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

Comparing AnyEvent-Fork/Fork.pm (file contents):
Revision 1.3 by root, Tue Apr 2 18:00:04 2013 UTC vs.
Revision 1.42 by root, Mon Apr 8 05:44:23 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 can either implement it yourself
		40	in whatever way you like, use some message-passing module such
		41	as L<AnyEvent::MP>, some pipe such as L<AnyEvent::ZeroMQ>, use
		42	L<AnyEvent::Handle> on both sides to send e.g. JSON or Storable messages,
		43	and so on.
		44
		45	=head2 COMPARISON TO OTHER MODULES
		46
		47	There is an abundance of modules on CPAN that do "something fork", such as
		48	L<Parallel::ForkManager>, L<AnyEvent::ForkManager>, L<AnyEvent::Worker>
		49	or L<AnyEvent::Subprocess>. There are modules that implement their own
		50	process management, such as L<AnyEvent::DBI>.
		51
		52	The problems that all these modules try to solve are real, however, none
		53	of them (from what I have seen) tackle the very real problems of unwanted
		54	memory sharing, efficiency, not being able to use event processing or
		55	similar modules in the processes they create.
		56
		57	This module doesn't try to replace any of them - instead it tries to solve
		58	the problem of creating processes with a minimum of fuss and overhead (and
		59	also luxury). Ideally, most of these would use AnyEvent::Fork internally,
		60	except they were written before AnyEvent:Fork was available, so obviously
		61	had to roll their own.
		62
24	=head1 PROBLEM STATEMENT	63	=head2 PROBLEM STATEMENT
25		64
26	There are two ways to implement parallel processing on UNIX like operating	65	There are two traditional ways to implement parallel processing on UNIX
27	systems - fork and process, and fork+exec and process. They have different	66	like operating systems - fork and process, and fork+exec and process. They
28	advantages and disadvantages that I describe below, together with how this	67	have different advantages and disadvantages that I describe below,
29	module tries to mitigate the disadvantages.	68	together with how this module tries to mitigate the disadvantages.
30		69
31	=over 4	70	=over 4
32		71
33	=item Forking from a big process can be very slow (a 5GB process needs	72	=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	73
		74	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	75	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.	76	in some circumstances (e.g. when vfork is used), fork+exec can be much
		77	faster.
37		78
38	This module can help here by telling a small(er) helper process to fork,	79	This module can help here by telling a small(er) helper process to fork,
39	or fork+exec instead.	80	which is faster then forking the main process, and also uses vfork where
		81	possible. This gives the speed of vfork, with the flexibility of fork.
40		82
41	=item Forking usually creates a copy-on-write copy of the parent	83	=item Forking usually creates a copy-on-write copy of the parent
42	process. Memory (for example, modules or data files that have been	84	process.
43	will not take additional memory). When exec'ing a new process, modules	85
44	and data files might need to be loaded again, at extra cpu and memory	86	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	87	memory after a fork. When exec'ing a new process, modules and data files
		88	might need to be loaded again, at extra CPU and memory cost. But when
		89	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	90	them and replaces them by new data, the child processes will retain the
47	will retain the memory even if it isn't used.	91	old version even if it isn't used, which can suddenly and unexpectedly
		92	increase memory usage when freeing memory.
		93
		94	The trade-off is between more sharing with fork (which can be good or
		95	bad), and no sharing with exec.
48		96
49	This module allows the main program to do a controlled fork, and allows	97	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	98	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	99	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	100	risking to share large dynamic data structures that will blow up child
53	memory usage.	101	memory usage.
54		102
		103	In other words, this module puts you into control over what is being
		104	shared and what isn't, at all times.
		105
55	=item Exec'ing a new perl process might be difficult and slow. For	106	=item Exec'ing a new perl process might be difficult.
		107
56	example, it is not easy to find the correct path to the perl interpreter,	108	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	109	interpreter - C<$^X> might not be a perl interpreter at all.
58	might run into problems when perl is upgraded for example.
59		110
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	111	This module tries hard to identify the correct path to the perl
62	interpreter. With a cooperative main program, exec'ing the interpreter	112	interpreter. With a cooperative main program, exec'ing the interpreter
63	might not even be necessary.	113	might not even be necessary, but even without help from the main program,
		114	it will still work when used from a module.
64		115
		116	=item Exec'ing a new perl process might be slow, as all necessary modules
		117	have to be loaded from disk again, with no guarantees of success.
		118
		119	Long running processes might run into problems when perl is upgraded
		120	and modules are no longer loadable because they refer to a different
		121	perl version, or parts of a distribution are newer than the ones already
		122	loaded.
		123
		124	This module supports creating pre-initialised perl processes to be used as
		125	a template for new processes.
