[Xorg] DRI merging

[Alan Cox]

On Sul, 2004-06-13 at 20:35, Jon Smirl wrote:
> The work that would be wasted is extending the XAA 2D drivers to use the 3D
> hardware to accelerate render.

Lots of hardware can do render without 3D operations. Even my 
TV capture/playback card has blit-with-alpha on it. Extending existing
XAA drivers to do render better via DRI may make sense in some cases and
in shorter than two years.

As it is everyone is having to merge DRI and Xorg themselves right now
to get a working useful tree for 2D and 3D. Getting all the code in the
right places makes that work a lot easier, and means for example I can
viably attack S3virge and SiS6326 support. (Which you'll want for
mesa-solo eventually if only to understand the limits of hardware with
only primitive 3D support).

> There are three main solutions to mode/cursors problem that no one can agree on:
> 1) leave fbdev in charge of mode setting and cursor, port it around to other
> architectures.

For Linux I think everyone has accepted that you have to have a basic
fb driver simply to do hotplug. Xorg however has to run everywhere.

> 3) Add a couple of IOCTLs to DRM to support modes/cursors. Do as much of the
> work as possible in user space and just pass final register values to the
> IOCTLs.

DRM supports two platforms of the entire X world. In most of the
proprietary cases the frame buffer layer is vendor controlled and may or
may not be doing this. I guess Alan Coopersmith can comment on this for
Solaris x86 for example.