diff options
Diffstat (limited to 'drivers/infiniband/ulp/ipoib')
| -rw-r--r-- | drivers/infiniband/ulp/ipoib/Kconfig | 8 | ||||
| -rw-r--r-- | drivers/infiniband/ulp/ipoib/ipoib.h | 22 | ||||
| -rw-r--r-- | drivers/infiniband/ulp/ipoib/ipoib_cm.c | 53 | ||||
| -rw-r--r-- | drivers/infiniband/ulp/ipoib/ipoib_ethtool.c | 21 | ||||
| -rw-r--r-- | drivers/infiniband/ulp/ipoib/ipoib_fs.c | 50 | ||||
| -rw-r--r-- | drivers/infiniband/ulp/ipoib/ipoib_ib.c | 91 | ||||
| -rw-r--r-- | drivers/infiniband/ulp/ipoib/ipoib_main.c | 195 | ||||
| -rw-r--r-- | drivers/infiniband/ulp/ipoib/ipoib_multicast.c | 42 | ||||
| -rw-r--r-- | drivers/infiniband/ulp/ipoib/ipoib_netlink.c | 13 | ||||
| -rw-r--r-- | drivers/infiniband/ulp/ipoib/ipoib_verbs.c | 11 | ||||
| -rw-r--r-- | drivers/infiniband/ulp/ipoib/ipoib_vlan.c | 12 |
11 files changed, 251 insertions, 267 deletions
diff --git a/drivers/infiniband/ulp/ipoib/Kconfig b/drivers/infiniband/ulp/ipoib/Kconfig index 7af68604af77..254e31a90a66 100644 --- a/drivers/infiniband/ulp/ipoib/Kconfig +++ b/drivers/infiniband/ulp/ipoib/Kconfig @@ -2,7 +2,7 @@ config INFINIBAND_IPOIB tristate "IP-over-InfiniBand" depends on NETDEVICES && INET && (IPV6 || IPV6=n) - ---help--- + help Support for the IP-over-InfiniBand protocol (IPoIB). This transports IP packets over InfiniBand so you can use your IB device as a fancy NIC. @@ -13,7 +13,7 @@ config INFINIBAND_IPOIB_CM bool "IP-over-InfiniBand Connected Mode support" depends on INFINIBAND_IPOIB default n - ---help--- + help This option enables support for IPoIB connected mode. After enabling this option, you need to switch to connected mode through /sys/class/net/ibXXX/mode to actually create @@ -28,7 +28,7 @@ config INFINIBAND_IPOIB_DEBUG bool "IP-over-InfiniBand debugging" if EXPERT depends on INFINIBAND_IPOIB default y - ---help--- + help This option causes debugging code to be compiled into the IPoIB driver. The output can be turned on via the debug_level and mcast_debug_level module parameters (which @@ -42,7 +42,7 @@ config INFINIBAND_IPOIB_DEBUG config INFINIBAND_IPOIB_DEBUG_DATA bool "IP-over-InfiniBand data path debugging" depends on INFINIBAND_IPOIB_DEBUG - ---help--- + help This option compiles debugging code into the data path of the IPoIB driver. The output can be turned on via the data_debug_level module parameter; however, even with output diff --git a/drivers/infiniband/ulp/ipoib/ipoib.h b/drivers/infiniband/ulp/ipoib/ipoib.h index 2aa3457a30ce..35e9c8a330e2 100644 --- a/drivers/infiniband/ulp/ipoib/ipoib.h +++ b/drivers/infiniband/ulp/ipoib/ipoib.h @@ -377,8 +377,12 @@ struct ipoib_dev_priv { struct ipoib_rx_buf *rx_ring; struct ipoib_tx_buf *tx_ring; + /* cyclic ring variables for managing tx_ring, for UD only */ unsigned int tx_head; unsigned int tx_tail; + /* cyclic ring variables for counting overall outstanding send WRs */ + unsigned int global_tx_head; + unsigned int global_tx_tail; struct ib_sge tx_sge[MAX_SKB_FRAGS + 1]; struct ib_ud_wr tx_wr; struct ib_wc send_wc[MAX_SEND_CQE]; @@ -407,9 +411,9 @@ struct ipoib_dev_priv { struct dentry *path_dentry; #endif u64 hca_caps; + u64 kernel_caps; struct ipoib_ethtool_st ethtool; unsigned int max_send_sge; - bool sm_fullmember_sendonly_support; const struct net_device_ops *rn_ops; }; @@ -451,7 +455,7 @@ struct ipoib_neigh { struct list_head list; struct ipoib_neigh __rcu *hnext; struct rcu_head rcu; - atomic_t refcnt; + refcount_t refcnt; unsigned long alive; }; @@ -461,7 +465,7 @@ struct ipoib_neigh { void ipoib_neigh_dtor(struct ipoib_neigh *neigh); static inline void ipoib_neigh_put(struct ipoib_neigh *neigh) { - if (atomic_dec_and_test(&neigh->refcnt)) + if (refcount_dec_and_test(&neigh->refcnt)) ipoib_neigh_dtor(neigh); } struct ipoib_neigh *ipoib_neigh_get(struct net_device *dev, u8 *daddr); @@ -498,9 +502,9 @@ void ipoib_reap_ah(struct work_struct *work); struct ipoib_path *__path_find(struct net_device *dev, void *gid); void ipoib_mark_paths_invalid(struct net_device *dev); void ipoib_flush_paths(struct net_device *dev); -struct net_device *ipoib_intf_alloc(struct ib_device *hca, u8 port, +struct net_device *ipoib_intf_alloc(struct ib_device *hca, u32 port, const char *format); -int ipoib_intf_init(struct ib_device *hca, u8 port, const char *format, +int ipoib_intf_init(struct ib_device *hca, u32 port, const char *format, struct net_device *dev); void ipoib_ib_tx_timer_func(struct timer_list *t); void ipoib_ib_dev_flush_light(struct work_struct *work); @@ -511,7 +515,7 @@ void ipoib_ib_dev_cleanup(struct net_device *dev); int ipoib_ib_dev_open_default(struct net_device *dev); int ipoib_ib_dev_open(struct net_device *dev); -int ipoib_ib_dev_stop(struct net_device *dev); +void ipoib_ib_dev_stop(struct net_device *dev); void ipoib_ib_dev_up(struct net_device *dev); void ipoib_ib_dev_down(struct net_device *dev); int ipoib_ib_dev_stop_default(struct net_device *dev); @@ -523,7 +527,7 @@ void ipoib_mcast_send(struct net_device *dev, u8 *daddr, struct sk_buff *skb); void ipoib_mcast_restart_task(struct work_struct *work); void ipoib_mcast_start_thread(struct net_device *dev); -int ipoib_mcast_stop_thread(struct net_device *dev); +void ipoib_mcast_stop_thread(struct net_device *dev); void ipoib_mcast_dev_down(struct net_device *dev); void ipoib_mcast_dev_flush(struct net_device *dev); @@ -674,8 +678,6 @@ void ipoib_cm_handle_rx_wc(struct net_device *dev, struct ib_wc *wc); void ipoib_cm_handle_tx_wc(struct net_device *dev, struct ib_wc *wc); #else -struct ipoib_cm_tx; - #define ipoib_max_conn_qp 0 static inline int ipoib_cm_admin_enabled(struct net_device *dev) @@ -838,6 +840,4 @@ extern int ipoib_debug_level; #define IPOIB_QPN(ha) (be32_to_cpup((__be32 *) ha) & 0xffffff) -extern const char ipoib_driver_version[]; - #endif /* _IPOIB_H */ diff --git a/drivers/infiniband/ulp/ipoib/ipoib_cm.c b/drivers/infiniband/ulp/ipoib/ipoib_cm.c index c59e00a0881f..b610d36295bb 100644 --- a/drivers/infiniband/ulp/ipoib/ipoib_cm.c +++ b/drivers/infiniband/ulp/ipoib/ipoib_cm.c @@ -465,7 +465,7 @@ static int ipoib_cm_req_handler(struct ib_cm_id *cm_id, goto err_qp; } - psn = prandom_u32() & 0xffffff; + psn = get_random_u32() & 0xffffff; ret = ipoib_cm_modify_rx_qp(dev, cm_id, p->qp, psn); if (ret) goto err_modify; @@ -512,13 +512,13 @@ static int ipoib_cm_rx_handler(struct ib_cm_id *cm_id, return ipoib_cm_req_handler(cm_id, event); case IB_CM_DREQ_RECEIVED: ib_send_cm_drep(cm_id, NULL, 