ViewVC Help

View File | Revision Log | Show Annotations | Download File

/cvs/IO-AIO/AIO.pm

Comparing IO-AIO/AIO.pm (file contents):
Revision 1.14 by root, Mon Jul 11 02:53:59 2005 UTC vs.
Revision 1.172 by root, Thu Jan 7 18:08:21 2010 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
16	aio_read $fh, 30000, 1024, $buffer, 0, sub {	17	aio_read $fh, 30000, 1024, $buffer, 0, sub {
17	$_[0] > 0 or die "read error: $!";	18	$_[0] > 0 or die "read error: $!";
18	};	19	};
19		20
20	# Event	21	# version 2+ has request and group objects
		22	use IO::AIO 2;
		23
		24	aioreq_pri 4; # give next request a very high priority
		25	my $req = aio_unlink "/tmp/file", sub { };
		26	$req->cancel; # cancel request if still in queue
		27
		28	my $grp = aio_group sub { print "all stats done\n" };
		29	add $grp aio_stat "..." for ...;
		30
		31	# AnyEvent integration (EV, Event, Glib, Tk, POE, urxvt, pureperl...)
		32	use AnyEvent::AIO;
		33
		34	# EV integration
		35	my $aio_w = EV::io IO::AIO::poll_fileno, EV::READ, \&IO::AIO::poll_cb;
		36
		37	# Event integration
21	Event->io (fd => IO::AIO::poll_fileno,	38	Event->io (fd => IO::AIO::poll_fileno,
22	poll => 'r',	39	poll => 'r',
23	cb => \&IO::AIO::poll_cb);	40	cb => \&IO::AIO::poll_cb);
24		41
25	# Glib/Gtk2	42	# Glib/Gtk2 integration
26	add_watch Glib::IO IO::AIO::poll_fileno,	43	add_watch Glib::IO IO::AIO::poll_fileno,
27	\&IO::AIO::poll_cb;	44	in => sub { IO::AIO::poll_cb; 1 };
28		45
29	# Tk	46	# Tk integration
30	Tk::Event::IO->fileevent (IO::AIO::poll_fileno, "",	47	Tk::Event::IO->fileevent (IO::AIO::poll_fileno, "",
31	readable => \&IO::AIO::poll_cb);	48	readable => \&IO::AIO::poll_cb);
32		49
33	# Danga::Socket	50	# Danga::Socket integration
34	Danga::Socket->AddOtherFds (IO::AIO::poll_fileno =>	51	Danga::Socket->AddOtherFds (IO::AIO::poll_fileno =>
35	\&IO::AIO::poll_cb);	52	\&IO::AIO::poll_cb);
36		53
37
38	=head1 DESCRIPTION	54	=head1 DESCRIPTION
39		55
40	This module implements asynchronous I/O using whatever means your	56	This module implements asynchronous I/O using whatever means your
41	operating system supports.	57	operating system supports. It is implemented as an interface to C<libeio>
		58	(L<http://software.schmorp.de/pkg/libeio.html>).
42		59
		60	Asynchronous means that operations that can normally block your program
		61	(e.g. reading from disk) will be done asynchronously: the operation
		62	will still block, but you can do something else in the meantime. This
		63	is extremely useful for programs that need to stay interactive even
		64	when doing heavy I/O (GUI programs, high performance network servers
		65	etc.), but can also be used to easily do operations in parallel that are
		66	normally done sequentially, e.g. stat'ing many files, which is much faster
		67	on a RAID volume or over NFS when you do a number of stat operations
		68	concurrently.
		69
		70	While most of this works on all types of file descriptors (for
		71	example sockets), using these functions on file descriptors that
		72	support nonblocking operation (again, sockets, pipes etc.) is
		73	very inefficient. Use an event loop for that (such as the L<EV>
		74	module): IO::AIO will naturally fit into such an event loop itself.
		75
43	Currently, a number of threads are started that execute your read/writes	76	In this version, a number of threads are started that execute your
44	and signal their completion. You don't need thread support in your libc or	77	requests and signal their completion. You don't need thread support
45	perl, and the threads created by this module will not be visible to the	78	in perl, and the threads created by this module will not be visible
46	pthreads library. In the future, this module might make use of the native	79	to perl. In the future, this module might make use of the native aio
47	aio functions available on many operating systems. However, they are often	80	functions available on many operating systems. However, they are often
48	not well-supported (Linux doesn't allow them on normal files currently,	81	not well-supported or restricted (GNU/Linux doesn't allow them on normal
49	for example), and they would only support aio_read and aio_write, so the	82	files currently, for example), and they would only support aio_read and
50	remaining functionality would have to be implemented using threads anyway.	83	aio_write, so the remaining functionality would have to be implemented
		84	using threads anyway.
51		85
52	Although the module will work with in the presence of other threads, it is	86	Although the module will work in the presence of other (Perl-) threads,
53	currently not reentrant, so use appropriate locking yourself.	87	it is currently not reentrant in any way, so use appropriate locking
		88	yourself, always call C<poll_cb> from within the same thread, or never
		89	call C<poll_cb> (or other C<aio_> functions) recursively.
		90
		91	=head2 EXAMPLE
		92
		93	This is a simple example that uses the EV module and loads
		94	F</etc/passwd> asynchronously:
		95
		96	use Fcntl;
		97	use EV;
		98	use IO::AIO;
		99
		100	# register the IO::AIO callback with EV
		101	my $aio_w = EV::io IO::AIO::poll_fileno, EV::READ, \&IO::AIO::poll_cb;
		102
		103	# queue the request to open /etc/passwd
		104	aio_open "/etc/passwd", O_RDONLY, 0, sub {
		105	my $fh = shift
		106	or die "error while opening: $!";
		107
		108	# stat'ing filehandles is generally non-blocking
		109	my $size = -s $fh;
		110
		111	# queue a request to read the file
		112	my $contents;
		113	aio_read $fh, 0, $size, $contents, 0, sub {
		114	$_[0] == $size
		115	or die "short read: $!";
		116
		117	close $fh;
		118
		119	# file contents now in $contents
		120	print $contents;
		121
		122	# exit event loop and program
		123	EV::unloop;
		124	};
		125	};
		126
		127	# possibly queue up other requests, or open GUI windows,
		128	# check for sockets etc. etc.
		129
		130	# process events as long as there are some:
		131	EV::loop;
		132
		133	=head1 REQUEST ANATOMY AND LIFETIME
		134
		135	Every C<aio_*> function creates a request. which is a C data structure not
		136	directly visible to Perl.
		137
		138	If called in non-void context, every request function returns a Perl
		139	object representing the request. In void context, nothing is returned,
		140	which saves a bit of memory.
		141
		142	The perl object is a fairly standard ref-to-hash object. The hash contents
		143	are not used by IO::AIO so you are free to store anything you like in it.
		144
		145	During their existance, aio requests travel through the following states,
		146	in order:
		147
		148	=over 4
		149
		150	=item ready
		151
		152	Immediately after a request is created it is put into the ready state,
		153	waiting for a thread to execute it.
		154
		155	=item execute
		156
		157	A thread has accepted the request for processing and is currently
		158	executing it (e.g. blocking in read).
		159
		160	=item pending
		161
		162	The request has been executed and is waiting for result processing.
		163
		164	While request submission and execution is fully asynchronous, result
		165	processing is not and relies on the perl interpreter calling C<poll_cb>
		166	(or another function with the same effect).
		167
		168	=item result
		169
		170	The request results are processed synchronously by C<poll_cb>.
		171
		172	The C<poll_cb> function will process all outstanding aio requests by
		173	calling their callbacks, freeing memory associated with them and managing
		174	any groups they are contained in.
		175
		176	=item done
		177
		178	Request has reached the end of its lifetime and holds no resources anymore
		179	(except possibly for the Perl object, but its connection to the actual
		180	aio request is severed and calling its methods will either do nothing or
		181	result in a runtime error).
