diff options
| author | Matt Arsenault <Matthew.Arsenault@amd.com> | 2025-09-27 08:24:20 +0900 |
|---|---|---|
| committer | GitHub <noreply@github.com> | 2025-09-27 08:24:20 +0900 |
| commit | 597f93d36b035faeb63f4ba0d61a8b8e25eddaab (patch) | |
| tree | 4fbec98ac9fb08af17a8eab2a663aac47aa85614 /llvm/tools/llvm-diff/lib/DiffConsumer.h | |
| parent | f8d547fa8a038d0adf5c07708d11bbf1d8db3100 (diff) | |
| download | llvm-597f93d36b035faeb63f4ba0d61a8b8e25eddaab.zip llvm-597f93d36b035faeb63f4ba0d61a8b8e25eddaab.tar.gz llvm-597f93d36b035faeb63f4ba0d61a8b8e25eddaab.tar.bz2 | |
AMDGPU: Check if immediate is legal for av_mov_b32_imm_pseudo (#160819)
This is primarily to avoid folding a frame index materialized
into an SGPR into the pseudo; this would end up looking like:
%sreg = s_mov_b32 %stack.0
%av_32 = av_mov_b32_imm_pseudo %sreg
Which is not useful.
Match the check used for the b64 case. This is limited to the
pseudo to avoid regression due to gfx908's special case - it
is expecting to pass here with v_accvgpr_write_b32 for illegal
cases, and stay in the intermediate state with an sgpr input.
This avoids regressions in a future patch.
Diffstat (limited to 'llvm/tools/llvm-diff/lib/DiffConsumer.h')
0 files changed, 0 insertions, 0 deletions