threads::shared - Perl extension for sharing data structures between threads |
threads::shared - Perl extension for sharing data structures between threads
This document describes threads::shared version 1.18
use threads; use threads::shared;
my $var :shared; $var = $scalar_value; $var = $shared_ref_value; $var = share($simple_unshared_ref_value);
my ($scalar, @array, %hash); share($scalar); share(@array); share(%hash); my $bar = &share([]); $hash{bar} = &share({});
{ lock(%hash); ... }
cond_wait($scalar); cond_timedwait($scalar, time() + 30); cond_broadcast(@array); cond_signal(%hash);
my $lockvar :shared; # condition var != lock var cond_wait($var, $lockvar); cond_timedwait($var, time()+30, $lockvar);
By default, variables are private to each thread, and each newly created thread gets a private copy of each existing variable. This module allows you to share variables across different threads (and pseudo-forks on Win32). It is used together with the the threads manpage module.
share
, cond_wait
, cond_timedwait
, cond_signal
, cond_broadcast
,
is_shared
Note that if this module is imported when the threads manpage has not yet been loaded, then these functions all become no-ops. This makes it possible to write modules that will work in both threaded and non-threaded environments.
share
takes a value and marks it as shared. You can share a scalar, array,
hash, scalar ref, array ref, or hash ref. share
will return the shared
rvalue, but always as a reference.
A variable can also be marked as shared at compile time by using the
:shared
attribute: my $var :shared;
.
Due to problems with Perl's prototyping, if you want to share a newly created
reference, you need to use the &share([])
and &share({})
syntax.
The only values that can be assigned to a shared scalar are other scalar values, or shared refs:
my $var :shared; $var = 1; # ok $var = []; # error $var = &share([]); # ok
share
will traverse up references exactly one level. share(\$a)
is
equivalent to share($a)
, while share(\\$a)
is not. This means that you
must create nested shared data structures by first creating individual shared
leaf nodes, and then adding them to a shared hash or array.
my %hash :shared; $hash{'meaning'} = &share([]); $hash{'meaning'}[0] = &share({}); $hash{'meaning'}[0]{'life'} = 42;
is_shared
checks if the specified variable is shared or not. If shared,
returns the variable's internal ID (similar to
refaddr()). Otherwise, returns undef
.
if (is_shared($var)) { print("\$var is shared\n"); } else { print("\$var is not shared\n"); }
lock
places a lock on a variable until the lock goes out of scope. If the
variable is locked by another thread, the lock
call will block until it's
available. Multiple calls to lock
by the same thread from within
dynamically nested scopes are safe -- the variable will remain locked until
the outermost lock on the variable goes out of scope.
Locking a container object, such as a hash or array, doesn't lock the elements
of that container. For example, if a thread does a lock(@a)
, any other
thread doing a lock($a[12])
won't block.
lock()
follows references exactly one level. lock(\$a)
is equivalent
to lock($a)
, while lock(\\$a)
is not.
Note that you cannot explicitly unlock a variable; you can only wait for the lock to go out of scope. This is most easily accomplished by locking the variable inside a block.
my $var :shared; { lock($var); # $var is locked from here to the end of the block ... } # $var is now unlocked
If you need more fine-grained control over shared variable access, see the Thread::Semaphore manpage.
cond_wait
function takes a locked variable as a parameter, unlocks
the variable, and blocks until another thread does a cond_signal
or
cond_broadcast
for that same locked variable. The variable that
cond_wait
blocked on is relocked after the cond_wait
is satisfied. If
there are multiple threads cond_wait
ing on the same variable, all but one
will re-block waiting to reacquire the lock on the variable. (So if you're only
using cond_wait
for synchronisation, give up the lock as soon as possible).
The two actions of unlocking the variable and entering the blocked wait state
are atomic, the two actions of exiting from the blocked wait state and
re-locking the variable are not.
In its second form, cond_wait
takes a shared, unlocked variable followed
by a shared, locked variable. The second variable is unlocked and thread
execution suspended until another thread signals the first variable.
It is important to note that the variable can be notified even if no thread
cond_signal
or cond_broadcast
on the variable. It is therefore
important to check the value of the variable and go back to waiting if the
requirement is not fulfilled. For example, to pause until a shared counter
drops to zero:
{ lock($counter); cond_wait($count) until $counter == 0; }
cond_timedwait
takes a locked variable and an
absolute timeout as parameters, unlocks the variable, and blocks until the
timeout is reached or another thread signals the variable. A false value is
returned if the timeout is reached, and a true value otherwise. In either
case, the variable is re-locked upon return.
