[PATCH] D147732: [AMDGPU] Add type mangling for {read, write, readfirst, perm}lane intrinsics

[Yaxun Liu via Phabricator via cfe-commits]

yaxunl added a comment.

In D147732#4343146 <https://reviews.llvm.org/D147732#4343146>, @jrbyrnes wrote:

> Hi @yaxunl thanks for the comment.
>
> I looked into AutoUpgrade, and it catches upgrades of this type (unmangled -> mangled) by default (via remangleIntrinsicFunction). I've included tests to demonstrate. Is this the desired effect, or are you asking to blacklist these intrinstics from being upgraded?

That is the desired effect. Thanks.


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D147732/new/

https://reviews.llvm.org/D147732