[PATCH] D147996: [X86] combineConcatVectorOps - remove FADD/FSUB/FMUL handling (2-1)

Mon Apr 17 18:37:23 PDT 2023

xiangzhangllvm added a comment.

In D147996#4276233 <https://reviews.llvm.org/D147996#4276233>, @LuoYuanke wrote:

> In D147996#4276231 <https://reviews.llvm.org/D147996#4276231>, @xiangzhangllvm wrote:
>
>> In D147996#4273314 <https://reviews.llvm.org/D147996#4273314>, @RKSimon wrote:
>>
>>> LGTM if its causing regressions, but I'd appreciate any time you can spend on PR60441
>>
>> OK, many thanks! Let me double confirm, the PR60441 is https://reviews.llvm.org/D60441, right?
>
> I think PR60441 is https://github.com/llvm/llvm-project/issues/60441.

For https://godbolt.org/z/19564zseG in PR60441 <https://github.com/llvm/llvm-project/issues/60441> I found GCC vectorize it at very early stage (at early mid-end stage we can see the diff of vectorization)

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D147996/new/

https://reviews.llvm.org/D147996