[llvm] [AMDGPU] (x or y) xor -1 -> x nor y (PR #130264)

[Nicolai Hähnle via llvm-commits]

nhaehnle wrote:

> > As for the rest, I think the use of ballot/inverse.ballot here is very reasonable. Our calling conventions generally pass i1 as 32-bit values,
> 
> We're working on fixing that to pass as a lane mask for callable functions (this won't change for entries)

That's good. How quickly will that be ready?

https://github.com/llvm/llvm-project/pull/130264