ViewVC Help

View File | Revision Log | Show Annotations | Download File

/cvs/IO-AIO/AIO.pm

Comparing IO-AIO/AIO.pm (file contents):
Revision 1.57 by root, Sun Oct 22 01:28:31 2006 UTC vs.
Revision 1.129 by root, Tue Jun 17 23:38:10 2008 UTC

…		…
5	=head1 SYNOPSIS	5	=head1 SYNOPSIS
6		6
7	use IO::AIO;	7	use IO::AIO;
8		8
9	aio_open "/etc/passwd", O_RDONLY, 0, sub {	9	aio_open "/etc/passwd", O_RDONLY, 0, sub {
10	my ($fh) = @_;	10	my $fh = shift
		11	or die "/etc/passwd: $!";
11	...	12	...
12	};	13	};
13		14
14	aio_unlink "/tmp/file", sub { };	15	aio_unlink "/tmp/file", sub { };
15		16
…		…
18	};	19	};
19		20
20	# version 2+ has request and group objects	21	# version 2+ has request and group objects
21	use IO::AIO 2;	22	use IO::AIO 2;
22		23
		24	aioreq_pri 4; # give next request a very high priority
23	my $req = aio_unlink "/tmp/file", sub { };	25	my $req = aio_unlink "/tmp/file", sub { };
24	$req->cancel; # cancel request if still in queue	26	$req->cancel; # cancel request if still in queue
25		27
26	my $grp = aio_group sub { print "all stats done\n" };	28	my $grp = aio_group sub { print "all stats done\n" };
27	add $grp aio_stat "..." for ...;	29	add $grp aio_stat "..." for ...;
28		30
		31	# AnyEvent integration (EV, Event, Glib, Tk, POE, urxvt, pureperl...)
		32	use AnyEvent::AIO;
		33
29	# AnyEvent integration	34	# EV integration
30	open my $fh, "<&=" . IO::AIO::poll_fileno or die "$!";	35	my $w = EV::io IO::AIO::poll_fileno, EV::READ, \&IO::AIO::poll_cb;
31	my $w = AnyEvent->io (fh => $fh, poll => 'r', cb => sub { IO::AIO::poll_cb });
32		36
33	# Event integration	37	# Event integration
34	Event->io (fd => IO::AIO::poll_fileno,	38	Event->io (fd => IO::AIO::poll_fileno,
35	poll => 'r',	39	poll => 'r',
36	cb => \&IO::AIO::poll_cb);	40	cb => \&IO::AIO::poll_cb);
…		…
50	=head1 DESCRIPTION	54	=head1 DESCRIPTION
51		55
52	This module implements asynchronous I/O using whatever means your	56	This module implements asynchronous I/O using whatever means your
53	operating system supports.	57	operating system supports.
54		58
		59	Asynchronous means that operations that can normally block your program
		60	(e.g. reading from disk) will be done asynchronously: the operation
		61	will still block, but you can do something else in the meantime. This
		62	is extremely useful for programs that need to stay interactive even
		63	when doing heavy I/O (GUI programs, high performance network servers
		64	etc.), but can also be used to easily do operations in parallel that are
		65	normally done sequentially, e.g. stat'ing many files, which is much faster
		66	on a RAID volume or over NFS when you do a number of stat operations
		67	concurrently.
		68
		69	While most of this works on all types of file descriptors (for
		70	example sockets), using these functions on file descriptors that
		71	support nonblocking operation (again, sockets, pipes etc.) is very
		72	inefficient. Use an event loop for that (such as the L<Event|Event>
		73	module): IO::AIO will naturally fit into such an event loop itself.
		74
55	Currently, a number of threads are started that execute your read/writes	75	In this version, a number of threads are started that execute your
56	and signal their completion. You don't need thread support in your libc or	76	requests and signal their completion. You don't need thread support
57	perl, and the threads created by this module will not be visible to the	77	in perl, and the threads created by this module will not be visible
58	pthreads library. In the future, this module might make use of the native	78	to perl. In the future, this module might make use of the native aio
59	aio functions available on many operating systems. However, they are often	79	functions available on many operating systems. However, they are often
60	not well-supported (Linux doesn't allow them on normal files currently,	80	not well-supported or restricted (GNU/Linux doesn't allow them on normal
61	for example), and they would only support aio_read and aio_write, so the	81	files currently, for example), and they would only support aio_read and
62	remaining functionality would have to be implemented using threads anyway.	82	aio_write, so the remaining functionality would have to be implemented
		83	using threads anyway.
63		84
64	Although the module will work with in the presence of other threads, it is	85	Although the module will work in the presence of other (Perl-) threads,
65	currently not reentrant, so use appropriate locking yourself, always call	86	it is currently not reentrant in any way, so use appropriate locking
66	C<poll_cb> from within the same thread, or never call C<poll_cb> (or other	87	yourself, always call C<poll_cb> from within the same thread, or never
67	C<aio_> functions) recursively.	88	call C<poll_cb> (or other C<aio_> functions) recursively.
		89
		90	=head2 EXAMPLE
		91
		92	This is a simple example that uses the Event module and loads
		93	F</etc/passwd> asynchronously:
		94
		95	use Fcntl;
		96	use Event;
		97	use IO::AIO;
		98
		99	# register the IO::AIO callback with Event
		100	Event->io (fd => IO::AIO::poll_fileno,
		101	poll => 'r',
		102	cb => \&IO::AIO::poll_cb);
		103
		104	# queue the request to open /etc/passwd
		105	aio_open "/etc/passwd", O_RDONLY, 0, sub {
		106	my $fh = shift
		107	or die "error while opening: $!";
		108
		109	# stat'ing filehandles is generally non-blocking
		110	my $size = -s $fh;
		111
		112	# queue a request to read the file
		113	my $contents;
		114	aio_read $fh, 0, $size, $contents, 0, sub {
		115	$_[0] == $size
		116	or die "short read: $!";
		117
		118	close $fh;
		119
		120	# file contents now in $contents
		121	print $contents;
		122
		123	# exit event loop and program
		124	Event::unloop;
		125	};
		126	};
		127
		128	# possibly queue up other requests, or open GUI windows,
		129	# check for sockets etc. etc.
		130
		131	# process events as long as there are some:
		132	Event::loop;
		133
		134	=head1 REQUEST ANATOMY AND LIFETIME
		135
		136	Every C<aio_*> function creates a request. which is a C data structure not
		137	directly visible to Perl.
		138
		139	If called in non-void context, every request function returns a Perl
		140	object representing the request. In void context, nothing is returned,
		141	which saves a bit of memory.
		142
		143	The perl object is a fairly standard ref-to-hash object. The hash contents
		144	are not used by IO::AIO so you are free to store anything you like in it.
		145
		146	During their existance, aio requests travel through the following states,
		147	in order:
		148
		149	=over 4
		150
		151	=item ready
		152
		153	Immediately after a request is created it is put into the ready state,
		154	waiting for a thread to execute it.
		155
		156	=item execute
		157
		158	A thread has accepted the request for processing and is currently
		159	executing it (e.g. blocking in read).
		160
		161	=item pending
		162
		163	The request has been executed and is waiting for result processing.
		164
		165	While request submission and execution is fully asynchronous, result
		166	processing is not and relies on the perl interpreter calling C<poll_cb>
		167	(or another function with the same effect).
		168
		169	=item result
		170
		171	The request results are processed synchronously by C<poll_cb>.
		172
		173	The C<poll_cb> function will process all outstanding aio requests by
		174	calling their callbacks, freeing memory associated with them and managing
		175	any groups they are contained in.
		176
		177	=item done
		178
		179	Request has reached the end of its lifetime and holds no resources anymore
		180	(except possibly for the Perl object, but its connection to the actual
		181	aio request is severed and calling its methods will either do nothing or
		182	result in a runtime error).
