Re: [PATCH 5/5] nvme: use __blk_mq_complete_request in timeout path

[jianchao.wang]

On 06/25/2018 02:07 AM, Sagi Grimberg wrote:
> 
>> Hi Christoph
>>
>> Thanks for your kindly response.
>>
>> On 06/20/2018 10:39 PM, Christoph Hellwig wrote:
>>>> diff --git a/drivers/nvme/host/pci.c b/drivers/nvme/host/pci.c
>>>> index 73a97fc..2a161f6 100644
>>>> --- a/drivers/nvme/host/pci.c
>>>> +++ b/drivers/nvme/host/pci.c
>>>> @@ -1203,6 +1203,7 @@ static enum blk_eh_timer_return nvme_timeout(struct request *req, bool reserved)
>>>> ÂÂÂÂÂÂÂÂÂ nvme_warn_reset(dev, csts);
>>>> ÂÂÂÂÂÂÂÂÂ nvme_dev_disable(dev, false);
>>>> ÂÂÂÂÂÂÂÂÂ nvme_reset_ctrl(&dev->ctrl);
>>>> +ÂÂÂÂÂÂÂ __blk_mq_complete_request(req);
>>>> ÂÂÂÂÂÂÂÂÂ return BLK_EH_DONE;
>>>> ÂÂÂÂÂ }
>>>> Â @@ -1213,6 +1214,11 @@ static enum blk_eh_timer_return nvme_timeout(struct request *req, bool reserved)
>>>> ÂÂÂÂÂÂÂÂÂ dev_warn(dev->ctrl.device,
>>>> ÂÂÂÂÂÂÂÂÂÂÂÂÂÂ "I/O %d QID %d timeout, completion polled\n",
>>>> ÂÂÂÂÂÂÂÂÂÂÂÂÂÂ req->tag, nvmeq->qid);
>>>> +ÂÂÂÂÂÂÂ /*
>>>> +ÂÂÂÂÂÂÂÂ * nvme_end_request will invoke blk_mq_complete_request,
>>>> +ÂÂÂÂÂÂÂÂ * it will do nothing for this timed out request.
>>>> +ÂÂÂÂÂÂÂÂ */
>>>> +ÂÂÂÂÂÂÂ __blk_mq_complete_request(req);
>>>
>>> And this clearly is bogus. We want to iterate over the tagetset
>>> and cancel all requests, not do that manually here.
>>>
>>> That was the whole point of the original change.
>>>
>>
>> For nvme-pci, we indeed have an issue that when nvme_reset_work->nvme_dev_disable returns, timeout path maybe still
>> running and the nvme_dev_disable invoked by timeout path will race with the nvme_reset_work.
>> However, the hole is still there right now w/o my changes, but just narrower.
> 
> Given the amount of fixes (and fixes of fixes) we had in the timeout handler, maybe it'd be a good idea to step back and take a another look?
> 
> Won't it be better to avoid disabling the device and return
> BLK_EH_RESET_TIMER if we are not aborting in the timeout handler?
> 

Yes, that would be an ideal status for nvme-pci.
But we have to depend on the timeout handler to handle the timed out request from nvme_reset_work.

Thanks
Jianchao