* BUG: scsi/qla2xxx: BUG_ON(blk_queued_rq(req) is triggered in blk_finish_request
@ 2018-05-23 3:55 jianchao.wang
2018-05-24 1:50 ` jianchao.wang
0 siblings, 1 reply; 3+ messages in thread
From: jianchao.wang @ 2018-05-23 3:55 UTC (permalink / raw)
To: qla2xxx-upstream, himanshu.madhani, jthumshirn
Cc: linux-scsi, linux-kernel, jejb, Martin K. Petersen, Junxiao Bi
Hi all
Our customer met a panic triggered by BUG_ON in blk_finish_request.
>From the dmesg log, the BUG_ON was triggered after command abort occurred many times.
There is a race condition in the following scenario.
cpu A cpu B
kworker interrupt
scmd_eh_abort_handler()
-> scsi_try_to_abort_cmd()
-> qla2xxx_eh_abort()
-> qla2x00_eh_wait_on_command() qla2x00_status_entry()
-> qla2x00_sp_compl()
-> qla2x00_sp_free_dma()
-> scsi_queue_insert()
-> __scsi_queue_insert()
-> blk_requeue_request()
-> blk_clear_rq_complete()
-> scsi_done
-> blk_complete_request
-> blk_mark_rq_complete
-> elv_requeue_request() -> __blk_complete_request()
-> __elv_add_request()
// req will be queued here
BLK_SOFTIRQ
scsi_softirq_done()
-> scsi_finish_command()
-> scsi_io_completion()
-> scsi_end_request()
-> blk_finish_request() // BUG_ON(blk_queued_rq(req)) !!!
The issue will not be triggered most of time, because the request is marked as complete by timeout path.
So the scsi_done from qla2x00_sp_compl does nothing.
But as the scenario above, if the complete state has been cleaned by blk_requeue_request, we will get
the request both requeued and completed, and then BUG_ON(blk_queued_rq(req)) in blk_finish_request comes up.
Is there any solution for this in qla2xxx driver side ?
Thanks
Jianchao
^ permalink raw reply [flat|nested] 3+ messages in thread
* Re: BUG: scsi/qla2xxx: BUG_ON(blk_queued_rq(req) is triggered in blk_finish_request
2018-05-23 3:55 BUG: scsi/qla2xxx: BUG_ON(blk_queued_rq(req) is triggered in blk_finish_request jianchao.wang
@ 2018-05-24 1:50 ` jianchao.wang
[not found] ` <SN4PR0701MB363286637D0317358D61ECBFF56A0@SN4PR0701MB3632.namprd07.prod.outlook.com>
0 siblings, 1 reply; 3+ messages in thread
From: jianchao.wang @ 2018-05-24 1:50 UTC (permalink / raw)
To: qla2xxx-upstream, himanshu.madhani, jthumshirn
Cc: linux-scsi, linux-kernel, jejb, Martin K. Petersen, Junxiao Bi
Would anyone please take a look at this ?
Thanks in advance
Jianchao
On 05/23/2018 11:55 AM, jianchao.wang wrote:
>
>
> Hi all
>
> Our customer met a panic triggered by BUG_ON in blk_finish_request.
>>From the dmesg log, the BUG_ON was triggered after command abort occurred many times.
> There is a race condition in the following scenario.
>
> cpu A cpu B
> kworker interrupt
>
> scmd_eh_abort_handler()
> -> scsi_try_to_abort_cmd()
> -> qla2xxx_eh_abort()
> -> qla2x00_eh_wait_on_command() qla2x00_status_entry()
> -> qla2x00_sp_compl()
> -> qla2x00_sp_free_dma()
> -> scsi_queue_insert()
> -> __scsi_queue_insert()
> -> blk_requeue_request()
> -> blk_clear_rq_complete()
> -> scsi_done
> -> blk_complete_request
> -> blk_mark_rq_complete
> -> elv_requeue_request() -> __blk_complete_request()
> -> __elv_add_request()
> // req will be queued here
> BLK_SOFTIRQ
> scsi_softirq_done()
> -> scsi_finish_command()
> -> scsi_io_completion()
> -> scsi_end_request()
> -> blk_finish_request() // BUG_ON(blk_queued_rq(req)) !!!
>
> The issue will not be triggered most of time, because the request is marked as complete by timeout path.
> So the scsi_done from qla2x00_sp_compl does nothing.
> But as the scenario above, if the complete state has been cleaned by blk_requeue_request, we will get
> the request both requeued and completed, and then BUG_ON(blk_queued_rq(req)) in blk_finish_request comes up.
>
> Is there any solution for this in qla2xxx driver side ?