		126
65	=item Forking might be impossible when a program is running. For example,	127	=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		128
		129	For example, POSIX makes it almost impossible to fork from a
		130	multi-threaded program while doing anything useful in the child - in
		131	fact, if your perl program uses POSIX threads (even indirectly via
		132	e.g. L<IO::AIO> or L<threads>), you cannot call fork on the perl level
		133	anymore without risking corruption issues on a number of operating
		134	systems.
		135
71	This module can safely fork helper processes at any time, by caling	136	This module can safely fork helper processes at any time, by calling
72	fork+exec in C, in a POSIX-compatible way.	137	fork+exec in C, in a POSIX-compatible way (via L<Proc::FastSpawn>).
73		138
74	=item Parallel processing with fork might be inconvenient or difficult	139	=item Parallel processing with fork might be inconvenient or difficult
		140	to implement. Modules might not work in both parent and child.
		141
75	to implement. For example, when a program uses an event loop and creates	142	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	143	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	144	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	145	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.	146	or the main program also does, leading to problems.
80		147
81	This module only lets the main program create pools by forking (because	148	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	149	"unsafe module" category, or just about anything that communicates with
83	pools are created by fork+exec, after which such modules can again be	150	the external world, such as network libraries and file I/O modules, which
84	loaded.	151	usually don't like being copied and then allowed to continue in two
		152	processes.
		153
		154	With this module only the main program is allowed to create new processes
		155	by forking (because only the main program can know when it is still safe
		156	to do so) - all other processes are created via fork+exec, which makes it
		157	possible to use modules such as event loops or window interfaces safely.
85		158
86	=back	159	=back
		160
		161	=head1 EXAMPLES
		162
		163	=head2 Create a single new process, tell it to run your worker function.
		164
		165	AnyEvent::Fork
		166	->new
		167	->require ("MyModule")
		168	->run ("MyModule::worker, sub {
		169	my ($master_filehandle) = @_;
		170
		171	# now $master_filehandle is connected to the
		172	# $slave_filehandle in the new process.
		173	});
		174
		175	C<MyModule> might look like this:
		176
		177	package MyModule;
		178
		179	sub worker {
		180	my ($slave_filehandle) = @_;
		181
		182	# now $slave_filehandle is connected to the $master_filehandle
		183	# in the original prorcess. have fun!
		184	}
		185
		186	=head2 Create a pool of server processes all accepting on the same socket.
		187
		188	# create listener socket
		189	my $listener = ...;
		190
		191	# create a pool template, initialise it and give it the socket
		192	my $pool = AnyEvent::Fork
		193	->new
		194	->require ("Some::Stuff", "My::Server")
		195	->send_fh ($listener);
		196
		197	# now create 10 identical workers
		198	for my $id (1..10) {
		199	$pool
		200	->fork
		201	->send_arg ($id)
		202	->run ("My::Server::run");
		203	}
		204
		205	# now do other things - maybe use the filehandle provided by run
		206	# to wait for the processes to die. or whatever.
		207
		208	C<My::Server> might look like this:
		209
		210	package My::Server;
		211
		212	sub run {
		213	my ($slave, $listener, $id) = @_;
		214
		215	close $slave; # we do not use the socket, so close it to save resources
		216
		217	# we could go ballistic and use e.g. AnyEvent here, or IO::AIO,
		218	# or anything we usually couldn't do in a process forked normally.
		219	while (my $socket = $listener->accept) {
		220	# do sth. with new socket
		221	}
		222	}
		223
		224	=head2 use AnyEvent::Fork as a faster fork+exec
		225
		226	This runs C</bin/echo hi>, with stdandard output redirected to /tmp/log
		227	and standard error redirected to the communications socket. It is usually
		228	faster than fork+exec, but still lets you prepare the environment.
		229
		230	open my $output, ">/tmp/log" or die "$!";
		231
		232	AnyEvent::Fork
		233	->new
		234	->eval ('
		235	# compile a helper function for later use
		236	sub run {
		237	my ($fh, $output, @cmd) = @_;
		238
		239	# perl will clear close-on-exec on STDOUT/STDERR
		240	open STDOUT, ">&", $output or die;
		241	open STDERR, ">&", $fh or die;
		242
		243	exec @cmd;
		244	}
		245	')
		246	->send_fh ($output)
		247	->send_arg ("/bin/echo", "hi")
		248	->run ("run", my $cv = AE::cv);
		249
		250	my $stderr = $cv->recv;
87		251
88	=head1 CONCEPTS	252	=head1 CONCEPTS
89		253
90	This module can create new processes either by executing a new perl	254	This module can create new processes either by executing a new perl
91	process, or by forking from an existing "template" process.	255	process, or by forking from an existing "template" process.
