projects / platform / upstream / llvm.git / commit
? search:
summary | shortlog | log | commit | commitdiff | tree
(parent: c7545de) | patch
[AMDGPU] Make V_SAT_PK_U8_I16 a True16 Instruction
authorJoe Nash <Joseph.Nash@amd.com>
Fri, 7 Oct 2022 18:56:36 +0000 (14:56 -0400)
committerJoe Nash <Joseph.Nash@amd.com>
Mon, 10 Oct 2022 14:33:49 +0000 (10:33 -0400)
commitebb258d3b0785f6dcc65e1f277d0690891ddc94d
treed5cb7b63027124ff0ecda3fc35e5c19e3da23183tree | snapshot
parentc7545de9b4b7ea2b1a482c70baaecbd155c07bdecommit | diff
[AMDGPU] Make V_SAT_PK_U8_I16 a True16 Instruction

The return type is two u8 packed into a 16 bit VGPR, so this instruction
should be True16.

Reviewed By: dp

Differential Revision: https://reviews.llvm.org/D135478
llvm/lib/Target/AMDGPU/SIInstrInfo.td diff | blob | history
llvm/lib/Target/AMDGPU/VOP1Instructions.td diff | blob | history
llvm/test/MC/AMDGPU/gfx11_asm_vop1_t16_err.s diff | blob | history
llvm/test/MC/AMDGPU/gfx11_asm_vop1_t16_promote.s diff | blob | history
Domain: System / Toolchain;