>
> Thanks
> Jianchao
>
>
^ permalink raw reply [flat|nested] 3+ messages in thread
* Re: BUG: scsi/qla2xxx: BUG_ON(blk_queued_rq(req) is triggered in blk_finish_request
[not found] ` <SN4PR0701MB363286637D0317358D61ECBFF56A0@SN4PR0701MB3632.namprd07.prod.outlook.com>
@ 2018-05-24 9:09 ` jianchao.wang
0 siblings, 0 replies; 3+ messages in thread
From: jianchao.wang @ 2018-05-24 9:09 UTC (permalink / raw)
To: Madhani, Himanshu, Dept-Eng QLA2xxx Upstream, jthumshirn
Cc: linux-scsi, linux-kernel, jejb, Martin K. Petersen, Junxiao Bi
Hi Himanshu
Thanks god get response from your side finally :)
I get qla2xxx-upstream@qlogic.com from the maintainer information of driver/scsi/qla2xxx
Looks like it should be modified.
Thanks
Jianchao
On 05/24/2018 12:21 PM, Madhani, Himanshu wrote:
> Hi,
>
> Yes. We'll look at this issue.
>
> Thanks,
> Himanshu
>
>> -----Original Message-----
>> From: jianchao.wang [mailto:jianchao.w.wang@oracle.com]
>> Sent: Wednesday, May 23, 2018 6:51 PM
>> To: Dept-Eng QLA2xxx Upstream <qla2xxx-upstream@cavium.com>; Madhani,
>> Himanshu <Himanshu.Madhani@cavium.com>; jthumshirn@suse.de
>> Cc: linux-scsi@vger.kernel.org; linux-kernel@vger.kernel.org;
>> jejb@linux.vnet.ibm.com; Martin K. Petersen <martin.petersen@oracle.com>;
>> Junxiao Bi <junxiao.bi@oracle.com>
>> Subject: Re: BUG: scsi/qla2xxx: BUG_ON(blk_queued_rq(req) is triggered in
>> blk_finish_request
>>
>> Would anyone please take a look at this ?
>>
>> Thanks in advance
>> Jianchao
>>
>> On 05/23/2018 11:55 AM, jianchao.wang wrote:
>>>
>>>
>>> Hi all
>>>
>>> Our customer met a panic triggered by BUG_ON in blk_finish_request.
>>> >From the dmesg log, the BUG_ON was triggered after command abort
>> occurred many times.
>>> There is a race condition in the following scenario.
>>>
>>> cpu A cpu B
>>> kworker interrupt
>>>
>>> scmd_eh_abort_handler()
>>> -> scsi_try_to_abort_cmd()
>>> -> qla2xxx_eh_abort()
>>> -> qla2x00_eh_wait_on_command() qla2x00_status_entry()
>>> -> qla2x00_sp_compl()
>>> -> qla2x00_sp_free_dma()
>>> -> scsi_queue_insert()
>>> -> __scsi_queue_insert()
>>> -> blk_requeue_request()
>>> -> blk_clear_rq_complete()
>>> -> scsi_done
>>> -> blk_complete_request
>>> -> blk_mark_rq_complete
>>> -> elv_requeue_request() -> __blk_complete_request()
>>> -> __elv_add_request()
>>> // req will be queued here
>>> BLK_SOFTIRQ
>>> scsi_softirq_done()
>>> -> scsi_finish_command()
>>> -> scsi_io_completion()
>>> -> scsi_end_request()
>>> -> blk_finish_request() //
>> BUG_ON(blk_queued_rq(req)) !!!
>>>
>>> The issue will not be triggered most of time, because the request is marked as
>> complete by timeout path.
>>> So the scsi_done from qla2x00_sp_compl does nothing.
>>> But as the scenario above, if the complete state has been cleaned by
>> blk_requeue_request, we will get
>>> the request both requeued and completed, and then
>> BUG_ON(blk_queued_rq(req)) in blk_finish_request comes up.
>>>
>>> Is there any solution for this in qla2xxx driver side ?
>>>
>>> Thanks
>>> Jianchao
>>>
>>>
^ permalink raw reply [flat|nested] 3+ messages in thread
end of thread, other threads:[~2018-05-24 9:09 UTC | newest]
Thread overview: 3+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2018-05-23 3:55 BUG: scsi/qla2xxx: BUG_ON(blk_queued_rq(req) is triggered in blk_finish_request jianchao.wang
2018-05-24 1:50 ` jianchao.wang
[not found] ` <SN4PR0701MB363286637D0317358D61ECBFF56A0@SN4PR0701MB3632.namprd07.prod.outlook.com>
2018-05-24 9:09 ` jianchao.wang
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).