108	needed the first time. Forking from this process shares the memory used	272	needed the first time. Forking from this process shares the memory used
109	for the perl interpreter with the new process, but loading modules takes	273	for the perl interpreter with the new process, but loading modules takes
110	time, and the memory is not shared with anything else.	274	time, and the memory is not shared with anything else.
111		275
112	This is ideal for when you only need one extra process of a kind, with the	276	This is ideal for when you only need one extra process of a kind, with the
113	option of starting and stipping it on demand.	277	option of starting and stopping it on demand.
		278
		279	Example:
		280
		281	AnyEvent::Fork
		282	->new
		283	->require ("Some::Module")
		284	->run ("Some::Module::run", sub {
		285	my ($fork_fh) = @_;
		286	});
114		287
115	=item fork a new template process, load code, then fork processes off of	288	=item fork a new template process, load code, then fork processes off of
116	it and run the code	289	it and run the code
117		290
118	When you need to have a bunch of processes that all execute the same (or	291	When you need to have a bunch of processes that all execute the same (or
…		…
124	modules you loaded) is shared between the processes, and each new process	297	modules you loaded) is shared between the processes, and each new process
125	consumes relatively little memory of its own.	298	consumes relatively little memory of its own.
126		299
127	The disadvantage of this approach is that you need to create a template	300	The disadvantage of this approach is that you need to create a template
128	process for the sole purpose of forking new processes from it, but if you	301	process for the sole purpose of forking new processes from it, but if you
129	only need a fixed number of proceses you can create them, and then destroy	302	only need a fixed number of processes you can create them, and then destroy
130	the template process.	303	the template process.
		304
		305	Example:
		306
		307	my $template = AnyEvent::Fork->new->require ("Some::Module");
		308
		309	for (1..10) {
		310	$template->fork->run ("Some::Module::run", sub {
		311	my ($fork_fh) = @_;
		312	});
		313	}
		314
		315	# at this point, you can keep $template around to fork new processes
		316	# later, or you can destroy it, which causes it to vanish.
131		317
132	=item execute a new perl interpreter, load some code, run it	318	=item execute a new perl interpreter, load some code, run it
133		319
134	This is relatively slow, and doesn't allow you to share memory between	320	This is relatively slow, and doesn't allow you to share memory between
135	multiple processes.	321	multiple processes.
…		…
137	The only advantage is that you don't have to have a template process	323	The only advantage is that you don't have to have a template process
138	hanging around all the time to fork off some new processes, which might be	324	hanging around all the time to fork off some new processes, which might be
139	an advantage when there are long time spans where no extra processes are	325	an advantage when there are long time spans where no extra processes are
140	needed.	326	needed.
141		327
		328	Example:
		329
		330	AnyEvent::Fork
		331	->new_exec
		332	->require ("Some::Module")
		333	->run ("Some::Module::run", sub {
		334	my ($fork_fh) = @_;
		335	});
		336
142	=back	337	=back
143		338
144	=head1 FUNCTIONS	339	=head1 THE C<AnyEvent::Fork> CLASS
		340
		341	This module exports nothing, and only implements a single class -
		342	C<AnyEvent::Fork>.
		343
		344	There are two class constructors that both create new processes - C<new>
		345	and C<new_exec>. The C<fork> method creates a new process by forking an
		346	existing one and could be considered a third constructor.
		347
		348	Most of the remaining methods deal with preparing the new process, by
		349	loading code, evaluating code and sending data to the new process. They
		350	usually return the process object, so you can chain method calls.
		351
		352	If a process object is destroyed before calling its C<run> method, then
		353	the process simply exits. After C<run> is called, all responsibility is
		354	passed to the specified function.
		355
		356	As long as there is any outstanding work to be done, process objects
		357	resist being destroyed, so there is no reason to store them unless you
		358	need them later - configure and forget works just fine.
