apr-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Neil Conway <...@eecs.berkeley.edu>
Subject Re: Questions about apr_poll
Date Wed, 06 Jan 2010 00:20:00 GMT
On Tue, Jan 5, 2010 at 12:49 AM, Paul Querna <paul@querna.org> wrote:
> On Tue, Jan 5, 2010 at 12:16 AM, Neil Conway <nrc@cs.berkeley.edu> wrote:
>> From glancing at the code, it seems that the select provider has a
>> fixed limit on the pollset size, but kqueue, epoll and event port do
>> not, for example. Is that accurate? Perhaps the apr_pollset
>> documentation can be updated accordingly.
>
> Yes, this is correct.

Attached is a patch against trunk that updates the documentation for
apr_poll to describe this behavior. I also fixed a few minor doxygen
issues along the way.

> pollset_wakeup shouldn't exist imo, its place is in something like the
> apr_events_* api that was proposed and never adopted at one point, but
> it was added anyways because of a use case needed by tomcat nc (?)

I agree that a higher-level events API would be useful, but in the
absence of that, pollcb_wakeup() would be pretty useful: wakeup() + a
secondary event queue is a reasonable substitute for a general-purpose
event notification system. IMHO a lot of people will basically need to
roll this functionality by hand in the absence of a builtin
pollcb_wakeup().

If a patch for pollcb_wakeup() would be considered or accepted, I can
look into writing one.

Neil

Mime
View raw message