		183
		184	=back
68		185
69	=cut	186	=cut
70		187
71	package IO::AIO;	188	package IO::AIO;
		189
		190	use Carp ();
72		191
73	no warnings;	192	no warnings;
74	use strict 'vars';	193	use strict 'vars';
75		194
76	use base 'Exporter';	195	use base 'Exporter';
77		196
78	BEGIN {	197	BEGIN {
79	our $VERSION = '2.0';	198	our $VERSION = '3.04';
80		199
81	our @EXPORT = qw(aio_sendfile aio_read aio_write aio_open aio_close aio_stat	200	our @AIO_REQ = qw(aio_sendfile aio_read aio_write aio_open aio_close
82	aio_lstat aio_unlink aio_rmdir aio_readdir aio_scandir aio_symlink	201	aio_stat aio_lstat aio_unlink aio_rmdir aio_readdir
83	aio_fsync aio_fdatasync aio_readahead aio_rename aio_link aio_move	202	aio_scandir aio_symlink aio_readlink aio_sync aio_fsync
84	aio_group);	203	aio_fdatasync aio_pathsync aio_readahead
85	our @EXPORT_OK = qw(poll_fileno poll_cb min_parallel max_parallel max_outstanding nreqs);	204	aio_rename aio_link aio_move aio_copy aio_group
		205	aio_nop aio_mknod aio_load aio_rmtree aio_mkdir aio_chown
		206	aio_chmod aio_utime aio_truncate);
		207
		208	our @EXPORT = (@AIO_REQ, qw(aioreq_pri aioreq_nice));
		209	our @EXPORT_OK = qw(poll_fileno poll_cb poll_wait flush
		210	min_parallel max_parallel max_idle
		211	nreqs nready npending nthreads
		212	max_poll_time max_poll_reqs);
86		213
87	@IO::AIO::GRP::ISA = 'IO::AIO::REQ';	214	@IO::AIO::GRP::ISA = 'IO::AIO::REQ';
88		215
89	require XSLoader;	216	require XSLoader;
90	XSLoader::load ("IO::AIO", $VERSION);	217	XSLoader::load ("IO::AIO", $VERSION);
91	}	218	}
92		219
93	=head1 FUNCTIONS	220	=head1 FUNCTIONS
94		221
95	=head2 AIO FUNCTIONS	222	=head2 AIO REQUEST FUNCTIONS
96		223
97	All the C<aio_*> calls are more or less thin wrappers around the syscall	224	All the C<aio_*> calls are more or less thin wrappers around the syscall
98	with the same name (sans C<aio_>). The arguments are similar or identical,	225	with the same name (sans C<aio_>). The arguments are similar or identical,
99	and they all accept an additional (and optional) C<$callback> argument	226	and they all accept an additional (and optional) C<$callback> argument
100	which must be a code reference. This code reference will get called with	227	which must be a code reference. This code reference will get called with
…		…
103	syscall has been executed asynchronously.	230	syscall has been executed asynchronously.
104		231
105	All functions expecting a filehandle keep a copy of the filehandle	232	All functions expecting a filehandle keep a copy of the filehandle
106	internally until the request has finished.	233	internally until the request has finished.
107		234
108	All requests return objects of type L<IO::AIO::REQ> that allow further	235	All functions return request objects of type L<IO::AIO::REQ> that allow
109	manipulation of those requests while they are in-flight.	236	further manipulation of those requests while they are in-flight.
110		237
111	The pathnames you pass to these routines I<must> be absolute and	238	The pathnames you pass to these routines I<must> be absolute and
112	encoded in byte form. The reason for the former is that at the time the	239	encoded as octets. The reason for the former is that at the time the
113	request is being executed, the current working directory could have	240	request is being executed, the current working directory could have
114	changed. Alternatively, you can make sure that you never change the	241	changed. Alternatively, you can make sure that you never change the
115	current working directory.	242	current working directory anywhere in the program and then use relative
		243	paths.
116		244
117	To encode pathnames to byte form, either make sure you either: a)	245	To encode pathnames as octets, either make sure you either: a) always pass
118	always pass in filenames you got from outside (command line, readdir	246	in filenames you got from outside (command line, readdir etc.) without
119	etc.), b) are ASCII or ISO 8859-1, c) use the Encode module and encode	247	tinkering, b) are ASCII or ISO 8859-1, c) use the Encode module and encode
120	your pathnames to the locale (or other) encoding in effect in the user	248	your pathnames to the locale (or other) encoding in effect in the user
121	environment, d) use Glib::filename_from_unicode on unicode filenames or e)	249	environment, d) use Glib::filename_from_unicode on unicode filenames or e)
122	use something else.	250	use something else to ensure your scalar has the correct contents.
		251
		252	This works, btw. independent of the internal UTF-8 bit, which IO::AIO
		253	handles correctly wether it is set or not.
123		254
124	=over 4	255	=over 4
		256
		257	=item $prev_pri = aioreq_pri [$pri]
		258
		259	Returns the priority value that would be used for the next request and, if
		260	C<$pri> is given, sets the priority for the next aio request.
		261
		262	The default priority is C<0>, the minimum and maximum priorities are C<-4>
		263	and C<4>, respectively. Requests with higher priority will be serviced
		264	first.
		265
		266	The priority will be reset to C<0> after each call to one of the C<aio_*>
		267	functions.
		268
		269	Example: open a file with low priority, then read something from it with
		270	higher priority so the read request is serviced before other low priority
		271	open requests (potentially spamming the cache):
		272
		273	aioreq_pri -3;
		274	aio_open ..., sub {
		275	return unless $_[0];
		276
		277	aioreq_pri -2;
		278	aio_read $_[0], ..., sub {
		279	...
		280	};
		281	};
		282
		283
		284	=item aioreq_nice $pri_adjust
		285
		286	Similar to C<aioreq_pri>, but subtracts the given value from the current
		287	priority, so the effect is cumulative.
		288
125		289
126	=item aio_open $pathname, $flags, $mode, $callback->($fh)	290	=item aio_open $pathname, $flags, $mode, $callback->($fh)
127		291
128	Asynchronously open or create a file and call the callback with a newly	292	Asynchronously open or create a file and call the callback with a newly
129	created filehandle for the file.	293	created filehandle for the file.
…		…
135	list. They are the same as used by C<sysopen>.	299	list. They are the same as used by C<sysopen>.
136		300
137	Likewise, C<$mode> specifies the mode of the newly created file, if it	301	Likewise, C<$mode> specifies the mode of the newly created file, if it
138	didn't exist and C<O_CREAT> has been given, just like perl's C<sysopen>,	302	didn't exist and C<O_CREAT> has been given, just like perl's C<sysopen>,
139	except that it is mandatory (i.e. use C<0> if you don't create new files,	303	except that it is mandatory (i.e. use C<0> if you don't create new files,
140	and C<0666> or C<0777> if you do).	304	and C<0666> or C<0777> if you do). Note that the C<$mode> will be modified
		305	by the umask in effect then the request is being executed, so better never
		306	change the umask.
141		307
142	Example:	308	Example:
143		309
144	aio_open "/etc/passwd", O_RDONLY, 0, sub {	310	aio_open "/etc/passwd", O_RDONLY, 0, sub {
145	if ($_[0]) {	311	if ($_[0]) {
…		…
148	} else {	314	} else {
149	die "open failed: $!\n";	315	die "open failed: $!\n";
150	}	316	}
151	};	317	};
152		318
		319
153	=item aio_close $fh, $callback->($status)	320	=item aio_close $fh, $callback->($status)
154		321
155	Asynchronously close a file and call the callback with the result	322	Asynchronously close a file and call the callback with the result
156	code. I<WARNING:> although accepted, you should not pass in a perl	323	code.
157	filehandle here, as perl will likely close the file descriptor another
158	time when the filehandle is destroyed. Normally, you can safely call perls
159	C<close> or just let filehandles go out of scope.
160		324
161	This is supposed to be a bug in the API, so that might change. It's	325	Unfortunately, you can't do this to perl. Perl I<insists> very strongly on
162	therefore best to avoid this function.	326	closing the file descriptor associated with the filehandle itself.
		327
		328	Therefore, C<aio_close> will not close the filehandle - instead it will
		329	use dup2 to overwrite the file descriptor with the write-end of a pipe
		330	(the pipe fd will be created on demand and will be cached).
		331
		332	Or in other words: the file descriptor will be closed, but it will not be
		333	free for reuse until the perl filehandle is closed.
