From mboxrd@z Thu Jan 1 00:00:00 1970 From: "Martin K. Petersen" Subject: Re: [PATCH 4/6] block: Optionally snapshot page contents to provide stable pages during write Date: Wed, 16 Jan 2013 22:26:36 -0500 Message-ID: References: <20130115054235.1563.12967.stgit@blackbox.djwong.org> <20130115054305.1563.48197.stgit@blackbox.djwong.org> <20130116020000.GA25806@quack.suse.cz> <20130117030132.GK6426@blackbox.djwong.org> Mime-Version: 1.0 Content-Type: text/plain Cc: Jan Kara , axboe@kernel.dk, lucho@ionkov.net, ericvh@gmail.com, tytso@mit.edu, viro@zeniv.linux.org.uk, rminnich@sandia.gov, martin.petersen@oracle.com, neilb@suse.de, david@fromorbit.com, gnehzuil.liu@gmail.com, linux-kernel@vger.kernel.org, hch@infradead.org, linux-fsdevel@vger.kernel.org, Andy Lutomirski , adilger.kernel@dilger.ca, bharrosh@panasas.com, jlayton@samba.org, linux-ext4@vger.kernel.org, hirofumi@mail.parknet.co.jp To: "Darrick J. Wong" Return-path: In-Reply-To: <20130117030132.GK6426@blackbox.djwong.org> (Darrick J. Wong's message of "Wed, 16 Jan 2013 19:01:32 -0800") Sender: linux-kernel-owner@vger.kernel.org List-Id: linux-ext4.vger.kernel.org >>>>> "Darrick" == Darrick J Wong writes: >> Umm, why did you move this hunk? Darrick> I moved it so that the integrity data are generated against the Darrick> contents of the bounce buffer because the write paths don't Darrick> wait for writeback to finish if the snapshotting mode is Darrick> enabled, which means (I think) that programs can wander in and Darrick> scribble on the original page in between bio_integrity_prep and Darrick> blk_queue_bounce. It's also important that bio_integrity_prep() is run after partition remapping has taken place. But it looks like that's still the case after your change. -- Martin K. Petersen Oracle Linux Engineering