ViewVC Help

View File | Revision Log | Show Annotations | Download File

/cvs/IO-AIO/AIO.pm

Comparing IO-AIO/AIO.pm (file contents):
Revision 1.6 by root, Sun Jul 10 22:19:48 2005 UTC vs.
Revision 1.171 by root, Sat Jan 2 14:24:32 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', async => 1,	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
		50	# Danga::Socket integration
		51	Danga::Socket->AddOtherFds (IO::AIO::poll_fileno =>
		52	\&IO::AIO::poll_cb);
		53
33	=head1 DESCRIPTION	54	=head1 DESCRIPTION
34		55
35	This module implements asynchronous I/O using whatever means your	56	This module implements asynchronous I/O using whatever means your
36	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>).
37		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
38	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
39	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
40	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
41	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
42	aio functions available on many operating systems. However, they are often	80	functions available on many operating systems. However, they are often
43	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
44	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
45	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.
46		85
47	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,
48	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
49		184
50	=cut	185	=cut
51		186
52	package IO::AIO;	187	package IO::AIO;
53		188
		189	use Carp ();
		190
		191	use common::sense;
		192
54	use base 'Exporter';	193	use base 'Exporter';
55		194
56	use Fcntl ();
57
58	BEGIN {	195	BEGIN {
59	$VERSION = 0.2;	196	our $VERSION = '3.4';
60		197
61	@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
62	aio_fsync aio_fdatasync aio_readahead);	199	aio_stat aio_lstat aio_unlink aio_rmdir aio_readdir aio_readdirx
63	@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);
		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';
64		217
65	require XSLoader;	218	require XSLoader;
66	XSLoader::load IO::AIO, $VERSION;	219	XSLoader::load ("IO::AIO", $VERSION);
67	}	220	}
68		221
69	=head1 FUNCTIONS	222	=head1 FUNCTIONS
70		223
71	=head2 AIO FUNCTIONS	224	=head2 AIO REQUEST FUNCTIONS
72		225
73	All the C<aio_*> calls are more or less thin wrappers around the syscall	226	All the C<aio_*> calls are more or less thin wrappers around the syscall
74	with the same name (sans C<aio_>). The arguments are similar or identical,	227	with the same name (sans C<aio_>). The arguments are similar or identical,
75	and they all accept an additional C<$callback> argument which must be	228	and they all accept an additional (and optional) C<$callback> argument
76	a code reference. This code reference will get called with the syscall	229	which must be a code reference. This code reference will get called with
77	return code (e.g. most syscalls return C<-1> on error, unlike perl, which	230	the syscall return code (e.g. most syscalls return C<-1> on error, unlike
78	usually delivers "false") as it's sole argument when the given syscall has	231	perl, which usually delivers "false") as its sole argument after the given
79	been executed asynchronously.	232	syscall has been executed asynchronously.
80		233
81	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.
82		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
83	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
84	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
85	directory could have changed. Alternatively, you can make sure that you	242	request is being executed, the current working directory could have
86	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.
87		256
88	=over 4	257	=over 4
89		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
90	=item aio_open $pathname, $flags, $mode, $callback	292	=item aio_open $pathname, $flags, $mode, $callback->($fh)
91		293
92	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
93	created filehandle for the file.	295	created filehandle for the file.
94		296
95	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,
96	for an explanation.	298	for an explanation.
97		299
98	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
99	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.
100		309
101	Example:	310	Example:
102		311
103	aio_open "/etc/passwd", O_RDONLY, 0, sub {	312	aio_open "/etc/passwd", O_RDONLY, 0, sub {
104	if ($_[0]) {	313	if ($_[0]) {
…		…
107	} else {	316	} else {
108	die "open failed: $!\n";	317	die "open failed: $!\n";
109	}	318	}
110	};	319	};
111		320
		321
112	=item aio_close $fh, $callback	322	=item aio_close $fh, $callback->($status)
113		323
114	Asynchronously close a file and call the callback with the result	324	Asynchronously close a file and call the callback with the result
115	code. I<WARNING:> although accepted, you should not pass in a perl	325	code.
116	filehandle here, as perl will likely close the file descriptor itself when
117	the filehandle is destroyed. Normally, you can safely call perls C<close>
118	or just let filehandles go out of scope.
119		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
120	=item aio_read $fh,$offset,$length, $data,$dataoffset,$callback	339	=item aio_read $fh,$offset,$length, $data,$dataoffset, $callback->($retval)
121		340
122	=item aio_write $fh,$offset,$length, $data,$dataoffset,$callback	341	=item aio_write $fh,$offset,$length, $data,$dataoffset, $callback->($retval)
123		342
124	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
125	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<$dataoffset>
126	callback without the actual number of bytes read (or -1 on error, just	345	and calls the callback without the actual number of bytes read (or -1 on
127	like the syscall).	346	error, just like the syscall).
128		347
		348	C<aio_read> will, like C<sysread>, shrink or grow the C<$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
129	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
130	offset C<0> within the scalar:	366	offset C<0> within the scalar:
131		367
132	aio_read $fh, 7, 15, $buffer, 0, sub {	368	aio_read $fh, 7, 15, $buffer, 0, sub {
133	$_[0] >= 0 or die "read error: $!";	369	$_[0] > 0 or die "read error: $!";
134	print "read <$buffer>\n";	370	print "read $_[0] bytes: <$buffer>\n";
135	};	371	};
136		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
137	=item aio_readahead $fh,$offset,$length, $callback	399	=item aio_readahead $fh,$offset,$length, $callback->($retval)
138		400
139	Asynchronously reads the specified byte range into the page cache, using
140	the C<readahead> syscall. If that syscall doesn't exist the status will be
141	C<-1> and C<$!> is set to ENOSYS.
142
143	readahead() populates the page cache with data from a file so that	401	C<aio_readahead> populates the page cache with data from a file so that
144	subsequent reads from that file will not block on disk I/O. The C<$offset>	402	subsequent reads from that file will not block on disk I/O. The C<$offset>
145	argument specifies the starting point from which data is to be read and	403	argument specifies the starting point from which data is to be read and
146	C<$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
147	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
148	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
149	(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
150	file. The current file offset of the file is left unchanged.	408	file. The current file offset of the file is left unchanged.
151		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
152	=item aio_stat $fh_or_path, $callback	414	=item aio_stat $fh_or_path, $callback->($status)
153		415
154	=item aio_lstat $fh, $callback	416	=item aio_lstat $fh, $callback->($status)
155		417
156	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
157	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 _>
158	or C<-s _> etc...	420	or C<-s _> etc...
