[llvm] [AMDGPU] Document & Finalize GFX12 Memory Model (PR #98599)

[Nicolai Hähnle via llvm-commits]

nhaehnle wrote:

> I think we could replace those SCOPE_DEV with SCOPE_SE
>
> [...]
>
> AFAIK L1 forwards everything to L2, so SCOPE_SE vs DEV isn't any less efficient, it'll reach L2 in any case.

I'm not entirely sure that's true, and in any case I think we shouldn't rely on such details. Let's just have the scopes follow the semantics we want. So like you said, and agent scope release should do a SCOPE_DEV writeback, an agent scope acquire should do a SCOPE_DEV invalidate.

https://github.com/llvm/llvm-project/pull/98599