[all-commits] [llvm/llvm-project] 3e6afa: [CostModel][AArch64] Precommit tests for LD1 singl...

[zino benaissa via All-commits]

  Branch: refs/heads/main
  Home:   https://github.com/llvm/llvm-project
  Commit: 3e6afa77b4ec8b5fb5e0b6349f08e4476892c0f8
      https://github.com/llvm/llvm-project/commit/3e6afa77b4ec8b5fb5e0b6349f08e4476892c0f8
  Author: Sjoerd Meijer <smeijer at nvidia.com>
  Date:   2023-01-19 (Thu, 19 Jan 2023)

  Changed paths:
    M llvm/test/Analysis/CostModel/AArch64/insert-extract.ll

  Log Message:
  -----------
  [CostModel][AArch64] Precommit tests for LD1 single-element to lane. NFC.


  Commit: 68f45796edbdc4331a6be0b6e9d58f01d8b1fba0
      https://github.com/llvm/llvm-project/commit/68f45796edbdc4331a6be0b6e9d58f01d8b1fba0
  Author: Zino Benaissa <zbenaissa at nvidia.com>
  Date:   2023-01-19 (Thu, 19 Jan 2023)

  Changed paths:
    M llvm/lib/Target/AArch64/AArch64ISelLowering.cpp
    A llvm/test/CodeGen/AArch64/sve-fixed-vector-zext.ll

  Log Message:
  -----------
  [AARCH64][SVE] Do not optimize vector conversions

shuffle_vector instructions are serialized targeting SVE fixed vectors, see
https://reviews.llvm.org/D139111. This patch disables
optimizeExtendOrTruncateConversion peepholes that generates shuffle_vector.

Differential Revision: https://reviews.llvm.org/D141439


Compare: https://github.com/llvm/llvm-project/compare/04d4f4b3d4e4...68f45796edbd