ipoib_dbg(priv, "bringing up interface\n");
- napi_enable(&priv->napi);
set_bit(IPOIB_FLAG_ADMIN_UP, &priv->flags);
if (ipoib_pkey_dev_delay_open(dev))
return 0;
- if (ipoib_ib_dev_open(dev)) {
- napi_disable(&priv->napi);
- return -EINVAL;
- }
+ if (ipoib_ib_dev_open(dev))
+ goto err_disable;
- if (ipoib_ib_dev_up(dev)) {
- ipoib_ib_dev_stop(dev, 1);
- napi_disable(&priv->napi);
- return -EINVAL;
- }
+ if (ipoib_ib_dev_up(dev))
+ goto err_stop;
if (!test_bit(IPOIB_FLAG_SUBINTERFACE, &priv->flags)) {
struct ipoib_dev_priv *cpriv;
netif_start_queue(dev);
return 0;
+
+err_stop:
+ ipoib_ib_dev_stop(dev, 1);
+
+err_disable:
+ clear_bit(IPOIB_FLAG_ADMIN_UP, &priv->flags);
+
+ return -EINVAL;
}
static int ipoib_stop(struct net_device *dev)
ipoib_dbg(priv, "stopping interface\n");
clear_bit(IPOIB_FLAG_ADMIN_UP, &priv->flags);
- napi_disable(&priv->napi);
netif_stop_queue(dev);
- /*
- * Now flush workqueue to make sure a scheduled task doesn't
- * bring our internal state back up.
- */
- flush_workqueue(ipoib_workqueue);
-
- ipoib_ib_dev_down(dev, 1);
- ipoib_ib_dev_stop(dev, 1);
+ ipoib_ib_dev_down(dev, 0);
+ ipoib_ib_dev_stop(dev, 0);
if (!test_bit(IPOIB_FLAG_SUBINTERFACE, &priv->flags)) {
struct ipoib_dev_priv *cpriv;
spin_lock_irq(&priv->lock);
list_for_each_entry_safe(path, tp, &priv->path_list, list) {
- ipoib_dbg(priv, "mark path LID 0x%04x GID " IPOIB_GID_FMT " invalid\n",
+ ipoib_dbg(priv, "mark path LID 0x%04x GID %pI6 invalid\n",
be16_to_cpu(path->pathrec.dlid),
- IPOIB_GID_ARG(path->pathrec.dgid));
+ path->pathrec.dgid.raw);
path->valid = 0;
}
struct ipoib_dev_priv *priv = netdev_priv(dev);
struct ipoib_path *path, *tp;
LIST_HEAD(remove_list);
+ unsigned long flags;
- spin_lock_irq(&priv->tx_lock);
- spin_lock(&priv->lock);
+ netif_tx_lock_bh(dev);
+ spin_lock_irqsave(&priv->lock, flags);
list_splice_init(&priv->path_list, &remove_list);
list_for_each_entry_safe(path, tp, &remove_list, list) {
if (path->query)
ib_sa_cancel_query(path->query_id, path->query);
- spin_unlock(&priv->lock);
- spin_unlock_irq(&priv->tx_lock);
+ spin_unlock_irqrestore(&priv->lock, flags);
+ netif_tx_unlock_bh(dev);
wait_for_completion(&path->done);
path_free(dev, path);
- spin_lock_irq(&priv->tx_lock);
- spin_lock(&priv->lock);
+ netif_tx_lock_bh(dev);
+ spin_lock_irqsave(&priv->lock, flags);
}
- spin_unlock(&priv->lock);
- spin_unlock_irq(&priv->tx_lock);
+
+ spin_unlock_irqrestore(&priv->lock, flags);
+ netif_tx_unlock_bh(dev);
}
static void path_rec_completion(int status,
struct net_device *dev = path->dev;
struct ipoib_dev_priv *priv = netdev_priv(dev);
struct ipoib_ah *ah = NULL;
- struct ipoib_ah *old_ah;
+ struct ipoib_ah *old_ah = NULL;
struct ipoib_neigh *neigh, *tn;
struct sk_buff_head skqueue;
struct sk_buff *skb;
unsigned long flags;
if (!status)
- ipoib_dbg(priv, "PathRec LID 0x%04x for GID " IPOIB_GID_FMT "\n",
- be16_to_cpu(pathrec->dlid), IPOIB_GID_ARG(pathrec->dgid));
+ ipoib_dbg(priv, "PathRec LID 0x%04x for GID %pI6\n",
+ be16_to_cpu(pathrec->dlid), pathrec->dgid.raw);
else
- ipoib_dbg(priv, "PathRec status %d for GID " IPOIB_GID_FMT "\n",
- status, IPOIB_GID_ARG(path->pathrec.dgid));