159		421
…		…
169	aio_stat "/etc/passwd", sub {	431	aio_stat "/etc/passwd", sub {
170	$_[0] and die "stat failed: $!";	432	$_[0] and die "stat failed: $!";
171	print "size is ", -s _, "\n";	433	print "size is ", -s _, "\n";
172	};	434	};
173		435
		436
		437	=item aio_utime $fh_or_path, $atime, $mtime, $callback->($status)
		438
		439	Works like perl's C<utime> function (including the special case of $atime
		440	and $mtime being undef). Fractional times are supported if the underlying
		441	syscalls support them.
		442
		443	When called with a pathname, uses utimes(2) if available, otherwise
		444	utime(2). If called on a file descriptor, uses futimes(2) if available,
		445	otherwise returns ENOSYS, so this is not portable.
		446
		447	Examples:
		448
		449	# set atime and mtime to current time (basically touch(1)):
		450	aio_utime "path", undef, undef;
		451	# set atime to current time and mtime to beginning of the epoch:
		452	aio_utime "path", time, undef; # undef==0
		453
		454
		455	=item aio_chown $fh_or_path, $uid, $gid, $callback->($status)
		456
		457	Works like perl's C<chown> function, except that C<undef> for either $uid
		458	or $gid is being interpreted as "do not change" (but -1 can also be used).
		459
		460	Examples:
		461
		462	# same as "chown root path" in the shell:
		463	aio_chown "path", 0, -1;
		464	# same as above:
		465	aio_chown "path", 0, undef;
		466
		467
		468	=item aio_truncate $fh_or_path, $offset, $callback->($status)
		469
		470	Works like truncate(2) or ftruncate(2).
		471
		472
		473	=item aio_chmod $fh_or_path, $mode, $callback->($status)
		474
		475	Works like perl's C<chmod> function.
		476
		477
174	=item aio_unlink $pathname, $callback	478	=item aio_unlink $pathname, $callback->($status)
175		479
176	Asynchronously unlink (delete) a file and call the callback with the	480	Asynchronously unlink (delete) a file and call the callback with the
177	result code.	481	result code.
178		482
		483
		484	=item aio_mknod $path, $mode, $dev, $callback->($status)
		485
		486	[EXPERIMENTAL]
		487
		488	Asynchronously create a device node (or fifo). See mknod(2).
		489
		490	The only (POSIX-) portable way of calling this function is:
		491
		492	aio_mknod $path, IO::AIO::S_IFIFO | $mode, 0, sub { ...
		493
		494
		495	=item aio_link $srcpath, $dstpath, $callback->($status)
		496
		497	Asynchronously create a new link to the existing object at C<$srcpath> at
		498	the path C<$dstpath> and call the callback with the result code.
		499
		500
		501	=item aio_symlink $srcpath, $dstpath, $callback->($status)
		502
		503	Asynchronously create a new symbolic link to the existing object at C<$srcpath> at
		504	the path C<$dstpath> and call the callback with the result code.
		505
		506
		507	=item aio_readlink $path, $callback->($link)
		508
		509	Asynchronously read the symlink specified by C<$path> and pass it to
		510	the callback. If an error occurs, nothing or undef gets passed to the
		511	callback.
		512
		513
		514	=item aio_rename $srcpath, $dstpath, $callback->($status)
		515
		516	Asynchronously rename the object at C<$srcpath> to C<$dstpath>, just as
		517	rename(2) and call the callback with the result code.
		518
		519
		520	=item aio_mkdir $pathname, $mode, $callback->($status)
		521
		522	Asynchronously mkdir (create) a directory and call the callback with
		523	the result code. C<$mode> will be modified by the umask at the time the
		524	request is executed, so do not change your umask.
		525
		526
		527	=item aio_rmdir $pathname, $callback->($status)
		528
		529	Asynchronously rmdir (delete) a directory and call the callback with the
		530	result code.
		531
		532
		533	=item aio_readdir $pathname, $callback->($entries)
		534
		535	Unlike the POSIX call of the same name, C<aio_readdir> reads an entire
		536	directory (i.e. opendir + readdir + closedir). The entries will not be
		537	sorted, and will B<NOT> include the C<.> and C<..> entries.
		538
		539	The callback is passed a single argument which is either C<undef> or an
		540	array-ref with the filenames.
		541
		542
		543	=item aio_readdirx $pathname, $flags, $callback->($entries, $flags)
		544
		545	Quite similar to C<aio_readdir>, but the C<$flags> argument allows to tune
		546	behaviour and output format. In case of an error, C<$entries> will be
		547	C<undef>.
		548
		549	The flags are a combination of the following constants, ORed together (the
		550	flags will also be passed to the callback, possibly modified):
		551
		552	=over 4
		553
		554	=item IO::AIO::READDIR_DENTS
		555
		556	When this flag is off, then the callback gets an arrayref with of names
		557	only (as with C<aio_readdir>), otherwise it gets an arrayref with
		558	C<[$name, $type, $inode]> arrayrefs, each describing a single directory
		559	entry in more detail.
		560
		561	C<$name> is the name of the entry.
		562
		563	C<$type> is one of the C<IO::AIO::DT_xxx> constants:
		564
		565	C<IO::AIO::DT_UNKNOWN>, C<IO::AIO::DT_FIFO>, C<IO::AIO::DT_CHR>, C<IO::AIO::DT_DIR>,
		566	C<IO::AIO::DT_BLK>, C<IO::AIO::DT_REG>, C<IO::AIO::DT_LNK>, C<IO::AIO::DT_SOCK>,
		567	C<IO::AIO::DT_WHT>.
		568
		569	C<IO::AIO::DT_UNKNOWN> means just that: readdir does not know. If you need to
		570	know, you have to run stat yourself. Also, for speed reasons, the C<$type>
		571	scalars are read-only: you can not modify them.
		572
		573	C<$inode> is the inode number (which might not be exact on systems with 64
		574	bit inode numbers and 32 bit perls). This field has unspecified content on
		575	systems that do not deliver the inode information.
		576
		577	=item IO::AIO::READDIR_DIRS_FIRST
		578
		579	When this flag is set, then the names will be returned in an order where
		580	likely directories come first. This is useful when you need to quickly
		581	find directories, or you want to find all directories while avoiding to
		582	stat() each entry.
		583
		584	If the system returns type information in readdir, then this is used
		585	to find directories directly. Otherwise, likely directories are files
		586	beginning with ".", or otherwise files with no dots, of which files with
		587	short names are tried first.