		182
		183	=back
54		184
55	=cut	185	=cut
56		186
57	package IO::AIO;	187	package IO::AIO;
58		188
		189	use Carp ();
		190
		191	use common::sense;
		192
59	use base 'Exporter';	193	use base 'Exporter';
60		194
61	use Fcntl ();
62
63	BEGIN {	195	BEGIN {
64	$VERSION = 0.3;	196	our $VERSION = '3.4';
65		197
66	@EXPORT = qw(aio_read aio_write aio_open aio_close aio_stat aio_lstat aio_unlink	198	our @AIO_REQ = qw(aio_sendfile aio_read aio_write aio_open aio_close
67	aio_fsync aio_fdatasync aio_readahead);	199	aio_stat aio_lstat aio_unlink aio_rmdir aio_readdir aio_readdirx
68	@EXPORT_OK = qw(poll_fileno poll_cb min_parallel max_parallel max_outstanding nreqs);	200	aio_scandir aio_symlink aio_readlink aio_sync aio_fsync
		201	aio_fdatasync aio_sync_file_range aio_pathsync aio_readahead
		202	aio_rename aio_link aio_move aio_copy aio_group
		203	aio_nop aio_mknod aio_load aio_rmtree aio_mkdir aio_chown
		204	aio_chmod aio_utime aio_truncate
		205	aio_msync aio_mtouch aio_statvfs);
		206
		207	our @EXPORT = (@AIO_REQ, qw(aioreq_pri aioreq_nice));
		208	our @EXPORT_OK = qw(poll_fileno poll_cb poll_wait flush
		209	min_parallel max_parallel max_idle
		210	nreqs nready npending nthreads
		211	max_poll_time max_poll_reqs
		212	sendfile fadvise);
		213
		214	push @AIO_REQ, qw(aio_busy); # not exported
		215
		216	@IO::AIO::GRP::ISA = 'IO::AIO::REQ';
69		217
70	require XSLoader;	218	require XSLoader;
71	XSLoader::load IO::AIO, $VERSION;	219	XSLoader::load ("IO::AIO", $VERSION);
72	}	220	}
73		221
74	=head1 FUNCTIONS	222	=head1 FUNCTIONS
75		223
76	=head2 AIO FUNCTIONS	224	=head2 AIO REQUEST FUNCTIONS
77		225
78	All the C<aio_*> calls are more or less thin wrappers around the syscall	226	All the C<aio_*> calls are more or less thin wrappers around the syscall
79	with the same name (sans C<aio_>). The arguments are similar or identical,	227	with the same name (sans C<aio_>). The arguments are similar or identical,
80	and they all accept an additional (and optional) C<$callback> argument	228	and they all accept an additional (and optional) C<$callback> argument
81	which must be a code reference. This code reference will get called with	229	which must be a code reference. This code reference will get called with
82	the syscall return code (e.g. most syscalls return C<-1> on error, unlike	230	the syscall return code (e.g. most syscalls return C<-1> on error, unlike
83	perl, which usually delivers "false") as it's sole argument when the given	231	perl, which usually delivers "false") as its sole argument after the given
84	syscall has been executed asynchronously.	232	syscall has been executed asynchronously.
85		233
86	All functions that expect a filehandle will also accept a file descriptor.	234	All functions expecting a filehandle keep a copy of the filehandle
		235	internally until the request has finished.
87		236
		237	All functions return request objects of type L<IO::AIO::REQ> that allow
		238	further manipulation of those requests while they are in-flight.
		239
88	The filenames you pass to these routines I<must> be absolute. The reason	240	The pathnames you pass to these routines I<must> be absolute and
89	is that at the time the request is being executed, the current working	241	encoded as octets. The reason for the former is that at the time the
90	directory could have changed. Alternatively, you can make sure that you	242	request is being executed, the current working directory could have
91	never change the current working directory.	243	changed. Alternatively, you can make sure that you never change the
		244	current working directory anywhere in the program and then use relative
		245	paths.
		246
		247	To encode pathnames as octets, either make sure you either: a) always pass
		248	in filenames you got from outside (command line, readdir etc.) without
		249	tinkering, b) are ASCII or ISO 8859-1, c) use the Encode module and encode
		250	your pathnames to the locale (or other) encoding in effect in the user
		251	environment, d) use Glib::filename_from_unicode on unicode filenames or e)
		252	use something else to ensure your scalar has the correct contents.
		253
		254	This works, btw. independent of the internal UTF-8 bit, which IO::AIO
		255	handles correctly whether it is set or not.
92		256
93	=over 4	257	=over 4
94		258
		259	=item $prev_pri = aioreq_pri [$pri]
		260
		261	Returns the priority value that would be used for the next request and, if
		262	C<$pri> is given, sets the priority for the next aio request.
		263
		264	The default priority is C<0>, the minimum and maximum priorities are C<-4>
		265	and C<4>, respectively. Requests with higher priority will be serviced
		266	first.
		267
		268	The priority will be reset to C<0> after each call to one of the C<aio_*>
		269	functions.
		270
		271	Example: open a file with low priority, then read something from it with
		272	higher priority so the read request is serviced before other low priority
		273	open requests (potentially spamming the cache):
		274
		275	aioreq_pri -3;
		276	aio_open ..., sub {
		277	return unless $_[0];
		278
		279	aioreq_pri -2;
		280	aio_read $_[0], ..., sub {
		281	...
		282	};
		283	};
		284
		285
		286	=item aioreq_nice $pri_adjust
		287
		288	Similar to C<aioreq_pri>, but subtracts the given value from the current
		289	priority, so the effect is cumulative.
		290
		291
95	=item aio_open $pathname, $flags, $mode, $callback	292	=item aio_open $pathname, $flags, $mode, $callback->($fh)
96		293
97	Asynchronously open or create a file and call the callback with a newly	294	Asynchronously open or create a file and call the callback with a newly
98	created filehandle for the file.	295	created filehandle for the file.
99		296
100	The pathname passed to C<aio_open> must be absolute. See API NOTES, above,	297	The pathname passed to C<aio_open> must be absolute. See API NOTES, above,
101	for an explanation.	298	for an explanation.
102		299
103	The C<$mode> argument is a bitmask. See the C<Fcntl> module for a	300	The C<$flags> argument is a bitmask. See the C<Fcntl> module for a
104	list. They are the same as used in C<sysopen>.	301	list. They are the same as used by C<sysopen>.
		302
		303	Likewise, C<$mode> specifies the mode of the newly created file, if it
		304	didn't exist and C<O_CREAT> has been given, just like perl's C<sysopen>,
		305	except that it is mandatory (i.e. use C<0> if you don't create new files,
		306	and C<0666> or C<0777> if you do). Note that the C<$mode> will be modified
		307	by the umask in effect then the request is being executed, so better never
		308	change the umask.
105		309
106	Example:	310	Example:
107		311
108	aio_open "/etc/passwd", O_RDONLY, 0, sub {	312	aio_open "/etc/passwd", O_RDONLY, 0, sub {
109	if ($_[0]) {	313	if ($_[0]) {
…		…
112	} else {	316	} else {
113	die "open failed: $!\n";	317	die "open failed: $!\n";
114	}	318	}
115	};	319	};
116		320
		321
117	=item aio_close $fh, $callback	322	=item aio_close $fh, $callback->($status)
118		323
119	Asynchronously close a file and call the callback with the result	324	Asynchronously close a file and call the callback with the result
120	code. I<WARNING:> although accepted, you should not pass in a perl	325	code.
121	filehandle here, as perl will likely close the file descriptor itself when
122	the filehandle is destroyed. Normally, you can safely call perls C<close>
123	or just let filehandles go out of scope.
124		326
		327	Unfortunately, you can't do this to perl. Perl I<insists> very strongly on
		328	closing the file descriptor associated with the filehandle itself.
		329
		330	Therefore, C<aio_close> will not close the filehandle - instead it will
		331	use dup2 to overwrite the file descriptor with the write-end of a pipe
		332	(the pipe fd will be created on demand and will be cached).
		333
		334	Or in other words: the file descriptor will be closed, but it will not be
		335	free for reuse until the perl filehandle is closed.
		336
		337	=cut
		338
125	=item aio_read $fh,$offset,$length, $data,$dataoffset,$callback	339	=item aio_read $fh,$offset,$length, $data,$dataoffset, $callback->($retval)
126		340
127	=item aio_write $fh,$offset,$length, $data,$dataoffset,$callback	341	=item aio_write $fh,$offset,$length, $data,$dataoffset, $callback->($retval)
128		342
129	Reads or writes C<length> bytes from the specified C<fh> and C<offset>	343	Reads or writes C<$length> bytes from or to the specified C<$fh> and
130	into the scalar given by C<data> and offset C<dataoffset> and calls the	344	C<$offset> into the scalar given by C<$data> and offset C<$dataoffset>
131	callback without the actual number of bytes read (or -1 on error, just	345	and calls the callback without the actual number of bytes read (or -1 on
132	like the syscall).	346	error, just like the syscall).
133		347
		348	C<aio_read> will, like C<sysread>, shrink or grow the C<$data> scalar to
		349	offset plus the actual number of bytes read.
		350
		351	If C<$offset> is undefined, then the current file descriptor offset will
		352	be used (and updated), otherwise the file descriptor offset will not be
		353	changed by these calls.
		354
		355	If C<$length> is undefined in C<aio_write>, use the remaining length of
		356	C<$data>.
		357
		358	If C<$dataoffset> is less than zero, it will be counted from the end of
		359	C<$data>.
		360
		361	The C<$data> scalar I<MUST NOT> be modified in any way while the request
		362	is outstanding. Modifying it can result in segfaults or World War III (if
		363	the necessary/optional hardware is installed).
