projects / platform / upstream / llvm.git / commit
? search:
summary | shortlog | log | commit | commitdiff | tree
(parent: a85bd1b) | patch
AMDGPU/MC: Add assembler tests for v2f32 and v2b32 with imm operand
authorPetar Avramovic <Petar.Avramovic@amd.com>
Wed, 8 Feb 2023 12:39:56 +0000 (13:39 +0100)
committerPetar Avramovic <Petar.Avramovic@amd.com>
Thu, 9 Feb 2023 11:16:46 +0000 (12:16 +0100)
commit524c89e837ec19529f6190d1def7cb7cd345166b
treeecada415b3c22c34110c46b40da275bbf3450d2btree | snapshot
parenta85bd1b11934fc707e7d36a04a92b7ac4e45d35bcommit | diff
AMDGPU/MC: Add assembler tests for v2f32 and v2b32 with imm operand

Add test coverage for https://github.com/llvm/llvm-project/issues/60563.
D142636 introduced a bug: incorrect disassembly of floating point inline
constants for v_pk_mov_b32, v_pk_add_f32, v_pk_mul_f32 and v_pk_fma_f32.
Precommit for D143574.

Differential Revision: https://reviews.llvm.org/D143573
llvm/test/MC/AMDGPU/gfx90a_asm_features.s diff | blob | history
llvm/test/MC/Disassembler/AMDGPU/gfx90a_features.txt diff | blob | history
Domain: System / Toolchain;