[all-commits] [llvm/llvm-project] e3c1a2: [𝘀𝗽𝗿] initial version

[Florian Hahn via All-commits]

  Branch: refs/heads/users/fhahn/lv-reverse-mask-up-front-not-when-creating-vector-pointer
  Home:   https://github.com/llvm/llvm-project
  Commit: e3c1a20a1b5a115a75d83f5783ba64927607f427
      https://github.com/llvm/llvm-project/commit/e3c1a20a1b5a115a75d83f5783ba64927607f427
  Author: Florian Hahn <flo at fhahn.com>
  Date:   2023-11-17 (Fri, 17 Nov 2023)

  Changed paths:
    M llvm/lib/Transforms/Vectorize/LoopVectorize.cpp
    M llvm/test/Transforms/LoopVectorize/AArch64/sve-vector-reverse-mask4.ll
    M llvm/test/Transforms/LoopVectorize/AArch64/vector-reverse-mask4.ll
    M llvm/test/Transforms/LoopVectorize/X86/masked_load_store.ll

  Log Message:
  -----------
  [𝘀𝗽𝗿] initial version

Created using spr 1.3.4


  Commit: 1f729e0bc7a7012fa3a7618100f0865e4e32fc0d
      https://github.com/llvm/llvm-project/commit/1f729e0bc7a7012fa3a7618100f0865e4e32fc0d
  Author: Florian Hahn <flo at fhahn.com>
  Date:   2023-11-17 (Fri, 17 Nov 2023)

  Changed paths:
    M llvm/test/Transforms/LoopVectorize/X86/masked_load_store.ll

  Log Message:
  -----------
  [LV] Reverse mask up front, not when creating vector pointer.

Reverse mask early on when populating BlockInMask. This will enable
separating mask management and address computation from the memory
recipes in the future and is also needed to enable explicit unrolling in
VPlan.

Pull Request: https://github.com/llvm/llvm-project/pull/72163


Compare: https://github.com/llvm/llvm-project/compare/035b334598b4...1f729e0bc7a7