		334
		335	=cut
163		336
164	=item aio_read $fh,$offset,$length, $data,$dataoffset, $callback->($retval)	337	=item aio_read $fh,$offset,$length, $data,$dataoffset, $callback->($retval)
165		338
166	=item aio_write $fh,$offset,$length, $data,$dataoffset, $callback->($retval)	339	=item aio_write $fh,$offset,$length, $data,$dataoffset, $callback->($retval)
167		340
168	Reads or writes C<length> bytes from the specified C<fh> and C<offset>	341	Reads or writes C<$length> bytes from the specified C<$fh> and C<$offset>
169	into the scalar given by C<data> and offset C<dataoffset> and calls the	342	into the scalar given by C<$data> and offset C<$dataoffset> and calls the
170	callback without the actual number of bytes read (or -1 on error, just	343	callback without the actual number of bytes read (or -1 on error, just
171	like the syscall).	344	like the syscall).
172		345
		346	If C<$offset> is undefined, then the current file descriptor offset will
		347	be used (and updated), otherwise the file descriptor offset will not be
		348	changed by these calls.
		349
		350	If C<$length> is undefined in C<aio_write>, use the remaining length of C<$data>.
		351
		352	If C<$dataoffset> is less than zero, it will be counted from the end of
		353	C<$data>.
		354
173	The C<$data> scalar I<MUST NOT> be modified in any way while the request	355	The C<$data> scalar I<MUST NOT> be modified in any way while the request
174	is outstanding. Modifying it can result in segfaults or WW3 (if the	356	is outstanding. Modifying it can result in segfaults or World War III (if
175	necessary/optional hardware is installed).	357	the necessary/optional hardware is installed).
176		358
177	Example: Read 15 bytes at offset 7 into scalar C<$buffer>, starting at	359	Example: Read 15 bytes at offset 7 into scalar C<$buffer>, starting at
178	offset C<0> within the scalar:	360	offset C<0> within the scalar:
179		361
180	aio_read $fh, 7, 15, $buffer, 0, sub {	362	aio_read $fh, 7, 15, $buffer, 0, sub {
181	$_[0] > 0 or die "read error: $!";	363	$_[0] > 0 or die "read error: $!";
182	print "read $_[0] bytes: <$buffer>\n";	364	print "read $_[0] bytes: <$buffer>\n";
183	};	365	};
184		366
185	=item aio_move $srcpath, $dstpath, $callback->($status)
186
187	Try to move the I<file> (directories not supported as either source or
188	destination) from C<$srcpath> to C<$dstpath> and call the callback with
189	the C<0> (error) or C<-1> ok.
190
191	This is a composite request that tries to rename(2) the file first. If
192	rename files with C<EXDEV>, it creates the destination file with mode 0200
193	and copies the contents of the source file into it using C<aio_sendfile>,
194	followed by restoring atime, mtime, access mode and uid/gid, in that
195	order, and unlinking the C<$srcpath>.
196
197	If an error occurs, the partial destination file will be unlinked, if
198	possible, except when setting atime, mtime, access mode and uid/gid, where
199	errors are being ignored.
200
201	=cut
202
203	sub aio_move($$$) {
204	my ($src, $dst, $cb) = @_;
205
206	my $grp = aio_group;
207
208	add $grp aio_rename $src, $dst, sub {
209	if ($_[0] && $! == EXDEV) {
210	add $grp aio_open $src, O_RDONLY, 0, sub {
211	if (my $src_fh = $_[0]) {
212	my @stat = stat $src_fh;
213
214	add $grp aio_open $dst, O_WRONLY, 0200, sub {
215	if (my $dst_fh = $_[0]) {
216	add $grp aio_sendfile $dst_fh, $src_fh, 0, $stat[7], sub {
217	close $src_fh;
218
219	if ($_[0] == $stat[7]) {
220	utime $stat[8], $stat[9], $dst;
221	chmod $stat[2] & 07777, $dst_fh;
222	chown $stat[4], $stat[5], $dst_fh;
223	close $dst_fh;
224
225	add $grp aio_unlink $src, sub {
226	$cb->($_[0]);
227	};
228	} else {
229	my $errno = $!;
230	add $grp aio_unlink $dst, sub {
231	$! = $errno;
232	$cb->(-1);
233	};
234	}
235	};
236	} else {
237	$cb->(-1);
238	}
239	},
240
241	} else {
242	$cb->(-1);
243	}
244	};
245	} else {
246	$cb->($_[0]);
247	}
248	};
249
250	$grp
251	}
252		367
253	=item aio_sendfile $out_fh, $in_fh, $in_offset, $length, $callback->($retval)	368	=item aio_sendfile $out_fh, $in_fh, $in_offset, $length, $callback->($retval)
254		369
255	Tries to copy C<$length> bytes from C<$in_fh> to C<$out_fh>. It starts	370	Tries to copy C<$length> bytes from C<$in_fh> to C<$out_fh>. It starts
256	reading at byte offset C<$in_offset>, and starts writing at the current	371	reading at byte offset C<$in_offset>, and starts writing at the current
…		…
270	C<$in_fh> than are written, and there is no way to find out how many	385	C<$in_fh> than are written, and there is no way to find out how many
271	bytes have been read from C<aio_sendfile> alone, as C<aio_sendfile> only	386	bytes have been read from C<aio_sendfile> alone, as C<aio_sendfile> only
272	provides the number of bytes written to C<$out_fh>. Only if the result	387	provides the number of bytes written to C<$out_fh>. Only if the result
273	value equals C<$length> one can assume that C<$length> bytes have been	388	value equals C<$length> one can assume that C<$length> bytes have been
274	read.	389	read.
		390
275		391
276	=item aio_readahead $fh,$offset,$length, $callback->($retval)	392	=item aio_readahead $fh,$offset,$length, $callback->($retval)
277		393
278	C<aio_readahead> populates the page cache with data from a file so that	394	C<aio_readahead> populates the page cache with data from a file so that
279	subsequent reads from that file will not block on disk I/O. The C<$offset>	395	subsequent reads from that file will not block on disk I/O. The C<$offset>
…		…
285	file. The current file offset of the file is left unchanged.	401	file. The current file offset of the file is left unchanged.
286		402
287	If that syscall doesn't exist (likely if your OS isn't Linux) it will be	403	If that syscall doesn't exist (likely if your OS isn't Linux) it will be
288	emulated by simply reading the data, which would have a similar effect.	404	emulated by simply reading the data, which would have a similar effect.
289		405
		406
290	=item aio_stat $fh_or_path, $callback->($status)	407	=item aio_stat $fh_or_path, $callback->($status)
291		408
292	=item aio_lstat $fh, $callback->($status)	409	=item aio_lstat $fh, $callback->($status)
293		410
294	Works like perl's C<stat> or C<lstat> in void context. The callback will	411	Works like perl's C<stat> or C<lstat> in void context. The callback will
…		…
307	aio_stat "/etc/passwd", sub {	424	aio_stat "/etc/passwd", sub {
308	$_[0] and die "stat failed: $!";	425	$_[0] and die "stat failed: $!";
309	print "size is ", -s _, "\n";	426	print "size is ", -s _, "\n";
310	};	427	};
311		428
		429
		430	=item aio_utime $fh_or_path, $atime, $mtime, $callback->($status)
		431
		432	Works like perl's C<utime> function (including the special case of $atime
		433	and $mtime being undef). Fractional times are supported if the underlying
		434	syscalls support them.
		435
		436	When called with a pathname, uses utimes(2) if available, otherwise
		437	utime(2). If called on a file descriptor, uses futimes(2) if available,
		438	otherwise returns ENOSYS, so this is not portable.
		439
		440	Examples:
		441
		442	# set atime and mtime to current time (basically touch(1)):
		443	aio_utime "path", undef, undef;
		444	# set atime to current time and mtime to beginning of the epoch:
		445	aio_utime "path", time, undef; # undef==0
		446
		447
		448	=item aio_chown $fh_or_path, $uid, $gid, $callback->($status)
		449
		450	Works like perl's C<chown> function, except that C<undef> for either $uid
		451	or $gid is being interpreted as "do not change" (but -1 can also be used).
