[ViewVC] Diff of: cvs/AnyEvent-Fork/Fork.pm

Comparing AnyEvent-Fork/Fork.pm (file contents):
Revision 1.2 by root, Sun Mar 31 03:26:50 2013 UTC vs.
Revision 1.44 by root, Thu Apr 18 10:49:59 2013 UTC

1	=head1 NAME	1	=head1 NAME
2		2
3	AnyEvent::ProcessPool - manage pools of perl worker processes, exec'ed or fork'ed	3	AnyEvent::Fork - everything you wanted to use fork() for, but couldn't
4		4
5	=head1 SYNOPSIS	5	=head1 SYNOPSIS
6		6
7	use AnyEvent::ProcessPool;	7	use AnyEvent::Fork;
		8
		9	AnyEvent::Fork
		10	->new
		11	->require ("MyModule")
		12	->run ("MyModule::server", my $cv = AE::cv);
		13
		14	my $fh = $cv->recv;
8		15
9	=head1 DESCRIPTION	16	=head1 DESCRIPTION
10		17
11	This module allows you to create single worker processes but also worker	18	This module allows you to create new processes, without actually forking
12	pool that share memory, by forking from the main program, or exec'ing new	19	them from your current process (avoiding the problems of forking), but
13	perl interpreters from a module.	20	preserving most of the advantages of fork.
14		21
15	You create a new processes in a pool by specifying a function to call	22	It can be used to create new worker processes or new independent
16	with any combination of string values and file handles.	23	subprocesses for short- and long-running jobs, process pools (e.g. for use
		24	in pre-forked servers) but also to spawn new external processes (such as
		25	CGI scripts from a web server), which can be faster (and more well behaved)
		26	than using fork+exec in big processes.
17		27
18	A pool can have initialisation code which is executed before forking. The	28	Special care has been taken to make this module useful from other modules,
19	initialisation code is only executed once and the resulting process is	29	while still supporting specialised environments such as L<App::Staticperl>
20	cached, to be used as a template.	30	or L<PAR::Packer>.
21		31
22	Pools without such initialisation code don't cache an extra process.	32	=head2 WHAT THIS MODULE IS NOT
23		33
		34	This module only creates processes and lets you pass file handles and
		35	strings to it, and run perl code. It does not implement any kind of RPC -
		36	there is no back channel from the process back to you, and there is no RPC
		37	or message passing going on.
		38
		39	If you need some form of RPC, you could use the L<AnyEvent::Fork::RPC>
		40	companion module, which adds simple RPC/job queueing to a process created
		41	by this module.
		42
		43	Or you can implement it yourself in whatever way you like, use some
		44	message-passing module such as L<AnyEvent::MP>, some pipe such as
		45	L<AnyEvent::ZeroMQ>, use L<AnyEvent::Handle> on both sides to send
		46	e.g. JSON or Storable messages, and so on.
		47
		48	=head2 COMPARISON TO OTHER MODULES
		49
		50	There is an abundance of modules on CPAN that do "something fork", such as
		51	L<Parallel::ForkManager>, L<AnyEvent::ForkManager>, L<AnyEvent::Worker>
		52	or L<AnyEvent::Subprocess>. There are modules that implement their own
		53	process management, such as L<AnyEvent::DBI>.
		54
		55	The problems that all these modules try to solve are real, however, none
		56	of them (from what I have seen) tackle the very real problems of unwanted
		57	memory sharing, efficiency, not being able to use event processing or
		58	similar modules in the processes they create.
		59
		60	This module doesn't try to replace any of them - instead it tries to solve
		61	the problem of creating processes with a minimum of fuss and overhead (and
		62	also luxury). Ideally, most of these would use AnyEvent::Fork internally,
		63	except they were written before AnyEvent:Fork was available, so obviously
		64	had to roll their own.
		65
24	=head1 PROBLEM STATEMENT	66	=head2 PROBLEM STATEMENT
25		67
26	There are two ways to implement parallel processing on UNIX like operating	68	There are two traditional ways to implement parallel processing on UNIX
27	systems - fork and process, and fork+exec and process. They have different	69	like operating systems - fork and process, and fork+exec and process. They
28	advantages and disadvantages that I describe below, together with how this	70	have different advantages and disadvantages that I describe below,
29	module tries to mitigate the disadvantages.	71	together with how this module tries to mitigate the disadvantages.
30		72
31	=over 4	73	=over 4
32		74
33	=item Forking from a big process can be very slow (a 5GB process needs	75	=item Forking from a big process can be very slow.
34	0.05s to fork on my 3.6GHz amd64 GNU/Linux box for example). This overhead	76
		77	A 5GB process needs 0.05s to fork on my 3.6GHz amd64 GNU/Linux box. This
35	is often shared with exec (because you have to fork first), but in some	78	overhead is often shared with exec (because you have to fork first), but
36	circumstances (e.g. when vfork is used), fork+exec can be much faster.	79	in some circumstances (e.g. when vfork is used), fork+exec can be much
		80	faster.
37		81
38	This module can help here by telling a small(er) helper process to fork,	82	This module can help here by telling a small(er) helper process to fork,
39	or fork+exec instead.	83	which is faster then forking the main process, and also uses vfork where
		84	possible. This gives the speed of vfork, with the flexibility of fork.
