From: Jens Axboe <jens.axboe@oracle.com>
To: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Ingo Molnar <mingo@elte.hu>,
linux-kernel@vger.kernel.org,
Andrew Morton <akpm@linux-foundation.org>
Subject: Re: [bug] block subsystem related crash with latest -git
Date: Wed, 17 Oct 2007 19:29:32 +0200 [thread overview]
Message-ID: <20071017172932.GI15552@kernel.dk> (raw)
In-Reply-To: <20071017172158.GH15552@kernel.dk>
On Wed, Oct 17 2007, Jens Axboe wrote:
> On Wed, Oct 17 2007, Jens Axboe wrote:
> > On Wed, Oct 17 2007, Jens Axboe wrote:
> > > On Wed, Oct 17 2007, Linus Torvalds wrote:
> > > >
> > > >
> > > > On Wed, 17 Oct 2007, Ingo Molnar wrote:
> > > > >
> > > > > Jens, just got this crash on a testbox:
> > > >
> > > > The code in question is:
> > > >
> > > > mov %edx,0xc(%esp)
> > > > mov (%ebx),%edi
> > > > mov %edi,%edx
> > > > sub %eax,%edx
> > > > mov %edx,%eax
> > > > sar $0x5,%eax
> > > > shl $0xc,%eax
> > > > add 0x8(%ebx),%eax
> > > > cmp %eax,0xc(%esp)
> > > > je +126
> > > > mov 0x10(%esi),%eax <----- Oops
> > > > lea 0x10(%esi),%edx
> > > > test $0x1,%al
> > > > jne +76
> > > > mov %edi,(%esi)
> > > > mov %ebp,0xc(%esi)
> > > > mov 0x8(%ebx),%eax
> > > > mov %eax,0x4(%esi)
> > > >
> > > >
> > > > and it looks like %esi is overflowing from one page to the next one, ie:
> > > >
> > > > BUG: unable to handle kernel paging request at virtual address 7ca76000
> > > > ESI: 7ca75ff0
> > > >
> > > > and you caught this thanks to page-alloc debugging again.
> > > >
> > > > I think I can match that up with the source code: that's "sg_next()". It's
> > > > doing:
> > > >
> > > > sg++;
> > > >
> > > > if (unlikely(sg_is_chain(sg)))
> > > > sg = sg_chain_ptr(sg);
> > > >
> > > > return sg;
> > > >
> > > > and the oopsing instruction is that load of "sg->page" in the assembly
> > > > code:
> > > >
> > > > mov 0x10(%esi),%eax # %eax = sg->page
> > > > lea 0x10(%esi),%edx # %edx = sg+1;
> > > > test $0x1,%al # if (unlikely(sg_is_chain()))
> > > > jne +76
> > > >
> > > > Jens?
> > >
> > > Yep, that's what I came up with as well - I asked Ingo for a dump in
> > > private, but ended up just using ksymoops to decode the line.
> > >
> > > The way blk_rq_map_sg() operates is that it ends up doing a
> > >
> > > next_sg = sg_next(sg);
> > >
> > > even though sg may be the last entry. Perhaps this is crapping out,
> > > although if sg is a valid address, then sg + 1 should be as well.
> > > next_sg may end up being crap, in fact it will, but we'll never use that
> > > unless there are more entries to fill. And if there is, then both sg and
> > > next_sg were valid.
> > >
> > > So nothing in for-linus should fix it, I'll try and come up with an
> > > alternate way to assign next_sg so it's always valid.
> >
> > OK, the below should actually be safe, I don't know why I talked myself
> > into the next_sg stuff in the beginning. It's always safe to zero sg,
> > since it's a valid entry - nothing to save in ->page. Ingo, does this
> > work for you?
