From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1752394AbeBLUJl (ORCPT ); Mon, 12 Feb 2018 15:09:41 -0500 Received: from outbound-smtp24.blacknight.com ([81.17.249.192]:39164 "EHLO outbound-smtp24.blacknight.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751021AbeBLUJk (ORCPT ); Mon, 12 Feb 2018 15:09:40 -0500 Date: Mon, 12 Feb 2018 20:09:38 +0000 From: Mel Gorman To: Andreas Mohr Cc: Peter Zijlstra , Mike Galbraith , Matt Fleming , LKML Subject: Re: [PATCH 4/4] sched/fair: Do not migrate due to a sync wakeup on exit Message-ID: <20180212200938.sfofuhztrsbvsriy@techsingularity.net> References: <20180212145857.8056-1-mgorman@techsingularity.net> <20180212145857.8056-5-mgorman@techsingularity.net> <20180212200127.GA27522@rhlx01.hs-esslingen.de> MIME-Version: 1.0 Content-Type: text/plain; charset=iso-8859-15 Content-Disposition: inline In-Reply-To: <20180212200127.GA27522@rhlx01.hs-esslingen.de> User-Agent: NeoMutt/20170912 (1.9.0) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Mon, Feb 12, 2018 at 09:01:27PM +0100, Andreas Mohr wrote: > Hello, > > On Mon, Feb 12, 2018 at 02:58:57PM +0000, Mel Gorman wrote: > > diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c > > index 28c8d9c91955..50442697b455 100644 > > --- a/kernel/sched/fair.c > > +++ b/kernel/sched/fair.c > > @@ -5710,8 +5710,14 @@ wake_affine_idle(int this_cpu, int prev_cpu, int sync) > > if (idle_cpu(this_cpu) && cpus_share_cache(this_cpu, prev_cpu)) > > return idle_cpu(prev_cpu) ? prev_cpu : this_cpu; > > > > - if (sync && cpu_rq(this_cpu)->nr_running == 1) > > + if (sync && cpu_rq(this_cpu)->nr_running == 1) { > > + /* Avoid tasks exiting pulling parents to new nodes */ > > Semi-confusing wording (WTH are "pulling parents"? :) - possibly better variant?: > Avoid having exiting tasks pull their parents to new nodes That is better wording. If I don't replace the patch with Peter's version, I'll update it. > Good catch, it seems. > Thanks. -- Mel Gorman SUSE Labs