From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1751683AbeDVNIl (ORCPT ); Sun, 22 Apr 2018 09:08:41 -0400 Received: from mx2.suse.de ([195.135.220.15]:53290 "EHLO mx2.suse.de" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751296AbeDVNIk (ORCPT ); Sun, 22 Apr 2018 09:08:40 -0400 Date: Sun, 22 Apr 2018 07:08:36 -0600 From: Michal Hocko To: Tetsuo Handa Cc: rientjes@google.com, akpm@linux-foundation.org, aarcange@redhat.com, guro@fb.com, linux-kernel@vger.kernel.org, linux-mm@kvack.org Subject: Re: [patch v2] mm, oom: fix concurrent munlock and oom reaperunmap Message-ID: <20180422130836.GH17484@dhcp22.suse.cz> References: <20180419063556.GK17484@dhcp22.suse.cz> <20180420082349.GW17484@dhcp22.suse.cz> <20180420124044.GA17484@dhcp22.suse.cz> <201804221248.CHE35432.FtOMOLSHOFJFVQ@I-love.SAKURA.ne.jp> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <201804221248.CHE35432.FtOMOLSHOFJFVQ@I-love.SAKURA.ne.jp> User-Agent: Mutt/1.9.4 (2018-02-28) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Sun 22-04-18 12:48:13, Tetsuo Handa wrote: > David Rientjes wrote: > > How have you tested this? > > > > I'm wondering why you do not see oom killing of many processes if the > > victim is a very large process that takes a long time to free memory in > > exit_mmap() as I do because the oom reaper gives up trying to acquire > > mm->mmap_sem and just sets MMF_OOM_SKIP itself. > > > > We can call __oom_reap_task_mm() from exit_mmap() (or __mmput()) before > exit_mmap() holds mmap_sem for write. Then, at least memory which could > have been reclaimed if exit_mmap() did not hold mmap_sem for write will > be guaranteed to be reclaimed before MMF_OOM_SKIP is set. That might be an alternative way but I am really wondering whether this is the simplest way forward. -- Michal Hocko SUSE Labs