On 31/01/2017 08:50, Stefan Hajnoczi wrote:
>> I'm not sure a
>> comment in the NBD driver is the best place, because similar logic will
>> appear soon in other networked block drivers.
> Maybe add a new function that just does these two calls. I don't have a
> good name for it.
For now I'll ad
On Mon, Jan 30, 2017 at 04:18:10PM -0500, Paolo Bonzini wrote:
> On 30/01/2017 10:50, Stefan Hajnoczi wrote:
> > On Fri, Jan 20, 2017 at 05:43:12PM +0100, Paolo Bonzini wrote:
> >> +aio_co_wake(s->recv_coroutine[i]);
> >>
> >> -qemu_coroutine_enter(nbd_get_client_session(bs)->send_cor
On 30/01/2017 10:50, Stefan Hajnoczi wrote:
> On Fri, Jan 20, 2017 at 05:43:12PM +0100, Paolo Bonzini wrote:
>> +aio_co_wake(s->recv_coroutine[i]);
>>
>> -qemu_coroutine_enter(nbd_get_client_session(bs)->send_coroutine);
>> +/* We're woken up by the recv_coroutine itself. *
On Fri, Jan 20, 2017 at 05:43:12PM +0100, Paolo Bonzini wrote:
> +aio_co_wake(s->recv_coroutine[i]);
>
> -qemu_coroutine_enter(nbd_get_client_session(bs)->send_coroutine);
> +/* We're woken up by the recv_coroutine itself. */
> +qemu_coroutine_yield();
This relies on
In the client, read the reply headers from a coroutine, switching the
read side between the "read header" coroutine and the I/O coroutine that
reads the body of the reply.
In the server, if the server can read more requests it will create a new
"read request" coroutine as soon as a request has bee
