Re: memory access op ideas

[Jens Axboe <axboe@kernel.dk>]

On 4/22/22 6:52 AM, Hao Xu wrote:
> Hi Avi,
> ? 4/13/22 6:33 PM, Avi Kivity ??:
>> Unfortunately, only ideas, no patches. But at least the first seems very easy.
>>
>>
>> - IORING_OP_MEMCPY_IMMEDIATE - copy some payload included in the op itself (1-8 bytes) to a user memory location specified by the op.
>>
>>
>> Linked to another op, this can generate an in-memory notification useful for busy-waiters or the UMWAIT instruction
>>
>>
>> This would be useful for Seastar, which looks at a timer-managed memory location to check when to break computation loops.
>>
>>
>> - IORING_OP_MEMCPY - asynchronously copy memory
>>
>>
>> Some CPUs include a DMA engine, and io_uring is a perfect interface to exercise it. It may be difficult to find space for two iovecs though.
> 
> I have a question about the 'DMA' here, do you mean DMA device for
> memory copy? My understanding is you want async memcpy so that the
> cpu can relax when the specific hardware is doing memory copy. the
> thing is for cases like busy waiting or UMAIT, the length of the memory
> to be copied is usually small(otherwise we don't use busy waiting or
> UMAIT, right?). Then making it async by io_uring's iowq may introduce
> much more overhead(the context switch).

Nothing fast should use io-wq. But not sure why you think this would
need it, as long as you can start the operation in a sane fashion and
get notified when done, why would it need io-wq?

-- 
Jens Axboe