From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1752568AbZBOKOQ (ORCPT ); Sun, 15 Feb 2009 05:14:16 -0500 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1751696AbZBOKOA (ORCPT ); Sun, 15 Feb 2009 05:14:00 -0500 Received: from mx3.mail.elte.hu ([157.181.1.138]:50066 "EHLO mx3.mail.elte.hu" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751285AbZBOKN7 (ORCPT ); Sun, 15 Feb 2009 05:13:59 -0500 Date: Sun, 15 Feb 2009 11:13:51 +0100 From: Ingo Molnar To: Damien Wyart Cc: "Rafael J. Wysocki" , Linux Kernel Mailing List , Kernel Testers List Subject: Re: [Bug #12650] Strange load average and ksoftirqd behavior with 2.6.29-rc2-git1 Message-ID: <20090215101351.GA23274@elte.hu> References: <20090215080941.GA2295@localhost.localdomain> <20090215090026.GA31147@elte.hu> <20090215095128.GA3234@localhost.localdomain> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20090215095128.GA3234@localhost.localdomain> User-Agent: Mutt/1.5.18 (2008-05-17) X-ELTE-VirusStatus: clean X-ELTE-SpamScore: -1.5 X-ELTE-SpamLevel: X-ELTE-SpamCheck: no X-ELTE-SpamVersion: ELTE 2.0 X-ELTE-SpamCheck-Details: score=-1.5 required=5.9 tests=BAYES_00 autolearn=no SpamAssassin version=3.2.3 -1.5 BAYES_00 BODY: Bayesian spam probability is 0 to 1% [score: 0.0000] Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org * Damien Wyart wrote: > > > Load average is between 0.30 and 0.60 when the machine is idle, the > > > two ksoftirqd threads get a very high total running time in top. > > > This is on a P4 machine. On a recent laptop, the load is not as > > > high, but the ksoftirqd threads also get a very high total running > > > time. > > > Mind having a look at this anomaly with the function graph tracer? We are > > interested in a representative trace that shows the weird ksoftirq activities. > > [ which trace could possibly pinpoint their origin. ] > > > Here's the tracing quickstart: > > > http://redhat.com/~mingo/tip.git/tracing-quickstart.txt > > > Also attached below. Let me know if you have trouble getting a good trace, > > or if any of the steps were non-intuitive or burdensome to you. > > Thanks for your feedback & explanations; I will try to have a look at > all this material asap, I think tomorrow. Note that if the box you test this on is multi-core or HT, then interpreting traces is easier if there's just a single CPU to look at. In that case i'd suggest to reproduce with just a single core, by turning the second one off: echo 0 > /sys/devices/system/cpu/cpu1/online Or, if the problem only occurs with two cpus, restrict tracing to CPU#1: echo 2 > /debug/tracing/tracing_cpumask Ingo