From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1753977Ab0CCMJc (ORCPT ); Wed, 3 Mar 2010 07:09:32 -0500 Received: from 0122700014.0.fullrate.dk ([95.166.99.235]:40685 "EHLO kernel.dk" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753876Ab0CCMJ2 (ORCPT ); Wed, 3 Mar 2010 07:09:28 -0500 Date: Wed, 3 Mar 2010 13:09:27 +0100 From: Jens Axboe To: John Hughes Cc: Andrew Morton , bugzilla-daemon@bugzilla.kernel.org, bugme-daemon@bugzilla.kernel.org, linux-kernel@vger.kernel.org Subject: Re: [Bugme-new] [Bug 15426] New: Running many copies of bonnie++ on different filesystems seems to deadlock in sync Message-ID: <20100303120927.GQ5768@kernel.dk> References: <20100302161627.bf02d0a2.akpm@linux-foundation.org> <4B8E517E.3010702@Calva.COM> <20100303115015.GN5768@kernel.dk> <4B8E5805.30505@Calva.COM> <20100303120317.GP5768@kernel.dk> <4B8E59CC.2030004@Calva.COM> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <4B8E59CC.2030004@Calva.COM> Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Wed, Mar 03 2010, John Hughes wrote: > Jens Axboe wrote: >> On Wed, Mar 03 2010, John Hughes wrote: >> >>> Just a thought. What exactly is sync(2) supposed to do - block until >>> there are no more dirty pages, or block until all pages that were >>> dirty when the sync was done are clean? In other words is the >>> problem simply that pages are being dirtied faster than the sync is >>> writing them out? >>> >> >> Our sync is currently broken in that regard, since it'll wait for too >> long. We have a debated patch going, I have included it below. Any >> chance you could give it a whirl? >> >> The semantics of sync are supposed to be 'wait for dirty IO generated >> BEFORE this sync call'. >> > Ok, I will try this patch and get back to you. Looking at your bug report, I saw that you were on 2.6.32. If you are testing 2.6.33, let me know, the patch wont apply (I have another variant for that). -- Jens Axboe