diff options
| author | Simon Pilgrim <llvm-dev@redking.me.uk> | 2021-11-07 21:24:55 +0000 |
|---|---|---|
| committer | Simon Pilgrim <llvm-dev@redking.me.uk> | 2021-11-07 21:24:55 +0000 |
| commit | 55e4cd8485dc6308211f85c60559a373ae96873e (patch) | |
| tree | eccd8107960097db6aeb0561e7487f3bf205280a /llvm/lib/Transforms/Utils/LoopUtils.cpp | |
| parent | 54c563474ab69fa8fbbbf68da6fab84cb0ee6e6b (diff) | |
| download | llvm-55e4cd8485dc6308211f85c60559a373ae96873e.zip llvm-55e4cd8485dc6308211f85c60559a373ae96873e.tar.gz llvm-55e4cd8485dc6308211f85c60559a373ae96873e.tar.bz2 | |
[X86][AVX2] Recognise 256-bit truncation shuffles and mask 256-bit source
For v8i16 shuffle patterns that are lowered with AND+PACKUS, check to see if the sources are from a 256-bit vector and perform the masking using BLENDW at the 256-bit level.
With the test changes we can see more examples of duplicate XMM/YMM zero vectors (PR26018) :(
Diffstat (limited to 'llvm/lib/Transforms/Utils/LoopUtils.cpp')
0 files changed, 0 insertions, 0 deletions