From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1753764Ab3IPFwK (ORCPT ); Mon, 16 Sep 2013 01:52:10 -0400 Received: from merlin.infradead.org ([205.233.59.134]:41729 "EHLO merlin.infradead.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752858Ab3IPFwJ (ORCPT ); Mon, 16 Sep 2013 01:52:09 -0400 Date: Mon, 16 Sep 2013 07:52:02 +0200 From: Peter Zijlstra To: Vladimir Davydov Cc: Ingo Molnar , Paul Turner , linux-kernel@vger.kernel.org, devel@openvz.org Subject: Re: [PATCH 1/2] sched: calculate_imbalance: Fix local->avg_load > sds->avg_load case Message-ID: <20130916055202.GL21832@twins.programming.kicks-ass.net> References: <8f596cc6bc0e5e655119dc892c9bfcad26e971f4.1379252740.git.vdavydov@parallels.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <8f596cc6bc0e5e655119dc892c9bfcad26e971f4.1379252740.git.vdavydov@parallels.com> User-Agent: Mutt/1.5.21 (2012-12-30) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Sun, Sep 15, 2013 at 05:49:13PM +0400, Vladimir Davydov wrote: > In busiest->group_imb case we can come to calculate_imbalance() with > local->avg_load >= busiest->avg_load >= sds->avg_load. This can result > in imbalance overflow, because it is calculated as follows > > env->imbalance = min( > max_pull * busiest->group_power, > (sds->avg_load - local->avg_load) * local->group_power > ) / SCHED_POWER_SCALE; > > As a result we can end up constantly bouncing tasks from one cpu to > another if there are pinned tasks. > > Fix this by skipping the assignment and assuming imbalance=0 in case > local->avg_load > sds->avg_load. > -- > The bug can be caught by running 2*N cpuhogs pinned to two logical cpus > belonging to different cores on an HT-enabled machine with N logical > cpus: just look at se.nr_migrations growth. > > Signed-off-by: Vladimir Davydov > --- > kernel/sched/fair.c | 3 ++- > 1 file changed, 2 insertions(+), 1 deletion(-) > > diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c > index 9b3fe1c..507a8a9 100644 > --- a/kernel/sched/fair.c > +++ b/kernel/sched/fair.c > @@ -4896,7 +4896,8 @@ static inline void calculate_imbalance(struct lb_env *env, struct sd_lb_stats *s > * max load less than avg load(as we skip the groups at or below > * its cpu_power, while calculating max_load..) > */ > - if (busiest->avg_load < sds->avg_load) { > + if (busiest->avg_load <= sds->avg_load || > + local->avg_load >= sds->avg_load) { > env->imbalance = 0; > return fix_small_imbalance(env, sds); > } Why the = part? Surely 'busiest->avg_load < sds->avg_load || local->avg_load > sds->avg_load' avoids both underflows?