[PATCH] D158600: [InferAlignment] Enable InferAlignment pass by default

Nuno Lopes via Phabricator via llvm-commits llvm-commits at lists.llvm.org
Fri Sep 22 02:55:48 PDT 2023


nlopes added a comment.

In D158600#4649041 <https://reviews.llvm.org/D158600#4649041>, @nikic wrote:

> In D158600#4649040 <https://reviews.llvm.org/D158600#4649040>, @nlopes wrote:
>
>> I'm confused...
>>
>> This pass regresses alignment information in a lot of tests. I've also observed in when compiling real programs.
>
> The test regressions in this diff are just an artifact of inference moving into a different pass (the tests run instcombine as part of the pipeline, while is no longer responsible for alignment inference). If you see end-to-end regressions in alignment information, that would be a problem though. Do you have an example?

Ah, I see.
No, didn't see end-to-end regressions. I was looking somewhere in the middle of the pipeline only.
But this may impact alias analysis, depending on how often and there this new pass runs. Instcombine runs more often than this pass. We need to monitor for regressions.


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D158600/new/

https://reviews.llvm.org/D158600



More information about the llvm-commits mailing list