From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1752933AbZHZTFy (ORCPT ); Wed, 26 Aug 2009 15:05:54 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1752907AbZHZTFx (ORCPT ); Wed, 26 Aug 2009 15:05:53 -0400 Received: from bombadil.infradead.org ([18.85.46.34]:41776 "EHLO bombadil.infradead.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752897AbZHZTFx (ORCPT ); Wed, 26 Aug 2009 15:05:53 -0400 Subject: Re: [RFC][PATCH -rt] sched: fully ignore RT tasks for CFS load-balancing From: Peter Zijlstra To: Darren Hart Cc: Thomas Gleixner , Ingo Molnar , linux-kernel In-Reply-To: <4A958480.3070207@us.ibm.com> References: <1251297875.18584.5.camel@twins> <4A958480.3070207@us.ibm.com> Content-Type: text/plain Content-Transfer-Encoding: 7bit Date: Wed, 26 Aug 2009 21:04:45 +0200 Message-Id: <1251313485.18584.33.camel@twins> Mime-Version: 1.0 X-Mailer: Evolution 2.26.1 Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Wed, 2009-08-26 at 11:52 -0700, Darren Hart wrote: > Peter Zijlstra wrote: > > Since -rt runs IRQs and SoftIRQs as RT tasks and load-balancing is done > > from softirq context, there is always at least one RT task (and very > > likely multiple) running when we balance. > > > > The current (and totally broken) interaction between RT tasks and CFS > > load-balancing makes it so that we'll try to evacuate a significant > > amount of tasks due to RT tasks being runnable. > > > > Solve this in another broken way by not accounting RT tasks at all. > > > > This will likely break another class of cases, but until we can properly > > fix this, we might as well do this. > > Peter, > > Do we know why this became a problem lately? I don't recall seeing load > balancing issues like this way back around 2.6.21 or 22 when CFS was > released. Is the cpu_load stuff fairly recent? The earlier problem solved by the previous patch was caused by tglx tinkering with the cpu_load computation. This particular problem I'm not sure about, afaict it should have been present a long long time.