From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1753821AbaGWM6Z (ORCPT ); Wed, 23 Jul 2014 08:58:25 -0400 Received: from cantor2.suse.de ([195.135.220.15]:39244 "EHLO mx2.suse.de" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751049AbaGWM6Y (ORCPT ); Wed, 23 Jul 2014 08:58:24 -0400 Date: Wed, 23 Jul 2014 13:58:18 +0100 From: Mel Gorman To: Johannes Weiner Cc: Andrew Morton , Michal Hocko , Linux Kernel , Linux-MM Subject: Re: [PATCH 1/2] mm: Avoid full RCU lookup of memcg for statistics updates Message-ID: <20140723125818.GT10819@suse.de> References: <1406114656-16350-1-git-send-email-mgorman@suse.de> <1406114656-16350-2-git-send-email-mgorman@suse.de> <20140723114449.GE1725@cmpxchg.org> MIME-Version: 1.0 Content-Type: text/plain; charset=iso-8859-15 Content-Disposition: inline In-Reply-To: <20140723114449.GE1725@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 Wed, Jul 23, 2014 at 07:44:49AM -0400, Johannes Weiner wrote: > On Wed, Jul 23, 2014 at 12:24:15PM +0100, Mel Gorman wrote: > > When updating memcg VM statistics like PGFAULT we take the rcu read > > lock and lookup the memcg. For statistic updates this is overkill > > when the process may not belong to a memcg. This patch adds a light > > check to check if a memcg potentially exists. It's race-prone in that > > some VM stats may be missed when a process first joins a memcg but > > that is not serious enough to justify a constant performance penalty. > > Tasks always belong to a memcg, the root group per default. There > isn't really any accounting that could be omitted. > Crap, ok, I had not taken that into account. The lookup of that cannot really be avoided. It's a pity because the stats on the root memcg are not likely to be that interesting. Thanks for reviewing. -- Mel Gorman SUSE Labs