From: NeilBrown <neilb@suse.de>
To: Joe Lawrence <joe.lawrence@stratus.com>
Cc: linux-raid@vger.kernel.org, kmo@daterainc.com
Subject: Re: [PATCH 2/2] md: raid1: use bio_rewind() before bio_copy_data()
Date: Tue, 9 Jul 2013 14:33:52 +1000 [thread overview]
Message-ID: <20130709143352.4a5d616c@notabene.brown> (raw)
In-Reply-To: <1373315114-16625-2-git-send-email-joe.lawrence@stratus.com>
[-- Attachment #1: Type: text/plain, Size: 2447 bytes --]
On Mon, 8 Jul 2013 16:25:14 -0400 Joe Lawrence <joe.lawrence@stratus.com>
wrote:
> bio_copy_data() cannot execute on a fully completed source (or
> destination) bio. MD RAID1 is recycling a completed read bio for RAID
> repair purposes and needs to fix up the source and destination bios for
> bio_copy_data(): in particular, the io-vector index needs to be rewound
> and the bios unchained.
>
> Suggested-by: NeilBrown <neilb@suse.de>
> Signed-off-by: Joe Lawrence <joe.lawrence@stratus.com>
Thanks Joe.
Jens: are you happy with this? If so is it OK for me to forward to Linus
and -stable for 3.10, or would you rather?
The follow-on patch which uses the so avoid a crash is
http://marc.info/?l=linux-raid&m=137331514131503&w=2
(This is sitting near the top of "git://neil.brown.name/md for-next" if you
want to cherry-pick it).
Thanks,
NeilBrown
> ---
> drivers/md/raid1.c | 28 +++++-----------------------
> 1 file changed, 5 insertions(+), 23 deletions(-)
>
> diff --git a/drivers/md/raid1.c b/drivers/md/raid1.c
> index 6e17f81..bb1e80f 100644
> --- a/drivers/md/raid1.c
> +++ b/drivers/md/raid1.c
> @@ -1861,7 +1861,6 @@ static int process_checks(struct r1bio *r1_bio)
> int j;
> struct bio *pbio = r1_bio->bios[primary];
> struct bio *sbio = r1_bio->bios[i];
> - int size;
>
> if (sbio->bi_end_io != end_sync_read)
> continue;
> @@ -1887,28 +1886,11 @@ static int process_checks(struct r1bio *r1_bio)
> rdev_dec_pending(conf->mirrors[i].rdev, mddev);
> continue;
> }
> - /* fixup the bio for reuse */
> - bio_reset(sbio);
> - sbio->bi_vcnt = vcnt;
> - sbio->bi_size = r1_bio->sectors << 9;
> - sbio->bi_sector = r1_bio->sector +
> - conf->mirrors[i].rdev->data_offset;
> - sbio->bi_bdev = conf->mirrors[i].rdev->bdev;
> - sbio->bi_end_io = end_sync_read;
> - sbio->bi_private = r1_bio;
> -
> - size = sbio->bi_size;
> - for (j = 0; j < vcnt ; j++) {
> - struct bio_vec *bi;
> - bi = &sbio->bi_io_vec[j];
> - bi->bv_offset = 0;
> - if (size > PAGE_SIZE)
> - bi->bv_len = PAGE_SIZE;
> - else
> - bi->bv_len = size;
> - size -= PAGE_SIZE;
> - }
> -
> + /* copy pbio io_vec data into sbio, but first reset io_vec
> + * index and unchain both */
> + bio_rewind(pbio);
> + bio_rewind(sbio);
> + pbio->bi_next = sbio->bi_next = NULL;
> bio_copy_data(sbio, pbio);
> }
> return 0;
[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 828 bytes --]
next prev parent reply other threads:[~2013-07-09 4:33 UTC|newest]
Thread overview: 10+ messages / expand[flat|nested] mbox.gz Atom feed top
2013-07-02 1:20 RAID1 repair GPF crash w/3.10-rc7 Joe Lawrence
2013-07-03 21:49 ` Joe Lawrence
2013-07-04 0:53 ` NeilBrown
2013-07-08 20:06 ` Joe Lawrence
2013-07-08 20:25 ` [PATCH 1/2] block: add bio_rewind() to reset bio_vec Joe Lawrence
2013-07-14 23:40 ` NeilBrown
2013-07-08 20:25 ` [PATCH 2/2] md: raid1: use bio_rewind() before bio_copy_data() Joe Lawrence
2013-07-09 4:33 ` NeilBrown [this message]
2013-07-17 6:12 ` RAID1 repair GPF crash w/3.10-rc7 NeilBrown
2013-07-17 21:01 ` Joe Lawrence
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20130709143352.4a5d616c@notabene.brown \
--to=neilb@suse.de \
--cc=joe.lawrence@stratus.com \
--cc=kmo@daterainc.com \
--cc=linux-raid@vger.kernel.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).