Oli Scherer

> Just out of curiosity I'd to ask if the previous approach was unsound? It fed a hir id query, which we haven't considered yet at all. We could have...

> Function pointers are handled by walking the final value of the const, and if we see a function pointer in there we add it to the monomorphization list. since...

> Or do we actually codegen all functions that are "reachable"? In that case we probably have to distinguish "reachable in const code" and "reachable in runtime code". Yes. These...

I was mistaken. While I recently dug into this, I mixed up two things. It is used separately (and I think we hope that monomorphization and reachability are in sync...

The MCP has been accepted

I think we need to guarantee at the tail call site, that no things with a destructor are still in scope.

Yea, this is exactly what I wanted. Do any of the new deps have problematic licenses? Otherwise seems fine to just add them to the list of allowed crates.

> Miri should probably keep installing its own hook, but it might be nice to use the static inside `rustc_*`. You could do this change in this PR, subtrees ftw

It would be possible, but the current implementation of the feature is very much in flux, so we'd like to avoid doing that in the near future. cc @fee1-dead

We could implement a hacky solution for non-generic impls methods today without relevant tech debt or complexity. We could allow adding `rustc_const_stable` and `rustc_const_unstable` to impl methods, allowing them to...