		452
		453	Examples:
		454
		455	# same as "chown root path" in the shell:
		456	aio_chown "path", 0, -1;
		457	# same as above:
		458	aio_chown "path", 0, undef;
		459
		460
		461	=item aio_truncate $fh_or_path, $offset, $callback->($status)
		462
		463	Works like truncate(2) or ftruncate(2).
		464
		465
		466	=item aio_chmod $fh_or_path, $mode, $callback->($status)
		467
		468	Works like perl's C<chmod> function.
		469
		470
312	=item aio_unlink $pathname, $callback->($status)	471	=item aio_unlink $pathname, $callback->($status)
313		472
314	Asynchronously unlink (delete) a file and call the callback with the	473	Asynchronously unlink (delete) a file and call the callback with the
315	result code.	474	result code.
316		475
		476
		477	=item aio_mknod $path, $mode, $dev, $callback->($status)
		478
		479	[EXPERIMENTAL]
		480
		481	Asynchronously create a device node (or fifo). See mknod(2).
		482
		483	The only (POSIX-) portable way of calling this function is:
		484
		485	aio_mknod $path, IO::AIO::S_IFIFO | $mode, 0, sub { ...
		486
		487
317	=item aio_link $srcpath, $dstpath, $callback->($status)	488	=item aio_link $srcpath, $dstpath, $callback->($status)
318		489
319	Asynchronously create a new link to the existing object at C<$srcpath> at	490	Asynchronously create a new link to the existing object at C<$srcpath> at
320	the path C<$dstpath> and call the callback with the result code.	491	the path C<$dstpath> and call the callback with the result code.
321		492
		493
322	=item aio_symlink $srcpath, $dstpath, $callback->($status)	494	=item aio_symlink $srcpath, $dstpath, $callback->($status)
323		495
324	Asynchronously create a new symbolic link to the existing object at C<$srcpath> at	496	Asynchronously create a new symbolic link to the existing object at C<$srcpath> at
325	the path C<$dstpath> and call the callback with the result code.	497	the path C<$dstpath> and call the callback with the result code.
326		498
		499
		500	=item aio_readlink $path, $callback->($link)
		501
		502	Asynchronously read the symlink specified by C<$path> and pass it to
		503	the callback. If an error occurs, nothing or undef gets passed to the
		504	callback.
		505
		506
327	=item aio_rename $srcpath, $dstpath, $callback->($status)	507	=item aio_rename $srcpath, $dstpath, $callback->($status)
328		508
329	Asynchronously rename the object at C<$srcpath> to C<$dstpath>, just as	509	Asynchronously rename the object at C<$srcpath> to C<$dstpath>, just as
330	rename(2) and call the callback with the result code.	510	rename(2) and call the callback with the result code.
331		511
		512
		513	=item aio_mkdir $pathname, $mode, $callback->($status)
		514
		515	Asynchronously mkdir (create) a directory and call the callback with
		516	the result code. C<$mode> will be modified by the umask at the time the
		517	request is executed, so do not change your umask.
		518
		519
332	=item aio_rmdir $pathname, $callback->($status)	520	=item aio_rmdir $pathname, $callback->($status)
333		521
334	Asynchronously rmdir (delete) a directory and call the callback with the	522	Asynchronously rmdir (delete) a directory and call the callback with the
335	result code.	523	result code.
		524
336		525
337	=item aio_readdir $pathname, $callback->($entries)	526	=item aio_readdir $pathname, $callback->($entries)
338		527
339	Unlike the POSIX call of the same name, C<aio_readdir> reads an entire	528	Unlike the POSIX call of the same name, C<aio_readdir> reads an entire
340	directory (i.e. opendir + readdir + closedir). The entries will not be	529	directory (i.e. opendir + readdir + closedir). The entries will not be
341	sorted, and will B<NOT> include the C<.> and C<..> entries.	530	sorted, and will B<NOT> include the C<.> and C<..> entries.
342		531
343	The callback a single argument which is either C<undef> or an array-ref	532	The callback a single argument which is either C<undef> or an array-ref
344	with the filenames.	533	with the filenames.
345		534
		535
		536	=item aio_load $path, $data, $callback->($status)
		537
		538	This is a composite request that tries to fully load the given file into
		539	memory. Status is the same as with aio_read.
		540
		541	=cut
		542
		543	sub aio_load($$;$) {
		544	my ($path, undef, $cb) = @_;
		545	my $data = \$_[1];
		546
		547	my $pri = aioreq_pri;
		548	my $grp = aio_group $cb;
		549
		550	aioreq_pri $pri;
		551	add $grp aio_open $path, O_RDONLY, 0, sub {
		552	my $fh = shift
		553	or return $grp->result (-1);
		554
		555	aioreq_pri $pri;
		556	add $grp aio_read $fh, 0, (-s $fh), $$data, 0, sub {
		557	$grp->result ($_[0]);
		558	};
		559	};
		560
		561	$grp
		562	}
		563
		564	=item aio_copy $srcpath, $dstpath, $callback->($status)
		565
		566	Try to copy the I<file> (directories not supported as either source or
		567	destination) from C<$srcpath> to C<$dstpath> and call the callback with
		568	the C<0> (error) or C<-1> ok.
		569
		570	This is a composite request that it creates the destination file with
		571	mode 0200 and copies the contents of the source file into it using
		572	C<aio_sendfile>, followed by restoring atime, mtime, access mode and
		573	uid/gid, in that order.
		574
		575	If an error occurs, the partial destination file will be unlinked, if
		576	possible, except when setting atime, mtime, access mode and uid/gid, where
		577	errors are being ignored.
		578
		579	=cut
		580
		581	sub aio_copy($$;$) {
		582	my ($src, $dst, $cb) = @_;
		583
		584	my $pri = aioreq_pri;
		585	my $grp = aio_group $cb;
		586
		587	aioreq_pri $pri;
		588	add $grp aio_open $src, O_RDONLY, 0, sub {
		589	if (my $src_fh = $_[0]) {
		590	my @stat = stat $src_fh;
		591
		592	aioreq_pri $pri;
		593	add $grp aio_open $dst, O_CREAT | O_WRONLY | O_TRUNC, 0200, sub {
		594	if (my $dst_fh = $_[0]) {
		595	aioreq_pri $pri;
		596	add $grp aio_sendfile $dst_fh, $src_fh, 0, $stat[7], sub {
		597	if ($_[0] == $stat[7]) {
		598	$grp->result (0);
		599	close $src_fh;
		600
		601	# those should not normally block. should. should.
		602	utime $stat[8], $stat[9], $dst;
		603	chmod $stat[2] & 07777, $dst_fh;
		604	chown $stat[4], $stat[5], $dst_fh;
		605
		606	aioreq_pri $pri;
		607	add $grp aio_close $dst_fh;
		608	} else {
		609	$grp->result (-1);
		610	close $src_fh;
		611	close $dst_fh;
		612
		613	aioreq $pri;
		614	add $grp aio_unlink $dst;
		615	}
		616	};
		617	} else {
		618	$grp->result (-1);
		619	}
		620	},
		621
		622	} else {
		623	$grp->result (-1);
		624	}
		625	};
		626
		627	$grp
		628	}
		629
		630	=item aio_move $srcpath, $dstpath, $callback->($status)
		631
		632	Try to move the I<file> (directories not supported as either source or
		633	destination) from C<$srcpath> to C<$dstpath> and call the callback with
		634	the C<0> (error) or C<-1> ok.
		635
		636	This is a composite request that tries to rename(2) the file first. If
		637	rename files with C<EXDEV>, it copies the file with C<aio_copy> and, if
		638	that is successful, unlinking the C<$srcpath>.
		639
		640	=cut
		641
		642	sub aio_move($$;$) {
		643	my ($src, $dst, $cb) = @_;
		644
		645	my $pri = aioreq_pri;
		646	my $grp = aio_group $cb;
		647
		648	aioreq_pri $pri;
		649	add $grp aio_rename $src, $dst, sub {
		650	if ($_[0] && $! == EXDEV) {
		651	aioreq_pri $pri;
		652	add $grp aio_copy $src, $dst, sub {
		653	$grp->result ($_[0]);
		654
		655	if (!$_[0]) {
		656	aioreq_pri $pri;
		657	add $grp aio_unlink $src;
		658	}
		659	};
		660	} else {
		661	$grp->result ($_[0]);
		662	}
		663	};
		664
		665	$grp
		666	}
		667
346	=item aio_scandir $path, $maxreq, $callback->($dirs, $nondirs)	668	=item aio_scandir $path, $maxreq, $callback->($dirs, $nondirs)
347		669
348	Scans a directory (similar to C<aio_readdir>) but additionally tries to	670	Scans a directory (similar to C<aio_readdir>) but additionally tries to
349	separate the entries of directory C<$path> into two sets of names, ones	671	efficiently separate the entries of directory C<$path> into two sets of
350	you can recurse into (directories or links to them), and ones you cannot	672	names, directories you can recurse into (directories), and ones you cannot
351	recurse into (everything else).	673	recurse into (everything else, including symlinks to directories).
