* [PATCH] scsi: core: fix race between handling STS_RESOURCE and completion
@ 2020-12-02 10:04 Ming Lei
2020-12-02 18:10 ` Bart Van Assche
2020-12-08 4:56 ` Martin K. Petersen
0 siblings, 2 replies; 6+ messages in thread
From: Ming Lei @ 2020-12-02 10:04 UTC (permalink / raw)
To: linux-scsi, Martin K . Petersen
Cc: Ming Lei, Hannes Reinecke, Sumit Saxena, Kashyap Desai,
Bart Van Assche, Ewan Milne, Long Li, chenxiang (M),
John Garry
When queuing IO request to LLD, STS_RESOURCE may be returned because:
- host in recovery or blocked
- target queue throttling or blocked
- LLD rejection
Any one of the above doesn't happen frequently enough.
BLK_STS_DEV_RESOURCE is returned to block layer for avoiding unnecessary
re-run queue, and it is just one small optimization. However, all
in-flight requests originated from this scsi device may be completed
just after reading 'sdev->device_busy', so BLK_STS_DEV_RESOURCE is
returned to block layer. And the current failed IO won't get chance
to be queued any more, since it is invisible at that time for either
scsi_run_queue_async() or blk-mq's RESTART.
Fix the issue by not returning BLK_STS_DEV_RESOURCE in this situation.
Cc: Hannes Reinecke <hare@suse.com>
Cc: Sumit Saxena <sumit.saxena@broadcom.com>
Cc: Kashyap Desai <kashyap.desai@broadcom.com>
Cc: Bart Van Assche <bvanassche@acm.org>
Cc: Ewan Milne <emilne@redhat.com>
Cc: Long Li <longli@microsoft.com>
Tested-by: "chenxiang (M)" <chenxiang66@hisilicon.com>
Reported-by: John Garry <john.garry@huawei.com>
Signed-off-by: Ming Lei <ming.lei@redhat.com>
---
drivers/scsi/scsi_lib.c | 3 +--
1 file changed, 1 insertion(+), 2 deletions(-)
diff --git a/drivers/scsi/scsi_lib.c b/drivers/scsi/scsi_lib.c
index 60c7a7d74852..03c6d0620bfd 100644
--- a/drivers/scsi/scsi_lib.c
+++ b/drivers/scsi/scsi_lib.c
@@ -1703,8 +1703,7 @@ static blk_status_t scsi_queue_rq(struct blk_mq_hw_ctx *hctx,
break;
case BLK_STS_RESOURCE:
case BLK_STS_ZONE_RESOURCE:
- if (atomic_read(&sdev->device_busy) ||
- scsi_device_blocked(sdev))
+ if (scsi_device_blocked(sdev))
ret = BLK_STS_DEV_RESOURCE;
break;
default:
--
2.28.0
^ permalink raw reply related [flat|nested] 6+ messages in thread
* Re: [PATCH] scsi: core: fix race between handling STS_RESOURCE and completion
2020-12-02 10:04 [PATCH] scsi: core: fix race between handling STS_RESOURCE and completion Ming Lei
@ 2020-12-02 18:10 ` Bart Van Assche
2020-12-03 1:03 ` Ming Lei
2020-12-08 4:56 ` Martin K. Petersen
1 sibling, 1 reply; 6+ messages in thread
From: Bart Van Assche @ 2020-12-02 18:10 UTC (permalink / raw)
To: Ming Lei, linux-scsi, Martin K . Petersen
Cc: Hannes Reinecke, Sumit Saxena, Kashyap Desai, Ewan Milne,
Long Li, chenxiang (M),
John Garry
On 12/2/20 2:04 AM, Ming Lei wrote:
> When queuing IO request to LLD, STS_RESOURCE may be returned because:
>
> - host in recovery or blocked
> - target queue throttling or blocked
> - LLD rejection
>
> Any one of the above doesn't happen frequently enough.
