From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mail-ed1-f69.google.com (mail-ed1-f69.google.com [209.85.208.69]) by kanga.kvack.org (Postfix) with ESMTP id 186018E0002 for ; Thu, 3 Jan 2019 13:13:33 -0500 (EST) Received: by mail-ed1-f69.google.com with SMTP id 39so34313059edq.13 for ; Thu, 03 Jan 2019 10:13:33 -0800 (PST) Received: from mx1.suse.de (mx2.suse.de. [195.135.220.15]) by mx.google.com with ESMTPS id u16si964396edm.213.2019.01.03.10.13.31 for (version=TLS1_2 cipher=ECDHE-RSA-AES128-GCM-SHA256 bits=128/128); Thu, 03 Jan 2019 10:13:31 -0800 (PST) Date: Thu, 3 Jan 2019 19:13:29 +0100 From: Michal Hocko Subject: Re: [RFC PATCH 0/3] mm: memcontrol: delayed force empty Message-ID: <20190103181329.GW31793@dhcp22.suse.cz> References: <1546459533-36247-1-git-send-email-yang.shi@linux.alibaba.com> <20190103101215.GH31793@dhcp22.suse.cz> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: Sender: owner-linux-mm@kvack.org List-ID: To: Yang Shi Cc: hannes@cmpxchg.org, akpm@linux-foundation.org, linux-mm@kvack.org, linux-kernel@vger.kernel.org On Thu 03-01-19 09:33:14, Yang Shi wrote: > > > On 1/3/19 2:12 AM, Michal Hocko wrote: > > On Thu 03-01-19 04:05:30, Yang Shi wrote: > > > Currently, force empty reclaims memory synchronously when writing to > > > memory.force_empty. It may take some time to return and the afterwards > > > operations are blocked by it. Although it can be interrupted by signal, > > > it still seems suboptimal. > > Why it is suboptimal? We are doing that operation on behalf of the > > process requesting it. What should anybody else pay for it? In other > > words why should we hide the overhead? > > Please see the below explanation. > > > > > > Now css offline is handled by worker, and the typical usecase of force > > > empty is before memcg offline. So, handling force empty in css offline > > > sounds reasonable. > > Hmm, so I guess you are talking about > > echo 1 > $MEMCG/force_empty > > rmdir $MEMCG > > > > and you are complaining that the operation takes too long. Right? Why do > > you care actually? > > We have some usecases which create and remove memcgs very frequently, and > the tasks in the memcg may just access the files which are unlikely accessed > by anyone else. So, we prefer force_empty the memcg before rmdir'ing it to > reclaim the page cache so that they don't get accumulated to incur > unnecessary memory pressure. Since the memory pressure may incur direct > reclaim to harm some latency sensitive applications. Yes, this makes sense to me. > And, the create/remove might be run in a script sequentially (there might be > a lot scripts or applications are run in parallel to do this), i.e. > mkdir cg1 > do something > echo 0 > cg1/memory.force_empty > rmdir cg1 > > mkdir cg2 > ... > > The creation of the afterwards memcg might be blocked by the force_empty for > long time if there are a lot page caches, so the overall throughput of the > system may get hurt. Is there any reason for your scripts to be strictly sequential here? In other words why cannot you offload those expensive operations to a detached context in _userspace_? -- Michal Hocko SUSE Labs