From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1754325AbZLXXxd (ORCPT ); Thu, 24 Dec 2009 18:53:33 -0500 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1753077AbZLXXxc (ORCPT ); Thu, 24 Dec 2009 18:53:32 -0500 Received: from lennier.cc.vt.edu ([198.82.162.213]:56900 "EHLO lennier.cc.vt.edu" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751338AbZLXXxb (ORCPT ); Thu, 24 Dec 2009 18:53:31 -0500 X-Mailer: exmh version 2.7.2 01/07/2005 with nmh-1.2 To: Carl Cc: linux-kernel@vger.kernel.org Subject: Re: CPU eaten by unknown code In-Reply-To: Your message of "Thu, 24 Dec 2009 16:12:21 +0200." From: Valdis.Kletnieks@vt.edu References: Mime-Version: 1.0 Content-Type: multipart/signed; boundary="==_Exmh_1261698808_3923P"; micalg=pgp-sha1; protocol="application/pgp-signature" Content-Transfer-Encoding: 7bit Date: Thu, 24 Dec 2009 18:53:28 -0500 Message-ID: <25319.1261698808@localhost> X-Mirapoint-Received-SPF: 128.173.34.103 localhost Valdis.Kletnieks@vt.edu 2 pass X-Mirapoint-IP-Reputation: reputation=neutral-1, source=Fixed, refid=n/a, actions=MAILHURDLE SPF TAG X-Junkmail-Info: (45) HELO_LOCALHOST X-Junkmail-Status: score=45/50, host=steiner.cc.vt.edu X-Junkmail-SD-Raw: score=unknown, refid=str=0001.0A020204.4B33FEF9.006A,ss=1,fgs=0, ip=0.0.0.0, so=2009-09-22 00:05:22, dmn=2009-09-10 00:05:08, mode=multiengine X-Junkmail-IWF: false Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org --==_Exmh_1261698808_3923P Content-Type: text/plain; charset=us-ascii On Thu, 24 Dec 2009 16:12:21 +0200, Carl said: > The loop takes 1.2 seconds for about 138 seconds. > The loop takes 2.4 second for about 45 seconds. > And this repeats forever. > > In other words, every 138 seconds something is "running" for > 45 seconds and it takes 50% of my CPU time! Thermal throttling? Is the fan functional? Can you monitor the CPU temperature while running your testcase? My second guess is something is causing a cpuspeed module to misfire and drop the CPU to a 50% clock speed even though there's load on the system. That one is easily testable - just do a 'grep MHz /proc/cpuinfo' during the 'fast' and 'slow' phases. --==_Exmh_1261698808_3923P Content-Type: application/pgp-signature -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.10 (GNU/Linux) Comment: Exmh version 2.5 07/13/2001 iD8DBQFLM/74cC3lWbTT17ARAgsmAJwPlIImHge7YoK4mLYkrWCrngn0RACg5lfT +YfkjCJoWIz9OR5/xQZMCy8= =fETf -----END PGP SIGNATURE----- --==_Exmh_1261698808_3923P--