[all-commits] [llvm/llvm-project] dc0365: InstCombine: Fold known-qnan results to a literal nan

[Matt Arsenault via All-commits]

  Branch: refs/heads/users/arsenm/instcombine/simplify-demanded-fpclass-allow-fold-to-qnan
  Home:   https://github.com/llvm/llvm-project
  Commit: dc03657035a9f32b58a94b5673f80e1e37e7e767
      https://github.com/llvm/llvm-project/commit/dc03657035a9f32b58a94b5673f80e1e37e7e767
  Author: Matt Arsenault <Matthew.Arsenault at amd.com>
  Date:   2026-01-15 (Thu, 15 Jan 2026)

  Changed paths:
    M llvm/lib/Transforms/InstCombine/InstCombineSimplifyDemanded.cpp
    M llvm/test/Transforms/InstCombine/simplify-demanded-fpclass-exp.ll
    M llvm/test/Transforms/InstCombine/simplify-demanded-fpclass-extractelement.ll
    M llvm/test/Transforms/InstCombine/simplify-demanded-fpclass-fadd.ll
    M llvm/test/Transforms/InstCombine/simplify-demanded-fpclass-fmul.ll
    M llvm/test/Transforms/InstCombine/simplify-demanded-fpclass-fpext.ll
    M llvm/test/Transforms/InstCombine/simplify-demanded-fpclass-fptrunc-round.ll
    M llvm/test/Transforms/InstCombine/simplify-demanded-fpclass-fptrunc.ll
    M llvm/test/Transforms/InstCombine/simplify-demanded-fpclass-frexp.ll
    M llvm/test/Transforms/InstCombine/simplify-demanded-fpclass-insertelement.ll
    M llvm/test/Transforms/InstCombine/simplify-demanded-fpclass-maximum.ll
    M llvm/test/Transforms/InstCombine/simplify-demanded-fpclass-maximumnum.ll
    M llvm/test/Transforms/InstCombine/simplify-demanded-fpclass-minimum.ll
    M llvm/test/Transforms/InstCombine/simplify-demanded-fpclass-minimumnum.ll
    M llvm/test/Transforms/InstCombine/simplify-demanded-fpclass-rounding-intrinsics.ll
    M llvm/test/Transforms/InstCombine/simplify-demanded-fpclass-shufflevector.ll
    M llvm/test/Transforms/InstCombine/simplify-demanded-fpclass-sqrt.ll
    M llvm/test/Transforms/InstCombine/simplify-demanded-fpclass.ll

  Log Message:
  -----------
  InstCombine: Fold known-qnan results to a literal nan

Previously we only considered fcNan to fold to qnan for canonicalizing
results, ignoring the simpler case where we know the nan is already
quiet.



To unsubscribe from these emails, change your notification settings at https://github.com/llvm/llvm-project/settings/notifications