From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1752946AbaBCQo5 (ORCPT ); Mon, 3 Feb 2014 11:44:57 -0500 Received: from cantor2.suse.de ([195.135.220.15]:36230 "EHLO mx2.suse.de" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752189AbaBCQo4 (ORCPT ); Mon, 3 Feb 2014 11:44:56 -0500 Date: Mon, 3 Feb 2014 17:44:55 +0100 From: Michal Hocko To: Johannes Weiner Cc: Andrew Morton , KAMEZAWA Hiroyuki , KOSAKI Motohiro , LKML , linux-mm@kvack.org Subject: Re: [RFC 4/5] memcg: make sure that memcg is not offline when charging Message-ID: <20140203164455.GL2495@dhcp22.suse.cz> References: <1387295130-19771-1-git-send-email-mhocko@suse.cz> <1387295130-19771-5-git-send-email-mhocko@suse.cz> <20140130172906.GE6963@cmpxchg.org> <20140203133313.GF2495@dhcp22.suse.cz> <20140203161823.GJ6963@cmpxchg.org> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20140203161823.GJ6963@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 Mon 03-02-14 11:18:23, Johannes Weiner wrote: > On Mon, Feb 03, 2014 at 02:33:13PM +0100, Michal Hocko wrote: > > On Thu 30-01-14 12:29:06, Johannes Weiner wrote: > > > On Tue, Dec 17, 2013 at 04:45:29PM +0100, Michal Hocko wrote: [...] > > > > In order to make this raceless we would need to hold rcu_read_lock since > > > > css_tryget until res_counter_charge. This is not so easy unfortunately > > > > because mem_cgroup_do_charge might sleep so we would need to do drop rcu > > > > lock and do css_tryget tricks after each reclaim. > > > > > > Yes, why not? > > > > Although css_tryget is cheap these days I thought that a simple flag > > check would be even heaper in this hot path. Changing the patch to use > > css_tryget rather than offline check is trivial if you really think it > > is better? > > You already changed it to do css_tryget() on every single charge. Fair point Thanks! -- Michal Hocko SUSE Labs