> >
> > diff --git a/block/ll_rw_blk.c b/block/ll_rw_blk.c
> > index 9e3f3cc..3935469 100644
> > --- a/block/ll_rw_blk.c
> > +++ b/block/ll_rw_blk.c
> > @@ -1322,8 +1322,8 @@ int blk_rq_map_sg(struct request_queue *q, struct request *rq,
> > struct scatterlist *sglist)
> > {
> > struct bio_vec *bvec, *bvprv;
> > - struct scatterlist *next_sg, *sg;
> > struct req_iterator iter;
> > + struct scatterlist *sg;
> > int nsegs, cluster;
> >
> > nsegs = 0;
> > @@ -1333,7 +1333,7 @@ int blk_rq_map_sg(struct request_queue *q, struct request *rq,
> > * for each bio in rq
> > */
> > bvprv = NULL;
> > - sg = next_sg = &sglist[0];
> > + sg = NULL;
> > rq_for_each_segment(bvec, rq, iter) {
> > int nbytes = bvec->bv_len;
> >
> > @@ -1349,8 +1349,10 @@ int blk_rq_map_sg(struct request_queue *q, struct request *rq,
> > sg->length += nbytes;
> > } else {
> > new_segment:
> > - sg = next_sg;
> > - next_sg = sg_next(sg);
> > + if (!sg)
> > + sg = sglist;
> > + else
> > + sg = sg_next(sg);
> >
> > memset(sg, 0, sizeof(*sg));
> > sg->page = bvec->bv_page;
> >
>
> Scratch that, it cannot work... I'll think up a different approach.
OK, it is fine, as long as the sglist is cleared initially. And I don't
think there's anyway around that, clearly I didn't think long enough
before including the memset() removal from Tomo.
Ingo, please try this rolled up version.
Linus, this should work. It would probably be best if you first did a
git revert on f5c0dde4c66421a3a2d7d6fa604a712c9b0744e5 and then applied
the ll_rw_blk.c bit alone. Do you want me to stuff that (revert + patch)
into a branch for you to pull?
diff --git a/block/ll_rw_blk.c b/block/ll_rw_blk.c
index 9e3f3cc..3935469 100644
--- a/block/ll_rw_blk.c
+++ b/block/ll_rw_blk.c
@@ -1322,8 +1322,8 @@ int blk_rq_map_sg(struct request_queue *q, struct request *rq,
struct scatterlist *sglist)
{
struct bio_vec *bvec, *bvprv;
- struct scatterlist *next_sg, *sg;
struct req_iterator iter;
+ struct scatterlist *sg;
int nsegs, cluster;
nsegs = 0;
@@ -1333,7 +1333,7 @@ int blk_rq_map_sg(struct request_queue *q, struct request *rq,
* for each bio in rq
*/
bvprv = NULL;
- sg = next_sg = &sglist[0];
+ sg = NULL;
rq_for_each_segment(bvec, rq, iter) {
int nbytes = bvec->bv_len;
@@ -1349,8 +1349,10 @@ int blk_rq_map_sg(struct request_queue *q, struct request *rq,
sg->length += nbytes;
} else {
new_segment:
- sg = next_sg;
- next_sg = sg_next(sg);
+ if (!sg)
+ sg = sglist;
+ else
+ sg = sg_next(sg);
memset(sg, 0, sizeof(*sg));
sg->page = bvec->bv_page;
diff --git a/drivers/scsi/scsi_lib.c b/drivers/scsi/scsi_lib.c
index 0c86be7..aac8a02 100644
--- a/drivers/scsi/scsi_lib.c
+++ b/drivers/scsi/scsi_lib.c
@@ -764,6 +764,8 @@ struct scatterlist *scsi_alloc_sgtable(struct scsi_cmnd *cmd, gfp_t gfp_mask)
