Re: [PATCH V2] nvme: Add support for ACRE Command Interrupted status

[Hannes Reinecke <hare@suse.de>]

On 12/3/19 10:00 PM, Keith Busch wrote:
> On Tue, Dec 03, 2019 at 05:38:04PM +0000, Meneghini, John wrote:
>> This is an update to say that I've tested this patch and it works as expected. 
>>
>> When the controller returns a Command Interrupted status the request is avoids nvme_failover_req()
>> and goes down the nvme_retry_req() path where the CRD is implemented and the command is 
>> retried after a delay.
>>
>> If the controllers returns Command Interrupted too many times, and nvme_req(req)->retries
>> runs down, this results in a device resource error returned to the block layer.  But I think we'll
>> have this problem with any error.  
> 
> Why is the controller returning the same error so many times? Are we
> not waiting the requested delay timed? If so, the controller told us
> retrying should be successful.
> 
> It is possible we kick the requeue list early if one command error
> has a valid CRD, but a subsequent retryable command does not. Is that
> what's happening?
> 
No. The problem is that CRD needs to be observed _in all cases_ whenever
we get a COMMAND INTERRUPTED status.
Currently we're only observing CRD when multipathing is _not_ active,
which is wrong.

And this is what the patch is trying to achieve.

> I'm just concerned because if we just skip counting the retry, a broken
> device could have the driver retry the same command indefinitely, which
> often leaves a task in an uninterruptible sleep state forever.
> 
Oh, but we're not skipping the retries; in fact we do _enable_ retry
counting with this patch as we're skipping the nvme_failover_req()
branch and moving to nvme_retry_req().

>>     diff --git a/drivers/nvme/host/core.c b/drivers/nvme/host/core.c
>>     index 9696404a6182..24dc9ed1a11b 100644
>>     --- a/drivers/nvme/host/core.c
>>     +++ b/drivers/nvme/host/core.c
>>     @@ -230,6 +230,8 @@ static blk_status_t nvme_error_status(u16 status)
>>                     return BLK_STS_NEXUS;
>>             case NVME_SC_HOST_PATH_ERROR:
>>                     return BLK_STS_TRANSPORT;
>>     +       case NVME_SC_CMD_INTERRUPTED:
>>     +               return BLK_STS_DEV_RESOURCE;
> 
> Just for the sake of keeping this change isloted to nvme, perhaps use an
> existing blk_status_t value that already maps to not path error, like
> BLK_STS_TARGET.
> 
I don't mind; whichever works for you.

Cheers,

Hannes
-- 
Dr. Hannes Reinecke		      Teamlead Storage & Networking
hare@suse.de			                  +49 911 74053 688
SUSE Software Solutions Germany GmbH, Maxfeldstr. 5, 90409 Nürnberg
HRB 36809 (AG Nürnberg), GF: Felix Imendörffer

_______________________________________________
linux-nvme mailing list
linux-nvme@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-nvme