From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S932789AbeCIXGd convert rfc822-to-8bit (ORCPT ); Fri, 9 Mar 2018 18:06:33 -0500 Received: from mx3-rdu2.redhat.com ([66.187.233.73]:59458 "EHLO mx1.redhat.com" rhost-flags-OK-OK-OK-FAIL) by vger.kernel.org with ESMTP id S932195AbeCIXGc (ORCPT ); Fri, 9 Mar 2018 18:06:32 -0500 Subject: Re: [PATCH v4] cpuset: Enable cpuset controller in default hierarchy To: Peter Zijlstra Cc: Mike Galbraith , Tejun Heo , Li Zefan , Johannes Weiner , Ingo Molnar , cgroups@vger.kernel.org, linux-kernel@vger.kernel.org, linux-doc@vger.kernel.org, kernel-team@fb.com, pjt@google.com, luto@amacapital.net, torvalds@linux-foundation.org, Roman Gushchin References: <1520609707-16582-1-git-send-email-longman@redhat.com> <1520613285.12489.36.camel@gmx.de> <1c3fe7b0-2600-c46d-1527-d3aaf024bb91@redhat.com> <1520619426.27998.18.camel@gmx.de> <55809fe4-98ba-5566-86ed-457acfef0e1c@redhat.com> <1520624424.27998.76.camel@gmx.de> <53de9683-01b7-bac4-8b70-dc1f93ede600@redhat.com> <20180309221736.GB5926@hirez.programming.kicks-ass.net> From: Waiman Long Organization: Red Hat Message-ID: Date: Fri, 9 Mar 2018 18:06:29 -0500 User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:52.0) Gecko/20100101 Thunderbird/52.2.0 MIME-Version: 1.0 In-Reply-To: <20180309221736.GB5926@hirez.programming.kicks-ass.net> Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: 8BIT Content-Language: en-US Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On 03/09/2018 05:17 PM, Peter Zijlstra wrote: > On Fri, Mar 09, 2018 at 03:43:34PM -0500, Waiman Long wrote: >> The isolcpus= parameter just reduce the cpus available to the rests of >> the system. The cpuset controller does look at that value and make >> adjustment accordingly, but it has no dependence on exclusive cpu/mem >> features of cpuset. > The isolcpus= boot param is donkey shit and needs to die. cpuset _used_ > to be able to fully replace it, but with the advent of cgroup 'feature' > this got lost. > > And instead of fixing it, you're making it _far_ worse. You completely > removed all the bits that allow repartitioning the scheduler domains. > > Mike is completely right, full NAK on any such approach. So you are talking about sched_relax_domain_level and sched_load_balance. I have not removed any bits. I just haven't exposed them yet. It does seem like these 2 control knobs are useful from the scheduling perspective. Do we also need cpu_exclusive or just the two sched control knobs are enough? Cheers, Longman