[all-commits] [llvm/llvm-project] e18b97: [AArch64][InstCombine] Simplify repeated complex p...

[Matthew Devereau via All-commits]

  Branch: refs/heads/main
  Home:   https://github.com/llvm/llvm-project
  Commit: e18b971685fb349299583d95716244f34f974ef8
      https://github.com/llvm/llvm-project/commit/e18b971685fb349299583d95716244f34f974ef8
  Author: Matt Devereau <matthew.devereau at arm.com>
  Date:   2023-01-16 (Mon, 16 Jan 2023)

  Changed paths:
    M llvm/lib/Target/AArch64/AArch64TargetTransformInfo.cpp
    M llvm/test/CodeGen/AArch64/sve-intrinsics-perm-select.ll
    A llvm/test/Transforms/InstCombine/AArch64/sve-intrinsic-dupqlane.ll

  Log Message:
  -----------
  [AArch64][InstCombine] Simplify repeated complex patterns in dupqlane

Repeated floating-point complex patterns in dupqlane such as (f32 a, f32 b, f32
a, f32 b) can be simplified to shufflevector(f64(a, b), undef, 0)