From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1757436Ab2BWXW1 (ORCPT ); Thu, 23 Feb 2012 18:22:27 -0500 Received: from mail.linuxfoundation.org ([140.211.169.12]:45834 "EHLO mail.linuxfoundation.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1755884Ab2BWXWZ (ORCPT ); Thu, 23 Feb 2012 18:22:25 -0500 Date: Thu, 23 Feb 2012 15:22:23 -0800 From: Andrew Morton To: Tejun Heo Cc: vgoyal@redhat.com, axboe@kernel.dk, hughd@google.com, avi@redhat.com, nate@cpanel.net, cl@linux-foundation.org, linux-kernel@vger.kernel.org, dpshah@google.com, ctalbott@google.com, rni@google.com Subject: Re: [PATCHSET] mempool, percpu, blkcg: fix percpu stat allocation and remove stats_lock Message-Id: <20120223152223.43c72ccb.akpm@linux-foundation.org> In-Reply-To: <20120223231204.GM22536@google.com> References: <1330036246-21633-1-git-send-email-tj@kernel.org> <20120223144336.58742e1b.akpm@linux-foundation.org> <20120223230123.GL22536@google.com> <20120223231204.GM22536@google.com> X-Mailer: Sylpheed 3.0.2 (GTK+ 2.20.1; x86_64-pc-linux-gnu) Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Thu, 23 Feb 2012 15:12:04 -0800 Tejun Heo wrote: > On Thu, Feb 23, 2012 at 03:01:23PM -0800, Tejun Heo wrote: > > Hmmm... going through the thread again, ah, okay, I forgot about that > > completely. Yeah, that is an actual problem. Both __GFP_WAIT which > > isn't GFP_KERNEL and GFP_KERNEL are valid use cases. I guess we'll be > > building async percpu pool in blkcg then. Great. :( > > Vivek, you win. :) Can you please refresh the async alloc patch on top > of blkcg-stacking branch? I'll rool that into this series and drop > the mempool stuff. I forget how those patches work, but they might be vulnerable to the same issue. If the block layer can handle the failed allocation attempt and retry at the next I/O event then I guess that would be acceptable; we'd lose a bit of statistical info occasionally, but who cares. But ISTR that we can't handle allocation failures here?