diff options
| author | Matt Arsenault <Matthew.Arsenault@amd.com> | 2025-01-30 10:55:21 +0700 |
|---|---|---|
| committer | GitHub <noreply@github.com> | 2025-01-30 10:55:21 +0700 |
| commit | 97a1f494a6778df24cfca817a30b73b1f0a7228a (patch) | |
| tree | d2d7e73c0a586516ce3d00ca9fbc6e445005ec37 /clang/lib/FrontendTool/ExecuteCompilerInvocation.cpp | |
| parent | edb7f6c0da17469f53f38b648380b49c4a5362fb (diff) | |
| download | llvm-97a1f494a6778df24cfca817a30b73b1f0a7228a.zip llvm-97a1f494a6778df24cfca817a30b73b1f0a7228a.tar.gz llvm-97a1f494a6778df24cfca817a30b73b1f0a7228a.tar.bz2 | |
DAG: Avoid breaking legal vector_shuffle with multiple uses (#123712)
Previously this combine would undo AMDGPU's new custom legalization of
wide vector shuffles into 2 element pieces. The comment also
states that this combine is only done before legalization,
but the case with a build_vector source was unconditional.
We probably don't want to do this if the multiple uses are full
scalarization of the vector, but this seems to work well enough.
Scalarizing extracts should have folded out pre-legalize.
Diffstat (limited to 'clang/lib/FrontendTool/ExecuteCompilerInvocation.cpp')
0 files changed, 0 insertions, 0 deletions