ViewVC Help

View File | Revision Log | Show Annotations | Download File

/cvs/IO-AIO/AIO.pm

Comparing IO-AIO/AIO.pm (file contents):
Revision 1.30 by root, Wed Aug 17 04:47:38 2005 UTC vs.
Revision 1.184 by root, Mon Nov 1 22:03:43 2010 UTC

…		…
4		4
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", IO::AIO::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
21	Event->io (fd => IO::AIO::poll_fileno,	22	use IO::AIO 2;
22	poll => 'r',
23	cb => \&IO::AIO::poll_cb);
24		23
25	# Glib/Gtk2	24	aioreq_pri 4; # give next request a very high priority
26	add_watch Glib::IO IO::AIO::poll_fileno,	25	my $req = aio_unlink "/tmp/file", sub { };
27	in => sub { IO::AIO::poll_cb; 1 };	26	$req->cancel; # cancel request if still in queue
28		27
29	# Tk	28	my $grp = aio_group sub { print "all stats done\n" };
30	Tk::Event::IO->fileevent (IO::AIO::poll_fileno, "",	29	add $grp aio_stat "..." for ...;
31	readable => \&IO::AIO::poll_cb);
32
33	# Danga::Socket
34	Danga::Socket->AddOtherFds (IO::AIO::poll_fileno =>
35	\&IO::AIO::poll_cb);
36
37		30
38	=head1 DESCRIPTION	31	=head1 DESCRIPTION
39		32
40	This module implements asynchronous I/O using whatever means your	33	This module implements asynchronous I/O using whatever means your
41	operating system supports.	34	operating system supports. It is implemented as an interface to C<libeio>
		35	(L<http://software.schmorp.de/pkg/libeio.html>).
42		36
		37	Asynchronous means that operations that can normally block your program
		38	(e.g. reading from disk) will be done asynchronously: the operation
		39	will still block, but you can do something else in the meantime. This
		40	is extremely useful for programs that need to stay interactive even
		41	when doing heavy I/O (GUI programs, high performance network servers
		42	etc.), but can also be used to easily do operations in parallel that are
		43	normally done sequentially, e.g. stat'ing many files, which is much faster
		44	on a RAID volume or over NFS when you do a number of stat operations
		45	concurrently.
		46
		47	While most of this works on all types of file descriptors (for
		48	example sockets), using these functions on file descriptors that
		49	support nonblocking operation (again, sockets, pipes etc.) is
		50	very inefficient. Use an event loop for that (such as the L<EV>
		51	module): IO::AIO will naturally fit into such an event loop itself.
		52
43	Currently, a number of threads are started that execute your read/writes	53	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	54	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	55	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	56	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	57	functions available on many operating systems. However, they are often
48	not well-supported (Linux doesn't allow them on normal files currently,	58	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	59	files currently, for example), and they would only support aio_read and
50	remaining functionality would have to be implemented using threads anyway.	60	aio_write, so the remaining functionality would have to be implemented
		61	using threads anyway.
51		62
52	Although the module will work with in the presence of other threads, it is	63	Although the module will work in the presence of other (Perl-) threads,
53	currently not reentrant, so use appropriate locking yourself, always call	64	it is currently not reentrant in any way, so use appropriate locking
54	C<poll_cb> from within the same thread, or never call C<poll_cb> (or other	65	yourself, always call C<poll_cb> from within the same thread, or never
55	C<aio_> functions) recursively.	66	call C<poll_cb> (or other C<aio_> functions) recursively.
		67
		68	=head2 EXAMPLE
		69
		70	This is a simple example that uses the EV module and loads
		71	F</etc/passwd> asynchronously:
		72
		73	use Fcntl;
		74	use EV;
		75	use IO::AIO;
		76
		77	# register the IO::AIO callback with EV
		78	my $aio_w = EV::io IO::AIO::poll_fileno, EV::READ, \&IO::AIO::poll_cb;
		79
		80	# queue the request to open /etc/passwd
		81	aio_open "/etc/passwd", IO::AIO::O_RDONLY, 0, sub {
		82	my $fh = shift
		83	or die "error while opening: $!";
		84
		85	# stat'ing filehandles is generally non-blocking
		86	my $size = -s $fh;
		87
		88	# queue a request to read the file
		89	my $contents;
		90	aio_read $fh, 0, $size, $contents, 0, sub {
		91	$_[0] == $size
		92	or die "short read: $!";
		93
		94	close $fh;
		95
		96	# file contents now in $contents
		97	print $contents;
		98
		99	# exit event loop and program
		100	EV::unloop;
		101	};
		102	};
		103
		104	# possibly queue up other requests, or open GUI windows,
		105	# check for sockets etc. etc.
		106
		107	# process events as long as there are some:
		108	EV::loop;
		109
		110	=head1 REQUEST ANATOMY AND LIFETIME
		111
		112	Every C<aio_*> function creates a request. which is a C data structure not
		113	directly visible to Perl.
		114
		115	If called in non-void context, every request function returns a Perl
		116	object representing the request. In void context, nothing is returned,
		117	which saves a bit of memory.
		118
		119	The perl object is a fairly standard ref-to-hash object. The hash contents
		120	are not used by IO::AIO so you are free to store anything you like in it.
		121
		122	During their existance, aio requests travel through the following states,
		123	in order:
		124
		125	=over 4
		126
		127	=item ready
		128
		129	Immediately after a request is created it is put into the ready state,
		130	waiting for a thread to execute it.
		131
		132	=item execute
		133
		134	A thread has accepted the request for processing and is currently
		135	executing it (e.g. blocking in read).
		136
		137	=item pending
		138
		139	The request has been executed and is waiting for result processing.
		140
		141	While request submission and execution is fully asynchronous, result
		142	processing is not and relies on the perl interpreter calling C<poll_cb>
		143	(or another function with the same effect).
		144
		145	=item result
		146
		147	The request results are processed synchronously by C<poll_cb>.
		148
		149	The C<poll_cb> function will process all outstanding aio requests by
		150	calling their callbacks, freeing memory associated with them and managing
		151	any groups they are contained in.
		152
		153	=item done
		154
		155	Request has reached the end of its lifetime and holds no resources anymore
		156	(except possibly for the Perl object, but its connection to the actual
		157	aio request is severed and calling its methods will either do nothing or
		158	result in a runtime error).