+ ipoib_dbg(priv, "PathRec status %d for GID %pI6\n",
+ status, path->pathrec.dgid.raw);
skb_queue_head_init(&skqueue);
spin_lock_irqsave(&priv->lock, flags);
- old_ah = path->ah;
- path->ah = ah;
-
if (ah) {
path->pathrec = *pathrec;
+ old_ah = path->ah;
+ path->ah = ah;
+
ipoib_dbg(priv, "created address handle %p for LID 0x%04x, SL %d\n",
ah, be16_to_cpu(pathrec->dlid), pathrec->sl);
{
struct ipoib_dev_priv *priv = netdev_priv(dev);
- ipoib_dbg(priv, "Start path record lookup for " IPOIB_GID_FMT "\n",
- IPOIB_GID_ARG(path->pathrec.dgid));
+ ipoib_dbg(priv, "Start path record lookup for %pI6\n",
+ path->pathrec.dgid.raw);
init_completion(&path->done);
path_rec_completion,
path, &path->query);
if (path->query_id < 0) {
- ipoib_warn(priv, "ib_sa_path_rec_get failed\n");
+ ipoib_warn(priv, "ib_sa_path_rec_get failed: %d\n", path->query_id);
path->query = NULL;
+ complete(&path->done);
return path->query_id;
}
struct ipoib_dev_priv *priv = netdev_priv(dev);
struct ipoib_path *path;
struct ipoib_neigh *neigh;
+ unsigned long flags;
- neigh = ipoib_neigh_alloc(skb->dst->neighbour, skb->dev);
+ neigh = ipoib_neigh_alloc(skb_dst(skb)->neighbour, skb->dev);
if (!neigh) {
++dev->stats.tx_dropped;
dev_kfree_skb_any(skb);
return;
}
- /*
- * We can only be called from ipoib_start_xmit, so we're
- * inside tx_lock -- no need to save/restore flags.
- */
- spin_lock(&priv->lock);
+ spin_lock_irqsave(&priv->lock, flags);
- path = __path_find(dev, skb->dst->neighbour->ha + 4);
+ path = __path_find(dev, skb_dst(skb)->neighbour->ha + 4);
if (!path) {
- path = path_rec_create(dev, skb->dst->neighbour->ha + 4);
+ path = path_rec_create(dev, skb_dst(skb)->neighbour->ha + 4);
if (!path)
goto err_path;
goto err_drop;
}
} else
- ipoib_send(dev, skb, path->ah, IPOIB_QPN(skb->dst->neighbour->ha));
+ ipoib_send(dev, skb, path->ah, IPOIB_QPN(skb_dst(skb)->neighbour->ha));
} else {
neigh->ah = NULL;
__skb_queue_tail(&neigh->queue, skb);
}
- spin_unlock(&priv->lock);
+ spin_unlock_irqrestore(&priv->lock, flags);
return;
err_list:
++dev->stats.tx_dropped;
dev_kfree_skb_any(skb);
- spin_unlock(&priv->lock);
+ spin_unlock_irqrestore(&priv->lock, flags);
}
static void ipoib_path_lookup(struct sk_buff *skb, struct net_device *dev)
struct ipoib_dev_priv *priv = netdev_priv(skb->dev);
/* Look up path record for unicasts */
- if (skb->dst->neighbour->ha[4] != 0xff) {
+ if (skb_dst(skb)->neighbour->ha[4] != 0xff) {
neigh_add_path(skb, dev);
return;
}
/* Add in the P_Key for multicasts */
- skb->dst->neighbour->ha[8] = (priv->pkey >> 8) & 0xff;
- skb->dst->neighbour->ha[9] = priv->pkey & 0xff;
- ipoib_mcast_send(dev, skb->dst->neighbour->ha + 4, skb);
+ skb_dst(skb)->neighbour->ha[8] = (priv->pkey >> 8) & 0xff;
+ skb_dst(skb)->neighbour->ha[9] = priv->pkey & 0xff;
+ ipoib_mcast_send(dev, skb_dst(skb)->neighbour->ha + 4, skb);
}
static void unicast_arp_send(struct sk_buff *skb, struct net_device *dev,
{
struct ipoib_dev_priv *priv = netdev_priv(dev);
struct ipoib_path *path;
+ unsigned long flags;
- /*
- * We can only be called from ipoib_start_xmit, so we're
- * inside tx_lock -- no need to save/restore flags.
