From mboxrd@z Thu Jan  1 00:00:00 1970
From: Mike Snitzer <snitzer@redhat.com>
Subject: Re: [PATCH 23/41] dm: implement REQ_FLUSH/FUA support for bio-based
 dm
Date: Tue, 7 Sep 2010 17:17:24 -0400
Message-ID: <20100907211724.GA22525@redhat.com>
References: <1283509796-1510-1-git-send-email-tj@kernel.org>
 <1283509796-1510-24-git-send-email-tj@kernel.org>
 <4C84CD18.6060906@redhat.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Cc: device-mapper development <dm-devel@redhat.com>,
	Tejun Heo <tj@kernel.org>, jaxboe@fusionio.com,
	linux-kernel@vger.kernel.org, linux-fsdevel@vger.kernel.org,
	linux-scsi@vger.kernel.org, linux-ide@vger.kernel.org,
	linux-raid@vger.kernel.org, hch@lst.de,
	konishi.ryusuke@lab.ntt.co.jp, James.Bottomley@suse.de,
	tytso@mit.edu, chris.mason@oracle.com, swhiteho@redhat.com,
	vst@vlnb.net, jack@suse.cz, rwheeler@redhat.com, hare@suse.de,
	neilb@suse.de, rusty@rustcorp.com.au, mst@redhat.com,
	k-ueda@ct.jp.nec.com, mpatocka@redhat.com
To: Milan Broz <mbroz@redhat.com>
Return-path: <linux-ide-owner@vger.kernel.org>
Content-Disposition: inline
In-Reply-To: <4C84CD18.6060906@redhat.com>
Sender: linux-ide-owner@vger.kernel.org
List-Id: linux-fsdevel.vger.kernel.org

On Mon, Sep 06 2010 at  7:14am -0400,
Milan Broz <mbroz@redhat.com> wrote:

> On 09/03/2010 12:29 PM, Tejun Heo wrote:
> 
> > +++ b/drivers/md/dm-crypt.c
> > @@ -1278,7 +1278,7 @@ static int crypt_map(struct dm_target *ti, struct bio *bio,
> >  	struct dm_crypt_io *io;
> >  	struct crypt_config *cc;
> >  
> > -	if (unlikely(bio_empty_barrier(bio))) {
> > +	if (bio->bi_rw & REQ_FLUSH) {
> >  		cc = ti->private;
> >  		bio->bi_bdev = cc->dev->bdev;
> >  		return DM_MAPIO_REMAPPED;
> 
> ...
> 
> > +++ b/drivers/md/dm.c
> > @@ -1400,14 +1391,22 @@ static void __split_and_process_bio(struct mapped_device *md, struct bio *bio)
> >  	ci.io->md = md;
> >  	spin_lock_init(&ci.io->endio_lock);
> >  	ci.sector = bio->bi_sector;
> > -	ci.sector_count = bio_sectors(bio);
> > -	if (unlikely(bio_empty_barrier(bio)))
> > +	if (!(bio->bi_rw & REQ_FLUSH))
> > +		ci.sector_count = bio_sectors(bio);
> > +	else {
> > +		/* all FLUSH bio's reaching here should be empty */
> > +		WARN_ON_ONCE(bio_has_data(bio));
> >  		ci.sector_count = 1;
> > +	}
> 
> 
> I would add BUG_ON(bio_has_data(bio)) either to dm-crypt target or directly to DM core
> in this path.

I agree, that WARN_ON_ONCE should be changed to BUG_ON.  This is a
guarantee that the block layer now provides so it seems correct to have
the DM core bug if that guarantee isn't actually provided.
 
> Note that empty barrier request bypass encryption layer now in dm-crypt, so if some bio
> with data payload reach it after the change, it causes data corruption
> (moreover plain data reach the disk directly).

Given the consequences, it wouldn't hurt to BUG_ON() in dm-crypt too.
It's redundant if the DM core will also BUG_ON() but it serves as a
dm-crypt safety-net/documentation.

Mike