> From: Ship Mints <shipmints@gmail.com>
> Date: Fri, 7 Feb 2025 07:32:48 -0500
> Cc: Andrew Cohen <acohen@ust.hk>, 63620@debbugs.gnu.org, monnier@iro.umontreal.ca,
>       michael.albinus@gmx.de
>
> On Fri, Feb 7, 2025 at 7:18 AM Eli Zaretskii <eliz@gnu.org> wrote:
>
>  > Date: Fri, 7 Feb 2025 19:43:32 +0800 (GMT+08:00)
>  > From: Andrew Cohen <acohen@ust.hk>
>  > Cc: 63620@debbugs.gnu.org, Stefan Monnier <monnier@iro.umontreal.ca>,
>  >       Eli Zaretskii <eliz@gnu.org>, michael.albinus@gmx.de
>  >
>  > Can I get any feedback about the file descriptor closure issue? From what
>  > I can tell without exposing it in lisp the way I did I won't be able to
>  > use blocking anyway.
>
>  But since there could be only one such descriptor at any given time,
>  why would any Lisp program need to know its value?  It sounds like we
>  need to record the descriptor internally, and then close it in C, no?
>  Maybe some D-Bus related primitive could do that?
>
> This is why I suggested that the lower-level C code return opaque cookies that can be used to send back to
> lower-level code to operate on them, such as closing an fd.

I'm asking why return such cookies to Lisp at all.