On 24/11/2014 11:02, Ming Lei wrote:
>> > Right, but "-1" is "-EPERM" which doesn't make much sense. I think the
>> > right thing to do is to return EAGAIN, and let the owner figure it out.
>> > For example, a SCSI device might return a "BUSY" status code.
> We can do that, but laio_submit() do
On Mon, Nov 24, 2014 at 5:47 PM, Paolo Bonzini wrote:
>
>
> On 22/11/2014 13:16, Ming Lei wrote:
>> > > +static int ioq_enqueue(struct qemu_laio_state *s, struct iocb *iocb)
>> > > {
>> > > unsigned int idx = s->io_q.idx;
>> > >
>> > > +if (unlikely(idx == s->io_q.size)) {
>> > > +
On 22/11/2014 13:16, Ming Lei wrote:
> > > +static int ioq_enqueue(struct qemu_laio_state *s, struct iocb *iocb)
> > > {
> > > unsigned int idx = s->io_q.idx;
> > >
> > > +if (unlikely(idx == s->io_q.size)) {
> > > +return -1;
> >
> > return -EAGAIN?
>
> It means the io queue is
On Tue, Nov 18, 2014 at 10:18 PM, Paolo Bonzini wrote:
>
>
>> @@ -137,6 +145,12 @@ static void qemu_laio_completion_bh(void *opaque)
>> }
>> }
>>
>> +static void qemu_laio_start_retry(struct qemu_laio_state *s)
>> +{
>> +if (s->io_q.idx)
>> +qemu_bh_schedule(s->io_q.retry);
>> +}
> @@ -137,6 +145,12 @@ static void qemu_laio_completion_bh(void *opaque)
> }
> }
>
> +static void qemu_laio_start_retry(struct qemu_laio_state *s)
> +{
> +if (s->io_q.idx)
> +qemu_bh_schedule(s->io_q.retry);
> +}
> +
> static void qemu_laio_completion_cb(EventNotifier *e)
>
In the enqueue path, we can't complete request, otherwise
"Co-routine re-entered recursively" may be caused, so this
patch fixes the issue with below ideas:
- for -EAGAIN or partial completion, retry the submision by
schedule an BH in following completion cb
- for part of c