>
> BLK_STS_DEV_RESOURCE is returned to block layer for avoiding unnecessary
> re-run queue, and it is just one small optimization. However, all
> in-flight requests originated from this scsi device may be completed
> just after reading 'sdev->device_busy', so BLK_STS_DEV_RESOURCE is
> returned to block layer. And the current failed IO won't get chance
> to be queued any more, since it is invisible at that time for either
> scsi_run_queue_async() or blk-mq's RESTART.
>
> Fix the issue by not returning BLK_STS_DEV_RESOURCE in this situation.
>
> Cc: Hannes Reinecke <hare@suse.com>
> Cc: Sumit Saxena <sumit.saxena@broadcom.com>
> Cc: Kashyap Desai <kashyap.desai@broadcom.com>
> Cc: Bart Van Assche <bvanassche@acm.org>
> Cc: Ewan Milne <emilne@redhat.com>
> Cc: Long Li <longli@microsoft.com>
> Tested-by: "chenxiang (M)" <chenxiang66@hisilicon.com>
> Reported-by: John Garry <john.garry@huawei.com>
> Signed-off-by: Ming Lei <ming.lei@redhat.com>
> ---
> drivers/scsi/scsi_lib.c | 3 +--
> 1 file changed, 1 insertion(+), 2 deletions(-)
>
> diff --git a/drivers/scsi/scsi_lib.c b/drivers/scsi/scsi_lib.c
> index 60c7a7d74852..03c6d0620bfd 100644
> --- a/drivers/scsi/scsi_lib.c
> +++ b/drivers/scsi/scsi_lib.c
> @@ -1703,8 +1703,7 @@ static blk_status_t scsi_queue_rq(struct blk_mq_hw_ctx *hctx,
> break;
> case BLK_STS_RESOURCE:
> case BLK_STS_ZONE_RESOURCE:
> - if (atomic_read(&sdev->device_busy) ||
> - scsi_device_blocked(sdev))
> + if (scsi_device_blocked(sdev))
> ret = BLK_STS_DEV_RESOURCE;
> break;
> default:
Since this patch modifies code introduced in commit 86ff7c2a80cd
("blk-mq: introduce BLK_STS_DEV_RESOURCE"), does this patch perhaps
needs a Fixes: tag?
Thanks,
Bart.
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: [PATCH] scsi: core: fix race between handling STS_RESOURCE and completion
2020-12-02 18:10 ` Bart Van Assche
@ 2020-12-03 1:03 ` Ming Lei
2020-12-07 3:32 ` chenxiang (M)
0 siblings, 1 reply; 6+ messages in thread
From: Ming Lei @ 2020-12-03 1:03 UTC (permalink / raw)
To: Bart Van Assche
Cc: linux-scsi, Martin K . Petersen, Hannes Reinecke, Sumit Saxena,
Kashyap Desai, Ewan Milne, Long Li, chenxiang (M),
John Garry
On Wed, Dec 02, 2020 at 10:10:30AM -0800, Bart Van Assche wrote:
> On 12/2/20 2:04 AM, Ming Lei wrote:
> > When queuing IO request to LLD, STS_RESOURCE may be returned because:
> >
> > - host in recovery or blocked
> > - target queue throttling or blocked
> > - LLD rejection
> >
> > Any one of the above doesn't happen frequently enough.
> >
> > BLK_STS_DEV_RESOURCE is returned to block layer for avoiding unnecessary
> > re-run queue, and it is just one small optimization. However, all
> > in-flight requests originated from this scsi device may be completed
> > just after reading 'sdev->device_busy', so BLK_STS_DEV_RESOURCE is
> > returned to block layer. And the current failed IO won't get chance
> > to be queued any more, since it is invisible at that time for either
> > scsi_run_queue_async() or blk-mq's RESTART.
> >
> > Fix the issue by not returning BLK_STS_DEV_RESOURCE in this situation.
