[PATCH] D89566: [LV] Epilogue Vectorization with Optimal Control Flow

Tue Dec 8 02:12:57 PST 2020

mivnay added a comment.

In D89566#2438061 <https://reviews.llvm.org/D89566#2438061>, @bmahjour wrote:

> In D89566#2437988 <https://reviews.llvm.org/D89566#2437988>, @nikic wrote:
>
>> Should this really be enabled by default given the current (lack of) cost modelling? Primarily concerned about the code size regression this causes.
>
> Please note that functions with optsize and minsize attributes are not affected. I'm ok with turning it back off (that was my preference as well), but curious about what has regressed (since that information can help with cost-modeling). Any comments @mivnay @xbolva00 ?

@bmahjour Thanks for the patch. I will run the experiments and get back to you...

Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D89566/new/

https://reviews.llvm.org/D89566