IB/ipoib: Fix packet loss after hardware address update

The neighbour ha field may get updated without destroying the
neighbour.  In this case, the ha field gets out of sync with the
address handle stored in ipoib_neigh->ah, with the result that
the ah field would point to an incorrect path, resulting in all
packets being lost.

Signed-off-by: Michael S. Tsirkin <mst@mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>

diff --git a/drivers/infiniband/ulp/ipoib/ipoib.h b/drivers/infiniband/ulp/ipoib/ipoib.h
index 3f89f5e190367ce6d82f6930d2372bd94bb8955c..474aa214ab57ed923b16b4376979ca5b63d28249 100644 (file)
--- a/drivers/infiniband/ulp/ipoib/ipoib.h
+++ b/drivers/infiniband/ulp/ipoib/ipoib.h
@@ -212,6 +212,7 @@ struct ipoib_path {
 
 struct ipoib_neigh {
        struct ipoib_ah    *ah;
+       union ib_gid        dgid;
        struct sk_buff_head queue;
 
        struct neighbour   *neighbour;

diff --git a/drivers/infiniband/ulp/ipoib/ipoib_main.c b/drivers/infiniband/ulp/ipoib/ipoib_main.c
index 1c6ea1c682a5dc7c4bd5bed14f67307ef27aa1a1..cf71d2a5515c1d734026827355680847a6a51893 100644 (file)
--- a/drivers/infiniband/ulp/ipoib/ipoib_main.c
+++ b/drivers/infiniband/ulp/ipoib/ipoib_main.c
@@ -404,6 +404,8 @@ static void path_rec_completion(int status,
                list_for_each_entry(neigh, &path->neigh_list, list) {
                        kref_get(&path->ah->ref);
                        neigh->ah = path->ah;
+                       memcpy(&neigh->dgid.raw, &path->pathrec.dgid.raw,
+                              sizeof(union ib_gid));
 
                        while ((skb = __skb_dequeue(&neigh->queue)))
                                __skb_queue_tail(&skqueue, skb);
@@ -510,6 +512,8 @@ static void neigh_add_path(struct sk_buff *skb, struct net_device *dev)
        if (path->ah) {
                kref_get(&path->ah->ref);
                neigh->ah = path->ah;
+               memcpy(&neigh->dgid.raw, &path->pathrec.dgid.raw,
+                      sizeof(union ib_gid));
 
                ipoib_send(dev, skb, path->ah,
                           be32_to_cpup((__be32 *) skb->dst->neighbour->ha));
@@ -633,6 +637,25 @@ static int ipoib_start_xmit(struct sk_buff *skb, struct net_device *dev)
                neigh = *to_ipoib_neigh(skb->dst->neighbour);
 
                if (likely(neigh->ah)) {
+                       if (unlikely(memcmp(&neigh->dgid.raw,
+                                           skb->dst->neighbour->ha + 4,
+                                           sizeof(union ib_gid)))) {
+                               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.
+                                */
+                               ipoib_put_ah(neigh->ah);
+                               list_del(&neigh->list);
+                               ipoib_neigh_free(neigh);
+                               spin_unlock(&priv->lock);
+                               ipoib_path_lookup(skb, dev);
+                               goto out;
+                       }
+
                        ipoib_send(dev, skb, neigh->ah,
                                   be32_to_cpup((__be32 *) skb->dst->neighbour->ha));
                        goto out;