linux-kernel - Re: [PATCH 07/10] block: kill merge_bvec

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <20150429080919.342fddfd@notabene.brown>
Date:	Wed, 29 Apr 2015 08:09:19 +1000
From:	NeilBrown <neilb@...e.de>
To:	Ming Lin <mlin@...nel.org>
Cc:	linux-kernel@...r.kernel.org, Christoph Hellwig <hch@....de>,
	Jens Axboe <axboe@...nel.dk>,
	Kent Overstreet <kent.overstreet@...il.com>,
	Dongsu Park <dongsu.park@...fitbricks.com>,
	Lars Ellenberg <drbd-dev@...ts.linbit.com>,
	drbd-user@...ts.linbit.com, Jiri Kosina <jkosina@...e.cz>,
	Yehuda Sadeh <yehuda@...tank.com>,
	Sage Weil <sage@...tank.com>, Alex Elder <elder@...nel.org>,
	ceph-devel@...r.kernel.org, Alasdair Kergon <agk@...hat.com>,
	Mike Snitzer <snitzer@...hat.com>, dm-devel@...hat.com,
	linux-raid@...r.kernel.org, Christoph Hellwig <hch@...radead.org>,
	"Martin K. Petersen" <martin.petersen@...cle.com>
Subject: Re: [PATCH 07/10] block: kill merge_bvec_fn() completely

On Mon, 27 Apr 2015 23:48:34 -0700 Ming Lin <mlin@...nel.org> wrote:

> From: Kent Overstreet <kent.overstreet@...il.com>
> 
> As generic_make_request() is now able to handle arbitrarily sized bios,
> it's no longer necessary for each individual block driver to define its
> own ->merge_bvec_fn() callback. Remove every invocation completely.

This patch it just a little premature I think.

md/raid5 still assumes read requests will mostly fit within a single chunk
(which merge_bvec_fn encourages) so they can be serviced without using the
stripe-cache.
You've just broken that assumption.

I think 'chunk_aligned_read' needs to get a loop using bio_split, a bit like
raid0, first.

Thanks,
NeilBrown

Content of type "application/pgp-signature" skipped