Like cond_wait
, this function may take a shared, locked variable as an
additional parameter; in this case the first parameter is an unlocked
condition variable protected by a distinct lock variable.
Again like cond_wait
, waking up and reacquiring the lock are not atomic,
and you should always check your desired condition after this function
returns. Since the timeout is an absolute value, however, it does not have to
be recalculated with each pass:
lock($var); my $abs = time() + 15; until ($ok = desired_condition($var)) { last if !cond_timedwait($var, $abs); } # we got it if $ok, otherwise we timed out!
cond_signal
function takes a locked variable as a parameter and
unblocks one thread that's cond_wait
ing on that variable. If more than one
thread is blocked in a cond_wait
on that variable, only one (and which one
is indeterminate) will be unblocked.
If there are no threads blocked in a cond_wait
on the variable, the signal
is discarded. By always locking before signaling, you can (with care), avoid
signaling before another thread has entered cond_wait().
cond_signal
will normally generate a warning if you attempt to use it on an
unlocked variable. On the rare occasions where doing this may be sensible, you
can suppress the warning with:
{ no warnings 'threads'; cond_signal($foo); }
cond_broadcast
function works similarly to cond_signal
.
cond_broadcast
, though, will unblock all the threads that are blocked in
a cond_wait
on the locked variable, rather than only one.
the threads::shared manpage exports a version of bless() that works on shared objects such that blessings propagate across threads.
# Create a shared 'foo' object my $foo; share($foo); $foo = &share({}); bless($foo, 'foo');
# Create a shared 'bar' object my $bar; share($bar); $bar = &share({}); bless($bar, 'bar');
# Put 'bar' inside 'foo' $foo->{'bar'} = $bar;
# Rebless the objects via a thread threads->create(sub { # Rebless the outer object bless($foo, 'yin');
# Cannot directly rebless the inner object #bless($foo->{'bar'}, 'yang');
# Retrieve and rebless the inner object my $obj = $foo->{'bar'}; bless($obj, 'yang'); $foo->{'bar'} = $obj;
})->join();
print(ref($foo), "\n"); # Prints 'yin' print(ref($foo->{'bar'}), "\n"); # Prints 'yang' print(ref($bar), "\n"); # Also prints 'yang'
the threads::shared manpage is designed to disable itself silently if threads are not available. This allows you to write modules and packages that can be used in both threaded and non-threaded applications.
If you want access to threads, you must use threads
before you
use threads::shared
. the threads manpage will emit a warning if you use it after
the threads::shared manpage.
When share
is used on arrays, hashes, array refs or hash refs, any data
they contain will be lost.
my @arr = qw(foo bar baz); share(@arr); # @arr is now empty (i.e., == ());
# Create a 'foo' object my $foo = { 'data' => 99 }; bless($foo, 'foo');
# Share the object share($foo); # Contents are now wiped out print("ERROR: \$foo is empty\n") if (! exists($foo->{'data'}));
Therefore, populate such variables after declaring them as shared. (Scalar and scalar refs are not affected by this problem.)
It is often not wise to share an object unless the class itself has been written to support sharing. For example, an object's destructor may get called multiple times, once for each thread's scope exit. Another danger is that the contents of hash-based objects will be lost due to the above mentioned limitation. See examples/class.pl (in the CPAN distribution of this module) for how to create a class that supports object sharing.
Does not support splice
on arrays!
Taking references to the elements of shared arrays and hashes does not autovivify the elements, and neither does slicing a shared array/hash over non-existent indices/keys autovivify the elements.
share()
allows you to share($hashref->{key})
without giving any
error message. But the $hashref->{key}
is not shared, causing the
error ``locking can only be used on shared values'' to occur when you attempt to
lock($hasref->{key})
.
View existing bug reports at, and submit any new bugs, problems, patches, etc. to: http://rt.cpan.org/Public/Dist/Display.html
the threads::shared manpage Discussion Forum on CPAN: http://www.cpanforum.com/dist/threads-shared
Annotated POD for the threads::shared manpage: http://annocpan.org/~JDHEDDEN/threads-shared-1.18/shared.pm
Source repository: http://code.google.com/p/threads-shared/
the threads manpage, the perlthrtut manpage
http://www.perl.com/pub/a/2002/06/11/threads.html and http://www.perl.com/pub/a/2002/09/04/threads.html
Perl threads mailing list: http://lists.cpan.org/showlist.cgi
Artur Bergman <sky AT crucially DOT net>
threads::shared is released under the same license as Perl.
Documentation borrowed from the old Thread.pm.
CPAN version produced by Jerry D. Hedden <jdhedden AT cpan DOT org>.
threads::shared - Perl extension for sharing data structures between threads |