[PATCH] D143256: [SDAG] Check fminnum for non zero operand.

[Sam Parker via Phabricator via llvm-commits]

samparker created this revision.
samparker added reviewers: spatel, aheejin, dmgreen.
Herald added subscribers: pmatos, asb, StephenFan, hiraditya, jgravelle-google, sbc100, dschuff.
Herald added a project: All.
samparker requested review of this revision.
Herald added a project: LLVM.

Following on from D143106 <https://reviews.llvm.org/D143106>...

Currently, in TargetLowering, if the target does not support fminnum, we lower to fminimum if neither operand could be a NaN. But this isn't quite correct because fminnum and fminimum treat +/-0 differently; so, we need to prove that one of the operands isn't a zero.


Repository:
  rG LLVM Github Monorepo

https://reviews.llvm.org/D143256

Files:
  llvm/lib/CodeGen/SelectionDAG/TargetLowering.cpp
  llvm/test/CodeGen/ARM/lower-vmax.ll
  llvm/test/CodeGen/WebAssembly/f32.ll
  llvm/test/CodeGen/WebAssembly/f64.ll
  llvm/test/CodeGen/WebAssembly/simd-arith.ll

-------------- next part --------------
A non-text attachment was scrubbed...
Name: D143256.494568.patch
Type: text/x-patch
Size: 18039 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20230203/294e34f2/attachment.bin>