		588
		589	=item IO::AIO::READDIR_STAT_ORDER
		590
		591	When this flag is set, then the names will be returned in an order
		592	suitable for stat()'ing each one. That is, when you plan to stat()
		593	all files in the given directory, then the returned order will likely
		594	be fastest.
		595
		596	If both this flag and C<IO::AIO::READDIR_DIRS_FIRST> are specified, then
		597	the likely dirs come first, resulting in a less optimal stat order.
		598
		599	=item IO::AIO::READDIR_FOUND_UNKNOWN
		600
		601	This flag should not be set when calling C<aio_readdirx>. Instead, it
		602	is being set by C<aio_readdirx>, when any of the C<$type>'s found were
		603	C<IO::AIO::DT_UNKNOWN>. The absense of this flag therefore indicates that all
		604	C<$type>'s are known, which can be used to speed up some algorithms.
		605
		606	=back
		607
		608
		609	=item aio_load $path, $data, $callback->($status)
		610
		611	This is a composite request that tries to fully load the given file into
		612	memory. Status is the same as with aio_read.
		613
		614	=cut
		615
		616	sub aio_load($$;$) {
		617	my ($path, undef, $cb) = @_;
		618	my $data = \$_[1];
		619
		620	my $pri = aioreq_pri;
		621	my $grp = aio_group $cb;
		622
		623	aioreq_pri $pri;
		624	add $grp aio_open $path, O_RDONLY, 0, sub {
		625	my $fh = shift
		626	or return $grp->result (-1);
		627
		628	aioreq_pri $pri;
		629	add $grp aio_read $fh, 0, (-s $fh), $$data, 0, sub {
		630	$grp->result ($_[0]);
		631	};
		632	};
		633
		634	$grp
		635	}
		636
		637	=item aio_copy $srcpath, $dstpath, $callback->($status)
		638
		639	Try to copy the I<file> (directories not supported as either source or
		640	destination) from C<$srcpath> to C<$dstpath> and call the callback with
		641	a status of C<0> (ok) or C<-1> (error, see C<$!>).
		642
		643	This is a composite request that creates the destination file with
		644	mode 0200 and copies the contents of the source file into it using
		645	C<aio_sendfile>, followed by restoring atime, mtime, access mode and
		646	uid/gid, in that order.
		647
		648	If an error occurs, the partial destination file will be unlinked, if
		649	possible, except when setting atime, mtime, access mode and uid/gid, where
		650	errors are being ignored.
		651
		652	=cut
		653
		654	sub aio_copy($$;$) {
		655	my ($src, $dst, $cb) = @_;
		656
		657	my $pri = aioreq_pri;
		658	my $grp = aio_group $cb;
		659
		660	aioreq_pri $pri;
		661	add $grp aio_open $src, O_RDONLY, 0, sub {
		662	if (my $src_fh = $_[0]) {
		663	my @stat = stat $src_fh; # hmm, might block over nfs?
		664
		665	aioreq_pri $pri;
		666	add $grp aio_open $dst, O_CREAT | O_WRONLY | O_TRUNC, 0200, sub {
		667	if (my $dst_fh = $_[0]) {
		668	aioreq_pri $pri;
		669	add $grp aio_sendfile $dst_fh, $src_fh, 0, $stat[7], sub {
		670	if ($_[0] == $stat[7]) {
		671	$grp->result (0);
		672	close $src_fh;
		673
		674	my $ch = sub {
		675	aioreq_pri $pri;
		676	add $grp aio_chmod $dst_fh, $stat[2] & 07777, sub {
		677	aioreq_pri $pri;
		678	add $grp aio_chown $dst_fh, $stat[4], $stat[5], sub {
		679	aioreq_pri $pri;
		680	add $grp aio_close $dst_fh;
		681	}
		682	};
		683	};
		684
		685	aioreq_pri $pri;
		686	add $grp aio_utime $dst_fh, $stat[8], $stat[9], sub {
		687	if ($_[0] < 0 && $! == ENOSYS) {
		688	aioreq_pri $pri;
		689	add $grp aio_utime $dst, $stat[8], $stat[9], $ch;
		690	} else {
		691	$ch->();
		692	}
		693	};
		694	} else {
		695	$grp->result (-1);
		696	close $src_fh;
		697	close $dst_fh;
		698
		699	aioreq $pri;
		700	add $grp aio_unlink $dst;
		701	}
		702	};
		703	} else {
		704	$grp->result (-1);
		705	}
		706	},
		707
		708	} else {
		709	$grp->result (-1);
		710	}
		711	};
		712
		713	$grp
		714	}
		715
		716	=item aio_move $srcpath, $dstpath, $callback->($status)
		717
		718	Try to move the I<file> (directories not supported as either source or
		719	destination) from C<$srcpath> to C<$dstpath> and call the callback with
		720	a status of C<0> (ok) or C<-1> (error, see C<$!>).
		721
		722	This is a composite request that tries to rename(2) the file first; if
		723	rename fails with C<EXDEV>, it copies the file with C<aio_copy> and, if
		724	that is successful, unlinks the C<$srcpath>.
		725
		726	=cut
		727
		728	sub aio_move($$;$) {
		729	my ($src, $dst, $cb) = @_;
		730
		731	my $pri = aioreq_pri;
		732	my $grp = aio_group $cb;
		733
		734	aioreq_pri $pri;
		735	add $grp aio_rename $src, $dst, sub {
		736	if ($_[0] && $! == EXDEV) {
		737	aioreq_pri $pri;
		738	add $grp aio_copy $src, $dst, sub {
		739	$grp->result ($_[0]);
		740
		741	if (!$_[0]) {
		742	aioreq_pri $pri;
		743	add $grp aio_unlink $src;
		744	}
		745	};
		746	} else {
		747	$grp->result ($_[0]);
		748	}
		749	};
		750
		751	$grp
		752	}
		753
		754	=item aio_scandir $path, $maxreq, $callback->($dirs, $nondirs)
		755
		756	Scans a directory (similar to C<aio_readdir>) but additionally tries to
		757	efficiently separate the entries of directory C<$path> into two sets of
		758	names, directories you can recurse into (directories), and ones you cannot
		759	recurse into (everything else, including symlinks to directories).
		760
		761	C<aio_scandir> is a composite request that creates of many sub requests_
		762	C<$maxreq> specifies the maximum number of outstanding aio requests that
		763	this function generates. If it is C<< <= 0 >>, then a suitable default
		764	will be chosen (currently 4).
		765
		766	On error, the callback is called without arguments, otherwise it receives
		767	two array-refs with path-relative entry names.
