From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1752919AbeDETpR (ORCPT ); Thu, 5 Apr 2018 15:45:17 -0400 Received: from mail-yb0-f174.google.com ([209.85.213.174]:43674 "EHLO mail-yb0-f174.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752822AbeDETpP (ORCPT ); Thu, 5 Apr 2018 15:45:15 -0400 X-Google-Smtp-Source: AIpwx48EYbtC1IXqn3ZhxWYvb8mPZHC/rlzRYLB2Xqmib+b6hIb798FtdiGJntJDsJkbKz0blPXqLg== Date: Thu, 5 Apr 2018 12:45:12 -0700 From: Tejun Heo To: Johannes Weiner Cc: Michal Hocko , vdavydov.dev@gmail.com, guro@fb.com, riel@surriel.com, akpm@linux-foundation.org, linux-kernel@vger.kernel.org, kernel-team@fb.com, cgroups@vger.kernel.org, linux-mm@kvack.org Subject: Re: [PATCH] mm: memcg: make sure memory.events is uptodate when waking pollers Message-ID: <20180405194512.GD3126663@devbig577.frc2.facebook.com> References: <20180324160901.512135-1-tj@kernel.org> <20180324160901.512135-2-tj@kernel.org> <20180404140855.GA28966@cmpxchg.org> <20180404141850.GC28966@cmpxchg.org> <20180404143447.GJ6312@dhcp22.suse.cz> <20180404165829.GA3126663@devbig577.frc2.facebook.com> <20180405175507.GA24817@cmpxchg.org> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20180405175507.GA24817@cmpxchg.org> User-Agent: Mutt/1.5.21 (2010-09-15) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Thu, Apr 05, 2018 at 01:55:16PM -0400, Johannes Weiner wrote: > From 4369ce161a9085aa408f2eca54f9de72909ee1b1 Mon Sep 17 00:00:00 2001 > From: Johannes Weiner > Date: Thu, 5 Apr 2018 11:53:55 -0400 > Subject: [PATCH] mm: memcg: make sure memory.events is uptodate when waking > pollers > > a983b5ebee57 ("mm: memcontrol: fix excessive complexity in memory.stat > reporting") added per-cpu drift to all memory cgroup stats and events > shown in memory.stat and memory.events. > > For memory.stat this is acceptable. But memory.events issues file > notifications, and somebody polling the file for changes will be > confused when the counters in it are unchanged after a wakeup. > > Luckily, the events in memory.events - MEMCG_LOW, MEMCG_HIGH, > MEMCG_MAX, MEMCG_OOM - are sufficiently rare and high-level that we > don't need per-cpu buffering for them: MEMCG_HIGH and MEMCG_MAX would > be the most frequent, but they're counting invocations of reclaim, > which is a complex operation that touches many shared cachelines. > > This splits memory.events from the generic VM events and tracks them > in their own, unbuffered atomic counters. That's also cleaner, as it > eliminates the ugly enum nesting of VM and cgroup events. > > Fixes: a983b5ebee57 ("mm: memcontrol: fix excessive complexity in memory.stat reporting") > Reported-by: Tejun Heo > Signed-off-by: Johannes Weiner Yeah, that works. FWIW, Acked-by: Tejun Heo Thanks. -- tejun