Re: [PATCH V4 0/5] blk-mq: improvement on handling IO during CPU hotplug

[John Garry <john.garry@huawei.com>]

>>>
>>
>> Yes, we share tags among all queues, but we generate the tag - known as IPTT
>> - in the LLDD now, as we can no longer use the request tag (as it is not
>> unique per all queues):
>>
>> https://github.com/hisilicon/kernel-dev/commit/087b95af374be6965583c1673032fb33bc8127e8#diff-f5d8fff19bc539a7387af5230d4e5771R188
>>
>> As I said, the branch is messy and I did have to fix 087b95af374.
>
> Firstly this way may waste lots of memory, especially the queue depth is
> big, such as, hisilicon V3's queue depth is 4096.
>
> Secondly, you have to deal with queue busy efficiently and correctly,
> for example, your real hw tags(IPTT) can be used up easily, and how
> will you handle these dispatched request?

I have not seen scenario of exhausted IPTT. And IPTT count is same as 
SCSI host.can_queue, so SCSI midlayer should ensure that this does not 
occur.

>
> Finally, you have to evaluate the performance effect, this is highly
> related with how to deal with out-of-IPTT.

Some figures from our previous testing:

Managed interrupt without exposing multiple queues: 3M IOPs
Managed interrupt with exposing multiple queues: 2.6M IOPs
No managed interrupts: 500K IOPs.

Now my problem is that I hear some people are relying on the 3M 
performance, even though it is experimental and unsafe. I need to follow 
up on this. I really don't want to keep that code.

>
> I'd suggest you to fix the stuff and post them out for review.

OK, I'll also check on adding that WARN you provided.

Thanks again,
John

>
> Thanks,
> Ming
>
>
> .
>