		768
		769	Example:
		770
		771	aio_scandir $dir, 0, sub {
		772	my ($dirs, $nondirs) = @_;
		773	print "real directories: @$dirs\n";
		774	print "everything else: @$nondirs\n";
		775	};
		776
		777	Implementation notes.
		778
		779	The C<aio_readdir> cannot be avoided, but C<stat()>'ing every entry can.
		780
		781	If readdir returns file type information, then this is used directly to
		782	find directories.
		783
		784	Otherwise, after reading the directory, the modification time, size etc.
		785	of the directory before and after the readdir is checked, and if they
		786	match (and isn't the current time), the link count will be used to decide
		787	how many entries are directories (if >= 2). Otherwise, no knowledge of the
		788	number of subdirectories will be assumed.
		789
		790	Then entries will be sorted into likely directories a non-initial dot
		791	currently) and likely non-directories (see C<aio_readdirx>). Then every
		792	entry plus an appended C</.> will be C<stat>'ed, likely directories first,
		793	in order of their inode numbers. If that succeeds, it assumes that the
		794	entry is a directory or a symlink to directory (which will be checked
		795	seperately). This is often faster than stat'ing the entry itself because
		796	filesystems might detect the type of the entry without reading the inode
		797	data (e.g. ext2fs filetype feature), even on systems that cannot return
		798	the filetype information on readdir.
		799
		800	If the known number of directories (link count - 2) has been reached, the
		801	rest of the entries is assumed to be non-directories.
		802
		803	This only works with certainty on POSIX (= UNIX) filesystems, which
		804	fortunately are the vast majority of filesystems around.
		805
		806	It will also likely work on non-POSIX filesystems with reduced efficiency
		807	as those tend to return 0 or 1 as link counts, which disables the
		808	directory counting heuristic.
		809
		810	=cut
		811
		812	sub aio_scandir($$;$) {
		813	my ($path, $maxreq, $cb) = @_;
		814
		815	my $pri = aioreq_pri;
		816
		817	my $grp = aio_group $cb;
		818
		819	$maxreq = 4 if $maxreq <= 0;
		820
		821	# stat once
		822	aioreq_pri $pri;
		823	add $grp aio_stat $path, sub {
		824	return $grp->result () if $_[0];
		825	my $now = time;
		826	my $hash1 = join ":", (stat _)[0,1,3,7,9];
		827
		828	# read the directory entries
		829	aioreq_pri $pri;
		830	add $grp aio_readdirx $path, READDIR_DIRS_FIRST, sub {
		831	my $entries = shift
		832	or return $grp->result ();
		833
		834	# stat the dir another time
		835	aioreq_pri $pri;
		836	add $grp aio_stat $path, sub {
		837	my $hash2 = join ":", (stat _)[0,1,3,7,9];
		838
		839	my $ndirs;
		840
		841	# take the slow route if anything looks fishy
		842	if ($hash1 ne $hash2 or (stat _)[9] == $now) {
		843	$ndirs = -1;
		844	} else {
		845	# if nlink == 2, we are finished
		846	# for non-posix-fs's, we rely on nlink < 2
		847	$ndirs = (stat _)[3] - 2
		848	or return $grp->result ([], $entries);
		849	}
		850
		851	my (@dirs, @nondirs);
		852
		853	my $statgrp = add $grp aio_group sub {
		854	$grp->result (\@dirs, \@nondirs);
		855	};
		856
		857	limit $statgrp $maxreq;
		858	feed $statgrp sub {
		859	return unless @$entries;
		860	my $entry = shift @$entries;
		861
		862	aioreq_pri $pri;
		863	add $statgrp aio_stat "$path/$entry/.", sub {
		864	if ($_[0] < 0) {
		865	push @nondirs, $entry;
		866	} else {
		867	# need to check for real directory
		868	aioreq_pri $pri;
		869	add $statgrp aio_lstat "$path/$entry", sub {
		870	if (-d _) {
		871	push @dirs, $entry;
		872
		873	unless (--$ndirs) {
		874	push @nondirs, @$entries;
		875	feed $statgrp;
		876	}
		877	} else {
		878	push @nondirs, $entry;
		879	}
		880	}
		881	}
		882	};
		883	};
		884	};
		885	};
		886	};
		887
		888	$grp
		889	}
		890
		891	=item aio_rmtree $path, $callback->($status)
		892
		893	Delete a directory tree starting (and including) C<$path>, return the
		894	status of the final C<rmdir> only. This is a composite request that
		895	uses C<aio_scandir> to recurse into and rmdir directories, and unlink
		896	everything else.
		897
		898	=cut
		899
		900	sub aio_rmtree;
		901	sub aio_rmtree($;$) {
		902	my ($path, $cb) = @_;
		903
		904	my $pri = aioreq_pri;
		905	my $grp = aio_group $cb;
		906
		907	aioreq_pri $pri;
		908	add $grp aio_scandir $path, 0, sub {
		909	my ($dirs, $nondirs) = @_;
		910
		911	my $dirgrp = aio_group sub {
		912	add $grp aio_rmdir $path, sub {
		913	$grp->result ($_[0]);
		914	};
		915	};
		916
		917	(aioreq_pri $pri), add $dirgrp aio_rmtree "$path/$_" for @$dirs;
		918	(aioreq_pri $pri), add $dirgrp aio_unlink "$path/$_" for @$nondirs;
		919
		920	add $grp $dirgrp;
		921	};
		922
		923	$grp
		924	}
		925
		926	=item aio_sync $callback->($status)
		927
		928	Asynchronously call sync and call the callback when finished.
		929
179	=item aio_fsync $fh, $callback	930	=item aio_fsync $fh, $callback->($status)
180		931
181	Asynchronously call fsync on the given filehandle and call the callback	932	Asynchronously call fsync on the given filehandle and call the callback
182	with the fsync result code.	933	with the fsync result code.
183		934
184	=item aio_fdatasync $fh, $callback	935	=item aio_fdatasync $fh, $callback->($status)
185		936
186	Asynchronously call fdatasync on the given filehandle and call the	937	Asynchronously call fdatasync on the given filehandle and call the
187	callback with the fdatasync result code.	938	callback with the fdatasync result code.
188		939
		940	If this call isn't available because your OS lacks it or it couldn't be
		941	detected, it will be emulated by calling C<fsync> instead.