40		85
41	=item Forking usually creates a copy-on-write copy of the parent	86	=item Forking usually creates a copy-on-write copy of the parent
42	process. Memory (for example, modules or data files that have been	87	process.
43	will not take additional memory). When exec'ing a new process, modules	88
44	and data files might need to be loaded again, at extra cpu and memory	89	For example, modules or data files that are loaded will not use additional
45	cost. Likewise when forking, all data structures are copied as well - if	90	memory after a fork. When exec'ing a new process, modules and data files
		91	might need to be loaded again, at extra CPU and memory cost. But when
		92	forking, literally all data structures are copied - if the program frees
46	the program frees them and replaces them by new data, the child processes	93	them and replaces them by new data, the child processes will retain the
47	will retain the memory even if it isn't used.	94	old version even if it isn't used, which can suddenly and unexpectedly
		95	increase memory usage when freeing memory.
		96
		97	The trade-off is between more sharing with fork (which can be good or
		98	bad), and no sharing with exec.
48		99
49	This module allows the main program to do a controlled fork, and allows	100	This module allows the main program to do a controlled fork, and allows
50	modules to exec processes safely at any time. When creating a custom	101	modules to exec processes safely at any time. When creating a custom
51	process pool you can take advantage of data sharing via fork without	102	process pool you can take advantage of data sharing via fork without
52	risking to share large dynamic data structures that will blow up child	103	risking to share large dynamic data structures that will blow up child
53	memory usage.	104	memory usage.
54		105
		106	In other words, this module puts you into control over what is being
		107	shared and what isn't, at all times.
		108
55	=item Exec'ing a new perl process might be difficult and slow. For	109	=item Exec'ing a new perl process might be difficult.
		110
56	example, it is not easy to find the correct path to the perl interpreter,	111	For example, it is not easy to find the correct path to the perl
57	and all modules have to be loaded from disk again. Long running processes	112	interpreter - C<$^X> might not be a perl interpreter at all.
58	might run into problems when perl is upgraded for example.
59		113
60	This module supports creating pre-initialised perl processes to be used
61	as template, and also tries hard to identify the correct path to the perl	114	This module tries hard to identify the correct path to the perl
62	interpreter. With a cooperative main program, exec'ing the interpreter	115	interpreter. With a cooperative main program, exec'ing the interpreter
63	might not even be necessary.	116	might not even be necessary, but even without help from the main program,
		117	it will still work when used from a module.
64		118
		119	=item Exec'ing a new perl process might be slow, as all necessary modules
		120	have to be loaded from disk again, with no guarantees of success.
		121
		122	Long running processes might run into problems when perl is upgraded
		123	and modules are no longer loadable because they refer to a different
		124	perl version, or parts of a distribution are newer than the ones already
		125	loaded.
		126
		127	This module supports creating pre-initialised perl processes to be used as
		128	a template for new processes.
		129
65	=item Forking might be impossible when a program is running. For example,	130	=item Forking might be impossible when a program is running.
66	POSIX makes it almost impossible to fork from a multithreaded program and
67	do anything useful in the child - strictly speaking, if your perl program
68	uses posix threads (even indirectly via e.g. L<IO::AIO> or L<threads>),
69	you cannot call fork on the perl level anymore, at all.
70		131
		132	For example, POSIX makes it almost impossible to fork from a
		133	multi-threaded program while doing anything useful in the child - in
		134	fact, if your perl program uses POSIX threads (even indirectly via
		135	e.g. L<IO::AIO> or L<threads>), you cannot call fork on the perl level
		136	anymore without risking corruption issues on a number of operating
		137	systems.
		138
71	This module can safely fork helper processes at any time, by caling	139	This module can safely fork helper processes at any time, by calling
72	fork+exec in C, in a POSIX-compatible way.	140	fork+exec in C, in a POSIX-compatible way (via L<Proc::FastSpawn>).
73		141
74	=item Parallel processing with fork might be inconvenient or difficult	142	=item Parallel processing with fork might be inconvenient or difficult
		143	to implement. Modules might not work in both parent and child.
		144
75	to implement. For example, when a program uses an event loop and creates	145	For example, when a program uses an event loop and creates watchers it
76	watchers it becomes very hard to use the event loop from a child	146	becomes very hard to use the event loop from a child program, as the
77	program, as the watchers already exist but are only meaningful in the	147	watchers already exist but are only meaningful in the parent. Worse, a
78	parent. Worse, a module might want to use such a system, not knowing	148	module might want to use such a module, not knowing whether another module
79	whether another module or the main program also does, leading to problems.	149	or the main program also does, leading to problems.
80		150
81	This module only lets the main program create pools by forking (because	151	Apart from event loops, graphical toolkits also commonly fall into the
82	only the main program can know when it is still safe to do so) - all other	152	"unsafe module" category, or just about anything that communicates with
83	pools are created by fork+exec, after which such modules can again be	153	the external world, such as network libraries and file I/O modules, which
84	loaded.	154	usually don't like being copied and then allowed to continue in two
		155	processes.