		159
		160	=back
56		161
57	=cut	162	=cut
58		163
59	package IO::AIO;	164	package IO::AIO;
60		165
61	no warnings;	166	use Carp ();
		167
		168	use common::sense;
62		169
63	use base 'Exporter';	170	use base 'Exporter';
64		171
65	use Fcntl ();
66
67	BEGIN {	172	BEGIN {
68	$VERSION = 1.2;	173	our $VERSION = '3.7';
69		174
70	@EXPORT = qw(aio_read aio_write aio_open aio_close aio_stat aio_lstat aio_unlink	175	our @AIO_REQ = qw(aio_sendfile aio_read aio_write aio_open aio_close
71	aio_rmdir aio_symlink aio_fsync aio_fdatasync aio_readahead);	176	aio_stat aio_lstat aio_unlink aio_rmdir aio_readdir aio_readdirx
72	@EXPORT_OK = qw(poll_fileno poll_cb min_parallel max_parallel max_outstanding nreqs);	177	aio_scandir aio_symlink aio_readlink aio_sync aio_fsync
		178	aio_fdatasync aio_sync_file_range aio_pathsync aio_readahead
		179	aio_rename aio_link aio_move aio_copy aio_group
		180	aio_nop aio_mknod aio_load aio_rmtree aio_mkdir aio_chown
		181	aio_chmod aio_utime aio_truncate
		182	aio_msync aio_mtouch aio_mlock aio_mlockall
		183	aio_statvfs);
		184
		185	our @EXPORT = (@AIO_REQ, qw(aioreq_pri aioreq_nice));
		186	our @EXPORT_OK = qw(poll_fileno poll_cb poll_wait flush
		187	min_parallel max_parallel max_idle
		188	nreqs nready npending nthreads
		189	max_poll_time max_poll_reqs
		190	sendfile fadvise madvise
		191	mmap munmap munlock munlockall);
		192
		193	push @AIO_REQ, qw(aio_busy); # not exported
		194
		195	@IO::AIO::GRP::ISA = 'IO::AIO::REQ';
73		196
74	require XSLoader;	197	require XSLoader;
75	XSLoader::load IO::AIO, $VERSION;	198	XSLoader::load ("IO::AIO", $VERSION);
76	}	199	}
77		200
78	=head1 FUNCTIONS	201	=head1 FUNCTIONS
79		202
		203	=head2 QUICK OVERVIEW
		204
		205	This section simply lists the prototypes of the most important functions
		206	for quick reference. See the following sections for function-by-function
		207	documentation.
		208
		209	aio_open $pathname, $flags, $mode, $callback->($fh)
		210	aio_close $fh, $callback->($status)
		211	aio_read $fh,$offset,$length, $data,$dataoffset, $callback->($retval)
		212	aio_write $fh,$offset,$length, $data,$dataoffset, $callback->($retval)
		213	aio_sendfile $out_fh, $in_fh, $in_offset, $length, $callback->($retval)
		214	aio_readahead $fh,$offset,$length, $callback->($retval)
		215	aio_stat $fh_or_path, $callback->($status)
		216	aio_lstat $fh, $callback->($status)
		217	aio_statvfs $fh_or_path, $callback->($statvfs)
		218	aio_utime $fh_or_path, $atime, $mtime, $callback->($status)
		219	aio_chown $fh_or_path, $uid, $gid, $callback->($status)
		220	aio_truncate $fh_or_path, $offset, $callback->($status)
		221	aio_chmod $fh_or_path, $mode, $callback->($status)
		222	aio_unlink $pathname, $callback->($status)
		223	aio_mknod $path, $mode, $dev, $callback->($status)
		224	aio_link $srcpath, $dstpath, $callback->($status)
		225	aio_symlink $srcpath, $dstpath, $callback->($status)
		226	aio_readlink $path, $callback->($link)
		227	aio_rename $srcpath, $dstpath, $callback->($status)
		228	aio_mkdir $pathname, $mode, $callback->($status)
		229	aio_rmdir $pathname, $callback->($status)
		230	aio_readdir $pathname, $callback->($entries)
		231	aio_readdirx $pathname, $flags, $callback->($entries, $flags)
		232	IO::AIO::READDIR_DENTS IO::AIO::READDIR_DIRS_FIRST
		233	IO::AIO::READDIR_STAT_ORDER IO::AIO::READDIR_FOUND_UNKNOWN
		234	aio_load $path, $data, $callback->($status)
		235	aio_copy $srcpath, $dstpath, $callback->($status)
		236	aio_move $srcpath, $dstpath, $callback->($status)
		237	aio_scandir $path, $maxreq, $callback->($dirs, $nondirs)
		238	aio_rmtree $path, $callback->($status)
		239	aio_sync $callback->($status)
		240	aio_fsync $fh, $callback->($status)
		241	aio_fdatasync $fh, $callback->($status)
		242	aio_sync_file_range $fh, $offset, $nbytes, $flags, $callback->($status)
		243	aio_pathsync $path, $callback->($status)
		244	aio_msync $scalar, $offset = 0, $length = undef, flags = 0, $callback->($status)
		245	aio_mtouch $scalar, $offset = 0, $length = undef, flags = 0, $callback->($status)
		246	aio_mlock $scalar, $offset = 0, $length = undef, $callback->($status)
		247	aio_mlockall $flags, $callback->($status)
		248	aio_group $callback->(...)
		249	aio_nop $callback->()
		250
		251	$prev_pri = aioreq_pri [$pri]
		252	aioreq_nice $pri_adjust
		253
		254	IO::AIO::poll_wait
		255	IO::AIO::poll_cb
		256	IO::AIO::poll
		257	IO::AIO::flush
		258	IO::AIO::max_poll_reqs $nreqs
		259	IO::AIO::max_poll_time $seconds
		260	IO::AIO::min_parallel $nthreads
		261	IO::AIO::max_parallel $nthreads
		262	IO::AIO::max_idle $nthreads
		263	IO::AIO::max_outstanding $maxreqs
		264	IO::AIO::nreqs
		265	IO::AIO::nready
		266	IO::AIO::npending
		267
		268	IO::AIO::sendfile $ofh, $ifh, $offset, $count
		269	IO::AIO::fadvise $fh, $offset, $len, $advice
		270	IO::AIO::madvise $scalar, $offset, $length, $advice
		271	IO::AIO::mprotect $scalar, $offset, $length, $protect
		272	IO::AIO::munlock $scalar, $offset = 0, $length = undef
		273	IO::AIO::munlockall
		274
80	=head2 AIO FUNCTIONS	275	=head2 AIO REQUEST FUNCTIONS
81		276
82	All the C<aio_*> calls are more or less thin wrappers around the syscall	277	All the C<aio_*> calls are more or less thin wrappers around the syscall
83	with the same name (sans C<aio_>). The arguments are similar or identical,	278	with the same name (sans C<aio_>). The arguments are similar or identical,
84	and they all accept an additional (and optional) C<$callback> argument	279	and they all accept an additional (and optional) C<$callback> argument
85	which must be a code reference. This code reference will get called with	280	which must be a code reference. This code reference will get called with
86	the syscall return code (e.g. most syscalls return C<-1> on error, unlike	281	the syscall return code (e.g. most syscalls return C<-1> on error, unlike
87	perl, which usually delivers "false") as it's sole argument when the given	282	perl, which usually delivers "false") as its sole argument after the given
88	syscall has been executed asynchronously.	283	syscall has been executed asynchronously.