		364
134	Example: Read 15 bytes at offset 7 into scalar C<$buffer>, strating at	365	Example: Read 15 bytes at offset 7 into scalar C<$buffer>, starting at
135	offset C<0> within the scalar:	366	offset C<0> within the scalar:
136		367
137	aio_read $fh, 7, 15, $buffer, 0, sub {	368	aio_read $fh, 7, 15, $buffer, 0, sub {
138	$_[0] > 0 or die "read error: $!";	369	$_[0] > 0 or die "read error: $!";
139	print "read $_[0] bytes: <$buffer>\n";	370	print "read $_[0] bytes: <$buffer>\n";
140	};	371	};
141		372
		373
		374	=item aio_sendfile $out_fh, $in_fh, $in_offset, $length, $callback->($retval)
		375
		376	Tries to copy C<$length> bytes from C<$in_fh> to C<$out_fh>. It starts
		377	reading at byte offset C<$in_offset>, and starts writing at the current
		378	file offset of C<$out_fh>. Because of that, it is not safe to issue more
		379	than one C<aio_sendfile> per C<$out_fh>, as they will interfere with each
		380	other.
		381
		382	This call tries to make use of a native C<sendfile> syscall to provide
		383	zero-copy operation. For this to work, C<$out_fh> should refer to a
		384	socket, and C<$in_fh> should refer to mmap'able file.
		385
		386	If a native sendfile cannot be found or it fails with C<ENOSYS>,
		387	C<ENOTSUP>, C<EOPNOTSUPP>, C<EAFNOSUPPORT>, C<EPROTOTYPE> or C<ENOTSOCK>,
		388	it will be emulated, so you can call C<aio_sendfile> on any type of
		389	filehandle regardless of the limitations of the operating system.
		390
		391	Please note, however, that C<aio_sendfile> can read more bytes from
		392	C<$in_fh> than are written, and there is no way to find out how many
		393	bytes have been read from C<aio_sendfile> alone, as C<aio_sendfile> only
		394	provides the number of bytes written to C<$out_fh>. Only if the result
		395	value equals C<$length> one can assume that C<$length> bytes have been
		396	read.
		397
		398
142	=item aio_readahead $fh,$offset,$length, $callback	399	=item aio_readahead $fh,$offset,$length, $callback->($retval)
143		400
144	Asynchronously reads the specified byte range into the page cache, using
145	the C<readahead> syscall. If that syscall doesn't exist the status will be
146	C<-1> and C<$!> is set to ENOSYS.
147
148	readahead() populates the page cache with data from a file so that	401	C<aio_readahead> populates the page cache with data from a file so that
149	subsequent reads from that file will not block on disk I/O. The C<$offset>	402	subsequent reads from that file will not block on disk I/O. The C<$offset>
150	argument specifies the starting point from which data is to be read and	403	argument specifies the starting point from which data is to be read and
151	C<$length> specifies the number of bytes to be read. I/O is performed in	404	C<$length> specifies the number of bytes to be read. I/O is performed in
152	whole pages, so that offset is effectively rounded down to a page boundary	405	whole pages, so that offset is effectively rounded down to a page boundary
153	and bytes are read up to the next page boundary greater than or equal to	406	and bytes are read up to the next page boundary greater than or equal to
154	(off-set+length). aio_readahead() does not read beyond the end of the	407	(off-set+length). C<aio_readahead> does not read beyond the end of the
155	file. The current file offset of the file is left unchanged.	408	file. The current file offset of the file is left unchanged.
156		409
		410	If that syscall doesn't exist (likely if your OS isn't Linux) it will be
		411	emulated by simply reading the data, which would have a similar effect.
		412
		413
157	=item aio_stat $fh_or_path, $callback	414	=item aio_stat $fh_or_path, $callback->($status)
158		415
159	=item aio_lstat $fh, $callback	416	=item aio_lstat $fh, $callback->($status)
160		417
161	Works like perl's C<stat> or C<lstat> in void context. The callback will	418	Works like perl's C<stat> or C<lstat> in void context. The callback will
162	be called after the stat and the results will be available using C<stat _>	419	be called after the stat and the results will be available using C<stat _>
163	or C<-s _> etc...	420	or C<-s _> etc...
164		421
…		…
174	aio_stat "/etc/passwd", sub {	431	aio_stat "/etc/passwd", sub {
175	$_[0] and die "stat failed: $!";	432	$_[0] and die "stat failed: $!";
176	print "size is ", -s _, "\n";	433	print "size is ", -s _, "\n";
177	};	434	};
178		435
		436
		437	=item aio_statvfs $fh_or_path, $callback->($statvfs)
		438
		439	Works like the POSIX C<statvfs> or C<fstatvfs> syscalls, depending on
		440	whether a file handle or path was passed.
		441
		442	On success, the callback is passed a hash reference with the following
		443	members: C<bsize>, C<frsize>, C<blocks>, C<bfree>, C<bavail>, C<files>,
		444	C<ffree>, C<favail>, C<fsid>, C<flag> and C<namemax>. On failure, C<undef>
		445	is passed.
		446
		447	The following POSIX IO::AIO::ST_* constants are defined: C<ST_RDONLY> and
		448	C<ST_NOSUID>.
		449
		450	The following non-POSIX IO::AIO::ST_* flag masks are defined to
		451	their correct value when available, or to C<0> on systems that do
		452	not support them: C<ST_NODEV>, C<ST_NOEXEC>, C<ST_SYNCHRONOUS>,
		453	C<ST_MANDLOCK>, C<ST_WRITE>, C<ST_APPEND>, C<ST_IMMUTABLE>, C<ST_NOATIME>,
		454	C<ST_NODIRATIME> and C<ST_RELATIME>.
		455
		456	Example: stat C</wd> and dump out the data if successful.
		457
		458	aio_statvfs "/wd", sub {
		459	my $f = $_[0]
		460	or die "statvfs: $!";
		461
		462	use Data::Dumper;
		463	say Dumper $f;
		464	};
		465
		466	# result:
		467	{
		468	bsize => 1024,
		469	bfree => 4333064312,
		470	blocks => 10253828096,
		471	files => 2050765568,
		472	flag => 4096,
		473	favail => 2042092649,
		474	bavail => 4333064312,
		475	ffree => 2042092649,
		476	namemax => 255,
		477	frsize => 1024,
		478	fsid => 1810
		479	}
		480
		481
		482	=item aio_utime $fh_or_path, $atime, $mtime, $callback->($status)
		483
		484	Works like perl's C<utime> function (including the special case of $atime
		485	and $mtime being undef). Fractional times are supported if the underlying
		486	syscalls support them.
		487
		488	When called with a pathname, uses utimes(2) if available, otherwise
		489	utime(2). If called on a file descriptor, uses futimes(2) if available,
		490	otherwise returns ENOSYS, so this is not portable.
		491
		492	Examples:
		493
		494	# set atime and mtime to current time (basically touch(1)):
		495	aio_utime "path", undef, undef;
		496	# set atime to current time and mtime to beginning of the epoch:
		497	aio_utime "path", time, undef; # undef==0
		498
		499
		500	=item aio_chown $fh_or_path, $uid, $gid, $callback->($status)
		501
		502	Works like perl's C<chown> function, except that C<undef> for either $uid
		503	or $gid is being interpreted as "do not change" (but -1 can also be used).
		504
		505	Examples:
		506
		507	# same as "chown root path" in the shell:
		508	aio_chown "path", 0, -1;
		509	# same as above:
		510	aio_chown "path", 0, undef;
		511
		512
		513	=item aio_truncate $fh_or_path, $offset, $callback->($status)
		514
		515	Works like truncate(2) or ftruncate(2).
		516
		517
		518	=item aio_chmod $fh_or_path, $mode, $callback->($status)
		519
		520	Works like perl's C<chmod> function.
		521
		522
179	=item aio_unlink $pathname, $callback	523	=item aio_unlink $pathname, $callback->($status)
180		524
181	Asynchronously unlink (delete) a file and call the callback with the	525	Asynchronously unlink (delete) a file and call the callback with the
182	result code.	526	result code.
183		527
		528
		529	=item aio_mknod $path, $mode, $dev, $callback->($status)
		530
		531	[EXPERIMENTAL]
		532
		533	Asynchronously create a device node (or fifo). See mknod(2).
		534
		535	The only (POSIX-) portable way of calling this function is:
		536
		537	aio_mknod $path, IO::AIO::S_IFIFO | $mode, 0, sub { ...
		538
		539
		540	=item aio_link $srcpath, $dstpath, $callback->($status)
		541
		542	Asynchronously create a new link to the existing object at C<$srcpath> at
		543	the path C<$dstpath> and call the callback with the result code.