		942
		943	=item aio_sync_file_range $fh, $offset, $nbytes, $flags, $callback->($status)
		944
		945	Sync the data portion of the file specified by C<$offset> and C<$length>
		946	to disk (but NOT the metadata), by calling the Linux-specific
		947	sync_file_range call. If sync_file_range is not available or it returns
		948	ENOSYS, then fdatasync or fsync is being substituted.
		949
		950	C<$flags> can be a combination of C<IO::AIO::SYNC_FILE_RANGE_WAIT_BEFORE>,
		951	C<IO::AIO::SYNC_FILE_RANGE_WRITE> and
		952	C<IO::AIO::SYNC_FILE_RANGE_WAIT_AFTER>: refer to the sync_file_range
		953	manpage for details.
		954
		955	=item aio_pathsync $path, $callback->($status)
		956
		957	This request tries to open, fsync and close the given path. This is a
		958	composite request intended to sync directories after directory operations
		959	(E.g. rename). This might not work on all operating systems or have any
		960	specific effect, but usually it makes sure that directory changes get
		961	written to disc. It works for anything that can be opened for read-only,
		962	not just directories.
		963
		964	Future versions of this function might fall back to other methods when
		965	C<fsync> on the directory fails (such as calling C<sync>).
		966
		967	Passes C<0> when everything went ok, and C<-1> on error.
		968
		969	=cut
		970
		971	sub aio_pathsync($;$) {
		972	my ($path, $cb) = @_;
		973
		974	my $pri = aioreq_pri;
		975	my $grp = aio_group $cb;
		976
		977	aioreq_pri $pri;
		978	add $grp aio_open $path, O_RDONLY, 0, sub {
		979	my ($fh) = @_;
		980	if ($fh) {
		981	aioreq_pri $pri;
		982	add $grp aio_fsync $fh, sub {
		983	$grp->result ($_[0]);
		984
		985	aioreq_pri $pri;
		986	add $grp aio_close $fh;
		987	};
		988	} else {
		989	$grp->result (-1);
		990	}
		991	};
		992
		993	$grp
		994	}
		995
		996	=item aio_msync $scalar, $offset = 0, $length = undef, flags = 0, $callback->($status)
		997
		998	This is a rather advanced IO::AIO call, which only works on mmap(2)ed
		999	scalars (see the L<Sys::Mmap> or L<Mmap> modules for details on this, note
		1000	that the scalar must only be modified in-place while an aio operation is
		1001	pending on it).
		1002
		1003	It calls the C<msync> function of your OS, if available, with the memory
		1004	area starting at C<$offset> in the string and ending C<$length> bytes
		1005	later. If C<$length> is negative, counts from the end, and if C<$length>
		1006	is C<undef>, then it goes till the end of the string. The flags can be
		1007	a combination of C<IO::AIO::MS_ASYNC>, C<IO::AIO::MS_INVALIDATE> and
		1008	C<IO::AIO::MS_SYNC>.
		1009
		1010	=item aio_mtouch $scalar, $offset = 0, $length = undef, flags = 0, $callback->($status)
		1011
		1012	This is a rather advanced IO::AIO call, which works best on mmap(2)ed
		1013	scalars.
		1014
		1015	It touches (reads or writes) all memory pages in the specified
		1016	range inside the scalar. All caveats and parameters are the same
		1017	as for C<aio_msync>, above, except for flags, which must be either
		1018	C<0> (which reads all pages and ensures they are instantiated) or
		1019	C<IO::AIO::MT_MODIFY>, which modifies the memory page s(by reading and
		1020	writing an octet from it, which dirties the page).
		1021
		1022	=item aio_group $callback->(...)
		1023
		1024	This is a very special aio request: Instead of doing something, it is a
		1025	container for other aio requests, which is useful if you want to bundle
		1026	many requests into a single, composite, request with a definite callback
		1027	and the ability to cancel the whole request with its subrequests.
		1028
		1029	Returns an object of class L<IO::AIO::GRP>. See its documentation below
		1030	for more info.
		1031
		1032	Example:
		1033
		1034	my $grp = aio_group sub {
		1035	print "all stats done\n";
		1036	};
		1037
		1038	add $grp
		1039	(aio_stat ...),
		1040	(aio_stat ...),
		1041	...;
		1042
		1043	=item aio_nop $callback->()
		1044
		1045	This is a special request - it does nothing in itself and is only used for
		1046	side effects, such as when you want to add a dummy request to a group so
		1047	that finishing the requests in the group depends on executing the given
		1048	code.
		1049
		1050	While this request does nothing, it still goes through the execution
		1051	phase and still requires a worker thread. Thus, the callback will not
		1052	be executed immediately but only after other requests in the queue have
		1053	entered their execution phase. This can be used to measure request
		1054	latency.
		1055
		1056	=item IO::AIO::aio_busy $fractional_seconds, $callback->() *NOT EXPORTED*
		1057
		1058	Mainly used for debugging and benchmarking, this aio request puts one of
		1059	the request workers to sleep for the given time.
		1060
		1061	While it is theoretically handy to have simple I/O scheduling requests
		1062	like sleep and file handle readable/writable, the overhead this creates is
		1063	immense (it blocks a thread for a long time) so do not use this function
		1064	except to put your application under artificial I/O pressure.
		1065
189	=back	1066	=back
190		1067
		1068	=head2 IO::AIO::REQ CLASS
		1069
		1070	All non-aggregate C<aio_*> functions return an object of this class when
		1071	called in non-void context.
		1072
		1073	=over 4
		1074
		1075	=item cancel $req
		1076
		1077	Cancels the request, if possible. Has the effect of skipping execution
		1078	when entering the B<execute> state and skipping calling the callback when
		1079	entering the the B<result> state, but will leave the request otherwise
		1080	untouched (with the exception of readdir). That means that requests that
		1081	currently execute will not be stopped and resources held by the request
		1082	will not be freed prematurely.
		1083
		1084	=item cb $req $callback->(...)
		1085
		1086	Replace (or simply set) the callback registered to the request.
		1087
		1088	=back
		1089
		1090	=head2 IO::AIO::GRP CLASS
		1091
		1092	This class is a subclass of L<IO::AIO::REQ>, so all its methods apply to
		1093	objects of this class, too.
		1094
		1095	A IO::AIO::GRP object is a special request that can contain multiple other
		1096	aio requests.
