From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S934552AbdDFKmO (ORCPT ); Thu, 6 Apr 2017 06:42:14 -0400 Received: from mx2.suse.de ([195.135.220.15]:44865 "EHLO mx2.suse.de" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S932442AbdDFKmH (ORCPT ); Thu, 6 Apr 2017 06:42:07 -0400 Date: Thu, 6 Apr 2017 12:42:04 +0200 From: Michal Hocko To: Peter Zijlstra Cc: Srikar Dronamraju , Ingo Molnar , LKML , Mel Gorman , Rik van Riel Subject: Re: [PATCH] sched: Fix numabalancing to work with isolated cpus Message-ID: <20170406104204.GI5497@dhcp22.suse.cz> References: <1491326848-5748-1-git-send-email-srikar@linux.vnet.ibm.com> <20170405125743.GB7258@dhcp22.suse.cz> <20170405152215.GA6019@linux.vnet.ibm.com> <20170405164437.GT6035@dhcp22.suse.cz> <20170406071950.GA5843@linux.vnet.ibm.com> <20170406073436.GD5497@dhcp22.suse.cz> <20170406092329.52zzkxlkvitnh44c@hirez.programming.kicks-ass.net> <20170406101348.GH5497@dhcp22.suse.cz> <20170406102957.xz53avexxyiim4az@hirez.programming.kicks-ass.net> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20170406102957.xz53avexxyiim4az@hirez.programming.kicks-ass.net> User-Agent: Mutt/1.5.23 (2014-03-12) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Thu 06-04-17 12:29:57, Peter Zijlstra wrote: > On Thu, Apr 06, 2017 at 12:13:49PM +0200, Michal Hocko wrote: > > On Thu 06-04-17 11:23:29, Peter Zijlstra wrote: > > > On Thu, Apr 06, 2017 at 09:34:36AM +0200, Michal Hocko wrote: > > [...] > > > > I would really like to see it confirmed by the scheduler maintainers and > > > > documented properly as well. What you are claiming here is rather > > > > surprising to my understanding of what isolcpus acutally is. > > > > > > isolcpus gets you a set of fully partitioned CPUs. What's surprising > > > about that? > > > > Well, I thought that all isolated cpus simply form their own scheduling > > domain which is isolated from the general workload on the system > > (kthreads, softirqs etc...). > > No, they all form their own 1 cpu partition. Is this something dictated by usecases which rely on isolcpus or rather nobody bothered to implement one scheduling domain? -- Michal Hocko SUSE Labs