[PATCH] D90901: [DAGCombiner] Don't fold ((fma (fneg X), Y, (fneg Z)) to fneg (fma X, Y, Z))

[Jim Lin via Phabricator via llvm-commits]

Jim updated this revision to Diff 345677.
Jim added a comment.
Herald added a subscriber: pengfei.

Check if it has no signed zeros flag (nsz) or no NoSignedZerosFPMath option
in getNegatedExpression for x86.


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D90901/new/

https://reviews.llvm.org/D90901

Files:
  llvm/lib/Target/X86/X86ISelLowering.cpp
  llvm/test/CodeGen/X86/avx2-fma-fneg-combine.ll
  llvm/test/CodeGen/X86/fma-fneg-combine.ll
  llvm/test/CodeGen/X86/fma-signed-zero.ll
  llvm/test/CodeGen/X86/fma_patterns.ll
  llvm/test/CodeGen/X86/fma_patterns_wide.ll

-------------- next part --------------
A non-text attachment was scrubbed...
Name: D90901.345677.patch
Type: text/x-patch
Size: 17427 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20210516/92d71c3b/attachment.bin>