		156
		157	With this module only the main program is allowed to create new processes
		158	by forking (because only the main program can know when it is still safe
		159	to do so) - all other processes are created via fork+exec, which makes it
		160	possible to use modules such as event loops or window interfaces safely.
85		161
86	=back	162	=back
87		163
		164	=head1 EXAMPLES
		165
		166	=head2 Create a single new process, tell it to run your worker function.
		167
		168	AnyEvent::Fork
		169	->new
		170	->require ("MyModule")
		171	->run ("MyModule::worker, sub {
		172	my ($master_filehandle) = @_;
		173
		174	# now $master_filehandle is connected to the
		175	# $slave_filehandle in the new process.
		176	});
		177
		178	C<MyModule> might look like this:
		179
		180	package MyModule;
		181
		182	sub worker {
		183	my ($slave_filehandle) = @_;
		184
		185	# now $slave_filehandle is connected to the $master_filehandle
		186	# in the original prorcess. have fun!
		187	}
		188
		189	=head2 Create a pool of server processes all accepting on the same socket.
		190
		191	# create listener socket
		192	my $listener = ...;
		193
		194	# create a pool template, initialise it and give it the socket
		195	my $pool = AnyEvent::Fork
		196	->new
		197	->require ("Some::Stuff", "My::Server")
		198	->send_fh ($listener);
		199
		200	# now create 10 identical workers
		201	for my $id (1..10) {
		202	$pool
		203	->fork
		204	->send_arg ($id)
		205	->run ("My::Server::run");
		206	}
		207
		208	# now do other things - maybe use the filehandle provided by run
		209	# to wait for the processes to die. or whatever.
		210
		211	C<My::Server> might look like this:
		212
		213	package My::Server;
		214
		215	sub run {
		216	my ($slave, $listener, $id) = @_;
		217
		218	close $slave; # we do not use the socket, so close it to save resources
		219
		220	# we could go ballistic and use e.g. AnyEvent here, or IO::AIO,
		221	# or anything we usually couldn't do in a process forked normally.
		222	while (my $socket = $listener->accept) {
		223	# do sth. with new socket
		224	}
		225	}
		226
		227	=head2 use AnyEvent::Fork as a faster fork+exec
		228
		229	This runs C</bin/echo hi>, with standard output redirected to F</tmp/log>
		230	and standard error redirected to the communications socket. It is usually
		231	faster than fork+exec, but still lets you prepare the environment.
		232
		233	open my $output, ">/tmp/log" or die "$!";
		234
		235	AnyEvent::Fork
		236	->new
		237	->eval ('
		238	# compile a helper function for later use
		239	sub run {
		240	my ($fh, $output, @cmd) = @_;
		241
		242	# perl will clear close-on-exec on STDOUT/STDERR
		243	open STDOUT, ">&", $output or die;
		244	open STDERR, ">&", $fh or die;
		245
		246	exec @cmd;
		247	}
		248	')
		249	->send_fh ($output)
		250	->send_arg ("/bin/echo", "hi")
		251	->run ("run", my $cv = AE::cv);
		252
		253	my $stderr = $cv->recv;
		254
		255	=head1 CONCEPTS
		256
		257	This module can create new processes either by executing a new perl
		258	process, or by forking from an existing "template" process.
		259
		260	Each such process comes with its own file handle that can be used to
		261	communicate with it (it's actually a socket - one end in the new process,
		262	one end in the main process), and among the things you can do in it are
		263	load modules, fork new processes, send file handles to it, and execute
		264	functions.
		265
		266	There are multiple ways to create additional processes to execute some
		267	jobs:
		268
88	=over 4	269	=over 4
89		270
90	=cut	271	=item fork a new process from the "default" template process, load code,
		272	run it
91		273
		274	This module has a "default" template process which it executes when it is
		275	needed the first time. Forking from this process shares the memory used
		276	for the perl interpreter with the new process, but loading modules takes
		277	time, and the memory is not shared with anything else.
		278
		279	This is ideal for when you only need one extra process of a kind, with the
		280	option of starting and stopping it on demand.
		281
		282	Example:
		283
		284	AnyEvent::Fork
		285	->new
		286	->require ("Some::Module")
		287	->run ("Some::Module::run", sub {
		288	my ($fork_fh) = @_;
		289	});
		290
		291	=item fork a new template process, load code, then fork processes off of
		292	it and run the code
		293
		294	When you need to have a bunch of processes that all execute the same (or
		295	very similar) tasks, then a good way is to create a new template process
		296	for them, loading all the modules you need, and then create your worker
		297	processes from this new template process.
		298
		299	This way, all code (and data structures) that can be shared (e.g. the
		300	modules you loaded) is shared between the processes, and each new process
		301	consumes relatively little memory of its own.
		302
		303	The disadvantage of this approach is that you need to create a template
		304	process for the sole purpose of forking new processes from it, but if you
		305	only need a fixed number of processes you can create them, and then destroy
		306	the template process.
		307
		308	Example:
		309
		310	my $template = AnyEvent::Fork->new->require ("Some::Module");
		311
		312	for (1..10) {
		313	$template->fork->run ("Some::Module::run", sub {
		314	my ($fork_fh) = @_;
		315	});
		316	}
		317
		318	# at this point, you can keep $template around to fork new processes
		319	# later, or you can destroy it, which causes it to vanish.