- */
- spin_lock(&priv->lock);
+ spin_lock_irqsave(&priv->lock, flags);
path = __path_find(dev, phdr->hwaddr + 4);
if (!path || !path->valid) {
- if (!path)
+ int new_path = 0;
+
+ if (!path) {
path = path_rec_create(dev, phdr->hwaddr + 4);
+ new_path = 1;
+ }
if (path) {
/* put pseudoheader back on for next time */
skb_push(skb, sizeof *phdr);
__skb_queue_tail(&path->queue, skb);
- if (path_rec_start(dev, path)) {
- spin_unlock(&priv->lock);
- path_free(dev, path);
+ if (!path->query && path_rec_start(dev, path)) {
+ spin_unlock_irqrestore(&priv->lock, flags);
+ if (new_path)
+ path_free(dev, path);
return;
} else
__path_add(dev, path);
dev_kfree_skb_any(skb);
}
- spin_unlock(&priv->lock);
+ spin_unlock_irqrestore(&priv->lock, flags);
return;
}
dev_kfree_skb_any(skb);
}
- spin_unlock(&priv->lock);
+ spin_unlock_irqrestore(&priv->lock, flags);
}
static int ipoib_start_xmit(struct sk_buff *skb, struct net_device *dev)
struct ipoib_neigh *neigh;
unsigned long flags;
- if (unlikely(!spin_trylock_irqsave(&priv->tx_lock, flags)))
- return NETDEV_TX_LOCKED;
-
- if (likely(skb->dst && skb->dst->neighbour)) {
- if (unlikely(!*to_ipoib_neigh(skb->dst->neighbour))) {
+ if (likely(skb_dst(skb) && skb_dst(skb)->neighbour)) {
+ if (unlikely(!*to_ipoib_neigh(skb_dst(skb)->neighbour))) {
ipoib_path_lookup(skb, dev);
- goto out;
+ return NETDEV_TX_OK;
}
- neigh = *to_ipoib_neigh(skb->dst->neighbour);
-
- if (neigh->ah)
- if (unlikely((memcmp(&neigh->dgid.raw,
- skb->dst->neighbour->ha + 4,
- sizeof(union ib_gid))) ||
- (neigh->dev != dev))) {
- spin_lock(&priv->lock);
- /*
- * It's safe to call ipoib_put_ah() inside
- * priv->lock here, because we know that
- * path->ah will always hold one more reference,
- * so ipoib_put_ah() will never do more than
- * decrement the ref count.
- */
+ neigh = *to_ipoib_neigh(skb_dst(skb)->neighbour);
+
+ if (unlikely((memcmp(&neigh->dgid.raw,
+ skb_dst(skb)->neighbour->ha + 4,
+ sizeof(union ib_gid))) ||
+ (neigh->dev != dev))) {
+ spin_lock_irqsave(&priv->lock, flags);
+ /*
+ * It's safe to call ipoib_put_ah() inside
+ * priv->lock here, because we know that
+ * path->ah will always hold one more reference,
+ * so ipoib_put_ah() will never do more than
+ * decrement the ref count.
