This is an archive of the discontinued LLVM Phabricator instance.

AMDGPU/MC: Add assembler tests for v2f32 and v2b32 with imm operand
ClosedPublic

Authored by Petar.Avramovic on Feb 8 2023, 4:58 AM.

Details

Summary

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.

Diff Detail

Event Timeline

Petar.Avramovic created this revision.Feb 8 2023, 4:58 AM
Herald added a project: Restricted Project. · View Herald TranscriptFeb 8 2023, 4:58 AM
Petar.Avramovic requested review of this revision.Feb 8 2023, 4:58 AM
Herald added a project: Restricted Project. · View Herald TranscriptFeb 8 2023, 4:58 AM
dp added inline comments.Feb 8 2023, 7:49 AM
llvm/test/MC/Disassembler/AMDGPU/gfx90a_features.txt
54

It is unusual to add invalid tests with the purpose of fixing them in a separate patch.
At a minimum, I think you should mention this in the description and make these patches dependent (stacked). Or they could be merged.
Aside from that, both patches look good, thanks.

dp accepted this revision.Feb 8 2023, 8:15 AM
This revision is now accepted and ready to land.Feb 8 2023, 8:15 AM