145		359
146	=over 4	360	=over 4
147		361
148	=cut	362	=cut
149		363
150	package AnyEvent::ProcessPool;	364	package AnyEvent::Fork;
151		365
152	use common::sense;	366	use common::sense;
153		367
154	use Socket ();	368	use Errno ();
155		369
156	use Proc::FastSpawn;
157	use AnyEvent;	370	use AnyEvent;
158	use AnyEvent::ProcessPool::Util;
159	use AnyEvent::Util ();	371	use AnyEvent::Util ();
160		372
161	BEGIN {	373	use IO::FDPass;
162	# require Exporter;
163	}
164		374
165	=item my $pool = new AnyEvent::ProcessPool key => value...	375	our $VERSION = 0.6;
166
167	Create a new process pool. The following named parameters are supported:
168		376
169	=over 4	377	=over 4
170		378
171	=back	379	=back
172		380
173	=cut	381	=cut
174		382
		383	# the early fork template process
		384	our $EARLY;
		385
175	# the template process	386	# the empty template process
176	our $template;	387	our $TEMPLATE;
177		388
178	sub _queue {	389	sub QUEUE() { 0 }
179	my ($pid, $fh) = @_;	390	sub FH() { 1 }
		391	sub WW() { 2 }
		392	sub PID() { 3 }
		393	sub CB() { 4 }
180		394
181	[	395	sub _new {
		396	my ($self, $fh, $pid) = @_;
		397
		398	AnyEvent::Util::fh_nonblocking $fh, 1;
		399
		400	$self = bless [
		401	[], # write queue - strings or fd's
		402	$fh,
		403	undef, # AE watcher
182	$pid,	404	$pid,
183	$fh,	405	], $self;
184	[],
185	undef
186	]
187	}
188		406
		407	$self
		408	}
		409
189	sub queue_cmd {	410	sub _cmd {
190	my $queue = shift;	411	my $self = shift;
191		412
192	push @{ $queue->[2] }, pack "N/a", pack "a (w/a)*", @_;	413	# ideally, we would want to use "a (w/a)*" as format string, but perl
		414	# versions from at least 5.8.9 to 5.16.3 are all buggy and can't unpack
		415	# it.
		416	push @{ $self->[QUEUE] }, pack "a L/a*", $_[0], $_[1];
193		417
194	$queue->[3] \|\|= AE::io $queue->[1], 1, sub {	418	$self->[WW] \|\|= AE::io $self->[FH], 1, sub {
		419	do {
		420	# send the next "thing" in the queue - either a reference to an fh,
		421	# or a plain string.
		422
195	if (ref $queue->[2][0]) {	423	if (ref $self->[QUEUE][0]) {
196	AnyEvent::ProcessPool::Util::fd_send fileno $queue->[1], fileno ${ $queue->[2][0] }	424	# send fh
		425	unless (IO::FDPass::send fileno $self->[FH], fileno ${ $self->[QUEUE][0] }) {
		426	return if $! == Errno::EAGAIN \|\| $! == Errno::EWOULDBLOCK;
		427	undef $self->[WW];
		428	die "AnyEvent::Fork: file descriptor send failure: $!";
		429	}
		430
197	and shift @{ $queue->[2] };	431	shift @{ $self->[QUEUE] };
		432
198	} else {	433	} else {
		434	# send string
199	my $len = syswrite $queue->[1], $queue->[2][0]	435	my $len = syswrite $self->[FH], $self->[QUEUE][0];
200	or do { undef $queue->[3]; die "AnyEvent::ProcessPool::queue write failure: $!" };	436
		437	unless ($len) {
		438	return if $! == Errno::EAGAIN \|\| $! == Errno::EWOULDBLOCK;
		439	undef $self->[3];
		440	die "AnyEvent::Fork: command write failure: $!";
		441	}
		442
201	substr $queue->[2][0], 0, $len, "";	443	substr $self->[QUEUE][0], 0, $len, "";
202	shift @{ $queue->[2] } unless length $queue->[2][0];	444	shift @{ $self->[QUEUE] } unless length $self->[QUEUE][0];
203	}	445	}
		446	} while @{ $self->[QUEUE] };
204		447
205	undef $queue->[3] unless @{ $queue->[2] };	448	# everything written
		449	undef $self->[WW];
		450
		451	# invoke run callback, if any
		452	$self->[CB]->($self->[FH]) if $self->[CB];
206	};	453	};
207	}
208		454
209	sub run_template {	455	() # make sure we don't leak the watcher
210	return if $template;	456	}
211		457
		458	# fork template from current process, used by AnyEvent::Fork::Early/Template
		459	sub _new_fork {
212	my ($fh, $slave) = AnyEvent::Util::portable_socketpair;	460	my ($fh, $slave) = AnyEvent::Util::portable_socketpair;
213	AnyEvent::Util::fh_nonblocking $fh, 1;	461	my $parent = $$;
214	fd_inherit fileno $slave;
215		462
216	my %env = %ENV;	463	my $pid = fork;
217	$env{PERL5LIB} = join ":", grep !ref, @INC;
218		464
219	my $pid = spawn	465	if ($pid eq 0) {
220	$^X,	466	require AnyEvent::Fork::Serve;
221	["perl", "-MAnyEvent::ProcessPool::Serve", "-e", "AnyEvent::ProcessPool::Serve::me", fileno $slave],	467	$AnyEvent::Fork::Serve::OWNER = $parent;
222	[map "$_=$env{$_}", keys %env],	468	close $fh;
223	or die "unable to spawn AnyEvent::ProcessPool server: $!";	469	$0 = "$_[1] of $parent";
		470	AnyEvent::Fork::Serve::serve ($slave);
		471	exit 0;
		472	} elsif (!$pid) {
		473	die "AnyEvent::Fork::Early/Template: unable to fork template process: $!";
		474	}
224		475
225	close $slave;	476	AnyEvent::Fork->_new ($fh, $pid)
226
227	$template = _queue $pid, $fh;
228
229	my ($a, $b) = AnyEvent::Util::portable_socketpair;
230
231	queue_cmd $template, "Iabc";
232	push @{ $template->[2] }, \$b;
233
234	use Coro::AnyEvent; Coro::AnyEvent::sleep 1;
235	undef $b;
236	die "x" . <$a>;
237	}	477	}
		478
		479	=item my $proc = new AnyEvent::Fork
		480
		481	Create a new "empty" perl interpreter process and returns its process
		482	object for further manipulation.