0); - /* Fall through */ + fallthrough; case IB_CM_REJ_RECEIVED: p = cm_id->context; priv = ipoib_priv(p->dev); if (ib_modify_qp(p->qp, &ipoib_cm_err_attr, IB_QP_STATE)) ipoib_warn(priv, "unable to move qp to error state\n"); - /* Fall through */ + fallthrough; default: return 0; } @@ -756,7 +756,8 @@ void ipoib_cm_send(struct net_device *dev, struct sk_buff *skb, struct ipoib_cm_ return; } - if ((priv->tx_head - priv->tx_tail) == ipoib_sendq_size - 1) { + if ((priv->global_tx_head - priv->global_tx_tail) == + ipoib_sendq_size - 1) { ipoib_dbg(priv, "TX ring 0x%x full, stopping kernel net queue\n", tx->qp->qp_num); netif_stop_queue(dev); @@ -786,7 +787,7 @@ void ipoib_cm_send(struct net_device *dev, struct sk_buff *skb, struct ipoib_cm_ } else { netif_trans_update(dev); ++tx->tx_head; - ++priv->tx_head; + ++priv->global_tx_head; } } @@ -820,10 +821,11 @@ void ipoib_cm_handle_tx_wc(struct net_device *dev, struct ib_wc *wc) netif_tx_lock(dev); ++tx->tx_tail; - ++priv->tx_tail; + ++priv->global_tx_tail; if (unlikely(netif_queue_stopped(dev) && - (priv->tx_head - priv->tx_tail) <= ipoib_sendq_size >> 1 && + ((priv->global_tx_head - priv->global_tx_tail) <= + ipoib_sendq_size >> 1) && test_bit(IPOIB_FLAG_ADMIN_UP, &priv->flags))) netif_wake_queue(dev); @@ -882,8 +884,8 @@ int ipoib_cm_dev_open(struct net_device *dev) goto err_cm; } - ret = ib_cm_listen(priv->cm.id, cpu_to_be64(IPOIB_CM_IETF_ID | priv->qp->qp_num), - 0); + ret = ib_cm_listen(priv->cm.id, + cpu_to_be64(IPOIB_CM_IETF_ID | priv->qp->qp_num)); if (ret) { pr_warn("%s: failed to listen on ID 0x%llx\n", priv->ca->name, IPOIB_CM_IETF_ID | priv->qp->qp_num); @@ -1120,12 +1122,8 @@ static int ipoib_cm_modify_tx_init(struct net_device *dev, struct ipoib_dev_priv *priv = ipoib_priv(dev); struct ib_qp_attr qp_attr; int qp_attr_mask, ret; - ret = ib_find_pkey(priv->ca, priv->port, priv->pkey, &qp_attr.pkey_index); - if (ret) { - ipoib_warn(priv, "pkey 0x%x not found: %d\n", priv->pkey, ret); - return ret; - } + qp_attr.pkey_index = priv->pkey_index; qp_attr.qp_state = IB_QPS_INIT; qp_attr.qp_access_flags = IB_ACCESS_LOCAL_WRITE; qp_attr.port_num = priv->port; @@ -1232,8 +1230,9 @@ timeout: dev_kfree_skb_any(tx_req->skb); netif_tx_lock_bh(p->dev); ++p->tx_tail; - ++priv->tx_tail; - if (unlikely(priv->tx_head - priv->tx_tail == ipoib_sendq_size >> 1) && + ++priv->global_tx_tail; + if (unlikely((priv->global_tx_head - priv->global_tx_tail) <= + ipoib_sendq_size >> 1) && netif_queue_stopped(p->dev) && test_bit(IPOIB_FLAG_ADMIN_UP, &priv->flags)) netif_wake_queue(p->dev); @@ -1504,20 +1503,20 @@ static void ipoib_cm_stale_task(struct work_struct *work) spin_unlock_irq(&priv->lock); } -static ssize_t show_mode(struct device *d, struct device_attribute *attr, +static ssize_t mode_show(struct device *d, struct device_attribute *attr, char *buf) { struct net_device *dev = to_net_dev(d); struct ipoib_dev_priv *priv = ipoib_priv(dev); if (test_bit(IPOIB_FLAG_ADMIN_CM, &priv->flags)) - return sprintf(buf, "connected\n"); + return sysfs_emit(buf, "connected\n"); else - return sprintf(buf, "datagram\n"); + return sysfs_emit(buf, "datagram\n"); } -static ssize_t set_mode(struct device *d, struct device_attribute *attr, - const char *buf, size_t count) +static ssize_t mode_store(struct device *d, struct device_attribute *attr, + const char *buf, size_t count) { struct net_device *dev = to_net_dev(d); int ret; @@ -1543,7 +1542,7 @@ static ssize_t set_mode(struct device *d, struct device_attribute *attr, return (!ret || ret == -EBUSY) ? count : ret; } -static DEVICE_ATTR(mode, S_IWUSR | S_IRUGO, show_mode, set_mode); +static DEVICE_ATTR_RW(mode); int ipoib_cm_add_mode_attr(struct net_device *dev) { @@ -1584,6 +1583,7 @@ int ipoib_cm_dev_init(struct net_device *dev) { struct ipoib_dev_priv *priv = ipoib_priv(dev); int max_srq_sge, i; + u8 addr; INIT_LIST_HEAD(&priv->cm.passive_ids); INIT_LIST_HEAD(&priv->cm.reap_list); @@ -1637,24 +1637,21 @@ int ipoib_cm_dev_init(struct net_device *dev) } } - priv->dev->dev_addr[0] = IPOIB_FLAGS_RC; + addr = IPOIB_FLAGS_RC; + dev_addr_mod(dev, 0, &addr, 1); return 0; } void ipoib_cm_dev_cleanup(struct net_device *dev) { struct ipoib_dev_priv *priv = ipoib_priv(dev); - int ret; if (!priv->cm.srq) return; ipoib_dbg(priv, "Cleanup ipoib connected mode.\n"); - ret = ib_destroy_srq(priv->cm.srq); - if (ret) - ipoib_warn(priv, "ib_destroy_srq failed: %d\n", ret); - + ib_destroy_srq(priv->cm.srq); priv->cm.srq = NULL; if (!priv->cm.srq_ring) return; diff --git a/drivers/infiniband/ulp/ipoib/ipoib_ethtool.c b/drivers/infiniband/ulp/ipoib/ipoib_ethtool.c index 63e4f9d15fd9..8af99b18d361 100644 --- a/drivers/infiniband/ulp/ipoib/ipoib_ethtool.c +++ b/drivers/infiniband/ulp/ipoib/ipoib_ethtool.c @@ -65,17 +65,16 @@ static void ipoib_get_drvinfo(struct net_device *netdev, ib_get_device_fw_str(priv->ca, drvinfo->fw_version); - strlcpy(drvinfo->bus_info, dev_name(priv->ca->dev.parent), + strscpy(drvinfo->bus_info, dev_name(priv->ca->dev.parent), sizeof(drvinfo->bus_info)); - strlcpy(drvinfo->version, ipoib_driver_version, - sizeof(drvinfo->version)); - - strlcpy(drvinfo->driver, "ib_ipoib", sizeof(drvinfo->driver)); + strscpy(drvinfo->driver, "ib_ipoib", sizeof(drvinfo->driver)); } static int ipoib_get_coalesce(struct net_device *dev, - struct ethtool_coalesce *coal) + struct ethtool_coalesce *coal, + struct kernel_ethtool_coalesce *kernel_coal, + struct netlink_ext_ack *extack) { struct ipoib_dev_priv *priv = ipoib_priv(dev); @@ -86,7 +85,9 @@ static int ipoib_get_coalesce(struct net_device *dev, } static int ipoib_set_coalesce(struct net_device *dev, - struct ethtool_coalesce *coal) + struct ethtool_coalesce *coal, + struct kernel_ethtool_coalesce *kernel_coal, + struct netlink_ext_ack *extack) { struct ipoib_dev_priv *priv = ipoib_priv(dev); int ret; @@ -169,6 +170,10 @@ static inline int ib_speed_enum_to_int(int speed) return SPEED_14000; case IB_SPEED_EDR: return SPEED_25000; + case IB_SPEED_HDR: + return SPEED_50000; + case IB_SPEED_NDR: + return SPEED_100000; } return SPEED_UNKNOWN; @@ -213,6 +218,8 @@ static int ipoib_get_link_ksettings(struct net_device *netdev, } static const struct ethtool_ops ipoib_ethtool_ops = { + .supported_coalesce_params = ETHTOOL_COALESCE_RX_USECS | + ETHTOOL_COALESCE_RX_MAX_FRAMES, .get_link_ksettings = ipoib_get_link_ksettings, .get_drvinfo = ipoib_get_drvinfo, .get_coalesce = ipoib_get_coalesce, diff --git