projects / mesa.git / commit
? search:
summary | shortlog | log | commit | commitdiff | tree
(parent: e652ca4) | patch
freedreno: a2xx: use nir_lower_alu_to_scalar instead of lowering pass
authorJonathan Marek <jonathan@marek.ca>
Thu, 1 Aug 2019 19:52:58 +0000 (15:52 -0400)
committerJonathan Marek <jonathan@marek.ca>
Fri, 2 Aug 2019 15:58:22 +0000 (15:58 +0000)
commitfb5c3db0ab0ae1b7944cc982192909165cb96772
tree0fb260386c8729067f1b4a0e72b13a7bded1676btree
parente652ca4e0b4dc773e0754993a5bea4fad1a2e8a1commit | diff
freedreno: a2xx: use nir_lower_alu_to_scalar instead of lowering pass

nir_lower_alu_to_scalar can now be used to only lower certain ops, so we
don't need the custom pass. And we can lower fall_equal/fany_nequal with
lower_vector_cmp instead.

Signed-off-by: Jonathan Marek <jonathan@marek.ca>
Reviewed-by: Rob Clark <robdclark@chromium.org>
src/gallium/drivers/freedreno/Makefile.sources diff | blob | history
src/gallium/drivers/freedreno/a2xx/ir2_nir.c diff | blob | history
src/gallium/drivers/freedreno/a2xx/ir2_nir_lower_scalar.c [deleted file] blob | history
src/gallium/drivers/freedreno/meson.build diff | blob | history