		544
		545
		546	=item aio_symlink $srcpath, $dstpath, $callback->($status)
		547
		548	Asynchronously create a new symbolic link to the existing object at C<$srcpath> at
		549	the path C<$dstpath> and call the callback with the result code.
		550
		551
		552	=item aio_readlink $path, $callback->($link)
		553
		554	Asynchronously read the symlink specified by C<$path> and pass it to
		555	the callback. If an error occurs, nothing or undef gets passed to the
		556	callback.
		557
		558
		559	=item aio_rename $srcpath, $dstpath, $callback->($status)
		560
		561	Asynchronously rename the object at C<$srcpath> to C<$dstpath>, just as
		562	rename(2) and call the callback with the result code.
		563
		564
		565	=item aio_mkdir $pathname, $mode, $callback->($status)
		566
		567	Asynchronously mkdir (create) a directory and call the callback with
		568	the result code. C<$mode> will be modified by the umask at the time the
		569	request is executed, so do not change your umask.
		570
		571
		572	=item aio_rmdir $pathname, $callback->($status)
		573
		574	Asynchronously rmdir (delete) a directory and call the callback with the
		575	result code.
		576
		577
		578	=item aio_readdir $pathname, $callback->($entries)
		579
		580	Unlike the POSIX call of the same name, C<aio_readdir> reads an entire
		581	directory (i.e. opendir + readdir + closedir). The entries will not be
		582	sorted, and will B<NOT> include the C<.> and C<..> entries.
		583
		584	The callback is passed a single argument which is either C<undef> or an
		585	array-ref with the filenames.
		586
		587
		588	=item aio_readdirx $pathname, $flags, $callback->($entries, $flags)
		589
		590	Quite similar to C<aio_readdir>, but the C<$flags> argument allows to tune
		591	behaviour and output format. In case of an error, C<$entries> will be
		592	C<undef>.
		593
		594	The flags are a combination of the following constants, ORed together (the
		595	flags will also be passed to the callback, possibly modified):
		596
		597	=over 4
		598
		599	=item IO::AIO::READDIR_DENTS
		600
		601	When this flag is off, then the callback gets an arrayref with of names
		602	only (as with C<aio_readdir>), otherwise it gets an arrayref with
		603	C<[$name, $type, $inode]> arrayrefs, each describing a single directory
		604	entry in more detail.
		605
		606	C<$name> is the name of the entry.
		607
		608	C<$type> is one of the C<IO::AIO::DT_xxx> constants:
		609
		610	C<IO::AIO::DT_UNKNOWN>, C<IO::AIO::DT_FIFO>, C<IO::AIO::DT_CHR>, C<IO::AIO::DT_DIR>,
		611	C<IO::AIO::DT_BLK>, C<IO::AIO::DT_REG>, C<IO::AIO::DT_LNK>, C<IO::AIO::DT_SOCK>,
		612	C<IO::AIO::DT_WHT>.
		613
		614	C<IO::AIO::DT_UNKNOWN> means just that: readdir does not know. If you need to
		615	know, you have to run stat yourself. Also, for speed reasons, the C<$type>
		616	scalars are read-only: you can not modify them.
		617
		618	C<$inode> is the inode number (which might not be exact on systems with 64
		619	bit inode numbers and 32 bit perls). This field has unspecified content on
		620	systems that do not deliver the inode information.
		621
		622	=item IO::AIO::READDIR_DIRS_FIRST
		623
		624	When this flag is set, then the names will be returned in an order where
		625	likely directories come first. This is useful when you need to quickly
		626	find directories, or you want to find all directories while avoiding to
		627	stat() each entry.
		628
		629	If the system returns type information in readdir, then this is used
		630	to find directories directly. Otherwise, likely directories are files
		631	beginning with ".", or otherwise files with no dots, of which files with
		632	short names are tried first.
		633
		634	=item IO::AIO::READDIR_STAT_ORDER
		635
		636	When this flag is set, then the names will be returned in an order
		637	suitable for stat()'ing each one. That is, when you plan to stat()
		638	all files in the given directory, then the returned order will likely
		639	be fastest.
		640
		641	If both this flag and C<IO::AIO::READDIR_DIRS_FIRST> are specified, then
		642	the likely dirs come first, resulting in a less optimal stat order.
		643
		644	=item IO::AIO::READDIR_FOUND_UNKNOWN
		645
		646	This flag should not be set when calling C<aio_readdirx>. Instead, it
		647	is being set by C<aio_readdirx>, when any of the C<$type>'s found were
		648	C<IO::AIO::DT_UNKNOWN>. The absense of this flag therefore indicates that all
		649	C<$type>'s are known, which can be used to speed up some algorithms.
		650
		651	=back
		652
		653
		654	=item aio_load $path, $data, $callback->($status)
		655
		656	This is a composite request that tries to fully load the given file into
		657	memory. Status is the same as with aio_read.
		658
		659	=cut
		660
		661	sub aio_load($$;$) {
		662	my ($path, undef, $cb) = @_;
		663	my $data = \$_[1];
		664
		665	my $pri = aioreq_pri;
		666	my $grp = aio_group $cb;
		667
		668	aioreq_pri $pri;
		669	add $grp aio_open $path, O_RDONLY, 0, sub {
		670	my $fh = shift
		671	or return $grp->result (-1);
		672
		673	aioreq_pri $pri;
		674	add $grp aio_read $fh, 0, (-s $fh), $$data, 0, sub {
		675	$grp->result ($_[0]);
		676	};
		677	};
		678
		679	$grp
		680	}
		681
		682	=item aio_copy $srcpath, $dstpath, $callback->($status)
		683
		684	Try to copy the I<file> (directories not supported as either source or
		685	destination) from C<$srcpath> to C<$dstpath> and call the callback with
		686	a status of C<0> (ok) or C<-1> (error, see C<$!>).
		687
		688	This is a composite request that creates the destination file with
		689	mode 0200 and copies the contents of the source file into it using
		690	C<aio_sendfile>, followed by restoring atime, mtime, access mode and
		691	uid/gid, in that order.
		692
		693	If an error occurs, the partial destination file will be unlinked, if
		694	possible, except when setting atime, mtime, access mode and uid/gid, where
		695	errors are being ignored.
		696
		697	=cut
		698
		699	sub aio_copy($$;$) {
		700	my ($src, $dst, $cb) = @_;
		701
		702	my $pri = aioreq_pri;
		703	my $grp = aio_group $cb;
		704
		705	aioreq_pri $pri;
		706	add $grp aio_open $src, O_RDONLY, 0, sub {
		707	if (my $src_fh = $_[0]) {
		708	my @stat = stat $src_fh; # hmm, might block over nfs?
		709
		710	aioreq_pri $pri;
		711	add $grp aio_open $dst, O_CREAT | O_WRONLY | O_TRUNC, 0200, sub {
		712	if (my $dst_fh = $_[0]) {
		713	aioreq_pri $pri;
		714	add $grp aio_sendfile $dst_fh, $src_fh, 0, $stat[7], sub {
		715	if ($_[0] == $stat[7]) {
		716	$grp->result (0);
		717	close $src_fh;
		718
		719	my $ch = sub {
		720	aioreq_pri $pri;
		721	add $grp aio_chmod $dst_fh, $stat[2] & 07777, sub {
		722	aioreq_pri $pri;
		723	add $grp aio_chown $dst_fh, $stat[4], $stat[5], sub {
		724	aioreq_pri $pri;
		725	add $grp aio_close $dst_fh;
		726	}
		727	};
		728	};
		729
		730	aioreq_pri $pri;
		731	add $grp aio_utime $dst_fh, $stat[8], $stat[9], sub {
		732	if ($_[0] < 0 && $! == ENOSYS) {
		733	aioreq_pri $pri;
		734	add $grp aio_utime $dst, $stat[8], $stat[9], $ch;
		735	} else {
		736	$ch->();
		737	}
		738	};
		739	} else {
		740	$grp->result (-1);
		741	close $src_fh;
		742	close $dst_fh;
		743
		744	aioreq $pri;
		745	add $grp aio_unlink $dst;
		746	}
		747	};
		748	} else {
		749	$grp->result (-1);
		750	}
		751	},
		752
		753	} else {
		754	$grp->result (-1);
		755	}
		756	};
		757
		758	$grp
		759	}
		760
		761	=item aio_move $srcpath, $dstpath, $callback->($status)
		762
		763	Try to move the I<file> (directories not supported as either source or
		764	destination) from C<$srcpath> to C<$dstpath> and call the callback with
		765	a status of C<0> (ok) or C<-1> (error, see C<$!>).