		483
		484	The new process is forked from a template process that is kept around
		485	for this purpose. When it doesn't exist yet, it is created by a call to
		486	C<new_exec> first and then stays around for future calls.
		487
		488	=cut
238		489
239	sub new {	490	sub new {
240	my $class = shift;	491	my $class = shift;
241		492
242	my $self = bless {	493	$TEMPLATE \|\|= $class->new_exec;
243	@_	494	$TEMPLATE->fork
244	}, $class;	495	}
245		496
246	run_template;	497	=item $new_proc = $proc->fork
		498
		499	Forks C<$proc>, creating a new process, and returns the process object
		500	of the new process.
		501
		502	If any of the C<send_> functions have been called before fork, then they
		503	will be cloned in the child. For example, in a pre-forked server, you
		504	might C<send_fh> the listening socket into the template process, and then
		505	keep calling C<fork> and C<run>.
		506
		507	=cut
		508
		509	sub fork {
		510	my ($self) = @_;
		511
		512	my ($fh, $slave) = AnyEvent::Util::portable_socketpair;
		513
		514	$self->send_fh ($slave);
		515	$self->_cmd ("f");
		516
		517	AnyEvent::Fork->_new ($fh)
		518	}
		519
		520	=item my $proc = new_exec AnyEvent::Fork
		521
		522	Create a new "empty" perl interpreter process and returns its process
		523	object for further manipulation.
		524
		525	Unlike the C<new> method, this method I<always> spawns a new perl process
		526	(except in some cases, see L<AnyEvent::Fork::Early> for details). This
		527	reduces the amount of memory sharing that is possible, and is also slower.
		528
		529	You should use C<new> whenever possible, except when having a template
		530	process around is unacceptable.
		531
		532	The path to the perl interpreter is divined using various methods - first
		533	C<$^X> is investigated to see if the path ends with something that sounds
		534	as if it were the perl interpreter. Failing this, the module falls back to
		535	using C<$Config::Config{perlpath}>.
		536
		537	=cut
		538
		539	sub new_exec {
		540	my ($self) = @_;
		541
		542	return $EARLY->fork
		543	if $EARLY;
		544
		545	# first find path of perl
		546	my $perl = $;
		547
		548	# first we try $^X, but the path must be absolute (always on win32), and end in sth.
		549	# that looks like perl. this obviously only works for posix and win32
		550	unless (
		551	($^O eq "MSWin32" \|\| $perl =~ m%^/%)
		552	&& $perl =~ m%[/\\]perl(?:[0-9]+(\.[0-9]+)+)?(\.exe)?$%i
		553	) {
		554	# if it doesn't look perlish enough, try Config
		555	require Config;
		556	$perl = $Config::Config{perlpath};
		557	$perl =~ s/(?:\Q$Config::Config{_exe}\E)?$/$Config::Config{_exe}/;
		558	}
		559
		560	require Proc::FastSpawn;
		561
		562	my ($fh, $slave) = AnyEvent::Util::portable_socketpair;
		563	Proc::FastSpawn::fd_inherit (fileno $slave);
		564
		565	# new fh's should always be set cloexec (due to $^F),
		566	# but hey, not on win32, so we always clear the inherit flag.
