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 kanga.kvack.org (kanga.kvack.org [205.233.56.17]) by smtp.lore.kernel.org (Postfix) with ESMTP id 2F1DFC433F5 for ; Thu, 17 Feb 2022 14:26:20 +0000 (UTC) Received: by kanga.kvack.org (Postfix) id 7FA7F6B0074; Thu, 17 Feb 2022 09:26:19 -0500 (EST) Received: by kanga.kvack.org (Postfix, from userid 40) id 784176B0075; Thu, 17 Feb 2022 09:26:19 -0500 (EST) X-Delivered-To: int-list-linux-mm@kvack.org Received: by kanga.kvack.org (Postfix, from userid 63042) id 6252D6B0078; Thu, 17 Feb 2022 09:26:19 -0500 (EST) X-Delivered-To: linux-mm@kvack.org Received: from forelay.hostedemail.com (smtprelay0191.hostedemail.com [216.40.44.191]) by kanga.kvack.org (Postfix) with ESMTP id 50D636B0074 for ; Thu, 17 Feb 2022 09:26:19 -0500 (EST) Received: from smtpin13.hostedemail.com (10.5.19.251.rfc1918.com [10.5.19.251]) by forelay02.hostedemail.com (Postfix) with ESMTP id 0C5F396F0E for ; Thu, 17 Feb 2022 14:26:19 +0000 (UTC) X-FDA: 79152496878.13.3F0598C Received: from p3plsmtpa07-02.prod.phx3.secureserver.net (p3plsmtpa07-02.prod.phx3.secureserver.net [173.201.192.231]) by imf31.hostedemail.com (Postfix) with ESMTP id A05CF20012 for ; Thu, 17 Feb 2022 14:26:18 +0000 (UTC) Received: from localhost ([82.17.115.212]) by :SMTPAUTH: with ESMTPA id Khjbnebuy0d8ZKhjcnszcD; Thu, 17 Feb 2022 07:26:17 -0700 X-CMAE-Analysis: v=2.4 cv=QJKt+iHL c=1 sm=1 tr=0 ts=620e5b09 a=9gipVNR6X1CoIeAWHwLoWw==:117 a=9gipVNR6X1CoIeAWHwLoWw==:17 a=IkcTkHD0fZMA:10 a=caFokCbLAISjWCXd3CkA:9 a=QEXdDO2ut3YA:10 X-SECURESERVER-ACCT: atomlin@atomlin.com Date: Thu, 17 Feb 2022 14:26:15 +0000 From: Aaron Tomlin To: Frederic Weisbecker Cc: Aaron Tomlin , Marcelo Tosatti , Christoph Lameter , tglx@linutronix.de, mingo@kernel.org, linux-kernel@vger.kernel.org, linux-mm@kvack.org, Phil Auld Subject: Re: [RFC PATCH] tick/sched: Ensure quiet_vmstat() is called when the idle tick was stopped too Message-ID: <20220217142615.xqtiydixvnumyvei@ava.usersys.com> References: <20220203214339.1889971-1-atomlin@redhat.com> <20220217124729.GA743618@lothringen> MIME-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Disposition: inline In-Reply-To: <20220217124729.GA743618@lothringen> X-CMAE-Envelope: MS4xfALsXlmYPNYwRtA/HuOcvZPn1r7QNGWJmyaFB2jLdhKjMRjmXVEu06hr1Y1VTV7KVvhxQGvNTbhGxBT/b7GFnuZprsaNclwWnRjHGWXo7TdfeYlwDDCW n4gBJuOudrTF8//Q7rmQfKDDT9Jsr7iZHy3e0qwesX7ssS+kVI3gZpM/ICK69Fd8hXprls7dOo43pVMyYzlneeBMWsiYmV8ci1+zId7cR/0VTbtsXe+ui2kY v2HQbf1/TJW2mrIj+eFlfPdGPl892bHVAc0nVS2Vtf+PWA67YMJYwpUSTovtZsjnoKb6g3A81Kww0dYxL4xYIU/XPVUHwW3UWC9//hxIJgV8QFg2dFPQdFOj VezZNgRasbvUJPWTFysdMeizHDEl3cwhYqI1j1zUD80il+clWZMmCCriKQMjG3JJEPcGNze+ X-Stat-Signature: wob3twnfyfanxd4heeut6ko7wf83za58 X-Rspamd-Server: rspam02 X-Rspamd-Queue-Id: A05CF20012 Authentication-Results: imf31.hostedemail.com; dkim=none; dmarc=none; spf=pass (imf31.hostedemail.com: domain of atomlin@atomlin.com designates 173.201.192.231 as permitted sender) smtp.mailfrom=atomlin@atomlin.com X-Rspam-User: X-HE-Tag: 1645107978-200043 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: On Thu 2022-02-17 13:47 +0100, Frederic Weisbecker wrote: > So, to make sure I understand, the issue is that with nohz_full, we may > well enter into the idle loop with the tick already stopped. We may also > exit from idle without restarting the tick (again only with nohz_full). And > so this can cause the vmstat to not be flushed upon idle entry. Right? Hi Frederic, Yes - this is exactly it. > > A customer provided some evidence which indicates that the idle tick was > > stopped; albeit, CPU-specific vmstat counters still remained populated. > > Thus one can only assume quiet_vmstat() was not invoked on return to the > > idle loop. > > > > Unfortunately, I suspect this divergence might erroneously prevent a > > reclaim attempt by kswapd. If the number of zone specific free pages are > > below their per-cpu drift value then zone_page_state_snapshot() is used to > > compute a more accurate view of the aforementioned statistic. > > Thus any task blocked on the NUMA node specific pfmemalloc_wait queue will > > be unable to make significant progress via direct reclaim unless it is > > killed after being woken up by kswapd (see throttle_direct_reclaim()). > > That being said, eventually reclaim should give up if the conditions are > > correct, no? > Now if quiet_vmstat() isn't called, the vmstat_work should fix this later, > right? Or does that happen too late perhaps? If I understand correctly, in the context of nohz_full, since such work is deferred, it will only be handled in a scenario when the periodic/or scheduling-clock tick is enabled i.e. the timer was reprogrammed on exit from idle. Kind regards, -- Aaron Tomlin