net: revert 8728c54 ("net: dev_pick_tx() fix")

commit 8728c54 ("net: dev_pick_tx() fix") and commit b6fe83e ("bonding: refine IFF_XMIT_DST_RELEASE capability") are quite incompatible : Queue selection is disabled because skb dst was dropped before entering bonding device. This causes major performance regression, mainly because TCP packets for a given flow can be sent to multiple queues. This is particularly visible when using the new FQ packet scheduler with MQ + FQ setup on the slaves. We can safely revert the first commit now that 416186f ("net: Split core bits of netdev_pick_tx into __netdev_pick_tx") properly caps the queue_index. Reported-by: Xi Wang <[email protected]> Diagnosed-by: Xi Wang <[email protected]> Signed-off-by: Eric Dumazet <[email protected]> Cc: Tom Herbert <[email protected]> Cc: Alexander Duyck <[email protected]> Cc: Denys Fedorysychenko <[email protected]> Signed-off-by: David S. Miller <[email protected]>
cluon · Aug 30, 2013 · 702821f · 702821f
1 parent 25ad611
commit 702821f
Showing 1 changed file with 3 additions and 8 deletions.
diff --git a/net/core/flow_dissector.c b/net/core/flow_dissector.c
@@ -346,14 +346,9 @@ u16 __netdev_pick_tx(struct net_device *dev, struct sk_buff *skb)
 		if (new_index < 0)
 			new_index = skb_tx_hash(dev, skb);
 
-		if (queue_index != new_index && sk) {
-			struct dst_entry *dst =
-				    rcu_dereference_check(sk->sk_dst_cache, 1);
-
-			if (dst && skb_dst(skb) == dst)
-				sk_tx_queue_set(sk, queue_index);
-
-		}
+		if (queue_index != new_index && sk &&
+		    rcu_access_pointer(sk->sk_dst_cache))
+			sk_tx_queue_set(sk, queue_index);
 
 		queue_index = new_index;
 	}