Re: [PATCH RFC 5.13 1/2] io_uring: add support for ns granularity of io_sq_thread_idle

[Pavel Begunkov <asml.silence@gmail.com>]

On 4/29/21 4:41 AM, Hao Xu wrote:
> 在 2021/4/28 下午10:16, Jens Axboe 写道:
>> On 4/28/21 8:07 AM, Pavel Begunkov wrote:
>>>> diff --git a/include/uapi/linux/io_uring.h b/include/uapi/linux/io_uring.h
>>>> index e1ae46683301..311532ff6ce3 100644
>>>> --- a/include/uapi/linux/io_uring.h
>>>> +++ b/include/uapi/linux/io_uring.h
>>>> @@ -98,6 +98,7 @@ enum {
>>>>   #define IORING_SETUP_CLAMP    (1U << 4)    /* clamp SQ/CQ ring sizes */
>>>>   #define IORING_SETUP_ATTACH_WQ    (1U << 5)    /* attach to existing wq */
>>>>   #define IORING_SETUP_R_DISABLED    (1U << 6)    /* start with ring disabled */
>>>> +#define IORING_SETUP_IDLE_NS    (1U << 7)    /* unit of thread_idle is nano second */
>>>>     enum {
>>>>       IORING_OP_NOP,
>>>> @@ -259,7 +260,7 @@ struct io_uring_params {
>>>>       __u32 cq_entries;
>>>>       __u32 flags;
>>>>       __u32 sq_thread_cpu;
>>>> -    __u32 sq_thread_idle;
>>>> +    __u64 sq_thread_idle;
>>>
>>> breaks userspace API
>>
>> And I don't think we need to. If you're using IDLE_NS, then the value
>> should by definition be small enough that it'd fit in 32-bits. If you
> I make it u64 since I thought users may want a full flexibility to set
> idle in nanosecond granularity(eg. (1e6 + 10) ns cannot be set by

It's a really weird user requiring such a precision. u32 allows up to
~1s, and if more is needed users can switch to ms mode, so in the worst
case the precision is 1/1000 of the desired value, more than enough.

> millisecond granularity). But I'm not sure if this deserve changing the
> userspace API.
 
That's not about deserve or not, we can't break ABI. Can be worked around,
e.g. by taking resv fields, but don't see a reason

>> need higher timeouts, don't set it and it's in usec instead.
>>
>> So I'd just leave this one alone.
>>
> 

-- 
Pavel Begunkov