[all-commits] [llvm/llvm-project] 0d15d4: [SLP] use operand index abstraction for number of ...

[RotateRight via All-commits]

  Branch: refs/heads/master
  Home:   https://github.com/llvm/llvm-project
  Commit: 0d15d4b6f43a3355c1d618766c8e550cfe1481d0
      https://github.com/llvm/llvm-project/commit/0d15d4b6f43a3355c1d618766c8e550cfe1481d0
  Author: Sanjay Patel <spatel at rotateright.com>
  Date:   2020-12-22 (Tue, 22 Dec 2020)

  Changed paths:
    M llvm/lib/Transforms/Vectorize/SLPVectorizer.cpp

  Log Message:
  -----------
  [SLP] use operand index abstraction for number of operands

I think this is NFC currently, but the bug would be exposed
when we allow binary intrinsics (maxnum, etc) as candidates
for reductions.

The code in matchAssociativeReduction() is using
OperationData::getNumberOfOperands() when comparing whether
the "EdgeToVisit" iterator is in-bounds, so this code must
use the same (potentially offset) operand value to set
the "EdgeToVisit".


  Commit: f6929c01952b3f144df620544ed937e801b9c945
      https://github.com/llvm/llvm-project/commit/f6929c01952b3f144df620544ed937e801b9c945
  Author: Sanjay Patel <spatel at rotateright.com>
  Date:   2020-12-22 (Tue, 22 Dec 2020)

  Changed paths:
    M llvm/test/Transforms/SLPVectorizer/X86/fmaxnum.ll
    M llvm/test/Transforms/SLPVectorizer/X86/fminnum.ll

  Log Message:
  -----------
  [SLP] add reduction tests for maxnum/minnum intrinsics; NFC


Compare: https://github.com/llvm/llvm-project/compare/57ffbe020af6...f6929c01952b