> Why do you want different things to pretend they're the same thing? If you want the language to be changed, you need to offer a reason why it would be good, not simply "why not".

My proposition was that because it could produce smaller binaries (and allow for better code optimization) is itself reason enough to make the change.

And I'm trying to challenge this idea that "functions have different names and therefore should have different addresses, just because".

Same logic applies as to why zero size objects shouldn't have at least 1byte. no_unique_address should be the default and not the opt-in exception IMO.

> Because the language is supposed to actually make sense.

I don't see a reason why a programming language couldn't be coherent if &foo != &bar is undefined behavior.

> If you want two different declarations with different names to denote the same entity... shouldn't the user actually, *explicitly* request that? What even is a declaration or a name if it doesn't represent a specific,
distinct entity?

A function is not an object, it is a function.
There's no reason to reason about functions as if they were objects.

> But they don't happen behind the user's back;

It already happens behind the user's back. You are not able to define what is the exact address of your function, or if one even exists at all. Why would removing the guarantee that it is unique would make a difference?


________________________________
From: Std-Proposals <std-proposals-bounces_at_[hidden]> on behalf of Jason McKesson via Std-Proposals <std-proposals_at_[hidden]>
Sent: Sunday, April 27, 2025 4:46:59 PM
To: std-proposals_at_[hidden] <std-proposals_at_lists.isocpp.org>
Cc: Jason McKesson <jmckesson_at_[hidden]>
Subject: [std-proposals] Fwd: Fwd: [std-discussion] Guarantees over addresses from function pointers created from lambda

On Sun, Apr 27, 2025 at 10:42 AM Tiago Freire <tmiguelf_at_[hidden]> wrote:
>
> > By the rules of C++, if you test the addresses of these two objects, they *must* be different.
>
> Again why? I know that is what it does now, but why is it important to behave this way?

Why do you want different things to pretend they're the same thing? If
you want the language to be changed, you need to offer a reason why it
would be good, not simply "why not".

> > Because they have different names which
> denote different entities *even if* the result of evaluating them is known by the compiler to be the same.
>
> Why would that matter?

Because the language is supposed to actually make sense. If you want
two different declarations with different names to denote the same
entity... shouldn't the user actually, *explicitly* request that? What
even is a declaration or a name if it doesn't represent a specific,
distinct entity?

We have a way to create a name that denotes an existing (object)
entity: references. But they don't happen behind the user's back;
there's special syntax for creating them.
--
Std-Proposals mailing list
Std-Proposals_at_[hidden]g
https://lists.isocpp.org/mailman/listinfo.cgi/std-proposals