a/drivers/infiniband/ulp/ipoib/ipoib_fs.c b/drivers/infiniband/ulp/ipoib/ipoib_fs.c index 64c19f6fa931..12ba7a0fe0b5 100644 --- a/drivers/infiniband/ulp/ipoib/ipoib_fs.c +++ b/drivers/infiniband/ulp/ipoib/ipoib_fs.c @@ -124,35 +124,14 @@ static int ipoib_mcg_seq_show(struct seq_file *file, void *iter_ptr) return 0; } -static const struct seq_operations ipoib_mcg_seq_ops = { +static const struct seq_operations ipoib_mcg_sops = { .start = ipoib_mcg_seq_start, .next = ipoib_mcg_seq_next, .stop = ipoib_mcg_seq_stop, .show = ipoib_mcg_seq_show, }; -static int ipoib_mcg_open(struct inode *inode, struct file *file) -{ - struct seq_file *seq; - int ret; - - ret = seq_open(file, &ipoib_mcg_seq_ops); - if (ret) - return ret; - - seq = file->private_data; - seq->private = inode->i_private; - - return 0; -} - -static const struct file_operations ipoib_mcg_fops = { - .owner = THIS_MODULE, - .open = ipoib_mcg_open, - .read = seq_read, - .llseek = seq_lseek, - .release = seq_release -}; +DEFINE_SEQ_ATTRIBUTE(ipoib_mcg); static void *ipoib_path_seq_start(struct seq_file *file, loff_t *pos) { @@ -229,35 +208,14 @@ static int ipoib_path_seq_show(struct seq_file *file, void *iter_ptr) return 0; } -static const struct seq_operations ipoib_path_seq_ops = { +static const struct seq_operations ipoib_path_sops = { .start = ipoib_path_seq_start, .next = ipoib_path_seq_next, .stop = ipoib_path_seq_stop, .show = ipoib_path_seq_show, }; -static int ipoib_path_open(struct inode *inode, struct file *file) -{ - struct seq_file *seq; - int ret; - - ret = seq_open(file, &ipoib_path_seq_ops); - if (ret) - return ret; - - seq = file->private_data; - seq->private = inode->i_private; - - return 0; -} - -static const struct file_operations ipoib_path_fops = { - .owner = THIS_MODULE, - .open = ipoib_path_open, - .read = seq_read, - .llseek = seq_lseek, - .release = seq_release -}; +DEFINE_SEQ_ATTRIBUTE(ipoib_path); void ipoib_create_debug_files(struct net_device *dev) { diff --git a/drivers/infiniband/ulp/ipoib/ipoib_ib.c b/drivers/infiniband/ulp/ipoib/ipoib_ib.c index c332b4761816..ed25061fac62 100644 --- a/drivers/infiniband/ulp/ipoib/ipoib_ib.c +++ b/drivers/infiniband/ulp/ipoib/ipoib_ib.c @@ -407,9 +407,11 @@ static void ipoib_ib_handle_tx_wc(struct net_device *dev, struct ib_wc *wc) dev_kfree_skb_any(tx_req->skb); ++priv->tx_tail; + ++priv->global_tx_tail; if (unlikely(netif_queue_stopped(dev) && - ((priv->tx_head - priv->tx_tail) <= ipoib_sendq_size >> 1) && + ((priv->global_tx_head - priv->global_tx_tail) <= + ipoib_sendq_size >> 1) && test_bit(IPOIB_FLAG_ADMIN_UP, &priv->flags))) netif_wake_queue(dev); @@ -571,7 +573,7 @@ int ipoib_send(struct net_device *dev, struct sk_buff *skb, unsigned int usable_sge = priv->max_send_sge - !!skb_headlen(skb); if (skb_is_gso(skb)) { - hlen = skb_transport_offset(skb) + tcp_hdrlen(skb); + hlen = skb_tcp_all_headers(skb); phead = skb->data; if (unlikely(!skb_pull(skb, hlen))) { ipoib_warn(priv, "linear data too small\n"); @@ -634,7 +636,8 @@ int ipoib_send(struct net_device *dev, struct sk_buff *skb, else priv->tx_wr.wr.send_flags &= ~IB_SEND_IP_CSUM; /* increase the tx_head after send success, but use it for queue state */ - if (priv->tx_head - priv->tx_tail == ipoib_sendq_size - 1) { + if ((priv->global_tx_head - priv->global_tx_tail) == + ipoib_sendq_size - 1) { ipoib_dbg(priv, "TX ring full, stopping kernel net queue\n"); netif_stop_queue(dev); } @@ -662,17 +665,17 @@ int ipoib_send(struct net_device *dev, struct sk_buff *skb, rc = priv->tx_head; ++priv->tx_head; + ++priv->global_tx_head; } return rc; } -static void __ipoib_reap_ah(struct net_device *dev) +static void ipoib_reap_dead_ahs(struct ipoib_dev_priv *priv) { - struct ipoib_dev_priv *priv = ipoib_priv(dev); struct ipoib_ah *ah, *tah; unsigned long flags; - netif_tx_lock_bh(dev); + netif_tx_lock_bh(priv->dev); spin_lock_irqsave(&priv->lock, flags); list_for_each_entry_safe(ah, tah, &priv->dead_ahs, list) @@ -683,37 +686,37 @@ static void __ipoib_reap_ah(struct net_device *dev) } spin_unlock_irqrestore(&priv->lock, flags); - netif_tx_unlock_bh(dev); + netif_tx_unlock_bh(priv->dev); } void ipoib_reap_ah(struct work_struct *work) { struct ipoib_dev_priv *priv = container_of(work, struct ipoib_dev_priv, ah_reap_task.work); - struct net_device *dev = priv->dev; - __ipoib_reap_ah(dev); + ipoib_reap_dead_ahs(priv); if (!test_bit(IPOIB_STOP_REAPER, &priv->flags)) queue_delayed_work(priv->wq, &priv->ah_reap_task, round_jiffies_relative(HZ)); } -static void ipoib_flush_ah(struct net_device *dev) +static void ipoib_start_ah_reaper(struct ipoib_dev_priv *priv) { - struct ipoib_dev_priv *priv = ipoib_priv(dev); - - cancel_delayed_work(&priv->ah_reap_task); - flush_workqueue(priv->wq); - ipoib_reap_ah(&priv->ah_reap_task.work); + clear_bit(IPOIB_STOP_REAPER, &priv->flags); + queue_delayed_work(priv->wq, &priv->ah_reap_task, + round_jiffies_relative(HZ)); } -static void ipoib_stop_ah(struct net_device *dev) +static void ipoib_stop_ah_reaper(struct ipoib_dev_priv *priv) { - struct ipoib_dev_priv *priv = ipoib_priv(dev); - set_bit(IPOIB_STOP_REAPER, &priv->flags); - ipoib_flush_ah(dev); + cancel_delayed_work(&priv->ah_reap_task); + /* + * After ipoib_stop_ah_reaper() we always go through + * ipoib_reap_dead_ahs() which ensures the work is really stopped and + * does a final flush out of the dead_ah's list + */ } static int recvs_pending(struct net_device *dev) @@ -807,6 +810,7 @@ int ipoib_ib_dev_stop_default(struct net_device *dev) ipoib_dma_unmap_tx(priv, tx_req); dev_kfree_skb_any(tx_req->skb); ++priv->tx_tail; + ++priv->global_tx_tail; } for (i = 0; i < ipoib_recvq_size; ++i) { @@ -841,18 +845,6 @@ timeout: return 0; } -int ipoib_ib_dev_stop(struct net_device *dev) -{ - struct ipoib_dev_priv *priv = ipoib_priv(dev); - - priv->rn_ops->ndo_stop(dev); - - clear_bit(IPOIB_FLAG_INITIALIZED, &priv->flags); - ipoib_flush_ah(dev); - - return 0; -} - int ipoib_ib_dev_open_default(struct net_device *dev) { struct ipoib_dev_priv *priv = ipoib_priv(dev); @@ -896,10 +888,7 @@ int ipoib_ib_dev_open(struct net_device *dev) return -1; } - clear_bit(IPOIB_STOP_REAPER, &priv->flags); - queue_delayed_work(priv->wq, &priv->ah_reap_task, - round_jiffies_relative(HZ)); - + ipoib_start_ah_reaper(priv); if (priv->rn_ops->ndo_open(dev)) { pr_warn("%s: Failed to open dev\n", dev->name); goto dev_stop; @@ -910,13 +899,20 @@ int ipoib_ib_dev_open(struct net_device *dev) return 0; dev_stop: - set_bit(IPOIB_STOP_REAPER, &priv->flags); - cancel_delayed_work(&priv->ah_reap_task); - set_bit(IPOIB_FLAG_INITIALIZED, &priv->flags); - ipoib_ib_dev_stop(dev); + ipoib_stop_ah_reaper(priv); return -1; } +void ipoib_ib_dev_stop(struct net_device *dev) +{ + struct ipoib_dev_priv *priv = ipoib_priv(dev); + + priv->rn_ops->ndo_stop(dev); + + clear_bit(IPOIB_FLAG_INITIALIZED, &priv->flags); + ipoib_stop_ah_reaper(priv); +} + void ipoib_pkey_dev_check_presence(struct net_device *dev) { struct ipoib_dev_priv *priv = ipoib_priv(dev); @@ -1061,13 +1057,11 @@ static bool ipoib_dev_addr_changed_valid(struct ipoib_dev_priv *priv) { union ib_gid search_gid; union ib_gid gid0; - union ib_gid *netdev_gid; int err; u16 index; - u8 port; + u32 port; bool ret = false; - netdev_gid = (union ib_gid *)(priv->dev->dev_addr + 4); if (rdma_query_gid(priv->ca, priv->port, 0, &gid0)) return false; @@ -1077,7 +1071,8 @@ static bool ipoib_dev_addr_changed_valid(struct ipoib_dev_priv *priv) * to do it later */ priv->local_gid.global.subnet_prefix = gid0.global.subnet_prefix; - netdev_gid->global.subnet_prefix = gid0.global.subnet_prefix; + dev_addr_mod(priv->dev, 4, (u8 *)&gid0.global.subnet_prefix, + sizeof(gid0.global.subnet_prefix)); search_gid.global.subnet_prefix = gid0.global.subnet_prefix; search_gid.global.interface_id = priv->local_gid.global.interface_id; @@ -1114,7 +1109,7 @@ static bool ipoib_dev_addr_changed_valid(struct ipoib_dev_priv *priv) * if he sets the device address back to be based on GID index 0, * he no longer wishs to control it. * - * If the user doesn't control the the device address, + * If the user doesn't control the device address, * IPOIB_FLAG_DEV_ADDR_SET is set and ib_find_gid failed it means * the port GUID has changed and GID at index 0 has changed * so we need to change priv->local_gid and priv->dev->dev_addr @@ -1139,8 +1134,8 @@ static bool ipoib_dev_addr_changed_valid(struct ipoib_dev_priv *priv) if (!test_bit(IPOIB_FLAG_DEV_ADDR_CTRL, &priv->flags)) { memcpy(&priv->local_gid, &gid0, sizeof(priv->local_gid)); - memcpy(priv->dev->dev_addr + 4, &gid0, - sizeof(priv->local_gid)); + dev_addr_mod(priv->dev, 4, (u8 *)&gid0, + sizeof(priv->local_gid)); ret = true; } } @@ -1227,7 +1222,7 @@ static void __ipoib_ib_dev_flush(struct ipoib_dev_priv *priv, ipoib_mcast_dev_flush(dev); if (oper_up) set_bit(IPOIB_FLAG_OPER_UP, &priv->flags); - ipoib_flush_ah(dev); + ipoib_reap_dead_ahs(priv); } if (level >= IPOIB_FLUSH_NORMAL) @@ -1302,7 +1297,7 @@ void ipoib_ib_dev_cleanup(struct net_device *dev) * the neighbor garbage collection is stopped and reaped. * That should all be done now, so make a final ah flush. */ - ipoib_stop_ah(dev); + ipoib_reap_dead_ahs(priv); clear_bit(IPOIB_PKEY_ASSIGNED, &priv->flags); diff --git a/drivers/infiniband/ulp/ipoib/ipoib_main.c b/drivers/infiniband/ulp/ipoib/ipoib_main.c index 4a0d3a9e72e1..ac25fc80fb33 100644 --- a/drivers/infiniband/ulp/ipoib/ipoib_main.c +++ b/drivers/infiniband/ulp/ipoib/ipoib_main.c @@ -52,10 +52,6 @@ #include <linux/inetdevice.h> #include <rdma/ib_cache.h> -#define DRV_VERSION "1.0.0" - -const char ipoib_driver_version[] = DRV_VERSION; - MODULE_AUTHOR("Roland Dreier"); MODULE_DESCRIPTION("IP-over-InfiniBand net driver"); MODULE_LICENSE("Dual BSD/GPL"); @@ -90,11 +86,11 @@ struct workqueue_struct *ipoib_workqueue; struct ib_sa_client ipoib_sa_client; -static void ipoib_add_one(struct ib_device *device); +static int ipoib_add_one(struct ib_device *device); static void ipoib_remove_one(struct ib_device *device, void *client_data); static void ipoib_neigh_reclaim(struct rcu_head *rp); static struct net_device *ipoib_get_net_dev_by_params( - struct ib_device *dev, u8 port, u16 pkey, + struct ib_device *dev, u32 port, u16 pkey, const union ib_gid *gid, const struct sockaddr *addr, void *client_data); static int ipoib_set_mac(struct net_device *dev, void *addr); @@ -145,8 +141,6 @@ int ipoib_open(struct net_device *dev) set_bit(IPOIB_FLAG_ADMIN_UP, &priv->flags); - priv->sm_fullmember_sendonly_support = false; - if (ipoib_ib_dev_open(dev)) { if (!test_bit(IPOIB_PKEY_ASSIGNED, &priv->flags)) return 0; @@ -170,8 +164,13 @@ int ipoib_open(struct net_device *dev) dev_change_flags(cpriv->dev, flags | IFF_UP, NULL); } up_read(&priv->vlan_rwsem); - } + } else if (priv->parent) { + struct ipoib_dev_priv *ppriv = ipoib_priv(priv->parent); + if (!test_bit(IPOIB_FLAG_ADMIN_UP, &ppriv->flags)) + ipoib_dbg(priv, "parent device %s is not up, so child device may be not functioning.\n", + ppriv->dev->name); + } netif_start_queue(dev); return 0; @@ -317,7 +316,7 @@ static bool ipoib_is_dev_match_addr_rcu(const struct sockaddr *addr, return false; } -/** +/* * Find the master net_device on top of the given net_device. * @dev: base IPoIB net_device * @@ -346,9 +345,10 @@ struct ipoib_walk_data { struct net_device *result; }; -static int ipoib_upper_walk(struct net_device *upper, void *_data) +static int ipoib_upper_walk(struct net_device *upper, + struct netdev_nested_priv *priv) { - struct ipoib_walk_data *data = _data; + struct ipoib_walk_data *data = (struct ipoib_walk_data *)priv->data; int ret = 0; if (ipoib_is_dev_match_addr_rcu(data->addr, upper)) { @@ -361,8 +361,9 @@ static int ipoib_upper_walk(struct net_device *upper, void *_data) } /** - * Find a net_device matching the given address, which is an upper device of - * the given net_device. + * ipoib_get_net_dev_match_addr - Find a net_device matching + * the given address, which is an upper device of the given net_device. + * * @addr: IP address to look for. * @dev: base IPoIB net_device * @@ -372,10 +373,12 @@ static int ipoib_upper_walk(struct net_device *upper, void *_data) static struct net_device *ipoib_get_net_dev_match_addr( const struct sockaddr *addr, struct net_device *dev) { + struct netdev_nested_priv priv; struct ipoib_walk_data data = { .addr = addr, }; + priv.data = (void *)&data; rcu_read_lock(); if (ipoib_is_dev_match_addr_rcu(addr, dev)) { dev_hold(dev); @@ -383,7 +386,7 @@ static struct net_device *ipoib_get_net_dev_match_addr( goto out; } - netdev_walk_all_upper_dev_rcu(dev, ipoib_upper_walk, &data); + netdev_walk_all_upper_dev_rcu(dev, ipoib_upper_walk, &priv); out: rcu_read_unlock(); return data.result; @@ -441,7 +444,7 @@ static int ipoib_match_gid_pkey_addr(struct ipoib_dev_priv *priv, /* Returns the number of matching net_devs found (between 0 and 2). Also * return the matching net_device in the @net_dev