This implements fast Z clear, Z compression, and HiZ support for r300->r500
GPUs.
It also allows cbzb clears when fast Z clears are being used for the ZB.
It requires a kernel with hyper-z support.
Thanks to Marek Olšák <maraeo@gmail.com>, who started this off, and Alex Deucher at AMD for providing lots of hints.
v2:
squashed zmask ram size fix]
squashed r300g/blitter: fix Z readback when compressed]
v3:
rebase around texture changes in master - .1 fix more bits
v4:
migrated to using u_mm in r300_texture to manage hiz/zmask rams consistently
disabled HiZ when using OQ
flush z-cache before turning hyper-z off
update hyper-z state on dsa state change
store depthclearvalue across cbzb clears and replace it afterwards.
Signed-off-by: Dave Airlie <airlied@redhat.com>
Currently r300g does not save vertex buffer on blitter calls.
It gets away with it because the current Mesa state tracker usually
resets vertex buffers on every draw calls.
However, this is wrong.
nvfx won't be lucky because it needs to use the blitter inside draw
calls.
The interface of util_draw_vertex_buffer looks a bit odd (calling code has to
set vertex elements but not vertex buffers) but due to the way cso state
handling generally works (can't re-bind original vertex element state easily
there) I guess that's ok for now.
Marek Olšák
Dave Airlie
Roland Scheidegger
Luca Barbieri
Michal Krol