if (unlikely(!sgl))
goto enomem;
+ memset(sgl, 0, sizeof(*sgl) * sgp->size);
+
/*
* first loop through, set initial index and return value
*/
--
Jens Axboe
next prev parent reply other threads:[~2007-10-17 17:29 UTC|newest]
Thread overview: 151+ messages / expand[flat|nested] mbox.gz Atom feed top
2007-10-17 15:46 [bug] block subsystem related crash with latest -git Ingo Molnar
2007-10-17 15:50 ` Ingo Molnar
2007-10-17 16:32 ` Jens Axboe
2007-10-17 16:50 ` Linus Torvalds
2007-10-17 16:59 ` Jens Axboe
2007-10-17 17:08 ` Jens Axboe
2007-10-17 17:21 ` Jens Axboe
2007-10-17 17:29 ` Jens Axboe [this message]
2007-10-17 17:34 ` Ingo Molnar
2007-10-17 17:36 ` Jens Axboe
2007-10-17 17:45 ` [bug] ata " Ingo Molnar
2007-10-17 17:53 ` Jens Axboe
2007-10-17 17:55 ` Jens Axboe
2007-10-17 17:58 ` Ingo Molnar
2007-10-17 18:37 ` Jens Axboe
2007-10-17 19:04 ` Ingo Molnar
2007-10-17 19:08 ` Jens Axboe
2007-10-17 19:14 ` Ingo Molnar
2007-10-17 19:17 ` Ingo Molnar
2007-10-17 19:25 ` Jens Axboe
2007-10-17 19:25 ` Jens Axboe
2007-10-17 19:09 ` Ingo Molnar
2007-10-17 19:28 ` Linus Torvalds
2007-10-17 19:35 ` Jens Axboe
2007-10-17 19:45 ` Linus Torvalds
2007-10-17 19:56 ` Jens Axboe
2007-10-17 20:06 ` Jens Axboe
2007-10-17 20:24 ` Linus Torvalds
2007-10-17 20:31 ` Jens Axboe
2007-10-17 21:11 ` Linus Torvalds
2007-10-17 23:00 ` FUJITA Tomonori
2007-10-18 1:07 ` Linus Torvalds
2007-10-18 1:14 ` Jeff Garzik
2007-10-18 1:19 ` David Miller
2007-10-18 1:36 ` Linus Torvalds
2007-10-18 1:49 ` David Miller
2007-10-18 3:44 ` Mark Lord
2007-10-18 4:01 ` Linus Torvalds
2007-10-18 4:05 ` Mark Lord
2007-10-18 4:14 ` Jeff Garzik
2007-10-18 4:18 ` Mark Lord
2007-10-18 4:31 ` Jeff Garzik
2007-10-18 4:41 ` Mark Lord
2007-10-18 4:53 ` Linus Torvalds
2007-10-18 7:05 ` Jens Axboe
2007-10-18 13:13 ` Mark Lord
2007-10-18 13:23 ` Jens Axboe
2007-10-18 13:32 ` Mark Lord
2007-10-18 13:34 ` Jens Axboe
2007-10-18 13:59 ` Mark Lord
2007-10-18 14:04 ` Jens Axboe
2007-10-18 4:45 ` Linus Torvalds
2007-10-18 4:54 ` Mark Lord
2007-10-18 5:09 ` Mark Lord
2007-10-18 4:20 ` Linus Torvalds
2007-10-18 5:25 ` Mark Lord
2007-10-18 5:34 ` Mark Lord
2007-10-18 5:45 ` Jeff Garzik
2007-10-18 7:09 ` Jens Axboe
2007-10-18 7:30 ` Jeff Garzik
2007-10-18 8:21 ` Jens Axboe
2007-10-18 11:55 ` David Miller
2007-10-18 11:57 ` Jens Axboe
2007-10-18 12:05 ` David Miller
2007-10-18 12:09 ` Jens Axboe
2007-10-18 12:15 ` Jens Axboe
2007-10-18 12:36 ` David Miller
2007-10-18 12:39 ` Jens Axboe
2007-10-18 12:58 ` Benny Halevy
2007-10-18 13:56 ` Jens Axboe
2007-10-18 14:05 ` Jens Axboe
2007-10-18 14:16 ` Benny Halevy
2007-10-18 14:38 ` Jens Axboe
2007-10-18 14:58 ` Olof Johansson
2007-10-18 15:25 ` Jens Axboe
2007-10-18 12:58 ` Jens Axboe