		1097
		1098	You create one by calling the C<aio_group> constructing function with a
		1099	callback that will be called when all contained requests have entered the
		1100	C<done> state:
		1101
		1102	my $grp = aio_group sub {
		1103	print "all requests are done\n";
		1104	};
		1105
		1106	You add requests by calling the C<add> method with one or more
		1107	C<IO::AIO::REQ> objects:
		1108
		1109	$grp->add (aio_unlink "...");
		1110
		1111	add $grp aio_stat "...", sub {
		1112	$_[0] or return $grp->result ("error");
		1113
		1114	# add another request dynamically, if first succeeded
		1115	add $grp aio_open "...", sub {
		1116	$grp->result ("ok");
		1117	};
		1118	};
		1119
		1120	This makes it very easy to create composite requests (see the source of
		1121	C<aio_move> for an application) that work and feel like simple requests.
		1122
		1123	=over 4
		1124
		1125	=item * The IO::AIO::GRP objects will be cleaned up during calls to
		1126	C<IO::AIO::poll_cb>, just like any other request.
		1127
		1128	=item * They can be canceled like any other request. Canceling will cancel not
		1129	only the request itself, but also all requests it contains.
		1130
		1131	=item * They can also can also be added to other IO::AIO::GRP objects.
		1132
		1133	=item * You must not add requests to a group from within the group callback (or
		1134	any later time).
		1135
		1136	=back
		1137
		1138	Their lifetime, simplified, looks like this: when they are empty, they
		1139	will finish very quickly. If they contain only requests that are in the
		1140	C<done> state, they will also finish. Otherwise they will continue to
		1141	exist.
		1142
		1143	That means after creating a group you have some time to add requests
		1144	(precisely before the callback has been invoked, which is only done within
		1145	the C<poll_cb>). And in the callbacks of those requests, you can add
		1146	further requests to the group. And only when all those requests have
		1147	finished will the the group itself finish.
		1148
		1149	=over 4
		1150
		1151	=item add $grp ...
		1152
		1153	=item $grp->add (...)
		1154
		1155	Add one or more requests to the group. Any type of L<IO::AIO::REQ> can
		1156	be added, including other groups, as long as you do not create circular
		1157	dependencies.
		1158
		1159	Returns all its arguments.
		1160
		1161	=item $grp->cancel_subs
		1162
		1163	Cancel all subrequests and clears any feeder, but not the group request
		1164	itself. Useful when you queued a lot of events but got a result early.
		1165
		1166	The group request will finish normally (you cannot add requests to the
		1167	group).
		1168
		1169	=item $grp->result (...)
		1170
		1171	Set the result value(s) that will be passed to the group callback when all
		1172	subrequests have finished and set the groups errno to the current value
		1173	of errno (just like calling C<errno> without an error number). By default,
		1174	no argument will be passed and errno is zero.
		1175
		1176	=item $grp->errno ([$errno])
		1177
		1178	Sets the group errno value to C<$errno>, or the current value of errno
		1179	when the argument is missing.
		1180
		1181	Every aio request has an associated errno value that is restored when
		1182	the callback is invoked. This method lets you change this value from its
		1183	default (0).
		1184
		1185	Calling C<result> will also set errno, so make sure you either set C<$!>
		1186	before the call to C<result>, or call c<errno> after it.
		1187
		1188	=item feed $grp $callback->($grp)
		1189
		1190	Sets a feeder/generator on this group: every group can have an attached
		1191	generator that generates requests if idle. The idea behind this is that,
		1192	although you could just queue as many requests as you want in a group,
		1193	this might starve other requests for a potentially long time. For example,
		1194	C<aio_scandir> might generate hundreds of thousands C<aio_stat> requests,
		1195	delaying any later requests for a long time.
		1196
		1197	To avoid this, and allow incremental generation of requests, you can
		1198	instead a group and set a feeder on it that generates those requests. The
		1199	feed callback will be called whenever there are few enough (see C<limit>,
		1200	below) requests active in the group itself and is expected to queue more
		1201	requests.
		1202
		1203	The feed callback can queue as many requests as it likes (i.e. C<add> does
		1204	not impose any limits).
		1205
		1206	If the feed does not queue more requests when called, it will be
		1207	automatically removed from the group.
		1208
		1209	If the feed limit is C<0> when this method is called, it will be set to
		1210	C<2> automatically.
		1211
		1212	Example:
		1213
		1214	# stat all files in @files, but only ever use four aio requests concurrently:
		1215
		1216	my $grp = aio_group sub { print "finished\n" };
		1217	limit $grp 4;
		1218	feed $grp sub {
		1219	my $file = pop @files
		1220	or return;
		1221
		1222	add $grp aio_stat $file, sub { ... };
		1223	};
		1224
		1225	=item limit $grp $num
		1226
		1227	Sets the feeder limit for the group: The feeder will be called whenever
		1228	the group contains less than this many requests.
		1229
		1230	Setting the limit to C<0> will pause the feeding process.
		1231
		1232	The default value for the limit is C<0>, but note that setting a feeder
		1233	automatically bumps it up to C<2>.
		1234
		1235	=back
		1236
191	=head2 SUPPORT FUNCTIONS	1237	=head2 SUPPORT FUNCTIONS
192		1238
		1239	=head3 EVENT PROCESSING AND EVENT LOOP INTEGRATION
		1240
193	=over 4	1241	=over 4
194		1242
195	=item $fileno = IO::AIO::poll_fileno	1243	=item $fileno = IO::AIO::poll_fileno
196		1244
197	Return the I<request result pipe filehandle>. This filehandle must be	1245	Return the I<request result pipe file descriptor>. This filehandle must be
198	polled for reading by some mechanism outside this module (e.g. Event	1246	polled for reading by some mechanism outside this module (e.g. EV, Glib,
199	or select, see below). If the pipe becomes readable you have to call	1247	select and so on, see below or the SYNOPSIS). If the pipe becomes readable
200	C<poll_cb> to check the results.	1248	you have to call C<poll_cb> to check the results.
201		1249
202	See C<poll_cb> for an example.	1250	See C<poll_cb> for an example.
203		1251
204	=item IO::AIO::poll_cb	1252	=item IO::AIO::poll_cb
205		1253
206	Process all outstanding events on the result pipe. You have to call this	1254	Process some outstanding events on the result pipe. You have to call this
207	regularly. Returns the number of events processed. Returns immediately	1255	regularly. Returns C<0> if all events could be processed, or C<-1> if it
208	when no events are outstanding.	1256	returned earlier for whatever reason. Returns immediately when no events
		1257	are outstanding. The amount of events processed depends on the settings of
		1258	C<IO::AIO::max_poll_req> and C<IO::AIO::max_poll_time>.
209		1259
210	You can use Event to multiplex, e.g.:	1260	If not all requests were processed for whatever reason, the filehandle
		1261	will still be ready when C<poll_cb> returns, so normally you don't have to
		1262	do anything special to have it called later.
