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 Received: from mail.kernel.org (mail.kernel.org [198.145.29.99]) by smtp.lore.kernel.org (Postfix) with ESMTP id 8D5FDC433F5 for ; Wed, 3 Nov 2021 07:08:38 +0000 (UTC) Received: from kanga.kvack.org (kanga.kvack.org [205.233.56.17]) by mail.kernel.org (Postfix) with ESMTP id 1E97A61156 for ; Wed, 3 Nov 2021 07:08:38 +0000 (UTC) DMARC-Filter: OpenDMARC Filter v1.4.1 mail.kernel.org 1E97A61156 Authentication-Results: mail.kernel.org; dmarc=fail (p=none dis=none) header.from=gmail.com Authentication-Results: mail.kernel.org; spf=pass smtp.mailfrom=kvack.org Received: by kanga.kvack.org (Postfix) id 667F16B006C; Wed, 3 Nov 2021 03:08:37 -0400 (EDT) Received: by kanga.kvack.org (Postfix, from userid 40) id 617D26B0071; Wed, 3 Nov 2021 03:08:37 -0400 (EDT) X-Delivered-To: int-list-linux-mm@kvack.org Received: by kanga.kvack.org (Postfix, from userid 63042) id 5080B940007; Wed, 3 Nov 2021 03:08:37 -0400 (EDT) X-Delivered-To: linux-mm@kvack.org Received: from forelay.hostedemail.com (smtprelay0243.hostedemail.com [216.40.44.243]) by kanga.kvack.org (Postfix) with ESMTP id 40EFD6B006C for ; Wed, 3 Nov 2021 03:08:37 -0400 (EDT) Received: from smtpin06.hostedemail.com (10.5.19.251.rfc1918.com [10.5.19.251]) by forelay05.hostedemail.com (Postfix) with ESMTP id F1C191801C1AD for ; Wed, 3 Nov 2021 07:08:36 +0000 (UTC) X-FDA: 78766741032.06.E44C51F Received: from mail-qv1-f51.google.com (mail-qv1-f51.google.com [209.85.219.51]) by imf12.hostedemail.com (Postfix) with ESMTP id 9888010000A8 for ; Wed, 3 Nov 2021 07:08:36 +0000 (UTC) Received: by mail-qv1-f51.google.com with SMTP id gh1so1965967qvb.8 for ; Wed, 03 Nov 2021 00:08:36 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20210112; h=mime-version:references:in-reply-to:from:date:message-id:subject:to :cc; bh=ttRK9jhUfRsR37wv0FMnPjvaQAeGbLSUMSc2dK17DwM=; b=AQYl/WVLl6K4JhlsVJAj0S0QriZMgR3JogNW0AiBQWKPu1V2FMi1JhlCE44FYYSD8V +8fdEa0dMK3BkBe/13iHOD1XnxUGxyLzJciAhKfo4Y3SDNIu3r6y6CzgVxZbNjbFMa/f c517lwdd1GJ+U/Xz3/QvkOjaS/DzAzTsrCiPeOFZhdehgGhWkkJYr7e77Zo5qPxmJhPt cgZDCagWrv6bBPrXkVFReN0G5rWSKqjQyNFFSM4tmPbd10VEkM9ZcLbz1nGQw3G1TUEx 4rwPKgGQjKPtmt816V/an99GloslpuSYEI3oaRTJk70JJ3UDTbEo126gHfOJUMZehdmG BU/Q== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20210112; h=x-gm-message-state:mime-version:references:in-reply-to:from:date :message-id:subject:to:cc; bh=ttRK9jhUfRsR37wv0FMnPjvaQAeGbLSUMSc2dK17DwM=; b=Aq72NAFv33cqU48T8JJH6vQrV4BBG37P1rGIWmYGxt9n+/UGuQjX16Qh5De7QD3jQ5 1UQ1Paano/cF2smL5taATRU6hXGuoAuwCN83zFwqesml1V2+O8OKPmaeitn6PsycAs66 mR3Ag9AEQOP8YeAEKS21X0LZO4b0PPnHgD9AjPo3lHDvo+yb8qe0Itqh3aqrf4oZHuwQ wq/OOhPOanP1/5Fgyp7V8kV+HRdBNR+QCepAAcwkTSScBAWjDTrFP1u+aCDJttW3XxB5 0a3OXvQ0jqgzX6hjL6BLrInkp2Tc8lqVSoWYlMx1Bl4lMzxC9XwcoULQQwmXFzpwy17G yzrA== X-Gm-Message-State: AOAM530baqiusZmuwR7jZxhvfBABqJ4Ky/CCMmJq3Q8StLBnofeuTsRs ZGEjkTA4t5dNEFRG2xkn0Wm16PCn7DHeqZqzgLE= X-Google-Smtp-Source: ABdhPJzH8UZanVAfKauDGAl2d+HeXHUgZdlmeFWd9l6TekjjfWRmiWD5CcGs3u2liEAYpB5r3jSz1ajPdK2F5x8FSuo= X-Received: by 2002:a05:6214:2b0f:: with SMTP id jx15mr20586340qvb.62.1635923316054; Wed, 03 Nov 2021 00:08:36 -0700 (PDT) MIME-Version: 1.0 References: <1634278612-17055-1-git-send-email-huangzhaoyang@gmail.com> In-Reply-To: From: Zhaoyang Huang Date: Wed, 3 Nov 2021 15:08:14 +0800 Message-ID: Subject: Re: [Resend PATCH] psi : calc cfs task memstall time more precisely To: Johannes Weiner Cc: Andrew Morton , Michal Hocko , Vladimir Davydov , Zhaoyang Huang , "open list:MEMORY MANAGEMENT" , LKML , Peter Zijlstra , Vincent Guittot Content-Type: text/plain; charset="UTF-8" X-Rspamd-Server: rspam03 X-Rspamd-Queue-Id: 9888010000A8 X-Stat-Signature: wqjuqcycpzqyzc16inwje87qjtnd8x56 Authentication-Results: imf12.hostedemail.com; dkim=pass header.d=gmail.com header.s=20210112 header.b="AQYl/WVL"; dmarc=pass (policy=none) header.from=gmail.com; spf=pass (imf12.hostedemail.com: domain of huangzhaoyang@gmail.com designates 209.85.219.51 as permitted sender) smtp.mailfrom=huangzhaoyang@gmail.com X-HE-Tag: 1635923316-713075 X-Bogosity: Ham, tests=bogofilter, spamicity=0.000000, version=1.2.4 Sender: owner-linux-mm@kvack.org