From mboxrd@z Thu Jan 1 00:00:00 1970 From: David Miller Subject: Re: [PATCH net-next] bnx2x: reduce skb truesize by 50% Date: Mon, 14 Nov 2011 14:21:17 -0500 (EST) Message-ID: <20111114.142117.1578594312371186159.davem@davemloft.net> References: <20111114.000811.780218767629149648.davem@davemloft.net> <1321251945.17837.55.camel@edumazet-laptop> <1321286265.2272.46.camel@edumazet-HP-Compaq-6005-Pro-SFF-PC> Mime-Version: 1.0 Content-Type: Text/Plain; charset=iso-8859-1 Content-Transfer-Encoding: QUOTED-PRINTABLE Cc: eilong@broadcom.com, bhutchings@solarflare.com, pstaszewski@itcare.pl, netdev@vger.kernel.org, tgraf@infradead.org, therbert@google.com, hadi@mojatatu.com, shemminger@vyatta.com To: eric.dumazet@gmail.com Return-path: Received: from shards.monkeyblade.net ([198.137.202.13]:44255 "EHLO shards.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753733Ab1KNTWk convert rfc822-to-8bit (ORCPT ); Mon, 14 Nov 2011 14:22:40 -0500 In-Reply-To: <1321286265.2272.46.camel@edumazet-HP-Compaq-6005-Pro-SFF-PC> Sender: netdev-owner@vger.kernel.org List-ID: =46rom: Eric Dumazet Date: Mon, 14 Nov 2011 16:57:45 +0100 > Le lundi 14 novembre 2011 =E0 07:25 +0100, Eric Dumazet a =E9crit : >> Le lundi 14 novembre 2011 =E0 00:08 -0500, David Miller a =E9crit : >>=20 >> > I fully support bringing this thing back to life :-) >>=20 >> I'll make extensive tests today and provide two patches when ready, = with >> all performance results. >>=20 >> Some prefetch() calls will be removed, since build_skb() provides >> already cache hot skb. >=20 > Impressive results : >=20 > before : 720.000 pps > after : 820.000 pps >=20 > [ One mono threaded application receiving UDP messages on a single > socket, asking IP_PKTINFO ancillary info ] Sweeeeeet.