From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1750915Ab2BOPnz (ORCPT ); Wed, 15 Feb 2012 10:43:55 -0500 Received: from mx1.redhat.com ([209.132.183.28]:42566 "EHLO mx1.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751433Ab2BOPnu (ORCPT ); Wed, 15 Feb 2012 10:43:50 -0500 Date: Wed, 15 Feb 2012 10:43:41 -0500 From: Vivek Goyal To: Tejun Heo Cc: Andrew Morton , KAMEZAWA Hiroyuki , avi@redhat.com, nate@cpanel.net, cl@linux-foundation.org, oleg@redhat.com, axboe@kernel.dk, linux-kernel@vger.kernel.org Subject: Re: [PATCHSET] block, mempool, percpu: implement percpu mempool and fix blkcg percpu alloc deadlock Message-ID: <20120215154341.GB27312@redhat.com> References: <20111222175834.66559c8b.akpm@linux-foundation.org> <20111223025629.GE12738@redhat.com> <20111226150531.3c22f2f0.kamezawa.hiroyu@jp.fujitsu.com> <20111227175249.GE17712@google.com> <20111228091402.c45a08f6.kamezawa.hiroyu@jp.fujitsu.com> <20111228004102.GK17712@google.com> <20120105012842.GQ31746@google.com> <20120209235845.GH19392@google.com> <20120210162658.GC3150@redhat.com> <20120213223156.GG12117@google.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20120213223156.GG12117@google.com> 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, Feb 13, 2012 at 02:31:56PM -0800, Tejun Heo wrote: > Hello, Vivek. > > On Fri, Feb 10, 2012 at 11:26:58AM -0500, Vivek Goyal wrote: > > The only difference is that by putting this logic in per cpu counters, > > we make it somewhat generic so that other users who can't do GFP_KERNEL > > allocation of per cpu data, can use it. I can live with that. > > Also, it has fallback mechanism while percpu data isn't there, so the > counts are guaranteed to be right. Probably doesn't matter all that > much for blkcg stats. Ok, I had missed that part. Nice. I have no objections to making this lazy allocation in per cpu counter idea work and migrate blkcg per cpu stats to use per cpu counters. If this patchset is not too intrusive, it might even be backportable to stable kernels. Andrew does not seem to like mempool idea, so instead of not making any progress, it is better to go with this idea. We need to fix this issue. More poeple might run into this. Thanks Vivek