		320
		321	=item execute a new perl interpreter, load some code, run it
		322
		323	This is relatively slow, and doesn't allow you to share memory between
		324	multiple processes.
		325
		326	The only advantage is that you don't have to have a template process
		327	hanging around all the time to fork off some new processes, which might be
		328	an advantage when there are long time spans where no extra processes are
		329	needed.
		330
		331	Example:
		332
		333	AnyEvent::Fork
		334	->new_exec
		335	->require ("Some::Module")
		336	->run ("Some::Module::run", sub {
		337	my ($fork_fh) = @_;
		338	});
		339
		340	=back
		341
		342	=head1 THE C<AnyEvent::Fork> CLASS
		343
		344	This module exports nothing, and only implements a single class -
		345	C<AnyEvent::Fork>.
		346
		347	There are two class constructors that both create new processes - C<new>
		348	and C<new_exec>. The C<fork> method creates a new process by forking an
		349	existing one and could be considered a third constructor.
		350
		351	Most of the remaining methods deal with preparing the new process, by
		352	loading code, evaluating code and sending data to the new process. They
		353	usually return the process object, so you can chain method calls.
		354
		355	If a process object is destroyed before calling its C<run> method, then
		356	the process simply exits. After C<run> is called, all responsibility is
		357	passed to the specified function.
		358
		359	As long as there is any outstanding work to be done, process objects
		360	resist being destroyed, so there is no reason to store them unless you
		361	need them later - configure and forget works just fine.
		362
		363	=over 4
		364
		365	=cut
		366
92	package AnyEvent::ProcessPool;	367	package AnyEvent::Fork;
93		368
94	use common::sense;	369	use common::sense;
95		370
96	use Socket ();	371	use Errno ();
97		372
98	use Proc::FastSpawn;
99	use AnyEvent;	373	use AnyEvent;
100	use AnyEvent::ProcessPool::Util;
101	use AnyEvent::Util ();	374	use AnyEvent::Util ();
102		375
103	BEGIN {	376	use IO::FDPass;
104	# require Exporter;
105	}
106		377
107	=item my $pool = new AnyEvent::ProcessPool key => value...	378	our $VERSION = 0.6;
108		379
109	Create a new process pool. The following named parameters are supported:	380	# the early fork template process
		381	our $EARLY;
110		382
111	=over 4
112
113	=back
114
115	=cut
116
117	# the template process	383	# the empty template process
118	our $template;	384	our $TEMPLATE;
119		385
120	sub _queue {	386	sub QUEUE() { 0 }
121	my ($pid, $fh) = @_;	387	sub FH() { 1 }
		388	sub WW() { 2 }
		389	sub PID() { 3 }
		390	sub CB() { 4 }
122		391
123	[	392	sub _new {
		393	my ($self, $fh, $pid) = @_;
		394
		395	AnyEvent::Util::fh_nonblocking $fh, 1;
		396
		397	$self = bless [
		398	[], # write queue - strings or fd's
		399	$fh,
		400	undef, # AE watcher
124	$pid,	401	$pid,
125	$fh,	402	], $self;
126	[],
127	undef
128	]
129	}
130		403
		404	$self
		405	}
		406
131	sub queue_cmd {	407	sub _cmd {
132	my ($queue, $cmd) = @_;	408	my $self = shift;
133		409
134	push @{ $queue->[2] }, pack "N/a", $cmd;	410	# ideally, we would want to use "a (w/a)*" as format string, but perl
		411	# versions from at least 5.8.9 to 5.16.3 are all buggy and can't unpack
		412	# it.
		413	push @{ $self->[QUEUE] }, pack "a L/a*", $_[0], $_[1];
135		414
136	$queue->[3] \|\|= AE::io $queue->[1], 1, sub {	415	$self->[WW] \|\|= AE::io $self->[FH], 1, sub {
		416	do {
		417	# send the next "thing" in the queue - either a reference to an fh,
		418	# or a plain string.
