From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1752765Ab1GLJsp (ORCPT ); Tue, 12 Jul 2011 05:48:45 -0400 Received: from cantor2.suse.de ([195.135.220.15]:33244 "EHLO mx2.suse.de" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752424Ab1GLJso (ORCPT ); Tue, 12 Jul 2011 05:48:44 -0400 Date: Tue, 12 Jul 2011 11:48:41 +0200 From: Michal Hocko To: Hiroyuki Kamezawa Cc: KAMEZAWA Hiroyuki , Christoph Hellwig , linux-mm@kvack.org, linux-kernel@vger.kernel.org, Andrew Morton , Hugh Dickins , Rik van Riel , Michel Lespinasse , Mel Gorman , Lutz Vieweg Subject: Re: [PATCH] mm: preallocate page before lock_page at filemap COW. (WasRe: [PATCH V2] mm: Do not keep page locked during page fault while charging it for memcg Message-ID: <20110712094841.GC10552@tiehlicka.suse.cz> References: <20110622120635.GB14343@tiehlicka.suse.cz> <20110622121516.GA28359@infradead.org> <20110622123204.GC14343@tiehlicka.suse.cz> <20110623150842.d13492cd.kamezawa.hiroyu@jp.fujitsu.com> <20110623074133.GA31593@tiehlicka.suse.cz> <20110623170811.16f4435f.kamezawa.hiroyu@jp.fujitsu.com> <20110623090204.GE31593@tiehlicka.suse.cz> <20110623190157.1bc8cbb9.kamezawa.hiroyu@jp.fujitsu.com> <20110624075742.GA10455@tiehlicka.suse.cz> 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 On Fri 24-06-11 20:46:29, Hiroyuki Kamezawa wrote: > 2011/6/24 Michal Hocko : > > Sorry, forgot to send my > > Reviewed-by: Michal Hocko > > > > Thanks. > > > I still have concerns about this way to handle the issue. See the follow > > up discussion in other thread (https://lkml.org/lkml/2011/6/23/135). > > > > Anyway I think that we do not have many other options to handle this. > > Either we unlock, charge, lock&restes or we preallocate, fault in > > > I agree. > > > Or am I missing some other ways how to do it? What do others think about > > these approaches? > > > > Yes, I'd like to hear other mm specialists' suggestion. and I'll think > other way, again. > Anyway, memory reclaim with holding a lock_page() can cause big latency > or starvation especially when memcg is used. It's better to avoid it. Is there any intereset in discussing this or the email just got lost? Just for reference preallocation patch from Kamezawa is already in the Andrew's tree. -- Michal Hocko SUSE Labs SUSE LINUX s.r.o. Lihovarska 1060/12 190 00 Praha 9 Czech Republic