_complete(cmd, mgmt_status(status)); - mgmt_pending_remove(cmd); - -unlock: - hci_dev_unlock(hdev); -} - static int remove_device(struct sock *sk, struct hci_dev *hdev, void *data, u16 len) { struct mgmt_cp_remove_device *cp = data; - struct mgmt_pending_cmd *cmd; - struct hci_request req; int err; BT_DBG("%s", hdev->name); - hci_req_init(&req, hdev); - hci_dev_lock(hdev); - cmd = mgmt_pending_add(sk, MGMT_OP_REMOVE_DEVICE, hdev, data, len); - if (!cmd) { - err = -ENOMEM; - goto unlock; - } - - cmd->cmd_complete = addr_cmd_complete; - if (bacmp(&cp->addr.bdaddr, BDADDR_ANY)) { struct hci_conn_params *params; u8 addr_type; if (!bdaddr_type_is_valid(cp->addr.type)) { - err = cmd->cmd_complete(cmd, - MGMT_STATUS_INVALID_PARAMS); - mgmt_pending_remove(cmd); + err = mgmt_cmd_complete(sk, hdev->id, + MGMT_OP_REMOVE_DEVICE, + MGMT_STATUS_INVALID_PARAMS, + &cp->addr, sizeof(cp->addr)); goto unlock; } @@ -6330,13 +6016,15 @@ static int remove_device(struct sock *sk, struct hci_dev *hdev, &cp->addr.bdaddr, cp->addr.type); if (err) { - err = cmd->cmd_complete(cmd, - MGMT_STATUS_INVALID_PARAMS); - mgmt_pending_remove(cmd); + err = mgmt_cmd_complete(sk, hdev->id, + MGMT_OP_REMOVE_DEVICE, + MGMT_STATUS_INVALID_PARAMS, + &cp->addr, + sizeof(cp->addr)); goto unlock; } - __hci_update_page_scan(&req); + hci_update_page_scan(hdev); device_removed(sk, hdev, &cp->addr.bdaddr, cp->addr.type); @@ -6351,33 +6039,36 @@ static int remove_device(struct sock *sk, struct hci_dev *hdev, * hci_conn_params_lookup. */ if (!hci_is_identity_address(&cp->addr.bdaddr, addr_type)) { - err = cmd->cmd_complete(cmd, - MGMT_STATUS_INVALID_PARAMS); - mgmt_pending_remove(cmd); + err = mgmt_cmd_complete(sk, hdev->id, + MGMT_OP_REMOVE_DEVICE, + MGMT_STATUS_INVALID_PARAMS, + &cp->addr, sizeof(cp->addr)); goto unlock; } params = hci_conn_params_lookup(hdev, &cp->addr.bdaddr, addr_type); if (!params) { - err = cmd->cmd_complete(cmd, - MGMT_STATUS_INVALID_PARAMS); - mgmt_pending_remove(cmd); + err = mgmt_cmd_complete(sk, hdev->id, + MGMT_OP_REMOVE_DEVICE, + MGMT_STATUS_INVALID_PARAMS, + &cp->addr, sizeof(cp->addr)); goto unlock; } if (params->auto_connect == HCI_AUTO_CONN_DISABLED || params->auto_connect == HCI_AUTO_CONN_EXPLICIT) { - err = cmd->cmd_complete(cmd, - MGMT_STATUS_INVALID_PARAMS); - mgmt_pending_remove(cmd); + err = mgmt_cmd_complete(sk, hdev->id, + MGMT_OP_REMOVE_DEVICE, + MGMT_STATUS_INVALID_PARAMS, + &cp->addr, sizeof(cp->addr)); goto unlock; } list_del(¶ms->action); list_del(¶ms->list); kfree(params); - __hci_update_background_scan(&req); + hci_update_background_scan(hdev); device_removed(sk, hdev, &cp->addr.bdaddr, cp->addr.type); } else { @@ -6385,9 +6076,10 @@ static int remove_device(struct sock *sk, struct hci_dev *hdev, struct bdaddr_list *b, *btmp; if (cp->addr.type) { - err = cmd->cmd_complete(cmd, - MGMT_STATUS_INVALID_PARAMS); - mgmt_pending_remove(cmd); + err = mgmt_cmd_complete(sk, hdev->id, + MGMT_OP_REMOVE_DEVICE, + MGMT_STATUS_INVALID_PARAMS, + &cp->addr, sizeof(cp->addr)); goto unlock; } @@ -6397,7 +6089,7 @@ static int remove_device(struct sock *sk, struct hci_dev *hdev, kfree(b); } - __hci_update_page_scan(&req); + hci_update_page_scan(hdev); list_for_each_entry_safe(p, tmp, &hdev->le_conn_params, list) { if (p->auto_connect == HCI_AUTO_CONN_DISABLED) @@ -6414,20 +6106,13 @@ static int remove_device(struct sock *sk, struct hci_dev *hdev, BT_DBG("All LE connection parameters were removed"); - __hci_update_background_scan(&req); + hci_update_background_scan(hdev); } complete: - err = hci_req_run(&req, remove_device_complete); - if (err < 0) { - /* ENODATA means no HCI commands were needed (e.g. if - * the adapter is powered off). - */ - if (err == -ENODATA) - err = cmd->cmd_complete(cmd, MGMT_STATUS_SUCCESS); - mgmt_pending_remove(cmd); - } - + err = mgmt_cmd_complete(sk, hdev->id, MGMT_OP_REMOVE_DEVICE, + MGMT_STATUS_SUCCESS, &cp->addr, + sizeof(cp->addr)); unlock: hci_dev_unlock(hdev); return err; @@ -7016,17 +6701,19 @@ static bool tlv_data_is_valid(struct hci_dev *hdev, u32 adv_flags, u8 *data, int i, cur_len; bool flags_managed = false; bool tx_power_managed = false; - u32 flags_params = MGMT_ADV_FLAG_DISCOV | MGMT_ADV_FLAG_LIMITED_DISCOV | - MGMT_ADV_FLAG_MANAGED_FLAGS; - if (is_adv_data && (adv_flags & flags_params)) { - flags_managed = true; - max_len -= 3; - } + if (is_adv_data) { + if (adv_flags & (MGMT_ADV_FLAG_DISCOV | + MGMT_ADV_FLAG_LIMITED_DISCOV | + MGMT_ADV_FLAG_MANAGED_FLAGS)) { + flags_managed = true; + max_len -= 3; + } - if (is_adv_data && (adv_flags & MGMT_ADV_FLAG_TX_POWER)) { - tx_power_managed = true; - max_len -= 3; + if (adv_flags & MGMT_ADV_FLAG_TX_POWER) { + tx_power_managed = true; + max_len -= 3; + } } if (len > max_len) @@ -7155,6 +6842,10 @@ static int add_advertising(struct sock *sk, struct hci_dev *hdev, return mgmt_cmd_status(sk, hdev->id, MGMT_OP_ADD_ADVERTISING, status); + if (cp->instance < 1 || cp->instance > HCI_MAX_ADV_INSTANCES) + return mgmt_cmd_status(sk, hdev->id, MGMT_OP_ADD_ADVERTISING, + MGMT_STATUS_INVALID_PARAMS); + flags = __le32_to_cpu(cp->flags); timeout = __le16_to_cpu(cp->timeout); duration = __le16_to_cpu(cp->duration); @@ -7369,6 +7060,62 @@ unlock: return err; } +static u8 tlv_data_max_len(u32 adv_flags, bool is_adv_data) +{ + u8 max_len = HCI_MAX_AD_LENGTH; + + if (is_adv_data) { + if (adv_flags & (MGMT_ADV_FLAG_DISCOV | + MGMT_ADV_FLAG_LIMITED_DISCOV | + MGMT_ADV_FLAG_MANAGED_FLAGS)) + max_len -= 3; + + if (adv_flags & MGMT_ADV_FLAG_TX_POWER) + max_len -= 3; + } + + return max_len; +} + +static int get_adv_size_info(struct sock *sk, struct hci_dev *hdev, + void *data, u16 data_len) +{ + struct mgmt_cp_get_adv_size_info *cp = data; + struct mgmt_rp_get_adv_size_info rp; + u32 flags, supported_flags; + int err; + + BT_DBG("%s", hdev->name); + + if (!lmp_le_capable(hdev)) + return mgmt_cmd_status(sk, hdev->id, MGMT_OP_GET_ADV_SIZE_INFO, + MGMT_STATUS_REJECTED); + + if (cp->instance < 1 || cp->instance > HCI_MAX_ADV_INSTANCES) + return mgmt_cmd_status(sk, hdev->id, MGMT_OP_GET_ADV_SIZE_INFO, + MGMT_STATUS_INVALID_PARAMS); + + flags = __le32_to_cpu(cp->flags); + + /* The current implementation only supports a subset of the specified + * flags. + */ + supported_flags = get_supported_adv_flags(hdev); + if (flags & ~supported_flags) + return mgmt_cmd_status(sk, hdev->id, MGMT_OP_GET_ADV_SIZE_INFO, + MGMT_STATUS_INVALID_PARAMS); + + rp.instance = cp->instance; + rp.flags = cp->flags; + rp.max_adv_data_len = tlv_data_max_len(flags, true); + rp.max_scan_rsp_len = tlv_data_max_len(flags, false); + + err = mgmt_cmd_complete(sk, hdev->id, MGMT_OP_GET_ADV_SIZE_INFO, + MGMT_STATUS_SUCCESS, &rp, sizeof(rp)); + + return err; +} + static const struct hci_mgmt_handler mgmt_handlers[] = { { NULL }, /* 0x0000 (no command) */ { read_version, MGMT_READ_VERSION_SIZE, @@ -7456,6 +7203,7 @@ static const struct hci_mgmt_handler mgmt_handlers[] = { { add_advertising, MGMT_ADD_ADVERTISING_SIZE, HCI_MGMT_VAR_LEN }, { remove_advertising, MGMT_REMOVE_ADVERTISING_SIZE }, + { get_adv_size_info, MGMT_GET_ADV_SIZE_INFO_SIZE }, }; void mgmt_index_added(struct hci_dev *hdev) @@ -7526,9 +7274,8 @@ void mgmt_index_removed(struct hci_dev *hdev) } /* This function requires the caller holds hdev->lock */ -static void restart_le_actions(struct hci_request *req) +static void restart_le_actions(struct hci_dev *hdev) { - struct hci_dev *hdev = req->hdev; struct hci_conn_params *p; list_for_each_entry(p, &hdev->le_conn_params, list) { @@ -7549,8 +7296,6 @@ static void restart_le_actions(struct hci_request *req) break; } } - - __hci_update_background_scan(req); } static void powered_complete(struct hci_dev *hdev, u8 status, u16 opcode) @@ -7560,12 +7305,8 @@ static void powered_complete(struct hci_dev *hdev, u8 status, u16 opcode) BT_DBG("status 0x%02x", status); if (!status) { - /* Register the available SMP channels (BR/EDR and LE) only - * when successfully powering on the controller. This late - * registration is required so that LE SMP can clearly - * decide if the public address or static address is used. - */ - smp_register(hdev); + restart_le_actions(hdev); + hci_update_background_scan(hdev); } hci_dev_lock(hdev); @@ -7644,8 +7385,6 @@ static int powered_update_hci(struct hci_dev *hdev) hdev->cur_adv_instance) schedule_adv_instance(&req, hdev->cur_adv_instance, true); - - restart_le_actions(&req); } link_sec = hci_dev_test_flag(hdev, HCI_LINK_SECURITY); @@ -7677,6 +7416,13 @@ int mgmt_powered(struct hci_dev *hdev, u8 powered) return 0; if (powered) { + /* Register the available SMP channels (BR/EDR and LE) only + * when successfully powering on the controller. This late + * registration is required so that LE SMP can clearly + * decide if the public address or static address is used. + */ + smp_register(hdev); + if (powered_update_hci(hdev) == 0) return 0; @@ -8452,7 +8198,7 @@ static void restart_le_scan(struct hci_dev *hdev) hdev->discovery.scan_duration)) return; - queue_delayed_work(hdev->workqueue, &hdev->le_scan_restart, + queue_delayed_work(hdev->req_workqueue, &hdev->le_scan_restart, DISCOV_LE_RESTART_DELAY); } diff --git a/net/bridge/br_if.c b/net/bridge/br_if.c index ec02f58..8d1d4a2 100644 --- a/net/bridge/br_if.c +++ b/net/bridge/br_if.c @@ -493,7 +493,7 @@ int br_add_if(struct net_bridge *br, struct net_device *dev) dev->priv_flags |= IFF_BRIDGE_PORT; - err = netdev_master_upper_dev_link(dev, br->dev); + err = netdev_master_upper_dev_link(dev, br->dev, NULL, NULL); if (err) goto err5; diff --git a/net/core/dev.c b/net/core/dev.c index ae00b89..d1706e8 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -96,6 +96,7 @@ #include <linux/skbuff.h> #include <net/net_namespace.h> #include <net/sock.h> +#include <net/busy_poll.h> #include <linux/rtnetlink.h> #include <linux/stat.h> #include <net/dst.h> @@ -182,8 +183,8 @@ EXPORT_SYMBOL(dev_base_lock); /* protects napi_hash addition/deletion and napi_gen_id */ static DEFINE_SPINLOCK(napi_hash_lock); -static unsigned int napi_gen_id; -static DEFINE_HASHTABLE(napi_hash, 8); +static unsigned int napi_gen_id = NR_CPUS; +static DEFINE_READ_MOSTLY_HASHTABLE(napi_hash, 8); static seqcount_t devnet_rename_seq; @@ -3021,7 +3022,9 @@ struct netdev_queue *netdev_pick_tx(struct net_device *dev, int queue_index = 0; #ifdef CONFIG_XPS - if (skb->sender_cpu == 0) + u32 sender_cpu = skb->sender_cpu - 1; + + if (sender_cpu >= (u32)NR_CPUS) skb->sender_cpu = raw_smp_processor_id() + 1; #endif @@ -4353,6 +4356,7 @@ static gro_result_t napi_skb_finish(gro_result_t ret, struct sk_buff *skb) gro_result_t napi_gro_receive(struct napi_struct *napi, struct sk_buff *skb) { + skb_mark_napi_id(skb, napi); trace_napi_gro_receive_entry(skb); skb_gro_reset_offset(skb); @@ -4386,7 +4390,10 @@ struct sk_buff *napi_get_frags(struct napi_struct *napi) if (!skb) { skb = napi_alloc_skb(napi, GRO_MAX_HEAD); - napi->skb = skb; + if (skb) { + napi->skb = skb; + skb_mark_napi_id(skb, napi); + } } return skb; } @@ -4661,7 +4668,7 @@ void napi_complete_done(struct napi_struct *n, int work_done) EXPORT_SYMBOL(napi_complete_done); /* must be called under rcu_read_lock(), as we dont take a reference */ -struct napi_struct *napi_by_id(unsigned int napi_id) +static struct napi_struct *napi_by_id(unsigned int napi_id) { unsigned int hash = napi_id % HASH_SIZE(napi_hash); struct napi_struct *napi; @@ -4672,43 +4679,101 @@ struct napi_struct *napi_by_id(unsigned int napi_id) return NULL; } -EXPORT_SYMBOL_GPL(napi_by_id); -void napi_hash_add(struct napi_struct *napi) +#if defined(CONFIG_NET_RX_BUSY_POLL) +#define BUSY_POLL_BUDGET 8 +bool sk_busy_loop(struct sock *sk, int nonblock) { - if (!test_and_set_bit(NAPI_STATE_HASHED, &napi->state)) { + unsigned long end_time = !nonblock ? sk_busy_loop_end_time(sk) : 0; + int (*busy_poll)(struct napi_struct *dev); + struct napi_struct *napi; + int rc = false; - spin_lock(&napi_hash_lock); + rcu_read_lock(); - /* 0 is not a valid id, we also skip an id that is taken - * we expect both events to be extremely rare - */ - napi->napi_id = 0; - while (!napi->napi_id) { - napi->napi_id = ++napi_gen_id; - if (napi_by_id(napi->napi_id)) - napi->napi_id = 0; + napi = napi_by_id(sk->sk_napi_id); + if (!napi) + goto out; + + /* Note: ndo_busy_poll method is optional in linux-4.5 */ + busy_poll = napi->dev->netdev_ops->ndo_busy_poll; + + do { + rc = 0; + local_bh_disable(); + if (busy_poll) { + rc = busy_poll(napi); + } else if (napi_schedule_prep(napi)) { + void *have = netpoll_poll_lock(napi); + + if (test_bit(NAPI_STATE_SCHED, &napi->state)) { + rc = napi->poll(napi, BUSY_POLL_BUDGET); + trace_napi_poll(napi); + if (rc == BUSY_POLL_BUDGET) { + napi_complete_done(napi, rc); + napi_schedule(napi); + } + } + netpoll_poll_unlock(have); } + if (rc > 0) + NET_ADD_STATS_BH(sock_net(sk), + LINUX_MIB_BUSYPOLLRXPACKETS, rc); + local_bh_enable(); - hlist_add_head_rcu(&napi->napi_hash_node, - &napi_hash[napi->napi_id % HASH_SIZE(napi_hash)]); + if (rc == LL_FLUSH_FAILED) + break; /* permanent failure */ - spin_unlock(&napi_hash_lock); - } + cpu_relax(); + } while (!nonblock && skb_queue_empty(&sk->sk_receive_queue) && + !need_resched() && !busy_loop_timeout(end_time)); + + rc = !skb_queue_empty(&sk->sk_receive_queue); +out: + rcu_read_unlock(); + return rc; +} +EXPORT_SYMBOL(sk_busy_loop); + +#endif /* CONFIG_NET_RX_BUSY_POLL */ + +void napi_hash_add(struct napi_struct *napi) +{ + if (test_bit(NAPI_STATE_NO_BUSY_POLL, &napi->state) || + test_and_set_bit(NAPI_STATE_HASHED, &napi->state)) + return; + + spin_lock(&napi_hash_lock); + + /* 0..NR_CPUS+1 range is reserved for sender_cpu use */ + do { + if (unlikely(++napi_gen_id < NR_CPUS + 1)) + napi_gen_id = NR_CPUS + 1; + } while (napi_by_id(napi_gen_id)); + napi->napi_id = napi_gen_id; + + hlist_add_head_rcu(&napi->napi_hash_node, + &napi_hash[napi->napi_id % HASH_SIZE(napi_hash)]); + + spin_unlock(&napi_hash_lock); } EXPORT_SYMBOL_GPL(napi_hash_add); /* Warning : caller is responsible to make sure rcu grace period * is respected before freeing memory containing @napi */ -void napi_hash_del(struct napi_struct *napi) +bool napi_hash_del(struct napi_struct *napi) { + bool rcu_sync_needed = false; + spin_lock(&napi_hash_lock); - if (test_and_clear_bit(NAPI_STATE_HASHED, &napi->state)) + if (test_and_clear_bit(NAPI_STATE_HASHED, &napi->state)) { + rcu_sync_needed = true; hlist_del_rcu(&napi->napi_hash_node); - + } spin_unlock(&napi_hash_lock); + return rcu_sync_needed; } EXPORT_SYMBOL_GPL(napi_hash_del); @@ -4744,6 +4809,7 @@ void netif_napi_add(struct net_device *dev, struct napi_struct *napi, napi->poll_owner = -1; #endif set_bit(NAPI_STATE_SCHED, &napi->state); + napi_hash_add(napi); } EXPORT_SYMBOL(netif_napi_add); @@ -4763,8 +4829,12 @@ void napi_disable(struct napi_struct *n) } EXPORT_SYMBOL(napi_disable); +/* Must be called in process context */ void netif_napi_del(struct napi_struct *napi) { + might_sleep(); + if (napi_hash_del(napi)) + synchronize_net(); list_del_init(&napi->dev_list); napi_free_frags(napi); @@ -5351,7 +5421,7 @@ static void __netdev_adjacent_dev_unlink_neighbour(struct net_device *dev, static int __netdev_upper_dev_link(struct net_device *dev, struct net_device *upper_dev, bool master, - void *private) + void *upper_priv, void *upper_info) { struct netdev_notifier_changeupper_info changeupper_info; struct netdev_adjacent *i, *j, *to_i, *to_j; @@ -5375,6 +5445,7 @@ static int __netdev_upper_dev_link(struct net_device *dev, changeupper_info.upper_dev = upper_dev; changeupper_info.master = master; changeupper_info.linking = true; + changeupper_info.upper_info = upper_info; ret = call_netdevice_notifiers_info(NETDEV_PRECHANGEUPPER, dev, &changeupper_info.info); @@ -5382,7 +5453,7 @@ static int __netdev_upper_dev_link(struct net_device *dev, if (ret) return ret; - ret = __netdev_adjacent_dev_link_neighbour(dev, upper_dev, private, + ret = __netdev_adjacent_dev_link_neighbour(dev, upper_dev, upper_priv, master); if (ret) return ret; @@ -5420,8 +5491,12 @@ static int __netdev_upper_dev_link(struct net_device *dev, goto rollback_lower_mesh; } - call_netdevice_notifiers_info(NETDEV_CHANGEUPPER, dev, - &changeupper_info.info); + ret = call_netdevice_notifiers_info(NETDEV_CHANGEUPPER, dev, + &changeupper_info.info); + ret = notifier_to_errno(ret); + if (ret) + goto rollback_lower_mesh; + return 0; rollback_lower_mesh: @@ -5475,7 +5550,7 @@ rollback_mesh: int netdev_upper_dev_link(struct net_device *dev, struct net_device *upper_dev) { - return __netdev_upper_dev_link(dev, upper_dev, false, NULL); + return __netdev_upper_dev_link(dev, upper_dev, false, NULL, NULL); } EXPORT_SYMBOL(netdev_upper_dev_link); @@ -5483,6 +5558,8 @@ EXPORT_SYMBOL(netdev_upper_dev_link); * netdev_master_upper_dev_link - Add a master link to the upper device * @dev: device * @upper_dev: new upper device + * @upper_priv: upper device private + * @upper_info: upper info to be passed down via notifier * * Adds a link to device which is upper to this one. In this case, only * one master upper device can be linked, although other non-master devices @@ -5491,20 +5568,14 @@ EXPORT_SYMBOL(netdev_upper_dev_link); * counts are adjusted and the function returns zero. */ int netdev_master_upper_dev_link(struct net_device *dev, - struct net_device *upper_dev) + struct net_device *upper_dev, + void *upper_priv, void *upper_info) { - return __netdev_upper_dev_link(dev, upper_dev, true, NULL); + return __netdev_upper_dev_link(dev, upper_dev, true, + upper_priv, upper_info); } EXPORT_SYMBOL(netdev_master_upper_dev_link); -int netdev_master_upper_dev_link_private(struct net_device *dev, - struct net_device *upper_dev, - void *private) -{ - return __netdev_upper_dev_link(dev, upper_dev, true, private); -} -EXPORT_SYMBOL(netdev_master_upper_dev_link_private); - /** * netdev_upper_dev_unlink - Removes a link to upper device * @dev: device @@ -5685,6 +5756,26 @@ int dev_get_nest_level(struct net_device *dev, } EXPORT_SYMBOL(dev_get_nest_level); +/** + * netdev_lower_change - Dispatch event about lower device state change + * @lower_dev: device + * @lower_state_info: state to dispatch + * + * Send NETDEV_CHANGELOWERSTATE to netdev notifiers with info. + * The caller must hold the RTNL lock. + */ +void netdev_lower_state_changed(struct net_device *lower_dev, + void *lower_state_info) +{ + struct netdev_notifier_changelowerstate_info changelowerstate_info; + + ASSERT_RTNL(); + changelowerstate_info.lower_state_info = lower_state_info; + call_netdevice_notifiers_info(NETDEV_CHANGELOWERSTATE, lower_dev, + &changelowerstate_info.info); +} +EXPORT_SYMBOL(netdev_lower_state_changed); + static void dev_change_rx_flags(struct net_device *dev, int flags) { const struct net_device_ops *ops = dev->netdev_ops; @@ -7164,11 +7255,13 @@ EXPORT_SYMBOL(alloc_netdev_mqs); * This function does the last stage of destroying an allocated device * interface. The reference to the device object is released. * If this is the last reference then it will be freed. + * Must be called in process context. */ void free_netdev(struct net_device *dev) { struct napi_struct *p, *n; + might_sleep(); netif_free_tx_queues(dev); #ifdef CONFIG_SYSFS kvfree(dev->_rx); diff --git a/net/core/net-traces.c b/net/core/net-traces.c index adef015..92da5e4 100644 --- a/net/core/net-traces.c +++ b/net/core/net-traces.c @@ -32,6 +32,10 @@ #include <trace/events/sock.h> #include <trace/events/udp.h> #include <trace/events/fib.h> +#if IS_ENABLED(CONFIG_IPV6) +#include <trace/events/fib6.h> +EXPORT_TRACEPOINT_SYMBOL_GPL(fib6_table_lookup); +#endif EXPORT_TRACEPOINT_SYMBOL_GPL(kfree_skb); diff --git a/net/core/sock.c b/net/core/sock.c index e31dfce..7965ef4 100644 --- a/net/core/sock.c +++ b/net/core/sock.c @@ -2282,7 +2282,7 @@ static void sock_def_wakeup(struct sock *sk) rcu_read_lock(); wq = rcu_dereference(sk->sk_wq); - if (wq_has_sleeper(wq)) + if (skwq_has_sleeper(wq)) wake_up_interruptible_all(&wq->wait); rcu_read_unlock(); } @@ -2293,7 +2293,7 @@ static void sock_def_error_report(struct sock *sk) rcu_read_lock(); wq = rcu_dereference(sk->sk_wq); - if (wq_has_sleeper(wq)) + if (skwq_has_sleeper(wq)) wake_up_interruptible_poll(&wq->wait, POLLERR); sk_wake_async(sk, SOCK_WAKE_IO, POLL_ERR); rcu_read_unlock(); @@ -2305,7 +2305,7 @@ static void sock_def_readable(struct sock *sk) rcu_read_lock(); wq = rcu_dereference(sk->sk_wq); - if (wq_has_sleeper(wq)) + if (skwq_has_sleeper(wq)) wake_up_interruptible_sync_poll(&wq->wait, POLLIN | POLLPRI | POLLRDNORM | POLLRDBAND); sk_wake_async(sk, SOCK_WAKE_WAITD, POLL_IN); @@ -2323,7 +2323,7 @@ static void sock_def_write_space(struct sock *sk) */ if ((atomic_read(&sk->sk_wmem_alloc) << 1) <= sk->sk_sndbuf) { wq = rcu_dereference(sk->sk_wq); - if (wq_has_sleeper(wq)) + if (skwq_has_sleeper(wq)) wake_up_interruptible_sync_poll(&wq->wait, POLLOUT | POLLWRNORM | POLLWRBAND); diff --git a/net/core/stream.c b/net/core/stream.c index b96f7a7..159516a 100644 --- a/net/core/stream.c +++ b/net/core/stream.c @@ -35,7 +35,7 @@ void sk_stream_write_space(struct sock *sk) rcu_read_lock(); wq = rcu_dereference(sk->sk_wq); - if (wq_has_sleeper(wq)) + if (skwq_has_sleeper(wq)) wake_up_interruptible_poll(&wq->wait, POLLOUT | POLLWRNORM | POLLWRBAND); if (wq && wq->fasync_list && !(sk->sk_shutdown & SEND_SHUTDOWN)) diff --git a/net/dccp/output.c b/net/dccp/output.c index 4ce912e..b66c84d 100644 --- a/net/dccp/output.c +++ b/net/dccp/output.c @@ -201,7 +201,7 @@ void dccp_write_space(struct sock *sk) rcu_read_lock(); wq = rcu_dereference(sk->sk_wq); - if (wq_has_sleeper(wq)) + if (skwq_has_sleeper(wq)) wake_up_interruptible(&wq->wait); /* Should agree with poll, otherwise some programs break */ if (sock_writeable(sk)) diff --git a/net/dsa/dsa.c b/net/dsa/dsa.c index 1eba07f..b7448c8 100644 --- a/net/dsa/dsa.c +++ b/net/dsa/dsa.c @@ -21,8 +21,10 @@ #include <linux/of_mdio.h> #include <linux/of_platform.h> #include <linux/of_net.h> +#include <linux/of_gpio.h> #include <linux/sysfs.h> #include <linux/phy_fixed.h> +#include <linux/gpio/consumer.h> #include "dsa_priv.h" char dsa_driver_version[] = "0.1"; @@ -688,6 +690,9 @@ static int dsa_of_probe(struct device *dev) const char *port_name; int chip_index, port_index; const unsigned int *sw_addr, *port_reg; + int gpio; + enum of_gpio_flags of_flags; + unsigned long flags; u32 eeprom_len; int ret; @@ -766,6 +771,19 @@ static int dsa_of_probe(struct device *dev) put_device(cd->host_dev); cd->host_dev = &mdio_bus_switch->dev; } + gpio = of_get_named_gpio_flags(child, "reset-gpios", 0, + &of_flags); + if (gpio_is_valid(gpio)) { + flags = (of_flags == OF_GPIO_ACTIVE_LOW ? + GPIOF_ACTIVE_LOW : 0); + ret = devm_gpio_request_one(dev, gpio, flags, + "switch_reset"); + if (ret) + goto out_free_chip; + + cd->reset = gpio_to_desc(gpio); + gpiod_direction_output(cd->reset, 0); + } for_each_available_child_of_node(child, port) { port_reg = of_get_property(port, "reg", NULL); diff --git a/net/ipv4/ip_gre.c b/net/ipv4/ip_gre.c index 6145214..04a48c0 100644 --- a/net/ipv4/ip_gre.c +++ b/net/ipv4/ip_gre.c @@ -24,7 +24,6 @@ #include <linux/tcp.h> #include <linux/udp.h> #include <linux/if_arp.h> -#include <linux/mroute.h> #include <linux/if_vlan.h> #include <linux/init.h> #include <linux/in6.h> diff --git a/net/ipv4/ip_output.c b/net/ipv4/ip_output.c index 4233cbe..e0b94cd 100644 --- a/net/ipv4/ip_output.c +++ b/net/ipv4/ip_output.c @@ -76,7 +76,6 @@ #include <linux/igmp.h> #include <linux/netfilter_ipv4.h> #include <linux/netfilter_bridge.h> -#include <linux/mroute.h> #include <linux/netlink.h> #include <linux/tcp.h> diff --git a/net/ipv4/ip_tunnel.c b/net/ipv4/ip_tunnel.c index cbb51f3..0f6e9ee 100644 --- a/net/ipv4/ip_tunnel.c +++ b/net/ipv4/ip_tunnel.c @@ -30,7 +30,6 @@ #include <linux/tcp.h> #include <linux/udp.h> #include <linux/if_arp.h> -#include <linux/mroute.h> #include <linux/init.h> #include <linux/in6.h> #include <linux/inetdevice.h> diff --git a/net/ipv4/ip_tunnel_core.c b/net/ipv4/ip_tunnel_core.c index 6cb9009..1db8418 100644 --- a/net/ipv4/ip_tunnel_core.c +++ b/net/ipv4/ip_tunnel_core.c @@ -24,7 +24,6 @@ #include <linux/netdevice.h> #include <linux/in.h> #include <linux/if_arp.h> -#include <linux/mroute.h> #include <linux/init.h> #include <linux/in6.h> #include <linux/inetdevice.h> diff --git a/net/ipv4/ip_vti.c b/net/ipv4/ip_vti.c index 4d8f0b6..02d9c21 100644 --- a/net/ipv4/ip_vti.c +++ b/net/ipv4/ip_vti.c @@ -30,7 +30,6 @@ #include <linux/tcp.h> #include <linux/udp.h> #include <linux/if_arp.h> -#include <linux/mroute.h> #include <linux/init.h> #include <linux/netfilter_ipv4.h> #include <linux/if_ether.h> diff --git a/net/ipv4/ipconfig.c b/net/ipv4/ipconfig.c index 0bc7412..67f7c9d 100644 --- a/net/ipv4/ipconfig.c +++ b/net/ipv4/ipconfig.c @@ -65,15 +65,6 @@ #include <net/checksum.h> #include <asm/processor.h> -/* Define this to allow debugging output */ -#undef IPCONFIG_DEBUG - -#ifdef IPCONFIG_DEBUG -#define DBG(x) printk x -#else -#define DBG(x) do { } while(0) -#endif - #if defined(CONFIG_IP_PNP_DHCP) #define IPCONFIG_DHCP #endif @@ -227,7 +218,7 @@ static int __init ic_open_devs(void) if (dev->mtu >= 364) able |= IC_BOOTP; else - pr_warn("DHCP/BOOTP: Ignoring device %s, MTU %d too small", + pr_warn("DHCP/BOOTP: Ignoring device %s, MTU %d too small\n", dev->name, dev->mtu); if (!(dev->flags & IFF_NOARP)) able |= IC_RARP; @@ -254,8 +245,8 @@ static int __init ic_open_devs(void) else d->xid = 0; ic_proto_have_if |= able; - DBG(("IP-Config: %s UP (able=%d, xid=%08x)\n", - dev->name, able, d->xid)); + pr_debug("IP-Config: %s UP (able=%d, xid=%08x)\n", + dev->name, able, d->xid); } } @@ -311,7 +302,7 @@ static void __init ic_close_devs(void) next = d->next; dev = d->dev; if (dev != ic_dev && !netdev_uses_dsa(dev)) { - DBG(("IP-Config: Downing %s\n", dev->name)); + pr_debug("IP-Config: Downing %s\n", dev->name); dev_change_flags(dev, d->flags); } kfree(d); @@ -464,7 +455,8 @@ static int __init ic_defaults(void) &ic_myaddr); return -1; } - printk("IP-Config: Guessing netmask %pI4\n", &ic_netmask); + pr_notice("IP-Config: Guessing netmask %pI4\n", + &ic_netmask); } return 0; @@ -675,9 +667,7 @@ ic_dhcp_init_options(u8 *options) u8 *e = options; int len; -#ifdef IPCONFIG_DEBUG - printk("DHCP: Sending message type %d\n", mt); -#endif + pr_debug("DHCP: Sending message type %d\n", mt); memcpy(e, ic_bootp_cookie, 4); /* RFC1048 Magic Cookie */ e += 4; @@ -847,7 +837,8 @@ static void __init ic_bootp_send_if(struct ic_device *d, unsigned long jiffies_d else if (dev->type == ARPHRD_FDDI) b->htype = ARPHRD_ETHER; else { - printk("Unknown ARP type 0x%04x for device %s\n", dev->type, dev->name); + pr_warn("Unknown ARP type 0x%04x for device %s\n", dev->type, + dev->name); b->htype = dev->type; /* can cause undefined behavior */ } @@ -904,14 +895,12 @@ static void __init ic_do_bootp_ext(u8 *ext) int i; __be16 mtu; -#ifdef IPCONFIG_DEBUG u8 *c; - printk("DHCP/BOOTP: Got extension %d:",*ext); + pr_debug("DHCP/BOOTP: Got extension %d:", *ext); for (c=ext+2; c<ext+2+ext[1]; c++) - printk(" %02x", *c); - printk("\n"); -#endif + pr_debug(" %02x", *c); + pr_debug("\n"); switch (*ext++) { case 1: /* Subnet mask */ @@ -1080,9 +1069,7 @@ static int __init ic_bootp_recv(struct sk_buff *skb, struct net_device *dev, str } } -#ifdef IPCONFIG_DEBUG - printk("DHCP: Got message type %d\n", mt); -#endif + pr_debug("DHCP: Got message type %d\n", mt); switch (mt) { case DHCPOFFER: @@ -1095,10 +1082,8 @@ static int __init ic_bootp_recv(struct sk_buff *skb, struct net_device *dev, str /* Let's accept that offer. */ ic_myaddr = b->your_ip; ic_servaddr = server_id; -#ifdef IPCONFIG_DEBUG - printk("DHCP: Offered address %pI4 by server %pI4\n", - &ic_myaddr, &b->iph.saddr); -#endif + pr_debug("DHCP: Offered address %pI4 by server %pI4\n", + &ic_myaddr, &b->iph.saddr); /* The DHCP indicated server address takes * precedence over the bootp header one if * they are different. @@ -1295,11 +1280,10 @@ static int __init ic_dynamic(void) return -1; } - printk("IP-Config: Got %s answer from %pI4, ", + pr_info("IP-Config: Got %s answer from %pI4, my address is %pI4\n", ((ic_got_reply & IC_RARP) ? "RARP" - : (ic_proto_enabled & IC_USE_DHCP) ? "DHCP" : "BOOTP"), - &ic_addrservaddr); - pr_cont("my address is %pI4\n", &ic_myaddr); + : (ic_proto_enabled & IC_USE_DHCP) ? "DHCP" : "BOOTP"), + &ic_addrservaddr, &ic_myaddr); return 0; } @@ -1426,7 +1410,7 @@ static int __init ip_auto_config(void) if (!ic_enable) return 0; - DBG(("IP-Config: Entered.\n")); + pr_debug("IP-Config: Entered.