		419
137	if (ref $queue->[2][0]) {	420	if (ref $self->[QUEUE][0]) {
138	AnyEvent::ProcessPool::Util::fd_send fileno $queue->[1], fileno ${ $queue->[2][0] }	421	# send fh
		422	unless (IO::FDPass::send fileno $self->[FH], fileno ${ $self->[QUEUE][0] }) {
		423	return if $! == Errno::EAGAIN \|\| $! == Errno::EWOULDBLOCK;
		424	undef $self->[WW];
		425	die "AnyEvent::Fork: file descriptor send failure: $!";
		426	}
		427
139	and shift @{ $queue->[2] };	428	shift @{ $self->[QUEUE] };
		429
140	} else {	430	} else {
		431	# send string
141	my $len = syswrite $queue->[1], $queue->[2][0]	432	my $len = syswrite $self->[FH], $self->[QUEUE][0];
142	or do { undef $queue->[3]; die "AnyEvent::ProcessPool::queue write failure: $!" };	433
		434	unless ($len) {
		435	return if $! == Errno::EAGAIN \|\| $! == Errno::EWOULDBLOCK;
		436	undef $self->[3];
		437	die "AnyEvent::Fork: command write failure: $!";
		438	}
		439
143	substr $queue->[2][0], 0, $len, "";	440	substr $self->[QUEUE][0], 0, $len, "";
144	shift @{ $queue->[2] } unless length $queue->[2][0];	441	shift @{ $self->[QUEUE] } unless length $self->[QUEUE][0];
145	}	442	}
		443	} while @{ $self->[QUEUE] };
146		444
147	undef $queue->[3] unless @{ $queue->[2] };	445	# everything written
		446	undef $self->[WW];
		447
		448	# invoke run callback, if any
		449	$self->[CB]->($self->[FH]) if $self->[CB];
148	};	450	};
149	}
150		451
151	sub run_template {	452	() # make sure we don't leak the watcher
152	return if $template;	453	}
153		454
		455	# fork template from current process, used by AnyEvent::Fork::Early/Template
		456	sub _new_fork {
154	my ($fh, $slave) = AnyEvent::Util::portable_socketpair;	457	my ($fh, $slave) = AnyEvent::Util::portable_socketpair;
155	AnyEvent::Util::fh_nonblocking $fh, 1;	458	my $parent = $$;
156	fd_inherit fileno $slave;
157		459
158	my %env = %ENV;	460	my $pid = fork;
159	$env{PERL5LIB} = join ":", grep !ref, @INC;
160		461
161	my $pid = spawn	462	if ($pid eq 0) {
162	$^X,	463	require AnyEvent::Fork::Serve;
163	["perl", "-MAnyEvent::ProcessPool::Serve", "-e", "AnyEvent::ProcessPool::Serve::me", fileno $slave],	464	$AnyEvent::Fork::Serve::OWNER = $parent;
164	[map "$_=$env{$_}", keys %env],	465	close $fh;
165	or die "unable to spawn AnyEvent::ProcessPool server: $!";	466	$0 = "$_[1] of $parent";
		467	AnyEvent::Fork::Serve::serve ($slave);
		468	exit 0;
		469	} elsif (!$pid) {
		470	die "AnyEvent::Fork::Early/Template: unable to fork template process: $!";
		471	}
166		472
167	close $slave;	473	AnyEvent::Fork->_new ($fh, $pid)
168
169	$template = _queue $pid, $fh;
170
171	my ($a, $b) = AnyEvent::Util::portable_socketpair;
172
173	queue_cmd $template, "Iabc";
174	push @{ $template->[2] }, \$b;
175
176	use Coro::AnyEvent; Coro::AnyEvent::sleep 1;
177	undef $b;
178	die "x" . <$a>;
179	}	474	}
		475
		476	=item my $proc = new AnyEvent::Fork
		477
		478	Create a new "empty" perl interpreter process and returns its process
		479	object for further manipulation.
		480
		481	The new process is forked from a template process that is kept around
		482	for this purpose. When it doesn't exist yet, it is created by a call to
		483	C<new_exec> first and then stays around for future calls.
		484
		485	=cut
180		486
181	sub new {	487	sub new {
182	my $class = shift;	488	my $class = shift;
183		489
184	my $self = bless {	490	$TEMPLATE \|\|= $class->new_exec;
185	@_	491	$TEMPLATE->fork
186	}, $class;	492	}
187		493
188	run_template;	494	=item $new_proc = $proc->fork
		495
		496	Forks C<$proc>, creating a new process, and returns the process object
		497	of the new process.
		498
		499	If any of the C<send_> functions have been called before fork, then they
		500	will be cloned in the child. For example, in a pre-forked server, you
		501	might C<send_fh> the listening socket into the template process, and then
		502	keep calling C<fork> and C<run>.
		503
		504	=cut
		505
		506	sub fork {
		507	my ($self) = @_;
		508
		509	my ($fh, $slave) = AnyEvent::Util::portable_socketpair;
		510
		511	$self->send_fh ($slave);
		512	$self->_cmd ("f");
		513
		514	AnyEvent::Fork->_new ($fh)
		515	}
		516
		517	=item my $proc = new_exec AnyEvent::Fork
		518
		519	Create a new "empty" perl interpreter process and returns its process
		520	object for further manipulation.
		521
		522	Unlike the C<new> method, this method I<always> spawns a new perl process
		523	(except in some cases, see L<AnyEvent::Fork::Early> for details). This
		524	reduces the amount of memory sharing that is possible, and is also slower.
		525
		526	You should use C<new> whenever possible, except when having a template
		527	process around is unacceptable.
		528
		529	The path to the perl interpreter is divined using various methods - first
		530	C<$^X> is investigated to see if the path ends with something that sounds
		531	as if it were the perl interpreter. Failing this, the module falls back to
		532	using C<$Config::Config{perlpath}>.
		533
		534	=cut
		535
		536	sub new_exec {
		537	my ($self) = @_;
		538
		539	return $EARLY->fork
		540	if $EARLY;
		541
		542	# first find path of perl
		543	my $perl = $;
		544
		545	# first we try $^X, but the path must be absolute (always on win32), and end in sth.
