From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1757541AbXFSGwZ (ORCPT ); Tue, 19 Jun 2007 02:52:25 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1754354AbXFSGwS (ORCPT ); Tue, 19 Jun 2007 02:52:18 -0400 Received: from pentafluge.infradead.org ([213.146.154.40]:56901 "EHLO pentafluge.infradead.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753881AbXFSGwR (ORCPT ); Tue, 19 Jun 2007 02:52:17 -0400 Subject: Re: [-RT] multiple streams have degraded performance From: Peter Zijlstra To: Vernon Mauery Cc: LKML , Ingo Molnar , Jeff Garzik In-Reply-To: <200706182212.22633.vernux@us.ibm.com> References: <200706182212.22633.vernux@us.ibm.com> Content-Type: text/plain Date: Tue, 19 Jun 2007 08:51:38 +0200 Message-Id: <1182235898.7348.361.camel@twins> Mime-Version: 1.0 X-Mailer: Evolution 2.10.1 Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org On Mon, 2007-06-18 at 22:12 -0700, Vernon Mauery wrote: > In looking at the performance characteristics of my network I found that > 2.6.21.5-rt15 suffers from degraded thoughput with multiple threads. The > test that I did this with is simply invoking 1, 2, 4, and 8 instances of > netperf at a time and measuring the total throughput. I have two 4-way > machines connected with 10GbE cards. I tested several kernels (some older > and some newer) and found that the only thing in common was that with -RT > kernels the performance went down with concurrent streams. > > While the test was showing the numbers for receiving as well as sending, the > receiving numbers are not reliable because that machine was running a -RT > kernel for these tests. > > I was just wondering if anyone had seen this problem before or would have any > idea on where to start hunting for the solution. could you enable CONFIG_LOCK_STAT echo 0 > /proc/lock_stat and report the output of (preferably not 80 column wrapped): grep : /proc/lock_stat | head or otherwise if there are any highly contended network locks listed? - Peter