> >
> > Cc: Hannes Reinecke <hare@suse.com>
> > Cc: Sumit Saxena <sumit.saxena@broadcom.com>
> > Cc: Kashyap Desai <kashyap.desai@broadcom.com>
> > Cc: Bart Van Assche <bvanassche@acm.org>
> > Cc: Ewan Milne <emilne@redhat.com>
> > Cc: Long Li <longli@microsoft.com>
> > Tested-by: "chenxiang (M)" <chenxiang66@hisilicon.com>
> > Reported-by: John Garry <john.garry@huawei.com>
> > Signed-off-by: Ming Lei <ming.lei@redhat.com>
> > ---
> > drivers/scsi/scsi_lib.c | 3 +--
> > 1 file changed, 1 insertion(+), 2 deletions(-)
> >
> > diff --git a/drivers/scsi/scsi_lib.c b/drivers/scsi/scsi_lib.c
> > index 60c7a7d74852..03c6d0620bfd 100644
> > --- a/drivers/scsi/scsi_lib.c
> > +++ b/drivers/scsi/scsi_lib.c
> > @@ -1703,8 +1703,7 @@ static blk_status_t scsi_queue_rq(struct blk_mq_hw_ctx *hctx,
> > break;
> > case BLK_STS_RESOURCE:
> > case BLK_STS_ZONE_RESOURCE:
> > - if (atomic_read(&sdev->device_busy) ||
> > - scsi_device_blocked(sdev))
> > + if (scsi_device_blocked(sdev))
> > ret = BLK_STS_DEV_RESOURCE;
> > break;
> > default:
>
> Since this patch modifies code introduced in commit 86ff7c2a80cd ("blk-mq:
> introduce BLK_STS_DEV_RESOURCE"), does this patch perhaps needs a Fixes:
> tag?
This same race exists before commit 86ff7c2a80cd, so I think the 'Fixes:' tag
is misleading.
Thanks,
Ming
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: [PATCH] scsi: core: fix race between handling STS_RESOURCE and completion
2020-12-03 1:03 ` Ming Lei
@ 2020-12-07 3:32 ` chenxiang (M)
2020-12-07 7:51 ` Ming Lei
0 siblings, 1 reply; 6+ messages in thread
From: chenxiang (M) @ 2020-12-07 3:32 UTC (permalink / raw)
To: Ming Lei, Bart Van Assche
Cc: linux-scsi, Martin K . Petersen, Hannes Reinecke, Sumit Saxena,
Kashyap Desai, Ewan Milne, Long Li, John Garry
在 2020/12/3 9:03, Ming Lei 写道:
> On Wed, Dec 02, 2020 at 10:10:30AM -0800, Bart Van Assche wrote:
>> On 12/2/20 2:04 AM, Ming Lei wrote:
>>> When queuing IO request to LLD, STS_RESOURCE may be returned because:
>>>
>>> - host in recovery or blocked
>>> - target queue throttling or blocked
>>> - LLD rejection
>>>
>>> Any one of the above doesn't happen frequently enough.
>>>
>>> BLK_STS_DEV_RESOURCE is returned to block layer for avoiding unnecessary
>>> re-run queue, and it is just one small optimization. However, all
>>> in-flight requests originated from this scsi device may be completed
>>> just after reading 'sdev->device_busy', so BLK_STS_DEV_RESOURCE is
>>> returned to block layer. And the current failed IO won't get chance
>>> to be queued any more, since it is invisible at that time for either
>>> scsi_run_queue_async() or blk-mq's RESTART.
>>>
>>> Fix the issue by not returning BLK_STS_DEV_RESOURCE in this situation.