		1263
		1264	Example: Install an Event watcher that automatically calls
		1265	IO::AIO::poll_cb with high priority (more examples can be found in the
		1266	SYNOPSIS section, at the top of this document):
211		1267
212	Event->io (fd => IO::AIO::poll_fileno,	1268	Event->io (fd => IO::AIO::poll_fileno,
213	poll => 'r', async => 1,	1269	poll => 'r', async => 1,
214	cb => \&IO::AIO::poll_cb);	1270	cb => \&IO::AIO::poll_cb);
215		1271
		1272	=item IO::AIO::max_poll_reqs $nreqs
		1273
		1274	=item IO::AIO::max_poll_time $seconds
		1275
		1276	These set the maximum number of requests (default C<0>, meaning infinity)
		1277	that are being processed by C<IO::AIO::poll_cb> in one call, respectively
		1278	the maximum amount of time (default C<0>, meaning infinity) spent in
		1279	C<IO::AIO::poll_cb> to process requests (more correctly the mininum amount
		1280	of time C<poll_cb> is allowed to use).
		1281
		1282	Setting C<max_poll_time> to a non-zero value creates an overhead of one
		1283	syscall per request processed, which is not normally a problem unless your
		1284	callbacks are really really fast or your OS is really really slow (I am
		1285	not mentioning Solaris here). Using C<max_poll_reqs> incurs no overhead.
		1286
		1287	Setting these is useful if you want to ensure some level of
		1288	interactiveness when perl is not fast enough to process all requests in
		1289	time.
		1290
		1291	For interactive programs, values such as C<0.01> to C<0.1> should be fine.
		1292
		1293	Example: Install an Event watcher that automatically calls
		1294	IO::AIO::poll_cb with low priority, to ensure that other parts of the
		1295	program get the CPU sometimes even under high AIO load.
		1296
		1297	# try not to spend much more than 0.1s in poll_cb
		1298	IO::AIO::max_poll_time 0.1;
		1299
		1300	# use a low priority so other tasks have priority
		1301	Event->io (fd => IO::AIO::poll_fileno,
		1302	poll => 'r', nice => 1,
		1303	cb => &IO::AIO::poll_cb);
		1304
216	=item IO::AIO::poll_wait	1305	=item IO::AIO::poll_wait
217		1306
		1307	If there are any outstanding requests and none of them in the result
218	Wait till the result filehandle becomes ready for reading (simply does a	1308	phase, wait till the result filehandle becomes ready for reading (simply
219	select on the filehandle. This is useful if you want to synchronously wait	1309	does a C<select> on the filehandle. This is useful if you want to
220	for some requests to finish).	1310	synchronously wait for some requests to finish).
221		1311
222	See C<nreqs> for an example.	1312	See C<nreqs> for an example.
223		1313
		1314	=item IO::AIO::poll
		1315
		1316	Waits until some requests have been handled.
		1317
		1318	Returns the number of requests processed, but is otherwise strictly
		1319	equivalent to:
		1320
		1321	IO::AIO::poll_wait, IO::AIO::poll_cb
		1322
224	=item IO::AIO::nreqs	1323	=item IO::AIO::flush
225		1324
226	Returns the number of requests currently outstanding.	1325	Wait till all outstanding AIO requests have been handled.
227		1326
228	Example: wait till there are no outstanding requests anymore:	1327	Strictly equivalent to:
229		1328
230	IO::AIO::poll_wait, IO::AIO::poll_cb	1329	IO::AIO::poll_wait, IO::AIO::poll_cb
231	while IO::AIO::nreqs;	1330	while IO::AIO::nreqs;
232		1331
		1332	=back
		1333
		1334	=head3 CONTROLLING THE NUMBER OF THREADS
		1335
		1336	=over
		1337
233	=item IO::AIO::min_parallel $nthreads	1338	=item IO::AIO::min_parallel $nthreads
234		1339
235	Set the minimum number of AIO threads to C<$nthreads>. The default is	1340	Set the minimum number of AIO threads to C<$nthreads>. The current
236	C<1>, which means a single asynchronous operation can be done at one time	1341	default is C<8>, which means eight asynchronous operations can execute
237	(the number of outstanding operations, however, is unlimited).	1342	concurrently at any one time (the number of outstanding requests,
		1343	however, is unlimited).
238		1344
		1345	IO::AIO starts threads only on demand, when an AIO request is queued and
		1346	no free thread exists. Please note that queueing up a hundred requests can
		1347	create demand for a hundred threads, even if it turns out that everything
		1348	is in the cache and could have been processed faster by a single thread.
		1349
239	It is recommended to keep the number of threads low, as some Linux	1350	It is recommended to keep the number of threads relatively low, as some
240	kernel versions will scale negatively with the number of threads (higher	1351	Linux kernel versions will scale negatively with the number of threads
241	parallelity => MUCH higher latency). With current Linux 2.6 versions, 4-32	1352	(higher parallelity => MUCH higher latency). With current Linux 2.6
242	threads should be fine.	1353	versions, 4-32 threads should be fine.
243		1354
244	Under normal circumstances you don't need to call this function, as this	1355	Under most circumstances you don't need to call this function, as the
245	module automatically starts some threads (the exact number might change,	1356	module selects a default that is suitable for low to moderate load.
246	and is currently 4).
247		1357
248	=item IO::AIO::max_parallel $nthreads	1358	=item IO::AIO::max_parallel $nthreads
249		1359
250	Sets the maximum number of AIO threads to C<$nthreads>. If more than	1360	Sets the maximum number of AIO threads to C<$nthreads>. If more than the
251	the specified number of threads are currently running, kill them. This	1361	specified number of threads are currently running, this function kills
252	function blocks until the limit is reached.	1362	them. This function blocks until the limit is reached.
		1363
		1364	While C<$nthreads> are zero, aio requests get queued but not executed
		1365	until the number of threads has been increased again.
253		1366
254	This module automatically runs C<max_parallel 0> at program end, to ensure	1367	This module automatically runs C<max_parallel 0> at program end, to ensure
255	that all threads are killed and that there are no outstanding requests.	1368	that all threads are killed and that there are no outstanding requests.
256		1369
257	Under normal circumstances you don't need to call this function.	1370	Under normal circumstances you don't need to call this function.
258		1371
		1372	=item IO::AIO::max_idle $nthreads
		1373
		1374	Limit the number of threads (default: 4) that are allowed to idle (i.e.,
		1375	threads that did not get a request to process within 10 seconds). That
		1376	means if a thread becomes idle while C<$nthreads> other threads are also
		1377	idle, it will free its resources and exit.