+ */
+ if (neigh->ah)
ipoib_put_ah(neigh->ah);
- list_del(&neigh->list);
- ipoib_neigh_free(dev, neigh);
- spin_unlock(&priv->lock);
- ipoib_path_lookup(skb, dev);
- goto out;
- }
+ list_del(&neigh->list);
+ ipoib_neigh_free(dev, neigh);
+ spin_unlock_irqrestore(&priv->lock, flags);
+ ipoib_path_lookup(skb, dev);
+ return NETDEV_TX_OK;
+ }
if (ipoib_cm_get(neigh)) {
if (ipoib_cm_up(neigh)) {
ipoib_cm_send(dev, skb, ipoib_cm_get(neigh));
- goto out;
+ return NETDEV_TX_OK;
}
} else if (neigh->ah) {
- ipoib_send(dev, skb, neigh->ah, IPOIB_QPN(skb->dst->neighbour->ha));
- goto out;
+ ipoib_send(dev, skb, neigh->ah, IPOIB_QPN(skb_dst(skb)->neighbour->ha));
+ return NETDEV_TX_OK;
}
if (skb_queue_len(&neigh->queue) < IPOIB_MAX_PATH_REC_QUEUE) {
- spin_lock(&priv->lock);
+ spin_lock_irqsave(&priv->lock, flags);
__skb_queue_tail(&neigh->queue, skb);
- spin_unlock(&priv->lock);
+ spin_unlock_irqrestore(&priv->lock, flags);
} else {
++dev->stats.tx_dropped;
dev_kfree_skb_any(skb);
if ((be16_to_cpup((__be16 *) skb->data) != ETH_P_ARP) &&
(be16_to_cpup((__be16 *) skb->data) != ETH_P_RARP)) {
- ipoib_warn(priv, "Unicast, no %s: type %04x, QPN %06x "
- IPOIB_GID_FMT "\n",
- skb->dst ? "neigh" : "dst",
+ ipoib_warn(priv, "Unicast, no %s: type %04x, QPN %06x %pI6\n",
+ skb_dst(skb) ? "neigh" : "dst",
be16_to_cpup((__be16 *) skb->data),
IPOIB_QPN(phdr->hwaddr),
- IPOIB_GID_RAW_ARG(phdr->hwaddr + 4));
+ phdr->hwaddr + 4);
dev_kfree_skb_any(skb);
++dev->stats.tx_dropped;
- goto out;
+ return NETDEV_TX_OK;
}
unicast_arp_send(skb, dev, phdr);
}
}
-out:
- spin_unlock_irqrestore(&priv->tx_lock, flags);
-
return NETDEV_TX_OK;
}
* destination address onto the front of the skb so we can
* figure out where to send the packet later.
*/
- if ((!skb->dst || !skb->dst->neighbour) && daddr) {
+ if ((!skb_dst(skb) || !skb_dst(skb)->neighbour) && daddr) {
struct ipoib_pseudoheader *phdr =
(struct ipoib_pseudoheader *) skb_push(skb, sizeof *phdr);
memcpy(phdr->hwaddr, daddr, INFINIBAND_ALEN);
else
return;
ipoib_dbg(priv,
- "neigh_cleanup for %06x " IPOIB_GID_FMT "\n",
+ "neigh_cleanup for %06x %pI6\n",
IPOIB_QPN(n->ha),
- IPOIB_GID_RAW_ARG(n->ha + 4));
+ n->ha + 4);
spin_lock_irqsave(&priv->lock, flags);
priv->lro.lro_mgr.ip_summed_aggr = CHECKSUM_UNNECESSARY;
}
+static const struct net_device_ops ipoib_netdev_ops = {
+ .ndo_open = ipoib_open,
+ .ndo_stop = ipoib_stop,
+ .ndo_change_mtu = ipoib_change_mtu,
+ .ndo_start_xmit = ipoib_start_xmit,
+ .ndo_tx_timeout = ipoib_timeout,
+ .ndo_set_multicast_list = ipoib_set_mcast_list,
+ .ndo_neigh_setup = ipoib_neigh_setup_dev,
+};
+
static void ipoib_setup(struct net_device *dev)
{
struct ipoib_dev_priv *priv = netdev_priv(dev);
- dev->open = ipoib_open;
- dev->stop = ipoib_stop;
- dev->change_mtu = ipoib_change_mtu;
- dev->hard_start_xmit = ipoib_start_xmit;
- dev->tx_timeout = ipoib_timeout;
+ dev->netdev_ops = &ipoib_netdev_ops;
dev->header_ops = &ipoib_header_ops;
- dev->set_multicast_list = ipoib_set_mcast_list;
- dev->neigh_setup = ipoib_neigh_setup_dev;
ipoib_set_ethtool_ops(dev);
dev->type = ARPHRD_INFINIBAND;
dev->tx_queue_len = ipoib_sendq_size * 2;
dev->features = (NETIF_F_VLAN_CHALLENGED |
- NETIF_F_LLTX |
NETIF_F_HIGHDMA);
+ dev->priv_flags &= ~IFF_XMIT_DST_RELEASE;
memcpy(dev->broadcast, ipv4_bcast_addr, INFINIBAND_ALEN);
ipoib_lro_setup(priv);
