[PATCH] D142254: [X86] Transform vector SET{LE/ULT/ULE} -> SETLT and SET{GE/UGT/UGE} -> SETGT if possible

[Noah Goldstein via Phabricator via llvm-commits]

goldstein.w.n added a comment.

In D142254#4074904 <https://reviews.llvm.org/D142254#4074904>, @goldstein.w.n wrote:

> In D142254#4070885 <https://reviews.llvm.org/D142254#4070885>, @goldstein.w.n wrote:
>
>> In D142254#4070259 <https://reviews.llvm.org/D142254#4070259>, @RKSimon wrote:
>>
>>> This feels like we could easily make this a generic DAG fold with a suitable TLI callback to help determine the preferred CondCode (or just signed/unsigned).
>>
>> We would need 2x TLI callback (for x86 at least) b.c of the `truncateAVX512SetCCNoBWI` that can be required.
>>
>> Are there any other arch that have preferences about which SETCC variant is used? If so I'm in favor of moving,
>> otherwise personally don't see the need, but what is the opinion of the room?
>
> @RKSimon thoughts?

@RKSimon ping.


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D142254/new/

https://reviews.llvm.org/D142254