[all-commits] [llvm/llvm-project] c3a10d: [AArch64] Disable consecutive store merging when N...

[Benjamin Maxwell via All-commits]

  Branch: refs/heads/main
  Home:   https://github.com/llvm/llvm-project
  Commit: c3a10dc8498b1e501f5a32b082b63b0c1fc499a5
      https://github.com/llvm/llvm-project/commit/c3a10dc8498b1e501f5a32b082b63b0c1fc499a5
  Author: Benjamin Maxwell <benjamin.maxwell at arm.com>
  Date:   2024-10-11 (Fri, 11 Oct 2024)

  Changed paths:
    M llvm/lib/Target/AArch64/AArch64ISelLowering.cpp
    M llvm/lib/Target/AArch64/AArch64ISelLowering.h
    A llvm/test/CodeGen/AArch64/consecutive-stores-of-faddv.ll

  Log Message:
  -----------
  [AArch64] Disable consecutive store merging when Neon is unavailable (#111519)

Lowering fixed-size BUILD_VECTORS without Neon may introduce stack
spills, leading to more stores/reloads than if the stores were not
merged. In some cases, it can also prevent using paired store
instructions.

In the future, we may want to relax when SVE is available, but
currently, the SVE lowerings for BUILD_VECTOR are limited to a few
specific cases.



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