		766
		767	This is a composite request that tries to rename(2) the file first; if
		768	rename fails with C<EXDEV>, it copies the file with C<aio_copy> and, if
		769	that is successful, unlinks the C<$srcpath>.
		770
		771	=cut
		772
		773	sub aio_move($$;$) {
		774	my ($src, $dst, $cb) = @_;
		775
		776	my $pri = aioreq_pri;
		777	my $grp = aio_group $cb;
		778
		779	aioreq_pri $pri;
		780	add $grp aio_rename $src, $dst, sub {
		781	if ($_[0] && $! == EXDEV) {
		782	aioreq_pri $pri;
		783	add $grp aio_copy $src, $dst, sub {
		784	$grp->result ($_[0]);
		785
		786	if (!$_[0]) {
		787	aioreq_pri $pri;
		788	add $grp aio_unlink $src;
		789	}
		790	};
		791	} else {
		792	$grp->result ($_[0]);
		793	}
		794	};
		795
		796	$grp
		797	}
		798
		799	=item aio_scandir $path, $maxreq, $callback->($dirs, $nondirs)
		800
		801	Scans a directory (similar to C<aio_readdir>) but additionally tries to
		802	efficiently separate the entries of directory C<$path> into two sets of
		803	names, directories you can recurse into (directories), and ones you cannot
		804	recurse into (everything else, including symlinks to directories).
		805
		806	C<aio_scandir> is a composite request that creates of many sub requests_
		807	C<$maxreq> specifies the maximum number of outstanding aio requests that
		808	this function generates. If it is C<< <= 0 >>, then a suitable default
		809	will be chosen (currently 4).
		810
		811	On error, the callback is called without arguments, otherwise it receives
		812	two array-refs with path-relative entry names.
		813
		814	Example:
		815
		816	aio_scandir $dir, 0, sub {
		817	my ($dirs, $nondirs) = @_;
		818	print "real directories: @$dirs\n";
		819	print "everything else: @$nondirs\n";
		820	};
		821
		822	Implementation notes.
		823
		824	The C<aio_readdir> cannot be avoided, but C<stat()>'ing every entry can.
		825
		826	If readdir returns file type information, then this is used directly to
		827	find directories.
		828
		829	Otherwise, after reading the directory, the modification time, size etc.
		830	of the directory before and after the readdir is checked, and if they
		831	match (and isn't the current time), the link count will be used to decide
		832	how many entries are directories (if >= 2). Otherwise, no knowledge of the
		833	number of subdirectories will be assumed.
		834
		835	Then entries will be sorted into likely directories a non-initial dot
		836	currently) and likely non-directories (see C<aio_readdirx>). Then every
		837	entry plus an appended C</.> will be C<stat>'ed, likely directories first,
		838	in order of their inode numbers. If that succeeds, it assumes that the
		839	entry is a directory or a symlink to directory (which will be checked
		840	seperately). This is often faster than stat'ing the entry itself because
		841	filesystems might detect the type of the entry without reading the inode
		842	data (e.g. ext2fs filetype feature), even on systems that cannot return
		843	the filetype information on readdir.
		844
		845	If the known number of directories (link count - 2) has been reached, the
		846	rest of the entries is assumed to be non-directories.
		847
		848	This only works with certainty on POSIX (= UNIX) filesystems, which
		849	fortunately are the vast majority of filesystems around.
		850
		851	It will also likely work on non-POSIX filesystems with reduced efficiency
		852	as those tend to return 0 or 1 as link counts, which disables the
		853	directory counting heuristic.
		854
		855	=cut
		856
		857	sub aio_scandir($$;$) {
		858	my ($path, $maxreq, $cb) = @_;
		859
		860	my $pri = aioreq_pri;
		861
		862	my $grp = aio_group $cb;
		863
		864	$maxreq = 4 if $maxreq <= 0;
		865
		866	# stat once
		867	aioreq_pri $pri;
		868	add $grp aio_stat $path, sub {
		869	return $grp->result () if $_[0];
		870	my $now = time;
		871	my $hash1 = join ":", (stat _)[0,1,3,7,9];
		872
		873	# read the directory entries
		874	aioreq_pri $pri;
		875	add $grp aio_readdirx $path, READDIR_DIRS_FIRST, sub {
		876	my $entries = shift
		877	or return $grp->result ();
		878
		879	# stat the dir another time
		880	aioreq_pri $pri;
		881	add $grp aio_stat $path, sub {
		882	my $hash2 = join ":", (stat _)[0,1,3,7,9];
		883
		884	my $ndirs;
		885
		886	# take the slow route if anything looks fishy
		887	if ($hash1 ne $hash2 or (stat _)[9] == $now) {
		888	$ndirs = -1;
		889	} else {
		890	# if nlink == 2, we are finished
		891	# for non-posix-fs's, we rely on nlink < 2
		892	$ndirs = (stat _)[3] - 2
		893	or return $grp->result ([], $entries);
		894	}
		895
		896	my (@dirs, @nondirs);
		897
		898	my $statgrp = add $grp aio_group sub {
		899	$grp->result (\@dirs, \@nondirs);
		900	};
		901
		902	limit $statgrp $maxreq;
		903	feed $statgrp sub {
		904	return unless @$entries;
		905	my $entry = shift @$entries;
		906
		907	aioreq_pri $pri;
		908	add $statgrp aio_stat "$path/$entry/.", sub {
		909	if ($_[0] < 0) {
		910	push @nondirs, $entry;
		911	} else {
		912	# need to check for real directory
		913	aioreq_pri $pri;
		914	add $statgrp aio_lstat "$path/$entry", sub {
		915	if (-d _) {
		916	push @dirs, $entry;
		917
		918	unless (--$ndirs) {
		919	push @nondirs, @$entries;
		920	feed $statgrp;
		921	}
		922	} else {
		923	push @nondirs, $entry;
		924	}
		925	}
		926	}
		927	};
		928	};
		929	};
		930	};
		931	};
		932
		933	$grp
		934	}
		935
		936	=item aio_rmtree $path, $callback->($status)
		937
		938	Delete a directory tree starting (and including) C<$path>, return the
		939	status of the final C<rmdir> only. This is a composite request that
		940	uses C<aio_scandir> to recurse into and rmdir directories, and unlink
		941	everything else.
		942
		943	=cut
		944
		945	sub aio_rmtree;
		946	sub aio_rmtree($;$) {
		947	my ($path, $cb) = @_;
		948
		949	my $pri = aioreq_pri;
		950	my $grp = aio_group $cb;
		951
		952	aioreq_pri $pri;
		953	add $grp aio_scandir $path, 0, sub {
		954	my ($dirs, $nondirs) = @_;
		955
		956	my $dirgrp = aio_group sub {
		957	add $grp aio_rmdir $path, sub {
		958	$grp->result ($_[0]);
		959	};
		960	};
		961
		962	(aioreq_pri $pri), add $dirgrp aio_rmtree "$path/$_" for @$dirs;
		963	(aioreq_pri $pri), add $dirgrp aio_unlink "$path/$_" for @$nondirs;
		964
		965	add $grp $dirgrp;
		966	};
		967
		968	$grp
		969	}
		970
		971	=item aio_sync $callback->($status)
		972
		973	Asynchronously call sync and call the callback when finished.
		974
184	=item aio_fsync $fh, $callback	975	=item aio_fsync $fh, $callback->($status)
185		976
186	Asynchronously call fsync on the given filehandle and call the callback	977	Asynchronously call fsync on the given filehandle and call the callback
187	with the fsync result code.	978	with the fsync result code.
188		979
189	=item aio_fdatasync $fh, $callback	980	=item aio_fdatasync $fh, $callback->($status)
190		981
191	Asynchronously call fdatasync on the given filehandle and call the	982	Asynchronously call fdatasync on the given filehandle and call the
192	callback with the fdatasync result code.	983	callback with the fdatasync result code.
193		984
		985	If this call isn't available because your OS lacks it or it couldn't be
		986	detected, it will be emulated by calling C<fsync> instead.
		987
		988	=item aio_sync_file_range $fh, $offset, $nbytes, $flags, $callback->($status)
		989
		990	Sync the data portion of the file specified by C<$offset> and C<$length>
		991	to disk (but NOT the metadata), by calling the Linux-specific
		992	sync_file_range call. If sync_file_range is not available or it returns
		993	ENOSYS, then fdatasync or fsync is being substituted.
		994
		995	C<$flags> can be a combination of C<IO::AIO::SYNC_FILE_RANGE_WAIT_BEFORE>,
		996	C<IO::AIO::SYNC_FILE_RANGE_WRITE> and
		997	C<IO::AIO::SYNC_FILE_RANGE_WAIT_AFTER>: refer to the sync_file_range
		998	manpage for details.