\n"); #ifdef IPCONFIG_DYNAMIC try_try_again: #endif @@ -1542,7 +1526,7 @@ static int __init ip_auto_config(void) pr_cont(", mtu=%d", ic_dev_mtu); for (i = 0; i < CONF_NAMESERVERS_MAX; i++) if (ic_nameservers[i] != NONE) { - pr_info(" nameserver%u=%pI4", + pr_cont(" nameserver%u=%pI4", i, &ic_nameservers[i]); break; } @@ -1585,7 +1569,7 @@ static int __init ic_proto_name(char *name) return 1; *v = 0; if (kstrtou8(client_id, 0, dhcp_client_identifier)) - DBG("DHCP: Invalid client identifier type\n"); + pr_debug("DHCP: Invalid client identifier type\n"); strncpy(dhcp_client_identifier + 1, v + 1, 251); *v = ','; } @@ -1644,7 +1628,7 @@ static int __init ip_auto_config_setup(char *addrs) if ((cp = strchr(ip, ':'))) *cp++ = '\0'; if (strlen(ip) > 0) { - DBG(("IP-Config: Parameter #%d: `%s'\n", num, ip)); + pr_debug("IP-Config: Parameter #%d: `%s'\n", num, ip); switch (num) { case 0: if ((ic_myaddr = in_aton(ip)) == ANY) @@ -1716,7 +1700,7 @@ static int __init vendor_class_identifier_setup(char *addrs) if (strlcpy(vendor_class_identifier, addrs, sizeof(vendor_class_identifier)) >= sizeof(vendor_class_identifier)) - pr_warn("DHCP: vendorclass too long, truncated to \"%s\"", + pr_warn("DHCP: vendorclass too long, truncated to \"%s\"\n", vendor_class_identifier); return 1; } diff --git a/net/ipv4/ipip.c b/net/ipv4/ipip.c index f34c31d..1f06729 100644 --- a/net/ipv4/ipip.c +++ b/net/ipv4/ipip.c @@ -103,7 +103,6 @@ #include <linux/tcp.h> #include <linux/udp.h> #include <linux/if_arp.h> -#include <linux/mroute.h> #include <linux/init.h> #include <linux/netfilter_ipv4.h> #include <linux/if_ether.h> diff --git a/net/ipv4/ipmr.c b/net/ipv4/ipmr.c index c3a3835..395e281 100644 --- a/net/ipv4/ipmr.c +++ b/net/ipv4/ipmr.c @@ -66,28 +66,7 @@ #include <net/netlink.h> #include <net/fib_rules.h> #include <linux/netconf.h> - -#if defined(CONFIG_IP_PIMSM_V1) || defined(CONFIG_IP_PIMSM_V2) -#define CONFIG_IP_PIMSM 1 -#endif - -struct mr_table { - struct list_head list; - possible_net_t net; - u32 id; - struct sock __rcu *mroute_sk; - struct timer_list ipmr_expire_timer; - struct list_head mfc_unres_queue; - struct list_head mfc_cache_array[MFC_LINES]; - struct vif_device vif_table[MAXVIFS]; - int maxvif; - atomic_t cache_resolve_queue_len; - bool mroute_do_assert; - bool mroute_do_pim; -#if defined(CONFIG_IP_PIMSM_V1) || defined(CONFIG_IP_PIMSM_V2) - int mroute_reg_vif_num; -#endif -}; +#include <net/nexthop.h> struct ipmr_rule { struct fib_rule common; @@ -103,11 +82,7 @@ struct ipmr_result { static DEFINE_RWLOCK(mrt_lock); -/* - * Multicast router control variables - */ - -#define VIF_EXISTS(_mrt, _idx) ((_mrt)->vif_table[_idx].dev != NULL) +/* Multicast router control variables */ /* Special spinlock for queue of unresolved entries */ static DEFINE_SPINLOCK(mfc_unres_lock); @@ -252,8 +227,8 @@ static int __net_init ipmr_rules_init(struct net *net) INIT_LIST_HEAD(&net->ipv4.mr_tables); mrt = ipmr_new_table(net, RT_TABLE_DEFAULT); - if (!mrt) { - err = -ENOMEM; + if (IS_ERR(mrt)) { + err = PTR_ERR(mrt); goto err1; } @@ -301,8 +276,13 @@ static int ipmr_fib_lookup(struct net *net, struct flowi4 *flp4, static int __net_init ipmr_rules_init(struct net *net) { - net->ipv4.mrt = ipmr_new_table(net, RT_TABLE_DEFAULT); - return net->ipv4.mrt ? 0 : -ENOMEM; + struct mr_table *mrt; + + mrt = ipmr_new_table(net, RT_TABLE_DEFAULT); + if (IS_ERR(mrt)) + return PTR_ERR(mrt); + net->ipv4.mrt = mrt; + return 0; } static void __net_exit ipmr_rules_exit(struct net *net) @@ -319,13 +299,17 @@ static struct mr_table *ipmr_new_table(struct net *net, u32 id) struct mr_table *mrt; unsigned int i; + /* "pimreg%u" should not exceed 16 bytes (IFNAMSIZ) */ + if (id != RT_TABLE_DEFAULT && id >= 1000000000) + return ERR_PTR(-EINVAL); + mrt = ipmr_get_table(net, id); if (mrt) return mrt; mrt = kzalloc(sizeof(*mrt), GFP_KERNEL); if (!mrt) - return NULL; + return ERR_PTR(-ENOMEM); write_pnet(&mrt->net, net); mrt->id = id; @@ -338,9 +322,7 @@ static struct mr_table *ipmr_new_table(struct net *net, u32 id) setup_timer(&mrt->ipmr_expire_timer, ipmr_expire_process, (unsigned long)mrt); -#ifdef CONFIG_IP_PIMSM mrt->mroute_reg_vif_num = -1; -#endif #ifdef CONFIG_IP_MROUTE_MULTIPLE_TABLES list_add_tail_rcu(&mrt->list, &net->ipv4.mr_tables); #endif @@ -387,8 +369,24 @@ static void ipmr_del_tunnel(struct net_device *dev, struct vifctl *v) } } -static -struct net_device *ipmr_new_tunnel(struct net *net, struct vifctl *v) +/* Initialize ipmr pimreg/tunnel in_device */ +static bool ipmr_init_vif_indev(const struct net_device *dev) +{ + struct in_device *in_dev; + + ASSERT_RTNL(); + + in_dev = __in_dev_get_rtnl(dev); + if (!in_dev) + return false; + ipv4_devconf_setall(in_dev); + neigh_parms_data_state_setall(in_dev->arp_parms); + IPV4_DEVCONF(in_dev->cnf, RP_FILTER) = 0; + + return true; +} + +static struct net_device *ipmr_new_tunnel(struct net *net, struct vifctl *v) { struct net_device *dev; @@ -399,7 +397,6 @@ struct net_device *ipmr_new_tunnel(struct net *net, struct vifctl *v) int err; struct ifreq ifr; struct ip_tunnel_parm p; - struct in_device *in_dev; memset(&p, 0, sizeof(p)); p.iph.daddr = v->vifc_rmt_addr.s_addr; @@ -424,15 +421,8 @@ struct net_device *ipmr_new_tunnel(struct net *net, struct vifctl *v) if (err == 0 && (dev = __dev_get_by_name(net, p.name)) != NULL) { dev->flags |= IFF_MULTICAST; - - in_dev = __in_dev_get_rtnl(dev); - if (!in_dev) + if (!ipmr_init_vif_indev(dev)) goto failure; - - ipv4_devconf_setall(in_dev); - neigh_parms_data_state_setall(in_dev->arp_parms); - IPV4_DEVCONF(in_dev->cnf, RP_FILTER) = 0; - if (dev_open(dev)) goto failure; dev_hold(dev); @@ -445,8 +435,7 @@ failure: return NULL; } -#ifdef CONFIG_IP_PIMSM - +#if defined(CONFIG_IP_PIMSM_V1) || defined(CONFIG_IP_PIMSM_V2) static netdev_tx_t reg_vif_xmit(struct sk_buff *skb, struct net_device *dev) { struct net *net = dev_net(dev); @@ -496,7 +485,6 @@ static void reg_vif_setup(struct net_device *dev) static struct net_device *ipmr_reg_vif(struct net *net, struct mr_table *mrt) { struct net_device *dev; - struct in_device *in_dev; char name[IFNAMSIZ]; if (mrt->id == RT_TABLE_DEFAULT) @@ -516,18 +504,8 @@ static struct net_device *ipmr_reg_vif(struct net *net, struct mr_table *mrt) return NULL; } - rcu_read_lock(); - in_dev = __in_dev_get_rcu(dev); - if (!in_dev) { - rcu_read_unlock(); + if (!ipmr_init_vif_indev(dev)) goto failure; - } - - ipv4_devconf_setall(in_dev); - neigh_parms_data_state_setall(in_dev->arp_parms); - IPV4_DEVCONF(in_dev->cnf, RP_FILTER) = 0; - rcu_read_unlock(); - if (dev_open(dev)) goto failure; @@ -539,13 +517,56 @@ failure: unregister_netdevice(dev); return NULL; } + +/* called with rcu_read_lock() */ +static int __pim_rcv(struct mr_table *mrt, struct sk_buff *skb, + unsigned int pimlen) +{ + struct net_device *reg_dev = NULL; + struct iphdr *encap; + + encap = (struct iphdr *)(skb_transport_header(skb) + pimlen); + /* Check that: + * a. packet is really sent to a multicast group + * b. packet is not a NULL-REGISTER + * c. packet is not truncated + */ + if (!ipv4_is_multicast(encap->daddr) || + encap->tot_len == 0 || + ntohs(encap->tot_len) + pimlen > skb->len) + return 1; + + read_lock(&mrt_lock); + if (mrt->mroute_reg_vif_num >= 0) + reg_dev = mrt->vif_table[mrt->mroute_reg_vif_num].dev; + read_unlock(&mrt_lock); + + if (!reg_dev) + return 1; + + skb->mac_header = skb->network_header; + skb_pull(skb, (u8 *)encap - skb->data); + skb_reset_network_header(skb); + skb->protocol = htons(ETH_P_IP); + skb->ip_summed = CHECKSUM_NONE; + + skb_tunnel_rx(skb, reg_dev, dev_net(reg_dev)); + + netif_rx(skb); + + return NET_RX_SUCCESS; +} +#else +static struct net_device *ipmr_reg_vif(struct net *net, struct mr_table *mrt) +{ + return NULL; +} #endif /** * vif_delete - Delete a VIF entry * @notify: Set to 1, if the caller is a notifier_call */ - static int vif_delete(struct mr_table *mrt, int vifi, int notify, struct list_head *head) { @@ -567,10 +588,8 @@ static int vif_delete(struct mr_table *mrt, int vifi, int notify, return -EADDRNOTAVAIL; } -#ifdef CONFIG_IP_PIMSM if (vifi == mrt->mroute_reg_vif_num) mrt->mroute_reg_vif_num = -1; -#endif if (vifi + 1 == mrt->maxvif) { int tmp; @@ -617,7 +636,6 @@ static inline void ipmr_cache_free(struct mfc_cache *c) /* Destroy an unresolved cache entry, killing queued skbs * and reporting error to netlink readers. */ - static void ipmr_destroy_unres(struct mr_table *mrt, struct mfc_cache *c) { struct net *net = read_pnet(&mrt->net); @@ -645,9 +663,7 @@ static void ipmr_destroy_unres(struct mr_table *mrt, struct mfc_cache *c) ipmr_cache_free(c); } - /* Timer process for the unresolved queue. */ - static void ipmr_expire_process(unsigned long arg) { struct mr_table *mrt = (struct mr_table *)arg; @@ -687,7 +703,6 @@ out: } /* Fill oifs list. It is called under write locked mrt_lock. */ - static void ipmr_update_thresholds(struct mr_table *mrt, struct mfc_cache *cache, unsigned char *ttls) { @@ -723,10 +738,10 @@ static int vif_add(struct net *net, struct mr_table *mrt, return -EADDRINUSE; switch (vifc->vifc_flags) { -#ifdef CONFIG_IP_PIMSM case VIFF_REGISTER: - /* - * Special Purpose VIF in PIM + if (!ipmr_pimsm_enabled()) + return -EINVAL; + /* Special Purpose VIF in PIM * All the packets will be sent to the daemon */ if (mrt->mroute_reg_vif_num >= 0) @@ -741,7 +756,6 @@ static int vif_add(struct net *net, struct mr_table *mrt, return err; } break; -#endif case VIFF_TUNNEL: dev = ipmr_new_tunnel(net, vifc); if (!dev) @@ -753,7 +767,6 @@ static int vif_add(struct net *net, struct mr_table *mrt, return err; } break; - case VIFF_USE_IFINDEX: case 0: if (vifc->vifc_flags == VIFF_USE_IFINDEX) { @@ -807,10 +820,8 @@ static int vif_add(struct net *net, struct mr_table *mrt, /* And finish update writing critical data */ write_lock_bh(&mrt_lock); v->dev = dev; -#ifdef CONFIG_IP_PIMSM if (v->flags & VIFF_REGISTER) mrt->mroute_reg_vif_num = vifi; -#endif if (vifi+1 > mrt->maxvif) mrt->maxvif = vifi+1; write_unlock_bh(&mrt_lock); @@ -875,9 +886,7 @@ skip: return ipmr_cache_find_any_parent(mrt, vifi); } -/* - * Allocate a multicast cache entry - */ +/* Allocate a multicast cache entry */ static struct mfc_cache *ipmr_cache_alloc(void) { struct mfc_cache *c = kmem_cache_zalloc(mrt_cachep, GFP_KERNEL); @@ -898,10 +907,7 @@ static struct mfc_cache *ipmr_cache_alloc_unres(void) return c; } -/* - * A cache entry has gone into a resolved state from queued - */ - +/* A cache entry has gone into a resolved state from queued */ static void ipmr_cache_resolve(struct net *net, struct mr_table *mrt, struct mfc_cache *uc, struct mfc_cache *c) { @@ -909,7 +915,6 @@ static void ipmr_cache_resolve(struct net *net, struct mr_table *mrt, struct nlmsgerr *e; /* Play the pending entries through our router */ - while ((skb = __skb_dequeue(&uc->mfc_un.unres.unresolved))) { if (ip_hdr(skb)->version == 0) { struct nlmsghdr *nlh = (struct nlmsghdr *)skb_pull(skb, sizeof(struct iphdr)); @@ -933,34 +938,29 @@ static void ipmr_cache_resolve(struct net *net, struct mr_table *mrt, } } -/* - * Bounce a cache query up to mrouted. We could use netlink for this but mrouted - * expects the following bizarre scheme. +/* Bounce a cache query up to mrouted. We could use netlink for this but mrouted + * expects the following bizarre scheme. * - * Called under mrt_lock. + * Called under mrt_lock. */ - static int ipmr_cache_report(struct mr_table *mrt, struct sk_buff *pkt, vifi_t vifi, int assert) { - struct sk_buff *skb; const int ihl = ip_hdrlen(pkt); + struct sock *mroute_sk; struct igmphdr *igmp; struct igmpmsg *msg; - struct sock *mroute_sk; + struct sk_buff *skb; int ret; -#ifdef CONFIG_IP_PIMSM if (assert == IGMPMSG_WHOLEPKT) skb = skb_realloc_headroom(pkt, sizeof(struct iphdr)); else -#endif skb = alloc_skb(128, GFP_ATOMIC); if (!skb) return -ENOBUFS; -#ifdef CONFIG_IP_PIMSM if (assert == IGMPMSG_WHOLEPKT) { /* Ugly, but we have no choice with this interface. * Duplicate old header, fix ihl, length etc. @@ -978,28 +978,23 @@ static int ipmr_cache_report(struct mr_table *mrt, ip_hdr(skb)->ihl = sizeof(struct iphdr) >> 2; ip_hdr(skb)->tot_len = htons(ntohs(ip_hdr(pkt)->tot_len) + sizeof(struct iphdr)); - } else -#endif - { - - /* Copy the IP header */ - - skb_set_network_header(skb, skb->len); - skb_put(skb, ihl); - skb_copy_to_linear_data(skb, pkt->data, ihl); - ip_hdr(skb)->protocol = 0; /* Flag to the kernel this is a route add */ - msg = (struct igmpmsg *)skb_network_header(skb); - msg->im_vif = vifi; - skb_dst_set(skb, dst_clone(skb_dst(pkt))); - - /* Add our header */ - - igmp = (struct igmphdr *)skb_put(skb, sizeof(struct igmphdr)); - igmp->type = - msg->im_msgtype = assert; - igmp->code = 0; - ip_hdr(skb)->tot_len = htons(skb->len); /* Fix the length */ - skb->transport_header = skb->network_header; + } else { + /* Copy the IP header */ + skb_set_network_header(skb, skb->len); + skb_put(skb, ihl); + skb_copy_to_linear_data(skb, pkt->data, ihl); + /* Flag to the kernel this is a route add */ + ip_hdr(skb)->protocol = 0; + msg = (struct igmpmsg *)skb_network_header(skb); + msg->im_vif = vifi; + skb_dst_set(skb, dst_clone(skb_dst(pkt))); + /* Add our header */ + igmp = (struct igmphdr *)skb_put(skb, sizeof(struct igmphdr)); + igmp->type = assert; + msg->im_msgtype = assert; + igmp->code = 0; + ip_hdr(skb)->tot_len = htons(skb->len); /* Fix the length */ + skb->transport_header = skb->network_header; } rcu_read_lock(); @@ -1011,7 +1006,6 @@ static int ipmr_cache_report(struct mr_table *mrt, } /* Deliver to mrouted */ - ret = sock_queue_rcv_skb(mroute_sk, skb); rcu_read_unlock(); if (ret < 0) { @@ -1022,12 +1016,9 @@ static int ipmr_cache_report(struct mr_table *mrt, return ret; } -/* - * Queue a packet for resolution. It gets locked cache entry! - */ - -static int -ipmr_cache_unresolved(struct mr_table *mrt, vifi_t vifi, struct sk_buff *skb) +/* Queue a packet for resolution. It gets locked cache entry! */ +static int ipmr_cache_unresolved(struct mr_table *mrt, vifi_t vifi, + struct sk_buff *skb) { bool found = false; int err; @@ -1045,7 +1036,6 @@ ipmr_cache_unresolved(struct mr_table *mrt, vifi_t vifi, struct sk_buff *skb) if (!found) { /* Create a new entry if allowable */ - if (atomic_read(&mrt->cache_resolve_queue_len) >= 10 || (c = ipmr_cache_alloc_unres()) == NULL) { spin_unlock_bh(&mfc_unres_lock); @@ -1055,13 +1045,11 @@ ipmr_cache_unresolved(struct mr_table *mrt, vifi_t vifi, struct sk_buff *skb) } /* Fill in the new cache entry */ - c->mfc_parent = -1; c->mfc_origin = iph->saddr; c->mfc_mcastgrp = iph->daddr; /* Reflect first query at mrouted. */ - err = ipmr_cache_report(mrt, skb, vifi, IGMPMSG_NOCACHE); if (err < 0) { /* If the report failed throw the cache entry @@ -1083,7 +1071,6 @@ ipmr_cache_unresolved(struct mr_table *mrt, vifi_t vifi, struct sk_buff *skb) } /* See if we can append the packet */ - if (c->mfc_un.unres.unresolved.qlen > 3) { kfree_skb(skb); err = -ENOBUFS; @@ -1096,9 +1083,7 @@ ipmr_cache_unresolved(struct mr_table *mrt, vifi_t vifi, struct sk_buff *skb) return err; } -/* - * MFC cache manipulation by user space mroute daemon - */ +/* MFC cache manipulation by user space mroute daemon */ static int ipmr_mfc_delete(struct mr_table *mrt, struct mfcctl *mfc, int parent) { @@ -1169,9 +1154,8 @@ static int ipmr_mfc_add(struct net *net, struct mr_table *mrt, list_add_rcu(&c->list, &mrt->mfc_cache_array[line]); - /* - * Check to see if we resolved a queued list. If so we - * need to send on the frames and tidy up. + /* Check to see if we resolved a queued list. If so we + * need to send on the frames and tidy up. */ found = false; spin_lock_bh(&mfc_unres_lock); @@ -1196,10 +1180,7 @@ static int ipmr_mfc_add(struct net *net, struct mr_table *mrt, return 0; } -/* - * Close the multicast socket, and clear the vif tables etc - */ - +/* Close the multicast socket, and clear the vif tables etc */ static void mroute_clean_tables(struct mr_table *mrt, bool all) { int i; @@ -1207,7 +1188,6 @@ static void mroute_clean_tables(struct mr_table *mrt, bool all) struct mfc_cache *c, *next; /* Shut down all active vif entries */ - for (i = 0; i < mrt->maxvif; i++) { if (!all && (mrt->vif_table[i].flags & VIFF_STATIC)) continue; @@ -1216,7 +1196,6 @@ static void mroute_clean_tables(struct mr_table *mrt, bool all) unregister_netdevice_many(&list); /* Wipe the cache */ - for (i = 0; i < MFC_LINES; i++) { list_for_each_entry_safe(c, next, &mrt->mfc_cache_array[i], list) { if (!all && (c->mfc_flags & MFC_STATIC)) @@ -1260,44 +1239,52 @@ static void mrtsock_destruct(struct sock *sk) rtnl_unlock(); } -/* - * Socket options and virtual interface manipulation. The whole - * virtual interface system is a complete heap, but unfortunately - * that's how BSD mrouted happens to think. Maybe one day with a proper - * MOSPF/PIM router set up we can clean this up. +/* Socket options and virtual interface manipulation. The whole + * virtual interface system is a complete heap, but unfortunately + * that's how BSD mrouted happens to think. Maybe one day with a proper + * MOSPF/PIM router set up we can clean this up. */ -int ip_mroute_setsockopt(struct sock *sk, int optname, char __user *optval, unsigned int optlen) +int ip_mroute_setsockopt(struct sock *sk, int optname, char __user *optval, + unsigned int optlen) { - int ret, parent = 0; - struct vifctl vif; - struct mfcctl mfc; struct net *net = sock_net(sk); + int val, ret = 0, parent = 0; struct mr_table *mrt; + struct vifctl vif; + struct mfcctl mfc; + u32 uval; + /* There's one exception to the lock - MRT_DONE which needs to unlock */ + rtnl_lock(); if (sk->sk_type != SOCK_RAW || - inet_sk(sk)->inet_num != IPPROTO_IGMP) - return -EOPNOTSUPP; + inet_sk(sk)->inet_num != IPPROTO_IGMP) { + ret = -EOPNOTSUPP; + goto out_unlock; + } mrt = ipmr_get_table(net, raw_sk(sk)->ipmr_table ? : RT_TABLE_DEFAULT); - if (!mrt) - return -ENOENT; - + if (!mrt) { + ret = -ENOENT; + goto out_unlock; + } if (optname != MRT_INIT) { if (sk != rcu_access_pointer(mrt->mroute_sk) && - !ns_capable(net->user_ns, CAP_NET_ADMIN)) - return -EACCES; + !ns_capable(net->user_ns, CAP_NET_ADMIN)) { + ret = -EACCES; + goto out_unlock; + } } switch (optname) { case MRT_INIT: - if (optlen != sizeof(int)) - return -EINVAL; - - rtnl_lock(); + if (optlen != sizeof(int)) { + ret = -EINVAL; + break; + } if (rtnl_dereference(mrt->mroute_sk)) { - rtnl_unlock(); - return -EADDRINUSE; + ret = -EADDRINUSE; + break; } ret = ip_ra_control(sk, 1, mrtsock_destruct); @@ -1308,129 +1295,133 @@ int ip_mroute_setsockopt(struct sock *sk, int optname, char __user *optval, unsi NETCONFA_IFINDEX_ALL, net->ipv4.devconf_all); } - rtnl_unlock(); - return ret; + break; case MRT_DONE: - if (sk != rcu_access_pointer(mrt->mroute_sk)) - return -EACCES; - return ip_ra_control(sk, 0, NULL); + if (sk != rcu_access_pointer(mrt->mroute_sk)) { + ret = -EACCES; + } else { + /* We need to unlock here because mrtsock_destruct takes + * care of rtnl itself and we can't change that due to + * the IP_ROUTER_ALERT setsockopt which runs without it. + */ + rtnl_unlock(); + ret = ip_ra_control(sk, 0, NULL); + goto out; + } + break; case MRT_ADD_VIF: case MRT_DEL_VIF: - if (optlen != sizeof(vif)) - return -EINVAL; - if (copy_from_user(&vif, optval, sizeof(vif))) - return -EFAULT; - if (vif.vifc_vifi >= MAXVIFS) - return -ENFILE; - rtnl_lock(); + if (optlen != sizeof(vif)) { + ret = -EINVAL; + break; + } + if (copy_from_user(&vif, optval, sizeof(vif))) { + ret = -EFAULT; + break; + } + if (vif.vifc_vifi >= MAXVIFS) { + ret = -ENFILE; + break; + } if (optname == MRT_ADD_VIF) { ret = vif_add(net, mrt, &vif, sk == rtnl_dereference(mrt->mroute_sk)); } else { ret = vif_delete(mrt, vif.vifc_vifi, 0, NULL); } - rtnl_unlock(); - return ret; - - /* - * Manipulate the forwarding caches. These live - * in a sort of kernel/user symbiosis. - */ + break; + /* Manipulate the forwarding caches. These live + * in a sort of kernel/user symbiosis. + */ case MRT_ADD_MFC: case MRT_DEL_MFC: parent = -1; case MRT_ADD_MFC_PROXY: case MRT_DEL_MFC_PROXY: - if (optlen != sizeof(mfc)) - return -EINVAL; - if (copy_from_user(&mfc, optval, sizeof(mfc))) - return -EFAULT; + if (optlen != sizeof(mfc)) { + ret = -EINVAL; + break; + } + if (copy_from_user(&mfc, optval, sizeof(mfc))) { + ret = -EFAULT; + break; + } if (parent == 0) parent = mfc.mfcc_parent; - rtnl_lock(); if (optname == MRT_DEL_MFC || optname == MRT_DEL_MFC_PROXY) ret = ipmr_mfc_delete(mrt, &mfc, parent); else ret = ipmr_mfc_add(net, mrt, &mfc, sk == rtnl_dereference(mrt->mroute_sk), parent); - rtnl_unlock(); - return ret; - /* - * Control PIM assert. - */ + break; + /* Control PIM assert. */ case MRT_ASSERT: - { - int v; - if (optlen != sizeof(v)) - return -EINVAL; - if (get_user(v, (int __user *)optval)) - return -EFAULT; - mrt->mroute_do_assert = v; - return 0; - } -#ifdef CONFIG_IP_PIMSM + if (optlen != sizeof(val)) { + ret = -EINVAL; + break; + } + if (get_user(val, (int __user *)optval)) { + ret = -EFAULT; + break; + } + mrt->mroute_do_assert = val; + break; case MRT_PIM: - { - int v; - - if (optlen != sizeof(v)) - return -EINVAL; - if (get_user(v, (int __user *)optval)) - return -EFAULT; - v = !!v; + if (!ipmr_pimsm_enabled()) { + ret = -ENOPROTOOPT; + break; + } + if (optlen != sizeof(val)) { + ret = -EINVAL; + break; + } + if (get_user(val, (int __user *)optval)) { + ret = -EFAULT; + break; + } - rtnl_lock(); - ret = 0; - if (v != mrt->mroute_do_pim) { - mrt->mroute_do_pim = v; - mrt->mroute_do_assert = v; + val = !!val; + if (val != mrt->mroute_do_pim) { + mrt->mroute_do_pim = val; + mrt->mroute_do_assert = val; } - rtnl_unlock(); - return ret; - } -#endif -#ifdef CONFIG_IP_MROUTE_MULTIPLE_TABLES + break; case MRT_TABLE: - { - u32 v; - - if (optlen != sizeof(u32)) - return -EINVAL; - if (get_user(v, (u32 __user *)optval)) - return -EFAULT; - - /* "pimreg%u" should not exceed 16 bytes (IFNAMSIZ) */ - if (v != RT_TABLE_DEFAULT && v >= 1000000000) - return -EINVAL; + if (!IS_BUILTIN(CONFIG_IP_MROUTE_MULTIPLE_TABLES)) { + ret = -ENOPROTOOPT; + break; + } + if (optlen != sizeof(uval)) { + ret = -EINVAL; + break; + } + if (get_user(uval, (u32 __user *)optval)) { + ret = -EFAULT; + break; + } - rtnl_lock(); - ret = 0; if (sk == rtnl_dereference(mrt->mroute_sk)) { ret = -EBUSY; } else { - if (!ipmr_new_table(net, v)) - ret = -ENOMEM; + mrt = ipmr_new_table(net, uval); + if (IS_ERR(mrt)) + ret = PTR_ERR(mrt); else - raw_sk(sk)->ipmr_table = v; + raw_sk(sk)->ipmr_table = uval; } - rtnl_unlock(); - return ret; - } -#endif - /* - * Spurious command, or MRT_VERSION which you cannot - * set. - */ + break; + /* Spurious command, or MRT_VERSION which you cannot set. */ default: - return -ENOPROTOOPT; + ret = -ENOPROTOOPT; } +out_unlock: + rtnl_unlock(); +out: + return ret; } -/* - * Getsock opt support for the multicast routing system. - */ - +/* Getsock opt support for the multicast routing system. */ int ip_mroute_getsockopt(struct sock *sk, int optname, char __user *optval, int __user *optlen) { int olr; @@ -1446,39 +1437,35 @@ int ip_mroute_getsockopt(struct sock *sk, int optname, char __user *optval, int if (!mrt) return -ENOENT; - if (optname != MRT_VERSION && -#ifdef CONFIG_IP_PIMSM - optname != MRT_PIM && -#endif - optname != MRT_ASSERT) + switch (optname) { + case MRT_VERSION: + val = 0x0305; + break; + case MRT_PIM: + if (!ipmr_pimsm_enabled()) + return -ENOPROTOOPT; + val = mrt->mroute_do_pim; + break; + case MRT_ASSERT: + val = mrt->mroute_do_assert; + break; + default: return -ENOPROTOOPT; + } if (get_user(olr, optlen)) return -EFAULT; - olr = min_t(unsigned int, olr, sizeof(int)); if (olr < 0) return -EINVAL; - if (put_user(olr, optlen)) return -EFAULT; - if (optname == MRT_VERSION) - val = 0x0305; -#ifdef CONFIG_IP_PIMSM - else if (optname == MRT_PIM) - val = mrt->mroute_do_pim; -#endif - else - val = mrt->mroute_do_assert; if (copy_to_user(optval, &val, olr)) return -EFAULT; return 0; } -/* - * The IP multicast ioctl support routines. - */ - +/* The IP multicast ioctl support routines. */ int ipmr_ioctl(struct sock *sk, int cmd, void __user *arg) { struct sioc_sg_req sr; @@ -1611,7 +1598,6 @@ int ipmr_compat_ioctl(struct sock *sk, unsigned int cmd, void __user *arg) } #endif - static int ipmr_device_event(struct notifier_block *this, unsigned long event, void *ptr) { struct net_device *dev = netdev_notifier_info_to_dev(ptr); @@ -1633,17 +1619,14 @@ static int ipmr_device_event(struct notifier_block *this, unsigned long event, v return NOTIFY_DONE; } - static struct notifier_block ip_mr_notifier = { .notifier_call = ipmr_device_event, }; -/* - * Encapsulate a packet by attaching a valid IPIP header to it. - * This avoids tunnel drivers and other mess and gives us the speed so - * important for multicast video. +/* Encapsulate a packet by attaching a valid IPIP header to it. + * This avoids tunnel drivers and other mess and gives us the speed so + * important for multicast video. */ - static void ip_encap(struct net *net, struct sk_buff *skb, __be32 saddr, __be32 daddr) { @@ -1685,9 +1668,7 @@ static inline int ipmr_forward_finish(struct net *net, struct sock *sk, return dst_output(net, sk, skb); } -/* - * Processing handlers for ipmr_forward - */ +/* Processing handlers for ipmr_forward */ static void ipmr_queue_xmit(struct net *net, struct mr_table *mrt, struct sk_buff *skb, struct mfc_cache *c, int vifi) @@ -1702,7 +1683,6 @@ static void ipmr_queue_xmit(struct net *net, struct mr_table *mrt, if (!vif->dev) goto out_free; -#ifdef CONFIG_IP_PIMSM if (vif->flags & VIFF_REGISTER) { vif->pkt_out++; vif->bytes_out += skb->len; @@ -1711,7 +1691,6 @@ static void ipmr_queue_xmit(struct net *net, struct mr_table *mrt, ipmr_cache_report(mrt, skb, vifi, IGMPMSG_WHOLEPKT); goto out_free; } -#endif if (vif->flags & VIFF_TUNNEL) { rt = ip_route_output_ports(net, &fl4, NULL, @@ -1738,7 +1717,6 @@ static void ipmr_queue_xmit(struct net *net, struct mr_table *mrt, * allow to send ICMP, so that packets will disappear * to blackhole. */ - IP_INC_STATS(net, IPSTATS_MIB_FRAGFAILS); ip_rt_put(rt); goto out_free; @@ -1770,8 +1748,7 @@ static void ipmr_queue_xmit(struct net *net, struct mr_table *mrt, IPCB(skb)->flags |= IPSKB_FORWARDED; - /* - * RFC1584 teaches, that DVMRP/PIM router must deliver packets locally + /* RFC1584 teaches, that DVMRP/PIM router must deliver packets locally * not only before forwarding, but after forwarding on all output * interfaces. It is clear, if mrouter runs a multicasting * program, it should receive packets not depending to what interface @@ -1802,7 +1779,6 @@ static int ipmr_find_vif(struct mr_table *mrt, struct net_device *dev) } /* "local" means that we should preserve one skb (for local delivery) */ - static void ip_mr_forward(struct net *net, struct mr_table *mrt, struct sk_buff *skb, struct mfc_cache *cache, int local) @@ -1827,9 +1803,7 @@ static void ip_mr_forward(struct net *net, struct mr_table *mrt, goto forward; } - /* - * Wrong interface: drop packet and (maybe) send PIM assert. - */ + /* Wrong interface: drop packet and (maybe) send PIM assert. */ if (mrt->vif_table[vif].dev != skb->dev) { if (rt_is_output_route(skb_rtable(skb))) { /* It is our own packet, looped back. @@ -1868,9 +1842,7 @@ forward: mrt->vif_table[vif].pkt_in++; mrt->vif_table[vif].bytes_in += skb->len; - /* - * Forward the frame - */ + /* Forward the frame */ if (cache->mfc_origin == htonl(INADDR_ANY) && cache->mfc_mcastgrp == htonl(INADDR_ANY)) { if (true_vifi >= 0 && @@ -1944,11 +1916,9 @@ static struct mr_table *ipmr_rt_fib_lookup(struct net *net, struct sk_buff *skb) return mrt; } -/* - * Multicast packets for forwarding arrive here - * Called with rcu_read_lock(); +/* Multicast packets for forwarding arrive here + * Called with rcu_read_lock(); */ - int ip_mr_input(struct sk_buff *skb) { struct mfc_cache *cache; @@ -1999,9 +1969,7 @@ int ip_mr_input(struct sk_buff *skb) vif); } - /* - * No usable cache entry - */ + /* No usable cache entry */ if (!cache) { int vif; @@ -2042,53 +2010,8 @@ dont_forward: return 0; } -#ifdef CONFIG_IP_PIMSM -/* called with rcu_read_lock() */ -static int __pim_rcv(struct mr_table *mrt, struct sk_buff *skb, - unsigned int pimlen) -{ - struct net_device *reg_dev = NULL; - struct iphdr *encap; - - encap = (struct iphdr *)(skb_transport_header(skb) + pimlen); - /* - * Check that: - * a. packet is really sent to a multicast group - * b. packet is not a NULL-REGISTER - * c. packet is not truncated - */ - if (!ipv4_is_multicast(encap->daddr) || - encap->tot_len == 0 || - ntohs(encap->tot_len) + pimlen > skb->len) - return 1; - - read_lock(&mrt_lock); - if (mrt->mroute_reg_vif_num >= 0) - reg_dev = mrt->vif_table[mrt->mroute_reg_vif_num].dev; - read_unlock(&mrt_lock); - - if (!reg_dev) - return 1; - - skb->mac_header = skb->network_header; - skb_pull(skb, (u8 *)encap - skb->data); - skb_reset_network_header(skb); - skb->protocol = htons(ETH_P_IP); - skb->ip_summed = CHECKSUM_NONE; - - skb_tunnel_rx(skb, reg_dev, dev_net(reg_dev)); - - netif_rx(skb); - - return NET_RX_SUCCESS; -} -#endif - #ifdef CONFIG_IP_PIMSM_V1 -/* - * Handle IGMP messages of PIMv1 - */ - +/* Handle IGMP messages of PIMv1 */ int pim_rcv_v1(struct sk_buff *skb) { struct igmphdr *pim; @@ -2249,8 +2172,6 @@ int ipmr_get_route(struct net *net, struct sk_buff *skb, } read_lock(&mrt_lock); - if (!nowait && (rtm->rtm_flags & RTM_F_NOTIFY)) - cache->mfc_flags |= MFC_NOTIFY; err = __ipmr_fill_mroute(mrt, skb, cache, rtm); read_unlock(&mrt_lock); rcu_read_unlock(); @@ -2412,10 +2333,133 @@ done: return skb->len; } +static const struct nla_policy rtm_ipmr_policy[RTA_MAX + 1] = { + [RTA_SRC] = { .type = NLA_U32 }, + [RTA_DST] = { .type = NLA_U32 }, + [RTA_IIF] = { .type = NLA_U32 }, + [RTA_TABLE] = { .type = NLA_U32 }, + [RTA_MULTIPATH] = { .len = sizeof(struct rtnexthop) }, +}; + +static bool ipmr_rtm_validate_proto(unsigned char rtm_protocol) +{ + switch (rtm_protocol) { + case RTPROT_STATIC: + case RTPROT_MROUTED: + return true; + } + return false; +} + +static int ipmr_nla_get_ttls(const struct nlattr *nla, struct mfcctl *mfcc) +{ + struct rtnexthop *rtnh = nla_data(nla); + int remaining = nla_len(nla), vifi = 0; + + while (rtnh_ok(rtnh, remaining)) { + mfcc->mfcc_ttls[vifi] = rtnh->rtnh_hops; + if (++vifi == MAXVIFS) + break; + rtnh = rtnh_next(rtnh, &remaining); + } + + return remaining > 0 ? -EINVAL : vifi; +} + +/* returns < 0 on error, 0 for ADD_MFC and 1 for ADD_MFC_PROXY */ +static int rtm_to_ipmr_mfcc(struct net *net, struct nlmsghdr *nlh, + struct mfcctl *mfcc, int *mrtsock, + struct mr_table **mrtret) +{ + struct net_device *dev = NULL; + u32 tblid = RT_TABLE_DEFAULT; + struct mr_table *mrt; + struct nlattr *attr; + struct rtmsg *rtm; + int ret, rem; + + ret = nlmsg_validate(nlh, sizeof(*rtm), RTA_MAX, rtm_ipmr_policy); + if (ret < 0) + goto out; + rtm = nlmsg_data(nlh); + + ret = -EINVAL; + if (rtm->rtm_family != RTNL_FAMILY_IPMR || rtm->rtm_dst_len != 32 || + rtm->rtm_type != RTN_MULTICAST || + rtm->rtm_scope != RT_SCOPE_UNIVERSE || + !ipmr_rtm_validate_proto(rtm->rtm_protocol)) + goto out; + + memset(mfcc, 0, sizeof(*mfcc)); + mfcc->mfcc_parent = -1; + ret = 0; + nlmsg_for_each_attr(attr, nlh, sizeof(struct rtmsg), rem) { + switch (nla_type(attr)) { + case RTA_SRC: + mfcc->mfcc_origin.s_addr = nla_get_be32(attr); + break; + case RTA_DST: + mfcc->mfcc_mcastgrp.s_addr = nla_get_be32(attr); + break; + case RTA_IIF: + dev = __dev_get_by_index(net, nla_get_u32(attr)); + if (!dev) { + ret = -ENODEV; + goto out; + } + break; + case RTA_MULTIPATH: + if (ipmr_nla_get_ttls(attr, mfcc) < 0) { + ret = -EINVAL; + goto out; + } + break; + case RTA_PREFSRC: + ret = 1; + break; + case RTA_TABLE: + tblid = nla_get_u32(attr); + break; + } + } + mrt = ipmr_get_table(net, tblid); + if (!mrt) { + ret = -ENOENT; + goto out; + } + *mrtret = mrt; + *mrtsock = rtm->rtm_protocol == RTPROT_MROUTED ? 