[all-commits] [llvm/llvm-project] 261075: llvm-reduce: Handle reducing FP values to nan

[Matt Arsenault via All-commits]

  Branch: refs/heads/main
  Home:   https://github.com/llvm/llvm-project
  Commit: 261075590b2bde348922b402bf982c33fb6475a8
      https://github.com/llvm/llvm-project/commit/261075590b2bde348922b402bf982c33fb6475a8
  Author: Matt Arsenault <Matthew.Arsenault at amd.com>
  Date:   2022-06-27 (Mon, 27 Jun 2022)

  Changed paths:
    M llvm/test/tools/llvm-reduce/remove-operands-fp.ll
    M llvm/tools/llvm-reduce/DeltaManager.cpp
    M llvm/tools/llvm-reduce/deltas/ReduceOperands.cpp
    M llvm/tools/llvm-reduce/deltas/ReduceOperands.h

  Log Message:
  -----------
  llvm-reduce: Handle reducing FP values to nan

Prefer 0/1 over NaN, but it may make more sense to invert this as FP
operations with nan inputs can universally be folded into something
else.