		999
		1000	=item aio_pathsync $path, $callback->($status)
		1001
		1002	This request tries to open, fsync and close the given path. This is a
		1003	composite request intended to sync directories after directory operations
		1004	(E.g. rename). This might not work on all operating systems or have any
		1005	specific effect, but usually it makes sure that directory changes get
		1006	written to disc. It works for anything that can be opened for read-only,
		1007	not just directories.
		1008
		1009	Future versions of this function might fall back to other methods when
		1010	C<fsync> on the directory fails (such as calling C<sync>).
		1011
		1012	Passes C<0> when everything went ok, and C<-1> on error.
		1013
		1014	=cut
		1015
		1016	sub aio_pathsync($;$) {
		1017	my ($path, $cb) = @_;
		1018
		1019	my $pri = aioreq_pri;
		1020	my $grp = aio_group $cb;
		1021
		1022	aioreq_pri $pri;
		1023	add $grp aio_open $path, O_RDONLY, 0, sub {
		1024	my ($fh) = @_;
		1025	if ($fh) {
		1026	aioreq_pri $pri;
		1027	add $grp aio_fsync $fh, sub {
		1028	$grp->result ($_[0]);
		1029
		1030	aioreq_pri $pri;
		1031	add $grp aio_close $fh;
		1032	};
		1033	} else {
		1034	$grp->result (-1);
		1035	}
		1036	};
		1037
		1038	$grp
		1039	}
		1040
		1041	=item aio_msync $scalar, $offset = 0, $length = undef, flags = 0, $callback->($status)
		1042
		1043	This is a rather advanced IO::AIO call, which only works on mmap(2)ed
		1044	scalars (see the L<Sys::Mmap> or L<Mmap> modules for details on this, note
		1045	that the scalar must only be modified in-place while an aio operation is
		1046	pending on it).
		1047
		1048	It calls the C<msync> function of your OS, if available, with the memory
		1049	area starting at C<$offset> in the string and ending C<$length> bytes
		1050	later. If C<$length> is negative, counts from the end, and if C<$length>
		1051	is C<undef>, then it goes till the end of the string. The flags can be
		1052	a combination of C<IO::AIO::MS_ASYNC>, C<IO::AIO::MS_INVALIDATE> and
		1053	C<IO::AIO::MS_SYNC>.
		1054
		1055	=item aio_mtouch $scalar, $offset = 0, $length = undef, flags = 0, $callback->($status)
		1056
		1057	This is a rather advanced IO::AIO call, which works best on mmap(2)ed
		1058	scalars.
		1059
		1060	It touches (reads or writes) all memory pages in the specified
		1061	range inside the scalar. All caveats and parameters are the same
		1062	as for C<aio_msync>, above, except for flags, which must be either
		1063	C<0> (which reads all pages and ensures they are instantiated) or
		1064	C<IO::AIO::MT_MODIFY>, which modifies the memory page s(by reading and
		1065	writing an octet from it, which dirties the page).
		1066
		1067	=item aio_group $callback->(...)
		1068
		1069	This is a very special aio request: Instead of doing something, it is a
		1070	container for other aio requests, which is useful if you want to bundle
		1071	many requests into a single, composite, request with a definite callback
		1072	and the ability to cancel the whole request with its subrequests.
		1073
		1074	Returns an object of class L<IO::AIO::GRP>. See its documentation below
		1075	for more info.
		1076
		1077	Example:
		1078
		1079	my $grp = aio_group sub {
		1080	print "all stats done\n";
		1081	};
		1082
		1083	add $grp
		1084	(aio_stat ...),
		1085	(aio_stat ...),
		1086	...;
		1087
		1088	=item aio_nop $callback->()
		1089
		1090	This is a special request - it does nothing in itself and is only used for
		1091	side effects, such as when you want to add a dummy request to a group so
		1092	that finishing the requests in the group depends on executing the given
		1093	code.
		1094
		1095	While this request does nothing, it still goes through the execution
		1096	phase and still requires a worker thread. Thus, the callback will not
		1097	be executed immediately but only after other requests in the queue have
		1098	entered their execution phase. This can be used to measure request
		1099	latency.
		1100
		1101	=item IO::AIO::aio_busy $fractional_seconds, $callback->() *NOT EXPORTED*
		1102
		1103	Mainly used for debugging and benchmarking, this aio request puts one of
		1104	the request workers to sleep for the given time.
		1105
		1106	While it is theoretically handy to have simple I/O scheduling requests
		1107	like sleep and file handle readable/writable, the overhead this creates is
		1108	immense (it blocks a thread for a long time) so do not use this function
		1109	except to put your application under artificial I/O pressure.
		1110
194	=back	1111	=back
195		1112
		1113	=head2 IO::AIO::REQ CLASS
		1114
		1115	All non-aggregate C<aio_*> functions return an object of this class when
		1116	called in non-void context.
		1117
		1118	=over 4
		1119
		1120	=item cancel $req
		1121
		1122	Cancels the request, if possible. Has the effect of skipping execution
		1123	when entering the B<execute> state and skipping calling the callback when
		1124	entering the the B<result> state, but will leave the request otherwise
		1125	untouched (with the exception of readdir). That means that requests that
		1126	currently execute will not be stopped and resources held by the request
		1127	will not be freed prematurely.
		1128
		1129	=item cb $req $callback->(...)
		1130
		1131	Replace (or simply set) the callback registered to the request.
		1132
		1133	=back
		1134
		1135	=head2 IO::AIO::GRP CLASS
		1136
		1137	This class is a subclass of L<IO::AIO::REQ>, so all its methods apply to
		1138	objects of this class, too.
		1139
		1140	A IO::AIO::GRP object is a special request that can contain multiple other
		1141	aio requests.
		1142
		1143	You create one by calling the C<aio_group> constructing function with a
		1144	callback that will be called when all contained requests have entered the
		1145	C<done> state:
		1146
		1147	my $grp = aio_group sub {
		1148	print "all requests are done\n";
		1149	};
		1150
		1151	You add requests by calling the C<add> method with one or more
		1152	C<IO::AIO::REQ> objects:
		1153
		1154	$grp->add (aio_unlink "...");
		1155
		1156	add $grp aio_stat "...", sub {
		1157	$_[0] or return $grp->result ("error");
		1158
		1159	# add another request dynamically, if first succeeded
		1160	add $grp aio_open "...", sub {
		1161	$grp->result ("ok");
		1162	};
		1163	};
		1164
		1165	This makes it very easy to create composite requests (see the source of
		1166	C<aio_move> for an application) that work and feel like simple requests.
		1167
		1168	=over 4
		1169
		1170	=item * The IO::AIO::GRP objects will be cleaned up during calls to
		1171	C<IO::AIO::poll_cb>, just like any other request.
		1172
		1173	=item * They can be canceled like any other request. Canceling will cancel not
		1174	only the request itself, but also all requests it contains.
		1175
		1176	=item * They can also can also be added to other IO::AIO::GRP objects.
		1177
		1178	=item * You must not add requests to a group from within the group callback (or
		1179	any later time).
		1180
		1181	=back
		1182
		1183	Their lifetime, simplified, looks like this: when they are empty, they
		1184	will finish very quickly. If they contain only requests that are in the
		1185	C<done> state, they will also finish. Otherwise they will continue to
		1186	exist.
		1187
		1188	That means after creating a group you have some time to add requests
		1189	(precisely before the callback has been invoked, which is only done within
		1190	the C<poll_cb>). And in the callbacks of those requests, you can add
		1191	further requests to the group. And only when all those requests have
		1192	finished will the the group itself finish.
		1193
		1194	=over 4
		1195
		1196	=item add $grp ...
		1197
		1198	=item $grp->add (...)
		1199
		1200	Add one or more requests to the group. Any type of L<IO::AIO::REQ> can
		1201	be added, including other groups, as long as you do not create circular
		1202	dependencies.
		1203
		1204	Returns all its arguments.
		1205
		1206	=item $grp->cancel_subs
		1207
		1208	Cancel all subrequests and clears any feeder, but not the group request
		1209	itself. Useful when you queued a lot of events but got a result early.
		1210
		1211	The group request will finish normally (you cannot add requests to the
		1212	group).
		1213
		1214	=item $grp->result (...)
		1215
		1216	Set the result value(s) that will be passed to the group callback when all
		1217	subrequests have finished and set the groups errno to the current value
		1218	of errno (just like calling C<errno> without an error number). By default,
		1219	no argument will be passed and errno is zero.
		1220
		1221	=item $grp->errno ([$errno])
		1222
		1223	Sets the group errno value to C<$errno>, or the current value of errno
		1224	when the argument is missing.
		1225
		1226	Every aio request has an associated errno value that is restored when
		1227	the callback is invoked. This method lets you change this value from its
		1228	default (0).