1 : 0; + if (dev) + mfcc->mfcc_parent = ipmr_find_vif(mrt, dev); + +out: + return ret; +} + +/* takes care of both newroute and delroute */ +static int ipmr_rtm_route(struct sk_buff *skb, struct nlmsghdr *nlh) +{ + struct net *net = sock_net(skb->sk); + int ret, mrtsock, parent; + struct mr_table *tbl; + struct mfcctl mfcc; + + mrtsock = 0; + tbl = NULL; + ret = rtm_to_ipmr_mfcc(net, nlh, &mfcc, &mrtsock, &tbl); + if (ret < 0) + return ret; + + parent = ret ? mfcc.mfcc_parent : -1; + if (nlh->nlmsg_type == RTM_NEWROUTE) + return ipmr_mfc_add(net, tbl, &mfcc, mrtsock, parent); + else + return ipmr_mfc_delete(tbl, &mfcc, parent); +} + #ifdef CONFIG_PROC_FS -/* - * The /proc interfaces to multicast routing : - * /proc/net/ip_mr_cache & /proc/net/ip_mr_vif +/* The /proc interfaces to multicast routing : + * /proc/net/ip_mr_cache & /proc/net/ip_mr_vif */ struct ipmr_vif_iter { struct seq_net_private p; @@ -2699,10 +2743,7 @@ static const struct net_protocol pim_protocol = { }; #endif - -/* - * Setup for IP multicast routing - */ +/* Setup for IP multicast routing */ static int __net_init ipmr_net_init(struct net *net) { int err; @@ -2752,8 +2793,6 @@ int __init ip_mr_init(void) sizeof(struct mfc_cache), 0, SLAB_HWCACHE_ALIGN | SLAB_PANIC, NULL); - if (!mrt_cachep) - return -ENOMEM; err = register_pernet_subsys(&ipmr_net_ops); if (err) @@ -2771,6 +2810,10 @@ int __init ip_mr_init(void) #endif rtnl_register(RTNL_FAMILY_IPMR, RTM_GETROUTE, NULL, ipmr_rtm_dumproute, NULL); + rtnl_register(RTNL_FAMILY_IPMR, RTM_NEWROUTE, + ipmr_rtm_route, NULL, NULL); + rtnl_register(RTNL_FAMILY_IPMR, RTM_DELROUTE, + ipmr_rtm_route, NULL, NULL); return 0; #ifdef CONFIG_IP_PIMSM_V2 diff --git a/net/ipv4/netfilter/arp_tables.c b/net/ipv4/netfilter/arp_tables.c index 11dccba..b488cac 100644 --- a/net/ipv4/netfilter/arp_tables.c +++ b/net/ipv4/netfilter/arp_tables.c @@ -38,13 +38,13 @@ MODULE_DESCRIPTION("arptables core"); /*#define DEBUG_ARP_TABLES_USER*/ #ifdef DEBUG_ARP_TABLES -#define dprintf(format, args...) printk(format , ## args) +#define dprintf(format, args...) pr_debug(format, ## args) #else #define dprintf(format, args...) #endif #ifdef DEBUG_ARP_TABLES_USER -#define duprintf(format, args...) printk(format , ## args) +#define duprintf(format, args...) pr_debug(format, ## args) #else #define duprintf(format, args...) #endif @@ -1905,7 +1905,7 @@ static int __init arp_tables_init(void) if (ret < 0) goto err4; - printk(KERN_INFO "arp_tables: (C) 2002 David S. Miller\n"); + pr_info("arp_tables: (C) 2002 David S. Miller\n"); return 0; err4: diff --git a/net/ipv4/netfilter/nf_conntrack_l3proto_ipv4.c b/net/ipv4/netfilter/nf_conntrack_l3proto_ipv4.c index 461ca92..e3c46e8 100644 --- a/net/ipv4/netfilter/nf_conntrack_l3proto_ipv4.c +++ b/net/ipv4/netfilter/nf_conntrack_l3proto_ipv4.c @@ -451,7 +451,7 @@ static int __init nf_conntrack_l3proto_ipv4_init(void) ret = nf_register_sockopt(&so_getorigdst); if (ret < 0) { - printk(KERN_ERR "Unable to register netfilter socket option\n"); + pr_err("Unable to register netfilter socket option\n"); return ret; } diff --git a/net/ipv4/netfilter/nf_nat_snmp_basic.c b/net/ipv4/netfilter/nf_nat_snmp_basic.c index ddb894a..c9b52c3 100644 --- a/net/ipv4/netfilter/nf_nat_snmp_basic.c +++ b/net/ipv4/netfilter/nf_nat_snmp_basic.c @@ -1048,7 +1048,7 @@ static int snmp_parse_mangle(unsigned char *msg, if (!asn1_uint_decode (&ctx, end, &vers)) return 0; if (debug > 1) - printk(KERN_DEBUG "bsalg: snmp version: %u\n", vers + 1); + pr_debug("bsalg: snmp version: %u\n", vers + 1); if (vers > 1) return 1; @@ -1064,10 +1064,10 @@ static int snmp_parse_mangle(unsigned char *msg, if (debug > 1) { unsigned int i; - printk(KERN_DEBUG "bsalg: community: "); + pr_debug("bsalg: community: "); for (i = 0; i < comm.len; i++) - printk("%c", comm.data[i]); - printk("\n"); + pr_cont("%c", comm.data[i]); + pr_cont("\n"); } kfree(comm.data); @@ -1091,9 +1091,9 @@ static int snmp_parse_mangle(unsigned char *msg, }; if (pdutype > SNMP_PDU_TRAP2) - printk(KERN_DEBUG "bsalg: bad pdu type %u\n", pdutype); + pr_debug("bsalg: bad pdu type %u\n", pdutype); else - printk(KERN_DEBUG "bsalg: pdu: %s\n", pdus[pdutype]); + pr_debug("bsalg: pdu: %s\n", pdus[pdutype]); } if (pdutype != SNMP_PDU_RESPONSE && pdutype != SNMP_PDU_TRAP1 && pdutype != SNMP_PDU_TRAP2) @@ -1119,7 +1119,7 @@ static int snmp_parse_mangle(unsigned char *msg, return 0; if (debug > 1) - printk(KERN_DEBUG "bsalg: request: id=0x%lx error_status=%u " + pr_debug("bsalg: request: id=0x%lx error_status=%u " "error_index=%u\n", req.id, req.error_status, req.error_index); } @@ -1145,13 +1145,13 @@ static int snmp_parse_mangle(unsigned char *msg, } if (debug > 1) { - printk(KERN_DEBUG "bsalg: object: "); + pr_debug("bsalg: object: "); for (i = 0; i < obj->id_len; i++) { if (i > 0) - printk("."); - printk("%lu", obj->id[i]); + pr_cont("."); + pr_cont("%lu", obj->id[i]); } - printk(": type=%u\n", obj->type); + pr_cont(": type=%u\n", obj->type); } diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c index cb7ca56..a800cee 100644 --- a/net/ipv4/tcp_output.c +++ b/net/ipv4/tcp_output.c @@ -2296,7 +2296,7 @@ void __tcp_push_pending_frames(struct sock *sk, unsigned int cur_mss, return; if (tcp_write_xmit(sk, cur_mss, nonagle, 0, - sk_gfp_atomic(sk, GFP_ATOMIC))) + sk_gfp_mask(sk, GFP_ATOMIC))) tcp_check_probe_timer(sk); } @@ -3352,8 +3352,9 @@ void tcp_send_ack(struct sock *sk) * tcp_transmit_skb() will set the ownership to this * sock. */ - buff = alloc_skb(MAX_TCP_HEADER, sk_gfp_atomic(sk, GFP_ATOMIC)); - if (!buff) { + buff = alloc_skb(MAX_TCP_HEADER, + sk_gfp_mask(sk, GFP_ATOMIC | __GFP_NOWARN)); + if (unlikely(!buff)) { inet_csk_schedule_ack(sk); inet_csk(sk)->icsk_ack.ato = TCP_ATO_MIN; inet_csk_reset_xmit_timer(sk, ICSK_TIME_DACK, @@ -3375,7 +3376,7 @@ void tcp_send_ack(struct sock *sk) /* Send it off, this clears delayed acks for us. */ skb_mstamp_get(&buff->skb_mstamp); - tcp_transmit_skb(sk, buff, 0, sk_gfp_atomic(sk, GFP_ATOMIC)); + tcp_transmit_skb(sk, buff, 0, (__force gfp_t)0); } EXPORT_SYMBOL_GPL(tcp_send_ack); @@ -3396,7 +3397,8 @@ static int tcp_xmit_probe_skb(struct sock *sk, int urgent, int mib) struct sk_buff *skb; /* We don't queue it, tcp_transmit_skb() sets ownership. */ - skb = alloc_skb(MAX_TCP_HEADER, sk_gfp_atomic(sk, GFP_ATOMIC)); + skb = alloc_skb(MAX_TCP_HEADER, + sk_gfp_mask(sk, GFP_ATOMIC | __GFP_NOWARN)); if (!skb) return -1; @@ -3409,7 +3411,7 @@ static int tcp_xmit_probe_skb(struct sock *sk, int urgent, int mib) tcp_init_nondata_skb(skb, tp->snd_una - !urgent, TCPHDR_ACK); skb_mstamp_get(&skb->skb_mstamp); NET_INC_STATS(sock_net(sk), mib); - return tcp_transmit_skb(sk, skb, 0, GFP_ATOMIC); + return tcp_transmit_skb(sk, skb, 0, (__force gfp_t)0); } void tcp_send_window_probe(struct sock *sk) diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c index 61f2685..6936d0d 100644 --- a/net/ipv6/addrconf.c +++ b/net/ipv6/addrconf.c @@ -5200,6 +5200,20 @@ int addrconf_sysctl_forward(struct ctl_table *ctl, int write, } static +int addrconf_sysctl_hop_limit(struct ctl_table *ctl, int write, + void __user *buffer, size_t *lenp, loff_t *ppos) +{ + struct ctl_table lctl; + int min_hl = 1, max_hl = 255; + + lctl = *ctl; + lctl.extra1 = &min_hl; + lctl.extra2 = &max_hl; + + return proc_dointvec_minmax(&lctl, write, buffer, lenp, ppos); +} + +static int addrconf_sysctl_mtu(struct ctl_table *ctl, int write, void __user *buffer, size_t *lenp, loff_t *ppos) { @@ -5454,7 +5468,7 @@ static struct addrconf_sysctl_table .data = &ipv6_devconf.hop_limit, .maxlen = sizeof(int), .mode = 0644, - .proc_handler = proc_dointvec, + .proc_handler = addrconf_sysctl_hop_limit, }, { .procname = "mtu", diff --git a/net/ipv6/ip6_gre.c b/net/ipv6/ip6_gre.c index 3c7b931..938d03c 100644 --- a/net/ipv6/ip6_gre.c +++ b/net/ipv6/ip6_gre.c @@ -24,7 +24,6 @@ #include <linux/tcp.h> #include <linux/udp.h> #include <linux/if_arp.h> -#include <linux/mroute.h> #include <linux/init.h> #include <linux/in6.h> #include <linux/inetdevice.h> diff --git a/net/ipv6/route.c b/net/ipv6/route.c index 826e6aa..c83b6a5 100644 --- a/net/ipv6/route.c +++ b/net/ipv6/route.c @@ -62,6 +62,7 @@ #include <net/lwtunnel.h> #include <net/ip_tunnels.h> #include <net/l3mdev.h> +#include <trace/events/fib6.h> #include <asm/uaccess.h> @@ -865,6 +866,9 @@ restart: } dst_use(&rt->dst, jiffies); read_unlock_bh(&table->tb6_lock); + + trace_fib6_table_lookup(net, rt, table->tb6_id, fl6); + return rt; } @@ -1078,6 +1082,8 @@ redo_rt6_select: read_unlock_bh(&table->tb6_lock); rt6_dst_from_metrics_check(rt); + + trace_fib6_table_lookup(net, rt, table->tb6_id, fl6); return rt; } else if (unlikely((fl6->flowi6_flags & FLOWI_FLAG_KNOWN_NH) && !(rt->rt6i_flags & RTF_GATEWAY))) { @@ -1101,6 +1107,8 @@ redo_rt6_select: uncached_rt = net->ipv6.ip6_null_entry; dst_hold(&uncached_rt->dst); + + trace_fib6_table_lookup(net, uncached_rt, table->tb6_id, fl6); return uncached_rt; } else { @@ -1125,6 +1133,7 @@ redo_rt6_select: dst_release(&rt->dst); } + trace_fib6_table_lookup(net, pcpu_rt, table->tb6_id, fl6); return pcpu_rt; } @@ -1474,6 +1483,7 @@ out: read_unlock_bh(&table->tb6_lock); + trace_fib6_table_lookup(net, rt, table->tb6_id, fl6); return rt; }; diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c index e7aab56..c16e3fb 100644 --- a/net/ipv6/tcp_ipv6.c +++ b/net/ipv6/tcp_ipv6.c @@ -1136,7 +1136,7 @@ static struct sock *tcp_v6_syn_recv_sock(const struct sock *sk, struct sk_buff * */ tcp_md5_do_add(newsk, (union tcp_md5_addr *)&newsk->sk_v6_daddr, AF_INET6, key->key, key->keylen, - sk_gfp_atomic(sk, GFP_ATOMIC)); + sk_gfp_mask(sk, GFP_ATOMIC)); } #endif @@ -1152,7 +1152,7 @@ static struct sock *tcp_v6_syn_recv_sock(const struct sock *sk, struct sk_buff * /* Clone pktoptions received with SYN, if we own the req */ if (ireq->pktopts) { newnp->pktoptions = skb_clone(ireq->pktopts, - sk_gfp_atomic(sk, GFP_ATOMIC)); + sk_gfp_mask(sk, GFP_ATOMIC)); consume_skb(ireq->pktopts); ireq->pktopts = NULL; if (newnp->pktoptions) @@ -1218,7 +1218,7 @@ static int tcp_v6_do_rcv(struct sock *sk, struct sk_buff *skb) --ANK (980728) */ if (np->rxopt.all) - opt_skb = skb_clone(skb, sk_gfp_atomic(sk, GFP_ATOMIC)); + opt_skb = skb_clone(skb, sk_gfp_mask(sk, GFP_ATOMIC)); if (sk->sk_state == TCP_ESTABLISHED) { /* Fast path */ struct dst_entry *dst = sk->sk_rx_dst; diff --git a/net/iucv/af_iucv.c b/net/iucv/af_iucv.c index 435608c..3ea4c98 100644 --- a/net/iucv/af_iucv.c +++ b/net/iucv/af_iucv.c @@ -303,7 +303,7 @@ static void iucv_sock_wake_msglim(struct sock *sk) rcu_read_lock(); wq = rcu_dereference(sk->sk_wq); - if (wq_has_sleeper(wq)) + if (skwq_has_sleeper(wq)) wake_up_interruptible_all(&wq->wait); sk_wake_async(sk, SOCK_WAKE_SPACE, POLL_OUT); rcu_read_unlock(); diff --git a/net/l2tp/l2tp_ppp.c b/net/l2tp/l2tp_ppp.c index 1ad18c5..d93f113 100644 --- a/net/l2tp/l2tp_ppp.c +++ b/net/l2tp/l2tp_ppp.c @@ -1862,5 +1862,5 @@ MODULE_AUTHOR("James Chapman <jchapman@katalix.com>"); MODULE_DESCRIPTION("PPP over L2TP over UDP"); MODULE_LICENSE("GPL"); MODULE_VERSION(PPPOL2TP_DRV_VERSION); -MODULE_ALIAS("pppox-proto-" __stringify(PX_PROTO_OL2TP)); +MODULE_ALIAS_NET_PF_PROTO(PF_PPPOX, PX_PROTO_OL2TP); MODULE_ALIAS_L2TP_PWTYPE(11); diff --git a/net/mac802154/rx.c b/net/mac802154/rx.c index 42e9672..446e130 100644 --- a/net/mac802154/rx.c +++ b/net/mac802154/rx.c @@ -217,8 +217,7 @@ __ieee802154_rx_handle_packet(struct ieee802154_local *local, break; } - if (skb) - kfree_skb(skb); + kfree_skb(skb); } static void diff --git a/net/mpls/af_mpls.c b/net/mpls/af_mpls.c index c70d750..4b3b9b3 100644 --- a/net/mpls/af_mpls.c +++ b/net/mpls/af_mpls.c @@ -96,22 +96,15 @@ bool mpls_pkt_too_big(const struct sk_buff *skb, unsigned int mtu) } EXPORT_SYMBOL_GPL(mpls_pkt_too_big); -static struct mpls_nh *mpls_select_multipath(struct mpls_route *rt, - struct sk_buff *skb, bool bos) +static u32 mpls_multipath_hash(struct mpls_route *rt, + struct sk_buff *skb, bool bos) { struct mpls_entry_decoded dec; struct mpls_shim_hdr *hdr; bool eli_seen = false; int label_index; - int nh_index = 0; u32 hash = 0; - /* No need to look further into packet if there's only - * one path - */ - if (rt->rt_nhn == 1) - goto out; - for (label_index = 0; label_index < MAX_MP_SELECT_LABELS && !bos; label_index++) { if (!pskb_may_pull(skb, sizeof(*hdr) * label_index)) @@ -165,7 +158,38 @@ static struct mpls_nh *mpls_select_multipath(struct mpls_route *rt, } } - nh_index = hash % rt->rt_nhn; + return hash; +} + +static struct mpls_nh *mpls_select_multipath(struct mpls_route *rt, + struct sk_buff *skb, bool bos) +{ + int alive = ACCESS_ONCE(rt->rt_nhn_alive); + u32 hash = 0; + int nh_index = 0; + int n = 0; + + /* No need to look further into packet if there's only + * one path + */ + if (rt->rt_nhn == 1) + goto out; + + if (alive <= 0) + return NULL; + + hash = mpls_multipath_hash(rt, skb, bos); + nh_index = hash % alive; + if (alive == rt->rt_nhn) + goto out; + for_nexthops(rt) { + if (nh->nh_flags & (RTNH_F_DEAD | RTNH_F_LINKDOWN)) + continue; + if (n == nh_index) + return nh; + n++; + } endfor_nexthops(rt); + out: return &rt->rt_nh[nh_index]; } @@ -365,6 +389,7 @@ static struct mpls_route *mpls_rt_alloc(int num_nh, u8 max_alen) GFP_KERNEL); if (rt) { rt->rt_nhn = num_nh; + rt->rt_nhn_alive = num_nh; rt->rt_max_alen = max_alen_aligned; } @@ -536,6 +561,16 @@ static int mpls_nh_assign_dev(struct net *net, struct mpls_route *rt, RCU_INIT_POINTER(nh->nh_dev, dev); + if (!(dev->flags & IFF_UP)) { + nh->nh_flags |= RTNH_F_DEAD; + } else { + unsigned int flags; + + flags = dev_get_flags(dev); + if (!(flags & (IFF_RUNNING | IFF_LOWER_UP))) + nh->nh_flags |= RTNH_F_LINKDOWN; + } + return 0; errout: @@ -570,6 +605,9 @@ static int mpls_nh_build_from_cfg(struct mpls_route_config *cfg, if (err) goto errout; + if (nh->nh_flags & (RTNH_F_DEAD | RTNH_F_LINKDOWN)) + rt->rt_nhn_alive--; + return 0; errout: @@ -577,8 +615,8 @@ errout: } static int mpls_nh_build(struct net *net, struct mpls_route *rt, - struct mpls_nh *nh, int oif, - struct nlattr *via, struct nlattr *newdst) + struct mpls_nh *nh, int oif, struct nlattr *via, + struct nlattr *newdst) { int err = -ENOMEM; @@ -681,11 +719,13 @@ static int mpls_nh_build_multi(struct mpls_route_config *cfg, goto errout; err = mpls_nh_build(cfg->rc_nlinfo.nl_net, rt, nh, - rtnh->rtnh_ifindex, nla_via, - nla_newdst); + rtnh->rtnh_ifindex, nla_via, nla_newdst); if (err) goto errout; + if (nh->nh_flags & (RTNH_F_DEAD | RTNH_F_LINKDOWN)) + rt->rt_nhn_alive--; + rtnh = rtnh_next(rtnh, &remaining); nhs++; } endfor_nexthops(rt); @@ -875,34 +915,74 @@ free: return ERR_PTR(err); } -static void mpls_ifdown(struct net_device *dev) +static void mpls_ifdown(struct net_device *dev, int event) { struct mpls_route __rcu **platform_label; struct net *net = dev_net(dev); - struct mpls_dev *mdev; unsigned index; platform_label = rtnl_dereference(net->mpls.platform_label); for (index = 0; index < net->mpls.platform_labels; index++) { struct mpls_route *rt = rtnl_dereference(platform_label[index]); + if (!rt) continue; - for_nexthops(rt) { + + change_nexthops(rt) { if (rtnl_dereference(nh->nh_dev) != dev) continue; - nh->nh_dev = NULL; + switch (event) { + case NETDEV_DOWN: + case NETDEV_UNREGISTER: + nh->nh_flags |= RTNH_F_DEAD; + /* fall through */ + case NETDEV_CHANGE: + nh->nh_flags |= RTNH_F_LINKDOWN; + ACCESS_ONCE(rt->rt_nhn_alive) = rt->rt_nhn_alive - 1; + break; + } + if (event == NETDEV_UNREGISTER) + RCU_INIT_POINTER(nh->nh_dev, NULL); } endfor_nexthops(rt); } - mdev = mpls_dev_get(dev); - if (!mdev) - return; - mpls_dev_sysctl_unregister(mdev); + return; +} + +static void mpls_ifup(struct net_device *dev, unsigned int nh_flags) +{ + struct mpls_route __rcu **platform_label; + struct net *net = dev_net(dev); + unsigned index; + int alive; + + platform_label = rtnl_dereference(net->mpls.platform_label); + for (index = 0; index < net->mpls.platform_labels; index++) { + struct mpls_route *rt = rtnl_dereference(platform_label[index]); + + if (!rt) + continue; + + alive = 0; + change_nexthops(rt) { + struct net_device *nh_dev = + rtnl_dereference(nh->nh_dev); + + if (!(nh->nh_flags & nh_flags)) { + alive++; + continue; + } + if (nh_dev != dev) + continue; + alive++; + nh->nh_flags &= ~nh_flags; + } endfor_nexthops(rt); - RCU_INIT_POINTER(dev->mpls_ptr, NULL); + ACCESS_ONCE(rt->rt_nhn_alive) = alive; + } - kfree_rcu(mdev, rcu); + return; } static int mpls_dev_notify(struct notifier_block *this, unsigned long event, @@ -910,9 +990,9 @@ static int mpls_dev_notify(struct notifier_block *this, unsigned long event, { struct net_device *dev = netdev_notifier_info_to_dev(ptr); struct mpls_dev *mdev; + unsigned int flags; - switch(event) { - case NETDEV_REGISTER: + if (event == NETDEV_REGISTER) { /* For now just support ethernet devices */ if ((dev->type == ARPHRD_ETHER) || (dev->type == ARPHRD_LOOPBACK)) { @@ -920,10 +1000,39 @@ static int mpls_dev_notify(struct notifier_block *this, unsigned long event, if (IS_ERR(mdev)) return notifier_from_errno(PTR_ERR(mdev)); } - break; + return NOTIFY_OK; + } + mdev = mpls_dev_get(dev); + if (!mdev) + return NOTIFY_OK; + + switch (event) { + case NETDEV_DOWN: + mpls_ifdown(dev, event); + break; + case NETDEV_UP: + flags = dev_get_flags(dev); + if (flags & (IFF_RUNNING | IFF_LOWER_UP)) + mpls_ifup(dev, RTNH_F_DEAD | RTNH_F_LINKDOWN); + else + mpls_ifup(dev, RTNH_F_DEAD); + break; + case NETDEV_CHANGE: + flags = dev_get_flags(dev); + if (flags & (IFF_RUNNING | IFF_LOWER_UP)) + mpls_ifup(dev, RTNH_F_DEAD | RTNH_F_LINKDOWN); + else + mpls_ifdown(dev, event); + break; case NETDEV_UNREGISTER: - mpls_ifdown(dev); + mpls_ifdown(dev, event); + mdev = mpls_dev_get(dev); + if (mdev) { + mpls_dev_sysctl_unregister(mdev); + RCU_INIT_POINTER(dev->mpls_ptr, NULL); + kfree_rcu(mdev, rcu); + } break; case NETDEV_CHANGENAME: mdev = mpls_dev_get(dev); @@ -1237,9 +1346,15 @@ static int mpls_dump_route(struct sk_buff *skb, u32 portid, u32 seq, int event, dev = rtnl_dereference(nh->nh_dev); if (dev && nla_put_u32(skb, RTA_OIF, dev->ifindex)) goto nla_put_failure; + if (nh->nh_flags & RTNH_F_LINKDOWN) + rtm->rtm_flags |= RTNH_F_LINKDOWN; + if (nh->nh_flags & RTNH_F_DEAD) + rtm->rtm_flags |= RTNH_F_DEAD; } else { struct rtnexthop *rtnh; struct nlattr *mp; + int dead = 0; + int linkdown = 0; mp = nla_nest_start(skb, RTA_MULTIPATH); if (!mp) @@ -1253,6 +1368,15 @@ static int mpls_dump_route(struct sk_buff *skb, u32 portid, u32 seq, int event, dev = rtnl_dereference(nh->nh_dev); if (dev) rtnh->rtnh_ifindex = dev->ifindex; + if (nh->nh_flags & RTNH_F_LINKDOWN) { + rtnh->rtnh_flags |= RTNH_F_LINKDOWN; + linkdown++; + } + if (nh->nh_flags & RTNH_F_DEAD) { + rtnh->rtnh_flags |= RTNH_F_DEAD; + dead++; + } + if (nh->nh_labels && nla_put_labels(skb, RTA_NEWDST, nh->nh_labels, nh->nh_label)) @@ -1266,6 +1390,11 @@ static int mpls_dump_route(struct sk_buff *skb, u32 portid, u32 seq, int event, rtnh->rtnh_len = nlmsg_get_pos(skb) - (void *)rtnh; } endfor_nexthops(rt); + if (linkdown == rt->rt_nhn) + rtm->rtm_flags |= RTNH_F_LINKDOWN; + if (dead == rt->rt_nhn) + rtm->rtm_flags |= RTNH_F_DEAD; + nla_nest_end(skb, mp); } diff --git a/net/mpls/internal.h b/net/mpls/internal.h index bde52ce..732a5c1 100644 --- a/net/mpls/internal.h +++ b/net/mpls/internal.h @@ -41,6 +41,7 @@ enum mpls_payload_type { struct mpls_nh { /* next hop label forwarding entry */ struct net_device __rcu *nh_dev; + unsigned int nh_flags; u32 nh_label[MAX_NEW_LABELS]; u8 nh_labels; u8 nh_via_alen; @@ -74,6 +75,7 @@ struct mpls_route { /* next hop label forwarding entry */ u8 rt_payload_type; u8 rt_max_alen; unsigned int rt_nhn; + unsigned int rt_nhn_alive; struct mpls_nh rt_nh[0]; }; diff --git a/net/openvswitch/vport-netdev.c b/net/openvswitch/vport-netdev.c index 6b0190b..6a6adf3 100644 --- a/net/openvswitch/vport-netdev.c +++ b/net/openvswitch/vport-netdev.c @@ -105,7 +105,7 @@ struct vport *ovs_netdev_link(struct vport *vport, const char *name) rtnl_lock(); err = netdev_master_upper_dev_link(vport->dev, - get_dpdev(vport->dp)); + get_dpdev(vport->dp), NULL, NULL); if (err) goto error_unlock; diff --git a/net/rxrpc/af_rxrpc.c b/net/rxrpc/af_rxrpc.c index 1f8a144..7e2d105 100644 --- a/net/rxrpc/af_rxrpc.c +++ b/net/rxrpc/af_rxrpc.c @@ -67,7 +67,7 @@ static void rxrpc_write_space(struct sock *sk) if (rxrpc_writable(sk)) { struct socket_wq *wq = rcu_dereference(sk->sk_wq); - if (wq_has_sleeper(wq)) + if (skwq_has_sleeper(wq)) wake_up_interruptible(&wq->wait); sk_wake_async(sk, SOCK_WAKE_SPACE, POLL_OUT); } diff --git a/net/sctp/socket.c b/net/sctp/socket.c index 03c8256..106bb09 100644 --- a/net/sctp/socket.c +++ b/net/sctp/socket.c @@ -6982,7 +6982,7 @@ void sctp_data_ready(struct sock *sk) rcu_read_lock(); wq = rcu_dereference(sk->sk_wq); - if (wq_has_sleeper(wq)) + if (skwq_has_sleeper(wq)) wake_up_interruptible_sync_poll(&wq->wait, POLLIN | POLLRDNORM | POLLRDBAND); sk_wake_async(sk, SOCK_WAKE_WAITD, POLL_IN); diff --git a/net/tipc/bcast.c b/net/tipc/bcast.c index 9dc239d..e401108 100644 --- a/net/tipc/bcast.c +++ b/net/tipc/bcast.c @@ -332,131 +332,15 @@ void tipc_bcast_remove_peer(struct net *net, struct tipc_link *rcv_l) tipc_sk_rcv(net, inputq); } -static int __tipc_nl_add_bc_link_stat(struct sk_buff *skb, - struct tipc_stats *stats) -{ - int i; - struct nlattr *nest; - - struct nla_map { - __u32 key; - __u32 val; - }; - - struct nla_map map[] = { - {TIPC_NLA_STATS_RX_INFO, stats->recv_info}, - {TIPC_NLA_STATS_RX_FRAGMENTS, stats->recv_fragments}, - {TIPC_NLA_STATS_RX_FRAGMENTED, stats->recv_fragmented}, - {TIPC_NLA_STATS_RX_BUNDLES, stats->recv_bundles}, - {TIPC_NLA_STATS_RX_BUNDLED, stats->recv_bundled}, - {TIPC_NLA_STATS_TX_INFO, stats->sent_info}, - {TIPC_NLA_STATS_TX_FRAGMENTS, stats->sent_fragments}, - {TIPC_NLA_STATS_TX_FRAGMENTED, stats->sent_fragmented}, - {TIPC_NLA_STATS_TX_BUNDLES, stats->sent_bundles}, - {TIPC_NLA_STATS_TX_BUNDLED, stats->sent_bundled}, - {TIPC_NLA_STATS_RX_NACKS, stats->recv_nacks}, - {TIPC_NLA_STATS_RX_DEFERRED, stats->deferred_recv}, - {TIPC_NLA_STATS_TX_NACKS, stats->sent_nacks}, - {TIPC_NLA_STATS_TX_ACKS, stats->sent_acks}, - {TIPC_NLA_STATS_RETRANSMITTED, stats->retransmitted}, - {TIPC_NLA_STATS_DUPLICATES, stats->duplicates}, - {TIPC_NLA_STATS_LINK_CONGS, stats->link_congs}, - {TIPC_NLA_STATS_MAX_QUEUE, stats->max_queue_sz}, - {TIPC_NLA_STATS_AVG_QUEUE, stats->queue_sz_counts ? - (stats->accu_queue_sz / stats->queue_sz_counts) : 0} - }; - - nest = nla_nest_start(skb, TIPC_NLA_LINK_STATS); - if (!nest) - return -EMSGSIZE; - - for (i = 0; i < ARRAY_SIZE(map); i++) - if (nla_put_u32(skb, map[i].key, map[i].val)) - goto msg_full; - - nla_nest_end(skb, nest); - - return 0; -msg_full: - nla_nest_cancel(skb, nest); - - return -EMSGSIZE; -} - -int tipc_nl_add_bc_link(struct net *net, struct tipc_nl_msg *msg) -{ - int err; - void *hdr; - struct nlattr *attrs; - struct nlattr *prop; - struct tipc_net *tn = net_generic(net, tipc_net_id); - struct tipc_link *bcl = tn->bcl; - - if (!bcl) - return 0; - - tipc_bcast_lock(net); - - hdr = genlmsg_put(msg->skb, msg->portid, msg->seq, &tipc_genl_family, - NLM_F_MULTI, TIPC_NL_LINK_GET); - if (!hdr) - return -EMSGSIZE; - - attrs = nla_nest_start(msg->skb, TIPC_NLA_LINK); - if (!attrs) - goto msg_full; - - /* The broadcast link is always up */ - if (nla_put_flag(msg->skb, TIPC_NLA_LINK_UP)) - goto attr_msg_full; - - if (nla_put_flag(msg->skb, TIPC_NLA_LINK_BROADCAST)) - goto attr_msg_full; - if (nla_put_string(msg->skb, TIPC_NLA_LINK_NAME, bcl->name)) - goto attr_msg_full; - if (nla_put_u32(msg->skb, TIPC_NLA_LINK_RX, bcl->rcv_nxt)) - goto attr_msg_full; - if (nla_put_u32(msg->skb, TIPC_NLA_LINK_TX, bcl->snd_nxt)) - goto attr_msg_full; - - prop = nla_nest_start(msg->skb, TIPC_NLA_LINK_PROP); - if (!prop) - goto attr_msg_full; - if (nla_put_u32(msg->skb, TIPC_NLA_PROP_WIN, bcl->window)) - goto prop_msg_full; - nla_nest_end(msg->skb, prop); - - err = __tipc_nl_add_bc_link_stat(msg->skb, &bcl->stats); - if (err) - goto attr_msg_full; - - tipc_bcast_unlock(net); - nla_nest_end(msg->skb, attrs); - genlmsg_end(msg->skb, hdr); - - return 0; - -prop_msg_full: - nla_nest_cancel(msg->skb, prop); -attr_msg_full: - nla_nest_cancel(msg->skb, attrs); -msg_full: - tipc_bcast_unlock(net); - genlmsg_cancel(msg->skb, hdr); - - return -EMSGSIZE; -} - int tipc_bclink_reset_stats(struct net *net) { - struct tipc_net *tn = net_generic(net, tipc_net_id); - struct tipc_link *bcl = tn->bcl; + struct tipc_link *l = tipc_bc_sndlink(net); - if (!bcl) + if (!l) return -ENOPROTOOPT; tipc_bcast_lock(net); - memset(&bcl->stats, 0, sizeof(bcl->stats)); + tipc_link_reset_stats(l); tipc_bcast_unlock(net); return 0; } @@ -530,9 +414,7 @@ enomem: void tipc_bcast_reinit(struct net *net) { - struct tipc_bc_base *b = tipc_bc_base(net); - - msg_set_prevnode(b->link->pmsg, tipc_own_addr(net)); + tipc_link_reinit(tipc_bc_sndlink(net), tipc_own_addr(net)); } void tipc_bcast_stop(struct net *net) diff --git a/net/tipc/bcast.h b/net/tipc/bcast.h index 2855b93..1944c6c 100644 --- a/net/tipc/bcast.h +++ b/net/tipc/bcast.h @@ -43,6 +43,7 @@ struct tipc_node; struct tipc_msg; struct tipc_nl_msg; struct tipc_node_map; +extern const char tipc_bclink_name[]; int tipc_bcast_init(struct net *net); void tipc_bcast_reinit(struct net *net); diff --git a/net/tipc/bearer.c