		546	# that looks like perl. this obviously only works for posix and win32
		547	unless (
		548	($^O eq "MSWin32" \|\| $perl =~ m%^/%)
		549	&& $perl =~ m%[/\\]perl(?:[0-9]+(\.[0-9]+)+)?(\.exe)?$%i
		550	) {
		551	# if it doesn't look perlish enough, try Config
		552	require Config;
		553	$perl = $Config::Config{perlpath};
		554	$perl =~ s/(?:\Q$Config::Config{_exe}\E)?$/$Config::Config{_exe}/;
		555	}
		556
		557	require Proc::FastSpawn;
		558
		559	my ($fh, $slave) = AnyEvent::Util::portable_socketpair;
		560	Proc::FastSpawn::fd_inherit (fileno $slave);
		561
		562	# new fh's should always be set cloexec (due to $^F),
		563	# but hey, not on win32, so we always clear the inherit flag.
		564	Proc::FastSpawn::fd_inherit (fileno $fh, 0);
		565
		566	# quick. also doesn't work in win32. of course. what did you expect
		567	#local $ENV{PERL5LIB} = join ":", grep !ref, @INC;
		568	my %env = %ENV;
		569	$env{PERL5LIB} = join +($^O eq "MSWin32" ? ";" : ":"), grep !ref, @INC;
		570
		571	my $pid = Proc::FastSpawn::spawn (
		572	$perl,
		573	["perl", "-MAnyEvent::Fork::Serve", "-e", "AnyEvent::Fork::Serve::me", fileno $slave, $$],
		574	[map "$_=$env{$_}", keys %env],
		575	) or die "unable to spawn AnyEvent::Fork server: $!";
		576
		577	$self->_new ($fh, $pid)
		578	}
		579
		580	=item $pid = $proc->pid
		581
		582	Returns the process id of the process I<iff it is a direct child of the
		583	process running AnyEvent::Fork>, and C<undef> otherwise.
		584
		585	Normally, only processes created via C<< AnyEvent::Fork->new_exec >> and
		586	L<AnyEvent::Fork::Template> are direct children, and you are responsible
		587	to clean up their zombies when they die.
		588
		589	All other processes are not direct children, and will be cleaned up by
		590	AnyEvent::Fork itself.
		591
		592	=cut
		593
		594	sub pid {
		595	$_[0][PID]
		596	}
		597
		598	=item $proc = $proc->eval ($perlcode, @args)
		599
		600	Evaluates the given C<$perlcode> as ... Perl code, while setting C<@_> to
		601	the strings specified by C<@args>, in the "main" package.
		602
		603	This call is meant to do any custom initialisation that might be required
		604	(for example, the C<require> method uses it). It's not supposed to be used
		605	to completely take over the process, use C<run> for that.
		606
		607	The code will usually be executed after this call returns, and there is no
		608	way to pass anything back to the calling process. Any evaluation errors
		609	will be reported to stderr and cause the process to exit.
		610
		611	If you want to execute some code (that isn't in a module) to take over the
		612	process, you should compile a function via C<eval> first, and then call
		613	it via C<run>. This also gives you access to any arguments passed via the
		614	C<send_xxx> methods, such as file handles. See the L<use AnyEvent::Fork as
		615	a faster fork+exec> example to see it in action.
		616
		617	Returns the process object for easy chaining of method calls.
		618
		619	=cut
		620
		621	sub eval {
		622	my ($self, $code, @args) = @_;
		623
		624	$self->_cmd (e => pack "(w/a)", $code, @args);
189		625
190	$self	626	$self
191	}	627	}
192		628
		629	=item $proc = $proc->require ($module, ...)
		630
		631	Tries to load the given module(s) into the process
		632
		633	Returns the process object for easy chaining of method calls.
		634
		635	=cut
		636
		637	sub require {
		638	my ($self, @modules) = @_;
		639
		640	s%::%/%g for @modules;
		641	$self->eval ('require "$_.pm" for @_', @modules);
		642
		643	$self
		644	}
		645
		646	=item $proc = $proc->send_fh ($handle, ...)
		647
		648	Send one or more file handles (I<not> file descriptors) to the process,
		649	to prepare a call to C<run>.
		650
		651	The process object keeps a reference to the handles until they have
		652	been passed over to the process, so you must not explicitly close the
		653	handles. This is most easily accomplished by simply not storing the file
		654	handles anywhere after passing them to this method - when AnyEvent::Fork
		655	is finished using them, perl will automatically close them.
		656
		657	Returns the process object for easy chaining of method calls.
		658
		659	Example: pass a file handle to a process, and release it without
		660	closing. It will be closed automatically when it is no longer used.
		661
		662	$proc->send_fh ($my_fh);
		663	undef $my_fh; # free the reference if you want, but DO NOT CLOSE IT
		664
		665	=cut
		666
		667	sub send_fh {
		668	my ($self, @fh) = @_;
		669
		670	for my $fh (@fh) {
		671	$self->_cmd ("h");
		672	push @{ $self->[QUEUE] }, \$fh;
		673	}
		674
		675	$self
		676	}
		677
		678	=item $proc = $proc->send_arg ($string, ...)
		679
		680	Send one or more argument strings to the process, to prepare a call to
		681	C<run>. The strings can be any octet strings.
		682
		683	The protocol is optimised to pass a moderate number of relatively short
		684	strings - while you can pass up to 4GB of data in one go, this is more
		685	meant to pass some ID information or other startup info, not big chunks of
		686	data.
