On Wed, Jun 29, 2011 at 03:06:33PM +0200, Gerd Hoffmann wrote:
> Hi,
>
> >+ case QXL_IO_FLUSH_SURFACES:
> >+ dprint(d, 1, "QXL_IO_FLUSH_SURFACES (%d) entry (%s, s#=%d,
> >res#=%d)\n",
> >+ val, qxl_mode_to_string(d->mode), d->guest_surfaces.count,
> >+ d->num_free_res);
> >+ qemu_spice_stop(&d->ssd);
> >+ qemu_spice_start(&d->ssd);
> >+ dprint(d, 1, "QXL_IO_FLUSH_SURFACES exit (%s, s#=%d, res#=%d,%p)\n",
> >+ qxl_mode_to_string(d->mode), d->guest_surfaces.count,
> >+ d->num_free_res, d->last_release);
> >+ break;
>
> This should be async as we'll go sleep and wait for the spice server
> thread finish in qemu_spice_stop().
Yeah, I meant to do that, forgot, thanks for catching it.
>
> >+ case QXL_IO_FLUSH_RELEASE: {
> >+ QXLReleaseRing *ring =&d->ram->release_ring;
> >+ if (ring->prod - ring->cons + 1 == ring->num_items) {
> >+ // TODO - "return" a value to the guest and let it loop?
> ^^^^
> Hmm.
So the story goes: I wrote this, but didn't actually see this happen in
practice,
particularily since the driver empties the release ring. The simplest would be
to
replace it with some fprintf(stderr)
>
> cheers,
> Gerd
