Messages in this thread Patch in this message |  | | | From | Dmitry Monakhov <> | | Subject | Re: [PATCH 1/2] blkdev: fix merge_bvec_fn return value checks | | Date | Wed, 03 Mar 2010 06:49:43 +0300 |
| |
Jens Axboe <jens.axboe@oracle.com> writes:
> On Sat, Feb 27 2010, Dmitry Monakhov wrote: >> merge_bvec_fn() returns bvec->bv_len on success. So we have to check >> against this value. But in case of fs_optimization merge we compare >> with wrong value. This patch must be included in >> b428cd6da7e6559aca69aa2e3a526037d3f20403 >> But accidentally i've forgot to add this in the initial patch. >> To make things straight let's replace all such checks. >> In fact this makes code easy to understand. > > Agree, applied. Ohh.. as you already know this patch break dm-layer. Sorry. This is because dm->merge may return more than requested. So correct check must test against less what requested. Correct patch attached.
From 145fb49bf2251f445ca29c5218333367448932d6 Mon Sep 17 00:00:00 2001 From: Dmitry Monakhov <dmonakhov@openvz.org> Date: Wed, 3 Mar 2010 06:28:06 +0300 Subject: [PATCH] blkdev: fix merge_bvec_fn return value checks v2 merge_bvec_fn() returns bvec->bv_len on success. So we have to check against this value. But in case of fs_optimization merge we compare with wrong value. This patch must be included in b428cd6da7e6559aca69aa2e3a526037d3f20403 But accidentally i've forgot to add this in the initial patch. To make things straight let's replace all such checks. In fact this makes code easy to understand.
Signed-off-by: Dmitry Monakhov <dmonakhov@openvz.org> --- fs/bio.c | 4 ++-- 1 files changed, 2 insertions(+), 2 deletions(-) diff --git a/fs/bio.c b/fs/bio.c index 88094af..975657a 100644 --- a/fs/bio.c +++ b/fs/bio.c @@ -557,7 +557,7 @@ static int __bio_add_page(struct request_queue *q, struct bio *bio, struct page .bi_rw = bio->bi_rw, }; - if (q->merge_bvec_fn(q, &bvm, prev) < len) { + if (q->merge_bvec_fn(q, &bvm, prev) < prev->bv_len) { prev->bv_len -= len; return 0; } @@ -611,7 +611,7 @@ static int __bio_add_page(struct request_queue *q, struct bio *bio, struct page * merge_bvec_fn() returns number of bytes it can accept * at this offset */ - if (q->merge_bvec_fn(q, &bvm, bvec) < len) { + if (q->merge_bvec_fn(q, &bvm, bvec) < bvec->bv_len) { bvec->bv_page = NULL; bvec->bv_len = 0; bvec->bv_offset = 0; -- 1.6.6
|  |