From mboxrd@z Thu Jan 1 00:00:00 1970 From: Eric Dumazet Subject: Re: 10GBE performance drop with net.ipv4.tcp_timestamps=0 Date: Wed, 20 Jun 2012 11:17:34 +0200 Message-ID: <1340183854.4604.847.camel@edumazet-glaptop> References: <4FE0EA33.1000309@profihost.ag> <1340141493.4604.773.camel@edumazet-glaptop> <4FE17522.5050005@profihost.ag> <1340176940.4604.801.camel@edumazet-glaptop> <4FE1881B.7030800@profihost.ag> <1340181676.4604.838.camel@edumazet-glaptop> <4FE192A1.6000000@profihost.ag> <4FE193FA.5030802@profihost.ag> Mime-Version: 1.0 Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: 7bit Cc: Linux Netdev List To: Stefan Priebe - Profihost AG Return-path: Received: from mail-bk0-f46.google.com ([209.85.214.46]:37439 "EHLO mail-bk0-f46.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751726Ab2FTJRj (ORCPT ); Wed, 20 Jun 2012 05:17:39 -0400 Received: by bkcji2 with SMTP id ji2so5941346bkc.19 for ; Wed, 20 Jun 2012 02:17:37 -0700 (PDT) In-Reply-To: <4FE193FA.5030802@profihost.ag> Sender: netdev-owner@vger.kernel.org List-ID: On Wed, 2012-06-20 at 11:12 +0200, Stefan Priebe - Profihost AG wrote: > Am 20.06.2012 11:06, schrieb Stefan Priebe - Profihost AG: > >> You seem to have a switch or something that drops packets in this case. > >> You could try to rate limit to 9Gb/s and see if it is better. > > Sadly i can't rate limit to 9Gbit/s on the switch. > > > >> Here, I roughly have same bandwidth with tcp_timestamps on or off, with > >> ixgbe cards and net-next kernels. > > Mhm strange. Do you have any vague idea what could cause this? Any wrong > > reordering of the packets without tcp_timestamps? > > I've now done another test without the switch. So both systems where > direct attached and still the same. Without tcp_timstamps speed drops to > 2-4Gbit/s. > > Stefan If you exchange sender/receiver role between linux kernel versions, is it the same problem ?