From mboxrd@z Thu Jan 1 00:00:00 1970 From: Eric Dumazet Subject: Re: [PATCH v5] rfs: Receive Flow Steering Date: Tue, 20 Apr 2010 17:39:22 +0200 Message-ID: <1271777962.7895.25.camel@edumazet-laptop> References: <1271452358.16881.4486.camel@edumazet-laptop> <1271520633.16881.4754.camel@edumazet-laptop> <20100419.130905.210660275.davem@davemloft.net> <20100419.132318.192086187.davem@davemloft.net> <1271709121.3845.94.camel@edumazet-laptop> <1271743164.3845.128.camel@edumazet-laptop> Mime-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: QUOTED-PRINTABLE Cc: Changli Gao , David Miller , netdev@vger.kernel.org To: Tom Herbert Return-path: Received: from mail-bw0-f219.google.com ([209.85.218.219]:64665 "EHLO mail-bw0-f219.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1750923Ab0DTPjb (ORCPT ); Tue, 20 Apr 2010 11:39:31 -0400 Received: by bwz19 with SMTP id 19so96619bwz.21 for ; Tue, 20 Apr 2010 08:39:30 -0700 (PDT) In-Reply-To: Sender: netdev-owner@vger.kernel.org List-ID: Le mardi 20 avril 2010 =C3=A0 08:04 -0700, Tom Herbert a =C3=A9crit : > Maybe for the purposes of RPS, but hash collisions could definitely b= e > an issue in RFS. If two active connections hit the same rps_flow > entry this may cause thrashing of those connections between CPUs. I > think your patch may increase the probability of this happening. >=20 Good point. I'll make a gathering of tcp tuples on a busy server over a day and try to compute number of clashes we can get with and without the addr/port swapping.