From mboxrd@z Thu Jan 1 00:00:00 1970 From: Felix von Leitner Subject: Re: bizarre network timing problem Date: Sat, 3 Nov 2007 00:23:32 +0100 Message-ID: <20071102232332.GA5739@codeblau.de> References: <20071017205127.GA21334@codeblau.de> <47167BDE.4000103@redhat.com> <20071017220019.GA22765@codeblau.de> <471689BF.2040909@hp.com> <20071018094230.GA2978@codeblau.de> <4717964A.8080100@hp.com> <20071102221146.GA4354@codeblau.de> <472BA5BC.6000901@hp.com> <20071102223856.GA5121@codeblau.de> <472BAB90.8020306@hp.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Cc: Chuck Ebbert , linux-kernel@vger.kernel.org, Netdev To: Rick Jones Return-path: Received: from ioctl.codeblau.de ([80.190.240.67]:35709 "EHLO codeblau.de" rhost-flags-OK-OK-OK-FAIL) by vger.kernel.org with ESMTP id S1752770AbXKBXXe (ORCPT ); Fri, 2 Nov 2007 19:23:34 -0400 Content-Disposition: inline In-Reply-To: <472BAB90.8020306@hp.com> Sender: netdev-owner@vger.kernel.org List-Id: netdev.vger.kernel.org Thus spake Rick Jones (rick.jones2@hp.com): > Past performance is no guarantee of current correctness :) And over an > Ethernet, there will be a very different set of both timings and TCP > segment sizes compared to loopback. > My guess is that you will find setting the lo mtu to 1500 a very > interesting experiment. Setting the MTU on lo to 1500 eliminates the problem and gives me double digit MB/sec throughput. Felix