[all-commits] [llvm/llvm-project] a4c828: [AArch64] Precommit tests to check more ld1r vecto...

[sjoerdmeijer via All-commits]

  Branch: refs/heads/main
  Home:   https://github.com/llvm/llvm-project
  Commit: a4c828a9e2b08c3606046aff30ebeecb558f226f
      https://github.com/llvm/llvm-project/commit/a4c828a9e2b08c3606046aff30ebeecb558f226f
  Author: Sjoerd Meijer <smeijer at nvidia.com>
  Date:   2023-03-01 (Wed, 01 Mar 2023)

  Changed paths:
    A llvm/test/CodeGen/AArch64/neon-vector-splat.ll

  Log Message:
  -----------
  [AArch64] Precommit tests to check more ld1r vector splat patterns in D145004.


  Commit: 2b462eb98d6777a2fb9385eea74bab9a5a6d3b95
      https://github.com/llvm/llvm-project/commit/2b462eb98d6777a2fb9385eea74bab9a5a6d3b95
  Author: Sjoerd Meijer <smeijer at nvidia.com>
  Date:   2023-03-01 (Wed, 01 Mar 2023)

  Changed paths:
    M llvm/lib/Target/AArch64/AArch64InstrInfo.td
    M llvm/test/CodeGen/AArch64/neon-vector-splat.ll

  Log Message:
  -----------
  [AArch64] More patterns to generate LD1R vector splats

We are missing patterns to generate vector splats using LD1R. A shuffle vector
with all 0s is a vector splat if the operands are a load and undef for which
we can generate a LD1R.

Differential Revision: https://reviews.llvm.org/D145004


Compare: https://github.com/llvm/llvm-project/compare/daf41890709b...2b462eb98d67