Files
clang-p2996/llvm/test/Transforms/VectorCombine/X86
Simon Pilgrim 1ad4f887bd [CostModel][X86] Improve accuracy of vector non-uniform shift costs on XOP/AVX2 targets
By llvm-mca analysis, Haswell/Broadwell has a non-uniform vector shift recip-throughput cost of the AVX2 targets at 2 for both 128 and 256-bit vectors - XOP capable targets have better 128-bit vector shifts so improve the fallback in those cases.
2021-05-24 14:18:21 +01:00
..