spin_lock_init(&priv->lock);
- spin_lock_init(&priv->tx_lock);
- mutex_init(&priv->mcast_mutex);
mutex_init(&priv->vlan_mutex);
INIT_LIST_HEAD(&priv->path_list);
INIT_DELAYED_WORK(&priv->pkey_poll_task, ipoib_pkey_poll);
INIT_DELAYED_WORK(&priv->mcast_task, ipoib_mcast_join_task);
+ INIT_WORK(&priv->carrier_on_task, ipoib_mcast_carrier_on_task);
INIT_WORK(&priv->flush_light, ipoib_ib_dev_flush_light);
INIT_WORK(&priv->flush_normal, ipoib_ib_dev_flush_normal);
INIT_WORK(&priv->flush_heavy, ipoib_ib_dev_flush_heavy);
return device_create_file(&dev->dev, &dev_attr_pkey);
}
+int ipoib_set_dev_features(struct ipoib_dev_priv *priv, struct ib_device *hca)
+{
+ struct ib_device_attr *device_attr;
+ int result = -ENOMEM;
+
+ device_attr = kmalloc(sizeof *device_attr, GFP_KERNEL);
+ if (!device_attr) {
+ printk(KERN_WARNING "%s: allocation of %zu bytes failed\n",
+ hca->name, sizeof *device_attr);
+ return result;
+ }
+
+ result = ib_query_device(hca, device_attr);
+ if (result) {
+ printk(KERN_WARNING "%s: ib_query_device failed (ret = %d)\n",
+ hca->name, result);
+ kfree(device_attr);
+ return result;
+ }
+ priv->hca_caps = device_attr->device_cap_flags;
+
+ kfree(device_attr);
+
+ if (priv->hca_caps & IB_DEVICE_UD_IP_CSUM) {
+ set_bit(IPOIB_FLAG_CSUM, &priv->flags);
+ priv->dev->features |= NETIF_F_SG | NETIF_F_IP_CSUM;
+ }
+
+ if (lro)
+ priv->dev->features |= NETIF_F_LRO;
+
+ if (priv->dev->features & NETIF_F_SG && priv->hca_caps & IB_DEVICE_UD_TSO)
+ priv->dev->features |= NETIF_F_TSO;
+
+ return 0;
+}
+
+
static struct net_device *ipoib_add_port(const char *format,
struct ib_device *hca, u8 port)
{
struct ipoib_dev_priv *priv;
- struct ib_device_attr *device_attr;
struct ib_port_attr attr;
int result = -ENOMEM;
goto device_init_failed;
}
- device_attr = kmalloc(sizeof *device_attr, GFP_KERNEL);
- if (!device_attr) {
- printk(KERN_WARNING "%s: allocation of %zu bytes failed\n",
- hca->name, sizeof *device_attr);
- goto device_init_failed;
- }
-
- result = ib_query_device(hca, device_attr);
- if (result) {
- printk(KERN_WARNING "%s: ib_query_device failed (ret = %d)\n",
- hca->name, result);
- kfree(device_attr);
+ if (ipoib_set_dev_features(priv, hca))
goto device_init_failed;
- }
- priv->hca_caps = device_attr->device_cap_flags;
-
- kfree(device_attr);
-
- if (priv->hca_caps & IB_DEVICE_UD_IP_CSUM) {
- set_bit(IPOIB_FLAG_CSUM, &priv->flags);
- priv->dev->features |= NETIF_F_SG | NETIF_F_IP_CSUM;
- }
-
- if (lro)
- priv->dev->features |= NETIF_F_LRO;
/*
* Set the full membership bit, so that we join the right
goto event_failed;
}
- if (priv->dev->features & NETIF_F_SG && priv->hca_caps & IB_DEVICE_UD_TSO)
- priv->dev->features |= NETIF_F_TSO;
-
result = register_netdev(priv->dev);
if (result) {
printk(KERN_WARNING "%s: couldn't register ipoib port %d; error %d\n",
register_failed:
ib_unregister_event_handler(&priv->event_handler);
- flush_scheduled_work();
+ flush_workqueue(ipoib_workqueue);
event_failed:
ipoib_dev_cleanup(priv->dev);
list_for_each_entry_safe(priv, tmp, dev_list, list) {
ib_unregister_event_handler(&priv->event_handler);
- flush_scheduled_work();
+
+ rtnl_lock();
+ dev_change_flags(priv->dev, priv->dev->flags & ~IFF_UP);
+ rtnl_unlock();
+
+ flush_workqueue(ipoib_workqueue);
unregister_netdev(priv->dev);
ipoib_dev_cleanup(priv->dev);