		1229
		1230	Calling C<result> will also set errno, so make sure you either set C<$!>
		1231	before the call to C<result>, or call c<errno> after it.
		1232
		1233	=item feed $grp $callback->($grp)
		1234
		1235	Sets a feeder/generator on this group: every group can have an attached
		1236	generator that generates requests if idle. The idea behind this is that,
		1237	although you could just queue as many requests as you want in a group,
		1238	this might starve other requests for a potentially long time. For example,
		1239	C<aio_scandir> might generate hundreds of thousands C<aio_stat> requests,
		1240	delaying any later requests for a long time.
		1241
		1242	To avoid this, and allow incremental generation of requests, you can
		1243	instead a group and set a feeder on it that generates those requests. The
		1244	feed callback will be called whenever there are few enough (see C<limit>,
		1245	below) requests active in the group itself and is expected to queue more
		1246	requests.
		1247
		1248	The feed callback can queue as many requests as it likes (i.e. C<add> does
		1249	not impose any limits).
		1250
		1251	If the feed does not queue more requests when called, it will be
		1252	automatically removed from the group.
		1253
		1254	If the feed limit is C<0> when this method is called, it will be set to
		1255	C<2> automatically.
		1256
		1257	Example:
		1258
		1259	# stat all files in @files, but only ever use four aio requests concurrently:
		1260
		1261	my $grp = aio_group sub { print "finished\n" };
		1262	limit $grp 4;
		1263	feed $grp sub {
		1264	my $file = pop @files
		1265	or return;
		1266
		1267	add $grp aio_stat $file, sub { ... };
		1268	};
		1269
		1270	=item limit $grp $num
		1271
		1272	Sets the feeder limit for the group: The feeder will be called whenever
		1273	the group contains less than this many requests.
		1274
		1275	Setting the limit to C<0> will pause the feeding process.
		1276
		1277	The default value for the limit is C<0>, but note that setting a feeder
		1278	automatically bumps it up to C<2>.
		1279
		1280	=back
		1281
196	=head2 SUPPORT FUNCTIONS	1282	=head2 SUPPORT FUNCTIONS
197		1283
		1284	=head3 EVENT PROCESSING AND EVENT LOOP INTEGRATION
		1285
198	=over 4	1286	=over 4
199		1287
200	=item $fileno = IO::AIO::poll_fileno	1288	=item $fileno = IO::AIO::poll_fileno
201		1289
202	Return the I<request result pipe filehandle>. This filehandle must be	1290	Return the I<request result pipe file descriptor>. This filehandle must be
203	polled for reading by some mechanism outside this module (e.g. Event	1291	polled for reading by some mechanism outside this module (e.g. EV, Glib,
204	or select, see below). If the pipe becomes readable you have to call	1292	select and so on, see below or the SYNOPSIS). If the pipe becomes readable
205	C<poll_cb> to check the results.	1293	you have to call C<poll_cb> to check the results.
206		1294
207	See C<poll_cb> for an example.	1295	See C<poll_cb> for an example.
208		1296
209	=item IO::AIO::poll_cb	1297	=item IO::AIO::poll_cb
210		1298
211	Process all outstanding events on the result pipe. You have to call this	1299	Process some outstanding events on the result pipe. You have to call this
212	regularly. Returns the number of events processed. Returns immediately	1300	regularly. Returns C<0> if all events could be processed, or C<-1> if it
213	when no events are outstanding.	1301	returned earlier for whatever reason. Returns immediately when no events
		1302	are outstanding. The amount of events processed depends on the settings of
		1303	C<IO::AIO::max_poll_req> and C<IO::AIO::max_poll_time>.
214		1304
215	You can use Event to multiplex, e.g.:	1305	If not all requests were processed for whatever reason, the filehandle
		1306	will still be ready when C<poll_cb> returns, so normally you don't have to
		1307	do anything special to have it called later.
		1308
		1309	Example: Install an Event watcher that automatically calls
		1310	IO::AIO::poll_cb with high priority (more examples can be found in the
		1311	SYNOPSIS section, at the top of this document):
216		1312
217	Event->io (fd => IO::AIO::poll_fileno,	1313	Event->io (fd => IO::AIO::poll_fileno,
218	poll => 'r', async => 1,	1314	poll => 'r', async => 1,
219	cb => \&IO::AIO::poll_cb);	1315	cb => \&IO::AIO::poll_cb);
220		1316
		1317	=item IO::AIO::max_poll_reqs $nreqs
		1318
		1319	=item IO::AIO::max_poll_time $seconds
		1320
		1321	These set the maximum number of requests (default C<0>, meaning infinity)
		1322	that are being processed by C<IO::AIO::poll_cb> in one call, respectively
		1323	the maximum amount of time (default C<0>, meaning infinity) spent in
		1324	C<IO::AIO::poll_cb> to process requests (more correctly the mininum amount
		1325	of time C<poll_cb> is allowed to use).
		1326
		1327	Setting C<max_poll_time> to a non-zero value creates an overhead of one
		1328	syscall per request processed, which is not normally a problem unless your
		1329	callbacks are really really fast or your OS is really really slow (I am
		1330	not mentioning Solaris here). Using C<max_poll_reqs> incurs no overhead.
		1331
		1332	Setting these is useful if you want to ensure some level of
		1333	interactiveness when perl is not fast enough to process all requests in
		1334	time.
		1335
		1336	For interactive programs, values such as C<0.01> to C<0.1> should be fine.
		1337
		1338	Example: Install an Event watcher that automatically calls
		1339	IO::AIO::poll_cb with low priority, to ensure that other parts of the
		1340	program get the CPU sometimes even under high AIO load.
		1341
		1342	# try not to spend much more than 0.1s in poll_cb
		1343	IO::AIO::max_poll_time 0.1;
		1344
		1345	# use a low priority so other tasks have priority
		1346	Event->io (fd => IO::AIO::poll_fileno,
		1347	poll => 'r', nice => 1,
		1348	cb => &IO::AIO::poll_cb);
		1349
221	=item IO::AIO::poll_wait	1350	=item IO::AIO::poll_wait
222		1351
		1352	If there are any outstanding requests and none of them in the result
223	Wait till the result filehandle becomes ready for reading (simply does a	1353	phase, wait till the result filehandle becomes ready for reading (simply
224	select on the filehandle. This is useful if you want to synchronously wait	1354	does a C<select> on the filehandle. This is useful if you want to
225	for some requests to finish).	1355	synchronously wait for some requests to finish).
226		1356
227	See C<nreqs> for an example.	1357	See C<nreqs> for an example.
228		1358
		1359	=item IO::AIO::poll
		1360
		1361	Waits until some requests have been handled.
		1362
		1363	Returns the number of requests processed, but is otherwise strictly
		1364	equivalent to:
		1365
		1366	IO::AIO::poll_wait, IO::AIO::poll_cb
		1367
229	=item IO::AIO::nreqs	1368	=item IO::AIO::flush
230		1369
231	Returns the number of requests currently outstanding.	1370	Wait till all outstanding AIO requests have been handled.
232		1371
233	Example: wait till there are no outstanding requests anymore:	1372	Strictly equivalent to:
234		1373
235	IO::AIO::poll_wait, IO::AIO::poll_cb	1374	IO::AIO::poll_wait, IO::AIO::poll_cb
236	while IO::AIO::nreqs;	1375	while IO::AIO::nreqs;
237		1376
		1377	=back
		1378
		1379	=head3 CONTROLLING THE NUMBER OF THREADS
		1380
		1381	=over
		1382
		1383	=item IO::AIO::min_parallel $nthreads
		1384
		1385	Set the minimum number of AIO threads to C<$nthreads>. The current
		1386	default is C<8>, which means eight asynchronous operations can execute
		1387	concurrently at any one time (the number of outstanding requests,
		1388	however, is unlimited).
		1389
		1390	IO::AIO starts threads only on demand, when an AIO request is queued and
		1391	no free thread exists. Please note that queueing up a hundred requests can
		1392	create demand for a hundred threads, even if it turns out that everything
		1393	is in the cache and could have been processed faster by a single thread.
		1394
		1395	It is recommended to keep the number of threads relatively low, as some
		1396	Linux kernel versions will scale negatively with the number of threads
		1397	(higher parallelity => MUCH higher latency). With current Linux 2.6
		1398	versions, 4-32 threads should be fine.
		1399
		1400	Under most circumstances you don't need to call this function, as the
		1401	module selects a default that is suitable for low to moderate load.
		1402
		1403	=item IO::AIO::max_parallel $nthreads
		1404
		1405	Sets the maximum number of AIO threads to C<$nthreads>. If more than the
		1406	specified number of threads are currently running, this function kills
		1407	them. This function blocks until the limit is reached.
