2009-09-03 15:49:40

by Mark McLoughlin

Subject: [PATCH resend] block: silently error unsupported empty barriers too

With 2.6.31-rc5 in a KVM guest using dm and virtio_blk, we see the
following errors:

end_request: I/O error, dev vda, sector 0
end_request: I/O error, dev vda, sector 0

The errors go away if dm stops submitting empty barriers, by reverting:

commit 52b1fd5a27c625c78373e024bf570af3c9d44a79
Author: Mikulas Patocka <[email protected]>
dm: send empty barriers to targets in dm_flush

We should error all barriers, even empty barriers, on devices like
virtio_blk which don't support them.

See also:

https://bugzilla.redhat.com/514901

Signed-off-by: Mark McLoughlin <[email protected]>
Cc: Rusty Russell <[email protected]>
Cc: Mikulas Patocka <[email protected]>
Cc: Alasdair G Kergon <[email protected]>
Cc: Neil Brown <[email protected]>
Cc: Christoph Hellwig <[email protected]>
---
block/blk-core.c | 3 +--
1 files changed, 1 insertions(+), 2 deletions(-)

diff --git a/block/blk-core.c b/block/blk-core.c
index e3299a7..35ad2bb 100644
--- a/block/blk-core.c
+++ b/block/blk-core.c
@@ -1163,8 +1163,7 @@ static int __make_request(struct request_queue *q, struct bio *bio)
const int unplug = bio_unplug(bio);
int rw_flags;

- if (bio_barrier(bio) && bio_has_data(bio) &&
- (q->next_ordered == QUEUE_ORDERED_NONE)) {
+ if (bio_barrier(bio) && (q->next_ordered == QUEUE_ORDERED_NONE)) {
bio_endio(bio, -EOPNOTSUPP);
return 0;
}
--
1.6.4

2009-10-15 00:24:38

by Alasdair G Kergon

Subject: Re: [dm-devel] [PATCH resend] block: silently error unsupported empty barriers too

On Thu, Sep 03, 2009 at 04:47:50PM +0100, Mark McLoughlin wrote:
> We should error all barriers, even empty barriers, on devices like
> virtio_blk which don't support them.

Acked-by: Alasdair G Kergon <[email protected]>

2009-10-15 19:12:13

by Mikulas Patocka

Subject: Re: [dm-devel] [PATCH resend] block: silently error unsupported empty barriers too

On Thu, 15 Oct 2009, Alasdair G Kergon wrote:

> On Thu, Sep 03, 2009 at 04:47:50PM +0100, Mark McLoughlin wrote:
> > We should error all barriers, even empty barriers, on devices like
> > virtio_blk which don't support them.
>
> Acked-by: Alasdair G Kergon <[email protected]>

Acked-by: Mikulas Patocka <[email protected]>

2009-10-23 20:40:59

by Mike Snitzer

Subject: Re: [PATCH resend] block: silently error unsupported empty barriers too

On Thu, Sep 03 2009 at 11:47am -0400,
Mark McLoughlin <[email protected]> wrote:

> With 2.6.31-rc5 in a KVM guest using dm and virtio_blk, we see the
> following errors:
>
> end_request: I/O error, dev vda, sector 0
> end_request: I/O error, dev vda, sector 0
>
> The errors go away if dm stops submitting empty barriers, by reverting:
>
> commit 52b1fd5a27c625c78373e024bf570af3c9d44a79
> Author: Mikulas Patocka <[email protected]>
> dm: send empty barriers to targets in dm_flush
>
> We should error all barriers, even empty barriers, on devices like
> virtio_blk which don't support them.
>
> See also:
>
> https://bugzilla.redhat.com/514901
>
> Signed-off-by: Mark McLoughlin <[email protected]>
> Cc: Rusty Russell <[email protected]>
> Cc: Mikulas Patocka <[email protected]>
> Cc: Alasdair G Kergon <[email protected]>
> Cc: Neil Brown <[email protected]>
> Cc: Christoph Hellwig <[email protected]>
> ---
> block/blk-core.c | 3 +--
> 1 files changed, 1 insertions(+), 2 deletions(-)
>
> diff --git a/block/blk-core.c b/block/blk-core.c
> index e3299a7..35ad2bb 100644
> --- a/block/blk-core.c
> +++ b/block/blk-core.c
> @@ -1163,8 +1163,7 @@ static int __make_request(struct request_queue *q, struct bio *bio)
> const int unplug = bio_unplug(bio);
> int rw_flags;
>
> - if (bio_barrier(bio) && bio_has_data(bio) &&
> - (q->next_ordered == QUEUE_ORDERED_NONE)) {
> + if (bio_barrier(bio) && (q->next_ordered == QUEUE_ORDERED_NONE)) {
> bio_endio(bio, -EOPNOTSUPP);
> return 0;
> }

Jens,

Any chance we can get this fix upstream for 2.6.32? I refreshed the
patch to work on 2.6.32-rc5:


From: Mark McLoughlin <[email protected]>

With 2.6.32-rc5 in a KVM guest using dm and virtio_blk, we see the
following errors:

end_request: I/O error, dev vda, sector 0
end_request: I/O error, dev vda, sector 0

The errors go away if dm stops submitting empty barriers, by reverting:

commit 52b1fd5a27c625c78373e024bf570af3c9d44a79
Author: Mikulas Patocka <[email protected]>
dm: send empty barriers to targets in dm_flush

We should silently error all barriers, even empty barriers, on devices
like virtio_blk which don't support them.

See also:

https://bugzilla.redhat.com/514901

Signed-off-by: Mark McLoughlin <[email protected]>
Signed-off-by: Mike Snitzer <[email protected]>
Acked-by: Alasdair G Kergon <[email protected]>
Acked-by: Mikulas Patocka <[email protected]>
Cc: Rusty Russell <[email protected]>
Cc: Neil Brown <[email protected]>
Cc: Christoph Hellwig <[email protected]>
---
block/blk-core.c | 2 +-
1 files changed, 1 insertions(+), 1 deletions(-)

diff --git a/block/blk-core.c b/block/blk-core.c
index ac0fa10..71da511 100644
--- a/block/blk-core.c
+++ b/block/blk-core.c
@@ -1161,7 +1161,7 @@ static int __make_request(struct request_queue *q, struct bio *bio)
const unsigned int ff = bio->bi_rw & REQ_FAILFAST_MASK;
int rw_flags;

- if (bio_rw_flagged(bio, BIO_RW_BARRIER) && bio_has_data(bio) &&
+ if (bio_rw_flagged(bio, BIO_RW_BARRIER) &&
(q->next_ordered == QUEUE_ORDERED_NONE)) {
bio_endio(bio, -EOPNOTSUPP);
return 0;