89		284
90	All functions expecting a filehandle keep a copy of the filehandle	285	All functions expecting a filehandle keep a copy of the filehandle
91	internally until the request has finished.	286	internally until the request has finished.
92		287
		288	All functions return request objects of type L<IO::AIO::REQ> that allow
		289	further manipulation of those requests while they are in-flight.
		290
93	The pathnames you pass to these routines I<must> be absolute and	291	The pathnames you pass to these routines I<must> be absolute and
94	encoded in byte form. The reason for the former is that at the time the	292	encoded as octets. The reason for the former is that at the time the
95	request is being executed, the current working directory could have	293	request is being executed, the current working directory could have
96	changed. Alternatively, you can make sure that you never change the	294	changed. Alternatively, you can make sure that you never change the
97	current working directory.	295	current working directory anywhere in the program and then use relative
		296	paths.
98		297
99	To encode pathnames to byte form, either make sure you either: a)	298	To encode pathnames as octets, either make sure you either: a) always pass
100	always pass in filenames you got from outside (command line, readdir	299	in filenames you got from outside (command line, readdir etc.) without
101	etc.), b) are ASCII or ISO 8859-1, c) use the Encode module and encode	300	tinkering, b) are ASCII or ISO 8859-1, c) use the Encode module and encode
102	your pathnames to the locale (or other) encoding in effect in the user	301	your pathnames to the locale (or other) encoding in effect in the user
103	environment, d) use Glib::filename_from_unicode on unicode filenames or e)	302	environment, d) use Glib::filename_from_unicode on unicode filenames or e)
104	use something else.	303	use something else to ensure your scalar has the correct contents.
		304
		305	This works, btw. independent of the internal UTF-8 bit, which IO::AIO
		306	handles correctly whether it is set or not.
105		307
106	=over 4	308	=over 4
107		309
		310	=item $prev_pri = aioreq_pri [$pri]
		311
		312	Returns the priority value that would be used for the next request and, if
		313	C<$pri> is given, sets the priority for the next aio request.
		314
		315	The default priority is C<0>, the minimum and maximum priorities are C<-4>
		316	and C<4>, respectively. Requests with higher priority will be serviced
		317	first.
		318
		319	The priority will be reset to C<0> after each call to one of the C<aio_*>
		320	functions.
		321
		322	Example: open a file with low priority, then read something from it with
		323	higher priority so the read request is serviced before other low priority
		324	open requests (potentially spamming the cache):
		325
		326	aioreq_pri -3;
		327	aio_open ..., sub {
		328	return unless $_[0];
		329
		330	aioreq_pri -2;
		331	aio_read $_[0], ..., sub {
		332	...
		333	};
		334	};
		335
		336
		337	=item aioreq_nice $pri_adjust
		338
		339	Similar to C<aioreq_pri>, but subtracts the given value from the current
		340	priority, so the effect is cumulative.
		341
		342
108	=item aio_open $pathname, $flags, $mode, $callback	343	=item aio_open $pathname, $flags, $mode, $callback->($fh)
109		344
110	Asynchronously open or create a file and call the callback with a newly	345	Asynchronously open or create a file and call the callback with a newly
111	created filehandle for the file.	346	created filehandle for the file.
112		347
113	The pathname passed to C<aio_open> must be absolute. See API NOTES, above,	348	The pathname passed to C<aio_open> must be absolute. See API NOTES, above,
…		…
117	list. They are the same as used by C<sysopen>.	352	list. They are the same as used by C<sysopen>.
118		353
119	Likewise, C<$mode> specifies the mode of the newly created file, if it	354	Likewise, C<$mode> specifies the mode of the newly created file, if it
120	didn't exist and C<O_CREAT> has been given, just like perl's C<sysopen>,	355	didn't exist and C<O_CREAT> has been given, just like perl's C<sysopen>,
121	except that it is mandatory (i.e. use C<0> if you don't create new files,	356	except that it is mandatory (i.e. use C<0> if you don't create new files,
122	and C<0666> or C<0777> if you do).	357	and C<0666> or C<0777> if you do). Note that the C<$mode> will be modified
		358	by the umask in effect then the request is being executed, so better never
		359	change the umask.
123		360
124	Example:	361	Example:
125		362
126	aio_open "/etc/passwd", O_RDONLY, 0, sub {	363	aio_open "/etc/passwd", IO::AIO::O_RDONLY, 0, sub {
127	if ($_[0]) {	364	if ($_[0]) {
128	print "open successful, fh is $_[0]\n";	365	print "open successful, fh is $_[0]\n";
129	...	366	...
130	} else {	367	} else {
131	die "open failed: $!\n";	368	die "open failed: $!\n";
132	}	369	}
133	};	370	};
134		371
		372
135	=item aio_close $fh, $callback	373	=item aio_close $fh, $callback->($status)
136		374
137	Asynchronously close a file and call the callback with the result	375	Asynchronously close a file and call the callback with the result
138	code. I<WARNING:> although accepted, you should not pass in a perl	376	code.
139	filehandle here, as perl will likely close the file descriptor another
140	time when the filehandle is destroyed. Normally, you can safely call perls
141	C<close> or just let filehandles go out of scope.
142		377
143	This is supposed to be a bug in the API, so that might change. It's	378	Unfortunately, you can't do this to perl. Perl I<insists> very strongly on
144	therefore best to avoid this function.	379	closing the file descriptor associated with the filehandle itself.
145		380
		381	Therefore, C<aio_close> will not close the filehandle - instead it will
		382	use dup2 to overwrite the file descriptor with the write-end of a pipe
		383	(the pipe fd will be created on demand and will be cached).
		384
		385	Or in other words: the file descriptor will be closed, but it will not be
		386	free for reuse until the perl filehandle is closed.
		387
		388	=cut
		389
146	=item aio_read $fh,$offset,$length, $data,$dataoffset,$callback	390	=item aio_read $fh,$offset,$length, $data,$dataoffset, $callback->($retval)
147		391
148	=item aio_write $fh,$offset,$length, $data,$dataoffset,$callback	392	=item aio_write $fh,$offset,$length, $data,$dataoffset, $callback->($retval)
149		393
150	Reads or writes C<length> bytes from the specified C<fh> and C<offset>	394	Reads or writes C<$length> bytes from or to the specified C<$fh> and
151	into the scalar given by C<data> and offset C<dataoffset> and calls the	395	C<$offset> into the scalar given by C<$data> and offset C<$dataoffset>
152	callback without the actual number of bytes read (or -1 on error, just	396	and calls the callback without the actual number of bytes read (or -1 on
153	like the syscall).	397	error, just like the syscall).
		398
		399	C<aio_read> will, like C<sysread>, shrink or grow the C<$data> scalar to
		400	offset plus the actual number of bytes read.
		401
		402	If C<$offset> is undefined, then the current file descriptor offset will
		403	be used (and updated), otherwise the file descriptor offset will not be
		404	changed by these calls.
