[clang] [llvm] [AMDGPU] Emit a waitcnt instruction after each memory instruction (PR #68932)

[Jun Wang via llvm-commits]

jwanggit86 wrote:

> > So, while it's possible to create a combined option, using a separate option also makes sense. Do we generally try to avoid creating new command-line options?
> 
> Looking again, I see they are different and unrelated. I don't really understand why we have amdgpu-waitcnt-forcezero, I'm not sure I've ever used it. I always expected it to behave like this flag.

So do you still think the new code should be integrated with the existing pass, or should it be separate?

https://github.com/llvm/llvm-project/pull/68932