352		674
353	C<aio_scandir> is a composite request that consists of many sub	675	C<aio_scandir> is a composite request that creates of many sub requests_
354	requests. C<$maxreq> specifies the maximum number of outstanding aio	676	C<$maxreq> specifies the maximum number of outstanding aio requests that
355	requests that this function generates. If it is C<< <= 0 >>, then a	677	this function generates. If it is C<< <= 0 >>, then a suitable default
356	suitable default will be chosen (currently 8).	678	will be chosen (currently 4).
357		679
358	On error, the callback is called without arguments, otherwise it receives	680	On error, the callback is called without arguments, otherwise it receives
359	two array-refs with path-relative entry names.	681	two array-refs with path-relative entry names.
360		682
361	Example:	683	Example:
…		…
395	as those tend to return 0 or 1 as link counts, which disables the	717	as those tend to return 0 or 1 as link counts, which disables the
396	directory counting heuristic.	718	directory counting heuristic.
397		719
398	=cut	720	=cut
399		721
400	sub aio_scandir($$$) {	722	sub aio_scandir($$;$) {
401	my ($path, $maxreq, $cb) = @_;	723	my ($path, $maxreq, $cb) = @_;
402		724
		725	my $pri = aioreq_pri;
		726
403	my $grp = aio_group;	727	my $grp = aio_group $cb;
404		728
405	$maxreq = 8 if $maxreq <= 0;	729	$maxreq = 4 if $maxreq <= 0;
406		730
407	# stat once	731	# stat once
		732	aioreq_pri $pri;
408	add $grp aio_stat $path, sub {	733	add $grp aio_stat $path, sub {
409	return $cb->() if $_[0];	734	return $grp->result () if $_[0];
410	my $now = time;	735	my $now = time;
411	my $hash1 = join ":", (stat _)[0,1,3,7,9];	736	my $hash1 = join ":", (stat _)[0,1,3,7,9];
412		737
413	# read the directory entries	738	# read the directory entries
		739	aioreq_pri $pri;
414	add $grp aio_readdir $path, sub {	740	add $grp aio_readdir $path, sub {
415	my $entries = shift	741	my $entries = shift
416	or return $cb->();	742	or return $grp->result ();
417		743
418	# stat the dir another time	744	# stat the dir another time
		745	aioreq_pri $pri;
419	add $grp aio_stat $path, sub {	746	add $grp aio_stat $path, sub {
420	my $hash2 = join ":", (stat _)[0,1,3,7,9];	747	my $hash2 = join ":", (stat _)[0,1,3,7,9];
421		748
422	my $ndirs;	749	my $ndirs;
423		750
…		…
426	$ndirs = -1;	753	$ndirs = -1;
427	} else {	754	} else {
428	# if nlink == 2, we are finished	755	# if nlink == 2, we are finished
429	# on non-posix-fs's, we rely on nlink < 2	756	# on non-posix-fs's, we rely on nlink < 2
430	$ndirs = (stat _)[3] - 2	757	$ndirs = (stat _)[3] - 2
431	or return $cb->([], $entries);	758	or return $grp->result ([], $entries);
432	}	759	}
433		760
434	# sort into likely dirs and likely nondirs	761	# sort into likely dirs and likely nondirs
435	# dirs == files without ".", short entries first	762	# dirs == files without ".", short entries first
436	$entries = [map $_->[0],	763	$entries = [map $_->[0],
…		…
438	map [$_, sprintf "%s%04d", (/.\./ ? "1" : "0"), length],	765	map [$_, sprintf "%s%04d", (/.\./ ? "1" : "0"), length],
439	@$entries];	766	@$entries];
440		767
441	my (@dirs, @nondirs);	768	my (@dirs, @nondirs);
442		769
443	my ($statcb, $schedcb);	770	my $statgrp = add $grp aio_group sub {
444	my $nreq = 0;	771	$grp->result (\@dirs, \@nondirs);
		772	};
445		773
446	$schedcb = sub {	774	limit $statgrp $maxreq;
447	if (@$entries) {	775	feed $statgrp sub {
448	if ($nreq < $maxreq) {	776	return unless @$entries;
449	my $ent = pop @$entries;	777	my $entry = pop @$entries;
		778
		779	aioreq_pri $pri;
		780	add $statgrp aio_stat "$path/$entry/.", sub {
		781	if ($_[0] < 0) {
		782	push @nondirs, $entry;
		783	} else {
		784	# need to check for real directory
		785	aioreq_pri $pri;
		786	add $statgrp aio_lstat "$path/$entry", sub {
		787	if (-d _) {
		788	push @dirs, $entry;
		789
		790	unless (--$ndirs) {
		791	push @nondirs, @$entries;
		792	feed $statgrp;
		793	}
		794	} else {
		795	push @nondirs, $entry;
		796	}
450	$nreq++;	797	}
451	add $grp aio_stat "$path/$ent/.", sub { $statcb->($_[0], $ent) };
452	}	798	}
453	} elsif (!$nreq) {
454	# finished
455	undef $statcb;
456	undef $schedcb;
457	$cb->(\@dirs, \@nondirs) if $cb;
458	undef $cb;
459	}	799	};
460	};	800	};
461	$statcb = sub {
462	my ($status, $entry) = @_;
463
464	if ($status < 0) {
465	$nreq--;
466	push @nondirs, $entry;
467	&$schedcb;
468	} else {
469	# need to check for real directory
470	add $grp aio_lstat "$path/$entry", sub {
471	$nreq--;
472
473	if (-d _) {
474	push @dirs, $entry;
475
476	if (!--$ndirs) {
477	push @nondirs, @$entries;
478	$entries = [];
479	}
480	} else {
481	push @nondirs, $entry;
482	}
483
484	&$schedcb;
485	}
486	}
487	};
488
489	&$schedcb while @$entries && $nreq < $maxreq;
490	};	801	};
491	};	802	};
492	};	803	};
493		804
494	$grp	805	$grp
495	}	806	}
496		807
		808	=item aio_rmtree $path, $callback->($status)
		809
		810	Delete a directory tree starting (and including) C<$path>, return the
		811	status of the final C<rmdir> only. This is a composite request that
		812	uses C<aio_scandir> to recurse into and rmdir directories, and unlink
		813	everything else.
		814
		815	=cut
		816
		817	sub aio_rmtree;
		818	sub aio_rmtree($;$) {
		819	my ($path, $cb) = @_;
		820
		821	my $pri = aioreq_pri;
		822	my $grp = aio_group $cb;
		823
		824	aioreq_pri $pri;
		825	add $grp aio_scandir $path, 0, sub {
		826	my ($dirs, $nondirs) = @_;
		827
		828	my $dirgrp = aio_group sub {
		829	add $grp aio_rmdir $path, sub {
		830	$grp->result ($_[0]);
		831	};
		832	};
		833
		834	(aioreq_pri $pri), add $dirgrp aio_rmtree "$path/$_" for @$dirs;
		835	(aioreq_pri $pri), add $dirgrp aio_unlink "$path/$_" for @$nondirs;
		836
		837	add $grp $dirgrp;
		838	};
		839
		840	$grp
		841	}
		842
		843	=item aio_sync $callback->($status)
		844
		845	Asynchronously call sync and call the callback when finished.
		846
497	=item aio_fsync $fh, $callback->($status)	847	=item aio_fsync $fh, $callback->($status)
498		848
499	Asynchronously call fsync on the given filehandle and call the callback	849	Asynchronously call fsync on the given filehandle and call the callback
500	with the fsync result code.	850	with the fsync result code.
501		851
…		…
505	callback with the fdatasync result code.	855	callback with the fdatasync result code.
506		856
507	If this call isn't available because your OS lacks it or it couldn't be	857	If this call isn't available because your OS lacks it or it couldn't be
508	detected, it will be emulated by calling C<fsync> instead.	858	detected, it will be emulated by calling C<fsync> instead.
