From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1752697AbaBMA27 (ORCPT ); Wed, 12 Feb 2014 19:28:59 -0500 Received: from mail-qa0-f45.google.com ([209.85.216.45]:63856 "EHLO mail-qa0-f45.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751381AbaBMA25 (ORCPT ); Wed, 12 Feb 2014 19:28:57 -0500 Date: Wed, 12 Feb 2014 19:28:53 -0500 From: Tejun Heo To: Hugh Dickins Cc: Michal Hocko , Johannes Weiner , Filipe Brandenburger , Li Zefan , Andrew Morton , Greg Thelen , Michel Lespinasse , Markus Blank-Burian , Shawn Bohrer , cgroups@vger.kernel.org, linux-kernel@vger.kernel.org, linux-mm@kvack.org Subject: Re: [PATCH 2/2] cgroup: bring back kill_cnt to order css destruction Message-ID: <20140213002853.GC2916@htj.dyndns.org> References: <20140207164321.GE6963@cmpxchg.org> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: 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 Hello, Hugh. On Wed, Feb 12, 2014 at 03:06:26PM -0800, Hugh Dickins wrote: > Sometimes the cleanup after memcg hierarchy testing gets stuck in > mem_cgroup_reparent_charges(), unable to bring non-kmem usage down to 0. > > There may turn out to be several causes, but a major cause is this: the > workitem to offline parent can get run before workitem to offline child; > parent's mem_cgroup_reparent_charges() circles around waiting for the > child's pages to be reparented to its lrus, but it's holding cgroup_mutex > which prevents the child from reaching its mem_cgroup_reparent_charges(). > > Further testing showed that an ordered workqueue for cgroup_destroy_wq > is not always good enough: percpu_ref_kill_and_confirm's call_rcu_sched > stage on the way can mess up the order before reaching the workqueue. > > Instead bring back v3.11's css kill_cnt, repurposing it to make sure > that offline_css() is not called for parent before it has been called > for all children. > > Fixes: e5fca243abae ("cgroup: use a dedicated workqueue for cgroup destruction") > Signed-off-by: Hugh Dickins > Reviewed-by: Filipe Brandenburger > Cc: stable@vger.kernel.org # v3.10+ (but will need extra care) > --- > This is an alternative to Filipe's 1/2: there's no need for both, > but each has its merits. I prefer Filipe's, which is much easier to > understand: this one made more sense in v3.11, when it was just a matter > of extending the use of css_kill_cnt; but might be preferred if offlining > children before parent is thought to be a good idea generally. Not that your implementation is bad or anything but the patch itself somehow makes me cringe a bit. It's probably just because it has to add to the already overly complicated offline path. Guaranteeing strict offline ordering might be a good idea but at least for the immediate bug fix, I agree that the memcg specific fix seems better suited. Let's apply that one and reconsider this one if it turns out we do need strict offline reordering. Thanks a lot! -- tejun