Re: Re: [PATCH v7] io_uring: Statistics of the true utilization of sq threads.

[Xiaobing Li <xiaobing.li@samsung.com>]

On 1/29/24 14:00, Jens Axboe wrote:
>On 1/29/24 12:18 AM, Xiaobing Li wrote:
>> On 1/18/24 19:34, Jens Axboe wrote:
>>>> diff --git a/io_uring/sqpoll.h b/io_uring/sqpoll.h
>>>> index 8df37e8c9149..c14c00240443 100644
>>>> --- a/io_uring/sqpoll.h
>>>> +++ b/io_uring/sqpoll.h
>>>> @@ -16,6 +16,7 @@ struct io_sq_data {
>>>>  	pid_t			task_pid;
>>>>  	pid_t			task_tgid;
>>>>  
>>>> +	long long			work_time;
>>>>  	unsigned long		state;
>>>>  	struct completion	exited;
>>>>  };
>>>
>>> Probably just make that an u64.
>>>
>>> As Pavel mentioned, I think we really need to consider if fdinfo is the
>>> appropriate API for this. It's fine if you're running stuff directly and
>>> you're just curious, but it's a very cumbersome API in general as you
>>> need to know the pid of the task holding the ring, the fd of the ring,
>>> and then you can get it as a textual description. If this is something
>>> that is deemed useful, would it not make more sense to make it
>>> programatically available in addition, or even exclusively?
>> 
>> Hi, Jens and Pavel
>> sorry for the late reply.
>> 
>> I've tried some other methods, but overall, I haven't found a more suitable 
>> method than fdinfo.
>> If you think it is troublesome to obtain the PID,  then I can provide
>>  a shell script to output the total_time and work_time of all sqpoll threads 
>>  to the terminal, so that we do not have to manually obtain the PID of each 
>>  thread (the script can be placed in tools/ include/io_uring).
>> 
>> eg:
>> 
>> PID    WorkTime(us)   TotalTime(us)   COMMAND
>> 9330   1106578        2215321         iou-sqp-9329
>> 9454   1510658        1715321         iou-sqp-9453
>> 9478   165785         223219          iou-sqp-9477
>> 9587   106578         153217          iou-sqp-9586
>> 
>> What do you think of this solution?
>
>I don't think it's a great interface, but at the same time, I don't feel
>that strongly about it and perhaps bundling a script that outputs the
>above in liburing would be Good Enough. I'm a bit reluctant to add a
>stats API via io_uring_register() just for this.
>
>So maybe spin a v8 with the s/long long/u64 change and include your
>script as well?
 
ok, I'll send out a v8.

--
Xiaobing Li