[PATCH] D134336: [AArch64] Lower multiplication by a constant int to madd

Mon Sep 26 02:28:02 PDT 2022

Allen marked an inline comment as done.
Allen added a comment.

hi, @efriedma

- base on commit 7328ff75ba <https://reviews.llvm.org/rG7328ff75bab0d559645927ddaeab39bf0f4b8d03>,  I add this patch and test the performace change, **there is no notable regression case**, and **502.gcc_r**  is benefit from this MR base on tsv110 target
- ps: the 548.exchange2_r compile fail as it is a fortran  source code, see https://www.spec.org/cpu2017/Docs/overview.html#benchmarks

F24701519: image.png <https://reviews.llvm.org/F24701519>

================
Comment at: llvm/test/CodeGen/AArch64/machine-outliner-throw.ll:1
+; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
 ; RUN: llc -verify-machineinstrs -enable-machine-outliner -mtriple=aarch64-arm-none-eabi < %s | FileCheck %s
----------------
efriedma wrote:
> I think this NOTE was intentionally dropped from the test because some of the CHECK lines are manually written?
Done, thanks

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D134336/new/

https://reviews.llvm.org/D134336