[PATCH] D56011: [x86] lower extracted fadd/fsub to horizontal vector math

[Sanjay Patel via Phabricator via llvm-commits]

spatel updated this revision to Diff 180181.
spatel added a comment.
This revision is now accepted and ready to land.

Patch updated:
Include cost model and SLP test changes. Should we be overriding the cost of scalar FADD/FSUB, so we don't have these diffs?

Without that, it looks like we fall back to this questionable logic:

  // If the operation is custom lowered, then assume that the code is twice
  // as expensive.


CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D56011/new/

https://reviews.llvm.org/D56011

Files:
  lib/Target/X86/X86ISelLowering.cpp
  test/Analysis/CostModel/X86/arith-fp.ll
  test/Analysis/CostModel/X86/reduction.ll
  test/CodeGen/X86/haddsub-undef.ll
  test/CodeGen/X86/haddsub.ll
  test/Transforms/SLPVectorizer/X86/addsub.ll
  test/Transforms/SLPVectorizer/X86/hadd.ll
  test/Transforms/SLPVectorizer/X86/hsub.ll

-------------- next part --------------
A non-text attachment was scrubbed...
Name: D56011.180181.patch
Type: text/x-patch
Size: 62367 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20190104/db20298c/attachment-0001.bin>