From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1751607Ab0JDDIm (ORCPT ); Sun, 3 Oct 2010 23:08:42 -0400 Received: from mailout-de.gmx.net ([213.165.64.23]:45136 "HELO mail.gmx.net" rhost-flags-OK-OK-OK-FAIL) by vger.kernel.org with SMTP id S1751506Ab0JDDIl (ORCPT ); Sun, 3 Oct 2010 23:08:41 -0400 X-Authenticated: #14349625 X-Provags-ID: V01U2FsdGVkX1/jXNRODb/j9Lh4k6xbn/S2eOQmw5c8xYMBQL5Z0D 3DEAkmsFe0tc+r Subject: Re: [PATCH 0/3][RFC] Improve load balancing when tasks have large weight differential From: Mike Galbraith To: Nikhil Rao Cc: Ingo Molnar , Peter Zijlstra , Venkatesh Pallipadi , linux-kernel@vger.kernel.org In-Reply-To: References: <1285633798-26886-1-git-send-email-ncrao@google.com> <1285682273.7469.3.camel@marge.simson.net> <1285724758.7440.11.camel@marge.simson.net> Content-Type: text/plain Date: Mon, 04 Oct 2010 05:08:37 +0200 Message-Id: <1286161717.7410.12.camel@marge.simson.net> Mime-Version: 1.0 X-Mailer: Evolution 2.24.1.1 Content-Transfer-Encoding: 7bit X-Y-GMX-Trusted: 0 Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Sorry for the late reply. (fired up your patchlet bright and early so it didn't rot in my inbox any longer;) On Wed, 2010-09-29 at 12:32 -0700, Nikhil Rao wrote: > On Tue, Sep 28, 2010 at 6:45 PM, Mike Galbraith wrote: > > On Tue, 2010-09-28 at 14:15 -0700, Nikhil Rao wrote: > > > >> Thanks for running this. I've not been able to reproduce what you are > >> seeing on the few test machines that I have (different combinations of > >> MC, CPU and NODE domains). Can you please give me more info about > >> your setup? > > > > It's a plain-jane Q6600 box, so has only MC and CPU domains. > > > > It doesn't necessarily _instantly_ "stick", can take a couple tries, or > > a little time. > > The closest I have is a quad-core dual-socket machine (MC, CPU > domains). And I'm having trouble reproducing it on that machine as > well :-( I ran 5 soaker threads (one of them niced to -15) for a few > hours and didn't see the problem. Can you please give me some trace > data & schedstats to work with? Booting with isolcpus or offlining the excess should help. > Looking at the patch/code, I suspect active migration on the CPU > scheduling domain pushes the nice 0 task (running on the same socket > as the nice -15 task) to the other socket. This leaves you with an > idle core on the nice -15 socket, and with soaker threads there is no > way to come back to a 100% utilized state. One possible explanation is > the group capacity for a sched group in the CPU sched domain is > rounded to 1 (instead of 2). I have a patch below that throws a hammer > at the problem and uses group weight instead of group capacity (this > is experimental, will refine it if it works). Can you please see if > that solves the problem? Nope, didn't help. I'll poke at it, but am squabbling elsewhere atm. -Mike