From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1757000Ab2DYR6r (ORCPT ); Wed, 25 Apr 2012 13:58:47 -0400 Received: from mga02.intel.com ([134.134.136.20]:46634 "EHLO mga02.intel.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1756717Ab2DYR6q (ORCPT ); Wed, 25 Apr 2012 13:58:46 -0400 X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="4.67,352,1309762800"; d="scan'208";a="133735705" Subject: Re: load balancing regression since commit 367456c7 From: Tim Chen To: Peter Zijlstra Cc: Suresh Siddha , Alex Shi , Ying , linux-kernel In-Reply-To: <1335375798.28150.271.camel@twins> References: <1334106376.19157.89.camel@schen9-DESK> <1334664553.28150.87.camel@twins> <1334681054.3796.28.camel@schen9-DESK> <1334930421.2463.60.camel@laptop> <1334940042.3796.48.camel@schen9-DESK> <1334940837.2463.70.camel@laptop> <1334942012.3796.50.camel@schen9-DESK> <1334943202.2463.71.camel@laptop> <1335365763.28150.267.camel@twins> <1335375537.3796.55.camel@schen9-DESK> <1335375798.28150.271.camel@twins> Content-Type: text/plain; charset="UTF-8" Date: Wed, 25 Apr 2012 10:58:46 -0700 Message-ID: <1335376726.3796.58.camel@schen9-DESK> Mime-Version: 1.0 X-Mailer: Evolution 2.32.3 (2.32.3-1.fc14) Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Wed, 2012-04-25 at 19:43 +0200, Peter Zijlstra wrote: > Gargh.. lost the change to kernel/sched/features.h, now included. > > Sorry for that. > > --- > Subject: sched: Fix more load-balance fallout > From: Peter Zijlstra > Date: Tue Apr 17 13:38:40 CEST 2012 > > Commits 367456c756a6 ("sched: Ditch per cgroup task lists for > load-balancing") and 5d6523ebd ("sched: Fix load-balance wreckage") > left some more wreckage. > > By setting loop_max unconditionally to ->nr_running load-balancing > could take a lot of time on very long runqueues (hackbench!). So keep > the sysctl as max limit of the amount of tasks we'll iterate. > > Furthermore, the min load filter for migration completely fails with > cgroups since inequality in per-cpu state can easily lead to such > small loads :/ > > Furthermore the change to add new tasks to the tail of the queue > instead of the head seems to have some effect.. not quite sure I > understand why. > > Combined these fixes solve the huge hackbench regression reported by > Tim when hackbench is ran in a cgroup. > > Reported-by: Tim Chen > Signed-off-by: Peter Zijlstra > Link: http://lkml.kernel.org/r/1335365763.28150.267.camel@twins The patch fixed the regression for me. Acked-by: Tim Chen