diff options
| author | peterbell10 <peterbell10@openai.com> | 2024-11-01 15:02:43 +0000 |
|---|---|---|
| committer | GitHub <noreply@github.com> | 2024-11-01 08:02:43 -0700 |
| commit | b74e588e1f460eb48ceb1a30cf8ac870b7537dcc (patch) | |
| tree | 7132f0fe1173cc1402ca321e9b6f6c07341eb9af /llvm/lib/Object/COFFObjectFile.cpp | |
| parent | 58f525a23ca44f1acdb80ff803fb7b95acf8c7d3 (diff) | |
| download | llvm-b74e588e1f460eb48ceb1a30cf8ac870b7537dcc.zip llvm-b74e588e1f460eb48ceb1a30cf8ac870b7537dcc.tar.gz llvm-b74e588e1f460eb48ceb1a30cf8ac870b7537dcc.tar.bz2 | |
[NVPTX] Don't use stack memory when bitcasting to/from v2i8 (#113928)
`v2i8` is an unsupported type, so we hit the default legalization rules
which perform the bitcast in stack memory and is very inefficient on
GPU.
This adds a custom lowering where we pack `v2i8` into `i16` and from
there use another bitcast node to reach the final desired type. And also
the inverse unpacking `i16` into `v2i8`.
Diffstat (limited to 'llvm/lib/Object/COFFObjectFile.cpp')
0 files changed, 0 insertions, 0 deletions