From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org X-Spam-Level: X-Spam-Status: No, score=-5.4 required=3.0 tests=BAYES_00, HEADER_FROM_DIFFERENT_DOMAINS,MAILING_LIST_MULTI,NICE_REPLY_A,SPF_HELO_NONE, SPF_PASS,USER_AGENT_SANE_1 autolearn=no autolearn_force=no version=3.4.0 Received: from mail.kernel.org (mail.kernel.org [198.145.29.99]) by smtp.lore.kernel.org (Postfix) with ESMTP id 5753AC4346E for ; Sun, 27 Sep 2020 08:38:00 +0000 (UTC) Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by mail.kernel.org (Postfix) with ESMTP id 0C482207BC for ; Sun, 27 Sep 2020 08:38:00 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1730506AbgI0Ih7 (ORCPT ); Sun, 27 Sep 2020 04:37:59 -0400 Received: from mga17.intel.com ([192.55.52.151]:9825 "EHLO mga17.intel.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1730438AbgI0Ih6 (ORCPT ); Sun, 27 Sep 2020 04:37:58 -0400 IronPort-SDR: 3jh06MT2h0ga7JMdr88kpcm7+lR9i2PFPOZqhG4hu3QiLZRl8UL5Ag6l02HL/1t3XDFq31pZSK TWYurtEz9JRQ== X-IronPort-AV: E=McAfee;i="6000,8403,9756"; a="141861087" X-IronPort-AV: E=Sophos;i="5.77,309,1596524400"; d="scan'208";a="141861087" X-Amp-Result: SKIPPED(no attachment in message) X-Amp-File-Uploaded: False Received: from fmsmga004.fm.intel.com ([10.253.24.48]) by fmsmga107.fm.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 26 Sep 2020 22:56:06 -0700 IronPort-SDR: 9oqP1bHZM0aubjyxN/MQ/IcRCDT3cfShUUigwZ7pc3L/Nx73DkcCao9w/Pqj6wpt1leXxKeXLL EsC6f2fuNeYA== X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="5.77,309,1596524400"; d="scan'208";a="337726177" Received: from cli6-desk1.ccr.corp.intel.com (HELO [10.239.161.135]) ([10.239.161.135]) by fmsmga004.fm.intel.com with ESMTP; 26 Sep 2020 22:56:03 -0700 Subject: Re: [RFC PATCH v2] sched/fair: select idle cpu from idle cpumask in sched domain To: Vincent Guittot Cc: Mel Gorman , Ingo Molnar , Peter Zijlstra , Juri Lelli , Dietmar Eggemann , Steven Rostedt , Ben Segall , Valentin Schneider , Tim Chen , linux-kernel , Qais Yousef , Jiang Biao References: <78d608f2-b974-e940-da32-b37777bc405a@linux.intel.com> <8a86b085-b445-b1c2-9b46-6346d923abf0@linux.intel.com> <20200925164514.GA6432@vingu-book> From: "Li, Aubrey" Message-ID: Date: Sun, 27 Sep 2020 13:56:02 +0800 User-Agent: Mozilla/5.0 (Windows NT 10.0; WOW64; rv:68.0) Gecko/20100101 Thunderbird/68.9.0 MIME-Version: 1.0 In-Reply-To: <20200925164514.GA6432@vingu-book> Content-Type: text/plain; charset=utf-8 Content-Language: en-US Content-Transfer-Encoding: 8bit Precedence: bulk List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On 2020/9/26 0:45, Vincent Guittot wrote: > Le vendredi 25 sept. 2020 à 17:21:46 (+0800), Li, Aubrey a écrit : >> Hi Vicent, >> >> On 2020/9/24 21:09, Vincent Guittot wrote: >>>>>> >>>>>> Would you mind share uperf(netperf load) result on your side? That's the >>>>>> workload I have seen the most benefit this patch contributed under heavy >>>>>> load level. >>>>> >>>>> with uperf, i've got the same kind of result as sched pipe >>>>> tip/sched/core: Throughput 24.83Mb/s (+/- 0.09%) >>>>> with this patch: Throughput 19.02Mb/s (+/- 0.71%) which is a 23% >>>>> regression as for sched pipe >>>>> >>>> In case this is caused by the logic error in this patch(sorry again), did >>>> you see any improvement in patch V2? Though it does not helps for nohz=off >>>> case, just want to know if it helps or does not help at all on arm platform. >>> >>> With the v2 which rate limit the update of the cpumask (but doesn't >>> support sched_idle stask), I don't see any performance impact: >> >> I agree we should go the way with cpumask update rate limited. >> >> And I think no performance impact for sched-pipe is expected, as this workload >> has only 2 threads and the platform has 8 cores, so mostly previous cpu is >> returned, and even if select_idle_sibling is called, select_idle_core is hit >> and rarely call select_idle_cpu. > > my platform is not smt so select_idle_core is nop. Nevertheless select_idle_cpu > is almost never called because prev is idle and selected before calling it in > our case > >> >> But I'm more curious why there is 23% performance penalty? So for this patch, if >> you revert this change but keep cpumask updated, is 23% penalty still there? >> >> - cpumask_and(cpus, sched_domain_span(sd), p->cpus_ptr); >> + cpumask_and(cpus, sds_idle_cpus(sd->shared), p->cpus_ptr); > > I was about to say that reverting this line should not change anything because > we never reach this point but it does in fact. And after looking at a trace, > I can see that the 2 threads of perf bench sched pipe are on the same CPU and > that the sds_idle_cpus(sd->shared) is always empty. In fact, the rq->curr is > not yet idle and still point to the cfs task when you call update_idle_cpumask(). > This means that once cleared, the bit will never be set > You can remove the test in update_idle_cpumask() which is called either when > entering idle or when there is only sched_idle tasks that are runnable. > > @@ -6044,8 +6044,7 @@ void update_idle_cpumask(struct rq *rq) > sd = rcu_dereference(per_cpu(sd_llc, cpu)); > if (!sd || !sd->shared) > goto unlock; > - if (!available_idle_cpu(cpu) || !sched_idle_cpu(cpu)) > - goto unlock; > + > cpumask_set_cpu(cpu, sds_idle_cpus(sd->shared)); > unlock: > rcu_read_unlock(); > > With this fix, the performance decrease is only 2% > >> >> I just wonder if it's caused by the atomic ops as you have two cache domains with >> sd_llc(?). Do you have a x86 machine to make a comparison? It's hard for me to find >> an ARM machine but I'll try. >> >> Also, for uperf(task thread num = cpu num) workload, how is it on patch v2? no any >> performance impact? > > with v2 :  Throughput 24.97Mb/s (+/- 0.07%) so there is no perf regression > Thanks Vincent, let me try to refine this patch. -Aubrey