view release on metacpan or  search on metacpan

local/lib/perl5/IO/Async/Loop.pm  view on Meta::CPAN

   my $count = 0;

   my $timequeue = $self->{timequeue};
   $count += $timequeue->fire if $timequeue;

   my $deferrals = $self->{deferrals};
   $self->{deferrals} = [];

   foreach my $code ( @$deferrals ) {
      $code->();
      $count++;
   }

   my $childwatches = $self->{childwatches};
   if( !HAVE_SIGNALS and keys %$childwatches ) {
      _reap_children( $childwatches );
   }

   return $count;
}

=head1 EXTENSIONS

An Extension is a Perl module that provides extra methods in the
C<IO::Async::Loop> or other packages. They are intended to provide extra
functionality that easily integrates with the rest of the code.

Certain base methods take an C<extensions> parameter; an ARRAY reference
containing a list of extension names. If such a list is passed to a method, it
will immediately call a method whose name is that of the base method, prefixed
by the first extension name in the list, separated by C<_>. If the
C<extensions> list contains more extension names, it will be passed the
remaining ones in another C<extensions> parameter.

For example,

 $loop->connect(
    extensions => [qw( FOO BAR )],
    %args
 )

will become

 $loop->FOO_connect(
    extensions => [qw( BAR )],
    %args
 )

This is provided so that extension modules, such as L<IO::Async::SSL> can
easily be invoked indirectly, by passing extra arguments to C<connect> methods
or similar, without needing every module to be aware of the C<SSL> extension.
This functionality is generic and not limited to C<SSL>; other extensions may
also use it.

The following methods take an C<extensions> parameter:

 $loop->connect
 $loop->listen

If an extension C<listen> method is invoked, it will be passed a C<listener>
parameter even if one was not provided to the original C<< $loop->listen >>
call, and it will not receive any of the C<on_*> event callbacks. It should
use the C<acceptor> parameter on the C<listener> object.

=cut

=head1 STALL WATCHDOG

A well-behaved L<IO::Async> program should spend almost all of its time
blocked on input using the underlying C<IO::Async::Loop> instance. The stall
watchdog is an optional debugging feature to help detect CPU spinlocks and
other bugs, where control is not returned to the loop every so often.

If the watchdog is enabled and an event handler consumes more than a given
amount of real time before returning to the event loop, it will be interrupted
by printing a stack trace and terminating the program. The watchdog is only in
effect while the loop itself is not blocking; it won't fail simply because the
loop instance is waiting for input or timers.

It is implemented using C<SIGALRM>, so if enabled, this signal will no longer
be available to user code. (Though in any case, most uses of C<alarm()> and
C<SIGALRM> are better served by one of the L<IO::Async::Timer> subclasses).

The following environment variables control its behaviour.

=over 4

=item IO_ASYNC_WATCHDOG => BOOL

Enables the stall watchdog if set to a non-zero value.

=item IO_ASYNC_WATCHDOG_INTERVAL => INT

Watchdog interval, in seconds, to pass to the C<alarm(2)> call. Defaults to 10
seconds.

=item IO_ASYNC_WATCHDOG_SIGABRT => BOOL

If enabled, the watchdog signal handler will raise a C<SIGABRT>, which usually
has the effect of breaking out of a running program in debuggers such as
F<gdb>. If not set then the process is terminated by throwing an exception with
C<die>.

=back

=cut

=head1 AUTHOR

Paul Evans <leonerd@leonerd.org.uk>

=cut

0x55AA;