From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1756837AbZHFVoh (ORCPT ); Thu, 6 Aug 2009 17:44:37 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1756779AbZHFVoh (ORCPT ); Thu, 6 Aug 2009 17:44:37 -0400 Received: from bombadil.infradead.org ([18.85.46.34]:43809 "EHLO bombadil.infradead.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1756393AbZHFVog (ORCPT ); Thu, 6 Aug 2009 17:44:36 -0400 Date: Thu, 6 Aug 2009 17:44:33 -0400 From: Christoph Hellwig To: Jan Kara Cc: Jens Axboe , linux-kernel@vger.kernel.org, linux-fsdevel@vger.kernel.org, chris.mason@oracle.com, david@fromorbit.com, hch@infradead.org, akpm@linux-foundation.org, yanmin_zhang@linux.intel.com, richard@rsk.demon.co.uk, damien.wyart@free.fr, fweisbec@gmail.com, Alan.Brunelle@hp.com Subject: Re: [PATCH 2/9] writeback: switch to per-bdi threads for flushing data Message-ID: <20090806214433.GA866@infradead.org> References: <1248989044-21605-1-git-send-email-jens.axboe@oracle.com> <1248989044-21605-3-git-send-email-jens.axboe@oracle.com> <20090805163532.GC27505@duck.suse.cz> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20090805163532.GC27505@duck.suse.cz> User-Agent: Mutt/1.5.19 (2009-01-05) X-SRS-Rewrite: SMTP reverse-path rewritten from by bombadil.infradead.org See http://www.infradead.org/rpr.html Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Wed, Aug 05, 2009 at 06:35:32PM +0200, Jan Kara wrote: > Jens, sorry about bitching about this again but you're silently changing > substantial locking assumptions without writing it *anywhere* and without > arguing it's safe. > Originally, generic_sync_sb_inodes() from writeback path have been > called with > a) s_umount_sem held > b) sb->s_count elevated > The second still seems to be true since, if I'm right, we pass here > non-NULL sb only from sync_filesystem() and that takes care of the > superblock reference. So that is just a matter of documenting this fact > before the function. We'll defintively need to keep both to prevent races vs unmount. And with a NULL superblock passed I'm not even sure how we can take care of it.