		405
		406	If C<$length> is undefined in C<aio_write>, use the remaining length of
		407	C<$data>.
		408
		409	If C<$dataoffset> is less than zero, it will be counted from the end of
		410	C<$data>.
		411
		412	The C<$data> scalar I<MUST NOT> be modified in any way while the request
		413	is outstanding. Modifying it can result in segfaults or World War III (if
		414	the necessary/optional hardware is installed).
154		415
155	Example: Read 15 bytes at offset 7 into scalar C<$buffer>, starting at	416	Example: Read 15 bytes at offset 7 into scalar C<$buffer>, starting at
156	offset C<0> within the scalar:	417	offset C<0> within the scalar:
157		418
158	aio_read $fh, 7, 15, $buffer, 0, sub {	419	aio_read $fh, 7, 15, $buffer, 0, sub {
159	$_[0] > 0 or die "read error: $!";	420	$_[0] > 0 or die "read error: $!";
160	print "read $_[0] bytes: <$buffer>\n";	421	print "read $_[0] bytes: <$buffer>\n";
161	};	422	};
162		423
		424
		425	=item aio_sendfile $out_fh, $in_fh, $in_offset, $length, $callback->($retval)
		426
		427	Tries to copy C<$length> bytes from C<$in_fh> to C<$out_fh>. It starts
		428	reading at byte offset C<$in_offset>, and starts writing at the current
		429	file offset of C<$out_fh>. Because of that, it is not safe to issue more
		430	than one C<aio_sendfile> per C<$out_fh>, as they will interfere with each
		431	other.
		432
		433	This call tries to make use of a native C<sendfile> syscall to provide
		434	zero-copy operation. For this to work, C<$out_fh> should refer to a
		435	socket, and C<$in_fh> should refer to an mmap'able file.
		436
		437	If a native sendfile cannot be found or it fails with C<ENOSYS>,
		438	C<ENOTSUP>, C<EOPNOTSUPP>, C<EAFNOSUPPORT>, C<EPROTOTYPE> or C<ENOTSOCK>,
		439	it will be emulated, so you can call C<aio_sendfile> on any type of
		440	filehandle regardless of the limitations of the operating system.
		441
		442	Please note, however, that C<aio_sendfile> can read more bytes from
		443	C<$in_fh> than are written, and there is no way to find out how many
		444	bytes have been read from C<aio_sendfile> alone, as C<aio_sendfile> only
		445	provides the number of bytes written to C<$out_fh>. Only if the result
		446	value equals C<$length> one can assume that C<$length> bytes have been
		447	read.
		448
		449
163	=item aio_readahead $fh,$offset,$length, $callback	450	=item aio_readahead $fh,$offset,$length, $callback->($retval)
164		451
165	C<aio_readahead> populates the page cache with data from a file so that	452	C<aio_readahead> populates the page cache with data from a file so that
166	subsequent reads from that file will not block on disk I/O. The C<$offset>	453	subsequent reads from that file will not block on disk I/O. The C<$offset>
167	argument specifies the starting point from which data is to be read and	454	argument specifies the starting point from which data is to be read and
168	C<$length> specifies the number of bytes to be read. I/O is performed in	455	C<$length> specifies the number of bytes to be read. I/O is performed in
…		…
172	file. The current file offset of the file is left unchanged.	459	file. The current file offset of the file is left unchanged.
173		460
174	If that syscall doesn't exist (likely if your OS isn't Linux) it will be	461	If that syscall doesn't exist (likely if your OS isn't Linux) it will be
175	emulated by simply reading the data, which would have a similar effect.	462	emulated by simply reading the data, which would have a similar effect.
176		463
		464
177	=item aio_stat $fh_or_path, $callback	465	=item aio_stat $fh_or_path, $callback->($status)
178		466
179	=item aio_lstat $fh, $callback	467	=item aio_lstat $fh, $callback->($status)
180		468
181	Works like perl's C<stat> or C<lstat> in void context. The callback will	469	Works like perl's C<stat> or C<lstat> in void context. The callback will
182	be called after the stat and the results will be available using C<stat _>	470	be called after the stat and the results will be available using C<stat _>
183	or C<-s _> etc...	471	or C<-s _> etc...
184		472
…		…
194	aio_stat "/etc/passwd", sub {	482	aio_stat "/etc/passwd", sub {
195	$_[0] and die "stat failed: $!";	483	$_[0] and die "stat failed: $!";
196	print "size is ", -s _, "\n";	484	print "size is ", -s _, "\n";
197	};	485	};
198		486
		487
		488	=item aio_statvfs $fh_or_path, $callback->($statvfs)
		489
		490	Works like the POSIX C<statvfs> or C<fstatvfs> syscalls, depending on
		491	whether a file handle or path was passed.
		492
		493	On success, the callback is passed a hash reference with the following
		494	members: C<bsize>, C<frsize>, C<blocks>, C<bfree>, C<bavail>, C<files>,
		495	C<ffree>, C<favail>, C<fsid>, C<flag> and C<namemax>. On failure, C<undef>
		496	is passed.
		497
		498	The following POSIX IO::AIO::ST_* constants are defined: C<ST_RDONLY> and
		499	C<ST_NOSUID>.
		500
		501	The following non-POSIX IO::AIO::ST_* flag masks are defined to
		502	their correct value when available, or to C<0> on systems that do
		503	not support them: C<ST_NODEV>, C<ST_NOEXEC>, C<ST_SYNCHRONOUS>,
		504	C<ST_MANDLOCK>, C<ST_WRITE>, C<ST_APPEND>, C<ST_IMMUTABLE>, C<ST_NOATIME>,
		505	C<ST_NODIRATIME> and C<ST_RELATIME>.
		506
		507	Example: stat C</wd> and dump out the data if successful.
		508
		509	aio_statvfs "/wd", sub {
		510	my $f = $_[0]
		511	or die "statvfs: $!";
		512
		513	use Data::Dumper;
		514	say Dumper $f;
		515	};
		516
		517	# result:
		518	{
		519	bsize => 1024,
		520	bfree => 4333064312,
		521	blocks => 10253828096,
		522	files => 2050765568,
		523	flag => 4096,
		524	favail => 2042092649,
		525	bavail => 4333064312,
		526	ffree => 2042092649,
		527	namemax => 255,
		528	frsize => 1024,
		529	fsid => 1810
		530	}
		531
		532
		533	=item aio_utime $fh_or_path, $atime, $mtime, $callback->($status)
		534
		535	Works like perl's C<utime> function (including the special case of $atime
		536	and $mtime being undef). Fractional times are supported if the underlying
		537	syscalls support them.
		538
		539	When called with a pathname, uses utimes(2) if available, otherwise
		540	utime(2). If called on a file descriptor, uses futimes(2) if available,
		541	otherwise returns ENOSYS, so this is not portable.