		1408
		1409	While C<$nthreads> are zero, aio requests get queued but not executed
		1410	until the number of threads has been increased again.
		1411
		1412	This module automatically runs C<max_parallel 0> at program end, to ensure
		1413	that all threads are killed and that there are no outstanding requests.
		1414
		1415	Under normal circumstances you don't need to call this function.
		1416
		1417	=item IO::AIO::max_idle $nthreads
		1418
		1419	Limit the number of threads (default: 4) that are allowed to idle (i.e.,
		1420	threads that did not get a request to process within 10 seconds). That
		1421	means if a thread becomes idle while C<$nthreads> other threads are also
		1422	idle, it will free its resources and exit.
		1423
		1424	This is useful when you allow a large number of threads (e.g. 100 or 1000)
		1425	to allow for extremely high load situations, but want to free resources
		1426	under normal circumstances (1000 threads can easily consume 30MB of RAM).
		1427
		1428	The default is probably ok in most situations, especially if thread
		1429	creation is fast. If thread creation is very slow on your system you might
		1430	want to use larger values.
		1431
		1432	=item IO::AIO::max_outstanding $maxreqs
		1433
		1434	This is a very bad function to use in interactive programs because it
		1435	blocks, and a bad way to reduce concurrency because it is inexact: Better
		1436	use an C<aio_group> together with a feed callback.
		1437
		1438	Sets the maximum number of outstanding requests to C<$nreqs>. If you
		1439	do queue up more than this number of requests, the next call to the
		1440	C<poll_cb> (and C<poll_some> and other functions calling C<poll_cb>)
		1441	function will block until the limit is no longer exceeded.
		1442
		1443	The default value is very large, so there is no practical limit on the
		1444	number of outstanding requests.
		1445
		1446	You can still queue as many requests as you want. Therefore,
		1447	C<max_outstanding> is mainly useful in simple scripts (with low values) or
		1448	as a stop gap to shield against fatal memory overflow (with large values).
		1449
		1450	=back
		1451
		1452	=head3 STATISTICAL INFORMATION
		1453
		1454	=over
		1455
238	=item IO::AIO::flush	1456	=item IO::AIO::nreqs
239		1457
240	Wait till all outstanding AIO requests have been handled.	1458	Returns the number of requests currently in the ready, execute or pending
		1459	states (i.e. for which their callback has not been invoked yet).
241		1460
242	Strictly equivalent to:	1461	Example: wait till there are no outstanding requests anymore:
243		1462
244	IO::AIO::poll_wait, IO::AIO::poll_cb	1463	IO::AIO::poll_wait, IO::AIO::poll_cb
245	while IO::AIO::nreqs;	1464	while IO::AIO::nreqs;
246		1465
		1466	=item IO::AIO::nready
		1467
		1468	Returns the number of requests currently in the ready state (not yet
		1469	executed).
		1470
247	=item IO::AIO::poll	1471	=item IO::AIO::npending
248		1472
249	Waits until some requests have been handled.	1473	Returns the number of requests currently in the pending state (executed,
250		1474	but not yet processed by poll_cb).
251	Strictly equivalent to:
252
253	IO::AIO::poll_wait, IO::AIO::poll_cb
254	if IO::AIO::nreqs;
255
256	=item IO::AIO::min_parallel $nthreads
257
258	Set the minimum number of AIO threads to C<$nthreads>. The default is
259	C<1>, which means a single asynchronous operation can be done at one time
260	(the number of outstanding operations, however, is unlimited).
261
262	It is recommended to keep the number of threads low, as some Linux
263	kernel versions will scale negatively with the number of threads (higher
264	parallelity => MUCH higher latency). With current Linux 2.6 versions, 4-32
265	threads should be fine.
266
267	Under normal circumstances you don't need to call this function, as this
268	module automatically starts some threads (the exact number might change,
269	and is currently 4).
270
271	=item IO::AIO::max_parallel $nthreads
272
273	Sets the maximum number of AIO threads to C<$nthreads>. If more than
274	the specified number of threads are currently running, kill them. This
275	function blocks until the limit is reached.
276
277	This module automatically runs C<max_parallel 0> at program end, to ensure
278	that all threads are killed and that there are no outstanding requests.
279
280	Under normal circumstances you don't need to call this function.
281
282	=item $oldnreqs = IO::AIO::max_outstanding $nreqs
283
284	Sets the maximum number of outstanding requests to C<$nreqs>. If you
285	try to queue up more than this number of requests, the caller will block until
286	some requests have been handled.
287
288	The default is very large, so normally there is no practical limit. If you
289	queue up many requests in a loop it it often improves speed if you set
290	this to a relatively low number, such as C<100>.
291
292	Under normal circumstances you don't need to call this function.
293		1475
294	=back	1476	=back
295		1477
		1478	=head3 MISCELLANEOUS FUNCTIONS
		1479
		1480	IO::AIO implements some functions that might be useful, but are not
		1481	asynchronous.
		1482
		1483	=over 4
		1484
		1485	=item IO::AIO::sendfile $ofh, $ifh, $offset, $count
		1486
		1487	Calls the C<eio_sendfile_sync> function, which is like C<aio_sendfile>,
		1488	but is blocking (this makes most sense if you know the input data is
		1489	likely cached already and the output filehandle is set to non-blocking
		1490	operations).
		1491
		1492	Returns the number of bytes copied, or C<-1> on error.
		1493
		1494	=item IO::AIO::fadvise $fh, $offset, $len, $advice
		1495
		1496	Simply calls the C<posix_fadvise> function (see it's
		1497	manpage for details). The following advice constants are
		1498	avaiable: C<IO::AIO::FADV_NORMAL>, C<IO::AIO::FADV_SEQUENTIAL>,
		1499	C<IO::AIO::FADV_RANDOM>, C<IO::AIO::FADV_NOREUSE>,
		1500	C<IO::AIO::FADV_WILLNEED>, C<IO::AIO::FADV_DONTNEED>.
		1501
		1502	On systems that do not implement C<posix_fadvise>, this function returns
		1503	ENOSYS, otherwise the return value of C<posix_fadvise>.
		1504
		1505	=back
		1506
296	=cut	1507	=cut
297		1508
298	# support function to convert a fd into a perl filehandle
299	sub _fd2fh {
300	return undef if $_[0] < 0;
301
302	# try to be perl5.6-compatible
303	local *AIO_FH;
304	open AIO_FH, "+<&=$_[0]"
305	or return undef;
306
307	*AIO_FH
308	}
309
310	min_parallel 4;	1509	min_parallel 8;
311		1510
312	END {	1511	END { flush }
313	max_parallel 0;
314	}
315		1512
316	1;	1513	1;
317		1514
		1515	=head2 FORK BEHAVIOUR
		1516
		1517	This module should do "the right thing" when the process using it forks:
		1518
		1519	Before the fork, IO::AIO enters a quiescent state where no requests
		1520	can be added in other threads and no results will be processed. After
		1521	the fork the parent simply leaves the quiescent state and continues
		1522	request/result processing, while the child frees the request/result queue
		1523	(so that the requests started before the fork will only be handled in the
		1524	parent). Threads will be started on demand until the limit set in the
		1525	parent process has been reached again.
		1526
		1527	In short: the parent will, after a short pause, continue as if fork had
		1528	not been called, while the child will act as if IO::AIO has not been used
		1529	yet.
		1530
		1531	=head2 MEMORY USAGE
		1532
		1533	Per-request usage:
		1534
		1535	Each aio request uses - depending on your architecture - around 100-200
		1536	bytes of memory. In addition, stat requests need a stat buffer (possibly
		1537	a few hundred bytes), readdir requires a result buffer and so on. Perl
		1538	scalars and other data passed into aio requests will also be locked and
		1539	will consume memory till the request has entered the done state.
		1540
		1541	This is not awfully much, so queuing lots of requests is not usually a
		1542	problem.
		1543
		1544	Per-thread usage:
		1545
		1546	In the execution phase, some aio requests require more memory for
		1547	temporary buffers, and each thread requires a stack and other data
		1548	structures (usually around 16k-128k, depending on the OS).
		1549
		1550	=head1 KNOWN BUGS
		1551
		1552	Known bugs will be fixed in the next release.
		1553
318	=head1 SEE ALSO	1554	=head1 SEE ALSO
319		1555
320	L<Coro>, L<Linux::AIO>.	1556	L<AnyEvent::AIO> for easy integration into event loops, L<Coro::AIO> for a
		1557	more natural syntax.
321		1558
322	=head1 AUTHOR	1559	=head1 AUTHOR
323		1560
324	Marc Lehmann <schmorp@schmorp.de>	1561	Marc Lehmann <schmorp@schmorp.de>
325	http://home.schmorp.de/	1562	http://home.schmorp.de/

Diff Legend

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