From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mail-wr0-f172.google.com ([209.85.128.172]:43390 "EHLO mail-wr0-f172.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753203AbeBTTjw (ORCPT ); Tue, 20 Feb 2018 14:39:52 -0500 Received: by mail-wr0-f172.google.com with SMTP id u49so13035888wrc.10 for ; Tue, 20 Feb 2018 11:39:51 -0800 (PST) MIME-Version: 1.0 In-Reply-To: <4964697.hB4CnsZvNY@natalenko.name> References: <20180219195652.242663-1-edumazet@google.com> <1519141172.55655.21.camel@gmail.com> <1519153062.55655.24.camel@gmail.com> <4964697.hB4CnsZvNY@natalenko.name> From: Eric Dumazet Date: Tue, 20 Feb 2018 11:39:49 -0800 Message-ID: Subject: Re: [PATCH net-next 0/6] tcp: remove non GSO code To: Oleksandr Natalenko Cc: Eric Dumazet , "David S . Miller" , netdev , Neal Cardwell , Yuchung Cheng , Soheil Hassas Yeganeh Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: quoted-printable Sender: netdev-owner@vger.kernel.org List-ID: On Tue, Feb 20, 2018 at 11:35 AM, Oleksandr Natalenko wrote: > Hi. > > On =C3=BAter=C3=BD 20. =C3=BAnora 2018 19:57:42 CET Eric Dumazet wrote: >> Actually timer drifts are not horrible (at least on my lab hosts) >> >> But BBR has a pessimistic way to sense the burst size, as it is tied to >> TSO/GSO being there. >> >> Following patch helps a lot. > > Not really, at least if applied to v4.15.4. Still getting 2 Gbps less bet= ween > VMs if using BBR instead of Reno. > > Am I doing something wrong? I am not trying to compare BBR and Reno on a lossless link. Reno is running as fast as possible and will win when bufferbloat is not an issue. If bufferbloat is not an issue, simply use Reno and be happy ;) My patch helps BBR only, I thought it was obvious ;)