On Tue, Jul 10, 2012 at 11:15:23AM +0100, Chris Wilson wrote: > In order to support snoopable memory on non-LLC architectures (so that > we can bind vgem objects into the i915 GATT for example), we have to > avoid the prefetcher on the GPU from crossing memory domains and so > prevent allocation of a snoopable PTE immediately following an uncached > PTE. To do that, we need to extend the range allocator with support for > tracking and segregating different node colours. > > This will be used by i915 to segregate memory domains within the GTT. > > v2: Now with more drm_mm helpers and less driver interference. > > Signed-off-by: Chris Wilson <ch...@chris-wilson.co.uk> > Cc: Dave Airlie <airl...@redhat.com > Cc: Benjamin Herrenschmidt <b...@kernel.crashing.org> > Cc: Ben Skeggs <bske...@redhat.com> > Cc: Jerome Glisse <jgli...@redhat.com> > Cc: Alex Deucher <alexander.deuc...@amd.com> > Cc: Daniel Vetter <daniel.vet...@ffwll.ch> > Cc: dri-de...@lists.freedesktop.org
Imo we should ditch the rather useless best_match and maybe also fold the _range variants into the generic ones for most cases, but that's stuff for other patches. So Reviewed-by: Daniel Vetter <daniel.vet...@ffwll.ch> Cheers, Daniel -- Daniel Vetter Mail: dan...@ffwll.ch Mobile: +41 (0)79 365 57 48 _______________________________________________ Intel-gfx mailing list Intel-gfx@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/intel-gfx