b/net/tipc/bearer.c index 648f2a6..802ffad 100644 --- a/net/tipc/bearer.c +++ b/net/tipc/bearer.c @@ -71,7 +71,7 @@ static const struct nla_policy tipc_nl_media_policy[TIPC_NLA_MEDIA_MAX + 1] = { [TIPC_NLA_MEDIA_PROP] = { .type = NLA_NESTED } }; -static void bearer_disable(struct net *net, struct tipc_bearer *b_ptr); +static void bearer_disable(struct net *net, struct tipc_bearer *b); /** * tipc_media_find - locates specified media object by name @@ -107,13 +107,13 @@ static struct tipc_media *media_find_id(u8 type) void tipc_media_addr_printf(char *buf, int len, struct tipc_media_addr *a) { char addr_str[MAX_ADDR_STR]; - struct tipc_media *m_ptr; + struct tipc_media *m; int ret; - m_ptr = media_find_id(a->media_id); + m = media_find_id(a->media_id); - if (m_ptr && !m_ptr->addr2str(a, addr_str, sizeof(addr_str))) - ret = scnprintf(buf, len, "%s(%s)", m_ptr->name, addr_str); + if (m && !m->addr2str(a, addr_str, sizeof(addr_str))) + ret = scnprintf(buf, len, "%s(%s)", m->name, addr_str); else { u32 i; @@ -175,13 +175,13 @@ static int bearer_name_validate(const char *name, struct tipc_bearer *tipc_bearer_find(struct net *net, const char *name) { struct tipc_net *tn = net_generic(net, tipc_net_id); - struct tipc_bearer *b_ptr; + struct tipc_bearer *b; u32 i; for (i = 0; i < MAX_BEARERS; i++) { - b_ptr = rtnl_dereference(tn->bearer_list[i]); - if (b_ptr && (!strcmp(b_ptr->name, name))) - return b_ptr; + b = rtnl_dereference(tn->bearer_list[i]); + if (b && (!strcmp(b->name, name))) + return b; } return NULL; } @@ -189,24 +189,24 @@ struct tipc_bearer *tipc_bearer_find(struct net *net, const char *name) void tipc_bearer_add_dest(struct net *net, u32 bearer_id, u32 dest) { struct tipc_net *tn = net_generic(net, tipc_net_id); - struct tipc_bearer *b_ptr; + struct tipc_bearer *b; rcu_read_lock(); - b_ptr = rcu_dereference_rtnl(tn->bearer_list[bearer_id]); - if (b_ptr) - tipc_disc_add_dest(b_ptr->link_req); + b = rcu_dereference_rtnl(tn->bearer_list[bearer_id]); + if (b) + tipc_disc_add_dest(b->link_req); rcu_read_unlock(); } void tipc_bearer_remove_dest(struct net *net, u32 bearer_id, u32 dest) { struct tipc_net *tn = net_generic(net, tipc_net_id); - struct tipc_bearer *b_ptr; + struct tipc_bearer *b; rcu_read_lock(); - b_ptr = rcu_dereference_rtnl(tn->bearer_list[bearer_id]); - if (b_ptr) - tipc_disc_remove_dest(b_ptr->link_req); + b = rcu_dereference_rtnl(tn->bearer_list[bearer_id]); + if (b) + tipc_disc_remove_dest(b->link_req); rcu_read_unlock(); } @@ -218,8 +218,8 @@ static int tipc_enable_bearer(struct net *net, const char *name, struct nlattr *attr[]) { struct tipc_net *tn = net_generic(net, tipc_net_id); - struct tipc_bearer *b_ptr; - struct tipc_media *m_ptr; + struct tipc_bearer *b; + struct tipc_media *m; struct tipc_bearer_names b_names; char addr_string[16]; u32 bearer_id; @@ -255,31 +255,31 @@ static int tipc_enable_bearer(struct net *net, const char *name, return -EINVAL; } - m_ptr = tipc_media_find(b_names.media_name); - if (!m_ptr) { + m = tipc_media_find(b_names.media_name); + if (!m) { pr_warn("Bearer <%s> rejected, media <%s> not registered\n", name, b_names.media_name); return -EINVAL; } if (priority == TIPC_MEDIA_LINK_PRI) - priority = m_ptr->priority; + priority = m->priority; restart: bearer_id = MAX_BEARERS; with_this_prio = 1; for (i = MAX_BEARERS; i-- != 0; ) { - b_ptr = rtnl_dereference(tn->bearer_list[i]); - if (!b_ptr) { + b = rtnl_dereference(tn->bearer_list[i]); + if (!b) { bearer_id = i; continue; } - if (!strcmp(name, b_ptr->name)) { + if (!strcmp(name, b->name)) { pr_warn("Bearer <%s> rejected, already enabled\n", name); return -EINVAL; } - if ((b_ptr->priority == priority) && + if ((b->priority == priority) && (++with_this_prio > 2)) { if (priority-- == 0) { pr_warn("Bearer <%s> rejected, duplicate priority\n", @@ -297,35 +297,35 @@ restart: return -EINVAL; } - b_ptr = kzalloc(sizeof(*b_ptr), GFP_ATOMIC); - if (!b_ptr) + b = kzalloc(sizeof(*b), GFP_ATOMIC); + if (!b) return -ENOMEM; - strcpy(b_ptr->name, name); - b_ptr->media = m_ptr; - res = m_ptr->enable_media(net, b_ptr, attr); + strcpy(b->name, name); + b->media = m; + res = m->enable_media(net, b, attr); if (res) { pr_warn("Bearer <%s> rejected, enable failure (%d)\n", name, -res); return -EINVAL; } - b_ptr->identity = bearer_id; - b_ptr->tolerance = m_ptr->tolerance; - b_ptr->window = m_ptr->window; - b_ptr->domain = disc_domain; - b_ptr->net_plane = bearer_id + 'A'; - b_ptr->priority = priority; + b->identity = bearer_id; + b->tolerance = m->tolerance; + b->window = m->window; + b->domain = disc_domain; + b->net_plane = bearer_id + 'A'; + b->priority = priority; - res = tipc_disc_create(net, b_ptr, &b_ptr->bcast_addr); + res = tipc_disc_create(net, b, &b->bcast_addr); if (res) { - bearer_disable(net, b_ptr); + bearer_disable(net, b); pr_warn("Bearer <%s> rejected, discovery object creation failed\n", name); return -EINVAL; } - rcu_assign_pointer(tn->bearer_list[bearer_id], b_ptr); + rcu_assign_pointer(tn->bearer_list[bearer_id], b); pr_info("Enabled bearer <%s>, discovery domain %s, priority %u\n", name, @@ -336,11 +336,11 @@ restart: /** * tipc_reset_bearer - Reset all links established over this bearer */ -static int tipc_reset_bearer(struct net *net, struct tipc_bearer *b_ptr) +static int tipc_reset_bearer(struct net *net, struct tipc_bearer *b) { - pr_info("Resetting bearer <%s>\n", b_ptr->name); - tipc_node_delete_links(net, b_ptr->identity); - tipc_disc_reset(net, b_ptr); + pr_info("Resetting bearer <%s>\n", b->name); + tipc_node_delete_links(net, b->identity); + tipc_disc_reset(net, b); return 0; } @@ -349,26 +349,26 @@ static int tipc_reset_bearer(struct net *net, struct tipc_bearer *b_ptr) * * Note: This routine assumes caller holds RTNL lock. */ -static void bearer_disable(struct net *net, struct tipc_bearer *b_ptr) +static void bearer_disable(struct net *net, struct tipc_bearer *b) { struct tipc_net *tn = net_generic(net, tipc_net_id); u32 i; - pr_info("Disabling bearer <%s>\n", b_ptr->name); - b_ptr->media->disable_media(b_ptr); + pr_info("Disabling bearer <%s>\n", b->name); + b->media->disable_media(b); - tipc_node_delete_links(net, b_ptr->identity); - RCU_INIT_POINTER(b_ptr->media_ptr, NULL); - if (b_ptr->link_req) - tipc_disc_delete(b_ptr->link_req); + tipc_node_delete_links(net, b->identity); + RCU_INIT_POINTER(b->media_ptr, NULL); + if (b->link_req) + tipc_disc_delete(b->link_req); for (i = 0; i < MAX_BEARERS; i++) { - if (b_ptr == rtnl_dereference(tn->bearer_list[i])) { + if (b == rtnl_dereference(tn->bearer_list[i])) { RCU_INIT_POINTER(tn->bearer_list[i], NULL); break; } } - kfree_rcu(b_ptr, rcu); + kfree_rcu(b, rcu); } int tipc_enable_l2_media(struct net *net, struct tipc_bearer *b, @@ -411,7 +411,7 @@ void tipc_disable_l2_media(struct tipc_bearer *b) /** * tipc_l2_send_msg - send a TIPC packet out over an L2 interface * @buf: the packet to be sent - * @b_ptr: the bearer through which the packet is to be sent + * @b: the bearer through which the packet is to be sent * @dest: peer destination address */ int tipc_l2_send_msg(struct net *net, struct sk_buff *skb, @@ -532,14 +532,14 @@ void tipc_bearer_bc_xmit(struct net *net, u32 bearer_id, static int tipc_l2_rcv_msg(struct sk_buff *buf, struct net_device *dev, struct packet_type *pt, struct net_device *orig_dev) { - struct tipc_bearer *b_ptr; + struct tipc_bearer *b; rcu_read_lock(); - b_ptr = rcu_dereference_rtnl(dev->tipc_ptr); - if (likely(b_ptr)) { + b = rcu_dereference_rtnl(dev->tipc_ptr); + if (likely(b)) { if (likely(buf->pkt_type <= PACKET_BROADCAST)) { buf->next = NULL; - tipc_rcv(dev_net(dev), buf, b_ptr); + tipc_rcv(dev_net(dev), buf, b); rcu_read_unlock(); return NET_RX_SUCCESS; } @@ -564,13 +564,13 @@ static int tipc_l2_device_event(struct notifier_block *nb, unsigned long evt, { struct net_device *dev = netdev_notifier_info_to_dev(ptr); struct net *net = dev_net(dev); - struct tipc_bearer *b_ptr; + struct tipc_bearer *b; - b_ptr = rtnl_dereference(dev->tipc_ptr); - if (!b_ptr) + b = rtnl_dereference(dev->tipc_ptr); + if (!b) return NOTIFY_DONE; - b_ptr->mtu = dev->mtu; + b->mtu = dev->mtu; switch (evt) { case NETDEV_CHANGE: @@ -578,16 +578,16 @@ static int tipc_l2_device_event(struct notifier_block *nb, unsigned long evt, break; case NETDEV_GOING_DOWN: case NETDEV_CHANGEMTU: - tipc_reset_bearer(net, b_ptr); + tipc_reset_bearer(net, b); break; case NETDEV_CHANGEADDR: - b_ptr->media->raw2addr(b_ptr, &b_ptr->addr, + b->media->raw2addr(b, &b->addr, (char *)dev->dev_addr); - tipc_reset_bearer(net, b_ptr); + tipc_reset_bearer(net, b); break; case NETDEV_UNREGISTER: case NETDEV_CHANGENAME: - bearer_disable(dev_net(dev), b_ptr); + bearer_disable(dev_net(dev), b); break; } return NOTIFY_OK; @@ -623,13 +623,13 @@ void tipc_bearer_cleanup(void) void tipc_bearer_stop(struct net *net) { struct tipc_net *tn = net_generic(net, tipc_net_id); - struct tipc_bearer *b_ptr; + struct tipc_bearer *b; u32 i; for (i = 0; i < MAX_BEARERS; i++) { - b_ptr = rtnl_dereference(tn->bearer_list[i]); - if (b_ptr) { - bearer_disable(net, b_ptr); + b = rtnl_dereference(tn->bearer_list[i]); + if (b) { + bearer_disable(net, b); tn->bearer_list[i] = NULL; } } diff --git a/net/tipc/bearer.h b/net/tipc/bearer.h index 552185b..e318205 100644 --- a/net/tipc/bearer.h +++ b/net/tipc/bearer.h @@ -103,11 +103,11 @@ struct tipc_bearer; */ struct tipc_media { int (*send_msg)(struct net *net, struct sk_buff *buf, - struct tipc_bearer *b_ptr, + struct tipc_bearer *b, struct tipc_media_addr *dest); - int (*enable_media)(struct net *net, struct tipc_bearer *b_ptr, + int (*enable_media)(struct net *net, struct tipc_bearer *b, struct nlattr *attr[]); - void (*disable_media)(struct tipc_bearer *b_ptr); + void (*disable_media)(struct tipc_bearer *b); int (*addr2str)(struct tipc_media_addr *addr, char *strbuf, int bufsz); @@ -176,7 +176,7 @@ struct tipc_bearer_names { * TIPC routines available to supported media types */ -void tipc_rcv(struct net *net, struct sk_buff *skb, struct tipc_bearer *b_ptr); +void tipc_rcv(struct net *net, struct sk_buff *skb, struct tipc_bearer *b); /* * Routines made available to TIPC by supported media types diff --git a/net/tipc/core.h b/net/tipc/core.h index 18e95a8..5504d63 100644 --- a/net/tipc/core.h +++ b/net/tipc/core.h @@ -118,6 +118,11 @@ static inline int tipc_netid(struct net *net) return tipc_net(net)->net_id; } +static inline struct list_head *tipc_nodes(struct net *net) +{ + return &tipc_net(net)->node_list; +} + static inline u16 mod(u16 x) { return x & 0xffffu; diff --git a/net/tipc/discover.c b/net/tipc/discover.c index afe8c47..f1e738e 100644 --- a/net/tipc/discover.c +++ b/net/tipc/discover.c @@ -75,14 +75,14 @@ struct tipc_link_req { * tipc_disc_init_msg - initialize a link setup message * @net: the applicable net namespace * @type: message type (request or response) - * @b_ptr: ptr to bearer issuing message + * @b: ptr to bearer issuing message */ static void tipc_disc_init_msg(struct net *net, struct sk_buff *buf, u32 type, - struct tipc_bearer *b_ptr) + struct tipc_bearer *b) { struct tipc_net *tn = net_generic(net, tipc_net_id); struct tipc_msg *msg; - u32 dest_domain = b_ptr->domain; + u32 dest_domain = b->domain; msg = buf_msg(buf); tipc_msg_init(tn->own_addr, msg, LINK_CONFIG, type, @@ -92,16 +92,16 @@ static void tipc_disc_init_msg(struct net *net, struct sk_buff *buf, u32 type, msg_set_node_capabilities(msg, TIPC_NODE_CAPABILITIES); msg_set_dest_domain(msg, dest_domain); msg_set_bc_netid(msg, tn->net_id); - b_ptr->media->addr2msg(msg_media_addr(msg), &b_ptr->addr); + b->media->addr2msg(msg_media_addr(msg), &b->addr); } /** * disc_dupl_alert - issue node address duplication alert - * @b_ptr: pointer to bearer detecting duplication + * @b: pointer to bearer detecting duplication * @node_addr: duplicated node address * @media_addr: media address advertised by duplicated node */ -static void disc_dupl_alert(struct tipc_bearer *b_ptr, u32 node_addr, +static void disc_dupl_alert(struct tipc_bearer *b, u32 node_addr, struct tipc_media_addr *media_addr) { char node_addr_str[16]; @@ -111,7 +111,7 @@ static void disc_dupl_alert(struct tipc_bearer *b_ptr, u32 node_addr, tipc_media_addr_printf(media_addr_str, sizeof(media_addr_str), media_addr); pr_warn("Duplicate %s using %s seen on <%s>\n", node_addr_str, - media_addr_str, b_ptr->name); + media_addr_str, b->name); } /** @@ -261,13 +261,13 @@ exit: /** * tipc_disc_create - create object to send periodic link setup requests * @net: the applicable net namespace - * @b_ptr: ptr to bearer issuing requests + * @b: ptr to bearer issuing requests * @dest: destination address for request messages * @dest_domain: network domain to which links can be established * * Returns 0 if successful, otherwise -errno. */ -int tipc_disc_create(struct net *net, struct tipc_bearer *b_ptr, +int tipc_disc_create(struct net *net, struct tipc_bearer *b, struct tipc_media_addr *dest) { struct tipc_link_req *req; @@ -282,17 +282,17 @@ int tipc_disc_create(struct net *net, struct tipc_bearer *b_ptr, return -ENOMEM; } - tipc_disc_init_msg(net, req->buf, DSC_REQ_MSG, b_ptr); + tipc_disc_init_msg(net, req->buf, DSC_REQ_MSG, b); memcpy(&req->dest, dest, sizeof(*dest)); req->net = net; - req->bearer_id = b_ptr->identity; - req->domain = b_ptr->domain; + req->bearer_id = b->identity; + req->domain = b->domain; req->num_nodes = 0; req->timer_intv = TIPC_LINK_REQ_INIT; spin_lock_init(&req->lock); setup_timer(&req->timer, disc_timeout, (unsigned long)req); mod_timer(&req->timer, jiffies + req->timer_intv); - b_ptr->link_req = req; + b->link_req = req; skb = skb_clone(req->buf, GFP_ATOMIC); if (skb) tipc_bearer_xmit_skb(net, req->bearer_id, skb, &req->dest); @@ -313,19 +313,19 @@ void tipc_disc_delete(struct tipc_link_req *req) /** * tipc_disc_reset - reset object to send periodic link setup requests * @net: the applicable net namespace - * @b_ptr: ptr to bearer issuing requests + * @b: ptr to bearer issuing requests * @dest_domain: network domain to which links can be established */ -void tipc_disc_reset(struct net *net, struct tipc_bearer *b_ptr) +void tipc_disc_reset(struct net *net, struct tipc_bearer *b) { - struct tipc_link_req *req = b_ptr->link_req; + struct tipc_link_req *req = b->link_req; struct sk_buff *skb; spin_lock_bh(&req->lock); - tipc_disc_init_msg(net, req->buf, DSC_REQ_MSG, b_ptr); + tipc_disc_init_msg(net, req->buf, DSC_REQ_MSG, b); req->net = net; - req->bearer_id = b_ptr->identity; - req->domain = b_ptr->domain; + req->bearer_id = b->identity; + req->domain = b->domain; req->num_nodes = 0; req->timer_intv = TIPC_LINK_REQ_INIT; mod_timer(&req->timer, jiffies + req->timer_intv); diff --git a/net/tipc/link.c b/net/tipc/link.c index 91aea07..0c2944f 100644 --- a/net/tipc/link.c +++ b/net/tipc/link.c @@ -45,28 +45,156 @@ #include <linux/pkt_sched.h> +struct tipc_stats { + u32 sent_info; /* used in counting # sent packets */ + u32 recv_info; /* used in counting # recv'd packets */ + u32 sent_states; + u32 recv_states; + u32 sent_probes; + u32 recv_probes; + u32 sent_nacks; + u32 recv_nacks; + u32 sent_acks; + u32 sent_bundled; + u32 sent_bundles; + u32 recv_bundled; + u32 recv_bundles; + u32 retransmitted; + u32 sent_fragmented; + u32 sent_fragments; + u32 recv_fragmented; + u32 recv_fragments; + u32 link_congs; /* # port sends blocked by congestion */ + u32 deferred_recv; + u32 duplicates; + u32 max_queue_sz; /* send queue size high water mark */ + u32 accu_queue_sz; /* used for send queue size profiling */ + u32 queue_sz_counts; /* used for send queue size profiling */ + u32 msg_length_counts; /* used for message length profiling */ + u32 msg_lengths_total; /* used for message length profiling */ + u32 msg_length_profile[7]; /* used for msg. length profiling */ +}; + +/** + * struct tipc_link - TIPC link data structure + * @addr: network address of link's peer node + * @name: link name character string + * @media_addr: media address to use when sending messages over link + * @timer: link timer + * @net: pointer to namespace struct + * @refcnt: reference counter for permanent references (owner node & timer) + * @peer_session: link session # being used by peer end of link + * @peer_bearer_id: bearer id used by link's peer endpoint + * @bearer_id: local bearer id used by link + * @tolerance: minimum link continuity loss needed to reset link [in ms] + * @keepalive_intv: link keepalive timer interval + * @abort_limit: # of unacknowledged continuity probes needed to reset link + * @state: current state of link FSM + * @peer_caps: bitmap describing capabilities of peer node + * @silent_intv_cnt: # of timer intervals without any reception from peer + * @proto_msg: template for control messages generated by link + * @pmsg: convenience pointer to "proto_msg" field + * @priority: current link priority + * @net_plane: current link network plane ('A' through 'H') + * @backlog_limit: backlog queue congestion thresholds (indexed by importance) + * @exp_msg_count: # of tunnelled messages expected during link changeover + * @reset_rcv_checkpt: seq # of last acknowledged message at time of link reset + * @mtu: current maximum packet size for this link + * @advertised_mtu: advertised own mtu when link is being established + * @transmitq: queue for sent, non-acked messages + * @backlogq: queue for messages waiting to be sent + * @snt_nxt: next sequence number to use for outbound messages + * @last_retransmitted: sequence number of most recently retransmitted message + * @stale_count: # of identical retransmit requests made by peer + * @ackers: # of peers that needs to ack each packet before it can be released + * @acked: # last packet acked by a certain peer. Used for broadcast. + * @rcv_nxt: next sequence number to expect for inbound messages + * @deferred_queue: deferred queue saved OOS b'cast message received from node + * @unacked_window: # of inbound messages rx'd without ack'ing back to peer + * @inputq: buffer queue for messages to be delivered upwards + * @namedq: buffer queue for name table messages to be delivered upwards + * @next_out: ptr to first unsent outbound message in queue + * @wakeupq: linked list of wakeup msgs waiting for link congestion to abate + * @long_msg_seq_no: next identifier to use for outbound fragmented messages + * @reasm_buf: head of partially reassembled inbound message fragments + * @bc_rcvr: marks that this is a broadcast receiver link + * @stats: collects statistics regarding link activity + */ +struct tipc_link { + u32 addr; + char name[TIPC_MAX_LINK_NAME]; + struct tipc_media_addr *media_addr; + struct net *net; + + /* Management and link supervision data */ + u32 peer_session; + u32 peer_bearer_id; + u32 bearer_id; + u32 tolerance; + unsigned long keepalive_intv; + u32 abort_limit; + u32 state; + u16 peer_caps; + bool active; + u32 silent_intv_cnt; + struct { + unchar hdr[INT_H_SIZE]; + unchar body[TIPC_MAX_IF_NAME]; + } proto_msg; + struct tipc_msg *pmsg; + u32 priority; + char net_plane; + + /* Failover/synch */ + u16 drop_point; + struct sk_buff *failover_reasm_skb; + + /* Max packet negotiation */ + u16 mtu; + u16 advertised_mtu; + + /* Sending */ + struct sk_buff_head transmq; + struct sk_buff_head backlogq; + struct { + u16 len; + u16 limit; + } backlog[5]; + u16 snd_nxt; + u16 last_retransm; + u16 window; + u32 stale_count; + + /* Reception */ + u16 rcv_nxt; + u32 rcv_unacked; + struct sk_buff_head deferdq; + struct sk_buff_head *inputq; + struct sk_buff_head *namedq; + + /* Congestion handling */ + struct sk_buff_head wakeupq; + + /* Fragmentation/reassembly */ + struct sk_buff *reasm_buf; + + /* Broadcast */ + u16 ackers; + u16 acked; + struct tipc_link *bc_rcvlink; + struct tipc_link *bc_sndlink; + int nack_state; + bool bc_peer_is_up; + + /* Statistics */ + struct tipc_stats stats; +}; + /* * Error message prefixes */ static const char *link_co_err = "Link tunneling error, "; static const char *link_rst_msg = "Resetting link "; -static const char tipc_bclink_name[] = "broadcast-link"; - -static const struct nla_policy tipc_nl_link_policy[TIPC_NLA_LINK_MAX + 1] = { - [TIPC_NLA_LINK_UNSPEC] = { .type = NLA_UNSPEC }, - [TIPC_NLA_LINK_NAME] = { - .type = NLA_STRING, - .len = TIPC_MAX_LINK_NAME - }, - [TIPC_NLA_LINK_MTU] = { .type = NLA_U32 }, - [TIPC_NLA_LINK_BROADCAST] = { .type = NLA_FLAG }, - [TIPC_NLA_LINK_UP] = { .type = NLA_FLAG }, - [TIPC_NLA_LINK_ACTIVE] = { .type = NLA_FLAG }, - [TIPC_NLA_LINK_PROP] = { .type = NLA_NESTED }, - [TIPC_NLA_LINK_STATS] = { .type = NLA_NESTED }, - [TIPC_NLA_LINK_RX] = { .type = NLA_U32 }, - [TIPC_NLA_LINK_TX] = { .type = NLA_U32 } -}; /* Properties valid for media, bearar and link */ static const struct nla_policy tipc_nl_prop_policy[TIPC_NLA_PROP_MAX + 1] = { @@ -117,8 +245,7 @@ static int tipc_link_proto_rcv(struct tipc_link *l, struct sk_buff *skb, static void tipc_link_build_proto_msg(struct tipc_link *l, int mtyp, bool probe, u16 rcvgap, int tolerance, int priority, struct sk_buff_head *xmitq); -static void link_reset_statistics(struct tipc_link *l_ptr); -static void link_print(struct tipc_link *l_ptr, const char *str); +static void link_print(struct tipc_link *l, const char *str); static void tipc_link_build_nack_msg(struct tipc_link *l, struct sk_buff_head *xmitq); static void tipc_link_build_bc_init_msg(struct tipc_link *l, @@ -183,6 +310,36 @@ void tipc_link_set_active(struct tipc_link *l, bool active) l->active = active; } +u32 tipc_link_id(struct tipc_link *l) +{ + return l->peer_bearer_id << 16 | l->bearer_id; +} + +int tipc_link_window(struct tipc_link *l) +{ + return l->window; +} + +int tipc_link_prio(struct tipc_link *l) +{ + return l->priority; +} + +unsigned long tipc_link_tolerance(struct tipc_link *l) +{ + return l->tolerance; +} + +struct sk_buff_head *tipc_link_inputq(struct tipc_link *l) +{ + return l->inputq; +} + +char tipc_link_plane(struct tipc_link *l) +{ + return l->net_plane; +} + void tipc_link_add_bc_peer(struct tipc_link *snd_l, struct tipc_link *uc_l, struct sk_buff_head *xmitq) @@ -227,11 +384,31 @@ int tipc_link_mtu(struct tipc_link *l) return l->mtu; } +u16 tipc_link_rcv_nxt(struct tipc_link *l) +{ + return l->rcv_nxt; +} + +u16 tipc_link_acked(struct tipc_link *l) +{ + return l->acked; +} + +char *tipc_link_name(struct tipc_link *l) +{ + return l->name; +} + static u32 link_own_addr(struct tipc_link *l) { return msg_prevnode(l->pmsg); } +void tipc_link_reinit(struct tipc_link *l, u32 addr) +{ + msg_set_prevnode(l->pmsg, addr); +} + /** * tipc_link_create - create a new link * @n: pointer to associated node @@ -694,7 +871,7 @@ void tipc_link_reset(struct tipc_link *l) l->stats.recv_info = 0; l->stale_count = 0; l->bc_peer_is_up = false; - link_reset_statistics(l); + tipc_link_reset_stats(l); } /** @@ -1087,8 +1264,9 @@ drop: /* * Send protocol message to the other endpoint. */ -void tipc_link_proto_xmit(struct tipc_link *l, u32 msg_typ, int probe_msg, - u32 gap, u32 tolerance, u32 priority) +static void tipc_link_proto_xmit(struct tipc_link *l, u32 msg_typ, + int probe_msg, u32 gap, u32 tolerance, + u32 priority) { struct sk_buff *skb = NULL; struct sk_buff_head xmitq; @@ -1262,6 +1440,8 @@ static int tipc_link_proto_rcv(struct tipc_link *l, struct sk_buff *skb, /* fall thru' */ case ACTIVATE_MSG: + skb_linearize(skb); + hdr = buf_msg(skb); /* Complete own link name with peer's interface name */ if_name = strrchr(l->name, ':') + 1; @@ -1527,53 +1707,17 @@ void tipc_link_set_queue_limits(struct tipc_link *l, u32 win) l->backlog[TIPC_SYSTEM_IMPORTANCE].limit = max_bulk; } -/* tipc_link_find_owner - locate owner node of link by link's name - * @net: the applicable net namespace - * @name: pointer to link name string - * @bearer_id: pointer to index in 'node->links' array where the link was found. - * - * Returns pointer to node owning the link, or 0 if no matching link is found. - */ -static struct tipc_node *tipc_link_find_owner(struct net *net, - const char *link_name, - unsigned int *bearer_id) -{ - struct tipc_net *tn = net_generic(net, tipc_net_id); - struct tipc_link *l_ptr; - struct tipc_node *n_ptr; - struct tipc_node *found_node = NULL; - int i; - - *bearer_id = 0; - rcu_read_lock(); - list_for_each_entry_rcu(n_ptr, &tn->node_list, list) { - tipc_node_lock(n_ptr); - for (i = 0; i < MAX_BEARERS; i++) { - l_ptr = n_ptr->links[i].link; - if (l_ptr && !strcmp(l_ptr->name, link_name)) { - *bearer_id = i; - found_node = n_ptr; - break; - } - } - tipc_node_unlock(n_ptr); - if (found_node) - break; - } - rcu_read_unlock(); - - return found_node; -} - /** - * link_reset_statistics - reset link statistics - * @l_ptr: pointer to link + * link_reset_stats - reset link statistics + * @l: pointer to link */ -static void link_reset_statistics(struct tipc_link *l_ptr) +void tipc_link_reset_stats(struct tipc_link *l) { - memset(&l_ptr->stats, 0, sizeof(l_ptr->stats)); - l_ptr->stats.sent_info = l_ptr->snd_nxt; - l_ptr->stats.recv_info = l_ptr->rcv_nxt; + memset(&l->stats, 0, sizeof(l->stats)); + if (!link_is_bc_sndlink(l)) { + l->stats.sent_info = l->snd_nxt; + l->stats.recv_info = l->rcv_nxt; + } } static void link_print(struct tipc_link *l, const char *str) @@ -1626,84 +1770,6 @@ int tipc_nl_parse_link_prop(struct nlattr *prop, struct nlattr *props[]) return 0; } -int tipc_nl_link_set(struct sk_buff *skb, struct genl_info *info) -{ - int err; - int res = 0; - int bearer_id; - char *name; - struct tipc_link *link; - struct tipc_node *node; - struct nlattr *attrs[TIPC_NLA_LINK_MAX + 1]; - struct net *net = sock_net(skb->sk); - - if (!info->attrs[TIPC_NLA_LINK]) - return -EINVAL; - - err = nla_parse_nested(attrs, TIPC_NLA_LINK_MAX, - info->attrs[TIPC_NLA_LINK], - tipc_nl_link_policy); - if (err) - return err; - - if (!attrs[TIPC_NLA_LINK_NAME]) - return -EINVAL; - - name = nla_data(attrs[TIPC_NLA_LINK_NAME]); - - if (strcmp(name, tipc_bclink_name) == 0) - return tipc_nl_bc_link_set(net, attrs); - - node = tipc_link_find_owner(net, name, &bearer_id); - if (!node) - return -EINVAL; - - tipc_node_lock(node); - - link = node->links[bearer_id].link; - if (!link) { - res = -EINVAL; - goto out; - } - - if (attrs[TIPC_NLA_LINK_PROP]) { - struct nlattr *props[TIPC_NLA_PROP_MAX + 1]; - - err = tipc_nl_parse_link_prop(attrs[TIPC_NLA_LINK_PROP], - props); - if (err) { - res = err; - goto out; - } - - if (props[TIPC_NLA_PROP_TOL]) { - u32 tol; - - tol = nla_get_u32(props[TIPC_NLA_PROP_TOL]); - link->tolerance = tol; - tipc_link_proto_xmit(link, STATE_MSG, 0, 0, tol, 0); - } - if (props[TIPC_NLA_PROP_PRIO]) { - u32 prio; - - prio = nla_get_u32(props[TIPC_NLA_PROP_PRIO]); - link->priority = prio; - tipc_link_proto_xmit(link, STATE_MSG, 0, 0, 0, prio); - } - if (props[TIPC_NLA_PROP_WIN]) { - u32 win; - - win = nla_get_u32(props[TIPC_NLA_PROP_WIN]); - tipc_link_set_queue_limits(link, win); - } - } - -out: - tipc_node_unlock(node); - - return res; -} - static int __tipc_nl_add_stats(struct sk_buff *skb, struct tipc_stats *s) { int i; @@ -1770,8 +1836,8 @@ msg_full: } /* Caller should hold appropriate locks to protect the link */ -static int __tipc_nl_add_link(struct net *net, struct tipc_nl_msg *msg, - struct tipc_link *link, int nlflags) +int __tipc_nl_add_link(struct net *net, struct tipc_nl_msg *msg, + struct tipc_link *link, int nlflags) { int err; void *hdr; @@ -1840,198 +1906,134 @@ msg_full: return -EMSGSIZE; } -/* Caller should hold node lock */ -static int __tipc_nl_add_node_links(struct net *net, struct tipc_nl_msg *msg, - struct tipc_node *node, u32 *prev_link) +static int __tipc_nl_add_bc_link_stat(struct sk_buff *skb, + struct tipc_stats *stats) { - u32 i; - int err; - - for (i = *prev_link; i < MAX_BEARERS; i++) { - *prev_link = i; - - if (!node->links[i].link) - continue; + int i; + struct nlattr *nest; - err = __tipc_nl_add_link(net, msg, - node->links[i].link, NLM_F_MULTI); - if (err) - return err; - } - *prev_link = 0; + struct nla_map { + __u32 key; + __u32 val; + }; - return 0; -} + struct nla_map map[] = { + {TIPC_NLA_STATS_RX_INFO, stats->recv_info}, + {TIPC_NLA_STATS_RX_FRAGMENTS, stats->recv_fragments}, + {TIPC_NLA_STATS_RX_FRAGMENTED, stats->recv_fragmented}, + {TIPC_NLA_STATS_RX_BUNDLES, stats->recv_bundles}, + {TIPC_NLA_STATS_RX_BUNDLED, stats->recv_bundled}, + {TIPC_NLA_STATS_TX_INFO, stats->sent_info}, + {TIPC_NLA_STATS_TX_FRAGMENTS, stats->sent_fragments}, + {TIPC_NLA_STATS_TX_FRAGMENTED, stats->sent_fragmented}, + {TIPC_NLA_STATS_TX_BUNDLES, stats->sent_bundles}, + {TIPC_NLA_STATS_TX_BUNDLED, stats->sent_bundled}, + {TIPC_NLA_STATS_RX_NACKS, stats->recv_nacks}, + {TIPC_NLA_STATS_RX_DEFERRED, stats->deferred_recv}, + {TIPC_NLA_STATS_TX_NACKS, stats->sent_nacks}, + {TIPC_NLA_STATS_TX_ACKS, stats->sent_acks}, + {TIPC_NLA_STATS_RETRANSMITTED, stats->retransmitted}, + {TIPC_NLA_STATS_DUPLICATES, stats->duplicates}, + {TIPC_NLA_STATS_LINK_CONGS, stats->link_congs}, + {TIPC_NLA_STATS_MAX_QUEUE, stats->max_queue_sz}, + {TIPC_NLA_STATS_AVG_QUEUE, stats->queue_sz_counts ? + (stats->accu_queue_sz / stats->queue_sz_counts) : 0} + }; -int tipc_nl_link_dump(struct sk_buff *skb, struct netlink_callback *cb) -{ - struct net *net = sock_net(skb->sk); - struct tipc_net *tn = net_generic(net, tipc_net_id); - struct tipc_node *node; - struct tipc_nl_msg msg; - u32 prev_node = cb->args[0]; - u32 prev_link = cb->args[1]; - int done = cb->args[2]; - int err; + nest = nla_nest_start(skb, TIPC_NLA_LINK_STATS); + if (!nest) + return -EMSGSIZE; - if (done) - return 0; + for (i = 0; i < ARRAY_SIZE(map); i++) + if (nla_put_u32(skb, map[i].key, map[i].val)) + goto msg_full; - msg.skb = skb; - msg.portid = NETLINK_CB(cb->skb).portid; - msg.seq = cb->nlh->nlmsg_seq; - - rcu_read_lock(); - if (prev_node) { - node = tipc_node_find(net, prev_node); - if (!node) { - /* We never set seq or call nl_dump_check_consistent() - * this means that setting prev_seq here will cause the - * consistence check to fail in the netlink callback - * handler. Resulting in the last NLMSG_DONE message - * having the NLM_F_DUMP_INTR flag set. - */ - cb->prev_seq = 1; - goto out; - } - tipc_node_put(node); - - list_for_each_entry_continue_rcu(node, &tn->node_list, - list) { - tipc_node_lock(node); - err = __tipc_nl_add_node_links(net, &msg, node, - &prev_link); - tipc_node_unlock(node); - if (err) - goto out; - - prev_node = node->addr; - } - } else { - err = tipc_nl_add_bc_link(net, &msg); - if (err) - goto out; - - list_for_each_entry_rcu(node, &tn->node_list, list) { - tipc_node_lock(node); - err = __tipc_nl_add_node_links(net, &msg, node, - &prev_link); - tipc_node_unlock(node); - if (err) - goto out; - - prev_node = node->addr; - } - } - done = 1; -out: - rcu_read_unlock(); + nla_nest_end(skb, nest); - cb->args[0] = prev_node; - cb->args[1] = prev_link; - cb->args[2] = done; + return 0; +msg_full: + nla_nest_cancel(skb, nest); - return skb->len; + return -EMSGSIZE; } -int tipc_nl_link_get(struct sk_buff *skb, struct genl_info *info) +int tipc_nl_add_bc_link(struct net *net, struct tipc_nl_msg *msg) { - struct net *net = genl_info_net(info); - struct tipc_nl_msg msg; - char *name; int err; + void *hdr; + struct nlattr *attrs; + struct nlattr *prop; + struct tipc_net *tn = net_generic(net, tipc_net_id); + struct tipc_link *bcl = tn->bcl; - msg.portid = info->snd_portid; - msg.seq = info->snd_seq; - - if (!info->attrs[TIPC_NLA_LINK_NAME]) - return -EINVAL; - name = nla_data(info->attrs[TIPC_NLA_LINK_NAME]); + if (!bcl) + return 0; - msg.skb = nlmsg_new(NLMSG_GOODSIZE, GFP_KERNEL); - if (!msg.skb) - return -ENOMEM; + tipc_bcast_lock(net); - if (strcmp(name, tipc_bclink_name) == 0) { - err = tipc_nl_add_bc_link(net, &msg); - if (err) { - nlmsg_free(msg.skb); - return err; - } - } else { - int bearer_id; - struct tipc_node *node; - struct tipc_link *link; + hdr = genlmsg_put(msg->skb, msg->portid, msg->seq, &tipc_genl_family, + NLM_F_MULTI, TIPC_NL_LINK_GET); + if (!hdr) + return -EMSGSIZE; - node = tipc_link_find_owner(net, name, &bearer_id); - if (!node) - return -EINVAL; + attrs = nla_nest_start(msg->skb, TIPC_NLA_LINK); + if (!attrs) + goto msg_full; - tipc_node_lock(node); - link = node->links[bearer_id].link; - if (!link) { - tipc_node_unlock(node); - nlmsg_free(msg.skb); - return -EINVAL; - } + /* The broadcast link is always up */ + if (nla_put_flag(msg->skb, TIPC_NLA_LINK_UP)) + goto attr_msg_full; - err = __tipc_nl_add_link(net, &msg, link, 0); - tipc_node_unlock(node); - if (err) { - nlmsg_free(msg.skb); - return err; - } - } + if (nla_put_flag(msg->skb, TIPC_NLA_LINK_BROADCAST)) + goto attr_msg_full; + if (nla_put_string(msg->skb, TIPC_NLA_LINK_NAME, bcl->name)) + goto attr_msg_full; + if (nla_put_u32(msg->skb, TIPC_NLA_LINK_RX, bcl->rcv_nxt)) + goto attr_msg_full; + if (nla_put_u32(msg->skb, TIPC_NLA_LINK_TX, bcl->snd_nxt)) + goto attr_msg_full; - return genlmsg_reply(msg.skb, info); -} + prop = nla_nest_start(msg->skb, TIPC_NLA_LINK_PROP); + if (!prop) + goto attr_msg_full; + if (nla_put_u32(msg->skb, TIPC_NLA_PROP_WIN, bcl->window)) + goto prop_msg_full; + nla_nest_end(msg->skb, prop); -int tipc_nl_link_reset_stats(struct sk_buff *skb, struct genl_info *info) -{ - int err; - char *link_name; - unsigned int bearer_id; - struct tipc_link *link; - struct tipc_node *node; - struct nlattr *attrs[TIPC_NLA_LINK_MAX + 1]; - struct net *net = sock_net(skb->sk); - - if (!info->attrs[TIPC_NLA_LINK]) - return -EINVAL; - - err = nla_parse_nested(attrs, TIPC_NLA_LINK_MAX, - info->attrs[TIPC_NLA_LINK], - tipc_nl_link_policy); + err = __tipc_nl_add_bc_link_stat(msg->skb, &bcl->stats); if (err) - return err; - - if (!attrs[TIPC_NLA_LINK_NAME]) - return -EINVAL; - - link_name = nla_data(attrs[TIPC_NLA_LINK_NAME]); + goto attr_msg_full; - if (strcmp(link_name, tipc_bclink_name) == 0) { - err = tipc_bclink_reset_stats(net); - if (err) - return err; - return 0; - } + tipc_bcast_unlock(net); + nla_nest_end(msg->skb, attrs); + genlmsg_end(msg->skb, hdr); - node = tipc_link_find_owner(net, link_name, &bearer_id); - if (!node) - return -EINVAL; + return 0; - tipc_node_lock(node); +prop_msg_full: + nla_nest_cancel(msg->skb, prop); +attr_msg_full: + nla_nest_cancel(msg->skb, attrs); +msg_full: + tipc_bcast_unlock(net); + genlmsg_cancel(msg->skb, hdr); - link = node->links[bearer_id].link; - if (!link) { - tipc_node_unlock(node); - return -EINVAL; - } + return -EMSGSIZE; +} - link_reset_statistics(link); +void tipc_link_set_tolerance(struct tipc_link *l, u32 tol) +{ + l->tolerance = tol; + tipc_link_proto_xmit(l, STATE_MSG, 0, 0, tol, 0); +} - tipc_node_unlock(node); +void tipc_link_set_prio(struct tipc_link *l, u32 prio) +{ + l->priority = prio; + tipc_link_proto_xmit(l, STATE_MSG, 0, 0, 0, prio); +} - return 0; +void tipc_link_set_abort_limit(struct tipc_link *l, u32 limit) +{ + l->abort_limit = limit; } diff --git a/net/tipc/link.h b/net/tipc/link.h index 66d859b..b2ae0f4 100644 --- a/net/tipc/link.h +++ b/net/tipc/link.h @@ -45,10 +45,6 @@ */ #define ELINKCONG EAGAIN /* link congestion <=> resource unavailable */ -/* Out-of-range value for link sequence numbers - */ -#define INVALID_LINK_SEQ 0x10000 - /* Link FSM events: */ enum { @@ -75,151 +71,6 @@ enum { */ #define MAX_PKT_DEFAULT 1500 -struct tipc_stats { - u32 sent_info; /* used in counting # sent packets */ - u32 recv_info; /* used in counting # recv'd packets */ - u32 sent_states; - u32 recv_states; - u32 sent_probes; - u32 recv_probes; - u32 sent_nacks; - u32 recv_nacks; - u32 sent_acks; - u32 sent_bundled; - u32 sent_bundles; - u32 recv_bundled; - u32 recv_bundles; - u32 retransmitted; - u32 sent_fragmented; - u32 sent_fragments; - u32 recv_fragmented; - u32 recv_fragments; - u32 link_congs; /* # port sends blocked by congestion */ - u32 deferred_recv; - u32 duplicates; - u32 max_queue_sz; /* send queue size high water mark */ - u32 accu_queue_sz; /* used for send queue size profiling */ - u32 queue_sz_counts; /* used for send queue size profiling */ - u32 msg_length_counts; /* used for message length profiling */ - u32 msg_lengths_total; /* used for message length profiling */ - u32 msg_length_profile[7]; /* used for msg. length profiling */ -}; - -/** - * struct tipc_link - TIPC link data structure - * @addr: network address of link's peer node - * @name: link name character string - * @media_addr: media address to use when sending messages over link - * @timer: link timer - * @net: pointer to namespace struct - * @refcnt: reference counter for permanent references (owner node & timer) - * @peer_session: link session # being used by peer end of link - * @peer_bearer_id: bearer id used by link's peer endpoint - * @bearer_id: local bearer id used by link - * @tolerance: minimum link continuity loss needed to reset link [in ms] - * @keepalive_intv: link keepalive timer interval - * @abort_limit: # of unacknowledged continuity probes needed to reset link - * @state: current state of link FSM - * @peer_caps: bitmap describing capabilities of peer node - * @silent_intv_cnt: # of timer intervals without any reception from peer - * @proto_msg: template for control messages generated by link - * @pmsg: convenience pointer to "proto_msg" field - * @priority: current link priority - * @net_plane: current link network plane ('A' through 'H') - * @backlog_limit: backlog queue congestion thresholds (indexed by importance) - * @exp_msg_count: # of tunnelled messages expected during link changeover - * @reset_rcv_checkpt: seq # of last acknowledged message at time of link reset - * @mtu: current maximum packet size for this link - * @advertised_mtu: advertised own mtu when link is being established - * @transmitq: queue for sent, non-acked messages - * @backlogq: queue for messages waiting to be sent - * @snt_nxt: next sequence number to use for outbound messages - * @last_retransmitted: sequence number of most recently retransmitted message - * @stale_count: # of identical retransmit requests made by peer - * @ackers: # of peers that needs to ack each packet before it can be released - * @acked: # last packet acked by a certain peer. Used for broadcast. - * @rcv_nxt: next sequence number to expect for inbound messages - * @deferred_queue: deferred queue saved OOS b'cast message received from node - * @unacked_window: # of inbound messages rx'd without ack'ing back to peer - * @inputq: buffer queue for messages to be delivered upwards - * @namedq: buffer queue for name table messages to be delivered upwards - * @next_out: ptr to first unsent outbound message in queue - * @wakeupq: linked list of wakeup msgs waiting for link congestion to abate - * @long_msg_seq_no: next identifier to use for outbound fragmented messages - * @reasm_buf: head of partially reassembled inbound message fragments - * @bc_rcvr: marks that this is a broadcast receiver link - * @stats: collects statistics regarding link activity - */ -struct tipc_link { - u32 addr; - char name[TIPC_MAX_LINK_NAME]; - struct tipc_media_addr *media_addr; - struct net *net; - - /* Management and link supervision data */ - u32 peer_session; - u32 peer_bearer_id; - u32 bearer_id; - u32 tolerance; - unsigned long keepalive_intv; - u32 abort_limit; - u32 state; - u16 peer_caps; - bool active; - u32 silent_intv_cnt; - struct { - unchar hdr[INT_H_SIZE]; - unchar body[TIPC_MAX_IF_NAME]; - } proto_msg; - struct tipc_msg *pmsg; - u32 priority; - char net_plane; - - /* Failover/synch */ - u16 drop_point; - struct sk_buff *failover_reasm_skb; - - /* Max packet negotiation */ - u16 mtu; - u16 advertised_mtu; - - /* Sending */ - struct sk_buff_head transmq; - struct sk_buff_head backlogq; - struct { - u16 len; - u16 limit; - } backlog[5]; - u16 snd_nxt; - u16 last_retransm; - u16 window; - u32 stale_count; - - /* Reception */ - u16 rcv_nxt; - u32 rcv_unacked; - struct sk_buff_head deferdq; - struct sk_buff_head *inputq; - struct sk_buff_head *namedq; - - /* Congestion handling */ - struct sk_buff_head wakeupq; - - /* Fragmentation/reassembly */ - struct sk_buff *reasm_buf; - - /* Broadcast */ - u16 ackers; - u16 acked; - struct tipc_link *bc_rcvlink; - struct tipc_link *bc_sndlink; - int nack_state; - bool bc_peer_is_up; - - /* Statistics */ - struct tipc_stats stats; -}; - bool tipc_link_create(struct net *net, char *if_name, int bearer_id, int tolerance, char net_plane, u32 mtu, int priority, int window, u32 session, u32 ownnode, u32 peer, @@ -235,11 +86,11 @@ bool tipc_link_bc_create(struct net *net, u32 ownnode, u32 peer, struct sk_buff_head *namedq, struct tipc_link *bc_sndlink, struct tipc_link **link); +void tipc_link_reinit(struct tipc_link *l, u32 addr); void tipc_link_tnl_prepare(struct tipc_link *l, struct tipc_link *tnl, int mtyp, struct sk_buff_head *xmitq); void tipc_link_build_reset_msg(struct tipc_link *l, struct sk_buff_head *xmitq); int tipc_link_fsm_evt(struct tipc_link *l, int evt); -void tipc_link_reset_fragments(struct tipc_link *l_ptr); bool tipc_link_is_up(struct tipc_link *l); bool tipc_link_peer_is_down(struct tipc_link *l); bool tipc_link_is_reset(struct tipc_link *l); @@ -248,15 +99,25 @@ bool tipc_link_is_synching(struct tipc_link *l); bool tipc_link_is_failingover(struct tipc_link *l); bool tipc_link_is_blocked(struct tipc_link *l); void tipc_link_set_active(struct tipc_link *l, bool active); -void tipc_link_reset(struct tipc_link *l_ptr); -int tipc_link_xmit(struct tipc_link *link, struct sk_buff_head *list, +void tipc_link_reset(struct tipc_link *l); +void tipc_link_reset_stats(struct tipc_link *l); +int tipc_link_xmit(struct tipc_link *link, struct sk_buff_head *list, struct sk_buff_head *xmitq); +struct sk_buff_head *tipc_link_inputq(struct tipc_link *l); +u16 tipc_link_rcv_nxt(struct tipc_link *l); +u16 tipc_link_acked(struct tipc_link *l); +u32 tipc_link_id(struct tipc_link *l); +char *tipc_link_name(struct tipc_link *l); +char tipc_link_plane(struct tipc_link *l); +int tipc_link_prio(struct tipc_link *l); +int tipc_link_window(struct tipc_link *l); +unsigned long tipc_link_tolerance(struct tipc_link *l); +void tipc_link_set_tolerance(struct tipc_link *l, u32 tol); +void tipc_link_set_prio(struct tipc_link *l, u32 prio); +void tipc_link_set_abort_limit(struct tipc_link *l, u32 limit); void tipc_link_set_queue_limits(struct tipc_link *l, u32 window); - -int tipc_nl_link_dump(struct sk_buff *skb, struct netlink_callback *cb); -int tipc_nl_link_get(struct sk_buff *skb, struct genl_info *info); -int tipc_nl_link_set(struct sk_buff *skb, struct genl_info *info); -int tipc_nl_link_reset_stats(struct sk_buff *skb, struct genl_info *info); +int __tipc_nl_add_link(struct net *net, struct tipc_nl_msg *msg, + struct tipc_link *link, int nlflags); int tipc_nl_parse_link_prop(struct nlattr *prop, struct nlattr *props[]); int tipc_link_timeout(struct tipc_link *l, struct sk_buff_head *xmitq); int tipc_link_rcv(struct tipc_link *l, struct sk_buff *skb, diff --git a/net/tipc/name_distr.c b/net/tipc/name_distr.c index c07612b..ebe9d0f 100644 --- a/net/tipc/name_distr.c +++ b/net/tipc/name_distr.c @@ -84,31 +84,6 @@ static struct sk_buff *named_prepare_buf(struct net *net, u32 type, u32 size, return buf; } -void named_cluster_distribute(struct net *net, struct sk_buff *skb) -{ - struct tipc_net *tn = net_generic(net, tipc_net_id); - struct sk_buff *oskb; - struct tipc_node *node; - u32 dnode; - - rcu_read_lock(); - list_for_each_entry_rcu(node, &tn->node_list, list) { - dnode = node->addr; - if (in_own_node(net, dnode)) - continue; - if (!tipc_node_is_up(node)) - continue; - oskb = pskb_copy(skb, GFP_ATOMIC); - if (!oskb) - break; - msg_set_destnode(buf_msg(oskb), dnode); - tipc_node_xmit_skb(net, oskb, dnode, 0); - } - rcu_read_unlock(); - - kfree_skb(skb); -} - /** * tipc_named_publish - tell other nodes about a new publication by this node */ @@ -226,42 +201,6 @@ void tipc_named_node_up(struct net *net, u32 dnode) tipc_node_xmit(net, &head, dnode, 0); } -static void tipc_publ_subscribe(struct net *net, struct publication *publ, - u32 addr) -{ - struct tipc_node *node; - - if (in_own_node(net, addr)) - return; - - node = tipc_node_find(net, addr); - if (!node) { - pr_warn("Node subscription rejected, unknown node 0x%x\n", - addr); - return; - } - - tipc_node_lock(node); - list_add_tail(&publ->nodesub_list, &node->publ_list); - tipc_node_unlock(node); - tipc_node_put(node); -} - -static void tipc_publ_unsubscribe(struct net *net, struct publication *publ, - u32 addr) -{ - struct tipc_node *node; - - node = tipc_node_find(net, addr); - if (!node) - return; - - tipc_node_lock(node); - list_del_init(&publ->nodesub_list); - tipc_node_unlock(node); - tipc_node_put(node); -} - /** * tipc_publ_purge - remove publication associated with a failed node * @@ -277,7 +216,7 @@ static void tipc_publ_purge(struct net *net, struct publication *publ, u32 addr) p = tipc_nametbl_remove_publ(net, publ->type, publ->lower, publ->node, publ->ref, publ->key); if (p) - tipc_publ_unsubscribe(net, p, addr); + tipc_node_unsubscribe(net, &p->nodesub_list, addr); spin_unlock_bh(&tn->nametbl_lock); if (p != publ) { @@ -317,7 +256,7 @@ static bool tipc_update_nametbl(struct net *net, struct distr_item *i, TIPC_CLUSTER_SCOPE, node, ntohl(i->ref), ntohl(i->key)); if (publ) { - tipc_publ_subscribe(net, publ, node); + tipc_node_subscribe(net, &publ->nodesub_list, node); return true; } } else if (dtype == WITHDRAWAL) { @@ -326,7 +265,7 @@ static bool tipc_update_nametbl(struct net *net, struct distr_item *i, node, ntohl(i->ref), ntohl(i->key)); if (publ) { - tipc_publ_unsubscribe(net, publ, node); + tipc_node_unsubscribe(net, &publ->nodesub_list, node); kfree_rcu(publ, rcu); return true; } @@ -397,6 +336,7 @@ void tipc_named_rcv(struct net *net, struct sk_buff_head *inputq) spin_lock_bh(&tn->nametbl_lock); for (skb = skb_dequeue(inputq); skb; skb = skb_dequeue(inputq)) { + skb_linearize(skb); msg = buf_msg(skb); mtype = msg_type(msg); item = (struct distr_item *)msg_data(msg); diff --git a/net/tipc/name_distr.h b/net/tipc/name_distr.h index dd2d9fd..1264ba0 100644 --- a/net/tipc/name_distr.h +++ b/net/tipc/name_distr.h @@ -69,7 +69,6 @@ struct distr_item { struct sk_buff *tipc_named_publish(struct net *net, struct publication *publ); struct sk_buff *tipc_named_withdraw(struct net *net, struct publication *publ); -void named_cluster_distribute(struct net *net, struct sk_buff *buf); void tipc_named_node_up(struct net *net, u32 dnode); void tipc_named_rcv(struct net *net, struct sk_buff_head *msg_queue); void tipc_named_reinit(struct net *net); diff --git a/net/tipc/name_table.c b/net/tipc/name_table.c index 0f47f08..91fce70 100644 --- a/net/tipc/name_table.c +++ b/net/tipc/name_table.c @@ -42,6 +42,7 @@ #include "subscr.h" #include "bcast.h" #include "addr.h" +#include "node.h" #include <net/genetlink.h> #define TIPC_NAMETBL_SIZE 1024 /* must be a power of 2 */ @@ -677,7 +678,7 @@ struct publication *tipc_nametbl_publish(struct net *net, u32 type, u32 lower, spin_unlock_bh(&tn->nametbl_lock); if (buf) - named_cluster_distribute(net, buf); + tipc_node_broadcast(net, buf); return publ; } @@ -709,7 +710,7 @@ int tipc_nametbl_withdraw(struct net *net, u32 type, u32 lower, u32 ref, spin_unlock_bh(&tn->nametbl_lock); if (skb) { - named_cluster_distribute(net, skb); + tipc_node_broadcast(net, skb); return 1; } return 0; diff --git a/net/tipc/netlink.c b/net/tipc/netlink.c index 7f6475e..8975b01 100644 --- a/net/tipc/netlink.c +++ b/net/tipc/netlink.c @@ -101,18 +101,18 @@ static const struct genl_ops tipc_genl_v2_ops[] = { }, { .cmd = TIPC_NL_LINK_GET, - .doit = tipc_nl_link_get, - .dumpit = tipc_nl_link_dump, + .doit = tipc_nl_node_get_link, + .dumpit = tipc_nl_node_dump_link, .policy = tipc_nl_policy, }, { .cmd = TIPC_NL_LINK_SET, - .doit = tipc_nl_link_set, + .doit = tipc_nl_node_set_link, .policy = tipc_nl_policy, }, { .cmd = TIPC_NL_LINK_RESET_STATS, - .doit = tipc_nl_link_reset_stats, + .doit = tipc_nl_node_reset_link_stats, .policy = tipc_nl_policy, }, { diff --git a/net/tipc/netlink_compat.c b/net/tipc/netlink_compat.c index 1eadc95..2c016fd 100644 --- a/net/tipc/netlink_compat.c +++ b/net/tipc/netlink_compat.c @@ -1023,25 +1023,25 @@ static int tipc_nl_compat_handle(struct tipc_nl_compat_msg *msg) msg->req_type = TIPC_TLV_LINK_NAME; msg->rep_size = ULTRA_STRING_MAX_LEN; msg->rep_type = TIPC_TLV_ULTRA_STRING; - dump.dumpit = tipc_nl_link_dump; + dump.dumpit = tipc_nl_node_dump_link; dump.format = tipc_nl_compat_link_stat_dump; return tipc_nl_compat_dumpit(&dump, msg); case TIPC_CMD_GET_LINKS: msg->req_type = TIPC_TLV_NET_ADDR; msg->rep_size = ULTRA_STRING_MAX_LEN; - dump.dumpit = tipc_nl_link_dump; + dump.dumpit = tipc_nl_node_dump_link; dump.format = tipc_nl_compat_link_dump; return tipc_nl_compat_dumpit(&dump, msg); case TIPC_CMD_SET_LINK_TOL: case TIPC_CMD_SET_LINK_PRI: case TIPC_CMD_SET_LINK_WINDOW: msg->req_type = TIPC_TLV_LINK_CONFIG; - doit.doit = tipc_nl_link_set; + doit.doit = tipc_nl_node_set_link; doit.transcode = tipc_nl_compat_link_set; return tipc_nl_compat_doit(&doit, msg); case TIPC_CMD_RESET_LINK_STATS: msg->req_type = TIPC_TLV_LINK_NAME; - doit.doit = tipc_nl_link_reset_stats; + doit.doit = tipc_nl_node_reset_link_stats; doit.transcode = tipc_nl_compat_link_reset_stats; return tipc_nl_compat_doit(&doit, msg); case TIPC_CMD_SHOW_NAME_TABLE: diff --git a/net/tipc/node.c b/net/tipc/node.c index 20cddec..fa97d96 100644 --- a/net/tipc/node.c +++ b/net/tipc/node.c @@ -42,6 +42,84 @@ #include "bcast.h" #include "discover.h" +#define INVALID_NODE_SIG 0x10000 + +/* Flags used to take different actions according to flag type + * TIPC_NOTIFY_NODE_DOWN: notify node is down + * TIPC_NOTIFY_NODE_UP: notify node is up + * TIPC_DISTRIBUTE_NAME: publish or withdraw link state name type + */ +enum { + TIPC_NOTIFY_NODE_DOWN = (1 << 3), + TIPC_NOTIFY_NODE_UP = (1 << 4), + TIPC_NOTIFY_LINK_UP = (1 << 6), + TIPC_NOTIFY_LINK_DOWN = (1 << 7) +}; + +struct tipc_link_entry { + struct tipc_link *link; + spinlock_t lock; /* per link */ + u32 mtu; + struct sk_buff_head inputq; + struct tipc_media_addr maddr; +}; + +struct tipc_bclink_entry { + struct tipc_link *link; + struct sk_buff_head inputq1; + struct sk_buff_head arrvq; + struct sk_buff_head inputq2; + struct sk_buff_head namedq; +}; + +/** + * struct tipc_node - TIPC node structure + * @addr: network address of node + * @ref: reference counter to node object + * @lock: rwlock governing access to structure + * @net: the applicable net namespace + * @hash: links to adjacent nodes in unsorted hash chain + * @inputq: pointer to input queue containing messages for msg event + * @namedq: pointer to name table input queue with name table messages + * @active_links: bearer ids of active links, used as index into links[] array + * @links: array containing references to all links to node + * @action_flags: bit mask of different types of node actions + * @state: connectivity state vs peer node + * @sync_point: sequence number where synch/failover is finished + * @list: links to adjacent nodes in sorted list of cluster's nodes + * @working_links: number of working links to node (both active and standby) + * @link_cnt: number of links to node + * @capabilities: bitmap, indicating peer node's functional capabilities + * @signature: node instance identifier + * @link_id: local and remote bearer ids of changing link, if any + * @publ_list: list of publications + * @rcu: rcu struct for tipc_node + */ +struct tipc_node { + u32 addr; + struct kref kref; + rwlock_t lock; + struct net *net; + struct hlist_node hash; + int active_links[2]; + struct tipc_link_entry links[MAX_BEARERS]; + struct tipc_bclink_entry bc_entry; + int action_flags; + struct list_head list; + int state; + u16 sync_point; + int link_cnt; + u16 working_links; + u16 capabilities; + u32 signature; + u32 link_id; + struct list_head publ_list; + struct list_head conn_sks; + unsigned long keepalive_intv; + struct timer_list timer; + struct rcu_head rcu; +}; + /* Node FSM states and events: */ enum { @@ -75,6 +153,9 @@ static void node_lost_contact(struct tipc_node *n, struct sk_buff_head *inputq); static void tipc_node_delete(struct tipc_node *node); static void tipc_node_timeout(unsigned long data); static void tipc_node_fsm_evt(struct tipc_node *n, int evt); +static struct tipc_node *tipc_node_find(struct net *net, u32 addr); +static void tipc_node_put(struct tipc_node *node); +static bool tipc_node_is_up(struct tipc_node *n); struct tipc_sock_conn { u32 port; @@ -83,12 +164,54 @@ struct tipc_sock_conn { struct list_head list; }; +static const struct nla_policy tipc_nl_link_policy[TIPC_NLA_LINK_MAX + 1] = { + [TIPC_NLA_LINK_UNSPEC] = { .type = NLA_UNSPEC }, + [TIPC_NLA_LINK_NAME] = { + .type = NLA_STRING, + .len = TIPC_MAX_LINK_NAME + }, + [TIPC_NLA_LINK_MTU] = { .type = NLA_U32 }, + [TIPC_NLA_LINK_BROADCAST] = { .type = NLA_FLAG }, + [TIPC_NLA_LINK_UP] = { .type = NLA_FLAG }, + [TIPC_NLA_LINK_ACTIVE] = { .type = NLA_FLAG }, + [TIPC_NLA_LINK_PROP] = { .type = NLA_NESTED }, + [TIPC_NLA_LINK_STATS] = { .type = NLA_NESTED }, + [TIPC_NLA_LINK_RX] = { .type = NLA_U32 }, + [TIPC_NLA_LINK_TX] = { .type = NLA_U32 } +}; + static const struct nla_policy tipc_nl_node_policy[TIPC_NLA_NODE_MAX + 1] = { [TIPC_NLA_NODE_UNSPEC] = { .type = NLA_UNSPEC }, [TIPC_NLA_NODE_ADDR] = { .type = NLA_U32 }, [TIPC_NLA_NODE_UP] = { .type = NLA_FLAG } }; +static struct tipc_link *node_active_link(struct tipc_node *n, int sel) +{ + int bearer_id = n->active_links[sel & 1]; + + if (unlikely(bearer_id == INVALID_BEARER_ID)) + return NULL; + + return n->links[bearer_id].link; +} + +int tipc_node_get_mtu(struct net *net, u32 addr, u32 sel) +{ + struct tipc_node *n; + int bearer_id; + unsigned int mtu = MAX_MSG_SIZE; + + n = tipc_node_find(net, addr); + if (unlikely(!n)) + return mtu; + + bearer_id = n->active_links[sel & 1]; + if (likely(bearer_id != INVALID_BEARER_ID)) + mtu = n->links[bearer_id].mtu; + tipc_node_put(n); + return mtu; +} /* * A trivial power-of-two bitmask technique is used for speed, since this * operation is done for every incoming TIPC packet. The number of hash table @@ -107,7 +230,7 @@ static void tipc_node_kref_release(struct kref *kref) tipc_node_delete(node); } -void tipc_node_put(struct tipc_node *node) +static void tipc_node_put(struct tipc_node *node) { kref_put(&node->kref, tipc_node_kref_release); } @@ -120,7 +243,7 @@ static void tipc_node_get(struct tipc_node *node) /* * tipc_node_find - locate specified node object, if it exists */ -struct tipc_node *tipc_node_find(struct net *net, u32 addr) +static struct tipc_node *tipc_node_find(struct net *net, u32 addr) { struct tipc_net *tn = net_generic(net, tipc_net_id); struct tipc_node *node; @@ -141,66 +264,122 @@ struct tipc_node *tipc_node_find(struct net *net, u32 addr) return NULL; } +static void tipc_node_read_lock(struct tipc_node *n) +{ + read_lock_bh(&n->lock); +} + +static void tipc_node_read_unlock(struct tipc_node *n) +{ + read_unlock_bh(&n->lock); +} + +static void tipc_node_write_lock(struct tipc_node *n) +{ + write_lock_bh(&n->lock); +} + +static void tipc_node_write_unlock(struct tipc_node *n) +{ + struct net *net = n->net; + u32 addr = 0; + u32 flags = n->action_flags; + u32 link_id = 0; + struct list_head *publ_list; + + if (likely(!flags)) { + write_unlock_bh(&n->lock); + return; + } + + addr = n->addr; + link_id = n->link_id; + publ_list = &n->publ_list; + + n->action_flags &= ~(TIPC_NOTIFY_NODE_DOWN | TIPC_NOTIFY_NODE_UP | + TIPC_NOTIFY_LINK_DOWN | TIPC_NOTIFY_LINK_UP); + + write_unlock_bh(&n->lock); + + if (flags & TIPC_NOTIFY_NODE_DOWN) + tipc_publ_notify(net, publ_list, addr); + + if (flags & TIPC_NOTIFY_NODE_UP) + tipc_named_node_up(net, addr); + + if (flags & TIPC_NOTIFY_LINK_UP) + tipc_nametbl_publish(net, TIPC_LINK_STATE, addr, addr, + TIPC_NODE_SCOPE, link_id, addr); + + if (flags & TIPC_NOTIFY_LINK_DOWN) + tipc_nametbl_withdraw(net, TIPC_LINK_STATE, addr, + link_id, addr); +} + struct tipc_node *tipc_node_create(struct net *net, u32 addr, u16 capabilities) { struct tipc_net *tn = net_generic(net, tipc_net_id); - struct tipc_node *n_ptr, *temp_node; + struct tipc_node *n, *temp_node; + int i; spin_lock_bh(&tn->node_list_lock); - n_ptr = tipc_node_find(net, addr); - if (n_ptr) + n = tipc_node_find(net, addr); + if (n) goto exit; - n_ptr = kzalloc(sizeof(*n_ptr), GFP_ATOMIC); - if (!n_ptr) { + n = kzalloc(sizeof(*n), GFP_ATOMIC); + if (!n) { pr_warn("Node creation failed, no memory\n"); goto exit; } - n_ptr->addr = addr; - n_ptr->net = net; - n_ptr->capabilities = capabilities; - kref_init(&n_ptr->kref); - spin_lock_init(&n_ptr->lock); - INIT_HLIST_NODE(&n_ptr->hash); - INIT_LIST_HEAD(&n_ptr->list); - INIT_LIST_HEAD(&n_ptr->publ_list); - INIT_LIST_HEAD(&n_ptr->conn_sks); - skb_queue_head_init(&n_ptr->bc_entry.namedq); - skb_queue_head_init(&n_ptr->bc_entry.inputq1); - __skb_queue_head_init(&n_ptr->bc_entry.arrvq); - skb_queue_head_init(&n_ptr->bc_entry.inputq2); - hlist_add_head_rcu(&n_ptr->hash, &tn->node_htable[tipc_hashfn(addr)]); + n->addr = addr; + n->net = net; + n->capabilities = capabilities; + kref_init(&n->kref); + rwlock_init(&n->lock); + INIT_HLIST_NODE(&n->hash); + INIT_LIST_HEAD(&n->list); + INIT_LIST_HEAD(&n->publ_list); + INIT_LIST_HEAD(&n->conn_sks); + skb_queue_head_init(&n->bc_entry.namedq); + skb_queue_head_init(&n->bc_entry.inputq1); + __skb_queue_head_init(&n->bc_entry.arrvq); + skb_queue_head_init(&n->bc_entry.inputq2); + for (i = 0; i < MAX_BEARERS; i++) + spin_lock_init(&n->links[i].lock); + hlist_add_head_rcu(&n->hash, &tn->node_htable[tipc_hashfn(addr)]); list_for_each_entry_rcu(temp_node, &tn->node_list, list) { - if (n_ptr->addr < temp_node->addr) + if (n->addr < temp_node->addr) break; } - list_add_tail_rcu(&n_ptr->list, &temp_node->list); - n_ptr->state = SELF_DOWN_PEER_LEAVING; - n_ptr->signature = INVALID_NODE_SIG; - n_ptr->active_links[0] = INVALID_BEARER_ID; - n_ptr->active_links[1] = INVALID_BEARER_ID; - if (!tipc_link_bc_create(net, tipc_own_addr(net), n_ptr->addr, - U16_MAX, tipc_bc_sndlink(net)->window, - n_ptr->capabilities, - &n_ptr->bc_entry.inputq1, - &n_ptr->bc_entry.namedq, + list_add_tail_rcu(&n->list, &temp_node->list); + n->state = SELF_DOWN_PEER_LEAVING; + n->signature = INVALID_NODE_SIG; + n->active_links[0] = INVALID_BEARER_ID; + n->active_links[1] = INVALID_BEARER_ID; + if (!tipc_link_bc_create(net, tipc_own_addr(net), n->addr, + U16_MAX, + tipc_link_window(tipc_bc_sndlink(net)), + n->capabilities, + &n->bc_entry.inputq1, + &n->bc_entry.namedq, tipc_bc_sndlink(net), - &n_ptr->bc_entry.link)) { + &n->bc_entry.link)) { pr_warn("Broadcast rcv link creation failed, no memory\n"); - kfree(n_ptr); - n_ptr = NULL; + kfree(n); + n = NULL; goto exit; } - tipc_node_get(n_ptr); - setup_timer(&n_ptr->timer, tipc_node_timeout, (unsigned long)n_ptr); - n_ptr->keepalive_intv = U32_MAX; + tipc_node_get(n); + setup_timer(&n->timer, tipc_node_timeout, (unsigned long)n); + n->keepalive_intv = U32_MAX; exit: spin_unlock_bh(&tn->node_list_lock); - return n_ptr; + return n; } static void tipc_node_calculate_timer(struct tipc_node *n, struct tipc_link *l) { - unsigned long tol = l->tolerance; + unsigned long tol = tipc_link_tolerance(l); unsigned long intv = ((tol / 4) > 500) ? 