Jesse A. Tov
Also, sorry about the soundness problems. Eek!
From what I can tell, the ClosureMut*N* problem can be solved by making ClosureMut*N*::code_ptr() take `&mut self`. The `Send`/`Sync` problem looks harder, but I have a few ideas.
Actually, I’m a bit confused by > It erases all auto trait information attached to closure captures (read: Send and Sync) because currently the Closure…*N* types are `!Send` and `!Sync`....
> Yep, `FnPtrN` is not `!Sync`, and yet one can get a `&FnPtrN` from a `&ClosureN`, which seems fishy / likely UB I’m wondering whether the auto traits can be...
So, I believe #40 fixes the mutability erasure problem. I think it should be merged, but I’d like a second opinion first. The auto trait problem is tougher. #42 makes...
Hey, sorry for the slow response. And thanks for the benchmarks! I'm now available to look at and merge #10, if that will help. I’m also wondering if you have...
Hey, I'm sorry for being so slow. (In the future, it might work better to email me as @cerisier just did. My GitHub notifications are cooked and I don't use...
That sounds fine to me (low confidence), but I'm wondering if there's a reason not to move it to `core`.