On Friday, 27 January 2023 17:13:01 PST samuel ammonius via Std-Proposals
wrote:
> The word ABI was kind of misleading, but I just meant specifying
> the things that normally get in the way of compatibility, which are
> name mangling and calling member functions. The whole proposal
> will be in the form of "this in C++ will be equal to this in C". It won't
> specify things like registers.

We already have that solution and it's both the Itanium C++ ABI, its psABI,
which is used by all major compiler vendors except one, and the lower-level
SysV psABI that all Unix systems subscribe to.

That leaves exactly one major implementation not using a common, standardised
solution.

Plus, your proposal has absolutely no future in the C++ language going through
the committee if all the relevant vendors aren't already on-board yet. You
have to get them to agree to implement something. This is the case for every
single proposal to the language, though some are easier than others to justify
and get buy-in for.

Since you have to talk to the vendors anyway and there's exactly one that
needs to agree to the common ABI that already exists, why don't you contact
them and have them adopt it?

-- 
Thiago Macieira - thiago (AT) macieira.info - thiago (AT) kde.org
   Software Architect - Intel DCAI Cloud Engineering