[all-commits] [llvm/llvm-project] d8a4c6: [CodeGenPrepare] Consider target memory intrinics ...

[Jeffrey Byrnes via All-commits]

  Branch: refs/heads/main
  Home:   https://github.com/llvm/llvm-project
  Commit: d8a4c61fe4c11b16a496a089e2962473c6b1b183
      https://github.com/llvm/llvm-project/commit/d8a4c61fe4c11b16a496a089e2962473c6b1b183
  Author: Jeffrey Byrnes <jeffrey.byrnes at amd.com>
  Date:   2025-09-19 (Fri, 19 Sep 2025)

  Changed paths:
    M llvm/lib/CodeGen/CodeGenPrepare.cpp
    M llvm/test/CodeGen/AMDGPU/sink-addr-memory-intrinsics.ll

  Log Message:
  -----------
  [CodeGenPrepare] Consider target memory intrinics as memory use (#159638)

When deciding to sink address instructions into their uses, we check if
it is profitable to do so. The profitability check is based on the types
of uses of this address instruction -- if there are users which are not
memory instructions, then do not fold.

However, this profitability check wasn't considering target intrinsics,
which may be loads / stores.

This adds some logic to handle target memory intrinsics.



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