RFC: adding fd-passing to win32

[Lawrence D'Oliveiro]

On Mon, 22 Aug 2022 14:53:09 +0200, Ralf Habacker wrote:

> [Cygwin] adds an additional layer of problems, as described at 
> https://cygwin.com/cygwin-ug-net/highlights.html.

That, to me, is a remarkably long list of positive achievements. I was
particularly impressed with this:

    Our implementation allows select to function normally when given
    different types of file descriptors (sockets, pipes, handles, and a
    custom /dev/windows Windows messages pseudo-device).

Particularly note the limitations with Python’s asyncio event loops on
Windows
<https://docs.python.org/3/library/asyncio-platforms.html#windows>:
which come in two different flavours on that platform, neither of which
covers all the bases you would expect. But if they built it on Cygwin,
they could do it, POSIX-style.

>> In summary, current Windows implementations make it impossible to
>> implement a perfectly reliable fork, and occasional fork failures
>> are inevitable.
> 
> This was the decisive reason why 
> https://en.wikipedia.org/wiki/KDE_on_Cygwin was discontinued.

That’s not necessarily the best example, given that the KDE frameworks
are built on Qt, which already has a native Windows port. So they had
an easier path to try to achieve what they wanted.

The question is, how would this affect Windows apps trying to use D-Bus?