git.libre-soc.org Git - mesa.git/commit

author	Jason Ekstrand <jason.ekstrand@intel.com>
	Wed, 7 Nov 2018 19:43:40 +0000 (13:43 -0600)
committer	Jason Ekstrand <jason.ekstrand@intel.com>
	Wed, 5 Dec 2018 21:03:07 +0000 (15:03 -0600)
commit	dca6cd9ce65100896976e913bf72c2c68ea4e1a7
tree	069181199b9b4f00d8c44f1ae507412c28ccc81f	tree
parent	be98b1db3899121f2d0cea009ef6430b13589032	commit \| diff

nir: Make boolean conversions sized just like the others

Instead of a single i2b and b2i, we now have i2b32 and b2iN where N is
one if 8, 16, 32, or 64. This leads to having a few more opcodes but
now everything is consistent and booleans aren't a weird special case
anymore.

Reviewed-by: Connor Abbott <cwabbott0@gmail.com>

20 files changed:

src/amd/common/ac_nir_to_llvm.c		diff \| blob \| history
src/broadcom/compiler/nir_to_vir.c		diff \| blob \| history
src/compiler/glsl/glsl_to_nir.cpp		diff \| blob \| history
src/compiler/nir/nir.h		diff \| blob \| history
src/compiler/nir/nir_algebraic.py		diff \| blob \| history
src/compiler/nir/nir_builder.h		diff \| blob \| history
src/compiler/nir/nir_lower_idiv.c		diff \| blob \| history
src/compiler/nir/nir_lower_int64.c		diff \| blob \| history
src/compiler/nir/nir_opcodes.py		diff \| blob \| history
src/compiler/nir/nir_opcodes_c.py		diff \| blob \| history
src/compiler/nir/nir_opt_algebraic.py		diff \| blob \| history
src/compiler/nir/nir_opt_if.c		diff \| blob \| history
src/compiler/nir/nir_search.c		diff \| blob \| history
src/compiler/nir/nir_search.h		diff \| blob \| history
src/compiler/spirv/vtn_glsl450.c		diff \| blob \| history
src/freedreno/ir3/ir3_compiler_nir.c		diff \| blob \| history
src/gallium/drivers/vc4/vc4_program.c		diff \| blob \| history
src/intel/compiler/brw_fs_nir.cpp		diff \| blob \| history
src/intel/compiler/brw_vec4_nir.cpp		diff \| blob \| history
src/mesa/program/prog_to_nir.c		diff \| blob \| history

RSS Atom