509		859
		860	=item aio_pathsync $path, $callback->($status)
		861
		862	This request tries to open, fsync and close the given path. This is a
		863	composite request intended tosync directories after directory operations
		864	(E.g. rename). This might not work on all operating systems or have any
		865	specific effect, but usually it makes sure that directory changes get
		866	written to disc. It works for anything that can be opened for read-only,
		867	not just directories.
		868
		869	Passes C<0> when everything went ok, and C<-1> on error.
		870
		871	=cut
		872
		873	sub aio_pathsync($;$) {
		874	my ($path, $cb) = @_;
		875
		876	my $pri = aioreq_pri;
		877	my $grp = aio_group $cb;
		878
		879	aioreq_pri $pri;
		880	add $grp aio_open $path, O_RDONLY, 0, sub {
		881	my ($fh) = @_;
		882	if ($fh) {
		883	aioreq_pri $pri;
		884	add $grp aio_fsync $fh, sub {
		885	$grp->result ($_[0]);
		886
		887	aioreq_pri $pri;
		888	add $grp aio_close $fh;
		889	};
		890	} else {
		891	$grp->result (-1);
		892	}
		893	};
		894
		895	$grp
		896	}
		897
510	=item aio_group $callback->()	898	=item aio_group $callback->(...)
511
512	[EXPERIMENTAL]
513		899
514	This is a very special aio request: Instead of doing something, it is a	900	This is a very special aio request: Instead of doing something, it is a
515	container for other aio requests, which is useful if you want to bundle	901	container for other aio requests, which is useful if you want to bundle
516	many requests into a single, composite, request.	902	many requests into a single, composite, request with a definite callback
		903	and the ability to cancel the whole request with its subrequests.
517		904
518	Returns an object of class L<IO::AIO::GRP>. See its documentation below	905	Returns an object of class L<IO::AIO::GRP>. See its documentation below
519	for more info.	906	for more info.
520		907
521	Example:	908	Example:
…		…
527	add $grp	914	add $grp
528	(aio_stat ...),	915	(aio_stat ...),
529	(aio_stat ...),	916	(aio_stat ...),
530	...;	917	...;
531		918
		919	=item aio_nop $callback->()
		920
		921	This is a special request - it does nothing in itself and is only used for
		922	side effects, such as when you want to add a dummy request to a group so
		923	that finishing the requests in the group depends on executing the given
		924	code.
		925
		926	While this request does nothing, it still goes through the execution
		927	phase and still requires a worker thread. Thus, the callback will not
		928	be executed immediately but only after other requests in the queue have
		929	entered their execution phase. This can be used to measure request
		930	latency.
		931
532	=item IO::AIO::aio_sleep $fractional_seconds, $callback->() *NOT EXPORTED*	932	=item IO::AIO::aio_busy $fractional_seconds, $callback->() *NOT EXPORTED*
533		933
534	Mainly used for debugging and benchmarking, this aio request puts one of	934	Mainly used for debugging and benchmarking, this aio request puts one of
535	the request workers to sleep for the given time.	935	the request workers to sleep for the given time.
536		936
537	While it is theoretically handy to have simple I/O scheduling requests	937	While it is theoretically handy to have simple I/O scheduling requests
538	like sleep and file handle readable/writable, the overhead this creates	938	like sleep and file handle readable/writable, the overhead this creates is
539	is immense, so do not use this function except to put your application	939	immense (it blocks a thread for a long time) so do not use this function
540	under artificial I/O pressure.	940	except to put your application under artificial I/O pressure.
541		941
542	=back	942	=back
543		943
544	=head2 IO::AIO::REQ CLASS	944	=head2 IO::AIO::REQ CLASS
545		945
546	All non-aggregate C<aio_*> functions return an object of this class when	946	All non-aggregate C<aio_*> functions return an object of this class when
547	called in non-void context.	947	called in non-void context.
548		948
549	A request always moves through the following five states in its lifetime,
550	in order: B<ready> (request has been created, but has not been executed
551	yet), B<execute> (request is currently being executed), B<pending>
552	(request has been executed but callback has not been called yet),
553	B<result> (results are being processed synchronously, includes calling the
554	callback) and B<done> (request has reached the end of its lifetime and
555	holds no resources anymore).
556
557	=over 4	949	=over 4
558		950
559	=item $req->cancel	951	=item cancel $req
560		952
561	Cancels the request, if possible. Has the effect of skipping execution	953	Cancels the request, if possible. Has the effect of skipping execution
562	when entering the B<execute> state and skipping calling the callback when	954	when entering the B<execute> state and skipping calling the callback when
563	entering the the B<result> state, but will leave the request otherwise	955	entering the the B<result> state, but will leave the request otherwise
564	untouched. That means that requests that currently execute will not be	956	untouched. That means that requests that currently execute will not be
565	stopped and resources held by the request will not be freed prematurely.	957	stopped and resources held by the request will not be freed prematurely.
566		958
		959	=item cb $req $callback->(...)
		960
		961	Replace (or simply set) the callback registered to the request.
		962
567	=back	963	=back
568		964
569	=head2 IO::AIO::GRP CLASS	965	=head2 IO::AIO::GRP CLASS
570		966
571	This class is a subclass of L<IO::AIO::REQ>, so all its methods apply to	967	This class is a subclass of L<IO::AIO::REQ>, so all its methods apply to
…		…
585	You add requests by calling the C<add> method with one or more	981	You add requests by calling the C<add> method with one or more
586	C<IO::AIO::REQ> objects:	982	C<IO::AIO::REQ> objects:
587		983
588	$grp->add (aio_unlink "...");	984	$grp->add (aio_unlink "...");
589		985
590	add $grp aio_stat "...", sub { ... };	986	add $grp aio_stat "...", sub {
		987	$_[0] or return $grp->result ("error");
		988
		989	# add another request dynamically, if first succeeded
		990	add $grp aio_open "...", sub {
		991	$grp->result ("ok");
		992	};
		993	};
591		994
592	This makes it very easy to create composite requests (see the source of	995	This makes it very easy to create composite requests (see the source of
593	C<aio_move> for an application) that work and feel like simple requests.	996	C<aio_move> for an application) that work and feel like simple requests.
594		997
		998	=over 4
		999
595	The IO::AIO::GRP objects will be cleaned up during calls to	1000	=item * The IO::AIO::GRP objects will be cleaned up during calls to
596	C<IO::AIO::poll_cb>, just like any other request.	1001	C<IO::AIO::poll_cb>, just like any other request.
597		1002
598	They can be canceled like any other request. Canceling will cancel not	1003	=item * They can be canceled like any other request. Canceling will cancel not
599	just the request itself, but also all requests it contains.	1004	only the request itself, but also all requests it contains.
600		1005
601	They can also can also be added to other IO::AIO::GRP objects.	1006	=item * They can also can also be added to other IO::AIO::GRP objects.
		1007
		1008	=item * You must not add requests to a group from within the group callback (or
		1009	any later time).
		1010
		1011	=back
602		1012
603	Their lifetime, simplified, looks like this: when they are empty, they	1013	Their lifetime, simplified, looks like this: when they are empty, they
604	will finish very quickly. If they contain only requests that are in the	1014	will finish very quickly. If they contain only requests that are in the
605	C<done> state, they will also finish. Otherwise they will continue to	1015	C<done> state, they will also finish. Otherwise they will continue to
606	exist.	1016	exist.
…		…
610	group. And only when all those requests have finished will the the group	1020	group. And only when all those requests have finished will the the group
611	itself finish.	1021	itself finish.
612		1022
613	=over 4	1023	=over 4
614		1024
		1025	=item add $grp ...
		1026
615	=item $grp->add (...)	1027	=item $grp->add (...)
616
617	=item add $grp ...
618		1028
619	Add one or more requests to the group. Any type of L<IO::AIO::REQ> can	1029	Add one or more requests to the group. Any type of L<IO::AIO::REQ> can
620	be added, including other groups, as long as you do not create circular	1030	be added, including other groups, as long as you do not create circular
621	dependencies.	1031	dependencies.
622		1032
623	Returns all its arguments.	1033	Returns all its arguments.
624		1034
		1035	=item $grp->cancel_subs
		1036
		1037	Cancel all subrequests and clears any feeder, but not the group request
		1038	itself. Useful when you queued a lot of events but got a result early.
