On 24/11/2014 12:49, Ming Lei wrote:
> On Mon, Nov 24, 2014 at 7:38 PM, Paolo Bonzini <pbonz...@redhat.com> wrote:
>>
>>
>> On 24/11/2014 12:31, Ming Lei wrote:
>>> + /* don't submit until next completion for -EAGAIN of non plug case */
>>> + if (unlikely(!s->io_q.plugged)) {
>>> + return 0;
>>> + }
>>> +
>>> /* submit immediately if queue depth is above 2/3 */
>>> if (idx > s->io_q.size * 2 / 3) {
>>> return ioq_submit(s);
>> }
>>
>> return 0;
>>
>> so I fail to see why my proposal:
>>
>>
>> /* This is reached in two cases: queue not plugged but io_submit
>> * returned -EAGAIN, or queue plugged. In the latter case, start
>> * submitting some I/O if the queue is getting too full. In the
>> * former case, instead, wait until an I/O operation is completed.
>> */
>> if (likely(idx <= s->io_q.size * 2 / 3) || unlikely(!s->io_q.plugged) {
>> return 0;
>> }
>>
>> return ioq_submit(s);
>>
>> was wrong. Can you explain?
>
> I didn't say your proposal is wrong, and this patch is correct too
> without fat comment.
>
> The difference is only that this patch returns immediately in case
> of !s->io_q.plugged after putting the req into io queue.
There is no difference in the behavior of the code, right?
The maintainers can decide if they want a v5 of this patch.
Thanks,
Paolo
