[all-commits] [llvm/llvm-project] 03f852: [AArch64] Improve cost model for legal subvec inse...

[Graham Hunter via All-commits]

  Branch: refs/heads/main
  Home:   https://github.com/llvm/llvm-project
  Commit: 03f852f704c343abc4a36db194027461b7b35afa
      https://github.com/llvm/llvm-project/commit/03f852f704c343abc4a36db194027461b7b35afa
  Author: Graham Hunter <graham.hunter at arm.com>
  Date:   2024-03-04 (Mon, 04 Mar 2024)

  Changed paths:
    M llvm/lib/Target/AArch64/AArch64TargetTransformInfo.cpp
    M llvm/test/Analysis/CostModel/AArch64/sve-intrinsics.ll
    M llvm/test/Transforms/LoopUnroll/AArch64/scalable-vec-ins-ext.ll

  Log Message:
  -----------
  [AArch64] Improve cost model for legal subvec insert/extract (#81135)

Currently we model subvector inserts and extracts as shuffles,
potentially going as far as scalarizing. If the types are legal then
they can just be simple zip/unzip operations, or possible even no-ops.
Change the cost to a relatively small one to ensure that simple loops
featuring such operations between fixed and scalable vector types that
are effectively the same at a given sve width can be unrolled and
further optimized.



To unsubscribe from these emails, change your notification settings at https://github.com/llvm/llvm-project/settings/notifications