		1039
		1040	=item $grp->result (...)
		1041
		1042	Set the result value(s) that will be passed to the group callback when all
		1043	subrequests have finished and set the groups errno to the current value
		1044	of errno (just like calling C<errno> without an error number). By default,
		1045	no argument will be passed and errno is zero.
		1046
		1047	=item $grp->errno ([$errno])
		1048
		1049	Sets the group errno value to C<$errno>, or the current value of errno
		1050	when the argument is missing.
		1051
		1052	Every aio request has an associated errno value that is restored when
		1053	the callback is invoked. This method lets you change this value from its
		1054	default (0).
		1055
		1056	Calling C<result> will also set errno, so make sure you either set C<$!>
		1057	before the call to C<result>, or call c<errno> after it.
		1058
		1059	=item feed $grp $callback->($grp)
		1060
		1061	Sets a feeder/generator on this group: every group can have an attached
		1062	generator that generates requests if idle. The idea behind this is that,
		1063	although you could just queue as many requests as you want in a group,
		1064	this might starve other requests for a potentially long time. For
		1065	example, C<aio_scandir> might generate hundreds of thousands C<aio_stat>
		1066	requests, delaying any later requests for a long time.
		1067
		1068	To avoid this, and allow incremental generation of requests, you can
		1069	instead a group and set a feeder on it that generates those requests. The
		1070	feed callback will be called whenever there are few enough (see C<limit>,
		1071	below) requests active in the group itself and is expected to queue more
		1072	requests.
		1073
		1074	The feed callback can queue as many requests as it likes (i.e. C<add> does
		1075	not impose any limits).
		1076
		1077	If the feed does not queue more requests when called, it will be
		1078	automatically removed from the group.
		1079
		1080	If the feed limit is C<0>, it will be set to C<2> automatically.
		1081
		1082	Example:
		1083
		1084	# stat all files in @files, but only ever use four aio requests concurrently:
		1085
		1086	my $grp = aio_group sub { print "finished\n" };
		1087	limit $grp 4;
		1088	feed $grp sub {
		1089	my $file = pop @files
		1090	or return;
		1091
		1092	add $grp aio_stat $file, sub { ... };
		1093	};
		1094
		1095	=item limit $grp $num
		1096
		1097	Sets the feeder limit for the group: The feeder will be called whenever
		1098	the group contains less than this many requests.
		1099
		1100	Setting the limit to C<0> will pause the feeding process.
		1101
625	=back	1102	=back
626		1103
627	=head2 SUPPORT FUNCTIONS	1104	=head2 SUPPORT FUNCTIONS
		1105
		1106	=head3 EVENT PROCESSING AND EVENT LOOP INTEGRATION
628		1107
629	=over 4	1108	=over 4
630		1109
631	=item $fileno = IO::AIO::poll_fileno	1110	=item $fileno = IO::AIO::poll_fileno
632		1111
…		…
637		1116
638	See C<poll_cb> for an example.	1117	See C<poll_cb> for an example.
639		1118
640	=item IO::AIO::poll_cb	1119	=item IO::AIO::poll_cb
641		1120
642	Process all outstanding events on the result pipe. You have to call this	1121	Process some outstanding events on the result pipe. You have to call this
643	regularly. Returns the number of events processed. Returns immediately	1122	regularly. Returns C<0> if all events could be processed, or C<-1> if it
644	when no events are outstanding.	1123	returned earlier for whatever reason. Returns immediately when no events
		1124	are outstanding. The amount of events processed depends on the settings of
		1125	C<IO::AIO::max_poll_req> and C<IO::AIO::max_poll_time>.
		1126
		1127	If not all requests were processed for whatever reason, the filehandle
		1128	will still be ready when C<poll_cb> returns, so normally you don't have to
		1129	do anything special to have it called later.
645		1130
646	Example: Install an Event watcher that automatically calls	1131	Example: Install an Event watcher that automatically calls
647	IO::AIO::poll_cb with high priority:	1132	IO::AIO::poll_cb with high priority:
648		1133
649	Event->io (fd => IO::AIO::poll_fileno,	1134	Event->io (fd => IO::AIO::poll_fileno,
650	poll => 'r', async => 1,	1135	poll => 'r', async => 1,
651	cb => \&IO::AIO::poll_cb);	1136	cb => \&IO::AIO::poll_cb);
652		1137
		1138	=item IO::AIO::max_poll_reqs $nreqs
		1139
		1140	=item IO::AIO::max_poll_time $seconds
		1141
		1142	These set the maximum number of requests (default C<0>, meaning infinity)
		1143	that are being processed by C<IO::AIO::poll_cb> in one call, respectively
		1144	the maximum amount of time (default C<0>, meaning infinity) spent in
		1145	C<IO::AIO::poll_cb> to process requests (more correctly the mininum amount
		1146	of time C<poll_cb> is allowed to use).
		1147
		1148	Setting C<max_poll_time> to a non-zero value creates an overhead of one
		1149	syscall per request processed, which is not normally a problem unless your
		1150	callbacks are really really fast or your OS is really really slow (I am
		1151	not mentioning Solaris here). Using C<max_poll_reqs> incurs no overhead.
		1152
		1153	Setting these is useful if you want to ensure some level of
		1154	interactiveness when perl is not fast enough to process all requests in
		1155	time.
		1156
		1157	For interactive programs, values such as C<0.01> to C<0.1> should be fine.
		1158
		1159	Example: Install an Event watcher that automatically calls
		1160	IO::AIO::poll_cb with low priority, to ensure that other parts of the
		1161	program get the CPU sometimes even under high AIO load.
		1162
		1163	# try not to spend much more than 0.1s in poll_cb
		1164	IO::AIO::max_poll_time 0.1;
		1165
		1166	# use a low priority so other tasks have priority
		1167	Event->io (fd => IO::AIO::poll_fileno,
		1168	poll => 'r', nice => 1,
		1169	cb => &IO::AIO::poll_cb);
		1170
653	=item IO::AIO::poll_wait	1171	=item IO::AIO::poll_wait
654		1172
		1173	If there are any outstanding requests and none of them in the result
655	Wait till the result filehandle becomes ready for reading (simply does a	1174	phase, wait till the result filehandle becomes ready for reading (simply
656	C<select> on the filehandle. This is useful if you want to synchronously wait	1175	does a C<select> on the filehandle. This is useful if you want to
657	for some requests to finish).	1176	synchronously wait for some requests to finish).
658		1177
659	See C<nreqs> for an example.	1178	See C<nreqs> for an example.
660		1179
		1180	=item IO::AIO::poll
		1181
		1182	Waits until some requests have been handled.
		1183
		1184	Returns the number of requests processed, but is otherwise strictly
		1185	equivalent to:
		1186
		1187	IO::AIO::poll_wait, IO::AIO::poll_cb
		1188
661	=item IO::AIO::nreqs	1189	=item IO::AIO::flush
662		1190
663	Returns the number of requests currently outstanding (i.e. for which their	1191	Wait till all outstanding AIO requests have been handled.
664	callback has not been invoked yet).
665		1192
666	Example: wait till there are no outstanding requests anymore:	1193	Strictly equivalent to:
667		1194
668	IO::AIO::poll_wait, IO::AIO::poll_cb	1195	IO::AIO::poll_wait, IO::AIO::poll_cb
669	while IO::AIO::nreqs;	1196	while IO::AIO::nreqs;
670		1197
671	=item IO::AIO::flush	1198	=back
672		1199
673	Wait till all outstanding AIO requests have been handled.	1200	=head3 CONTROLLING THE NUMBER OF THREADS
674		1201
675	Strictly equivalent to:	1202	=over
676
677	IO::AIO::poll_wait, IO::AIO::poll_cb
678	while IO::AIO::nreqs;
679
680	=item IO::AIO::poll
681
682	Waits until some requests have been handled.
683
684	Strictly equivalent to:
685
686	IO::AIO::poll_wait, IO::AIO::poll_cb
687	if IO::AIO::nreqs;
688		1203
689	=item IO::AIO::min_parallel $nthreads	1204	=item IO::AIO::min_parallel $nthreads
690		1205
691	Set the minimum number of AIO threads to C<$nthreads>. The current default	1206	Set the minimum number of AIO threads to C<$nthreads>. The current
692	is C<4>, which means four asynchronous operations can be done at one time	1207	default is C<8>, which means eight asynchronous operations can execute
693	(the number of outstanding operations, however, is unlimited).	1208	concurrently at any one time (the number of outstanding requests,
		1209	however, is unlimited).
