projects / platform / upstream / mesa.git / commit
? search:
summary | shortlog | log | commit | commitdiff | tree
(parent: c3b86bc) | patch
agx: Handle 8-bit vecs
authorAlyssa Rosenzweig <alyssa@rosenzweig.io>
Thu, 6 Jul 2023 20:26:37 +0000 (16:26 -0400)
committerMarge Bot <emma+marge@anholt.net>
Fri, 11 Aug 2023 20:31:27 +0000 (20:31 +0000)
commit5f3d784c6c4476db58ebc4d0d39ec2efa5649c37
tree2ad1d48c2181e58339a5caff856d08262f94e98btree | snapshot
parentc3b86bcbbcebf08701859f9c56c5965d76d17adacommit | diff
agx: Handle 8-bit vecs

These should "just" work, promoting the 8-bit channels to 16-bit registers
internally, allowing us to use our 8-bit stores with 8-bit data vectors packed
in 16-bit registers. All other non-conversion ALU gets lowered by the previous
patch, this is just needed for simple things like nir_op_vec of lowered math
passed to a vectorized store.

Signed-off-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24635>
src/asahi/compiler/agx_compile.c diff | blob | history
Domain: Graphics System / GL;