		567	Proc::FastSpawn::fd_inherit (fileno $fh, 0);
		568
		569	# quick. also doesn't work in win32. of course. what did you expect
		570	#local $ENV{PERL5LIB} = join ":", grep !ref, @INC;
		571	my %env = %ENV;
		572	$env{PERL5LIB} = join +($^O eq "MSWin32" ? ";" : ":"), grep !ref, @INC;
		573
		574	my $pid = Proc::FastSpawn::spawn (
		575	$perl,
		576	["perl", "-MAnyEvent::Fork::Serve", "-e", "AnyEvent::Fork::Serve::me", fileno $slave, $$],
		577	[map "$_=$env{$_}", keys %env],
		578	) or die "unable to spawn AnyEvent::Fork server: $!";
		579
		580	$self->_new ($fh, $pid)
		581	}
		582
		583	=item $pid = $proc->pid
		584
		585	Returns the process id of the process I<iff it is a direct child of the
		586	process running AnyEvent::Fork>, and C<undef> otherwise.
		587
		588	Normally, only processes created via C<< AnyEvent::Fork->new_exec >> and
		589	L<AnyEvent::Fork::Template> are direct children, and you are responsible
		590	to clean up their zombies when they die.
		591
		592	All other processes are not direct children, and will be cleaned up by
		593	AnyEvent::Fork itself.
		594
		595	=cut
		596
		597	sub pid {
		598	$_[0][PID]
		599	}
		600
		601	=item $proc = $proc->eval ($perlcode, @args)
		602
		603	Evaluates the given C<$perlcode> as ... perl code, while setting C<@_> to
		604	the strings specified by C<@args>, in the "main" package.
		605
		606	This call is meant to do any custom initialisation that might be required
		607	(for example, the C<require> method uses it). It's not supposed to be used
		608	to completely take over the process, use C<run> for that.
		609
		610	The code will usually be executed after this call returns, and there is no
		611	way to pass anything back to the calling process. Any evaluation errors
		612	will be reported to stderr and cause the process to exit.
		613
		614	If you want to execute some code (that isn't in a module) to take over the
		615	process, you should compile a function via C<eval> first, and then call
		616	it via C<run>. This also gives you access to any arguments passed via the
		617	C<send_xxx> methods, such as file handles. See the L<use AnyEvent::Fork as
		618	a faster fork+exec> example to see it in action.
		619
		620	Returns the process object for easy chaining of method calls.
		621
		622	=cut
		623
		624	sub eval {
		625	my ($self, $code, @args) = @_;
		626
		627	$self->_cmd (e => pack "(w/a)", $code, @args);
247		628
248	$self	629	$self
249	}	630	}
250		631
		632	=item $proc = $proc->require ($module, ...)
		633
		634	Tries to load the given module(s) into the process
		635
		636	Returns the process object for easy chaining of method calls.
		637
		638	=cut
		639
		640	sub require {
		641	my ($self, @modules) = @_;
		642
		643	s%::%/%g for @modules;
		644	$self->eval ('require "$_.pm" for @_', @modules);
		645
		646	$self
		647	}
		648
		649	=item $proc = $proc->send_fh ($handle, ...)
		650
		651	Send one or more file handles (I<not> file descriptors) to the process,
		652	to prepare a call to C<run>.
		653
		654	The process object keeps a reference to the handles until they have
		655	been passed over to the process, so you must not explicitly close the
		656	handles. This is most easily accomplished by simply not storing the file
		657	handles anywhere after passing them to this method - when AnyEvent::Fork
		658	is finished using them, perl will automatically close them.
		659
		660	Returns the process object for easy chaining of method calls.
		661
		662	Example: pass a file handle to a process, and release it without
		663	closing. It will be closed automatically when it is no longer used.
		664
		665	$proc->send_fh ($my_fh);
		666	undef $my_fh; # free the reference if you want, but DO NOT CLOSE IT
		667
		668	=cut
		669
		670	sub send_fh {
		671	my ($self, @fh) = @_;
		672
		673	for my $fh (@fh) {
		674	$self->_cmd ("h");
		675	push @{ $self->[QUEUE] }, \$fh;
		676	}
		677
		678	$self
		679	}
		680
		681	=item $proc = $proc->send_arg ($string, ...)
		682
		683	Send one or more argument strings to the process, to prepare a call to
		684	C<run>. The strings can be any octet strings.
		685
		686	The protocol is optimised to pass a moderate number of relatively short
		687	strings - while you can pass up to 4GB of data in one go, this is more
		688	meant to pass some ID information or other startup info, not big chunks of
		689	data.
		690
		691	Returns the process object for easy chaining of method calls.
		692
		693	=cut
		694
		695	sub send_arg {
		696	my ($self, @arg) = @_;
		697
		698	$self->_cmd (a => pack "(w/a)", @arg);
		699
		700	$self
		701	}
		702
		703	=item $proc->run ($func, $cb->($fh))
		704
		705	Enter the function specified by the function name in C<$func> in the
		706	process. The function is called with the communication socket as first
		707	argument, followed by all file handles and string arguments sent earlier
		708	via C<send_fh> and C<send_arg> methods, in the order they were called.
