From mboxrd@z Thu Jan 1 00:00:00 1970 From: "Jerry Chu" Subject: Re: Socket buffer sizes with autotuning Date: Mon, 28 Apr 2008 19:20:47 -0700 Message-ID: References: <1e41a3230804240932u510609beh8fb577baaadeb9bd@mail.gmail.com> <20080424.234628.170849475.davem@davemloft.net> <1e41a3230804281221s11442a82j55c363116988a5cb@mail.gmail.com> <1e41a3230804281628r363f9d3cn7b0eee0bd6423e65@mail.gmail.com> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Cc: "David Miller" , netdev@vger.kernel.org, rick.jones2@hp.com To: "John Heffner" Return-path: Received: from smtp-out.google.com ([216.239.33.17]:35369 "EHLO smtp-out.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752484AbYD2CUw (ORCPT ); Mon, 28 Apr 2008 22:20:52 -0400 Received: from zps37.corp.google.com (zps37.corp.google.com [172.25.146.37]) by smtp-out.google.com with ESMTP id m3T2KnJ2008993 for ; Tue, 29 Apr 2008 03:20:49 +0100 Received: from hs-out-0708.google.com (hsc55.prod.google.com [10.44.3.55]) by zps37.corp.google.com with ESMTP id m3T2KmI7004868 for ; Mon, 28 Apr 2008 19:20:48 -0700 Received: by hs-out-0708.google.com with SMTP id 55so3999802hsc.10 for ; Mon, 28 Apr 2008 19:20:48 -0700 (PDT) In-Reply-To: <1e41a3230804281628r363f9d3cn7b0eee0bd6423e65@mail.gmail.com> Content-Disposition: inline Sender: netdev-owner@vger.kernel.org List-ID: Just for the record, in my netperf RR (1MB request/20B reply) over 1GE/tg3 test on top of 2.6.18, changing tcp_tso_win_divisor to 0 from the default of 3 gets the cwnd down to 695 from 1037. Applying the bound check (left * tp->mss_cache < 65536) will get cwnd down to 737, not 695. Still big improvement though. (Anyway this may be moot after I fix the real culprit.) Jerry On Mon, Apr 28, 2008 at 4:28 PM, John Heffner wrote: > > On Mon, Apr 28, 2008 at 1:38 PM, Jerry Chu wrote: > > > Try this series of patches, against net-next. > > > > Ah, you already know about this problem. Yes it does the trick. > > I had not actually known about this. (Just cooked up the patches > after I saw your mail.) This looks to me like a clear bug. After > looking carefully at the code there, it seems there was another small > problem, too, though not likely to have much effect. > > Dave, can you apply the patches I just sent? > > Thanks, > -John >