pbuffers, extensions to DRM

[Keith Whitwell]

Jon Smirl wrote:
> --- Michel Dänzer <michel@daenzer.net> wrote:
> 
>>I think at least the core of it definitely has to be in the kernel for
>>enforceability, cleanup on process death etc.
>>
> 
> 
> You're and expert on the radeon driver, want to help code this up? I'm still
> bogged down with the mode code and making the VM86 work right.

Have a look at the allocator in for the i830 driver in the agp-i865-0-1-branch 
of dri.  It's a version of the existing radeon agp memory manager, but one 
which integrates with clientside texture allocation (a la. common/texmem.c).

Keith