		709
		710	The process object becomes unusable on return from this function - any
		711	further method calls result in undefined behaviour.
		712
		713	The function name should be fully qualified, but if it isn't, it will be
		714	looked up in the C<main> package.
		715
		716	If the called function returns, doesn't exist, or any error occurs, the
		717	process exits.
		718
		719	Preparing the process is done in the background - when all commands have
		720	been sent, the callback is invoked with the local communications socket
		721	as argument. At this point you can start using the socket in any way you
		722	like.
		723
		724	If the communication socket isn't used, it should be closed on both sides,
		725	to save on kernel memory.
		726
		727	The socket is non-blocking in the parent, and blocking in the newly
		728	created process. The close-on-exec flag is set in both.
		729
		730	Even if not used otherwise, the socket can be a good indicator for the
		731	existence of the process - if the other process exits, you get a readable
		732	event on it, because exiting the process closes the socket (if it didn't
		733	create any children using fork).
		734
		735	Example: create a template for a process pool, pass a few strings, some
		736	file handles, then fork, pass one more string, and run some code.
		737
		738	my $pool = AnyEvent::Fork
		739	->new
		740	->send_arg ("str1", "str2")
		741	->send_fh ($fh1, $fh2);
		742
		743	for (1..2) {
		744	$pool
		745	->fork
		746	->send_arg ("str3")
		747	->run ("Some::function", sub {
		748	my ($fh) = @_;
		749
		750	# fh is nonblocking, but we trust that the OS can accept these
		751	# few octets anyway.
		752	syswrite $fh, "hi #$_\n";
		753
		754	# $fh is being closed here, as we don't store it anywhere
		755	});
		756	}
		757
		758	# Some::function might look like this - all parameters passed before fork
		759	# and after will be passed, in order, after the communications socket.
		760	sub Some::function {
		761	my ($fh, $str1, $str2, $fh1, $fh2, $str3) = @_;
		762
		763	print scalar <$fh>; # prints "hi #1\n" and "hi #2\n" in any order
		764	}
		765
		766	=cut
		767
		768	sub run {
		769	my ($self, $func, $cb) = @_;
		770
		771	$self->[CB] = $cb;
		772	$self->_cmd (r => $func);
		773	}
		774
251	=back	775	=back
		776
		777	=head1 PERFORMANCE
		778
		779	Now for some unscientific benchmark numbers (all done on an amd64
		780	GNU/Linux box). These are intended to give you an idea of the relative
		781	performance you can expect, they are not meant to be absolute performance
		782	numbers.
		783
		784	OK, so, I ran a simple benchmark that creates a socket pair, forks, calls
		785	exit in the child and waits for the socket to close in the parent. I did
		786	load AnyEvent, EV and AnyEvent::Fork, for a total process size of 5100kB.
		787
		788	2079 new processes per second, using manual socketpair + fork
		789
		790	Then I did the same thing, but instead of calling fork, I called
		791	AnyEvent::Fork->new->run ("CORE::exit") and then again waited for the
		792	socket form the child to close on exit. This does the same thing as manual
		793	socket pair + fork, except that what is forked is the template process
		794	(2440kB), and the socket needs to be passed to the server at the other end
		795	of the socket first.
		796
		797	2307 new processes per second, using AnyEvent::Fork->new
		798
		799	And finally, using C<new_exec> instead C<new>, using vforks+execs to exec
		800	a new perl interpreter and compile the small server each time, I get:
		801
		802	479 vfork+execs per second, using AnyEvent::Fork->new_exec
		803
		804	So how can C<< AnyEvent->new >> be faster than a standard fork, even
		805	though it uses the same operations, but adds a lot of overhead?
		806
		807	The difference is simply the process size: forking the 5MB process takes
		808	so much longer than forking the 2.5MB template process that the extra
		809	overhead introduced is canceled out.
		810
		811	If the benchmark process grows, the normal fork becomes even slower:
		812
		813	1340 new processes, manual fork of a 20MB process
		814	731 new processes, manual fork of a 200MB process
		815	235 new processes, manual fork of a 2000MB process
		816
		817	What that means (to me) is that I can use this module without having a bad
		818	conscience because of the extra overhead required to start new processes.
		819
		820	=head1 TYPICAL PROBLEMS
		821
		822	This section lists typical problems that remain. I hope by recognising
		823	them, most can be avoided.
		824
		825	=over 4
		826
		827	=item leaked file descriptors for exec'ed processes
		828
		829	POSIX systems inherit file descriptors by default when exec'ing a new
		830	process. While perl itself laudably sets the close-on-exec flags on new
		831	file handles, most C libraries don't care, and even if all cared, it's
		832	often not possible to set the flag in a race-free manner.
