[PATCH] D43733: [X86][SSE] Reduce FADD/FSUB/FMUL costs on later targets (PR36280)

[Simon Pilgrim via Phabricator via llvm-commits]

RKSimon created this revision.
RKSimon added reviewers: ABataev, craig.topper, spatel.

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.


Repository:
  rL LLVM

https://reviews.llvm.org/D43733

Files:
  lib/Target/X86/X86TargetTransformInfo.cpp
  test/Analysis/CostModel/X86/arith-fp.ll
  test/Analysis/CostModel/X86/intrinsic-cost.ll
  test/Transforms/SLPVectorizer/X86/PR36280.ll
  test/Transforms/SLPVectorizer/X86/cse.ll
  test/Transforms/SLPVectorizer/X86/horizontal.ll
  test/Transforms/SLPVectorizer/X86/reorder_phi.ll
  test/Transforms/SLPVectorizer/X86/simplebb.ll

-------------- next part --------------
A non-text attachment was scrubbed...
Name: D43733.135800.patch
Type: text/x-patch
Size: 42316 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20180224/82c9118f/attachment-0001.bin>