500 : tol / 4; unsigned long keepalive_intv = msecs_to_jiffies(intv); @@ -209,7 +388,7 @@ static void tipc_node_calculate_timer(struct tipc_node *n, struct tipc_link *l) n->keepalive_intv = keepalive_intv; /* Ensure link's abort limit corresponds to current interval */ - l->abort_limit = l->tolerance / jiffies_to_msecs(n->keepalive_intv); + tipc_link_set_abort_limit(l, tol / jiffies_to_msecs(n->keepalive_intv)); } static void tipc_node_delete(struct tipc_node *node) @@ -234,6 +413,42 @@ void tipc_node_stop(struct net *net) spin_unlock_bh(&tn->node_list_lock); } +void tipc_node_subscribe(struct net *net, struct list_head *subscr, u32 addr) +{ + struct tipc_node *n; + + if (in_own_node(net, addr)) + return; + + n = tipc_node_find(net, addr); + if (!n) { + pr_warn("Node subscribe rejected, unknown node 0x%x\n", addr); + return; + } + tipc_node_write_lock(n); + list_add_tail(subscr, &n->publ_list); + tipc_node_write_unlock(n); + tipc_node_put(n); +} + +void tipc_node_unsubscribe(struct net *net, struct list_head *subscr, u32 addr) +{ + struct tipc_node *n; + + if (in_own_node(net, addr)) + return; + + n = tipc_node_find(net, addr); + if (!n) { + pr_warn("Node unsubscribe rejected, unknown node 0x%x\n", addr); + return; + } + tipc_node_write_lock(n); + list_del_init(subscr); + tipc_node_write_unlock(n); + tipc_node_put(n); +} + int tipc_node_add_conn(struct net *net, u32 dnode, u32 port, u32 peer_port) { struct tipc_node *node; @@ -257,9 +472,9 @@ int tipc_node_add_conn(struct net *net, u32 dnode, u32 port, u32 peer_port) conn->port = port; conn->peer_port = peer_port; - tipc_node_lock(node); + tipc_node_write_lock(node); list_add_tail(&conn->list, &node->conn_sks); - tipc_node_unlock(node); + tipc_node_write_unlock(node); exit: tipc_node_put(node); return err; @@ -277,14 +492,14 @@ void tipc_node_remove_conn(struct net *net, u32 dnode, u32 port) if (!node) return; - tipc_node_lock(node); + tipc_node_write_lock(node); list_for_each_entry_safe(conn, safe, &node->conn_sks, list) { if (port != conn->port) continue; list_del(&conn->list); kfree(conn); } - tipc_node_unlock(node); + tipc_node_write_unlock(node); tipc_node_put(node); } @@ -301,14 +516,16 @@ static void tipc_node_timeout(unsigned long data) __skb_queue_head_init(&xmitq); for (bearer_id = 0; bearer_id < MAX_BEARERS; bearer_id++) { - tipc_node_lock(n); + tipc_node_read_lock(n); le = &n->links[bearer_id]; + spin_lock_bh(&le->lock); if (le->link) { /* Link tolerance may change asynchronously: */ tipc_node_calculate_timer(n, le->link); rc = tipc_link_timeout(le->link, &xmitq); } - tipc_node_unlock(n); + spin_unlock_bh(&le->lock); + tipc_node_read_unlock(n); tipc_bearer_xmit(n->net, bearer_id, &xmitq, &le->maddr); if (rc & TIPC_LINK_DOWN_EVT) tipc_node_link_down(n, bearer_id, false); @@ -340,16 +557,16 @@ static void __tipc_node_link_up(struct tipc_node *n, int bearer_id, n->working_links++; n->action_flags |= TIPC_NOTIFY_LINK_UP; - n->link_id = nl->peer_bearer_id << 16 | bearer_id; + n->link_id = tipc_link_id(nl); /* Leave room for tunnel header when returning 'mtu' to users: */ - n->links[bearer_id].mtu = nl->mtu - INT_H_SIZE; + n->links[bearer_id].mtu = tipc_link_mtu(nl) - INT_H_SIZE; tipc_bearer_add_dest(n->net, bearer_id, n->addr); tipc_bcast_inc_bearer_dst_cnt(n->net, bearer_id); pr_debug("Established link <%s> on network plane %c\n", - nl->name, nl->net_plane); + tipc_link_name(nl), tipc_link_plane(nl)); /* First link? => give it both slots */ if (!ol) { @@ -362,17 +579,17 @@ static void __tipc_node_link_up(struct tipc_node *n, int bearer_id, } /* Second link => redistribute slots */ - if (nl->priority > ol->priority) { - pr_debug("Old link <%s> becomes standby\n", ol->name); + if (tipc_link_prio(nl) > tipc_link_prio(ol)) { + pr_debug("Old link <%s> becomes standby\n", tipc_link_name(ol)); *slot0 = bearer_id; *slot1 = bearer_id; tipc_link_set_active(nl, true); tipc_link_set_active(ol, false); - } else if (nl->priority == ol->priority) { + } else if (tipc_link_prio(nl) == tipc_link_prio(ol)) { tipc_link_set_active(nl, true); *slot1 = bearer_id; } else { - pr_debug("New link <%s> is standby\n", nl->name); + pr_debug("New link <%s> is standby\n", tipc_link_name(nl)); } /* Prepare synchronization with first link */ @@ -387,9 +604,9 @@ static void __tipc_node_link_up(struct tipc_node *n, int bearer_id, static void tipc_node_link_up(struct tipc_node *n, int bearer_id, struct sk_buff_head *xmitq) { - tipc_node_lock(n); + tipc_node_write_lock(n); __tipc_node_link_up(n, bearer_id, xmitq); - tipc_node_unlock(n); + tipc_node_write_unlock(n); } /** @@ -402,7 +619,7 @@ static void __tipc_node_link_down(struct tipc_node *n, int *bearer_id, struct tipc_link_entry *le = &n->links[*bearer_id]; int *slot0 = &n->active_links[0]; int *slot1 = &n->active_links[1]; - int i, highest = 0; + int i, highest = 0, prio; struct tipc_link *l, *_l, *tnl; l = n->links[*bearer_id].link; @@ -411,12 +628,12 @@ static void __tipc_node_link_down(struct tipc_node *n, int *bearer_id, n->working_links--; n->action_flags |= TIPC_NOTIFY_LINK_DOWN; - n->link_id = l->peer_bearer_id << 16 | *bearer_id; + n->link_id = tipc_link_id(l); tipc_bearer_remove_dest(n->net, *bearer_id, n->addr); pr_debug("Lost link <%s> on network plane %c\n", - l->name, l->net_plane); + tipc_link_name(l), tipc_link_plane(l)); /* Select new active link if any available */ *slot0 = INVALID_BEARER_ID; @@ -427,10 +644,11 @@ static void __tipc_node_link_down(struct tipc_node *n, int *bearer_id, continue; if (_l == l) continue; - if (_l->priority < highest) + prio = tipc_link_prio(_l); + if (prio < highest) continue; - if (_l->priority > highest) { - highest = _l->priority; + if (prio > highest) { + highest = prio; *slot0 = i; *slot1 = i; continue; @@ -453,17 +671,17 @@ static void __tipc_node_link_down(struct tipc_node *n, int *bearer_id, tipc_bcast_dec_bearer_dst_cnt(n->net, *bearer_id); /* There is still a working link => initiate failover */ - tnl = node_active_link(n, 0); + *bearer_id = n->active_links[0]; + tnl = n->links[*bearer_id].link; tipc_link_fsm_evt(tnl, LINK_SYNCH_END_EVT); tipc_node_fsm_evt(n, NODE_SYNCH_END_EVT); - n->sync_point = tnl->rcv_nxt + (U16_MAX / 2 - 1); + n->sync_point = tipc_link_rcv_nxt(tnl) + (U16_MAX / 2 - 1); tipc_link_tnl_prepare(l, tnl, FAILOVER_MSG, xmitq); tipc_link_reset(l); tipc_link_fsm_evt(l, LINK_RESET_EVT); tipc_link_fsm_evt(l, LINK_FAILOVER_BEGIN_EVT); tipc_node_fsm_evt(n, NODE_FAILOVER_BEGIN_EVT); - *maddr = &n->links[tnl->bearer_id].maddr; - *bearer_id = tnl->bearer_id; + *maddr = &n->links[*bearer_id].maddr; } static void tipc_node_link_down(struct tipc_node *n, int bearer_id, bool delete) @@ -478,7 +696,7 @@ static void tipc_node_link_down(struct tipc_node *n, int bearer_id, bool delete) __skb_queue_head_init(&xmitq); - tipc_node_lock(n); + tipc_node_write_lock(n); if (!tipc_link_is_establishing(l)) { __tipc_node_link_down(n, &bearer_id, &xmitq, &maddr); if (delete) { @@ -490,12 +708,12 @@ static void tipc_node_link_down(struct tipc_node *n, int bearer_id, bool delete) /* Defuse pending tipc_node_link_up() */ tipc_link_fsm_evt(l, LINK_RESET_EVT); } - tipc_node_unlock(n); + tipc_node_write_unlock(n); tipc_bearer_xmit(n->net, bearer_id, &xmitq, maddr); tipc_sk_rcv(n->net, &le->inputq); } -bool tipc_node_is_up(struct tipc_node *n) +static bool tipc_node_is_up(struct tipc_node *n) { return n->active_links[0] != INVALID_BEARER_ID; } @@ -523,7 +741,7 @@ void tipc_node_check_dest(struct net *net, u32 onode, if (!n) return; - tipc_node_lock(n); + tipc_node_write_lock(n); le = &n->links[b->identity]; @@ -626,7 +844,7 @@ void tipc_node_check_dest(struct net *net, u32 onode, } memcpy(&le->maddr, maddr, sizeof(*maddr)); exit: - tipc_node_unlock(n); + tipc_node_write_unlock(n); if (reset && !tipc_link_is_reset(l)) tipc_node_link_down(n, b->identity, false); tipc_node_put(n); @@ -834,24 +1052,6 @@ illegal_evt: pr_err("Illegal node fsm evt %x in state %x\n", evt, state); } -bool tipc_node_filter_pkt(struct tipc_node *n, struct tipc_msg *hdr) -{ - int state = n->state; - - if (likely(state == SELF_UP_PEER_UP)) - return true; - - if (state == SELF_LEAVING_PEER_DOWN) - return false; - - if (state == SELF_DOWN_PEER_LEAVING) { - if (msg_peer_node_is_up(hdr)) - return false; - } - - return true; -} - static void node_lost_contact(struct tipc_node *n, struct sk_buff_head *inputq) { @@ -913,56 +1113,18 @@ int tipc_node_get_linkname(struct net *net, u32 bearer_id, u32 addr, if (bearer_id >= MAX_BEARERS) goto exit; - tipc_node_lock(node); + tipc_node_read_lock(node); link = node->links[bearer_id].link; if (link) { - strncpy(linkname, link->name, len); + strncpy(linkname, tipc_link_name(link), len); err = 0; } exit: - tipc_node_unlock(node); + tipc_node_read_unlock(node); tipc_node_put(node); return err; } -void tipc_node_unlock(struct tipc_node *node) -{ - struct net *net = node->net; - u32 addr = 0; - u32 flags = node->action_flags; - u32 link_id = 0; - struct list_head *publ_list; - - if (likely(!flags)) { - spin_unlock_bh(&node->lock); - return; - } - - addr = node->addr; - link_id = node->link_id; - publ_list = &node->publ_list; - - node->action_flags &= ~(TIPC_NOTIFY_NODE_DOWN | TIPC_NOTIFY_NODE_UP | - TIPC_NOTIFY_LINK_DOWN | TIPC_NOTIFY_LINK_UP); - - spin_unlock_bh(&node->lock); - - if (flags & TIPC_NOTIFY_NODE_DOWN) - tipc_publ_notify(net, publ_list, addr); - - if (flags & TIPC_NOTIFY_NODE_UP) - tipc_named_node_up(net, addr); - - if (flags & TIPC_NOTIFY_LINK_UP) - tipc_nametbl_publish(net, TIPC_LINK_STATE, addr, addr, - TIPC_NODE_SCOPE, link_id, addr); - - if (flags & TIPC_NOTIFY_LINK_DOWN) - tipc_nametbl_withdraw(net, TIPC_LINK_STATE, addr, - link_id, addr); - -} - /* Caller should hold node lock for the passed node */ static int __tipc_nl_add_node(struct tipc_nl_msg *msg, struct tipc_node *node) { @@ -997,20 +1159,6 @@ msg_full: return -EMSGSIZE; } -static struct tipc_link *tipc_node_select_link(struct tipc_node *n, int sel, - int *bearer_id, - struct tipc_media_addr **maddr) -{ - int id = n->active_links[sel & 1]; - - if (unlikely(id < 0)) - return NULL; - - *bearer_id = id; - *maddr = &n->links[id].maddr; - return n->links[id].link; -} - /** * tipc_node_xmit() is the general link level function for message sending * @net: the applicable net namespace @@ -1023,29 +1171,32 @@ static struct tipc_link *tipc_node_select_link(struct tipc_node *n, int sel, int tipc_node_xmit(struct net *net, struct sk_buff_head *list, u32 dnode, int selector) { - struct tipc_link *l = NULL; + struct tipc_link_entry *le = NULL; struct tipc_node *n; struct sk_buff_head xmitq; - struct tipc_media_addr *maddr; - int bearer_id; + int bearer_id = -1; int rc = -EHOSTUNREACH; __skb_queue_head_init(&xmitq); n = tipc_node_find(net, dnode); if (likely(n)) { - tipc_node_lock(n); - l = tipc_node_select_link(n, selector, &bearer_id, &maddr); - if (likely(l)) - rc = tipc_link_xmit(l, list, &xmitq); - tipc_node_unlock(n); - if (unlikely(rc == -ENOBUFS)) + tipc_node_read_lock(n); + bearer_id = n->active_links[selector & 1]; + if (bearer_id >= 0) { + le = &n->links[bearer_id]; + spin_lock_bh(&le->lock); + rc = tipc_link_xmit(le->link, list, &xmitq); + spin_unlock_bh(&le->lock); + } + tipc_node_read_unlock(n); + if (likely(!rc)) + tipc_bearer_xmit(net, bearer_id, &xmitq, &le->maddr); + else if (rc == -ENOBUFS) tipc_node_link_down(n, bearer_id, false); tipc_node_put(n); + return rc; } - if (likely(!rc)) { - tipc_bearer_xmit(net, bearer_id, &xmitq, maddr); - return 0; - } + if (likely(in_own_node(net, dnode))) { tipc_sk_rcv(net, list); return 0; @@ -1075,6 +1226,30 @@ int tipc_node_xmit_skb(struct net *net, struct sk_buff *skb, u32 dnode, return 0; } +void tipc_node_broadcast(struct net *net, struct sk_buff *skb) +{ + struct sk_buff *txskb; + struct tipc_node *n; + u32 dst; + + rcu_read_lock(); + list_for_each_entry_rcu(n, tipc_nodes(net), list) { + dst = n->addr; + if (in_own_node(net, dst)) + continue; + if (!tipc_node_is_up(n)) + continue; + txskb = pskb_copy(skb, GFP_ATOMIC); + if (!txskb) + break; + msg_set_destnode(buf_msg(txskb), dst); + tipc_node_xmit_skb(net, txskb, dst, 0); + } + rcu_read_unlock(); + + kfree_skb(skb); +} + /** * tipc_node_bc_rcv - process TIPC broadcast packet arriving from off-node * @net: the applicable net namespace @@ -1116,9 +1291,9 @@ static void tipc_node_bc_rcv(struct net *net, struct sk_buff *skb, int bearer_id /* Broadcast ACKs are sent on a unicast link */ if (rc & TIPC_LINK_SND_BC_ACK) { - tipc_node_lock(n); + tipc_node_read_lock(n); tipc_link_build_ack_msg(le->link, &xmitq); - tipc_node_unlock(n); + tipc_node_read_unlock(n); } if (!skb_queue_empty(&xmitq)) @@ -1151,30 +1326,30 @@ static bool tipc_node_check_state(struct tipc_node *n, struct sk_buff *skb, u16 oseqno = msg_seqno(hdr); u16 iseqno = msg_seqno(msg_get_wrapped(hdr)); u16 exp_pkts = msg_msgcnt(hdr); - u16 rcv_nxt, syncpt, dlv_nxt; + u16 rcv_nxt, syncpt, dlv_nxt, inputq_len; int state = n->state; struct tipc_link *l, *tnl, *pl = NULL; struct tipc_media_addr *maddr; - int i, pb_id; + int pb_id; l = n->links[bearer_id].link; if (!l) return false; - rcv_nxt = l->rcv_nxt; + rcv_nxt = tipc_link_rcv_nxt(l); if (likely((state == SELF_UP_PEER_UP) && (usr != TUNNEL_PROTOCOL))) return true; /* Find parallel link, if any */ - for (i = 0; i < MAX_BEARERS; i++) { - if ((i != bearer_id) && n->links[i].link) { - pl = n->links[i].link; + for (pb_id = 0; pb_id < MAX_BEARERS; pb_id++) { + if ((pb_id != bearer_id) && n->links[pb_id].link) { + pl = n->links[pb_id].link; break; } } - /* Update node accesibility if applicable */ + /* Check and update node accesibility if applicable */ if (state == SELF_UP_PEER_COMING) { if (!tipc_link_is_up(l)) return true; @@ -1187,8 +1362,12 @@ static bool tipc_node_check_state(struct tipc_node *n, struct sk_buff *skb, if (msg_peer_node_is_up(hdr)) return false; tipc_node_fsm_evt(n, PEER_LOST_CONTACT_EVT); + return true; } + if (state == SELF_LEAVING_PEER_DOWN) + return false; + /* Ignore duplicate packets */ if ((usr != LINK_PROTOCOL) && less(oseqno, rcv_nxt)) return true; @@ -1197,9 +1376,9 @@ static bool tipc_node_check_state(struct tipc_node *n, struct sk_buff *skb, if ((usr == TUNNEL_PROTOCOL) && (mtyp == FAILOVER_MSG)) { syncpt = oseqno + exp_pkts - 1; if (pl && tipc_link_is_up(pl)) { - pb_id = pl->bearer_id; __tipc_node_link_down(n, &pb_id, xmitq, &maddr); - tipc_skb_queue_splice_tail_init(pl->inputq, l->inputq); + tipc_skb_queue_splice_tail_init(tipc_link_inputq(pl), + tipc_link_inputq(l)); } /* If pkts arrive out of order, use lowest calculated syncpt */ if (less(syncpt, n->sync_point)) @@ -1232,19 +1411,18 @@ static bool tipc_node_check_state(struct tipc_node *n, struct sk_buff *skb, tipc_link_fsm_evt(l, LINK_SYNCH_BEGIN_EVT); tipc_node_fsm_evt(n, NODE_SYNCH_BEGIN_EVT); } - if (less(syncpt, n->sync_point)) - n->sync_point = syncpt; } /* Open tunnel link when parallel link reaches synch point */ - if ((n->state == NODE_SYNCHING) && tipc_link_is_synching(l)) { + if (n->state == NODE_SYNCHING) { if (tipc_link_is_synching(l)) { tnl = l; } else { tnl = pl; pl = l; } - dlv_nxt = pl->rcv_nxt - mod(skb_queue_len(pl->inputq)); + inputq_len = skb_queue_len(tipc_link_inputq(pl)); + dlv_nxt = tipc_link_rcv_nxt(pl) - inputq_len; if (more(dlv_nxt, n->sync_point)) { tipc_link_fsm_evt(tnl, LINK_SYNCH_END_EVT); tipc_node_fsm_evt(n, NODE_SYNCH_END_EVT); @@ -1304,22 +1482,32 @@ void tipc_rcv(struct net *net, struct sk_buff *skb, struct tipc_bearer *b) /* Ensure broadcast reception is in synch with peer's send state */ if (unlikely(usr == LINK_PROTOCOL)) tipc_bcast_sync_rcv(net, n->bc_entry.link, hdr); - else if (unlikely(n->bc_entry.link->acked != bc_ack)) + else if (unlikely(tipc_link_acked(n->bc_entry.link) != bc_ack)) tipc_bcast_ack_rcv(net, n->bc_entry.link, bc_ack); - tipc_node_lock(n); - - /* Is reception permitted at the moment ? */ - if (!tipc_node_filter_pkt(n, hdr)) - goto unlock; - - /* Check and if necessary update node state */ - if (likely(tipc_node_check_state(n, skb, bearer_id, &xmitq))) { - rc = tipc_link_rcv(le->link, skb, &xmitq); - skb = NULL; + /* Receive packet directly if conditions permit */ + tipc_node_read_lock(n); + if (likely((n->state == SELF_UP_PEER_UP) && (usr != TUNNEL_PROTOCOL))) { + spin_lock_bh(&le->lock); + if (le->link) { + rc = tipc_link_rcv(le->link, skb, &xmitq); + skb = NULL; + } + spin_unlock_bh(&le->lock); + } + tipc_node_read_unlock(n); + + /* Check/update node state before receiving */ + if (unlikely(skb)) { + tipc_node_write_lock(n); + if (tipc_node_check_state(n, skb, bearer_id, &xmitq)) { + if (le->link) { + rc = tipc_link_rcv(le->link, skb, &xmitq); + skb = NULL; + } + } + tipc_node_write_unlock(n); } -unlock: - tipc_node_unlock(n); if (unlikely(rc & TIPC_LINK_UP_EVT)) tipc_node_link_up(n, bearer_id, &xmitq); @@ -1384,15 +1572,15 @@ int tipc_nl_node_dump(struct sk_buff *skb, struct netlink_callback *cb) continue; } - tipc_node_lock(node); + tipc_node_read_lock(node); err = __tipc_nl_add_node(&msg, node); if (err) { last_addr = node->addr; - tipc_node_unlock(node); + tipc_node_read_unlock(node); goto out; } - tipc_node_unlock(node); + tipc_node_read_unlock(node); } done = 1; out: @@ -1402,3 +1590,314 @@ out: return skb->len; } + +/* tipc_node_find_by_name - locate owner node of link by link's name + * @net: the applicable net namespace + * @name: pointer to link name string + * @bearer_id: pointer to index in 'node->links' array where the link was found. + * + * Returns pointer to node owning the link, or 0 if no matching link is found. + */ +static struct tipc_node *tipc_node_find_by_name(struct net *net, + const char *link_name, + unsigned int *bearer_id) +{ + struct tipc_net *tn = net_generic(net, tipc_net_id); + struct tipc_link *l; + struct tipc_node *n; + struct tipc_node *found_node = NULL; + int i; + + *bearer_id = 0; + rcu_read_lock(); + list_for_each_entry_rcu(n, &tn->node_list, list) { + tipc_node_read_lock(n); + for (i = 0; i < MAX_BEARERS; i++) { + l = n->links[i].link; + if (l && !strcmp(tipc_link_name(l), link_name)) { + *bearer_id = i; + found_node = n; + break; + } + } + tipc_node_read_unlock(n); + if (found_node) + break; + } + rcu_read_unlock(); + + return found_node; +} + +int tipc_nl_node_set_link(struct sk_buff *skb, struct genl_info *info) +{ + int err; + int res = 0; + int bearer_id; + char *name; + struct tipc_link *link; + struct tipc_node *node; + struct nlattr *attrs[TIPC_NLA_LINK_MAX + 1]; + struct net *net = sock_net(skb->sk); + + if (!info->attrs[TIPC_NLA_LINK]) + return -EINVAL; + + err = nla_parse_nested(attrs, TIPC_NLA_LINK_MAX, + info->attrs[TIPC_NLA_LINK], + tipc_nl_link_policy); + if (err) + return err; + + if (!attrs[TIPC_NLA_LINK_NAME]) + return -EINVAL; + + name = nla_data(attrs[TIPC_NLA_LINK_NAME]); + + if (strcmp(name, tipc_bclink_name) == 0) + return tipc_nl_bc_link_set(net, attrs); + + node = tipc_node_find_by_name(net, name, &bearer_id); + if (!node) + return -EINVAL; + + tipc_node_read_lock(node); + + link = node->links[bearer_id].link; + if (!link) { + res = -EINVAL; + goto out; + } + + if (attrs[TIPC_NLA_LINK_PROP]) { + struct nlattr *props[TIPC_NLA_PROP_MAX + 1]; + + err = tipc_nl_parse_link_prop(attrs[TIPC_NLA_LINK_PROP], + props); + if (err) { + res = err; + goto out; + } + + if (props[TIPC_NLA_PROP_TOL]) { + u32 tol; + + tol = nla_get_u32(props[TIPC_NLA_PROP_TOL]); + tipc_link_set_tolerance(link, tol); + } + if (props[TIPC_NLA_PROP_PRIO]) { + u32 prio; + + prio = nla_get_u32(props[TIPC_NLA_PROP_PRIO]); + tipc_link_set_prio(link, prio); + } + if (props[TIPC_NLA_PROP_WIN]) { + u32 win; + + win = nla_get_u32(props[TIPC_NLA_PROP_WIN]); + tipc_link_set_queue_limits(link, win); + } + } + +out: + tipc_node_read_unlock(node); + + return res; +} + +int tipc_nl_node_get_link(struct sk_buff *skb, struct genl_info *info) +{ + struct net *net = genl_info_net(info); + struct tipc_nl_msg msg; + char *name; + int err; + + msg.portid = info->snd_portid; + msg.seq = info->snd_seq; + + if (!info->attrs[TIPC_NLA_LINK_NAME]) + return -EINVAL; + name = nla_data(info->attrs[TIPC_NLA_LINK_NAME]); + + msg.skb = nlmsg_new(NLMSG_GOODSIZE, GFP_KERNEL); + if (!msg.skb) + return -ENOMEM; + + if (strcmp(name, tipc_bclink_name) == 0) { + err = tipc_nl_add_bc_link(net, &msg); + if (err) { + nlmsg_free(msg.skb); + return err; + } + } else { + int bearer_id; + struct tipc_node *node; + struct tipc_link *link; + + node = tipc_node_find_by_name(net, name, &bearer_id); + if (!node) + return -EINVAL; + + tipc_node_read_lock(node); + link = node->links[bearer_id].link; + if (!link) { + tipc_node_read_unlock(node); + nlmsg_free(msg.skb); + return -EINVAL; + } + + err = __tipc_nl_add_link(net, &msg, link, 0); + tipc_node_read_unlock(node); + if (err) { + nlmsg_free(msg.skb); + return err; + } + } + + return genlmsg_reply(msg.skb, info); +} + +int tipc_nl_node_reset_link_stats(struct sk_buff *skb, struct genl_info *info) +{ + int err; + char *link_name; + unsigned int bearer_id; + struct tipc_link *link; + struct tipc_node *node; + struct nlattr *attrs[TIPC_NLA_LINK_MAX + 1]; + struct net *net = sock_net(skb->sk); + struct tipc_link_entry *le; + + if (!info->attrs[TIPC_NLA_LINK]) + return -EINVAL; + + err = nla_parse_nested(attrs, TIPC_NLA_LINK_MAX, + info->attrs[TIPC_NLA_LINK], + tipc_nl_link_policy); + if (err) + return err; + + if (!attrs[TIPC_NLA_LINK_NAME]) + return -EINVAL; + + link_name = nla_data(attrs[TIPC_NLA_LINK_NAME]); + + if (strcmp(link_name, tipc_bclink_name) == 0) { + err = tipc_bclink_reset_stats(net); + if (err) + return err; + return 0; + } + + node = tipc_node_find_by_name(net, link_name, &bearer_id); + if (!node) + return -EINVAL; + + le = &node->links[bearer_id]; + tipc_node_read_lock(node); + spin_lock_bh(&le->lock); + link = node->links[bearer_id].link; + if (!link) { + spin_unlock_bh(&le->lock); + tipc_node_read_unlock(node); + return -EINVAL; + } + tipc_link_reset_stats(link); + spin_unlock_bh(&le->lock); + tipc_node_read_unlock(node); + return 0; +} + +/* Caller should hold node lock */ +static int __tipc_nl_add_node_links(struct net *net, struct tipc_nl_msg *msg, + struct tipc_node *node, u32 *prev_link) +{ + u32 i; + int err; + + for (i = *prev_link; i < MAX_BEARERS; i++) { + *prev_link = i; + + if (!node->links[i].link) + continue; + + err = __tipc_nl_add_link(net, msg, + node->links[i].link, NLM_F_MULTI); + if (err) + return err; + } + *prev_link = 0; + + return 0; +} + +int tipc_nl_node_dump_link(struct sk_buff *skb, struct netlink_callback *cb) +{ + struct net *net = sock_net(skb->sk); + struct tipc_net *tn = net_generic(net, tipc_net_id); + struct tipc_node *node; + struct tipc_nl_msg msg; + u32 prev_node = cb->args[0]; + u32 prev_link = cb->args[1]; + int done = cb->args[2]; + int err; + + if (done) + return 0; + + msg.skb = skb; + msg.portid = NETLINK_CB(cb->skb).portid; + msg.seq = cb->nlh->nlmsg_seq; + + rcu_read_lock(); + if (prev_node) { + node = tipc_node_find(net, prev_node); + if (!node) { + /* We never set seq or call nl_dump_check_consistent() + * this means that setting prev_seq here will cause the + * consistence check to fail in the netlink callback + * handler. Resulting in the last NLMSG_DONE message + * having the NLM_F_DUMP_INTR flag set. + */ + cb->prev_seq = 1; + goto out; + } + tipc_node_put(node); + + list_for_each_entry_continue_rcu(node, &tn->node_list, + list) { + tipc_node_read_lock(node); + err = __tipc_nl_add_node_links(net, &msg, node, + &prev_link); + tipc_node_read_unlock(node); + if (err) + goto out; + + prev_node = node->addr; + } + } else { + err = tipc_nl_add_bc_link(net, &msg); + if (err) + goto out; + + list_for_each_entry_rcu(node, &tn->node_list, list) { + tipc_node_read_lock(node); + err = __tipc_nl_add_node_links(net, &msg, node, + &prev_link); + tipc_node_read_unlock(node); + if (err) + goto out; + + prev_node = node->addr; + } + } + done = 1; +out: + rcu_read_unlock(); + + cb->args[0] = prev_node; + cb->args[1] = prev_link; + cb->args[2] = done; + + return skb->len; +} diff --git a/net/tipc/node.h b/net/tipc/node.h index 6734562..f39d9d0 100644 --- a/net/tipc/node.h +++ b/net/tipc/node.h @@ -42,23 +42,6 @@ #include "bearer.h" #include "msg.h" -/* Out-of-range value for node signature */ -#define INVALID_NODE_SIG 0x10000 - -#define INVALID_BEARER_ID -1 - -/* Flags used to take different actions according to flag type - * TIPC_NOTIFY_NODE_DOWN: notify node is down - * TIPC_NOTIFY_NODE_UP: notify node is up - * TIPC_DISTRIBUTE_NAME: publish or withdraw link state name type - */ -enum { - TIPC_NOTIFY_NODE_DOWN = (1 << 3), - TIPC_NOTIFY_NODE_UP = (1 << 4), - TIPC_NOTIFY_LINK_UP = (1 << 6), - TIPC_NOTIFY_LINK_DOWN = (1 << 7) -}; - /* Optional capabilities supported by this code version */ enum { @@ -66,72 +49,8 @@ enum { }; #define TIPC_NODE_CAPABILITIES TIPC_BCAST_SYNCH +#define INVALID_BEARER_ID -1 -struct tipc_link_entry { - struct tipc_link *link; - u32 mtu; - struct sk_buff_head inputq; - struct tipc_media_addr maddr; -}; - -struct tipc_bclink_entry { - struct tipc_link *link; - struct sk_buff_head inputq1; - struct sk_buff_head arrvq; - struct sk_buff_head inputq2; - struct sk_buff_head namedq; -}; - -/** - * struct tipc_node - TIPC node structure - * @addr: network address of node - * @ref: reference counter to node object - * @lock: spinlock governing access to structure - * @net: the applicable net namespace - * @hash: links to adjacent nodes in unsorted hash chain - * @inputq: pointer to input queue containing messages for msg event - * @namedq: pointer to name table input queue with name table messages - * @active_links: bearer ids of active links, used as index into links[] array - * @links: array containing references to all links to node - * @action_flags: bit mask of different types of node actions - * @state: connectivity state vs peer node - * @sync_point: sequence number where synch/failover is finished - * @list: links to adjacent nodes in sorted list of cluster's nodes - * @working_links: number of working links to node (both active and standby) - * @link_cnt: number of links to node - * @capabilities: bitmap, indicating peer node's functional capabilities - * @signature: node instance identifier - * @link_id: local and remote bearer ids of changing link, if any - * @publ_list: list of publications - * @rcu: rcu struct for tipc_node - */ -struct tipc_node { - u32 addr; - struct kref kref; - spinlock_t lock; - struct net *net; - struct hlist_node hash; - int active_links[2]; - struct tipc_link_entry links[MAX_BEARERS]; - struct