From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: From: Greg Kroah-Hartman To: linux-kernel@vger.kernel.org Cc: Greg Kroah-Hartman , stable@vger.kernel.org, Eric Dumazet , Alexander Duyck , "David S. Miller" Subject: [ 013/135] net: fix multiqueue selection Date: Fri, 11 Oct 2013 12:38:14 -0700 Message-Id: <20131011193946.684090274@linuxfoundation.org> In-Reply-To: <20131011193945.181603934@linuxfoundation.org> References: <20131011193945.181603934@linuxfoundation.org> Sender: linux-kernel-owner@vger.kernel.org List-ID: 3.11-stable review patch. If anyone has any objections, please let me know. ------------------ From: Eric Dumazet [ Upstream commit 50d1784ee4683f073c0362ee360bfae7a3333d6c ] commit 416186fbf8c5b4e4465 ("net: Split core bits of netdev_pick_tx into __netdev_pick_tx") added a bug that disables caching of queue index in the socket. This is the source of packet reorders for TCP flows, and again this is happening more often when using FQ pacing. Old code was doing if (queue_index != old_index) sk_tx_queue_set(sk, queue_index); Alexander renamed the variables but forgot to change sk_tx_queue_set() 2nd parameter. if (queue_index != new_index) sk_tx_queue_set(sk, queue_index); This means we store -1 over and over in sk->sk_tx_queue_mapping Signed-off-by: Eric Dumazet Cc: Alexander Duyck Acked-by: Alexander Duyck Signed-off-by: David S. Miller Signed-off-by: Greg Kroah-Hartman --- net/core/flow_dissector.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) --- a/net/core/flow_dissector.c +++ b/net/core/flow_dissector.c @@ -348,7 +348,7 @@ u16 __netdev_pick_tx(struct net_device * if (queue_index != new_index && sk && rcu_access_pointer(sk->sk_dst_cache)) - sk_tx_queue_set(sk, queue_index); + sk_tx_queue_set(sk, new_index); queue_index = new_index; }