		542
		543	Examples:
		544
		545	# set atime and mtime to current time (basically touch(1)):
		546	aio_utime "path", undef, undef;
		547	# set atime to current time and mtime to beginning of the epoch:
		548	aio_utime "path", time, undef; # undef==0
		549
		550
		551	=item aio_chown $fh_or_path, $uid, $gid, $callback->($status)
		552
		553	Works like perl's C<chown> function, except that C<undef> for either $uid
		554	or $gid is being interpreted as "do not change" (but -1 can also be used).
		555
		556	Examples:
		557
		558	# same as "chown root path" in the shell:
		559	aio_chown "path", 0, -1;
		560	# same as above:
		561	aio_chown "path", 0, undef;
		562
		563
		564	=item aio_truncate $fh_or_path, $offset, $callback->($status)
		565
		566	Works like truncate(2) or ftruncate(2).
		567
		568
		569	=item aio_chmod $fh_or_path, $mode, $callback->($status)
		570
		571	Works like perl's C<chmod> function.
		572
		573
199	=item aio_unlink $pathname, $callback	574	=item aio_unlink $pathname, $callback->($status)
200		575
201	Asynchronously unlink (delete) a file and call the callback with the	576	Asynchronously unlink (delete) a file and call the callback with the
202	result code.	577	result code.
203		578
		579
		580	=item aio_mknod $path, $mode, $dev, $callback->($status)
		581
		582	[EXPERIMENTAL]
		583
		584	Asynchronously create a device node (or fifo). See mknod(2).
		585
		586	The only (POSIX-) portable way of calling this function is:
		587
		588	aio_mknod $path, IO::AIO::S_IFIFO | $mode, 0, sub { ...
		589
		590
		591	=item aio_link $srcpath, $dstpath, $callback->($status)
		592
		593	Asynchronously create a new link to the existing object at C<$srcpath> at
		594	the path C<$dstpath> and call the callback with the result code.
		595
		596
		597	=item aio_symlink $srcpath, $dstpath, $callback->($status)
		598
		599	Asynchronously create a new symbolic link to the existing object at C<$srcpath> at
		600	the path C<$dstpath> and call the callback with the result code.
		601
		602
		603	=item aio_readlink $path, $callback->($link)
		604
		605	Asynchronously read the symlink specified by C<$path> and pass it to
		606	the callback. If an error occurs, nothing or undef gets passed to the
		607	callback.
		608
		609
		610	=item aio_rename $srcpath, $dstpath, $callback->($status)
		611
		612	Asynchronously rename the object at C<$srcpath> to C<$dstpath>, just as
		613	rename(2) and call the callback with the result code.
		614
		615
		616	=item aio_mkdir $pathname, $mode, $callback->($status)
		617
		618	Asynchronously mkdir (create) a directory and call the callback with
		619	the result code. C<$mode> will be modified by the umask at the time the
		620	request is executed, so do not change your umask.
		621
		622
204	=item aio_rmdir $pathname, $callback	623	=item aio_rmdir $pathname, $callback->($status)
205		624
206	Asynchronously rmdir (delete) a directory and call the callback with the	625	Asynchronously rmdir (delete) a directory and call the callback with the
207	result code.	626	result code.
208		627
		628
		629	=item aio_readdir $pathname, $callback->($entries)
		630
		631	Unlike the POSIX call of the same name, C<aio_readdir> reads an entire
		632	directory (i.e. opendir + readdir + closedir). The entries will not be
		633	sorted, and will B<NOT> include the C<.> and C<..> entries.
		634
		635	The callback is passed a single argument which is either C<undef> or an
		636	array-ref with the filenames.
		637
		638
		639	=item aio_readdirx $pathname, $flags, $callback->($entries, $flags)
		640
		641	Quite similar to C<aio_readdir>, but the C<$flags> argument allows to tune
		642	behaviour and output format. In case of an error, C<$entries> will be
		643	C<undef>.
		644
		645	The flags are a combination of the following constants, ORed together (the
		646	flags will also be passed to the callback, possibly modified):
		647
		648	=over 4
		649
		650	=item IO::AIO::READDIR_DENTS
		651
		652	When this flag is off, then the callback gets an arrayref with of names
		653	only (as with C<aio_readdir>), otherwise it gets an arrayref with
		654	C<[$name, $type, $inode]> arrayrefs, each describing a single directory
		655	entry in more detail.
		656
		657	C<$name> is the name of the entry.
		658
		659	C<$type> is one of the C<IO::AIO::DT_xxx> constants:
		660
		661	C<IO::AIO::DT_UNKNOWN>, C<IO::AIO::DT_FIFO>, C<IO::AIO::DT_CHR>, C<IO::AIO::DT_DIR>,
		662	C<IO::AIO::DT_BLK>, C<IO::AIO::DT_REG>, C<IO::AIO::DT_LNK>, C<IO::AIO::DT_SOCK>,
		663	C<IO::AIO::DT_WHT>.
		664
		665	C<IO::AIO::DT_UNKNOWN> means just that: readdir does not know. If you need to
		666	know, you have to run stat yourself. Also, for speed reasons, the C<$type>
		667	scalars are read-only: you can not modify them.
		668
		669	C<$inode> is the inode number (which might not be exact on systems with 64
		670	bit inode numbers and 32 bit perls). This field has unspecified content on
		671	systems that do not deliver the inode information.
		672
		673	=item IO::AIO::READDIR_DIRS_FIRST
		674
		675	When this flag is set, then the names will be returned in an order where
		676	likely directories come first. This is useful when you need to quickly
		677	find directories, or you want to find all directories while avoiding to
		678	stat() each entry.
		679
		680	If the system returns type information in readdir, then this is used
		681	to find directories directly. Otherwise, likely directories are files
		682	beginning with ".", or otherwise files with no dots, of which files with
		683	short names are tried first.
		684
		685	=item IO::AIO::READDIR_STAT_ORDER
		686
		687	When this flag is set, then the names will be returned in an order
		688	suitable for stat()'ing each one. That is, when you plan to stat()
		689	all files in the given directory, then the returned order will likely
		690	be fastest.
		691
		692	If both this flag and C<IO::AIO::READDIR_DIRS_FIRST> are specified, then
		693	the likely dirs come first, resulting in a less optimal stat order.
		694
		695	=item IO::AIO::READDIR_FOUND_UNKNOWN
		696
		697	This flag should not be set when calling C<aio_readdirx>. Instead, it
		698	is being set by C<aio_readdirx>, when any of the C<$type>'s found were
		699	C<IO::AIO::DT_UNKNOWN>. The absense of this flag therefore indicates that all
		700	C<$type>'s are known, which can be used to speed up some algorithms.
		701
		702	=back
		703
		704
		705	=item aio_load $path, $data, $callback->($status)
		706
		707	This is a composite request that tries to fully load the given file into
		708	memory. Status is the same as with aio_read.
