[PATCH 00/10] First Glamor core acceleration attempt

[Keith Packard]

Markus Wick <markus at selfnet.de> writes:

> eg ironlake. But I more think about new GPU where the driver just isn't 
> in such a good state. It would be nice to use glamor on very new gpus 
> without requiring a full features gl3.3 driver.

Sure, having support for some older version would be good. However, I'd
like to avoid the combinatorial explosion of features here -- can we
pick a couple of reasonable 'profiles' for drivers and then just write
two versions of the necessary code? That way, it would be easier to test
as well; just flip the switch at detection time and you'd be able to run
the simpler code.

I'd suggest just picking a particular OpenGL version for the older
version, that way it's at least easy to explain what the requirements
are. Of course, we'd also have to pick a GLES version if we want to
support that too. Still, three paths would be easier to test than the
current adventure where every function should have a long list of
feature tests before doing anything.

-- 
keith.packard at intel.com
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 810 bytes
Desc: not available
URL: <http://lists.x.org/archives/xorg-devel/attachments/20140322/36606053/attachment.sig>