On Fri, 16 Dec 2005, John-Mark Gurney wrote:
Seán C. Farley wrote this message on Fri, Dec 16, 2005 at 16:09 -0600:
I may have missed it in the man page, but I am unable to find a way
to determine how many kevents are currently registered within a
kqueue. If there is no method for a count, how about a way to find
if a kqueue is empty or not. Besides tracking what events are still
within a kqueue, this would make for an easier way to write an event
loop. Currently, calling kevent() on an empty kqueue will still
block.
Why do you need this? What is the point of calling kevent if you
don't have an event to wake it up? Can you give us an example of
where you need to know that there isn't any events in the queue? (I
can't think of one, since using kqueue w/o events is pointless.) I
can't see how tracking the count will let you know which events are
still in the kqueue... (second to last sentence)
If you register a number of events to be handled by a single kevent-loop
and they are deleted (by kevent() or EV_ONESHOT), it would be nice for a
way to see if there are any more kevents registered in the kqueue when
deciding to exit the loop (or program) or performing some special
operation when all events are gone.
Also, I recommend that the man page mention that a kqueue may be
close(2)'d. The only reference that it can be is that the manual
page says the kqueue() call returns a descriptor.
I would think that it would need to be documented when you can't close
a descriptor.. Neither pipe nor socketpair documents that close can
be called on them...
Good point. It was just a thought for us beginners to kevents. The
other calls have a multitude of sources for documentation.
Seán
--
[EMAIL PROTECTED]
_______________________________________________
[email protected] mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-hackers
To unsubscribe, send any mail to "[EMAIL PROTECTED]"