		833
		834	That means some file descriptors can leak through. And since it isn't
		835	possible to know which file descriptors are "good" and "necessary" (or
		836	even to know which file descriptors are open), there is no good way to
		837	close the ones that might harm.
		838
		839	As an example of what "harm" can be done consider a web server that
		840	accepts connections and afterwards some module uses AnyEvent::Fork for the
		841	first time, causing it to fork and exec a new process, which might inherit
		842	the network socket. When the server closes the socket, it is still open
		843	in the child (which doesn't even know that) and the client might conclude
		844	that the connection is still fine.
		845
		846	For the main program, there are multiple remedies available -
		847	L<AnyEvent::Fork::Early> is one, creating a process early and not using
		848	C<new_exec> is another, as in both cases, the first process can be exec'ed
		849	well before many random file descriptors are open.
		850
		851	In general, the solution for these kind of problems is to fix the
		852	libraries or the code that leaks those file descriptors.
		853
		854	Fortunately, most of these leaked descriptors do no harm, other than
		855	sitting on some resources.
		856
		857	=item leaked file descriptors for fork'ed processes
		858
		859	Normally, L<AnyEvent::Fork> does start new processes by exec'ing them,
		860	which closes file descriptors not marked for being inherited.
		861
		862	However, L<AnyEvent::Fork::Early> and L<AnyEvent::Fork::Template> offer
		863	a way to create these processes by forking, and this leaks more file
		864	descriptors than exec'ing them, as there is no way to mark descriptors as
		865	"close on fork".
		866
		867	An example would be modules like L<EV>, L<IO::AIO> or L<Gtk2>. Both create
		868	pipes for internal uses, and L<Gtk2> might open a connection to the X
		869	server. L<EV> and L<IO::AIO> can deal with fork, but Gtk2 might have
		870	trouble with a fork.
		871
		872	The solution is to either not load these modules before use'ing
		873	L<AnyEvent::Fork::Early> or L<AnyEvent::Fork::Template>, or to delay
		874	initialising them, for example, by calling C<init Gtk2> manually.
		875
		876	=item exiting calls object destructors
		877
		878	This only applies to users of L<AnyEvent::Fork:Early> and
		879	L<AnyEvent::Fork::Template>, or when initialiasing code creates objects
		880	that reference external resources.
		881
		882	When a process created by AnyEvent::Fork exits, it might do so by calling
		883	exit, or simply letting perl reach the end of the program. At which point
		884	Perl runs all destructors.
		885
		886	Not all destructors are fork-safe - for example, an object that represents
		887	the connection to an X display might tell the X server to free resources,
		888	which is inconvenient when the "real" object in the parent still needs to
		889	use them.
		890
		891	This is obviously not a problem for L<AnyEvent::Fork::Early>, as you used
		892	it as the very first thing, right?
		893
		894	It is a problem for L<AnyEvent::Fork::Template> though - and the solution
		895	is to not create objects with nontrivial destructors that might have an
		896	effect outside of Perl.
		897
		898	=back
		899
		900	=head1 PORTABILITY NOTES
		901
		902	Native win32 perls are somewhat supported (AnyEvent::Fork::Early is a nop,
		903	and ::Template is not going to work), and it cost a lot of blood and sweat
		904	to make it so, mostly due to the bloody broken perl that nobody seems to
		905	care about. The fork emulation is a bad joke - I have yet to see something
		906	useful that you can do with it without running into memory corruption
		907	issues or other braindamage. Hrrrr.
		908
		909	Cygwin perl is not supported at the moment due to some hilarious
		910	shortcomings of its API - see L<IO::FDPoll> for more details.
		911
		912	=head1 SEE ALSO
		913
		914	L<AnyEvent::Fork::Early> (to avoid executing a perl interpreter),
		915	L<AnyEvent::Fork::Template> (to create a process by forking the main
		916	program at a convenient time).
252		917
253	=head1 AUTHOR	918	=head1 AUTHOR
254		919
255	Marc Lehmann <schmorp@schmorp.de>	920	Marc Lehmann <schmorp@schmorp.de>
256	http://home.schmorp.de/	921	http://home.schmorp.de/

Diff Legend

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

Comparing AnyEvent-Fork/Fork.pm (file contents): Revision 1.3 by root, Tue Apr 2 18:00:04 2013 UTC vs. Revision 1.42 by root, Mon Apr 8 05:44:23 2013 UTC

Diff Legend

Comparing AnyEvent-Fork/Fork.pm (file contents):
Revision 1.3 by root, Tue Apr 2 18:00:04 2013 UTC vs.
Revision 1.42 by root, Mon Apr 8 05:44:23 2013 UTC