diff options
| author | Matt Arsenault <Matthew.Arsenault@amd.com> | 2025-05-19 21:44:44 +0200 |
|---|---|---|
| committer | GitHub <noreply@github.com> | 2025-05-19 21:44:44 +0200 |
| commit | 36018494fdb9e92e0f61f6937e5ecd3a4472677f (patch) | |
| tree | 405e1015b7c5bdbde9354dcd25f76d38df31d145 /clang/lib/Serialization/ModuleManager.cpp | |
| parent | 2b7cc2b03ea858633016cd16a1630be7fc0db837 (diff) | |
| download | llvm-36018494fdb9e92e0f61f6937e5ecd3a4472677f.zip llvm-36018494fdb9e92e0f61f6937e5ecd3a4472677f.tar.gz llvm-36018494fdb9e92e0f61f6937e5ecd3a4472677f.tar.bz2 | |
AMDGPU: Check for subreg match when folding through reg_sequence (#140582)
We need to consider the use instruction's intepretation of the bits,
not the defined immediate without use context. This will regress
some cases where we previously coud match f64 inline constants. We
can restore them by either using pseudo instructions to materialize f64
constants, or recognizing reg_sequence decomposed into 32-bit pieces for them
(which essentially means recognizing every other input is a 0).
Fixes #139908
Diffstat (limited to 'clang/lib/Serialization/ModuleManager.cpp')
0 files changed, 0 insertions, 0 deletions