++ed by:
1 non-PAUSE user
Author image Nilson Santos Figueiredo Júnior


subs::parallel - enables subroutines to seamlessly run in parallel


Version 0.09


  use subs::parallel;
  sub foo : Parallel {
    # foo runs in parallel
  # subroutine named bar now runs in parallel
  my $foo = foo(); # returns immediately
  my $bar = bar(); # also returns immediately
  # now it might block waiting for both to finish
  if ($foo == $bar) {
  my $baz = parallelize { ... code ... }; # returns immediately
  print "baz: $baz\n"; # if it's still running, blocks until it finishes
  # it can be done to anonymous subs or any other coderefs too
  my $anon = sub { ... more code ... };
  my $parallel_coderef = parallelize_coderef($anon);

  my $foobar = $parallel_coderef->('arg'); # returns immediately
  # sub should return an object, no problem
  $foobar->do_something_else(); # blocks until it finishes running


This module gives Perl programmers the ability to easily and conveniently create multi-threaded applications through parallel subroutines.

Parallel subroutines are just plain old subroutines which happen to run in another thread. When called, they return immediately to the calling thread but keep running in another. When its return value is needed somewhere, the returned value is transparently fetched. If the thread is still running, the code blocks waiting for it to finish (since the program can't go on without a value and keep being transparent).

So, as it's possible to notice, the module interface aims to be as simple as possible. In fact, it works in such a way that, aside from the parallelizing directives, you wouldn't be able to tell it's a multi-threaded application. All the thread handling (which isn't that complicated, really) is done automagically.

It should work for anything that's thread safe - even for subroutines whose return values are not usually available across thread boundaries (for example, usually, you can't share an object, but this module makes it possible to return them without any problems, provided they're thread safe).


The main reason behind this module was the fact that Perl threads are not widely used. Some could argue that this might happen because Perl threads are not very practical to use. Others could say that the restrictions imposed by the use of threads (you can't pass arbitrary data structures between threads).

The latter issue is out of my reach. But, through the use of already existing features, it was possible to mask it a little bit. So you can return anything from your parallelized subroutines, but this is just some maybe unknown feature of threads-join>.

The first issue is the main aim of this module: provide an extremely simple way to make multi-threaded applications. Along this near-future reality of a majority of dual-core computers, multi-threading might become the factor that distiguishes good from bad software.

But I would be dishonest if I didn't note that some part of the effort was driven by the will to prove it could be done in Perl.


The most practical way of using this module's features is through attrubutes. This snippet ilustrates how to do it:

  use subs::parallel;
  sub foobar : Parallel {
        # do stuff

That way, foobar is declared as a parallel subroutine. This should have exactly the same effect as using the parallelize_sub below, but is cleaner and more convenient.


The parallelize, parallelize_sub and parallelize_coderef subroutines are exported. If you have problems with that, you're free to:

  use subs::parallel ();


All the three functions available work similarly, implementing the idea of parallelized subroutines as explained above. Nevertheless, here's a brief explanation of each of them.

parallelize { BLOCK }

Starts running the specified block of code in parallel immediately. The snippet below ilustrates this concept:

  use subs::parallel;

  my $foo = parallelize {
    # do stuff here

  my $bar = parallelize {
    # do even more stuff

  # now we wait for them to finish
  print "$foo, $bar\n";


Transforms the named subroutine into a parallel subroutine. After this modification, every it's called, it will run inside another thread and return immediately to the caller.

The snippet below ilustrates this concept:

  use subs::parallel;

  sub blocking_stuff {
          # do blocking stuff

  blocking_stuff(); # this blocks

  # now it returns immediately and the return values are discarded

Note that, you can parallelize named subroutines inside other packages:



Returns a parallelized version of the given CODEREF. This function is used internally to implement the other two functions explained above.

  my $code = sub { ... }
  my $parallel = parallelize_coderef($code);

  my $return = $parallel->(@args); # runs in another thread


The only known issue is between this module and Catalyst. When using Catalyst the Parallel attribute seems to be broken. This is probably due to Catalyst trying to process every possible attribute, but I'm not really sure. Patches are welcome.

There's a simple workaround. Instead of writing:

  sub foo : Parallel {
      # code here

Use the slightly more verbose:

  sub foo {
      # code here


That should work in all cases.


Currently, the parallel subroutines will always be called in scalar context. There are plans for changing this in the future but, for now, if multiple return values are needed, consider wrapping them around an array ref (this feature would require adding the Want module as a dependency).

In order for objects to survice across threads, both the calling and the executing thread must have its class properly included. This means that you can't require a module only inside a subroutine and then return an object blessed into that class to the calling thread.

You shouldn't pass to another thread/parallelized subroutines previous return values from other parallelized subroutines without reading their values. And, while you can get away with it, it's probably too much rope to hang yourself on when things start to get ugly. This might be changed in the future, but would require some sort of explicit synchronization through shared variables.

Also, there's no way to tell if a parallelized subroutine is still running or not. This probably will be changed in the future but, unfortunately, will require either a minimalistic approach using some simple shared variables or the use of Thread::Running, which would be another dependency and ends up using shared variables anyway.

The caller stack may behave in a non expected manner. When calling returned object's methods, there will be an extra level in the stack, representing AUTOLOAD magic - the third form of goto couldn't be used since that way if the user called a non-existant method, he wouldn't get the default warning. In the future, maybe this will be configurable so that if you need to rely on the caller stack, you can make this trade-off.

If the parallelized subroutine dies, you won't get any warnings on notices of any kind. It would be like it returned undef. This might change in the future.

GLOBAL::CORE::ref is overriden. This may be an issue if there are other modules in use which override it too. In the future, there may be a switch to disable this.


Besides the CAVEATS (which some people might consider to be bugs) there are no known bugs.

Please report any bugs or feature requests to bug-subs-parallel@rt.cpan.org, or through the web interface at http://rt.cpan.org/NoAuth/ReportBug.html?Queue=subs-parallel-0.09. I will be notified, and then you'll automatically be notified of progress on your bug as I make changes.


Nilson Santos F. Jr., <nilsonsfj@cpan.org>


Copyright (C) 2005-2015 Nilson Santos Figueiredo Junior.

This program is free software; you can redistribute it and/or modify it under the same terms as Perl itself.