tipc_bclink_entry bc_entry; - int action_flags; - struct list_head list; - int state; - u16 sync_point; - int link_cnt; - u16 working_links; - u16 capabilities; - u32 signature; - u32 link_id; - struct list_head publ_list; - struct list_head conn_sks; - unsigned long keepalive_intv; - struct timer_list timer; - struct rcu_head rcu; -}; - -struct tipc_node *tipc_node_find(struct net *net, u32 addr); -void tipc_node_put(struct tipc_node *node); void tipc_node_stop(struct net *net); void tipc_node_check_dest(struct net *net, u32 onode, struct tipc_bearer *bearer, @@ -139,50 +58,22 @@ void tipc_node_check_dest(struct net *net, u32 onode, struct tipc_media_addr *maddr, bool *respond, bool *dupl_addr); void tipc_node_delete_links(struct net *net, int bearer_id); -void tipc_node_attach_link(struct tipc_node *n_ptr, struct tipc_link *l_ptr); -void tipc_node_detach_link(struct tipc_node *n_ptr, struct tipc_link *l_ptr); -bool tipc_node_is_up(struct tipc_node *n); int tipc_node_get_linkname(struct net *net, u32 bearer_id, u32 node, char *linkname, size_t len); -void tipc_node_unlock(struct tipc_node *node); int tipc_node_xmit(struct net *net, struct sk_buff_head *list, u32 dnode, int selector); int tipc_node_xmit_skb(struct net *net, struct sk_buff *skb, u32 dest, u32 selector); +void tipc_node_subscribe(struct net *net, struct list_head *subscr, u32 addr); +void tipc_node_unsubscribe(struct net *net, struct list_head *subscr, u32 addr); +void tipc_node_broadcast(struct net *net, struct sk_buff *skb); int tipc_node_add_conn(struct net *net, u32 dnode, u32 port, u32 peer_port); void tipc_node_remove_conn(struct net *net, u32 dnode, u32 port); +int tipc_node_get_mtu(struct net *net, u32 addr, u32 sel); int tipc_nl_node_dump(struct sk_buff *skb, struct netlink_callback *cb); - -static inline void tipc_node_lock(struct tipc_node *node) -{ - spin_lock_bh(&node->lock); -} - -static inline struct tipc_link *node_active_link(struct tipc_node *n, int sel) -{ - int bearer_id = n->active_links[sel & 1]; - - if (unlikely(bearer_id == INVALID_BEARER_ID)) - return NULL; - - return n->links[bearer_id].link; -} - -static inline unsigned int tipc_node_get_mtu(struct net *net, u32 addr, u32 sel) -{ - struct tipc_node *n; - int bearer_id; - unsigned int mtu = MAX_MSG_SIZE; - - n = tipc_node_find(net, addr); - if (unlikely(!n)) - return mtu; - - bearer_id = n->active_links[sel & 1]; - if (likely(bearer_id != INVALID_BEARER_ID)) - mtu = n->links[bearer_id].mtu; - tipc_node_put(n); - return mtu; -} +int tipc_nl_node_dump_link(struct sk_buff *skb, struct netlink_callback *cb); +int tipc_nl_node_reset_link_stats(struct sk_buff *skb, struct genl_info *info); +int tipc_nl_node_get_link(struct sk_buff *skb, struct genl_info *info); +int tipc_nl_node_set_link(struct sk_buff *skb, struct genl_info *info); #endif diff --git a/net/tipc/socket.c b/net/tipc/socket.c index b53246f..69c2905 100644 --- a/net/tipc/socket.c +++ b/net/tipc/socket.c @@ -1491,7 +1491,7 @@ static void tipc_write_space(struct sock *sk) rcu_read_lock(); wq = rcu_dereference(sk->sk_wq); - if (wq_has_sleeper(wq)) + if (skwq_has_sleeper(wq)) wake_up_interruptible_sync_poll(&wq->wait, POLLOUT | POLLWRNORM | POLLWRBAND); rcu_read_unlock(); @@ -1508,7 +1508,7 @@ static void tipc_data_ready(struct sock *sk) rcu_read_lock(); wq = rcu_dereference(sk->sk_wq); - if (wq_has_sleeper(wq)) + if (skwq_has_sleeper(wq)) wake_up_interruptible_sync_poll(&wq->wait, POLLIN | POLLRDNORM | POLLRDBAND); rcu_read_unlock(); diff --git a/net/tipc/udp_media.c b/net/tipc/udp_media.c index 70c0327..6af78c6 100644 --- a/net/tipc/udp_media.c +++ b/net/tipc/udp_media.c @@ -48,7 +48,6 @@ #include <linux/tipc_netlink.h> #include "core.h" #include "bearer.h" -#include "msg.h" /* IANA assigned UDP port */ #define UDP_PORT_DEFAULT 6118 @@ -224,10 +223,6 @@ static int tipc_udp_recv(struct sock *sk, struct sk_buff *skb) { struct udp_bearer *ub; struct tipc_bearer *b; - int usr = msg_user(buf_msg(skb)); - - if ((usr == LINK_PROTOCOL) || (usr == NAME_DISTRIBUTOR)) - skb_linearize(skb); ub = rcu_dereference_sk_user_data(sk); if (!ub) { diff --git a/net/unix/af_unix.c b/net/unix/af_unix.c index 45aebd9..502e572 100644 --- a/net/unix/af_unix.c +++ b/net/unix/af_unix.c @@ -451,7 +451,7 @@ static void unix_write_space(struct sock *sk) rcu_read_lock(); if (unix_writable(sk)) { wq = rcu_dereference(sk->sk_wq); - if (wq_has_sleeper(wq)) + if (skwq_has_sleeper(wq)) wake_up_interruptible_sync_poll(&wq->wait, POLLOUT | POLLWRNORM | POLLWRBAND); sk_wake_async(sk, SOCK_WAKE_SPACE, POLL_OUT); @@ -2109,8 +2109,10 @@ static int unix_dgram_recvmsg(struct socket *sock, struct msghdr *msg, goto out_unlock; } - wake_up_interruptible_sync_poll(&u->peer_wait, - POLLOUT | POLLWRNORM | POLLWRBAND); + if (wq_has_sleeper(&u->peer_wait)) + wake_up_interruptible_sync_poll(&u->peer_wait, + POLLOUT | POLLWRNORM | + POLLWRBAND); if (msg->msg_name) unix_copy_addr(msg, skb->sk); diff --git a/net/vmw_vsock/Kconfig b/net/vmw_vsock/Kconfig index 14810ab..74e0bc8 100644 --- a/net/vmw_vsock/Kconfig +++ b/net/vmw_vsock/Kconfig @@ -26,3 +26,21 @@ config VMWARE_VMCI_VSOCKETS To compile this driver as a module, choose M here: the module will be called vmw_vsock_vmci_transport. If unsure, say N. + +config VIRTIO_VSOCKETS + tristate "virtio transport for Virtual Sockets" + depends on VSOCKETS && VIRTIO + select VIRTIO_VSOCKETS_COMMON + help + This module implements a virtio transport for Virtual Sockets. + + Enable this transport if your Virtual Machine runs on Qemu/KVM. + + To compile this driver as a module, choose M here: the module + will be called virtio_vsock_transport. If unsure, say N. + +config VIRTIO_VSOCKETS_COMMON + tristate + ---help--- + This option is selected by any driver which needs to access + the virtio_vsock. diff --git a/net/vmw_vsock/Makefile b/net/vmw_vsock/Makefile index 2ce52d7..cf4c294 100644 --- a/net/vmw_vsock/Makefile +++ b/net/vmw_vsock/Makefile @@ -1,5 +1,7 @@ obj-$(CONFIG_VSOCKETS) += vsock.o obj-$(CONFIG_VMWARE_VMCI_VSOCKETS) += vmw_vsock_vmci_transport.o +obj-$(CONFIG_VIRTIO_VSOCKETS) += virtio_transport.o +obj-$(CONFIG_VIRTIO_VSOCKETS_COMMON) += virtio_transport_common.o vsock-y += af_vsock.o vsock_addr.o diff --git a/net/vmw_vsock/af_vsock.c b/net/vmw_vsock/af_vsock.c index 7fd1220..77247a2 100644 --- a/net/vmw_vsock/af_vsock.c +++ b/net/vmw_vsock/af_vsock.c @@ -223,6 +223,17 @@ static struct sock *__vsock_find_bound_socket(struct sockaddr_vm *addr) return NULL; } +static struct sock *__vsock_find_unbound_socket(struct sockaddr_vm *addr) +{ + struct vsock_sock *vsk; + + list_for_each_entry(vsk, vsock_unbound_sockets, bound_table) + if (addr->svm_port == vsk->local_addr.svm_port) + return sk_vsock(vsk); + + return NULL; +} + static struct sock *__vsock_find_connected_socket(struct sockaddr_vm *src, struct sockaddr_vm *dst) { @@ -298,6 +309,21 @@ struct sock *vsock_find_bound_socket(struct sockaddr_vm *addr) } EXPORT_SYMBOL_GPL(vsock_find_bound_socket); +struct sock *vsock_find_unbound_socket(struct sockaddr_vm *addr) +{ + struct sock *sk; + + spin_lock_bh(&vsock_table_lock); + sk = __vsock_find_unbound_socket(addr); + if (sk) + sock_hold(sk); + + spin_unlock_bh(&vsock_table_lock); + + return sk; +} +EXPORT_SYMBOL_GPL(vsock_find_unbound_socket); + struct sock *vsock_find_connected_socket(struct sockaddr_vm *src, struct sockaddr_vm *dst) { @@ -532,6 +558,50 @@ static int __vsock_bind_stream(struct vsock_sock *vsk, return 0; } +int vsock_bind_dgram_generic(struct vsock_sock *vsk, struct sockaddr_vm *addr) +{ + static u32 port = LAST_RESERVED_PORT + 1; + struct sockaddr_vm new_addr; + + vsock_addr_init(&new_addr, addr->svm_cid, addr->svm_port); + + if (addr->svm_port == VMADDR_PORT_ANY) { + bool found = false; + unsigned int i; + + for (i = 0; i < MAX_PORT_RETRIES; i++) { + if (port <= LAST_RESERVED_PORT) + port = LAST_RESERVED_PORT + 1; + + new_addr.svm_port = port++; + + if (!__vsock_find_unbound_socket(&new_addr)) { + found = true; + break; + } + } + + if (!found) + return -EADDRNOTAVAIL; + } else { + /* If port is in reserved range, ensure caller + * has necessary privileges. + */ + if (addr->svm_port <= LAST_RESERVED_PORT && + !capable(CAP_NET_BIND_SERVICE)) { + return -EACCES; + } + + if (__vsock_find_unbound_socket(&new_addr)) + return -EADDRINUSE; + } + + vsock_addr_init(&vsk->local_addr, new_addr.svm_cid, new_addr.svm_port); + + return 0; +} +EXPORT_SYMBOL_GPL(vsock_bind_dgram_generic); + static int __vsock_bind_dgram(struct vsock_sock *vsk, struct sockaddr_vm *addr) { diff --git a/net/vmw_vsock/virtio_transport.c b/net/vmw_vsock/virtio_transport.c new file mode 100644 index 0000000..df65dca --- /dev/null +++ b/net/vmw_vsock/virtio_transport.c @@ -0,0 +1,466 @@ +/* + * virtio transport for vsock + * + * Copyright (C) 2013-2015 Red Hat, Inc. + * Author: Asias He <asias@redhat.com> + * Stefan Hajnoczi <stefanha@redhat.com> + * + * Some of the code is take from Gerd Hoffmann <kraxel@redhat.com>'s + * early virtio-vsock proof-of-concept bits. + * + * This work is licensed under the terms of the GNU GPL, version 2. + */ +#include <linux/spinlock.h> +#include <linux/module.h> +#include <linux/list.h> +#include <linux/virtio.h> +#include <linux/virtio_ids.h> +#include <linux/virtio_config.h> +#include <linux/virtio_vsock.h> +#include <net/sock.h> +#include <linux/mutex.h> +#include <net/af_vsock.h> + +static struct workqueue_struct *virtio_vsock_workqueue; +static struct virtio_vsock *the_virtio_vsock; +static DEFINE_MUTEX(the_virtio_vsock_mutex); /* protects the_virtio_vsock */ +static void virtio_vsock_rx_fill(struct virtio_vsock *vsock); + +struct virtio_vsock { + /* Virtio device */ + struct virtio_device *vdev; + /* Virtio virtqueue */ + struct virtqueue *vqs[VSOCK_VQ_MAX]; + /* Wait queue for send pkt */ + wait_queue_head_t queue_wait; + /* Work item to send pkt */ + struct work_struct tx_work; + /* Work item to recv pkt */ + struct work_struct rx_work; + /* Mutex to protect send pkt*/ + struct mutex tx_lock; + /* Mutex to protect recv pkt*/ + struct mutex rx_lock; + /* Number of recv buffers */ + int rx_buf_nr; + /* Number of max recv buffers */ + int rx_buf_max_nr; + /* Used for global tx buf limitation */ + u32 total_tx_buf; + /* Guest context id, just like guest ip address */ + u32 guest_cid; +}; + +static struct virtio_vsock *virtio_vsock_get(void) +{ + return the_virtio_vsock; +} + +static u32 virtio_transport_get_local_cid(void) +{ + struct virtio_vsock *vsock = virtio_vsock_get(); + + return vsock->guest_cid; +} + +static int +virtio_transport_send_pkt(struct vsock_sock *vsk, + struct virtio_vsock_pkt_info *info) +{ + u32 src_cid, src_port, dst_cid, dst_port; + int ret, in_sg = 0, out_sg = 0; + struct virtio_transport *trans; + struct virtio_vsock_pkt *pkt; + struct virtio_vsock *vsock; + struct scatterlist hdr, buf, *sgs[2]; + struct virtqueue *vq; + u32 pkt_len = info->pkt_len; + DEFINE_WAIT(wait); + + vsock = virtio_vsock_get(); + if (!vsock) + return -ENODEV; + + src_cid = virtio_transport_get_local_cid(); + src_port = vsk->local_addr.svm_port; + if (!info->remote_cid) { + dst_cid = vsk->remote_addr.svm_cid; + dst_port = vsk->remote_addr.svm_port; + } else { + dst_cid = info->remote_cid; + dst_port = info->remote_port; + } + + trans = vsk->trans; + vq = vsock->vqs[VSOCK_VQ_TX]; + + if (pkt_len > VIRTIO_VSOCK_DEFAULT_RX_BUF_SIZE) + pkt_len = VIRTIO_VSOCK_DEFAULT_RX_BUF_SIZE; + pkt_len = virtio_transport_get_credit(trans, pkt_len); + /* Do not send zero length OP_RW pkt*/ + if (pkt_len == 0 && info->op == VIRTIO_VSOCK_OP_RW) + return pkt_len; + + /* Respect global tx buf limitation */ + mutex_lock(&vsock->tx_lock); + while (pkt_len + vsock->total_tx_buf > VIRTIO_VSOCK_MAX_TX_BUF_SIZE) { + prepare_to_wait_exclusive(&vsock->queue_wait, &wait, + TASK_UNINTERRUPTIBLE); + mutex_unlock(&vsock->tx_lock); + schedule(); + mutex_lock(&vsock->tx_lock); + finish_wait(&vsock->queue_wait, &wait); + } + vsock->total_tx_buf += pkt_len; + mutex_unlock(&vsock->tx_lock); + + pkt = virtio_transport_alloc_pkt(vsk, info, pkt_len, + src_cid, src_port, + dst_cid, dst_port); + if (!pkt) { + mutex_lock(&vsock->tx_lock); + vsock->total_tx_buf -= pkt_len; + mutex_unlock(&vsock->tx_lock); + virtio_transport_put_credit(trans, pkt_len); + return -ENOMEM; + } + + pr_debug("%s:info->pkt_len= %d\n", __func__, info->pkt_len); + + /* Will be released in virtio_transport_send_pkt_work */ + sock_hold(&trans->vsk->sk); + virtio_transport_inc_tx_pkt(pkt); + + /* Put pkt in the virtqueue */ + sg_init_one(&hdr, &pkt->hdr, sizeof(pkt->hdr)); + sgs[out_sg++] = &hdr; + if (info->msg && info->pkt_len > 0) { + sg_init_one(&buf, pkt->buf, pkt->len); + sgs[out_sg++] = &buf; + } + + mutex_lock(&vsock->tx_lock); + while ((ret = virtqueue_add_sgs(vq, sgs, out_sg, in_sg, pkt, + GFP_KERNEL)) < 0) { + prepare_to_wait_exclusive(&vsock->queue_wait, &wait, + TASK_UNINTERRUPTIBLE); + mutex_unlock(&vsock->tx_lock); + schedule(); + mutex_lock(&vsock->tx_lock); + finish_wait(&vsock->queue_wait, &wait); + } + virtqueue_kick(vq); + mutex_unlock(&vsock->tx_lock); + + return pkt_len; +} + +static struct virtio_transport_pkt_ops virtio_ops = { + .send_pkt = virtio_transport_send_pkt, +}; + +static void virtio_vsock_rx_fill(struct virtio_vsock *vsock) +{ + int buf_len = VIRTIO_VSOCK_DEFAULT_RX_BUF_SIZE; + struct virtio_vsock_pkt *pkt; + struct scatterlist hdr, buf, *sgs[2]; + struct virtqueue *vq; + int ret; + + vq = vsock->vqs[VSOCK_VQ_RX]; + + do { + pkt = kzalloc(sizeof(*pkt), GFP_KERNEL); + if (!pkt) { + pr_debug("%s: fail to allocate pkt\n", __func__); + goto out; + } + + /* TODO: use mergeable rx buffer */ + pkt->buf = kmalloc(buf_len, GFP_KERNEL); + if (!pkt->buf) { + pr_debug("%s: fail to allocate pkt->buf\n", __func__); + goto err; + } + + sg_init_one(&hdr, &pkt->hdr, sizeof(pkt->hdr)); + sgs[0] = &hdr; + + sg_init_one(&buf, pkt->buf, buf_len); + sgs[1] = &buf; + ret = virtqueue_add_sgs(vq, sgs, 0, 2, pkt, GFP_KERNEL); + if (ret) + goto err; + vsock->rx_buf_nr++; + } while (vq->num_free); + if (vsock->rx_buf_nr > vsock->rx_buf_max_nr) + vsock->rx_buf_max_nr = vsock->rx_buf_nr; +out: + virtqueue_kick(vq); + return; +err: + virtqueue_kick(vq); + virtio_transport_free_pkt(pkt); + return; +} + +static void virtio_transport_send_pkt_work(struct work_struct *work) +{ + struct virtio_vsock *vsock = + container_of(work, struct virtio_vsock, tx_work); + struct virtio_vsock_pkt *pkt; + bool added = false; + struct virtqueue *vq; + unsigned int len; + struct sock *sk; + + vq = vsock->vqs[VSOCK_VQ_TX]; + mutex_lock(&vsock->tx_lock); + do { + virtqueue_disable_cb(vq); + while ((pkt = virtqueue_get_buf(vq, &len)) != NULL) { + sk = &pkt->trans->vsk->sk; + virtio_transport_dec_tx_pkt(pkt); + /* Release refcnt taken in virtio_transport_send_pkt */ + sock_put(sk); + vsock->total_tx_buf -= pkt->len; + virtio_transport_free_pkt(pkt); + added = true; + } + } while (!virtqueue_enable_cb(vq)); + mutex_unlock(&vsock->tx_lock); + + if (added) + wake_up(&vsock->queue_wait); +} + +static void virtio_transport_recv_pkt_work(struct work_struct *work) +{ + struct virtio_vsock *vsock = + container_of(work, struct virtio_vsock, rx_work); + struct virtio_vsock_pkt *pkt; + struct virtqueue *vq; + unsigned int len; + + vq = vsock->vqs[VSOCK_VQ_RX]; + mutex_lock(&vsock->rx_lock); + do { + virtqueue_disable_cb(vq); + while ((pkt = virtqueue_get_buf(vq, &len)) != NULL) { + pkt->len = len; + virtio_transport_recv_pkt(pkt); + vsock->rx_buf_nr--; + } + } while (!virtqueue_enable_cb(vq)); + + if (vsock->rx_buf_nr < vsock->rx_buf_max_nr / 2) + virtio_vsock_rx_fill(vsock); + mutex_unlock(&vsock->rx_lock); +} + +static void virtio_vsock_ctrl_done(struct virtqueue *vq) +{ +} + +static void virtio_vsock_tx_done(struct virtqueue *vq) +{ + struct virtio_vsock *vsock = vq->vdev->priv; + + if (!vsock) + return; + queue_work(virtio_vsock_workqueue, &vsock->tx_work); +} + +static void virtio_vsock_rx_done(struct virtqueue *vq) +{ + struct virtio_vsock *vsock = vq->vdev->priv; + + if (!vsock) + return; + queue_work(virtio_vsock_workqueue, &vsock->rx_work); +} + +static int +virtio_transport_socket_init(struct vsock_sock *vsk, struct vsock_sock *psk) +{ + struct virtio_transport *trans; + int ret; + + ret = virtio_transport_do_socket_init(vsk, psk); + if (ret) + return ret; + + trans = vsk->trans; + trans->ops = &virtio_ops; + return ret; +} + +static struct vsock_transport virtio_transport = { + .get_local_cid = virtio_transport_get_local_cid, + + .init = virtio_transport_socket_init, + .destruct = virtio_transport_destruct, + .release = virtio_transport_release, + .connect = virtio_transport_connect, + .shutdown = virtio_transport_shutdown, + + .dgram_bind = virtio_transport_dgram_bind, + .dgram_dequeue = virtio_transport_dgram_dequeue, + .dgram_enqueue = virtio_transport_dgram_enqueue, + .dgram_allow = virtio_transport_dgram_allow, + + .stream_dequeue = virtio_transport_stream_dequeue, + .stream_enqueue = virtio_transport_stream_enqueue, + .stream_has_data = virtio_transport_stream_has_data, + .stream_has_space = virtio_transport_stream_has_space, + .stream_rcvhiwat = virtio_transport_stream_rcvhiwat, + .stream_is_active = virtio_transport_stream_is_active, + .stream_allow = virtio_transport_stream_allow, + + .notify_poll_in = virtio_transport_notify_poll_in, + .notify_poll_out = virtio_transport_notify_poll_out, + .notify_recv_init = virtio_transport_notify_recv_init, + .notify_recv_pre_block = virtio_transport_notify_recv_pre_block, + .notify_recv_pre_dequeue = virtio_transport_notify_recv_pre_dequeue, + .notify_recv_post_dequeue = virtio_transport_notify_recv_post_dequeue, + .notify_send_init = virtio_transport_notify_send_init, + .notify_send_pre_block = virtio_transport_notify_send_pre_block, + .notify_send_pre_enqueue = virtio_transport_notify_send_pre_enqueue, + .notify_send_post_enqueue = virtio_transport_notify_send_post_enqueue, + + .set_buffer_size = virtio_transport_set_buffer_size, + .set_min_buffer_size = virtio_transport_set_min_buffer_size, + .set_max_buffer_size = virtio_transport_set_max_buffer_size, + .get_buffer_size = virtio_transport_get_buffer_size, + .get_min_buffer_size = virtio_transport_get_min_buffer_size, + .get_max_buffer_size = virtio_transport_get_max_buffer_size, +}; + +static int virtio_vsock_probe(struct virtio_device *vdev) +{ + vq_callback_t *callbacks[] = { + virtio_vsock_ctrl_done, + virtio_vsock_rx_done, + virtio_vsock_tx_done, + }; + const char *names[] = { + "ctrl", + "rx", + "tx", + }; + struct virtio_vsock *vsock = NULL; + u32 guest_cid; + int ret; + + ret = mutex_lock_interruptible(&the_virtio_vsock_mutex); + if (ret) + return ret; + + /* Only one virtio-vsock device per guest is supported */ + if (the_virtio_vsock) { + ret = -EBUSY; + goto out; + } + + vsock = kzalloc(sizeof(*vsock), GFP_KERNEL); + if (!vsock) { + ret = -ENOMEM; + goto out; + } + + vsock->vdev = vdev; + + ret = vsock->vdev->config->find_vqs(vsock->vdev, VSOCK_VQ_MAX, + vsock->vqs, callbacks, names); + if (ret < 0) + goto out; + + vdev->config->get(vdev, offsetof(struct virtio_vsock_config, guest_cid), + &guest_cid, sizeof(guest_cid)); + vsock->guest_cid = le32_to_cpu(guest_cid); + pr_debug("%s:guest_cid=%d\n", __func__, vsock->guest_cid); + + ret = vsock_core_init(&virtio_transport); + if (ret < 0) + goto out_vqs; + + vsock->rx_buf_nr = 0; + vsock->rx_buf_max_nr = 0; + + vdev->priv = the_virtio_vsock = vsock; + init_waitqueue_head(&vsock->queue_wait); + mutex_init(&vsock->tx_lock); + mutex_init(&vsock->rx_lock); + INIT_WORK(&vsock->rx_work, virtio_transport_recv_pkt_work); + INIT_WORK(&vsock->tx_work, virtio_transport_send_pkt_work); + + mutex_lock(&vsock->rx_lock); + virtio_vsock_rx_fill(vsock); + mutex_unlock(&vsock->rx_lock); + + mutex_unlock(&the_virtio_vsock_mutex); + return 0; + +out_vqs: + vsock->vdev->config->del_vqs(vsock->vdev); +out: + kfree(vsock); + mutex_unlock(&the_virtio_vsock_mutex); + return ret; +} + +static void virtio_vsock_remove(struct virtio_device *vdev) +{ + struct virtio_vsock *vsock = vdev->priv; + + mutex_lock(&the_virtio_vsock_mutex); + the_virtio_vsock = NULL; + vsock_core_exit(); + mutex_unlock(&the_virtio_vsock_mutex); + + kfree(vsock); +} + +static struct virtio_device_id id_table[] = { + { VIRTIO_ID_VSOCK, VIRTIO_DEV_ANY_ID }, + { 0 }, +}; + +static unsigned int features[] = { +}; + +static struct virtio_driver virtio_vsock_driver = { + .feature_table = features, + .feature_table_size = ARRAY_SIZE(features), + .driver.name = KBUILD_MODNAME, + .driver.owner = THIS_MODULE, + .id_table = id_table, + .probe = virtio_vsock_probe, + .remove = virtio_vsock_remove, +}; + +static int __init virtio_vsock_init(void) +{ + int ret; + + virtio_vsock_workqueue = alloc_workqueue("virtio_vsock", 0, 0); + if (!virtio_vsock_workqueue) + return -ENOMEM; + ret = register_virtio_driver(&virtio_vsock_driver); + if (ret) + destroy_workqueue(virtio_vsock_workqueue); + return ret; +} + +static void __exit virtio_vsock_exit(void) +{ + unregister_virtio_driver(&virtio_vsock_driver); + destroy_workqueue(virtio_vsock_workqueue); +} + +module_init(virtio_vsock_init); +module_exit(virtio_vsock_exit); +MODULE_LICENSE("GPL v2"); +MODULE_AUTHOR("Asias He"); +MODULE_DESCRIPTION("virtio transport for vsock"); +MODULE_DEVICE_TABLE(virtio, id_table); diff --git a/net/vmw_vsock/virtio_transport_common.c b/net/vmw_vsock/virtio_transport_common.c new file mode 100644 index 0000000..28f790d --- /dev/null +++ b/net/vmw_vsock/virtio_transport_common.c @@ -0,0 +1,1272 @@ +/* + * common code for virtio vsock + * + * Copyright (C) 2013-2015 Red Hat, Inc. + * Author: Asias He <asias@redhat.com> + * Stefan Hajnoczi <stefanha@redhat.com> + * + * This work is licensed under the terms of the GNU GPL, version 2. + */ +#include <linux/module.h> +#include <linux/ctype.h> +#include <linux/list.h> +#include <linux/virtio.h> +#include <linux/virtio_ids.h> +#include <linux/virtio_config.h> +#include <linux/virtio_vsock.h> +#include <linux/random.h> +#include <linux/cryptohash.h> + +#include <net/sock.h> +#include <net/af_vsock.h> + +#define COOKIEBITS 24 +#define COOKIEMASK (((u32)1 << COOKIEBITS) - 1) +#define VSOCK_TIMEOUT_INIT 4 + +#define SHA_MESSAGE_WORDS 16 +#define SHA_VSOCK_WORDS 5 + +static u32 vsockcookie_secret[2][SHA_MESSAGE_WORDS - SHA_VSOCK_WORDS + + SHA_DIGEST_WORDS]; + +static DEFINE_PER_CPU(__u32[SHA_MESSAGE_WORDS + SHA_DIGEST_WORDS + + SHA_WORKSPACE_WORDS], vsock_cookie_scratch); + +static u32 cookie_hash(u32 saddr, u32 daddr, u16 sport, u16 dport, + u32 count, int c) +{ + __u32 *tmp = this_cpu_ptr(vsock_cookie_scratch); + + memcpy(tmp + SHA_VSOCK_WORDS, vsockcookie_secret[c], + sizeof(vsockcookie_secret[c])); + tmp[0] = saddr; + tmp[1] = daddr; + tmp[2] = sport; + tmp[3] = dport; + tmp[4] = count; + sha_transform(tmp + SHA_MESSAGE_WORDS, (__u8 *)tmp, + tmp + SHA_MESSAGE_WORDS + SHA_DIGEST_WORDS); + + return tmp[17]; +} + +static u32 +virtio_vsock_secure_cookie(u32 saddr, u32 daddr, u32 sport, u32 dport, + u32 count) +{ + u32 h1, h2; + + h1 = cookie_hash(saddr, daddr, sport, dport, 0, 0); + h2 = cookie_hash(saddr, daddr, sport, dport, count, 1); + + return h1 + (count << COOKIEBITS) + (h2 & COOKIEMASK); +} + +static u32 +virtio_vsock_check_cookie(u32 saddr, u32 daddr, u32 sport, u32 dport, + u32 count, u32 cookie, u32 maxdiff) +{ + u32 diff; + u32 ret; + + cookie -= cookie_hash(saddr, daddr, sport, dport, 0, 0); + + diff = (count - (cookie >> COOKIEBITS)) & ((u32)-1 >> COOKIEBITS); + pr_debug("%s: diff=%x\n", __func__, diff); + if (diff >= maxdiff) + return (u32)-1; + + ret = (cookie - + cookie_hash(saddr, daddr, sport, dport, count - diff, 1)) + & COOKIEMASK; + pr_debug("%s: ret=%x\n", __func__, diff); + + return ret; +} + +void virtio_vsock_dumppkt(const char *func, const struct virtio_vsock_pkt *pkt) +{ + pr_debug("%s: pkt=%p, op=%d, len=%d, %d:%d---%d:%d, len=%d\n", + func, pkt, + le16_to_cpu(pkt->hdr.op), + le32_to_cpu(pkt->hdr.len), + le32_to_cpu(pkt->hdr.src_cid), + le32_to_cpu(pkt->hdr.src_port), + le32_to_cpu(pkt->hdr.dst_cid), + le32_to_cpu(pkt->hdr.dst_port), + pkt->len); +} +EXPORT_SYMBOL_GPL(virtio_vsock_dumppkt); + +struct virtio_vsock_pkt * +virtio_transport_alloc_pkt(struct vsock_sock *vsk, + struct virtio_vsock_pkt_info *info, + size_t len, + u32 src_cid, + u32 src_port, + u32 dst_cid, + u32 dst_port) +{ + struct virtio_transport *trans = vsk->trans; + struct virtio_vsock_pkt *pkt; + int err; + + BUG_ON(!trans); + + pkt = kzalloc(sizeof(*pkt), GFP_KERNEL); + if (!pkt) + return NULL; + + pkt->hdr.type = cpu_to_le16(info->type); + pkt->hdr.op = cpu_to_le16(info->op); + pkt->hdr.src_cid = cpu_to_le32(src_cid); + pkt->hdr.src_port = cpu_to_le32(src_port); + pkt->hdr.dst_cid = cpu_to_le32(dst_cid); + pkt->hdr.dst_port = cpu_to_le32(dst_port); + pkt->hdr.flags = cpu_to_le32(info->flags); + pkt->len = len; + pkt->trans = trans; + if (info->type == VIRTIO_VSOCK_TYPE_DGRAM) + pkt->hdr.len = cpu_to_le32(len + (info->dgram_len << 16)); + else if (info->type == VIRTIO_VSOCK_TYPE_STREAM) + pkt->hdr.len = cpu_to_le32(len); + + if (info->msg && len > 0) { + pkt->buf = kmalloc(len, GFP_KERNEL); + if (!pkt->buf) + goto out_pkt; + err = memcpy_from_msg(pkt->buf, info->msg, len); + if (err) + goto out; + } + + return pkt; + +out: + kfree(pkt->buf); +out_pkt: + kfree(pkt); + return NULL; +} +EXPORT_SYMBOL_GPL(virtio_transport_alloc_pkt); + +struct sock * +virtio_transport_get_pending(struct sock *listener, + struct virtio_vsock_pkt *pkt) +{ + struct vsock_sock *vlistener; + struct vsock_sock *vpending; + struct sockaddr_vm src; + struct sockaddr_vm dst; + struct sock *pending; + + vsock_addr_init(&src, le32_to_cpu(pkt->hdr.src_cid), le32_to_cpu(pkt->hdr.src_port)); + vsock_addr_init(&dst, le32_to_cpu(pkt->hdr.dst_cid), le32_to_cpu(pkt->hdr.dst_port)); + + vlistener = vsock_sk(listener); + list_for_each_entry(vpending, &vlistener->pending_links, + pending_links) { + if (vsock_addr_equals_addr(&src, &vpending->remote_addr) && + vsock_addr_equals_addr(&dst, &vpending->local_addr)) { + pending = sk_vsock(vpending); + sock_hold(pending); + return pending; + } + } + + return NULL; +} +EXPORT_SYMBOL_GPL(virtio_transport_get_pending); + +static void virtio_transport_inc_rx_pkt(struct virtio_vsock_pkt *pkt) +{ + pkt->trans->rx_bytes += pkt->len; +} + +static void virtio_transport_dec_rx_pkt(struct virtio_vsock_pkt *pkt) +{ + pkt->trans->rx_bytes -= pkt->len; + pkt->trans->fwd_cnt += pkt->len; +} + +void virtio_transport_inc_tx_pkt(struct virtio_vsock_pkt *pkt) +{ + mutex_lock(&pkt->trans->tx_lock); + pkt->hdr.fwd_cnt = cpu_to_le32(pkt->trans->fwd_cnt); + pkt->hdr.buf_alloc = cpu_to_le32(pkt->trans->buf_alloc); + mutex_unlock(&pkt->trans->tx_lock); +} +EXPORT_SYMBOL_GPL(virtio_transport_inc_tx_pkt); + +void virtio_transport_dec_tx_pkt(struct virtio_vsock_pkt *pkt) +{ +} +EXPORT_SYMBOL_GPL(virtio_transport_dec_tx_pkt); + +u32 virtio_transport_get_credit(struct virtio_transport *trans, u32 credit) +{ + u32 ret; + + mutex_lock(&trans->tx_lock); + ret = trans->peer_buf_alloc - (trans->tx_cnt - trans->peer_fwd_cnt); + if (ret > credit) + ret = credit; + trans->tx_cnt += ret; + mutex_unlock(&trans->tx_lock); + + pr_debug("%s: ret=%d, buf_alloc=%d, peer_buf_alloc=%d," + "tx_cnt=%d, fwd_cnt=%d, peer_fwd_cnt=%d\n", __func__, + ret, trans->buf_alloc, trans->peer_buf_alloc, + trans->tx_cnt, trans->fwd_cnt, trans->peer_fwd_cnt); + + return ret; +} +EXPORT_SYMBOL_GPL(virtio_transport_get_credit); + +void virtio_transport_put_credit(struct virtio_transport *trans, u32 credit) +{ + mutex_lock(&trans->tx_lock); + trans->tx_cnt -= credit; + mutex_unlock(&trans->tx_lock); +} +EXPORT_SYMBOL_GPL(virtio_transport_put_credit); + +static int virtio_transport_send_credit_update(struct vsock_sock *vsk, int type, struct virtio_vsock_hdr *hdr) +{ + struct virtio_transport *trans = vsk->trans; + struct virtio_vsock_pkt_info info = { + .op = VIRTIO_VSOCK_OP_CREDIT_UPDATE, + .type = type, + }; + + if (hdr && type == VIRTIO_VSOCK_TYPE_DGRAM) { + info.remote_cid = le32_to_cpu(hdr->src_cid); + info.remote_port = le32_to_cpu(hdr->src_port); + } + + pr_debug("%s: sk=%p send_credit_update\n", __func__, vsk); + return trans->ops->send_pkt(vsk, &info); +} + +static int virtio_transport_send_credit_request(struct vsock_sock *vsk, int type) +{ + struct virtio_transport *trans = vsk->trans; + struct virtio_vsock_pkt_info info = { + .op = VIRTIO_VSOCK_OP_CREDIT_REQUEST, + .type = type, + }; + + pr_debug("%s: sk=%p send_credit_request\n", __func__, vsk); + return trans->ops->send_pkt(vsk, &info); +} + +static