		709
		710	=cut
		711
		712	sub aio_load($$;$) {
		713	my ($path, undef, $cb) = @_;
		714	my $data = \$_[1];
		715
		716	my $pri = aioreq_pri;
		717	my $grp = aio_group $cb;
		718
		719	aioreq_pri $pri;
		720	add $grp aio_open $path, O_RDONLY, 0, sub {
		721	my $fh = shift
		722	or return $grp->result (-1);
		723
		724	aioreq_pri $pri;
		725	add $grp aio_read $fh, 0, (-s $fh), $$data, 0, sub {
		726	$grp->result ($_[0]);
		727	};
		728	};
		729
		730	$grp
		731	}
		732
		733	=item aio_copy $srcpath, $dstpath, $callback->($status)
		734
		735	Try to copy the I<file> (directories not supported as either source or
		736	destination) from C<$srcpath> to C<$dstpath> and call the callback with
		737	a status of C<0> (ok) or C<-1> (error, see C<$!>).
		738
		739	This is a composite request that creates the destination file with
		740	mode 0200 and copies the contents of the source file into it using
		741	C<aio_sendfile>, followed by restoring atime, mtime, access mode and
		742	uid/gid, in that order.
		743
		744	If an error occurs, the partial destination file will be unlinked, if
		745	possible, except when setting atime, mtime, access mode and uid/gid, where
		746	errors are being ignored.
		747
		748	=cut
		749
		750	sub aio_copy($$;$) {
		751	my ($src, $dst, $cb) = @_;
		752
		753	my $pri = aioreq_pri;
		754	my $grp = aio_group $cb;
		755
		756	aioreq_pri $pri;
		757	add $grp aio_open $src, O_RDONLY, 0, sub {
		758	if (my $src_fh = $_[0]) {
		759	my @stat = stat $src_fh; # hmm, might block over nfs?
		760
		761	aioreq_pri $pri;
		762	add $grp aio_open $dst, O_CREAT | O_WRONLY | O_TRUNC, 0200, sub {
		763	if (my $dst_fh = $_[0]) {
		764	aioreq_pri $pri;
		765	add $grp aio_sendfile $dst_fh, $src_fh, 0, $stat[7], sub {
		766	if ($_[0] == $stat[7]) {
		767	$grp->result (0);
		768	close $src_fh;
		769
		770	my $ch = sub {
		771	aioreq_pri $pri;
		772	add $grp aio_chmod $dst_fh, $stat[2] & 07777, sub {
		773	aioreq_pri $pri;
		774	add $grp aio_chown $dst_fh, $stat[4], $stat[5], sub {
		775	aioreq_pri $pri;
		776	add $grp aio_close $dst_fh;
		777	}
		778	};
		779	};
		780
		781	aioreq_pri $pri;
		782	add $grp aio_utime $dst_fh, $stat[8], $stat[9], sub {
		783	if ($_[0] < 0 && $! == ENOSYS) {
		784	aioreq_pri $pri;
		785	add $grp aio_utime $dst, $stat[8], $stat[9], $ch;
		786	} else {
		787	$ch->();
		788	}
		789	};
		790	} else {
		791	$grp->result (-1);
		792	close $src_fh;
		793	close $dst_fh;
		794
		795	aioreq $pri;
		796	add $grp aio_unlink $dst;
		797	}
		798	};
		799	} else {
		800	$grp->result (-1);
		801	}
		802	},
		803
		804	} else {
		805	$grp->result (-1);
		806	}
		807	};
		808
		809	$grp
		810	}
		811
		812	=item aio_move $srcpath, $dstpath, $callback->($status)
		813
		814	Try to move the I<file> (directories not supported as either source or
		815	destination) from C<$srcpath> to C<$dstpath> and call the callback with
		816	a status of C<0> (ok) or C<-1> (error, see C<$!>).
		817
		818	This is a composite request that tries to rename(2) the file first; if
		819	rename fails with C<EXDEV>, it copies the file with C<aio_copy> and, if
		820	that is successful, unlinks the C<$srcpath>.
		821
		822	=cut
		823
		824	sub aio_move($$;$) {
		825	my ($src, $dst, $cb) = @_;
		826
		827	my $pri = aioreq_pri;
		828	my $grp = aio_group $cb;
		829
		830	aioreq_pri $pri;
		831	add $grp aio_rename $src, $dst, sub {
		832	if ($_[0] && $! == EXDEV) {
		833	aioreq_pri $pri;
		834	add $grp aio_copy $src, $dst, sub {
		835	$grp->result ($_[0]);
		836
		837	if (!$_[0]) {
		838	aioreq_pri $pri;
		839	add $grp aio_unlink $src;
		840	}
		841	};
		842	} else {
		843	$grp->result ($_[0]);
		844	}
		845	};
		846
		847	$grp
		848	}
		849
		850	=item aio_scandir $path, $maxreq, $callback->($dirs, $nondirs)
		851
		852	Scans a directory (similar to C<aio_readdir>) but additionally tries to
		853	efficiently separate the entries of directory C<$path> into two sets of
		854	names, directories you can recurse into (directories), and ones you cannot
		855	recurse into (everything else, including symlinks to directories).
		856
		857	C<aio_scandir> is a composite request that creates of many sub requests_
		858	C<$maxreq> specifies the maximum number of outstanding aio requests that
		859	this function generates. If it is C<< <= 0 >>, then a suitable default
		860	will be chosen (currently 4).
		861
		862	On error, the callback is called without arguments, otherwise it receives
		863	two array-refs with path-relative entry names.
		864
		865	Example:
		866
		867	aio_scandir $dir, 0, sub {
		868	my ($dirs, $nondirs) = @_;
		869	print "real directories: @$dirs\n";
		870	print "everything else: @$nondirs\n";
		871	};
		872
		873	Implementation notes.
		874
		875	The C<aio_readdir> cannot be avoided, but C<stat()>'ing every entry can.
		876
		877	If readdir returns file type information, then this is used directly to
		878	find directories.
		879
		880	Otherwise, after reading the directory, the modification time, size etc.
		881	of the directory before and after the readdir is checked, and if they
		882	match (and isn't the current time), the link count will be used to decide
		883	how many entries are directories (if >= 2). Otherwise, no knowledge of the
		884	number of subdirectories will be assumed.
		885
		886	Then entries will be sorted into likely directories a non-initial dot
		887	currently) and likely non-directories (see C<aio_readdirx>). Then every
		888	entry plus an appended C</.> will be C<stat>'ed, likely directories first,
		889	in order of their inode numbers. If that succeeds, it assumes that the
		890	entry is a directory or a symlink to directory (which will be checked
		891	seperately). This is often faster than stat'ing the entry itself because
		892	filesystems might detect the type of the entry without reading the inode
		893	data (e.g. ext2fs filetype feature), even on systems that cannot return
		894	the filetype information on readdir.
		895
		896	If the known number of directories (link count - 2) has been reached, the
		897	rest of the entries is assumed to be non-directories.
		898
		899	This only works with certainty on POSIX (= UNIX) filesystems, which
		900	fortunately are the vast majority of filesystems around.
