net: revert 8728c544a9c ("net: dev_pick_tx() fix")
authorEric Dumazet <edumazet@google.com>
Thu, 29 Aug 2013 01:10:43 +0000 (18:10 -0700)
committerDavid S. Miller <davem@davemloft.net>
Fri, 30 Aug 2013 21:48:04 +0000 (17:48 -0400)
commit702821f4ea6f68db18aa1de7d8ed62c6ba586a64
treea0e3a4ba9577812f220224d7e6380146c193dc2e
parent25ad6117e73656071b38fd19fa67ae325471c758
net: revert 8728c544a9c ("net: dev_pick_tx() fix")

commit 8728c544a9cbdc ("net: dev_pick_tx() fix") and commit
b6fe83e9525a ("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 416186fbf8c5b
("net: Split core bits of netdev_pick_tx into __netdev_pick_tx")
properly caps the queue_index.

Reported-by: Xi Wang <xii@google.com>
Diagnosed-by: Xi Wang <xii@google.com>
Signed-off-by: Eric Dumazet <edumazet@google.com>
Cc: Tom Herbert <therbert@google.com>
Cc: Alexander Duyck <alexander.h.duyck@intel.com>
Cc: Denys Fedorysychenko <nuclearcat@nuclearcat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
net/core/flow_dissector.c