> It is important to note that, in the Itanium ABI, even if the module doesn’t currently have
> anything that needs static initialization, the translation unit importing that module will still
> generate the code to call the initializer, because we don’t want the addition of a static initializer
> to a shared object to result in an ABI break

this is false.

On 10/15/24 7:17 PM, Michael Spencer via SG15 wrote:
>
> On Tue, Oct 15, 2024 at 9:36 AM Daniel Ruoso via SG15 <sg15_at_[hidden]
> <mailto:sg15_at_[hidden]>> wrote:
>
> This is a follow-up from the discussion we had in the last meeting, on what
> is necessary for us to be able to properly support interface-unit-only
> module libraries and the impact it has on the itanium ABI.
>
> https://wg21.link/P3470R0 <https://wg21.link/P3470R0> as usual for the paper.
>
> daniel
>
>
> Thanks for writing this up. I plan to schedule this for Wrocław.
>
> Personally I'd like to see an implementation of this ABI before we actually make
> an associated language change, but it's good to have the discussion in SG15 now.
>
> - Michael Spencer
>
> _______________________________________________
> SG15 mailing list
> SG15_at_[hidden]
> https://lists.isocpp.org/mailman/listinfo.cgi/sg15

-- 
Nathan Sidwell