694		1210
695	IO::AIO starts threads only on demand, when an AIO request is queued and	1211	IO::AIO starts threads only on demand, when an AIO request is queued and
696	no free thread exists.	1212	no free thread exists. Please note that queueing up a hundred requests can
		1213	create demand for a hundred threads, even if it turns out that everything
		1214	is in the cache and could have been processed faster by a single thread.
697		1215
698	It is recommended to keep the number of threads low, as some Linux	1216	It is recommended to keep the number of threads relatively low, as some
699	kernel versions will scale negatively with the number of threads (higher	1217	Linux kernel versions will scale negatively with the number of threads
700	parallelity => MUCH higher latency). With current Linux 2.6 versions, 4-32	1218	(higher parallelity => MUCH higher latency). With current Linux 2.6
701	threads should be fine.	1219	versions, 4-32 threads should be fine.
702		1220
703	Under most circumstances you don't need to call this function, as the	1221	Under most circumstances you don't need to call this function, as the
704	module selects a default that is suitable for low to moderate load.	1222	module selects a default that is suitable for low to moderate load.
705		1223
706	=item IO::AIO::max_parallel $nthreads	1224	=item IO::AIO::max_parallel $nthreads
…		…
715	This module automatically runs C<max_parallel 0> at program end, to ensure	1233	This module automatically runs C<max_parallel 0> at program end, to ensure
716	that all threads are killed and that there are no outstanding requests.	1234	that all threads are killed and that there are no outstanding requests.
717		1235
718	Under normal circumstances you don't need to call this function.	1236	Under normal circumstances you don't need to call this function.
719		1237
		1238	=item IO::AIO::max_idle $nthreads
		1239
		1240	Limit the number of threads (default: 4) that are allowed to idle (i.e.,
		1241	threads that did not get a request to process within 10 seconds). That
		1242	means if a thread becomes idle while C<$nthreads> other threads are also
		1243	idle, it will free its resources and exit.
		1244
		1245	This is useful when you allow a large number of threads (e.g. 100 or 1000)
		1246	to allow for extremely high load situations, but want to free resources
		1247	under normal circumstances (1000 threads can easily consume 30MB of RAM).
		1248
		1249	The default is probably ok in most situations, especially if thread
		1250	creation is fast. If thread creation is very slow on your system you might
		1251	want to use larger values.
		1252
720	=item $oldnreqs = IO::AIO::max_outstanding $nreqs	1253	=item IO::AIO::max_outstanding $maxreqs
		1254
		1255	This is a very bad function to use in interactive programs because it
		1256	blocks, and a bad way to reduce concurrency because it is inexact: Better
		1257	use an C<aio_group> together with a feed callback.
721		1258
722	Sets the maximum number of outstanding requests to C<$nreqs>. If you	1259	Sets the maximum number of outstanding requests to C<$nreqs>. If you
723	try to queue up more than this number of requests, the caller will block until	1260	do queue up more than this number of requests, the next call to the
724	some requests have been handled.	1261	C<poll_cb> (and C<poll_some> and other functions calling C<poll_cb>)
		1262	function will block until the limit is no longer exceeded.
725		1263
726	The default is very large, so normally there is no practical limit. If you	1264	The default value is very large, so there is no practical limit on the
727	queue up many requests in a loop it often improves speed if you set	1265	number of outstanding requests.
728	this to a relatively low number, such as C<100>.
729		1266
730	Under normal circumstances you don't need to call this function.	1267	You can still queue as many requests as you want. Therefore,
		1268	C<max_outstanding> is mainly useful in simple scripts (with low values) or
		1269	as a stop gap to shield against fatal memory overflow (with large values).
731		1270
732	=back	1271	=back
733		1272
		1273	=head3 STATISTICAL INFORMATION
		1274
		1275	=over
		1276
		1277	=item IO::AIO::nreqs
		1278
		1279	Returns the number of requests currently in the ready, execute or pending
		1280	states (i.e. for which their callback has not been invoked yet).
		1281
		1282	Example: wait till there are no outstanding requests anymore:
		1283
		1284	IO::AIO::poll_wait, IO::AIO::poll_cb
		1285	while IO::AIO::nreqs;
		1286
		1287	=item IO::AIO::nready
		1288
		1289	Returns the number of requests currently in the ready state (not yet
		1290	executed).
		1291
		1292	=item IO::AIO::npending
		1293
		1294	Returns the number of requests currently in the pending state (executed,
		1295	but not yet processed by poll_cb).
		1296
		1297	=back
		1298
734	=cut	1299	=cut
735		1300
736	# support function to convert a fd into a perl filehandle
737	sub _fd2fh {
738	return undef if $_[0] < 0;
739
740	# try to generate nice filehandles
741	my $sym = "IO::AIO::fd#$_[0]";
742	local *$sym;
743
744	open *$sym, "+<&=$_[0]" # usually works under any unix
745	or open *$sym, "<&=$_[0]" # cygwin needs this
746	or open *$sym, ">&=$_[0]" # or this
747	or return undef;
748
749	*$sym
750	}
751
752	min_parallel 4;	1301	min_parallel 8;
753		1302
754	END {	1303	END { flush }
755	max_parallel 0;
756	}
757		1304
758	1;	1305	1;
759		1306
760	=head2 FORK BEHAVIOUR	1307	=head2 FORK BEHAVIOUR
761		1308
762	This module should do "the right thing" when the process using it forks:	1309	This module should do "the right thing" when the process using it forks:
763		1310
764	Before the fork, IO::AIO enters a quiescent state where no requests	1311	Before the fork, IO::AIO enters a quiescent state where no requests
765	can be added in other threads and no results will be processed. After	1312	can be added in other threads and no results will be processed. After
766	the fork the parent simply leaves the quiescent state and continues	1313	the fork the parent simply leaves the quiescent state and continues
767	request/result processing, while the child clears the request/result	1314	request/result processing, while the child frees the request/result queue
768	queue (so the requests started before the fork will only be handled in	1315	(so that the requests started before the fork will only be handled in the
769	the parent). Threads will be started on demand until the limit ste in the	1316	parent). Threads will be started on demand until the limit set in the
770	parent process has been reached again.	1317	parent process has been reached again.
771		1318
772	In short: the parent will, after a short pause, continue as if fork had	1319	In short: the parent will, after a short pause, continue as if fork had
773	not been called, while the child will act as if IO::AIO has not been used	1320	not been called, while the child will act as if IO::AIO has not been used
774	yet.	1321	yet.
775		1322
		1323	=head2 MEMORY USAGE
		1324
		1325	Per-request usage:
		1326
		1327	Each aio request uses - depending on your architecture - around 100-200
		1328	bytes of memory. In addition, stat requests need a stat buffer (possibly
		1329	a few hundred bytes), readdir requires a result buffer and so on. Perl
		1330	scalars and other data passed into aio requests will also be locked and
		1331	will consume memory till the request has entered the done state.
		1332
		1333	This is not awfully much, so queuing lots of requests is not usually a
		1334	problem.
		1335
		1336	Per-thread usage:
		1337
		1338	In the execution phase, some aio requests require more memory for
		1339	temporary buffers, and each thread requires a stack and other data
		1340	structures (usually around 16k-128k, depending on the OS).
		1341
		1342	=head1 KNOWN BUGS
		1343
		1344	Known bugs will be fixed in the next release.
		1345
776	=head1 SEE ALSO	1346	=head1 SEE ALSO
777		1347
778	L<Coro>, L<Linux::AIO> (obsolete).	1348	L<AnyEvent::AIO> for easy integration into event loops, L<Coro::AIO> for a
		1349	more natural syntax.
779		1350
780	=head1 AUTHOR	1351	=head1 AUTHOR
781		1352
782	Marc Lehmann <schmorp@schmorp.de>	1353	Marc Lehmann <schmorp@schmorp.de>
783	http://home.schmorp.de/	1354	http://home.schmorp.de/

Diff Legend

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