From mboxrd@z Thu Jan 1 00:00:00 1970 From: Eric Dumazet Subject: Re: TSQ accounting skb->truesize degrades throughput for large packets Date: Sat, 07 Sep 2013 10:21:34 -0700 Message-ID: <1378574494.26319.14.camel@edumazet-glaptop> References: <20130906101635.GI14104@zion.uk.xensource.com> <1378472268.31445.15.camel@edumazet-glaptop> <522A049A.7000105@citrix.com> <1378486840.31445.36.camel@edumazet-glaptop> Mime-Version: 1.0 Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: 7bit Cc: Wei Liu , Jonathan Davies , Ian Campbell , netdev@vger.kernel.org, xen-devel@lists.xenproject.org To: Zoltan Kiss Return-path: Received: from mail-pd0-f170.google.com ([209.85.192.170]:40752 "EHLO mail-pd0-f170.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751059Ab3IGRVf (ORCPT ); Sat, 7 Sep 2013 13:21:35 -0400 Received: by mail-pd0-f170.google.com with SMTP id x10so4514865pdj.29 for ; Sat, 07 Sep 2013 10:21:35 -0700 (PDT) In-Reply-To: <1378486840.31445.36.camel@edumazet-glaptop> Sender: netdev-owner@vger.kernel.org List-ID: On Fri, 2013-09-06 at 10:00 -0700, Eric Dumazet wrote: > On Fri, 2013-09-06 at 17:36 +0100, Zoltan Kiss wrote: > > > So I guess it would be good to revisit the default value of this > > setting. > > If ixgbe requires 3 TSO packets in TX ring to get line rate, you also > can tweak dev->gso_max_size from 65535 to 64000. Another idea would be to no longer use tcp_limit_output_bytes but max(sk_pacing_rate / 1000, 2*MSS) This means that number of packets in FQ would be limited to the equivalent of 1ms, so TCP could have faster response to packet losses : Retransmitted packets would not have to wait for prior packets being drained from FQ For a 8Gbps flow, 1Gbyte/s, sk_pacing_rate would be 2Gbyte, this would translate to ~2 Mbytes in Qdisc/TX ring. sk_pacing_rate was introduced in linux-3.12, but could be backported easily.