2007-10-18 13:32 ` Jens Axboe
2007-10-18 13:49 ` Benny Halevy
2007-10-18 13:55 ` Jens Axboe
2007-10-18 13:51 ` Mark Lord
2007-10-18 13:58 ` Jens Axboe
2007-10-18 14:03 ` Mark Lord
2007-10-18 14:10 ` Mark Lord
2007-10-18 14:13 ` Mark Lord
2007-10-18 14:14 ` Jens Axboe
2007-10-18 16:55 ` Linus Torvalds
2007-10-18 17:01 ` Jens Axboe
2007-10-18 17:10 ` Jens Axboe
2007-10-18 17:10 ` Arjan van de Ven
2007-10-18 17:14 ` Jens Axboe
2007-10-19 8:59 ` FUJITA Tomonori
2007-10-18 19:20 ` Jeff Garzik
2007-10-17 20:51 ` Ingo Molnar
2007-10-17 19:49 ` Jens Axboe
2007-10-17 20:05 ` Ingo Molnar
2007-10-17 20:10 ` Linus Torvalds
2007-10-18 7:07 ` Ingo Molnar
2007-10-18 7:10 ` Jens Axboe
2007-10-18 8:22 ` Jeff Garzik
2007-10-18 8:32 ` Jens Axboe
2007-10-18 8:38 ` Jeff Garzik
2007-10-18 8:51 ` Jeff Garzik
2007-10-18 9:01 ` Jeff Garzik
[not found] ` <bd58e4af0710180210tcc0d31ep9d05a0f2e9d6df29@mail.gmail.com>
2007-10-18 9:14 ` Jeff Garzik
2007-10-18 9:17 ` Jens Axboe
2007-10-18 9:32 ` Jeff Garzik
2007-10-18 9:41 ` Jens Axboe
2007-10-18 10:04 ` Jeff Garzik
2007-10-18 10:10 ` Jens Axboe
2007-10-18 10:13 ` Ingo Molnar
2007-10-18 10:16 ` Jens Axboe
2007-10-18 10:17 ` Jens Axboe
2007-10-18 10:49 ` Ingo Molnar
2007-10-18 10:50 ` Jeff Garzik
2007-10-18 10:56 ` Jens Axboe
2007-10-18 10:42 ` [PATCH] " Jeff Garzik
2007-10-18 10:54 ` Ingo Molnar
2007-10-18 11:02 ` Jeff Garzik
2007-10-18 11:40 ` Ingo Molnar
2007-10-18 14:52 ` Olof Johansson
2007-10-20 11:55 ` Torsten Kaiser
2007-10-18 11:03 ` Ingo Molnar
2007-10-18 11:05 ` Jens Axboe
2007-10-17 19:42 ` Linus Torvalds
2007-10-17 19:55 ` Jens Axboe
2007-10-17 18:08 ` Linus Torvalds
2007-10-17 18:13 ` Ingo Molnar
2007-10-17 17:56 ` [bug] block " Linus Torvalds
2007-10-17 18:02 ` Jens Axboe
2007-10-17 18:13 ` Linus Torvalds
2007-10-17 18:20 ` Jens Axboe
2007-10-17 18:58 ` Linus Torvalds
2007-10-17 19:03 ` Jens Axboe
2007-10-17 19:15 ` Linus Torvalds
2007-10-17 18:02 ` Ingo Molnar
2007-10-17 18:14 ` Linus Torvalds
2007-10-17 20:15 ` Luca Tettamanti
2007-10-17 17:30 ` Ingo Molnar
2007-10-17 17:31 ` Jens Axboe
2007-10-17 17:28 ` Ingo Molnar
2007-10-17 17:52 ` Linus Torvalds
2007-10-17 18:00 ` Jens Axboe
2007-10-17 18:18 ` Linus Torvalds
2007-10-17 18:22 ` Jens Axboe
2007-10-18 10:52 ` Benny Halevy
2007-10-18 10:55 ` Jens Axboe
2007-10-18 12:03 ` David Miller
2007-10-18 12:28 ` Jens Axboe
2007-10-17 18:22 ` Linus Torvalds
2007-10-17 18:40 ` Jens Axboe
2007-10-17 17:11 ` Ingo Molnar
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=20071017172932.GI15552@kernel.dk \
--to=jens.axboe@oracle.com \
--cc=akpm@linux-foundation.org \
--cc=linux-kernel@vger.kernel.org \
--cc=mingo@elte.hu \
--cc=torvalds@linux-foundation.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