Re: Asynchronous DNS

[Alain Schneble]

Andreas Schwab <address@hidden> writes:

> Lars Ingebrigtsen <address@hidden> writes:
>
>> It would be better if the application code didn't have to care, but that
>> means that either all those functions will have to block (getting us
>> back to square one, basically), or they'll have to set up a queue of
>> pending actions to be taken, which I think is rather too ambitious. 
>
> There's still the benefit of blocking in the Emacs event loop instead of
> a system call, with the former Emacs can still run other process
> filters.

That sounds interesting, but would it really be feasible to call out to
the event loop while in the blocking call?  I doubt so.  My
understanding was that calls to this new blocking function e.g. in
`set-process-coding-system' would really block the thread e.g. by a call
to gai_suspend.  Or didn't you refer to this "blocking function concept"
in your remark?