		687
		688	Returns the process object for easy chaining of method calls.
		689
		690	=cut
		691
		692	sub send_arg {
		693	my ($self, @arg) = @_;
		694
		695	$self->_cmd (a => pack "(w/a)", @arg);
		696
		697	$self
		698	}
		699
		700	=item $proc->run ($func, $cb->($fh))
		701
		702	Enter the function specified by the function name in C<$func> in the
		703	process. The function is called with the communication socket as first
		704	argument, followed by all file handles and string arguments sent earlier
		705	via C<send_fh> and C<send_arg> methods, in the order they were called.
		706
		707	The process object becomes unusable on return from this function - any
		708	further method calls result in undefined behaviour.
		709
		710	The function name should be fully qualified, but if it isn't, it will be
		711	looked up in the C<main> package.
		712
		713	If the called function returns, doesn't exist, or any error occurs, the
		714	process exits.
		715
		716	Preparing the process is done in the background - when all commands have
		717	been sent, the callback is invoked with the local communications socket
		718	as argument. At this point you can start using the socket in any way you
		719	like.
		720
		721	If the communication socket isn't used, it should be closed on both sides,
		722	to save on kernel memory.
		723
		724	The socket is non-blocking in the parent, and blocking in the newly
		725	created process. The close-on-exec flag is set in both.
		726
		727	Even if not used otherwise, the socket can be a good indicator for the
		728	existence of the process - if the other process exits, you get a readable
		729	event on it, because exiting the process closes the socket (if it didn't
		730	create any children using fork).
		731
		732	Example: create a template for a process pool, pass a few strings, some
		733	file handles, then fork, pass one more string, and run some code.
		734
		735	my $pool = AnyEvent::Fork
		736	->new
		737	->send_arg ("str1", "str2")
		738	->send_fh ($fh1, $fh2);
		739
		740	for (1..2) {
		741	$pool
		742	->fork
		743	->send_arg ("str3")
		744	->run ("Some::function", sub {
		745	my ($fh) = @_;
		746
		747	# fh is nonblocking, but we trust that the OS can accept these
		748	# few octets anyway.
		749	syswrite $fh, "hi #$_\n";
		750
		751	# $fh is being closed here, as we don't store it anywhere
		752	});
		753	}
		754
		755	# Some::function might look like this - all parameters passed before fork
		756	# and after will be passed, in order, after the communications socket.
		757	sub Some::function {
		758	my ($fh, $str1, $str2, $fh1, $fh2, $str3) = @_;
		759
		760	print scalar <$fh>; # prints "hi #1\n" and "hi #2\n" in any order
		761	}
		762
		763	=cut
		764
		765	sub run {
		766	my ($self, $func, $cb) = @_;
		767
		768	$self->[CB] = $cb;
		769	$self->_cmd (r => $func);
		770	}
		771
193	=back	772	=back
194		773
195	=head1 AUTHOR	774	=head1 PERFORMANCE
		775
		776	Now for some unscientific benchmark numbers (all done on an amd64
		777	GNU/Linux box). These are intended to give you an idea of the relative
		778	performance you can expect, they are not meant to be absolute performance
		779	numbers.
		780
		781	OK, so, I ran a simple benchmark that creates a socket pair, forks, calls
		782	exit in the child and waits for the socket to close in the parent. I did
		783	load AnyEvent, EV and AnyEvent::Fork, for a total process size of 5100kB.
		784
		785	2079 new processes per second, using manual socketpair + fork
		786
		787	Then I did the same thing, but instead of calling fork, I called
		788	AnyEvent::Fork->new->run ("CORE::exit") and then again waited for the
		789	socket form the child to close on exit. This does the same thing as manual
		790	socket pair + fork, except that what is forked is the template process
		791	(2440kB), and the socket needs to be passed to the server at the other end
		792	of the socket first.
		793
		794	2307 new processes per second, using AnyEvent::Fork->new
		795
		796	And finally, using C<new_exec> instead C<new>, using vforks+execs to exec
		797	a new perl interpreter and compile the small server each time, I get:
		798
		799	479 vfork+execs per second, using AnyEvent::Fork->new_exec
		800
		801	So how can C<< AnyEvent->new >> be faster than a standard fork, even
		802	though it uses the same operations, but adds a lot of overhead?
		803
		804	The difference is simply the process size: forking the 5MB process takes
		805	so much longer than forking the 2.5MB template process that the extra
		806	overhead is canceled out.
		807
		808	If the benchmark process grows, the normal fork becomes even slower:
		809
		810	1340 new processes, manual fork of a 20MB process
		811	731 new processes, manual fork of a 200MB process
		812	235 new processes, manual fork of a 2000MB process
		813
		814	What that means (to me) is that I can use this module without having a bad
		815	conscience because of the extra overhead required to start new processes.
		816
		817	=head1 TYPICAL PROBLEMS
		818
		819	This section lists typical problems that remain. I hope by recognising
		820	them, most can be avoided.