		901
		902	It will also likely work on non-POSIX filesystems with reduced efficiency
		903	as those tend to return 0 or 1 as link counts, which disables the
		904	directory counting heuristic.
		905
		906	=cut
		907
		908	sub aio_scandir($$;$) {
		909	my ($path, $maxreq, $cb) = @_;
		910
		911	my $pri = aioreq_pri;
		912
		913	my $grp = aio_group $cb;
		914
		915	$maxreq = 4 if $maxreq <= 0;
		916
		917	# stat once
		918	aioreq_pri $pri;
		919	add $grp aio_stat $path, sub {
		920	return $grp->result () if $_[0];
		921	my $now = time;
		922	my $hash1 = join ":", (stat _)[0,1,3,7,9];
		923
		924	# read the directory entries
		925	aioreq_pri $pri;
		926	add $grp aio_readdirx $path, READDIR_DIRS_FIRST, sub {
		927	my $entries = shift
		928	or return $grp->result ();
		929
		930	# stat the dir another time
		931	aioreq_pri $pri;
		932	add $grp aio_stat $path, sub {
		933	my $hash2 = join ":", (stat _)[0,1,3,7,9];
		934
		935	my $ndirs;
		936
		937	# take the slow route if anything looks fishy
		938	if ($hash1 ne $hash2 or (stat _)[9] == $now) {
		939	$ndirs = -1;
		940	} else {
		941	# if nlink == 2, we are finished
		942	# for non-posix-fs's, we rely on nlink < 2
		943	$ndirs = (stat _)[3] - 2
		944	or return $grp->result ([], $entries);
		945	}
		946
		947	my (@dirs, @nondirs);
		948
		949	my $statgrp = add $grp aio_group sub {
		950	$grp->result (\@dirs, \@nondirs);
		951	};
		952
		953	limit $statgrp $maxreq;
		954	feed $statgrp sub {
		955	return unless @$entries;
		956	my $entry = shift @$entries;
		957
		958	aioreq_pri $pri;
		959	add $statgrp aio_stat "$path/$entry/.", sub {
		960	if ($_[0] < 0) {
		961	push @nondirs, $entry;
		962	} else {
		963	# need to check for real directory
		964	aioreq_pri $pri;
		965	add $statgrp aio_lstat "$path/$entry", sub {
		966	if (-d _) {
		967	push @dirs, $entry;
		968
		969	unless (--$ndirs) {
		970	push @nondirs, @$entries;
		971	feed $statgrp;
		972	}
		973	} else {
		974	push @nondirs, $entry;
		975	}
		976	}
		977	}
		978	};
		979	};
		980	};
		981	};
		982	};
		983
		984	$grp
		985	}
		986
		987	=item aio_rmtree $path, $callback->($status)
		988
		989	Delete a directory tree starting (and including) C<$path>, return the
		990	status of the final C<rmdir> only. This is a composite request that
		991	uses C<aio_scandir> to recurse into and rmdir directories, and unlink
		992	everything else.
		993
		994	=cut
		995
		996	sub aio_rmtree;
		997	sub aio_rmtree($;$) {
		998	my ($path, $cb) = @_;
		999
		1000	my $pri = aioreq_pri;
		1001	my $grp = aio_group $cb;
		1002
		1003	aioreq_pri $pri;
		1004	add $grp aio_scandir $path, 0, sub {
		1005	my ($dirs, $nondirs) = @_;
		1006
		1007	my $dirgrp = aio_group sub {
		1008	add $grp aio_rmdir $path, sub {
		1009	$grp->result ($_[0]);
		1010	};
		1011	};
		1012
		1013	(aioreq_pri $pri), add $dirgrp aio_rmtree "$path/$_" for @$dirs;
		1014	(aioreq_pri $pri), add $dirgrp aio_unlink "$path/$_" for @$nondirs;
		1015
		1016	add $grp $dirgrp;
		1017	};
		1018
		1019	$grp
		1020	}
		1021
		1022	=item aio_sync $callback->($status)
		1023
		1024	Asynchronously call sync and call the callback when finished.
		1025
209	=item aio_fsync $fh, $callback	1026	=item aio_fsync $fh, $callback->($status)
210		1027
211	Asynchronously call fsync on the given filehandle and call the callback	1028	Asynchronously call fsync on the given filehandle and call the callback
212	with the fsync result code.	1029	with the fsync result code.
213		1030
214	=item aio_fdatasync $fh, $callback	1031	=item aio_fdatasync $fh, $callback->($status)
215		1032
216	Asynchronously call fdatasync on the given filehandle and call the	1033	Asynchronously call fdatasync on the given filehandle and call the
217	callback with the fdatasync result code.	1034	callback with the fdatasync result code.
218		1035
219	If this call isn't available because your OS lacks it or it couldn't be	1036	If this call isn't available because your OS lacks it or it couldn't be
220	detected, it will be emulated by calling C<fsync> instead.	1037	detected, it will be emulated by calling C<fsync> instead.
221		1038
		1039	=item aio_sync_file_range $fh, $offset, $nbytes, $flags, $callback->($status)
		1040
		1041	Sync the data portion of the file specified by C<$offset> and C<$length>
		1042	to disk (but NOT the metadata), by calling the Linux-specific
		1043	sync_file_range call. If sync_file_range is not available or it returns
		1044	ENOSYS, then fdatasync or fsync is being substituted.
		1045
		1046	C<$flags> can be a combination of C<IO::AIO::SYNC_FILE_RANGE_WAIT_BEFORE>,
		1047	C<IO::AIO::SYNC_FILE_RANGE_WRITE> and
		1048	C<IO::AIO::SYNC_FILE_RANGE_WAIT_AFTER>: refer to the sync_file_range
		1049	manpage for details.
		1050
		1051	=item aio_pathsync $path, $callback->($status)
		1052
		1053	This request tries to open, fsync and close the given path. This is a
		1054	composite request intended to sync directories after directory operations
		1055	(E.g. rename). This might not work on all operating systems or have any
		1056	specific effect, but usually it makes sure that directory changes get
		1057	written to disc. It works for anything that can be opened for read-only,
		1058	not just directories.
		1059
		1060	Future versions of this function might fall back to other methods when
		1061	C<fsync> on the directory fails (such as calling C<sync>).
		1062
		1063	Passes C<0> when everything went ok, and C<-1> on error.
		1064
		1065	=cut
		1066
		1067	sub aio_pathsync($;$) {
		1068	my ($path, $cb) = @_;
		1069
		1070	my $pri = aioreq_pri;
		1071	my $grp = aio_group $cb;
		1072
		1073	aioreq_pri $pri;
		1074	add $grp aio_open $path, O_RDONLY, 0, sub {
		1075	my ($fh) = @_;
		1076	if ($fh) {
		1077	aioreq_pri $pri;
		1078	add $grp aio_fsync $fh, sub {
		1079	$grp->result ($_[0]);
		1080
		1081	aioreq_pri $pri;
		1082	add $grp aio_close $fh;
		1083	};
		1084	} else {
		1085	$grp->result (-1);
		1086	}
		1087	};
		1088
		1089	$grp
		1090	}
		1091
		1092	=item aio_msync $scalar, $offset = 0, $length = undef, flags = 0, $callback->($status)
		1093
		1094	This is a rather advanced IO::AIO call, which only works on mmap(2)ed
		1095	scalars (see the C<IO::AIO::mmap> function, although it also works on data
		1096	scalars managed by the L<Sys::Mmap> or L<Mmap> modules, note that the
		1097	scalar must only be modified in-place while an aio operation is pending on
		1098	it).