ssize_t +virtio_transport_stream_do_dequeue(struct vsock_sock *vsk, + struct msghdr *msg, + size_t len) +{ + struct virtio_transport *trans = vsk->trans; + struct virtio_vsock_pkt *pkt; + size_t bytes, total = 0; + int err = -EFAULT; + + mutex_lock(&trans->rx_lock); + while (total < len && trans->rx_bytes > 0 && + !list_empty(&trans->rx_queue)) { + pkt = list_first_entry(&trans->rx_queue, + struct virtio_vsock_pkt, list); + + bytes = len - total; + if (bytes > pkt->len - pkt->off) + bytes = pkt->len - pkt->off; + + err = memcpy_to_msg(msg, pkt->buf + pkt->off, bytes); + if (err) + goto out; + total += bytes; + pkt->off += bytes; + if (pkt->off == pkt->len) { + virtio_transport_dec_rx_pkt(pkt); + list_del(&pkt->list); + virtio_transport_free_pkt(pkt); + } + } + mutex_unlock(&trans->rx_lock); + + /* Send a credit pkt to peer */ + virtio_transport_send_credit_update(vsk, VIRTIO_VSOCK_TYPE_STREAM, + NULL); + + return total; + +out: + mutex_unlock(&trans->rx_lock); + if (total) + err = total; + return err; +} + +ssize_t +virtio_transport_stream_dequeue(struct vsock_sock *vsk, + struct msghdr *msg, + size_t len, int flags) +{ + if (flags & MSG_PEEK) + return -EOPNOTSUPP; + + return virtio_transport_stream_do_dequeue(vsk, msg, len); +} +EXPORT_SYMBOL_GPL(virtio_transport_stream_dequeue); + +struct dgram_skb { + struct list_head list; + struct sk_buff *skb; + u16 id; +}; + +static struct dgram_skb *dgram_id_to_skb(struct virtio_transport *trans, + u16 id) +{ + struct dgram_skb *dgram_skb; + + list_for_each_entry(dgram_skb, &trans->incomplete_dgrams, list) { + if (dgram_skb->id == id) + return dgram_skb; + } + + return NULL; +} + +static void +virtio_transport_recv_dgram(struct sock *sk, + struct virtio_vsock_pkt *pkt) +{ + struct sk_buff *skb = NULL; + struct vsock_sock *vsk; + struct virtio_transport *trans; + size_t size; + u16 dgram_id, pkt_off, dgram_len, pkt_len; + u32 flags, len; + struct dgram_skb *dgram_skb; + + vsk = vsock_sk(sk); + trans = vsk->trans; + + /* len: dgram_len | pkt_len */ + len = le32_to_cpu(pkt->hdr.len); + dgram_len = len >> 16; + pkt_len = len & 0xFFFF; + + /* flags: dgram_id | pkt_off */ + flags = le32_to_cpu(pkt->hdr.flags); + dgram_id = flags >> 16; + pkt_off = flags & 0xFFFF; + + pr_debug("%s: dgram_len=%d, pkt_len=%d, id=%d, off=%d\n", __func__, + dgram_len, pkt_len, dgram_id, pkt_off); + + dgram_skb = dgram_id_to_skb(trans, dgram_id); + if (dgram_skb) { + /* This pkt is for a existing dgram */ + skb = dgram_skb->skb; + pr_debug("%s:found skb\n", __func__); + } + + /* Packet payload must be within datagram bounds */ + if (pkt_len > VIRTIO_VSOCK_DEFAULT_RX_BUF_SIZE) + goto drop; + if (pkt_len > dgram_len) + goto drop; + if (pkt_off > dgram_len) + goto drop; + if (dgram_len - pkt_off < pkt_len) + goto drop; + + if (!skb) { + /* This pkt is for a new dgram */ + pr_debug("%s:create skb\n", __func__); + + size = sizeof(pkt->hdr) + dgram_len; + /* Attach the packet to the socket's receive queue as an sk_buff. */ + dgram_skb = kzalloc(sizeof(struct dgram_skb), GFP_ATOMIC); + if (!dgram_skb) + goto drop; + + skb = alloc_skb(size, GFP_ATOMIC); + if (!skb) { + kfree(dgram_skb); + dgram_skb = NULL; + goto drop; + } + dgram_skb->id = dgram_id; + dgram_skb->skb = skb; + list_add_tail(&dgram_skb->list, &trans->incomplete_dgrams); + + /* sk_receive_skb() will do a sock_put(), so hold here. */ + sock_hold(sk); + skb_put(skb, size); + memcpy(skb->data, &pkt->hdr, sizeof(pkt->hdr)); + } + + memcpy(skb->data + sizeof(pkt->hdr) + pkt_off, pkt->buf, pkt_len); + + pr_debug("%s:C, off=%d, pkt_len=%d, dgram_len=%d\n", __func__, + pkt_off, pkt_len, dgram_len); + + /* We are done with this dgram */ + if (pkt_off + pkt_len == dgram_len) { + pr_debug("%s:dgram_id=%d is done\n", __func__, dgram_id); + list_del(&dgram_skb->list); + kfree(dgram_skb); + sk_receive_skb(sk, skb, 0); + } + virtio_transport_free_pkt(pkt); + return; + +drop: + if (dgram_skb) { + list_del(&dgram_skb->list); + kfree(dgram_skb); + kfree_skb(skb); + sock_put(sk); + } + virtio_transport_free_pkt(pkt); +} + +int +virtio_transport_dgram_dequeue(struct vsock_sock *vsk, + struct msghdr *msg, + size_t len, int flags) +{ + struct virtio_vsock_hdr *hdr; + struct sk_buff *skb; + int noblock; + int err; + int dgram_len; + + noblock = flags & MSG_DONTWAIT; + + if (flags & MSG_OOB || flags & MSG_ERRQUEUE) + return -EOPNOTSUPP; + + /* Retrieve the head sk_buff from the socket's receive queue. */ + err = 0; + skb = skb_recv_datagram(&vsk->sk, flags, noblock, &err); + if (err) + return err; + if (!skb) + return -EAGAIN; + + hdr = (struct virtio_vsock_hdr *)skb->data; + if (!hdr) + goto out; + + dgram_len = le32_to_cpu(hdr->len) >> 16; + /* Place the datagram payload in the user's iovec. */ + err = skb_copy_datagram_msg(skb, sizeof(*hdr), msg, dgram_len); + if (err) + goto out; + + if (msg->msg_name) { + /* Provide the address of the sender. */ + DECLARE_SOCKADDR(struct sockaddr_vm *, vm_addr, msg->msg_name); + vsock_addr_init(vm_addr, le32_to_cpu(hdr->src_cid), le32_to_cpu(hdr->src_port)); + msg->msg_namelen = sizeof(*vm_addr); + } + err = dgram_len; + + /* Send a credit pkt to peer */ + virtio_transport_send_credit_update(vsk, VIRTIO_VSOCK_TYPE_DGRAM, hdr); + + pr_debug("%s:done, recved =%d\n", __func__, dgram_len); +out: + skb_free_datagram(&vsk->sk, skb); + return err; +} +EXPORT_SYMBOL_GPL(virtio_transport_dgram_dequeue); + +s64 virtio_transport_stream_has_data(struct vsock_sock *vsk) +{ + struct virtio_transport *trans = vsk->trans; + s64 bytes; + + mutex_lock(&trans->rx_lock); + bytes = trans->rx_bytes; + mutex_unlock(&trans->rx_lock); + + return bytes; +} +EXPORT_SYMBOL_GPL(virtio_transport_stream_has_data); + +static s64 virtio_transport_has_space(struct vsock_sock *vsk) +{ + struct virtio_transport *trans = vsk->trans; + s64 bytes; + + bytes = trans->peer_buf_alloc - (trans->tx_cnt - trans->peer_fwd_cnt); + if (bytes < 0) + bytes = 0; + + return bytes; +} + +s64 virtio_transport_stream_has_space(struct vsock_sock *vsk) +{ + struct virtio_transport *trans = vsk->trans; + s64 bytes; + + mutex_lock(&trans->tx_lock); + bytes = virtio_transport_has_space(vsk); + mutex_unlock(&trans->tx_lock); + + pr_debug("%s: bytes=%lld\n", __func__, bytes); + + return bytes; +} +EXPORT_SYMBOL_GPL(virtio_transport_stream_has_space); + +int virtio_transport_do_socket_init(struct vsock_sock *vsk, + struct vsock_sock *psk) +{ + struct virtio_transport *trans; + + trans = kzalloc(sizeof(*trans), GFP_KERNEL); + if (!trans) + return -ENOMEM; + + vsk->trans = trans; + trans->vsk = vsk; + if (psk) { + struct virtio_transport *ptrans = psk->trans; + trans->buf_size = ptrans->buf_size; + trans->buf_size_min = ptrans->buf_size_min; + trans->buf_size_max = ptrans->buf_size_max; + trans->peer_buf_alloc = ptrans->peer_buf_alloc; + } else { + trans->buf_size = VIRTIO_VSOCK_DEFAULT_BUF_SIZE; + trans->buf_size_min = VIRTIO_VSOCK_DEFAULT_MIN_BUF_SIZE; + trans->buf_size_max = VIRTIO_VSOCK_DEFAULT_MAX_BUF_SIZE; + } + + trans->buf_alloc = trans->buf_size; + + pr_debug("%s: trans->buf_alloc=%d\n", __func__, trans->buf_alloc); + + mutex_init(&trans->rx_lock); + mutex_init(&trans->tx_lock); + INIT_LIST_HEAD(&trans->rx_queue); + INIT_LIST_HEAD(&trans->incomplete_dgrams); + + return 0; +} +EXPORT_SYMBOL_GPL(virtio_transport_do_socket_init); + +u64 virtio_transport_get_buffer_size(struct vsock_sock *vsk) +{ + struct virtio_transport *trans = vsk->trans; + + return trans->buf_size; +} +EXPORT_SYMBOL_GPL(virtio_transport_get_buffer_size); + +u64 virtio_transport_get_min_buffer_size(struct vsock_sock *vsk) +{ + struct virtio_transport *trans = vsk->trans; + + return trans->buf_size_min; +} +EXPORT_SYMBOL_GPL(virtio_transport_get_min_buffer_size); + +u64 virtio_transport_get_max_buffer_size(struct vsock_sock *vsk) +{ + struct virtio_transport *trans = vsk->trans; + + return trans->buf_size_max; +} +EXPORT_SYMBOL_GPL(virtio_transport_get_max_buffer_size); + +void virtio_transport_set_buffer_size(struct vsock_sock *vsk, u64 val) +{ + struct virtio_transport *trans = vsk->trans; + + if (val > VIRTIO_VSOCK_MAX_BUF_SIZE) + val = VIRTIO_VSOCK_MAX_BUF_SIZE; + if (val < trans->buf_size_min) + trans->buf_size_min = val; + if (val > trans->buf_size_max) + trans->buf_size_max = val; + trans->buf_size = val; + trans->buf_alloc = val; +} +EXPORT_SYMBOL_GPL(virtio_transport_set_buffer_size); + +void virtio_transport_set_min_buffer_size(struct vsock_sock *vsk, u64 val) +{ + struct virtio_transport *trans = vsk->trans; + + if (val > VIRTIO_VSOCK_MAX_BUF_SIZE) + val = VIRTIO_VSOCK_MAX_BUF_SIZE; + if (val > trans->buf_size) + trans->buf_size = val; + trans->buf_size_min = val; +} +EXPORT_SYMBOL_GPL(virtio_transport_set_min_buffer_size); + +void virtio_transport_set_max_buffer_size(struct vsock_sock *vsk, u64 val) +{ + struct virtio_transport *trans = vsk->trans; + + if (val > VIRTIO_VSOCK_MAX_BUF_SIZE) + val = VIRTIO_VSOCK_MAX_BUF_SIZE; + if (val < trans->buf_size) + trans->buf_size = val; + trans->buf_size_max = val; +} +EXPORT_SYMBOL_GPL(virtio_transport_set_max_buffer_size); + +int +virtio_transport_notify_poll_in(struct vsock_sock *vsk, + size_t target, + bool *data_ready_now) +{ + if (vsock_stream_has_data(vsk)) + *data_ready_now = true; + else + *data_ready_now = false; + + return 0; +} +EXPORT_SYMBOL_GPL(virtio_transport_notify_poll_in); + +int +virtio_transport_notify_poll_out(struct vsock_sock *vsk, + size_t target, + bool *space_avail_now) +{ + s64 free_space; + + free_space = vsock_stream_has_space(vsk); + if (free_space > 0) + *space_avail_now = true; + else if (free_space == 0) + *space_avail_now = false; + + return 0; +} +EXPORT_SYMBOL_GPL(virtio_transport_notify_poll_out); + +int virtio_transport_notify_recv_init(struct vsock_sock *vsk, + size_t target, struct vsock_transport_recv_notify_data *data) +{ + return 0; +} +EXPORT_SYMBOL_GPL(virtio_transport_notify_recv_init); + +int virtio_transport_notify_recv_pre_block(struct vsock_sock *vsk, + size_t target, struct vsock_transport_recv_notify_data *data) +{ + return 0; +} +EXPORT_SYMBOL_GPL(virtio_transport_notify_recv_pre_block); + +int virtio_transport_notify_recv_pre_dequeue(struct vsock_sock *vsk, + size_t target, struct vsock_transport_recv_notify_data *data) +{ + return 0; +} +EXPORT_SYMBOL_GPL(virtio_transport_notify_recv_pre_dequeue); + +int virtio_transport_notify_recv_post_dequeue(struct vsock_sock *vsk, + size_t target, ssize_t copied, bool data_read, + struct vsock_transport_recv_notify_data *data) +{ + return 0; +} +EXPORT_SYMBOL_GPL(virtio_transport_notify_recv_post_dequeue); + +int virtio_transport_notify_send_init(struct vsock_sock *vsk, + struct vsock_transport_send_notify_data *data) +{ + return 0; +} +EXPORT_SYMBOL_GPL(virtio_transport_notify_send_init); + +int virtio_transport_notify_send_pre_block(struct vsock_sock *vsk, + struct vsock_transport_send_notify_data *data) +{ + return 0; +} +EXPORT_SYMBOL_GPL(virtio_transport_notify_send_pre_block); + +int virtio_transport_notify_send_pre_enqueue(struct vsock_sock *vsk, + struct vsock_transport_send_notify_data *data) +{ + return 0; +} +EXPORT_SYMBOL_GPL(virtio_transport_notify_send_pre_enqueue); + +int virtio_transport_notify_send_post_enqueue(struct vsock_sock *vsk, + ssize_t written, struct vsock_transport_send_notify_data *data) +{ + return 0; +} +EXPORT_SYMBOL_GPL(virtio_transport_notify_send_post_enqueue); + +u64 virtio_transport_stream_rcvhiwat(struct vsock_sock *vsk) +{ + struct virtio_transport *trans = vsk->trans; + + return trans->buf_size; +} +EXPORT_SYMBOL_GPL(virtio_transport_stream_rcvhiwat); + +bool virtio_transport_stream_is_active(struct vsock_sock *vsk) +{ + return true; +} +EXPORT_SYMBOL_GPL(virtio_transport_stream_is_active); + +bool virtio_transport_stream_allow(u32 cid, u32 port) +{ + return true; +} +EXPORT_SYMBOL_GPL(virtio_transport_stream_allow); + +int virtio_transport_dgram_bind(struct vsock_sock *vsk, + struct sockaddr_vm *addr) +{ + return vsock_bind_dgram_generic(vsk, addr); +} +EXPORT_SYMBOL_GPL(virtio_transport_dgram_bind); + +bool virtio_transport_dgram_allow(u32 cid, u32 port) +{ + return true; +} +EXPORT_SYMBOL_GPL(virtio_transport_dgram_allow); + +int virtio_transport_connect(struct vsock_sock *vsk) +{ + struct virtio_transport *trans = vsk->trans; + struct virtio_vsock_pkt_info info = { + .op = VIRTIO_VSOCK_OP_REQUEST, + .type = VIRTIO_VSOCK_TYPE_STREAM, + }; + + pr_debug("%s: vsk=%p send_request\n", __func__, vsk); + return trans->ops->send_pkt(vsk, &info); +} +EXPORT_SYMBOL_GPL(virtio_transport_connect); + +int virtio_transport_shutdown(struct vsock_sock *vsk, int mode) +{ + struct virtio_transport *trans = vsk->trans; + struct virtio_vsock_pkt_info info = { + .op = VIRTIO_VSOCK_OP_SHUTDOWN, + .type = VIRTIO_VSOCK_TYPE_STREAM, + .flags = (mode & RCV_SHUTDOWN ? + VIRTIO_VSOCK_SHUTDOWN_RCV : 0) | + (mode & SEND_SHUTDOWN ? + VIRTIO_VSOCK_SHUTDOWN_SEND : 0), + }; + + pr_debug("%s: vsk=%p: send_shutdown\n", __func__, vsk); + return trans->ops->send_pkt(vsk, &info); +} +EXPORT_SYMBOL_GPL(virtio_transport_shutdown); + +void virtio_transport_release(struct vsock_sock *vsk) +{ + struct virtio_transport *trans = vsk->trans; + struct sock *sk = &vsk->sk; + struct dgram_skb *dgram_skb; + struct dgram_skb *dgram_skb_tmp; + + pr_debug("%s: vsk=%p\n", __func__, vsk); + + /* Tell other side to terminate connection */ + if (sk->sk_type == SOCK_STREAM && sk->sk_state == SS_CONNECTED) { + virtio_transport_shutdown(vsk, SHUTDOWN_MASK); + } + + /* Free incomplete dgrams */ + lock_sock(sk); + list_for_each_entry_safe(dgram_skb, dgram_skb_tmp, + &trans->incomplete_dgrams, list) { + list_del(&dgram_skb->list); + kfree_skb(dgram_skb->skb); + kfree(dgram_skb); + sock_put(sk); /* held in virtio_transport_recv_dgram() */ + } + release_sock(sk); +} +EXPORT_SYMBOL_GPL(virtio_transport_release); + +int +virtio_transport_dgram_enqueue(struct vsock_sock *vsk, + struct sockaddr_vm *remote_addr, + struct msghdr *msg, + size_t dgram_len) +{ + struct virtio_transport *trans = vsk->trans; + struct virtio_vsock_pkt_info info = { + .op = VIRTIO_VSOCK_OP_RW, + .type = VIRTIO_VSOCK_TYPE_DGRAM, + .msg = msg, + }; + size_t total_written = 0, pkt_off = 0, written; + u16 dgram_id; + + /* The max size of a single dgram we support is 64KB */ + if (dgram_len > VIRTIO_VSOCK_MAX_DGRAM_SIZE) + return -EMSGSIZE; + + info.dgram_len = dgram_len; + vsk->remote_addr = *remote_addr; + + dgram_id = trans->dgram_id++; + + /* TODO: To optimize, if we have enough credit to send the pkt already, + * do not ask the peer to send credit to use */ + virtio_transport_send_credit_request(vsk, VIRTIO_VSOCK_TYPE_DGRAM); + + while (total_written < dgram_len) { + info.pkt_len = dgram_len - total_written; + info.flags = dgram_id << 16 | pkt_off; + written = trans->ops->send_pkt(vsk, &info); + if (written < 0) + return -ENOMEM; + if (written == 0) { + /* TODO: if written = 0, we need a sleep & wakeup + * instead of sleep */ + pr_debug("%s: SHOULD WAIT written==0", __func__); + msleep(10); + } + total_written += written; + pkt_off += written; + pr_debug("%s:id=%d, dgram_len=%zu, off=%zu, total_written=%zu, written=%zu\n", + __func__, dgram_id, dgram_len, pkt_off, total_written, written); + } + + return dgram_len; +} +EXPORT_SYMBOL_GPL(virtio_transport_dgram_enqueue); + +ssize_t +virtio_transport_stream_enqueue(struct vsock_sock *vsk, + struct msghdr *msg, + size_t len) +{ + struct virtio_transport *trans = vsk->trans; + struct virtio_vsock_pkt_info info = { + .op = VIRTIO_VSOCK_OP_RW, + .type = VIRTIO_VSOCK_TYPE_STREAM, + .msg = msg, + .pkt_len = len, + }; + + return trans->ops->send_pkt(vsk, &info); +} +EXPORT_SYMBOL_GPL(virtio_transport_stream_enqueue); + +void virtio_transport_destruct(struct vsock_sock *vsk) +{ + struct virtio_transport *trans = vsk->trans; + + pr_debug("%s: vsk=%p\n", __func__, vsk); + kfree(trans); +} +EXPORT_SYMBOL_GPL(virtio_transport_destruct); + +static int virtio_transport_send_ack(struct vsock_sock *vsk, u32 cookie) +{ + struct virtio_transport *trans = vsk->trans; + struct virtio_vsock_pkt_info info = { + .op = VIRTIO_VSOCK_OP_ACK, + .type = VIRTIO_VSOCK_TYPE_STREAM, + .flags = cpu_to_le32(cookie), + }; + + pr_debug("%s: sk=%p send_offer\n", __func__, vsk); + return trans->ops->send_pkt(vsk, &info); +} + +static int virtio_transport_send_reset(struct vsock_sock *vsk, + struct virtio_vsock_pkt *pkt) +{ + struct virtio_transport *trans = vsk->trans; + struct virtio_vsock_pkt_info info = { + .op = VIRTIO_VSOCK_OP_RST, + .type = VIRTIO_VSOCK_TYPE_STREAM, + }; + + pr_debug("%s\n", __func__); + + /* Send RST only if the original pkt is not a RST pkt */ + if (le16_to_cpu(pkt->hdr.op) == VIRTIO_VSOCK_OP_RST) + return 0; + + return trans->ops->send_pkt(vsk, &info); +} + +static int +virtio_transport_recv_connecting(struct sock *sk, + struct virtio_vsock_pkt *pkt) +{ + struct vsock_sock *vsk = vsock_sk(sk); + int err; + int skerr; + u32 cookie; + + pr_debug("%s: vsk=%p\n", __func__, vsk); + switch (le16_to_cpu(pkt->hdr.op)) { + case VIRTIO_VSOCK_OP_RESPONSE: + cookie = le32_to_cpu(pkt->hdr.flags); + pr_debug("%s: got RESPONSE and send ACK, cookie=%x\n", __func__, cookie); + err = virtio_transport_send_ack(vsk, cookie); + if (err < 0) { + skerr = -err; + goto destroy; + } + sk->sk_state = SS_CONNECTED; + sk->sk_socket->state = SS_CONNECTED; + vsock_insert_connected(vsk); + sk->sk_state_change(sk); + break; + case VIRTIO_VSOCK_OP_INVALID: + pr_debug("%s: got invalid\n", __func__); + break; + case VIRTIO_VSOCK_OP_RST: + pr_debug("%s: got rst\n", __func__); + skerr = ECONNRESET; + err = 0; + goto destroy; + default: + pr_debug("%s: got def\n", __func__); + skerr = EPROTO; + err = -EINVAL; + goto destroy; + } + return 0; + +destroy: + virtio_transport_send_reset(vsk, pkt); + sk->sk_state = SS_UNCONNECTED; + sk->sk_err = skerr; + sk->sk_error_report(sk); + return err; +} + +static int +virtio_transport_recv_connected(struct sock *sk, + struct virtio_vsock_pkt *pkt) +{ + struct vsock_sock *vsk = vsock_sk(sk); + struct virtio_transport *trans = vsk->trans; + int err = 0; + + switch (le16_to_cpu(pkt->hdr.op)) { + case VIRTIO_VSOCK_OP_RW: + pkt->len = le32_to_cpu(pkt->hdr.len); + pkt->off = 0; + pkt->trans = trans; + + mutex_lock(&trans->rx_lock); + virtio_transport_inc_rx_pkt(pkt); + list_add_tail(&pkt->list, &trans->rx_queue); + mutex_unlock(&trans->rx_lock); + + sk->sk_data_ready(sk); + return err; + case VIRTIO_VSOCK_OP_CREDIT_UPDATE: + sk->sk_write_space(sk); + break; + case VIRTIO_VSOCK_OP_SHUTDOWN: + pr_debug("%s: got shutdown\n", __func__); + if (le32_to_cpu(pkt->hdr.flags) & VIRTIO_VSOCK_SHUTDOWN_RCV) + vsk->peer_shutdown |= RCV_SHUTDOWN; + if (le32_to_cpu(pkt->hdr.flags) & VIRTIO_VSOCK_SHUTDOWN_SEND) + vsk->peer_shutdown |= SEND_SHUTDOWN; + if (le32_to_cpu(pkt->hdr.flags)) + sk->sk_state_change(sk); + break; + case VIRTIO_VSOCK_OP_RST: + pr_debug("%s: got rst\n", __func__); + sock_set_flag(sk, SOCK_DONE); + vsk->peer_shutdown = SHUTDOWN_MASK; + if (vsock_stream_has_data(vsk) <= 0) + sk->sk_state = SS_DISCONNECTING; + sk->sk_state_change(sk); + break; + default: + err = -EINVAL; + break; + } + + virtio_transport_free_pkt(pkt); + return err; +} + +static int +virtio_transport_send_response(struct vsock_sock *vsk, + struct virtio_vsock_pkt *pkt) +{ + struct virtio_transport *trans = vsk->trans; + struct virtio_vsock_pkt_info info = { + .op = VIRTIO_VSOCK_OP_RESPONSE, + .type = VIRTIO_VSOCK_TYPE_STREAM, + .remote_cid = le32_to_cpu(pkt->hdr.src_cid), + .remote_port = le32_to_cpu(pkt->hdr.src_port), + }; + u32 cookie; + + cookie = virtio_vsock_secure_cookie(le32_to_cpu(pkt->hdr.src_cid), + le32_to_cpu(pkt->hdr.dst_cid), + le32_to_cpu(pkt->hdr.src_port), + le32_to_cpu(pkt->hdr.dst_port), + jiffies / (HZ * 60)); + info.flags = cpu_to_le32(cookie); + + pr_debug("%s: send_response, cookie=%x\n", __func__, le32_to_cpu(cookie)); + + return trans->ops->send_pkt(vsk, &info); +} + +/* Handle server socket */ +static int +virtio_transport_recv_listen(struct sock *sk, struct virtio_vsock_pkt *pkt) +{ + struct vsock_sock *vsk = vsock_sk(sk); + struct vsock_sock *vpending; + struct sock *pending; + int err; + u32 cookie; + + switch (le16_to_cpu(pkt->hdr.op)) { + case VIRTIO_VSOCK_OP_REQUEST: + err = virtio_transport_send_response(vsk, pkt); + if (err < 0) { + // FIXME vsk should be vpending + virtio_transport_send_reset(vsk, pkt); + return err; + } + break; + case VIRTIO_VSOCK_OP_ACK: + cookie = le32_to_cpu(pkt->hdr.flags); + err = virtio_vsock_check_cookie(le32_to_cpu(pkt->hdr.src_cid), + le32_to_cpu(pkt->hdr.dst_cid), + le32_to_cpu(pkt->hdr.src_port), + le32_to_cpu(pkt->hdr.dst_port), + jiffies / (HZ * 60), + le32_to_cpu(pkt->hdr.flags), + VSOCK_TIMEOUT_INIT); + pr_debug("%s: cookie=%x, err=%d\n", __func__, cookie, err); + if (err) + return err; + + /* So no pending socket are responsible for this pkt, create one */ + pr_debug("%s: create pending\n", __func__); + pending = __vsock_create(sock_net(sk), NULL, sk, GFP_KERNEL, + sk->sk_type, 0); + if (!pending) { + virtio_transport_send_reset(vsk, pkt); + return -ENOMEM; + } + sk->sk_ack_backlog++; + pending->sk_state = SS_CONNECTING; + + vpending = vsock_sk(pending); + vsock_addr_init(&vpending->local_addr, le32_to_cpu(pkt->hdr.dst_cid), + le32_to_cpu(pkt->hdr.dst_port)); + vsock_addr_init(&vpending->remote_addr, le32_to_cpu(pkt->hdr.src_cid), + le32_to_cpu(pkt->hdr.src_port)); + vsock_add_pending(sk, pending); + + pr_debug("%s: get pending\n", __func__); + pending = virtio_transport_get_pending(sk, pkt); + vpending = vsock_sk(pending); + lock_sock(pending); + switch (pending->sk_state) { + case SS_CONNECTING: + if (le16_to_cpu(pkt->hdr.op) != VIRTIO_VSOCK_OP_ACK) { + pr_debug("%s: op=%d != OP_ACK\n", __func__, + le16_to_cpu(pkt->hdr.op)); + virtio_transport_send_reset(vpending, pkt); + pending->sk_err = EPROTO; + pending->sk_state = SS_UNCONNECTED; + sock_put(pending); + } else { + pending->sk_state = SS_CONNECTED; + vsock_insert_connected(vpending); + + vsock_remove_pending(sk, pending); + vsock_enqueue_accept(sk, pending); + + sk->sk_data_ready(sk); + } + err = 0; + break; + default: + pr_debug("%s: sk->sk_ack_backlog=%d\n", __func__, + sk->sk_ack_backlog); + virtio_transport_send_reset(vpending, pkt); + err = -EINVAL; + break; + } + if (err < 0) + vsock_remove_pending(sk, pending); + release_sock(pending); + + /* Release refcnt obtained in virtio_transport_get_pending */ + sock_put(pending); + break; + default: + break; + } + + return 0; +} + +static void virtio_transport_space_update(struct sock *sk, + struct virtio_vsock_pkt *pkt) +{ + struct vsock_sock *vsk = vsock_sk(sk); + struct virtio_transport *trans = vsk->trans; + bool space_available; + + /* buf_alloc and fwd_cnt is always included in the hdr */ + mutex_lock(&trans->tx_lock); + trans->peer_buf_alloc = le32_to_cpu(pkt->hdr.buf_alloc); + trans->peer_fwd_cnt = le32_to_cpu(pkt->hdr.fwd_cnt); + space_available = virtio_transport_has_space(vsk); + mutex_unlock(&trans->tx_lock); + + if (space_available) + sk->sk_write_space(sk); +} + +/* We are under the virtio-vsock's vsock->rx_lock or + * vhost-vsock's vq->mutex lock */ +void virtio_transport_recv_pkt(struct virtio_vsock_pkt *pkt) +{ + struct virtio_transport *trans; + struct sockaddr_vm src, dst; + struct vsock_sock *vsk; + struct sock *sk; + + vsock_addr_init(&src, le32_to_cpu(pkt->hdr.src_cid), le32_to_cpu(pkt->hdr.src_port)); + vsock_addr_init(&dst, le32_to_cpu(pkt->hdr.dst_cid), le32_to_cpu(pkt->hdr.dst_port)); + + virtio_vsock_dumppkt(__func__, pkt); + + if (le16_to_cpu(pkt->hdr.type) == VIRTIO_VSOCK_TYPE_DGRAM) { + sk = vsock_find_unbound_socket(&dst); + if (!sk) + goto free_pkt; + + vsk = vsock_sk(sk); + trans = vsk->trans; + BUG_ON(!trans); + + virtio_transport_space_update(sk, pkt); + + lock_sock(sk); + switch (le16_to_cpu(pkt->hdr.op)) { + case VIRTIO_VSOCK_OP_CREDIT_UPDATE: + virtio_transport_free_pkt(pkt); + break; + case VIRTIO_VSOCK_OP_CREDIT_REQUEST: + virtio_transport_send_credit_update(vsk, VIRTIO_VSOCK_TYPE_DGRAM, + &pkt->hdr); + virtio_transport_free_pkt(pkt); + break; + case VIRTIO_VSOCK_OP_RW: + virtio_transport_recv_dgram(sk, pkt); + break; + default: + virtio_transport_free_pkt(pkt); + break; + } + release_sock(sk); + + /* Release refcnt obtained when we fetched this socket out of + * the unbound list. + */ + sock_put(sk); + return; + } else if (le16_to_cpu(pkt->hdr.type) == VIRTIO_VSOCK_TYPE_STREAM) { + /* The socket must be in connected or bound table + * otherwise send reset back + */ + sk = vsock_find_connected_socket(&src, &dst); + if (!sk) { + sk = vsock_find_bound_socket(&dst); + if (!sk) { + pr_debug("%s: can not find bound_socket\n", __func__); + virtio_vsock_dumppkt(__func__, pkt); + /* Ignore this pkt instead of sending reset back */ + /* TODO send a RST unless this packet is a RST (to avoid infinite loops) */ + goto free_pkt; + } + } + + vsk = vsock_sk(sk); + trans = vsk->trans; + BUG_ON(!trans); + + virtio_transport_space_update(sk, pkt); + + lock_sock(sk); + switch (sk->sk_state) { + case VSOCK_SS_LISTEN: + virtio_transport_recv_listen(sk, pkt); + virtio_transport_free_pkt(pkt); + break; + case SS_CONNECTING: + virtio_transport_recv_connecting(sk, pkt); + virtio_transport_free_pkt(pkt); + break; + case SS_CONNECTED: + virtio_transport_recv_connected(sk, pkt); + break; + default: + virtio_transport_free_pkt(pkt); + break; + } + release_sock(sk); + + /* Release refcnt obtained when we fetched this socket out of the + * bound or connected list. + */ + sock_put(sk); + } + return; + +free_pkt: + virtio_transport_free_pkt(pkt); +} +EXPORT_SYMBOL_GPL(virtio_transport_recv_pkt); + +void virtio_transport_free_pkt(struct virtio_vsock_pkt *pkt) +{ + kfree(pkt->buf); + kfree(pkt); +} +EXPORT_SYMBOL_GPL(virtio_transport_free_pkt); + +static int __init virtio_vsock_common_init(void) +{ + get_random_bytes(vsockcookie_secret, sizeof(vsockcookie_secret)); + return 0; +} + +static void __exit virtio_vsock_common_exit(void) +{ +} + +module_init(virtio_vsock_common_init); +module_exit(virtio_vsock_common_exit); +MODULE_LICENSE("GPL v2"); +MODULE_AUTHOR("Asias He"); +MODULE_DESCRIPTION("common code for virtio vsock"); diff --git a/net/vmw_vsock/vmci_transport.h b/net/vmw_vsock/vmci_transport.h index 2ad46f3..1820e74 100644 --- a/net/vmw_vsock/vmci_transport.h +++ b/net/vmw_vsock/vmci_transport.h @@ -121,7 +121,7 @@ struct vmci_transport { u64 queue_pair_max_size; u32 detach_sub_id; union vmci_transport_notify notify; - struct vmci_transport_notify_ops *notify_ops; + const struct vmci_transport_notify_ops *notify_ops; struct list_head elem; struct sock *sk; spinlock_t lock; /* protects sk. */ diff --git a/net/vmw_vsock/vmci_transport_notify.c b/net/vmw_vsock/vmci_transport_notify.c index 9b7f207..fd8cf02 100644 --- a/net/vmw_vsock/vmci_transport_notify.c +++ b/net/vmw_vsock/vmci_transport_notify.c @@ -661,7 +661,7 @@ static void vmci_transport_notify_pkt_process_negotiate(struct sock *sk) } /* Socket control packet based operations. */ -struct vmci_transport_notify_ops vmci_transport_notify_pkt_ops = { +const struct vmci_transport_notify_ops vmci_transport_notify_pkt_ops = { vmci_transport_notify_pkt_socket_init, vmci_transport_notify_pkt_socket_destruct, vmci_transport_notify_pkt_poll_in, diff --git a/net/vmw_vsock/vmci_transport_notify.h b/net/vmw_vsock/vmci_transport_notify.h index 7df7932..3c464d3 100644 --- a/net/vmw_vsock/vmci_transport_notify.h +++ b/net/vmw_vsock/vmci_transport_notify.h @@ -77,7 +77,8 @@ struct vmci_transport_notify_ops { void (*process_negotiate) (struct sock *sk); }; -extern struct vmci_transport_notify_ops vmci_transport_notify_pkt_ops; -extern struct vmci_transport_notify_ops vmci_transport_notify_pkt_q_state_ops; +extern const struct vmci_transport_notify_ops vmci_transport_notify_pkt_ops; +extern const +struct vmci_transport_notify_ops vmci_transport_notify_pkt_q_state_ops; #endif /* __VMCI_TRANSPORT_NOTIFY_H__ */ diff --git a/net/vmw_vsock/vmci_transport_notify_qstate.c b/net/vmw_vsock/vmci_transport_notify_qstate.c index dc9c792..21e591d 100644 --- a/net/vmw_vsock/vmci_transport_notify_qstate.c +++ b/net/vmw_vsock/vmci_transport_notify_qstate.c @@ -419,7 +419,7 @@ vmci_transport_notify_pkt_send_pre_enqueue( } /* Socket always on control packet based operations. */ -struct vmci_transport_notify_ops vmci_transport_notify_pkt_q_state_ops = { +const struct vmci_transport_notify_ops vmci_transport_notify_pkt_q_state_ops = { vmci_transport_notify_pkt_socket_init, vmci_transport_notify_pkt_socket_destruct, vmci_transport_notify_pkt_poll_in, |