Re: Re: [PATCH v9] io_uring: Statistics of the true utilization of sq threads.

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

On 2/27/24 14:36, Jens Axboe wrote:
>On 2/26/24 10:45 PM, Xiaobing Li wrote:
>> On 2/21/24 10:28, Jens Axboe wrote:
>>> On 2/20/24 7:04 PM, Xiaobing Li wrote:
>>>> On 2/19/24 14:42, Xiaobing Li wrote:
>>>>> diff --git a/io_uring/fdinfo.c b/io_uring/fdinfo.c
>>>>> index 976e9500f651..37afc5bac279 100644
>>>>> --- a/io_uring/fdinfo.c
>>>>> +++ b/io_uring/fdinfo.c
>>>>> @@ -55,6 +55,7 @@ __cold void io_uring_show_fdinfo(struct seq_file *m, struct file *f)
>>>>> 	struct io_ring_ctx *ctx = f->private_data;
>>>>> 	struct io_overflow_cqe *ocqe;
>>>>> 	struct io_rings *r = ctx->rings;
>>>>> +	struct rusage sq_usage;
>>>>> 	unsigned int sq_mask = ctx->sq_entries - 1, cq_mask = ctx->cq_entries - 1;
>>>>> 	unsigned int sq_head = READ_ONCE(r->sq.head);
>>>>> 	unsigned int sq_tail = READ_ONCE(r->sq.tail);
>>>>> @@ -64,6 +65,7 @@ __cold void io_uring_show_fdinfo(struct seq_file *m, struct file *f)
>>>>> 	unsigned int sq_shift = 0;
>>>>> 	unsigned int sq_entries, cq_entries;
>>>>> 	int sq_pid = -1, sq_cpu = -1;
>>>>> +	u64 sq_total_time = 0, sq_work_time = 0;
>>>>> 	bool has_lock;
>>>>> 	unsigned int i;
>>>>>
>>>>> @@ -147,10 +149,15 @@ __cold void io_uring_show_fdinfo(struct seq_file *m, struct file *f)
>>>>>
>>>>> 		sq_pid = sq->task_pid;
>>>>> 		sq_cpu = sq->sq_cpu;
>>>>> +		getrusage(sq->thread, RUSAGE_SELF, &sq_usage);
>>>>> +		sq_total_time = sq_usage.ru_stime.tv_sec * 1000000 + sq_usage.ru_stime.tv_usec;
>>>>> +		sq_work_time = sq->work_time;
>>>>> 	}
>>>>>
>>>>> 	seq_printf(m, "SqThread:\t%d\n", sq_pid);
>>>>> 	seq_printf(m, "SqThreadCpu:\t%d\n", sq_cpu);
>>>>> +	seq_printf(m, "SqTotalTime:\t%llu\n", sq_total_time);
>>>>> +	seq_printf(m, "SqWorkTime:\t%llu\n", sq_work_time);
>>>>> 	seq_printf(m, "UserFiles:\t%u\n", ctx->nr_user_files);
>>>>> 	for (i = 0; has_lock && i < ctx->nr_user_files; i++) {
>>>>> 		struct file *f = io_file_from_index(&ctx->file_table, i);
>>>>> diff --git a/io_uring/sqpoll.c b/io_uring/sqpoll.c
>>>>> index 65b5dbe3c850..006d7fc9cf92 100644
>>>>> --- a/io_uring/sqpoll.c
>>>>> +++ b/io_uring/sqpoll.c
>>>>> @@ -219,10 +219,22 @@ static bool io_sqd_handle_event(struct io_sq_data *sqd)
>>>>> 	return did_sig || test_bit(IO_SQ_THREAD_SHOULD_STOP, &sqd->state);
>>>>> }
>>>>>
>>>>> +static void io_sq_update_worktime(struct io_sq_data *sqd, struct rusage *start)
>>>>> +{
>>>>> +		struct rusage end;
>>>>> +
>>>>> +		getrusage(current, RUSAGE_SELF, &end);
>>>>> +		end.ru_stime.tv_sec -= start->ru_stime.tv_sec;
>>>>> +		end.ru_stime.tv_usec -= start->ru_stime.tv_usec;
>>>>> +
>>>>> +		sqd->work_time += end.ru_stime.tv_usec + end.ru_stime.tv_sec * 1000000;
>>>>> +}
>>>>> +
>>>>> static int io_sq_thread(void *data)
>>>>> {
>>>>> 	struct io_sq_data *sqd = data;
>>>>> 	struct io_ring_ctx *ctx;
>>>>> +	struct rusage start;
>>>>> 	unsigned long timeout = 0;
>>>>> 	char buf[TASK_COMM_LEN];
>>>>> 	DEFINE_WAIT(wait);
>>>>> @@ -251,6 +263,7 @@ static int io_sq_thread(void *data)
>>>>> 		}
>>>>>
>>>>> 		cap_entries = !list_is_singular(&sqd->ctx_list);
>>>>> +		getrusage(current, RUSAGE_SELF, &start);
>>>>> 		list_for_each_entry(ctx, &sqd->ctx_list, sqd_list) {
>>>>> 			int ret = __io_sq_thread(ctx, cap_entries);
>>>>>
>>>>> @@ -261,8 +274,10 @@ static int io_sq_thread(void *data)
>>>>> 			sqt_spin = true;
>>>>>
>>>>> 		if (sqt_spin || !time_after(jiffies, timeout)) {
>>>>> -			if (sqt_spin)
>>>>> +			if (sqt_spin) {
>>>>> +				io_sq_update_worktime(sqd, &start);
>>>>> 				timeout = jiffies + sqd->sq_thread_idle;
>>>>> +			}
>>>>> 			if (unlikely(need_resched())) {
>>>>> 				mutex_unlock(&sqd->lock);
>>>>> 				cond_resched();
>>>>> diff --git a/io_uring/sqpoll.h b/io_uring/sqpoll.h
>>>>> index 8df37e8c9149..4171666b1cf4 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;
>>>>>
>>>>> +	u64			work_time;
>>>>> 	unsigned long		state;
>>>>> 	struct completion	exited;
>>>>> };
>>>>  
>>>> Hi, Jens
>>>> I have modified the code according to your suggestions.
>>>> Do you have any other comments?
>>>
>>> Out of town this week, I'll check next week. But from a quick look,
>>> looks much better now.
>>  
>> Hi, Jens
>> Do you have time to check now?
>
>Can I ask you to resend it against for-6.9/io_uring? For some reason I
>don't see the original patch on the list.
 
ok, I'll resend a v10.

--
Xiaobing Li