Il giorno dom, 06/07/2008 alle 22.29 +0100, Matthew Allum ha scritto:
> Hi;
> 
> On Sun, 2008-07-06 at 22:26 +0200, Filippo Argiolas wrote:
> > 
> > Yup, you're right.. it was so simple! Now let me make some test to see
> > if it solves the black screen issue.
> Fingers crossed!

Eheh :).. I'm sorry but setting the clutter context as shared doesn't
seem to do the job.. I've tried also to set the gstreamergl xwindow to
the clutter one but still doesn't work :(
I'm pretty sure I'm doing something wrong but I'm not so experienced in
X and glx programming to understand what's going wrong.

> Yeah, I *think* so - difficult to really say however without seeing the
> code. It would indeed be nice if the gstreamer-gl sink just worked with
> Clutter though afaik it is really orientated to being the single
> renderer targetting a X drawable (its subcalsses the base X thingy
> iirc?) and working around that could be the difficult part. Im no
> gstreamer expert however.

I've created a git branch so you can take a look at the code if you
want:

git://people.freedesktop.org/~fargiolas/gst-plugins-gl

Just pull the "clutter" branch, relevant code is
gst/gl/gstglcluttersink.c and test/examples/cluttertest.c (note that's
still a bit rough because it's just some code I've written to
experiment)
I'm not sure I understand what you mean with "subclassing the X thingy"
but all the X stuff is in gst-libs/gst/gl/gstgldisplay.c, so you can
take a look there too if you want.
I'm a bit discouraged.. Will it be a great bottleneck to retrieve the
pixel data from video memory and create the cluttertexture from that?
something like clutter-gst?
Thanks!

Filippo

PS.

If it can be someway useful to understand the problem, moving the mouse
over the clutter window outputs this:
(tests/examples/cluttertest:6293): Clutter-CRITICAL **:
clutter_id_pool_lookup: assertion `id < id_pool->array->len' failed

Attachment: signature.asc
Description: Questa è una parte del messaggio firmata digitalmente

Reply via email to