On 12/4/18 10:13 AM, Sagi Grimberg wrote:
> 
>>>> +static int nvme_poll_irqdisable(struct nvme_queue *nvmeq, unsigned int 
>>>> tag)
>>>
>>> Do we still need to carry the tag around?
>>
>> Yes, the timeout handler polls for a specific tag.
> 
> Does it have to? the documentation suggests that we missed
> an interrupt, so it is probably waiting on the completion queue.
> 
> I'd say that it'd be better if the tag search would be implemented
> on the timeout handler alone so we don't have to pass the tag around
> for everyone... Thoughts?

Without that you don't know if that's the request that completed. You
have to be able to look that up from the timeout handler.

-- 
Jens Axboe

Reply via email to