On Mon, Apr 28, 2025 at 2:21 PM Jennifier Burnett via Std-Discussion
<std-discussion_at_[hidden]> wrote:
>
> >When I have no guarantee for unique address I can't write code like
> >this
>
> As has been stated previously in this thread, compilers and linkers already don't guarantee that functions will have unique addresses, so **you already can't safely write anything that requires that functions have unique addresses**.
>
> >Each address value is unique and checkable by CPU, this is observable behavior.
>
> Yes, and **compilers already violate this**. We aren't discussing a proposed change to the standard (or I'm not, at least), we're talking about something that real compilers are already doing today.

We *are* talking about a proposed change to the standard. Because the
standard makes it pretty clear that the "something that real compilers
are already doing today" is non-compliant behavior.

The question is whether the standard should align itself with
implementation deviation on this matter or not.