From mboxrd@z Thu Jan 1 00:00:00 1970 From: David Miller Subject: Re: [PATCH] net: revert 8728c544a9c ("net: dev_pick_tx() fix") Date: Fri, 30 Aug 2013 17:48:37 -0400 (EDT) Message-ID: <20130830.174837.1340852981940136289.davem@davemloft.net> References: <1377738643.8277.25.camel@edumazet-glaptop> Mime-Version: 1.0 Content-Type: Text/Plain; charset=us-ascii Content-Transfer-Encoding: 7bit Cc: netdev@vger.kernel.org, alexander.h.duyck@intel.com, xii@google.com, therbert@google.com, nuclearcat@nuclearcat.com To: eric.dumazet@gmail.com Return-path: Received: from shards.monkeyblade.net ([149.20.54.216]:58006 "EHLO shards.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1754615Ab3H3Vsq (ORCPT ); Fri, 30 Aug 2013 17:48:46 -0400 In-Reply-To: <1377738643.8277.25.camel@edumazet-glaptop> Sender: netdev-owner@vger.kernel.org List-ID: From: Eric Dumazet Date: Wed, 28 Aug 2013 18:10:43 -0700 > From: Eric Dumazet > > 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 > Diagnosed-by: Xi Wang > Signed-off-by: Eric Dumazet Applied, thanks Eric.