diff options
| author | Simon Pilgrim <llvm-dev@redking.me.uk> | 2021-05-25 17:18:21 +0100 |
|---|---|---|
| committer | Simon Pilgrim <llvm-dev@redking.me.uk> | 2021-05-25 17:31:45 +0100 |
| commit | def6269779aff0d4f1994cf73f7910add044ff72 (patch) | |
| tree | 6b69859d85092d3b188b41611ea6cdac099d32db /llvm/lib/Bitcode/Reader/BitcodeReader.cpp | |
| parent | 8e83ff58c9078c32bbc8d8ea456fa8fc0580761b (diff) | |
| download | llvm-def6269779aff0d4f1994cf73f7910add044ff72.zip llvm-def6269779aff0d4f1994cf73f7910add044ff72.tar.gz llvm-def6269779aff0d4f1994cf73f7910add044ff72.tar.bz2 | |
[CostModel][X86] Improve accuracy of 256-bit non-uniform vector shifts on AVX1
Determined from llvm-mca analysis, AVX1 capable targets have a higher throughput for VPBLENDVB and shuffle ops, making it cheaper to perform shift+shuffle/select shift patterns.
Diffstat (limited to 'llvm/lib/Bitcode/Reader/BitcodeReader.cpp')
0 files changed, 0 insertions, 0 deletions