Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1754560AbZGWWR0 (ORCPT ); Thu, 23 Jul 2009 18:17:26 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1753895AbZGWWRZ (ORCPT ); Thu, 23 Jul 2009 18:17:25 -0400 Received: from smtp-out.google.com ([216.239.45.13]:31936 "EHLO smtp-out.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753791AbZGWWRY (ORCPT ); Thu, 23 Jul 2009 18:17:24 -0400 DomainKey-Signature: a=rsa-sha1; s=beta; d=google.com; c=nofws; q=dns; h=mime-version:in-reply-to:references:date:message-id:subject:from:to: cc:content-type:content-transfer-encoding:x-system-of-record; b=CZq7CuG5DHBNXzgEPR+TvxzL+/wL//O531Ym0/n6JVjPuxMBVq3VJVHBAPDicGs/1 AdMLZcU0gP2A798LGUQ8g== MIME-Version: 1.0 In-Reply-To: <20090723075735.GA18878@in.ibm.com> References: <20090723075735.GA18878@in.ibm.com> Date: Thu, 23 Jul 2009 15:17:18 -0700 Message-ID: Subject: Re: CFS group scheduler fairness broken starting from 2.6.29-rc1 From: Ken Chen To: bharata@linux.vnet.ibm.com Cc: linux-kernel@vger.kernel.org, Ingo Molnar , Peter Zijlstra , Dhaval Giani , Srivatsa Vaddagiri , Balbir Singh Content-Type: text/plain; charset=ISO-8859-1 X-System-Of-Record: true Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Transfer-Encoding: 8bit X-MIME-Autoconverted: from base64 to 8bit by alpha.home.local id n6NMI0J7013584 Content-Length: 1347 Lines: 3 On Thu, Jul 23, 2009 at 12:57 AM, Bharata BRao wrote:> Hi,>> Group scheduler fainess is broken since 2.6.29-rc1. git bisect led me> to this commit:>> commit ec4e0e2fe018992d980910db901637c814575914> Author: Ken Chen > Date: ? Tue Nov 18 22:41:57 2008 -0800>> ? ?sched: fix inconsistency when redistribute per-cpu tg->cfs_rq shares>> ? ?Impact: make load-balancing more consistent> ....>> ======================================================================> ? ? ? ? ? ? ? ? ? ? ? ?% CPU time division b/n groups> Group ? ? ? ? ? 2.6.29-rc1 ? ? ? ? ? ? ?2.6.29-rc1 w/o the above patch> ======================================================================> a with 8 tasks ?44 ? ? ? ? ? ? ? ? ? ? ?31> b with 5 tasks ?32 ? ? ? ? ? ? ? ? ? ? ?34> c with 3 tasks ?22 ? ? ? ? ? ? ? ? ? ? ?34> ======================================================================> All groups had equal shares. What value did you use for each task_group's share? For very largevalue of tg->shares, it could be that all of the boost went to one CPUand subsequently causes load-balancer to shuffle tasks around. Do yousee any unexpected task migration? - Ken????{.n?+???????+%?????ݶ??w??{.n?+????{??G?????{ay?ʇڙ?,j??f???h?????????z_??(?階?ݢj"???m??????G????????????&???~???iO???z??v?^?m???? ????????I?