Thread - manipulate threads in Perl (for old code only)

Perl has two thread models.

In Perl 5.005 the thread model was that all data is implicitly shared and shared access to data has to be explicitly synchronized. This model is called 5005threads.

In Perl 5.6 a new model was introduced in which all is was thread local and shared access to data has to be explicitly declared. This model is called ithreads, for interpreter threads.

In Perl 5.6 the ithreads model was not available as a public API, only as an internal API that was available for extension writers, and to implement fork() emulation on Win32 platforms.

In Perl 5.8 the ithreads model became available through the CWthreads module.

Neither model is configured by default into Perl (except, as mentioned above, in Win32 ithreads are always available.) You can see your Perl's threading configuration by running CWperl -V and looking for the use...threads variables, or inside script by CWuse Config; and testing for CW$Config{use5005threads} and CW$Config{useithreads}.

For old code and interim backwards compatibility, the Thread module has been reworked to function as a frontend for both 5005threads and ithreads.

Note that the compatibility is not complete: because the data sharing models are directly opposed, anything to do with data sharing has to be thought differently. With the ithreads you must explicitly share() variables between the threads.

For new code the use of the CWThread module is discouraged and the direct use of the CWthreads and CWthreads::shared modules is encouraged instead.

Finally, note that there are many known serious problems with the 5005threads, one of the least of which is that regular expression match variables like CW$1 are not threadsafe, that is, they easily get corrupted by competing threads. Other problems include more insidious data corruption and mysterious crashes. You are seriously urged to use ithreads instead.

    use Thread;

    my $t = Thread->new(\&start_sub, @start_args);

    $result = $t->join;
    $result = $t->eval;
    $t->detach;

    if ($t->done) {
        $t->join;
    }

    if($t->equal($another_thread)) {
        # ...
    }

    yield();

    my $tid = Thread->self->tid;

    lock($scalar);
    lock(@array);
    lock(%hash);

    lock(\&sub);        # not available with ithreads

    $flags = $t->flags; # not available with ithreads

    my @list = Thread->list;    # not available with ithreads

    use Thread 'async';

The CWThread module provides multithreading support for perl.

$thread = Thread->new(\&start_sub)

$thread = Thread->new(\&start_sub, LIST)

CWnew starts a new thread of execution in the referenced subroutine. The optional list is passed as parameters to the subroutine. Execution continues in both the subroutine and the code after the CWnew call. CWThread->new returns a thread object representing the newly created thread.

lock VARIABLE

CWlock places a lock on a variable until the lock goes out of scope. If the variable is locked by another thread, the CWlock call will block until it's available. CWlock is recursive, so multiple calls to CWlock are safethe variable will remain locked until the outermost lock on the variable goes out of scope. Locks on variables only affect CWlock callsthey do not affect normal access to a variable. (Locks on subs are different, and covered in a bit.) If you really, really want locks to block access, then go ahead and tie them to something and manage this yourself. This is done on purpose. While managing access to variables is a good thing, Perl doesn't force you out of its living room... If a container object, such as a hash or array, is locked, all the elements of that container are not locked. For example, if a thread does a CWlock @a, any other thread doing a CWlock($a[12]) won't block. With 5005threads you may also CWlock a sub, using CWlock &sub. Any calls to that sub from another thread will block until the lock is released. This behaviour is not equivalent to declaring the sub with the CWlocked attribute. The CWlocked attribute serializes access to a subroutine, but allows different threads non-simultaneous access. CWlock &sub, on the other hand, will not allow any other thread access for the duration of the lock. Finally, CWlock will traverse up references exactly one level. CWlock(\$a) is equivalent to CWlock($a), while CWlock(\\$a) is not.

async BLOCK;

CWasync creates a thread to execute the block immediately following it. This block is treated as an anonymous sub, and so must have a semi-colon after the closing brace. Like CWThread->new, CWasync returns a thread object.

Thread->self

The CWThread->self function returns a thread object that represents the thread making the CWThread->self call.

cond_wait VARIABLE

The CWcond_wait function takes a locked variable as a parameter, unlocks the variable, and blocks until another thread does a CWcond_signal or CWcond_broadcast for that same locked variable. The variable that CWcond_wait blocked on is relocked after the CWcond_wait is satisfied. If there are multiple threads CWcond_waiting on the same variable, all but one will reblock waiting to reaquire the lock on the variable. (So if you're only using CWcond_wait for synchronization, give up the lock as soon as possible.)

cond_signal VARIABLE

The CWcond_signal function takes a locked variable as a parameter and unblocks one thread that's CWcond_waiting on that variable. If more than one thread is blocked in a CWcond_wait on that variable, only one (and which one is indeterminate) will be unblocked. If there are no threads blocked in a CWcond_wait on the variable, the signal is discarded.

cond_broadcast VARIABLE

The CWcond_broadcast function works similarly to CWcond_signal. CWcond_broadcast, though, will unblock all the threads that are blocked in a CWcond_wait on the locked variable, rather than only one.

yield

The CWyield function allows another thread to take control of the CPU. The exact results are implementation-dependent.

join

CWjoin waits for a thread to end and returns any values the thread exited with. CWjoin will block until the thread has ended, though it won't block if the thread has already terminated. If the thread being CWjoined CWdied, the error it died with will be returned at this time. If you don't want the thread performing the CWjoin to die as well, you should either wrap the CWjoin in an CWeval or use the CWeval thread method instead of CWjoin.

eval

The CWeval method wraps an CWeval around a CWjoin, and so waits for a thread to exit, passing along any values the thread might have returned. Errors, of course, get placed into CW$@. (Not available with ithreads.)

detach

CWdetach tells a thread that it is never going to be joined i.e. that all traces of its existence can be removed once it stops running. Errors in detached threads will not be visible anywhere - if you want to catch them, you should use CW$SIG{__DIE__} or something like that.

equal

CWequal tests whether two thread objects represent the same thread and returns true if they do.

tid

The CWtid method returns the tid of a thread. The tid is a monotonically increasing integer assigned when a thread is created. The main thread of a program will have a tid of zero, while subsequent threads will have tids assigned starting with one.

flags

The CWflags method returns the flags for the thread. This is the integer value corresponding to the internal flags for the thread, and the value may not be all that meaningful to you. (Not available with ithreads.)

done

The CWdone method returns true if the thread you're checking has finished, and false otherwise. (Not available with ithreads.)

The sequence number used to assign tids is a simple integer, and no checking is done to make sure the tid isn't currently in use. If a program creates more than 2**32 - 1 threads in a single run, threads may be assigned duplicate tids. This limitation may be lifted in a future version of Perl.

threads::shared (not available with 5005threads)

attributes, Thread::Queue, Thread::Semaphore, Thread::Specific (not available with ithreads)