tree c98771d34d9e7cb6b64dd20174c63ad471f7942c
parent 35b540858f9b41b527e0de08518b31d0f01907dd
author Simon Pilgrim <llvm-dev@redking.me.uk> 1519683017 +0000
committer Simon Pilgrim <llvm-dev@redking.me.uk> 1519683017 +0000

[X86][SSE] Reduce FADD/FSUB/FMUL costs on later targets (PR36280)

Agner's tables indicate that for SSE42+ targets (Core2 and later) we can reduce the FADD/FSUB/FMUL costs down to 1, which should fix the Himeno benchmark.

Note: the AVX512 FDIV costs look rather dodgy, but this isn't part of this patch.

Differential Revision: https://reviews.llvm.org/D43733

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@326133 91177308-0d34-0410-b5e6-96231b3b80d8