diff options
| author | Simon Pilgrim <llvm-dev@redking.me.uk> | 2023-04-20 19:38:40 +0100 |
|---|---|---|
| committer | Simon Pilgrim <llvm-dev@redking.me.uk> | 2023-04-20 19:38:51 +0100 |
| commit | 4060042384209c9e7077b7024a73c70ef0be0b91 (patch) | |
| tree | cdcab3f8df8ed2e7793fd58cd3531d71ad3ecc20 /llvm/lib/Analysis/ModuleSummaryAnalysis.cpp | |
| parent | 16949c5c48ab475d34547d4bbbe91df7efe73120 (diff) | |
| download | llvm-4060042384209c9e7077b7024a73c70ef0be0b91.zip llvm-4060042384209c9e7077b7024a73c70ef0be0b91.tar.gz llvm-4060042384209c9e7077b7024a73c70ef0be0b91.tar.bz2 | |
[CostModel][X86] Improve i8 and vXi8 MUL costs
We were treating vXi8 multiply as the sum of a trunc(mul(extend(),extend())) which diverged from the costs from llvm-mcaonce we extended beyond legal types
Use a modified version of the D103695 script to determine more accurate throughput/latency/codesize/size-latency cost estimates
Helps address some of the regressions identified in D148806
Diffstat (limited to 'llvm/lib/Analysis/ModuleSummaryAnalysis.cpp')
0 files changed, 0 insertions, 0 deletions