Jakob Bornecrantz wrote:
> On Mon, Apr 6, 2009 at 4:52 PM, José Fonseca <[email protected]> wrote:
>> On Sat, 2009-04-04 at 04:02 -0700, Jose Fonseca wrote:
>>> All the python state tracker requires is two functions: one that
>>> creates a pipe_screen and another that creates a pipe_context. You can
>>> implement these by provifing your version of  st_hardpipe_winsys.c,
>>> and linking against the "pyst" library.  However, the python state
>>> tracker  is headless (it does not know anything about X or any window
>>> system). So for integrating it with a DRI-based driver you need to
>>> include all the DRM/DDX glue code yourself...
>>>
>>> I can provide you skeleton code/SConscript that does the above.
>> Pushed.

Rock. This looks awesome.

> If you used the drm_api_hooks structs this could be reused for all
> drivers exporting that api.

Well, I'm going to have to tweak this a bit anyway, because r300 can't
work with DRI1 due to limitations further down the line and glisse,
airlied, and I agree that supporting DRI1 officially in Gallium just
doesn't sound fun. But this is pretty cool.

I wonder if it makes sense to directly bind the Python stuff to drm_api
rather than piggyback it on DRI2 or EGL...

~ C.

------------------------------------------------------------------------------
This SF.net email is sponsored by:
High Quality Requirements in a Collaborative Environment.
Download a free trial of Rational Requirements Composer Now!
http://p.sf.net/sfu/www-ibm-com
_______________________________________________
Mesa3d-dev mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/mesa3d-dev

Reply via email to