parameter, holding a * reference to the net_device, if the number of matches >= 1 */ -static int __ipoib_get_net_dev_by_params(struct list_head *dev_list, u8 port, +static int __ipoib_get_net_dev_by_params(struct list_head *dev_list, u32 port, u16 pkey_index, const union ib_gid *gid, const struct sockaddr *addr, @@ -466,7 +469,7 @@ static int __ipoib_get_net_dev_by_params(struct list_head *dev_list, u8 port, } static struct net_device *ipoib_get_net_dev_by_params( - struct ib_device *dev, u8 port, u16 pkey, + struct ib_device *dev, u32 port, u16 pkey, const union ib_gid *gid, const struct sockaddr *addr, void *client_data) { @@ -483,9 +486,6 @@ static struct net_device *ipoib_get_net_dev_by_params( if (ret) return NULL; - if (!dev_list) - return NULL; - /* See if we can find a unique device matching the L2 parameters */ matches = __ipoib_get_net_dev_by_params(dev_list, port, pkey_index, gid, NULL, &net_dev); @@ -509,7 +509,7 @@ static struct net_device *ipoib_get_net_dev_by_params( default: dev_warn_ratelimited(&dev->dev, "duplicate IP address detected\n"); - /* Fall through */ + fallthrough; case 1: return net_dev; } @@ -533,6 +533,7 @@ int ipoib_set_mode(struct net_device *dev, const char *buf) "will cause multicast packet drops\n"); netdev_update_features(dev); dev_set_mtu(dev, ipoib_cm_max_mtu(dev)); + netif_set_real_num_tx_queues(dev, 1); rtnl_unlock(); priv->tx_wr.wr.send_flags &= ~IB_SEND_IP_CSUM; @@ -544,6 +545,7 @@ int ipoib_set_mode(struct net_device *dev, const char *buf) clear_bit(IPOIB_FLAG_ADMIN_CM, &priv->flags); netdev_update_features(dev); dev_set_mtu(dev, min(priv->mcast_mtu, dev->mtu)); + netif_set_real_num_tx_queues(dev, dev->num_tx_queues); rtnl_unlock(); ipoib_flush_paths(dev); return (!rtnl_trylock()) ? -EBUSY : 0; @@ -740,7 +742,7 @@ void ipoib_flush_paths(struct net_device *dev) static void path_rec_completion(int status, struct sa_path_rec *pathrec, - void *path_ptr) + int num_prs, void *path_ptr) { struct ipoib_path *path = path_ptr; struct net_device *dev = path->dev; @@ -1185,12 +1187,19 @@ unref: static void ipoib_timeout(struct net_device *dev, unsigned int txqueue) { struct ipoib_dev_priv *priv = ipoib_priv(dev); + struct rdma_netdev *rn = netdev_priv(dev); + if (rn->tx_timeout) { + rn->tx_timeout(dev, txqueue); + return; + } ipoib_warn(priv, "transmit timeout: latency %d msecs\n", jiffies_to_msecs(jiffies - dev_trans_start(dev))); - ipoib_warn(priv, "queue stopped %d, tx_head %u, tx_tail %u\n", - netif_queue_stopped(dev), - priv->tx_head, priv->tx_tail); + ipoib_warn(priv, + "queue stopped %d, tx_head %u, tx_tail %u, global_tx_head %u, global_tx_tail %u\n", + netif_queue_stopped(dev), priv->tx_head, priv->tx_tail, + priv->global_tx_head, priv->global_tx_tail); + /* XXX reset QP, etc. */ } @@ -1279,7 +1288,7 @@ struct ipoib_neigh *ipoib_neigh_get(struct net_device *dev, u8 *daddr) neigh = rcu_dereference_bh(neigh->hnext)) { if (memcmp(daddr, neigh->daddr, INFINIBAND_ALEN) == 0) { /* found, take one ref on behalf of the caller */ - if (!atomic_inc_not_zero(&neigh->refcnt)) { + if (!refcount_inc_not_zero(&neigh->refcnt)) { /* deleted */ neigh = NULL; goto out_unlock; @@ -1374,7 +1383,7 @@ static struct ipoib_neigh *ipoib_neigh_ctor(u8 *daddr, INIT_LIST_HEAD(&neigh->list); ipoib_cm_set(neigh, NULL); /* one ref on behalf of the caller */ - atomic_set(&neigh->refcnt, 1); + refcount_set(&neigh->refcnt, 1); return neigh; } @@ -1406,7 +1415,7 @@ struct ipoib_neigh *ipoib_neigh_alloc(u8 *daddr, lockdep_is_held(&priv->lock))) { if (memcmp(daddr, neigh->daddr, INFINIBAND_ALEN) == 0) { /* found, take one ref on behalf of the caller */ - if (!atomic_inc_not_zero(&neigh->refcnt)) { + if (!refcount_inc_not_zero(&neigh->refcnt)) { /* deleted */ neigh = NULL; break; @@ -1421,7 +1430,7 @@ struct ipoib_neigh *ipoib_neigh_alloc(u8 *daddr, goto out_unlock; /* one ref on behalf of the hash table */ - atomic_inc(&neigh->refcnt); + refcount_inc(&neigh->refcnt); neigh->alive = jiffies; /* put in hash */ rcu_assign_pointer(neigh->hnext, @@ -1655,8 +1664,10 @@ static void ipoib_napi_add(struct net_device *dev) { struct ipoib_dev_priv *priv = ipoib_priv(dev); - netif_napi_add(dev, &priv->recv_napi, ipoib_rx_poll, IPOIB_NUM_WC); - netif_napi_add(dev, &priv->send_napi, ipoib_tx_poll, MAX_SEND_CQE); + netif_napi_add_weight(dev, &priv->recv_napi, ipoib_rx_poll, + IPOIB_NUM_WC); + netif_napi_add_weight(dev, &priv->send_napi, ipoib_tx_poll, + MAX_SEND_CQE); } static void ipoib_napi_del(struct net_device *dev) @@ -1687,6 +1698,7 @@ static void ipoib_dev_uninit_default(struct net_device *dev) static int ipoib_dev_init_default(struct net_device *dev) { struct ipoib_dev_priv *priv = ipoib_priv(dev); + u8 addr_mod[3]; ipoib_napi_add(dev); @@ -1705,7 +1717,7 @@ static int ipoib_dev_init_default(struct net_device *dev) goto out_rx_ring_cleanup; } - /* priv->tx_head, tx_tail & tx_outstanding are already 0 */ + /* priv->tx_head, tx_tail and global_tx_tail/head are already 0 */ if (ipoib_transport_dev_init(dev, priv->ca)) { pr_warn("%s: ipoib_transport_dev_init failed\n", @@ -1714,9 +1726,10 @@ static int ipoib_dev_init_default(struct net_device *dev) } /* after qp created set dev address */ - priv->dev->dev_addr[1] = (priv->qp->qp_num >> 16) & 0xff; - priv->dev->dev_addr[2] = (priv->qp->qp_num >> 8) & 0xff; - priv->dev->dev_addr[3] = (priv->qp->qp_num) & 0xff; + addr_mod[0] = (priv->qp->qp_num >> 16) & 0xff; + addr_mod[1] = (priv->qp->qp_num >> 8) & 0xff; + addr_mod[2] = (priv->qp->qp_num) & 0xff; + dev_addr_mod(priv->dev, 1, addr_mod, sizeof(addr_mod)); return 0; @@ -1736,10 +1749,10 @@ static int ipoib_ioctl(struct net_device *dev, struct ifreq *ifr, { struct ipoib_dev_priv *priv = ipoib_priv(dev); - if (!priv->rn_ops->ndo_do_ioctl) + if (!priv->rn_ops->ndo_eth_ioctl) return -EOPNOTSUPP; - return priv->rn_ops->ndo_do_ioctl(dev, ifr, cmd); + return priv->rn_ops->ndo_eth_ioctl(dev, ifr, cmd); } static int ipoib_dev_init(struct net_device *dev) @@ -1839,11 +1852,12 @@ static void ipoib_parent_unregister_pre(struct net_device *ndev) static void ipoib_set_dev_features(struct ipoib_dev_priv *priv) { priv->hca_caps = priv->ca->attrs.device_cap_flags; + priv->kernel_caps = priv->ca->attrs.kernel_cap_flags; if (priv->hca_caps & IB_DEVICE_UD_IP_CSUM) { priv->dev->hw_features |= NETIF_F_IP_CSUM | NETIF_F_RXCSUM; - if (priv->hca_caps & IB_DEVICE_UD_TSO) + if (priv->kernel_caps & IBK_UD_TSO) priv->dev->hw_features |= NETIF_F_TSO; priv->dev->features |= priv->dev->hw_features; @@ -1862,7 +1876,7 @@ static int ipoib_parent_init(struct net_device *ndev) priv->port); return result; } - priv->max_ib_mtu = ib_mtu_enum_to_int(attr.max_mtu); + priv->max_ib_mtu = rdma_mtu_from_attr(priv->ca, priv->port, &attr); result = ib_query_pkey(priv->ca, priv->port, 0, &priv->pkey); if (result) { @@ -1877,8 +1891,7 @@ static int ipoib_parent_init(struct net_device *ndev) priv->ca->name, priv->port, result); return result; } - memcpy(priv->dev->dev_addr + 4, priv->local_gid.raw, - sizeof(union ib_gid)); + dev_addr_mod(priv->dev, 4, priv->local_gid.raw, sizeof(union ib_gid)); SET_NETDEV_DEV(priv->dev, priv->ca->dev.parent); priv->dev->dev_port = priv->port - 1; @@ -1895,14 +1908,22 @@ static void ipoib_child_init(struct net_device *ndev) priv->max_ib_mtu = ppriv->max_ib_mtu; set_bit(IPOIB_FLAG_SUBINTERFACE, &priv->flags); - memcpy(priv->dev->dev_addr, ppriv->dev->dev_addr, INFINIBAND_ALEN); - memcpy(&priv->local_gid, &ppriv->local_gid, sizeof(priv->local_gid)); + if (memchr_inv(priv->dev->dev_addr, 0, INFINIBAND_ALEN)) + memcpy(&priv->local_gid, priv->dev->dev_addr + 4, + sizeof(priv->local_gid)); + else { + __dev_addr_set(priv->dev, ppriv->dev->dev_addr, + INFINIBAND_ALEN); + memcpy(&priv->local_gid, &ppriv->local_gid, + sizeof(priv->local_gid)); + } } static int ipoib_ndo_init(struct net_device *ndev) { struct ipoib_dev_priv *priv = ipoib_priv(ndev); int rc; + struct rdma_netdev *rn = netdev_priv(ndev); if (priv->parent) { ipoib_child_init(ndev); @@ -1915,6 +1936,7 @@ static int ipoib_ndo_init(struct net_device *ndev) /* MTU will be reset when mcast join happens */ ndev->mtu = IPOIB_UD_MTU(priv->max_ib_mtu); priv->mcast_mtu = priv->admin_mtu = ndev->mtu; + rn->mtu = priv->mcast_mtu; ndev->max_mtu = IPOIB_CM_MTU; ndev->neigh_priv_len = sizeof(struct ipoib_neigh); @@ -1977,7 +1999,8 @@ static void ipoib_ndo_uninit(struct net_device *dev) /* no more works over the priv->wq */ if (priv->wq) { - flush_workqueue(priv->wq); + /* See ipoib_mcast_carrier_on_task() */ + WARN_ON(test_bit(IPOIB_FLAG_OPER_UP, &priv->flags)); destroy_workqueue(priv->wq); priv->wq = NULL; } @@ -2058,7 +2081,7 @@ static const struct net_device_ops ipoib_netdev_ops_pf = { .ndo_set_vf_guid = ipoib_set_vf_guid, .ndo_set_mac_address = ipoib_set_mac, .ndo_get_stats64 = ipoib_get_stats, - .ndo_do_ioctl = ipoib_ioctl, + .ndo_eth_ioctl = ipoib_ioctl, }; static const struct net_device_ops ipoib_netdev_ops_vf = { @@ -2073,12 +2096,20 @@ static const struct net_device_ops ipoib_netdev_ops_vf = { .ndo_set_rx_mode = ipoib_set_mcast_list, .ndo_get_iflink = ipoib_get_iflink, .ndo_get_stats64 = ipoib_get_stats, - .ndo_do_ioctl = ipoib_ioctl, + .ndo_eth_ioctl = ipoib_ioctl, +}; + +static const struct net_device_ops ipoib_netdev_default_pf = { + .ndo_init = ipoib_dev_init_default, + .ndo_uninit = ipoib_dev_uninit_default, + .ndo_open = ipoib_ib_dev_open_default, + .ndo_stop = ipoib_ib_dev_stop_default, }; void ipoib_setup_common(struct net_device *dev) { dev->header_ops = &ipoib_header_ops; + dev->netdev_ops = &ipoib_netdev_default_pf; ipoib_set_ethtool_ops(dev); @@ -2128,14 +2159,7 @@ static void ipoib_build_priv(struct net_device *dev) INIT_DELAYED_WORK(&priv->neigh_reap_task, ipoib_reap_neigh); } -static const struct net_device_ops ipoib_netdev_default_pf = { - .ndo_init = ipoib_dev_init_default, - .ndo_uninit = ipoib_dev_uninit_default, - .ndo_open = ipoib_ib_dev_open_default, - .ndo_stop = ipoib_ib_dev_stop_default, -}; - -static struct net_device *ipoib_alloc_netdev(struct ib_device *hca, u8 port, +static struct net_device *ipoib_alloc_netdev(struct ib_device *hca, u32 port, const char *name) { struct net_device *dev; @@ -2152,7 +2176,7 @@ static struct net_device *ipoib_alloc_netdev(struct ib_device *hca, u8 port, return dev; } -int ipoib_intf_init(struct ib_device *hca, u8 port, const char *name, +int ipoib_intf_init(struct ib_device *hca, u32 port, const char *name, struct net_device *dev) { struct rdma_netdev *rn = netdev_priv(dev); @@ -2172,7 +2196,6 @@ int ipoib_intf_init(struct ib_device *hca, u8 port, const char *name, if (rc != -EOPNOTSUPP) goto out; - dev->netdev_ops = &ipoib_netdev_default_pf; rn->send = ipoib_send; rn->attach_mcast = ipoib_mcast_attach; rn->detach_mcast = ipoib_mcast_detach; @@ -2181,7 +2204,7 @@ int ipoib_intf_init(struct ib_device *hca, u8 port, const char *name, priv->rn_ops = dev->netdev_ops; - if (hca->attrs.device_cap_flags & IB_DEVICE_VIRTUAL_FUNCTION) + if (hca->attrs.kernel_cap_flags & IBK_VIRTUAL_FUNCTION) dev->netdev_ops = &ipoib_netdev_ops_vf; else dev->netdev_ops = &ipoib_netdev_ops_pf; @@ -2204,7 +2227,7 @@ out: return rc; } -struct net_device *ipoib_intf_alloc(struct ib_device *hca, u8 port, +struct net_device *ipoib_intf_alloc(struct ib_device *hca, u32 port, const char *name) { struct net_device *dev; @@ -2249,23 +2272,24 @@ void ipoib_intf_free(struct net_device *dev) kfree(priv); } -static ssize_t show_pkey(struct device *dev, - struct device_attribute *attr, char *buf) +static ssize_t pkey_show(struct device *dev, struct device_attribute *attr, + char *buf) { struct net_device *ndev = to_net_dev(dev); struct ipoib_dev_priv *priv = ipoib_priv(ndev); - return sprintf(buf, "0x%04x\n", priv->pkey); + return sysfs_emit(buf, "0x%04x\n", priv->pkey); } -static DEVICE_ATTR(pkey, S_IRUGO, show_pkey, NULL); +static DEVICE_ATTR_RO(pkey); -static ssize_t show_umcast(struct device *dev, - struct device_attribute *attr, char *buf) +static ssize_t umcast_show(struct device *dev, struct device_attribute *attr, + char *buf) { struct net_device *ndev = to_net_dev(dev); struct ipoib_dev_priv *priv = ipoib_priv(ndev); - return sprintf(buf, "%d\n", test_bit(IPOIB_FLAG_UMCAST, &priv->flags)); + return sysfs_emit(buf, "%d\n", + test_bit(IPOIB_FLAG_UMCAST, &priv->flags)); } void ipoib_set_umcast(struct net_device *ndev, int umcast_val) @@ -2280,9 +2304,8 @@ void ipoib_set_umcast(struct net_device *ndev, int umcast_val) clear_bit(IPOIB_FLAG_UMCAST, &priv->flags); } -static ssize_t set_umcast(struct device *dev, - struct device_attribute *attr, - const char *buf, size_t count) +static ssize_t umcast_store(struct device *dev, struct device_attribute *attr, + const char *buf, size_t count) { unsigned long umcast_val = simple_strtoul(buf, NULL, 0); @@ -2290,7 +2313,7 @@ static ssize_t set_umcast(struct device *dev, return count; } -static DEVICE_ATTR(umcast, S_IWUSR | S_IRUGO, show_umcast, set_umcast); +static DEVICE_ATTR_RW(umcast); int ipoib_add_umcast_attr(struct net_device *dev) { @@ -2307,7 +2330,7 @@ static void set_base_guid(struct ipoib_dev_priv *priv, union ib_gid *gid) memcpy(&priv->local_gid.global.interface_id, &gid->global.interface_id, sizeof(gid->global.interface_id)); - memcpy(netdev->dev_addr + 4, &priv->local_gid, sizeof(priv->local_gid)); + dev_addr_mod(netdev, 4, (u8 *)&priv->local_gid, sizeof(priv->local_gid)); clear_bit(IPOIB_FLAG_DEV_ADDR_SET, &priv->flags); netif_addr_unlock_bh(netdev); @@ -2361,9 +2384,9 @@ static int ipoib_set_mac(struct net_device *dev, void *addr) return 0; } -static ssize_t create_child(struct device *dev, - struct device_attribute *attr, - const char *buf, size_t count) +static ssize_t create_child_store(struct device *dev, + struct device_attribute *attr, + const char *buf, size_t count) { int pkey; int ret; @@ -2378,11 +2401,11 @@ static ssize_t create_child(struct device *dev, return ret ? ret : count; } -static DEVICE_ATTR(create_child, S_IWUSR, NULL, create_child); +static DEVICE_ATTR_WO(create_child); -static ssize_t delete_child(struct device *dev, - struct device_attribute *attr, - const char *buf, size_t count) +static ssize_t delete_child_store(struct device *dev, + struct device_attribute *attr, + const char *buf, size_t count) { int pkey; int ret; @@ -2398,7 +2421,7 @@ static ssize_t delete_child(struct device *dev, return ret ? ret : count; } -static DEVICE_ATTR(delete_child, S_IWUSR, NULL, delete_child); +static DEVICE_ATTR_WO(delete_child); int ipoib_add_pkey_attr(struct net_device *dev) { @@ -2435,7 +2458,7 @@ static ssize_t dev_id_show(struct device *dev, "\"%s\" wants to know my dev_id. Should it look at dev_port instead? See Documentation/ABI/testing/sysfs-class-net for more info.\n", current->comm); - return sprintf(buf, "%#x\n", ndev->dev_id); + return sysfs_emit(buf, "%#x\n", ndev->dev_id); } static DEVICE_ATTR_RO(dev_id); @@ -2446,7 +2469,7 @@ static int ipoib_intercept_dev_id_attr(struct net_device *dev) } static struct net_device *ipoib_add_port(const char *format, - struct ib_device *hca, u8 port) + struct ib_device *hca, u32 port) { struct rtnl_link_ops *ops = ipoib_get_link_ops(); struct rdma_netdev_alloc_params params; @@ -2469,6 +2492,8 @@ static struct net_device *ipoib_add_port(const char *format, /* call event handler to ensure pkey in sync */ queue_work(ipoib_workqueue, &priv->flush_heavy); + ndev->rtnl_link_ops = ipoib_get_link_ops(); + result = register_netdev(ndev); if (result) { pr_warn("%s: couldn't register ipoib port %d; error %d\n", @@ -2518,7 +2543,7 @@ sysfs_failed: return ERR_PTR(-ENOMEM); } -static void ipoib_add_one(struct ib_device *device) +static int ipoib_add_one(struct ib_device *device) { struct list_head *dev_list; struct net_device *dev; @@ -2528,7 +2553,7 @@ static void ipoib_add_one(struct ib_device *device) dev_list = kmalloc(sizeof(*dev_list), GFP_KERNEL); if (!dev_list) - return; + return -ENOMEM; INIT_LIST_HEAD(dev_list); @@ -2545,10 +2570,11 @@ static void ipoib_add_one(struct ib_device *device) if (!count) { kfree(dev_list); - return; + return -EOPNOTSUPP; } ib_set_client_data(device, &ipoib_client, dev_list); + return 0; } static void ipoib_remove_one(struct ib_device *device, void *client_data) @@ -2556,9 +2582,6 @@ static void ipoib_remove_one(struct ib_device *device, void *client_data) struct ipoib_dev_priv *priv, *tmp, *cpriv, *tcpriv; struct list_head *dev_list = client_data; - if (!dev_list) - return; - list_for_each_entry_safe(priv, tmp, dev_list, list) { LIST_HEAD(head); ipoib_parent_unregister_pre(priv->dev); diff --git a/drivers/infiniband/ulp/ipoib/ipoib_multicast.c b/drivers/infiniband/ulp/ipoib/ipoib_multicast.c index b9e9562f5034..5b3154503bf4 100644 --- a/drivers/infiniband/ulp/ipoib/ipoib_multicast.c +++ b/drivers/infiniband/ulp/ipoib/ipoib_multicast.c @@ -135,12 +135,11 @@ static void ipoib_mcast_free(struct ipoib_mcast *mcast) kfree(mcast); } -static struct ipoib_mcast *ipoib_mcast_alloc(struct net_device *dev, - int can_sleep) +static struct ipoib_mcast *ipoib_mcast_alloc(struct net_device *dev) { struct ipoib_mcast *mcast; - mcast = kzalloc(sizeof(*mcast), can_sleep ? GFP_KERNEL : GFP_ATOMIC); + mcast = kzalloc(sizeof(*mcast), GFP_ATOMIC); if (!mcast) return NULL; @@ -218,6 +217,7 @@ static int ipoib_mcast_join_finish(struct ipoib_mcast *mcast, struct rdma_ah_attr av; int ret; int set_qkey = 0; + int mtu; mcast->mcmember = *mcmember; @@ -240,13 +240,12 @@ static int ipoib_mcast_join_finish(struct ipoib_mcast *mcast, priv->broadcast->mcmember.flow_label = mcmember->flow_label; priv->broadcast->mcmember.hop_limit = mcmember->hop_limit; /* assume if the admin and the mcast are the same both can be changed */ + mtu = rdma_mtu_enum_to_int(priv->ca, priv->port, + priv->broadcast->mcmember.mtu); if (priv->mcast_mtu == priv->admin_mtu) - priv->admin_mtu = - priv->mcast_mtu = - IPOIB_UD_MTU(ib_mtu_enum_to_int(priv->broadcast->mcmember.mtu)); - else - priv->mcast_mtu = - IPOIB_UD_MTU(ib_mtu_enum_to_int(priv->broadcast->mcmember.mtu)); + priv->admin_mtu = IPOIB_UD_MTU(mtu); + priv->mcast_mtu = IPOIB_UD_MTU(mtu); + rn->mtu = priv->mcast_mtu; priv->qkey = be32_to_cpu(priv->broadcast->mcmember.qkey); spin_unlock_irq(&priv->lock); @@ -276,7 +275,7 @@ static int ipoib_mcast_join_finish(struct ipoib_mcast *mcast, memset(&av, 0, sizeof(av)); av.type = rdma_ah_find_type(priv->ca, priv->port); - rdma_ah_set_dlid(&av, be16_to_cpu(mcast->mcmember.mlid)), + rdma_ah_set_dlid(&av, be16_to_cpu(mcast->mcmember.mlid)); rdma_ah_set_port_num(&av, priv->port); rdma_ah_set_sl(&av, mcast->mcmember.sl); rdma_ah_set_static_rate(&av, mcast->mcmember.rate); @@ -335,15 +334,6 @@ void ipoib_mcast_carrier_on_task(struct work_struct *work) return; } /* - * Check if can send sendonly MCG's with sendonly-fullmember join state. - * It done here after the successfully join to the broadcast group, - * because the broadcast group must always be joined first and is always - * re-joined if the SM changes substantially. - */ - priv->sm_fullmember_sendonly_support = - ib_sa_sendonly_fullmem_support(&ipoib_sa_client, - priv->ca, priv->port); - /* * Take rtnl_lock to avoid racing with ipoib_stop() and * turning the carrier back on while a device is being * removed. However, ipoib_stop() will attempt to flush @@ -538,9 +528,7 @@ static int ipoib_mcast_join(struct net_device *dev, struct ipoib_mcast *mcast) * most closely emulates the behavior, from a user space * application perspective, of Ethernet multicast operation. */ - if (test_bit(IPOIB_MCAST_FLAG_SENDONLY, &mcast->flags) && - priv->sm_fullmember_sendonly_support) - /* SM supports sendonly-fullmember, otherwise fallback to full-member */ + if (test_bit(IPOIB_MCAST_FLAG_SENDONLY, &mcast->flags)) rec.join_state = SENDONLY_FULLMEMBER_JOIN; } spin_unlock_irq(&priv->lock); @@ -599,7 +587,7 @@ void ipoib_mcast_join_task(struct work_struct *work) if (!priv->broadcast) { struct ipoib_mcast *broadcast; - broadcast = ipoib_mcast_alloc(dev, 0); + broadcast = ipoib_mcast_alloc(dev); if (!broadcast) { ipoib_warn(priv, "failed to allocate broadcast group\n"); /* @@ -681,15 +669,13 @@ void ipoib_mcast_start_thread(struct net_device *dev) spin_unlock_irqrestore(&priv->lock, flags); } -int ipoib_mcast_stop_thread(struct net_device *dev) +void ipoib_mcast_stop_thread(struct net_device *dev) { struct ipoib_dev_priv *priv = ipoib_priv(dev); ipoib_dbg_mcast(priv, "stopping multicast thread\n"); cancel_delayed_work_sync(&priv->mcast_task); - - return 0; } static int ipoib_mcast_leave(struct net_device *dev, struct ipoib_mcast *mcast) @@ -782,7 +768,7 @@ void ipoib_mcast_send(struct net_device *dev, u8 *daddr, struct sk_buff *skb) ipoib_dbg_mcast(priv, "setting up send only multicast group for %pI6\n", mgid); - mcast = ipoib_mcast_alloc(dev, 0); + mcast = ipoib_mcast_alloc(dev); if (!mcast) { ipoib_warn(priv, "unable to allocate memory " "for multicast structure\n"); @@ -936,7 +922,7 @@ void ipoib_mcast_restart_task(struct work_struct *work) ipoib_dbg_mcast(priv, "adding multicast entry for mgid %pI6\n", mgid.raw); - nmcast = ipoib_mcast_alloc(dev, 0); + nmcast = ipoib_mcast_alloc(dev); if (!nmcast) { ipoib_warn(priv, "unable to allocate memory for multicast structure\n"); continue; diff --git a/drivers/infiniband/ulp/ipoib/ipoib_netlink.c b/drivers/infiniband/ulp/ipoib/ipoib_netlink.c index 38c984d16996..ea16ba5d8da6 100644 --- a/drivers/infiniband/ulp/ipoib/ipoib_netlink.c +++ b/drivers/infiniband/ulp/ipoib/ipoib_netlink.c @@ -32,7 +32,6 @@ #include <linux/netdevice.h> #include <linux/if_arp.h> /* For ARPHRD_xxx */ -#include <linux/module.h> #include <net/rtnetlink.h> #include "ipoib.h" @@ -144,6 +143,16 @@ static int ipoib_new_child_link(struct net *src_net, struct net_device *dev, return 0; } +static void ipoib_del_child_link(struct net_device *dev, struct list_head *head) +{ + struct ipoib_dev_priv *priv = ipoib_priv(dev); + + if (!priv->parent) + return; + + unregister_netdevice_queue(dev, head); +} + static size_t ipoib_get_size(const struct net_device *dev) { return nla_total_size(2) + /* IFLA_IPOIB_PKEY */ @@ -153,11 +162,13 @@ static size_t ipoib_get_size(const struct net_device *dev) static struct rtnl_link_ops ipoib_link_ops __read_mostly = { .kind = "ipoib", + .netns_refund = true, .maxtype = IFLA_IPOIB_MAX, .policy = ipoib_policy, .priv_size = sizeof(struct ipoib_dev_priv), .setup = ipoib_setup_common, .newlink = ipoib_new_child_link, + .dellink = ipoib_del_child_link, .changelink = ipoib_changelink, .get_size = ipoib_get_size, .fill_info = ipoib_fill_info, diff --git a/drivers/infiniband/ulp/ipoib/ipoib_verbs.c b/drivers/infiniband/ulp/ipoib/ipoib_verbs.c index b69304d28f06..368e5d77416d 100644 --- a/drivers/infiniband/ulp/ipoib/ipoib_verbs.c +++ b/drivers/infiniband/ulp/ipoib/ipoib_verbs.c @@ -158,6 +158,7 @@ int ipoib_transport_dev_init(struct net_device *dev, struct ib_device *ca) int ret, size, req_vec; int i; + static atomic_t counter; size = ipoib_recvq_size + 1; ret = ipoib_cm_dev_init(dev); @@ -171,8 +172,7 @@ int ipoib_transport_dev_init(struct net_device *dev, struct ib_device *ca) if (ret != -EOPNOTSUPP) return ret; - req_vec = (priv->port - 1) * 2; - + req_vec = atomic_inc_return(&counter) * 2; cq_attr.cqe = size; cq_attr.comp_vector = req_vec % priv->ca->num_comp_vectors; priv->recv_cq = ib_create_cq(priv->ca, ipoib_ib_rx_completion, NULL, @@ -197,15 +197,18 @@ int ipoib_transport_dev_init(struct net_device *dev, struct ib_device *ca) init_attr.send_cq = priv->send_cq; init_attr.recv_cq = priv->recv_cq; - if (priv->hca_caps & IB_DEVICE_UD_TSO) + if (priv->kernel_caps & IBK_UD_TSO) init_attr.create_flags |= IB_QP_CREATE_IPOIB_UD_LSO; - if (priv->hca_caps & IB_DEVICE_BLOCK_MULTICAST_LOOPBACK) + if (priv->kernel_caps & IBK_BLOCK_MULTICAST_LOOPBACK) init_attr.create_flags |= IB_QP_CREATE_BLOCK_MULTICAST_LOOPBACK; if (priv->hca_caps & IB_DEVICE_MANAGED_FLOW_STEERING) init_attr.create_flags |= IB_QP_CREATE_NETIF_QP; + if (priv->kernel_caps & IBK_RDMA_NETDEV_OPA) + init_attr.create_flags |= IB_QP_CREATE_NETDEV_USE; + priv->qp = ib_create_qp(priv->pd, &init_attr); if (IS_ERR(priv->qp)) { pr_warn("%s: failed to create QP\n", ca->name); diff --git a/drivers/infiniband/ulp/ipoib/ipoib_vlan.c b/drivers/infiniband/ulp/ipoib/ipoib_vlan.c index 8ac8e18fbe0c..4bd161e86f8d 100644 --- a/drivers/infiniband/ulp/ipoib/ipoib_vlan.c +++ b/drivers/infiniband/ulp/ipoib/ipoib_vlan.c @@ -30,7 +30,6 @@ * SOFTWARE. */ -#include <linux/module.h> #include <linux/sched/signal.h> #include <linux/init.h> @@ -40,15 +39,15 @@ #include "ipoib.h" -static ssize_t show_parent(struct device *d, struct device_attribute *attr, +static ssize_t parent_show(struct device *d, struct device_attribute *attr, char *buf) { struct net_device *dev = to_net_dev(d); struct ipoib_dev_priv *priv = ipoib_priv(dev); - return sprintf(buf, "%s\n", priv->parent->name); + return sysfs_emit(buf, "%s\n", priv->parent->name); } -static DEVICE_ATTR(parent, S_IRUGO, show_parent, NULL); +static DEVICE_ATTR_RO(parent); static bool is_child_unique(struct ipoib_dev_priv *ppriv, struct ipoib_dev_priv *priv) @@ -97,6 +96,7 @@ int __ipoib_vlan_add(struct ipoib_dev_priv *ppriv, struct ipoib_dev_priv *priv, { struct net_device *ndev = priv->dev; int result; + struct rdma_netdev *rn = netdev_priv(ndev); ASSERT_RTNL(); @@ -117,6 +117,8 @@ int __ipoib_vlan_add(struct ipoib_dev_priv *ppriv, struct ipoib_dev_priv *priv, goto out_early; } + rn->mtu = priv->mcast_mtu; + priv->parent = ppriv->dev; priv->pkey = pkey; priv->child_type = type; @@ -192,6 +194,8 @@ int ipoib_vlan_add(struct net_device *pdev, unsigned short pkey) } priv = ipoib_priv(ndev); + ndev->rtnl_link_ops = ipoib_get_link_ops(); + result = __ipoib_vlan_add(ppriv, priv, pkey, IPOIB_LEGACY_CHILD); if (result && ndev->reg_state == NETREG_UNINITIALIZED) |