		1378
		1379	This is useful when you allow a large number of threads (e.g. 100 or 1000)
		1380	to allow for extremely high load situations, but want to free resources
		1381	under normal circumstances (1000 threads can easily consume 30MB of RAM).
		1382
		1383	The default is probably ok in most situations, especially if thread
		1384	creation is fast. If thread creation is very slow on your system you might
		1385	want to use larger values.
		1386
259	=item $oldnreqs = IO::AIO::max_outstanding $nreqs	1387	=item IO::AIO::max_outstanding $maxreqs
		1388
		1389	This is a very bad function to use in interactive programs because it
		1390	blocks, and a bad way to reduce concurrency because it is inexact: Better
		1391	use an C<aio_group> together with a feed callback.
260		1392
261	Sets the maximum number of outstanding requests to C<$nreqs>. If you	1393	Sets the maximum number of outstanding requests to C<$nreqs>. If you
262	try to queue up more than this number of requests, the caller will block until	1394	do queue up more than this number of requests, the next call to the
263	some requests have been handled.	1395	C<poll_cb> (and C<poll_some> and other functions calling C<poll_cb>)
		1396	function will block until the limit is no longer exceeded.
264		1397
265	The default is very large, so normally there is no practical limit. If you	1398	The default value is very large, so there is no practical limit on the
266	queue up many requests in a loop it it often improves speed if you set	1399	number of outstanding requests.
267	this to a relatively low number, such as C<100>.
268		1400
269	Under normal circumstances you don't need to call this function.	1401	You can still queue as many requests as you want. Therefore,
		1402	C<max_outstanding> is mainly useful in simple scripts (with low values) or
		1403	as a stop gap to shield against fatal memory overflow (with large values).
270		1404
271	=back	1405	=back
272		1406
		1407	=head3 STATISTICAL INFORMATION
		1408
		1409	=over
		1410
		1411	=item IO::AIO::nreqs
		1412
		1413	Returns the number of requests currently in the ready, execute or pending
		1414	states (i.e. for which their callback has not been invoked yet).
		1415
		1416	Example: wait till there are no outstanding requests anymore:
		1417
		1418	IO::AIO::poll_wait, IO::AIO::poll_cb
		1419	while IO::AIO::nreqs;
		1420
		1421	=item IO::AIO::nready
		1422
		1423	Returns the number of requests currently in the ready state (not yet
		1424	executed).
		1425
		1426	=item IO::AIO::npending
		1427
		1428	Returns the number of requests currently in the pending state (executed,
		1429	but not yet processed by poll_cb).
		1430
		1431	=back
		1432
		1433	=head3 MISCELLANEOUS FUNCTIONS
		1434
		1435	IO::AIO implements some functions that might be useful, but are not
		1436	asynchronous.
		1437
		1438	=over 4
		1439
		1440	=item IO::AIO::sendfile $ofh, $ifh, $offset, $count
		1441
		1442	Calls the C<eio_sendfile_sync> function, which is like C<aio_sendfile>,
		1443	but is blocking (this makes most sense if you know the input data is
		1444	likely cached already and the output filehandle is set to non-blocking
		1445	operations).
		1446
		1447	Returns the number of bytes copied, or C<-1> on error.
		1448
		1449	=item IO::AIO::fadvise $fh, $offset, $len, $advice
		1450
		1451	Simply calls the C<posix_fadvise> function (see it's
		1452	manpage for details). The following advice constants are
		1453	avaiable: C<IO::AIO::FADV_NORMAL>, C<IO::AIO::FADV_SEQUENTIAL>,
		1454	C<IO::AIO::FADV_RANDOM>, C<IO::AIO::FADV_NOREUSE>,
		1455	C<IO::AIO::FADV_WILLNEED>, C<IO::AIO::FADV_DONTNEED>.
		1456
		1457	On systems that do not implement C<posix_fadvise>, this function returns
		1458	ENOSYS, otherwise the return value of C<posix_fadvise>.
		1459
		1460	=back
		1461
273	=cut	1462	=cut
274		1463
275	# support function to convert a fd into a perl filehandle
276	sub _fd2fh {
277	return undef if $_[0] < 0;
278
279	# try to be perl5.6-compatible
280	local *AIO_FH;
281	open AIO_FH, "+<&=$_[0]"
282	or return undef;
283
284	*AIO_FH
285	}
286
287	min_parallel 4;	1464	min_parallel 8;
288		1465
289	END {	1466	END { flush }
290	max_parallel 0;
291	}
292		1467
293	1;	1468	1;
294		1469
		1470	=head2 FORK BEHAVIOUR
		1471
		1472	This module should do "the right thing" when the process using it forks:
		1473
		1474	Before the fork, IO::AIO enters a quiescent state where no requests
		1475	can be added in other threads and no results will be processed. After
		1476	the fork the parent simply leaves the quiescent state and continues
		1477	request/result processing, while the child frees the request/result queue
		1478	(so that the requests started before the fork will only be handled in the
		1479	parent). Threads will be started on demand until the limit set in the
		1480	parent process has been reached again.
		1481
		1482	In short: the parent will, after a short pause, continue as if fork had
		1483	not been called, while the child will act as if IO::AIO has not been used
		1484	yet.
		1485
		1486	=head2 MEMORY USAGE
		1487
		1488	Per-request usage:
		1489
		1490	Each aio request uses - depending on your architecture - around 100-200
		1491	bytes of memory. In addition, stat requests need a stat buffer (possibly
		1492	a few hundred bytes), readdir requires a result buffer and so on. Perl
		1493	scalars and other data passed into aio requests will also be locked and
		1494	will consume memory till the request has entered the done state.
		1495
		1496	This is not awfully much, so queuing lots of requests is not usually a
		1497	problem.
		1498
		1499	Per-thread usage:
		1500
		1501	In the execution phase, some aio requests require more memory for
		1502	temporary buffers, and each thread requires a stack and other data
		1503	structures (usually around 16k-128k, depending on the OS).
		1504
		1505	=head1 KNOWN BUGS
		1506
		1507	Known bugs will be fixed in the next release.
		1508
295	=head1 SEE ALSO	1509	=head1 SEE ALSO
296		1510
297	L<Coro>, L<Linux::AIO>.	1511	L<AnyEvent::AIO> for easy integration into event loops, L<Coro::AIO> for a
		1512	more natural syntax.
298		1513
299	=head1 AUTHOR	1514	=head1 AUTHOR
300		1515
301	Marc Lehmann <schmorp@schmorp.de>	1516	Marc Lehmann <schmorp@schmorp.de>
302	http://home.schmorp.de/	1517	http://home.schmorp.de/

Diff Legend

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