		821
		822	=over 4
		823
		824	=item leaked file descriptors for exec'ed processes
		825
		826	POSIX systems inherit file descriptors by default when exec'ing a new
		827	process. While perl itself laudably sets the close-on-exec flags on new
		828	file handles, most C libraries don't care, and even if all cared, it's
		829	often not possible to set the flag in a race-free manner.
		830
		831	That means some file descriptors can leak through. And since it isn't
		832	possible to know which file descriptors are "good" and "necessary" (or
		833	even to know which file descriptors are open), there is no good way to
		834	close the ones that might harm.
		835
		836	As an example of what "harm" can be done consider a web server that
		837	accepts connections and afterwards some module uses AnyEvent::Fork for the
		838	first time, causing it to fork and exec a new process, which might inherit
		839	the network socket. When the server closes the socket, it is still open
		840	in the child (which doesn't even know that) and the client might conclude
		841	that the connection is still fine.
		842
		843	For the main program, there are multiple remedies available -
		844	L<AnyEvent::Fork::Early> is one, creating a process early and not using
		845	C<new_exec> is another, as in both cases, the first process can be exec'ed
		846	well before many random file descriptors are open.
		847
		848	In general, the solution for these kind of problems is to fix the
		849	libraries or the code that leaks those file descriptors.
		850
		851	Fortunately, most of these leaked descriptors do no harm, other than
		852	sitting on some resources.
		853
		854	=item leaked file descriptors for fork'ed processes
		855
		856	Normally, L<AnyEvent::Fork> does start new processes by exec'ing them,
		857	which closes file descriptors not marked for being inherited.
		858
		859	However, L<AnyEvent::Fork::Early> and L<AnyEvent::Fork::Template> offer
		860	a way to create these processes by forking, and this leaks more file
		861	descriptors than exec'ing them, as there is no way to mark descriptors as
		862	"close on fork".
		863
		864	An example would be modules like L<EV>, L<IO::AIO> or L<Gtk2>. Both create
		865	pipes for internal uses, and L<Gtk2> might open a connection to the X
		866	server. L<EV> and L<IO::AIO> can deal with fork, but Gtk2 might have
		867	trouble with a fork.
		868
		869	The solution is to either not load these modules before use'ing
		870	L<AnyEvent::Fork::Early> or L<AnyEvent::Fork::Template>, or to delay
		871	initialising them, for example, by calling C<init Gtk2> manually.
		872
		873	=item exiting calls object destructors
		874
		875	This only applies to users of L<AnyEvent::Fork:Early> and
		876	L<AnyEvent::Fork::Template>, or when initialising code creates objects
		877	that reference external resources.
		878
		879	When a process created by AnyEvent::Fork exits, it might do so by calling
		880	exit, or simply letting perl reach the end of the program. At which point
		881	Perl runs all destructors.
		882
		883	Not all destructors are fork-safe - for example, an object that represents
		884	the connection to an X display might tell the X server to free resources,
		885	which is inconvenient when the "real" object in the parent still needs to
		886	use them.
		887
		888	This is obviously not a problem for L<AnyEvent::Fork::Early>, as you used
		889	it as the very first thing, right?
		890
		891	It is a problem for L<AnyEvent::Fork::Template> though - and the solution
		892	is to not create objects with nontrivial destructors that might have an
		893	effect outside of Perl.
		894
		895	=back
		896
		897	=head1 PORTABILITY NOTES
		898
		899	Native win32 perls are somewhat supported (AnyEvent::Fork::Early is a nop,
		900	and ::Template is not going to work), and it cost a lot of blood and sweat
		901	to make it so, mostly due to the bloody broken perl that nobody seems to
		902	care about. The fork emulation is a bad joke - I have yet to see something
		903	useful that you can do with it without running into memory corruption
		904	issues or other braindamage. Hrrrr.
		905
		906	Cygwin perl is not supported at the moment due to some hilarious
		907	shortcomings of its API - see L<IO::FDPoll> for more details.
		908
		909	=head1 SEE ALSO
		910
		911	L<AnyEvent::Fork::Early> (to avoid executing a perl interpreter),
		912	L<AnyEvent::Fork::Template> (to create a process by forking the main
		913	program at a convenient time), L<AnyEvent::Fork::RPC> (for simple RPC to
		914	child processes).
		915
		916	=head1 AUTHOR AND CONTACT INFORMATION
196		917
197	Marc Lehmann <schmorp@schmorp.de>	918	Marc Lehmann <schmorp@schmorp.de>
198	http://home.schmorp.de/	919	http://software.schmorp.de/pkg/AnyEvent-Fork
199		920
200	=cut	921	=cut
201		922
202	1	923	1
203		924

Diff Legend

-–
+Removed lines
-+
+Added lines
-<
+Changed lines
->
+Changed lines

Comparing AnyEvent-Fork/Fork.pm (file contents): Revision 1.2 by root, Sun Mar 31 03:26:50 2013 UTC vs. Revision 1.44 by root, Thu Apr 18 10:49:59 2013 UTC

Diff Legend

Comparing AnyEvent-Fork/Fork.pm (file contents):
Revision 1.2 by root, Sun Mar 31 03:26:50 2013 UTC vs.
Revision 1.44 by root, Thu Apr 18 10:49:59 2013 UTC