>>>
>>> Cc: Hannes Reinecke <hare@suse.com>
>>> Cc: Sumit Saxena <sumit.saxena@broadcom.com>
>>> Cc: Kashyap Desai <kashyap.desai@broadcom.com>
>>> Cc: Bart Van Assche <bvanassche@acm.org>
>>> Cc: Ewan Milne <emilne@redhat.com>
>>> Cc: Long Li <longli@microsoft.com>
>>> Tested-by: "chenxiang (M)" <chenxiang66@hisilicon.com>
>>> Reported-by: John Garry <john.garry@huawei.com>
>>> Signed-off-by: Ming Lei <ming.lei@redhat.com>
>>> ---
>>> drivers/scsi/scsi_lib.c | 3 +--
>>> 1 file changed, 1 insertion(+), 2 deletions(-)
>>>
>>> diff --git a/drivers/scsi/scsi_lib.c b/drivers/scsi/scsi_lib.c
>>> index 60c7a7d74852..03c6d0620bfd 100644
>>> --- a/drivers/scsi/scsi_lib.c
>>> +++ b/drivers/scsi/scsi_lib.c
>>> @@ -1703,8 +1703,7 @@ static blk_status_t scsi_queue_rq(struct blk_mq_hw_ctx *hctx,
>>> break;
>>> case BLK_STS_RESOURCE:
>>> case BLK_STS_ZONE_RESOURCE:
>>> - if (atomic_read(&sdev->device_busy) ||
>>> - scsi_device_blocked(sdev))
>>> + if (scsi_device_blocked(sdev))
>>> ret = BLK_STS_DEV_RESOURCE;
>>> break;
>>> default:
>> Since this patch modifies code introduced in commit 86ff7c2a80cd ("blk-mq:
>> introduce BLK_STS_DEV_RESOURCE"), does this patch perhaps needs a Fixes:
>> tag?
> This same race exists before commit 86ff7c2a80cd, so I think the 'Fixes:' tag
> is misleading.
When reverted the patch "scsi: core: Only re-run queue in
scsi_end_request() if device queue is busy", it also solves the issue.
Does the issue is brought by the patch? If so, maybe adding
fixes("Fixes: ed5dd6a67d5e ("scsi: core: Only re-run queue in
scsi_end_request() if device queue is busy")") is more accuratte.
>
>
>
> Thanks,
> Ming
>
>
> .
>
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: [PATCH] scsi: core: fix race between handling STS_RESOURCE and completion
2020-12-07 3:32 ` chenxiang (M)
@ 2020-12-07 7:51 ` Ming Lei
0 siblings, 0 replies; 6+ messages in thread
From: Ming Lei @ 2020-12-07 7:51 UTC (permalink / raw)
To: chenxiang (M)
Cc: Bart Van Assche, linux-scsi, Martin K . Petersen,
Hannes Reinecke, Sumit Saxena, Kashyap Desai, Ewan Milne,
Long Li, John Garry
On Mon, Dec 07, 2020 at 11:32:54AM +0800, chenxiang (M) wrote:
>
>
> 在 2020/12/3 9:03, Ming Lei 写道:
> > On Wed, Dec 02, 2020 at 10:10:30AM -0800, Bart Van Assche wrote:
> > > On 12/2/20 2:04 AM, Ming Lei wrote:
> > > > When queuing IO request to LLD, STS_RESOURCE may be returned because:
> > > >
> > > > - host in recovery or blocked
> > > > - target queue throttling or blocked
> > > > - LLD rejection
> > > >
> > > > Any one of the above doesn't happen frequently enough.
> > > >
> > > > BLK_STS_DEV_RESOURCE is returned to block layer for avoiding unnecessary
> > > > re-run queue, and it is just one small optimization. However, all
> > > > in-flight requests originated from this scsi device may be completed
> > > > just after reading 'sdev->device_busy', so BLK_STS_DEV_RESOURCE is
> > > > returned to block layer. And the current failed IO won't get chance
> > > > to be queued any more, since it is invisible at that time for either
> > > > scsi_run_queue_async() or blk-mq's RESTART.
> > > >
> > > > Fix the issue by not returning BLK_STS_DEV_RESOURCE in this situation.