Precedence: bulk X-Loop: owner-majordomo@kvack.org List-ID: +Vincent Guittot On Wed, Nov 3, 2021 at 3:07 PM Zhaoyang Huang wrote: > > On Wed, Nov 3, 2021 at 3:47 AM Johannes Weiner wrote: > > > > CC peterz as well for rt and timekeeping magic > > > > On Fri, Oct 15, 2021 at 02:16:52PM +0800, Huangzhaoyang wrote: > > > From: Zhaoyang Huang > > > > > > In an EAS enabled system, there are two scenarios discordant to current design, > > > > > > 1. workload used to be heavy uneven among cores for sake of scheduler policy. > > > RT task usually preempts CFS task in little core. > > > 2. CFS task's memstall time is counted as simple as exit - entry so far, which > > > ignore the preempted time by RT, DL and Irqs. > > > > > > With these two constraints, the percpu nonidle time would be mainly consumed by > > > none CFS tasks and couldn't be averaged. Eliminating them by calc the time growth > > > via the proportion of cfs_rq's utilization on the whole rq. > > > > > > eg. > > > Here is the scenario which this commit want to fix, that is the rt and irq consume > > > some utilization of the whole rq. This scenario could be typical in a core > > > which is assigned to deal with all irqs. Furthermore, the rt task used to run on > > > little core under EAS. > > > > > > Binder:305_3-314 [002] d..1 257.880195: psi_memtime_fixup: original:30616,adjusted:25951,se:89,cfs:353,rt:139,dl:0,irq:18 > > > droid.phone-1525 [001] d..1 265.145492: psi_memtime_fixup: original:61616,adjusted:53492,se:55,cfs:225,rt:121,dl:0,irq:15 > > > > > > Signed-off-by: Zhaoyang Huang > > > --- > > > kernel/sched/psi.c | 20 +++++++++++++++++++- > > > 1 file changed, 19 insertions(+), 1 deletion(-) > > > > > > diff --git a/kernel/sched/psi.c b/kernel/sched/psi.c > > > index cc25a3c..754a836 100644 > > > --- a/kernel/sched/psi.c > > > +++ b/kernel/sched/psi.c > > > @@ -182,6 +182,8 @@ struct psi_group psi_system = { > > > > > > static void psi_avgs_work(struct work_struct *work); > > > > > > +static unsigned long psi_memtime_fixup(u32 growth); > > > + > > > static void group_init(struct psi_group *group) > > > { > > > int cpu; > > > @@ -492,6 +494,21 @@ static u64 window_update(struct psi_window *win, u64 now, u64 value) > > > return growth; > > > } > > > > > > +static unsigned long psi_memtime_fixup(u32 growth) > > > +{ > > > + struct rq *rq = task_rq(current); > > > + unsigned long growth_fixed = (unsigned long)growth; > > > + > > > + if (!(current->policy == SCHED_NORMAL || current->policy == SCHED_BATCH)) > > > + return growth_fixed; > > > + > > > + if (current->in_memstall) > > > + growth_fixed = div64_ul((1024 - rq->avg_rt.util_avg - rq->avg_dl.util_avg > > > + - rq->avg_irq.util_avg + 1) * growth, 1024); > > > + > > > + return growth_fixed; > > > +} > > > + > > > static void init_triggers(struct psi_group *group, u64 now) > > > { > > > struct psi_trigger *t; > > > @@ -658,6 +675,7 @@ static void record_times(struct psi_group_cpu *groupc, u64 now) > > > } > > > > > > if (groupc->state_mask & (1 << PSI_MEM_SOME)) { > > > + delta = psi_memtime_fixup(delta); > > > add vincent for advise on cpu load mechanism > > > Ok, so we want to deduct IRQ and RT preemption time from the memstall > > period of an active reclaimer, since it's technically not stalled on > > memory during this time but on CPU. > > > > However, we do NOT want to deduct IRQ and RT time from memstalls that > > are sleeping on refaults swapins, since they are not affected by what > > is going on on the CPU. > > > > Does util_avg capture that difference? I'm not confident it does - but > > correct me if I'm wrong. We need length of time during which and IRQ > > or an RT task preempted the old rq->curr, not absolute irq/rt length. > As far as my understanding, core's capacity = IRQ + DEADLINE + RT + > CFS. For a certain time period, all cfs tasks preempt each other > inside CFS's utilization. So the sleeping on refaults is counted in. > > > > (Btw, such preemption periods, in addition to being deducted from > > memory stalls, should probably also be added to CPU contention stalls, > > to make CPU pressure reporting more accurate as well.)