[all-commits] [llvm/llvm-project] 7a3d99: [X86][AVX] Add HADDPD test case for PR44694

[Simon Pilgrim via All-commits]

  Branch: refs/heads/master
  Home:   https://github.com/llvm/llvm-project
  Commit: 7a3d994880cfe1551b16b2369ba0fcf2c29d7a9e
      https://github.com/llvm/llvm-project/commit/7a3d994880cfe1551b16b2369ba0fcf2c29d7a9e
  Author: Simon Pilgrim <llvm-dev at redking.me.uk>
  Date:   2020-03-21 (Sat, 21 Mar 2020)

  Changed paths:
    M llvm/test/CodeGen/X86/haddsub-undef.ll

  Log Message:
  -----------
  [X86][AVX] Add HADDPD test case for PR44694


  Commit: 25eb9056d7fef511004952d176016948e3e09f40
      https://github.com/llvm/llvm-project/commit/25eb9056d7fef511004952d176016948e3e09f40
  Author: Simon Pilgrim <llvm-dev at redking.me.uk>
  Date:   2020-03-21 (Sat, 21 Mar 2020)

  Changed paths:
    M llvm/lib/Target/X86/X86ISelLowering.cpp
    M llvm/test/CodeGen/X86/haddsub-undef.ll
    M llvm/test/CodeGen/X86/var-permute-256.ll
    M llvm/test/CodeGen/X86/vector-reduce-mul.ll

  Log Message:
  -----------
  [X86] getTargetShuffleAndZeroables - add insert_subvector(undef, sub, c) handling.

We often widen xmm/ymm vectors to ymm/zmm by insertion into an undef base vector. By letting getTargetShuffleAndZeroables track the undef elts we can help avoid a lot of unnecessary cross-lane shuffles.

Fixes PR44694


Compare: https://github.com/llvm/llvm-project/compare/684ee2057f5d...25eb9056d7fe