> > > >
> > > > Cc: Hannes Reinecke <hare@suse.com>
> > > > Cc: Sumit Saxena <sumit.saxena@broadcom.com>
> > > > Cc: Kashyap Desai <kashyap.desai@broadcom.com>
> > > > Cc: Bart Van Assche <bvanassche@acm.org>
> > > > Cc: Ewan Milne <emilne@redhat.com>
> > > > Cc: Long Li <longli@microsoft.com>
> > > > Tested-by: "chenxiang (M)" <chenxiang66@hisilicon.com>
> > > > Reported-by: John Garry <john.garry@huawei.com>
> > > > Signed-off-by: Ming Lei <ming.lei@redhat.com>
> > > > ---
> > > > drivers/scsi/scsi_lib.c | 3 +--
> > > > 1 file changed, 1 insertion(+), 2 deletions(-)
> > > >
> > > > diff --git a/drivers/scsi/scsi_lib.c b/drivers/scsi/scsi_lib.c
> > > > index 60c7a7d74852..03c6d0620bfd 100644
> > > > --- a/drivers/scsi/scsi_lib.c
> > > > +++ b/drivers/scsi/scsi_lib.c
> > > > @@ -1703,8 +1703,7 @@ static blk_status_t scsi_queue_rq(struct blk_mq_hw_ctx *hctx,
> > > > break;
> > > > case BLK_STS_RESOURCE:
> > > > case BLK_STS_ZONE_RESOURCE:
> > > > - if (atomic_read(&sdev->device_busy) ||
> > > > - scsi_device_blocked(sdev))
> > > > + if (scsi_device_blocked(sdev))
> > > > ret = BLK_STS_DEV_RESOURCE;
> > > > break;
> > > > default:
> > > Since this patch modifies code introduced in commit 86ff7c2a80cd ("blk-mq:
> > > introduce BLK_STS_DEV_RESOURCE"), does this patch perhaps needs a Fixes:
> > > tag?
> > This same race exists before commit 86ff7c2a80cd, so I think the 'Fixes:' tag
> > is misleading.
>
> When reverted the patch "scsi: core: Only re-run queue in scsi_end_request()
> if device queue is busy", it also solves the issue.
> Does the issue is brought by the patch? If so, maybe adding fixes("Fixes:
> ed5dd6a67d5e ("scsi: core: Only re-run queue in scsi_end_request() if device
> queue is busy")") is more accuratte.
The debugfs log shows that the issue isn't related with restart or out
of budget because the single request stays in hctx->dispatch, and the flag of
RQF_DONTPREP is set too.
The scsi's restart(scsi_run_queue_async) from request completion is just
for handling requests in scheduler queue.
Thanks,
Ming
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: [PATCH] scsi: core: fix race between handling STS_RESOURCE and completion
2020-12-02 10:04 [PATCH] scsi: core: fix race between handling STS_RESOURCE and completion Ming Lei
2020-12-02 18:10 ` Bart Van Assche
@ 2020-12-08 4:56 ` Martin K. Petersen
1 sibling, 0 replies; 6+ messages in thread
From: Martin K. Petersen @ 2020-12-08 4:56 UTC (permalink / raw)
To: Ming Lei, linux-scsi
Cc: Martin K . Petersen, Kashyap Desai, Hannes Reinecke, John Garry,
Long Li, Sumit Saxena, chenxiang (M),
Bart Van Assche, Ewan Milne
On Wed, 2 Dec 2020 18:04:19 +0800, Ming Lei wrote:
> When queuing IO request to LLD, STS_RESOURCE may be returned because:
>
> - host in recovery or blocked
> - target queue throttling or blocked
> - LLD rejection
>
> Any one of the above doesn't happen frequently enough.
>
> [...]
Applied to 5.10/scsi-fixes, thanks!
[1/1] scsi: core: fix race between handling STS_RESOURCE and completion
https://git.kernel.org/mkp/scsi/c/673235f91531
--
Martin K. Petersen Oracle Linux Engineering
^ permalink raw reply [flat|nested] 6+ messages in thread
end of thread, other threads:[~2020-12-08 4:57 UTC | newest]
Thread overview: 6+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2020-12-02 10:04 [PATCH] scsi: core: fix race between handling STS_RESOURCE and completion Ming Lei
2020-12-02 18:10 ` Bart Van Assche
2020-12-03 1:03 ` Ming Lei
2020-12-07 3:32 ` chenxiang (M)
2020-12-07 7:51 ` Ming Lei
2020-12-08 4:56 ` Martin K. Petersen
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.