I've made CWG3139 for this.

https://cplusplus.github.io/CWG/issues/3139.html

Jens


On 10/10/25 21:00, Brian Bi wrote:
> Jens, can we have a core issue for this, please?
>
> On Wed, Oct 8, 2025 at 12:27 PM Brian Bi <bbi5291_at_[hidden] <mailto:bbi5291_at_[hidden]>> wrote:
>
> It seems we've just discovered the reason why the pre-CWG2801 wording was the way it was. Sigh.
>
> We should probably
>
> * restore the old wording,
> * then change the part that says "cv1 shall be the same cv-qualification as, or greater cv-qualification than, cv2" to "cv1 T1 shall be reference-compatible with cv2 T2". I think that fixes the original issue.
> * then, add a note so that in a few years we don't forget why this wording is here.
>
>
> On Wed, Oct 8, 2025 at 11:55 AM Ell via Std-Discussion <std-discussion_at_[hidden] <mailto:std-discussion_at_[hidden]>> wrote:
>
> AFAICT, after the changes from CWG 2801, none of the bullets in
> [dcl.init.ref]/5 allow initializing a (const/rv) reference from a
> similarly-typed bit field. I'm pretty sure that wasn't intended.
> --
> Std-Discussion mailing list
> Std-Discussion_at_[hidden] <mailto:Std-Discussion_at_[hidden]>
> https://lists.isocpp.org/mailman/listinfo.cgi/std-discussion <https://lists.isocpp.org/mailman/listinfo.cgi/std-discussion>
>
>
>
> --
> /Brian Bi/
>
>
>
> --
> /Brian Bi/