[libcxx-commits] [libcxx] [libc++] Optimize __hash_table copy constructors and assignment (PR #151951)

[Nikolas Klauser via libcxx-commits]

philnik777 wrote:

> This one also changes behavior, right?

Yes.

> I think before this change, the copy of an unordered_foo had the same iteration order as the original unordered_foo, and after this change it's no longer true.

It's actually the other way around: https://godbolt.org/z/z6MzPYd41

> (This is fine by the standard, but still nice to mention in release notes.)

I'm not against it, but I'm also not sure it's worth a note. We're basically changing from "the elements could be in any order" to "they're the same before and after the copy".

https://github.com/llvm/llvm-project/pull/151951