On Mon, Jul 18, 2011 at 06:43:31PM +0300, Alon Levy wrote:
> On Mon, Jul 18, 2011 at 05:28:26PM +0200, Christophe Fergeau wrote:
> > On Mon, Jul 18, 2011 at 05:51:30PM +0300, Alon Levy wrote:
> > > > @@ -1046,6 +1047,17 @@ void RedClient::on_agent_announce_capabilities(
> > > > // not sending the color depth through
> > > > send_agent_monitors_config, since
> > > > // it applies only for attached screens.
> > > > send_agent_display_config();
> > > > + } else if (!_auto_display_res) {
> > >
> > > Who sets _auto_display_res? does this affect windows guest agents? the
> > > comment
> > > below says linux but the test above seems to be not linux specific.
> >
> > No idea, I added this test to mirror the
> > if (_auto_display_res) {
> > send_agent_monitors_config();
> > }
> > in handle_init. This avoids changing behaviour when we decided to send the
> > monitors config in handle_init. If the comment is misleading, I can change
> > it, for example "Some agents don't support monitors/displays ..."
>
> I didn't care about the comment, but about the behavior. I want to be sure
> this doesn't change the behavior for the windows agent.
In the windows case, we'll go through the 1st block:
if (VD_AGENT_HAS_CAPABILITY(caps->caps, caps_size,
VD_AGENT_CAP_DISPLAY_CONFIG) && !_agent_disp_config_sent) {
// not sending the color depth through send_agent_monitors_config,
// sinc
// it applies only for attached screens.
send_agent_display_config();
}
The if (!_auto_display_res) test was added to avoid as much as possible
changing the old behaviour, we used to be waiting in this case, since we
sent a message in handle_init, it makes sense to still be waiting in this
case.
Christophe
pgpb1GXOK1Vke.pgp
Description: PGP signature
_______________________________________________ Spice-devel mailing list [email protected] http://lists.freedesktop.org/mailman/listinfo/spice-devel
