bug#58218: 29.0.50; url-retrieve-synchronously with timeout causes process-query-on-exit

[Lars Ingebrigtsen]

Phil Sainty <psainty@orcon.net.nz> writes:

> (url-retrieve-synchronously "http://www.example.com"; nil nil 0.01)
> nil
>
> M-x list-processes
> www.example.com -- open -- -- Main (network connection to
> www.example.com:80)
>
> C-x C-c
> Active processes exist; kill them and exit anyway? (yes or no)

Yes, it seems to be an accounting error in the code around
`url-http-mark-connection-as-busy' -- that function is called too late,
and marks the connection as busy after we've already exited, and then it
never marks the connection as non-busy...  Hm...

We never get to

http -> Marking connection as free: www.example.com:80 #<process 
www.example.com<1>>

because we never get the callback.  Hm...