		1099
		1100	It calls the C<msync> function of your OS, if available, with the memory
		1101	area starting at C<$offset> in the string and ending C<$length> bytes
		1102	later. If C<$length> is negative, counts from the end, and if C<$length>
		1103	is C<undef>, then it goes till the end of the string. The flags can be
		1104	a combination of C<IO::AIO::MS_ASYNC>, C<IO::AIO::MS_INVALIDATE> and
		1105	C<IO::AIO::MS_SYNC>.
		1106
		1107	=item aio_mtouch $scalar, $offset = 0, $length = undef, flags = 0, $callback->($status)
		1108
		1109	This is a rather advanced IO::AIO call, which works best on mmap(2)ed
		1110	scalars.
		1111
		1112	It touches (reads or writes) all memory pages in the specified
		1113	range inside the scalar. All caveats and parameters are the same
		1114	as for C<aio_msync>, above, except for flags, which must be either
		1115	C<0> (which reads all pages and ensures they are instantiated) or
		1116	C<IO::AIO::MT_MODIFY>, which modifies the memory page s(by reading and
		1117	writing an octet from it, which dirties the page).
		1118
		1119	=item aio_mlock $scalar, $offset = 0, $length = undef, $callback->($status)
		1120
		1121	This is a rather advanced IO::AIO call, which works best on mmap(2)ed
		1122	scalars.
		1123
		1124	It reads in all the pages of the underlying storage into memory (if any)
		1125	and locks them, so they are not getting swapped/paged out or removed.
		1126
		1127	If C<$length> is undefined, then the scalar will be locked till the end.
		1128
		1129	On systems that do not implement C<mlock>, this function returns C<-1>
		1130	and sets errno to C<ENOSYS>.
		1131
		1132	Note that the corresponding C<munlock> is synchronous and is
		1133	documented under L<MISCELLANEOUS FUNCTIONS>.
		1134
		1135	Example: open a file, mmap and mlock it - both will be undone when
		1136	C<$data> gets destroyed.
		1137
		1138	open my $fh, "<", $path or die "$path: $!";
		1139	my $data;
		1140	IO::AIO::mmap $data, -s $fh, IO::AIO::PROT_READ, IO::AIO::MAP_SHARED, $fh;
		1141	aio_mlock $data; # mlock in background
		1142
		1143	=item aio_mlockall $flags, $callback->($status)
		1144
		1145	Calls the C<mlockall> function with the given C<$flags> (a combination of
		1146	C<IO::AIO::MCL_CURRENT> and C<IO::AIO::MCL_FUTURE>).
		1147
		1148	On systems that do not implement C<mlockall>, this function returns C<-1>
		1149	and sets errno to C<ENOSYS>.
		1150
		1151	Note that the corresponding C<munlockall> is synchronous and is
		1152	documented under L<MISCELLANEOUS FUNCTIONS>.
		1153
		1154	Example: asynchronously lock all current and future pages into memory.
		1155
		1156	aio_mlockall IO::AIO::MCL_FUTURE;
		1157
		1158	=item aio_group $callback->(...)
		1159
		1160	This is a very special aio request: Instead of doing something, it is a
		1161	container for other aio requests, which is useful if you want to bundle
		1162	many requests into a single, composite, request with a definite callback
		1163	and the ability to cancel the whole request with its subrequests.
		1164
		1165	Returns an object of class L<IO::AIO::GRP>. See its documentation below
		1166	for more info.
		1167
		1168	Example:
		1169
		1170	my $grp = aio_group sub {
		1171	print "all stats done\n";
		1172	};
		1173
		1174	add $grp
		1175	(aio_stat ...),
		1176	(aio_stat ...),
		1177	...;
		1178
		1179	=item aio_nop $callback->()
		1180
		1181	This is a special request - it does nothing in itself and is only used for
		1182	side effects, such as when you want to add a dummy request to a group so
		1183	that finishing the requests in the group depends on executing the given
		1184	code.
		1185
		1186	While this request does nothing, it still goes through the execution
		1187	phase and still requires a worker thread. Thus, the callback will not
		1188	be executed immediately but only after other requests in the queue have
		1189	entered their execution phase. This can be used to measure request
		1190	latency.
		1191
		1192	=item IO::AIO::aio_busy $fractional_seconds, $callback->() *NOT EXPORTED*
		1193
		1194	Mainly used for debugging and benchmarking, this aio request puts one of
		1195	the request workers to sleep for the given time.
		1196
		1197	While it is theoretically handy to have simple I/O scheduling requests
		1198	like sleep and file handle readable/writable, the overhead this creates is
		1199	immense (it blocks a thread for a long time) so do not use this function
		1200	except to put your application under artificial I/O pressure.
		1201
222	=back	1202	=back
223		1203
		1204	=head2 IO::AIO::REQ CLASS
		1205
		1206	All non-aggregate C<aio_*> functions return an object of this class when
		1207	called in non-void context.
		1208
		1209	=over 4
		1210
		1211	=item cancel $req
		1212
		1213	Cancels the request, if possible. Has the effect of skipping execution
		1214	when entering the B<execute> state and skipping calling the callback when
		1215	entering the the B<result> state, but will leave the request otherwise
		1216	untouched (with the exception of readdir). That means that requests that
		1217	currently execute will not be stopped and resources held by the request
		1218	will not be freed prematurely.
		1219
		1220	=item cb $req $callback->(...)
		1221
		1222	Replace (or simply set) the callback registered to the request.
		1223
		1224	=back
		1225
		1226	=head2 IO::AIO::GRP CLASS
		1227
		1228	This class is a subclass of L<IO::AIO::REQ>, so all its methods apply to
		1229	objects of this class, too.
		1230
		1231	A IO::AIO::GRP object is a special request that can contain multiple other
		1232	aio requests.
		1233
		1234	You create one by calling the C<aio_group> constructing function with a
		1235	callback that will be called when all contained requests have entered the
		1236	C<done> state:
		1237
		1238	my $grp = aio_group sub {
		1239	print "all requests are done\n";
		1240	};
		1241
		1242	You add requests by calling the C<add> method with one or more
		1243	C<IO::AIO::REQ> objects:
		1244
		1245	$grp->add (aio_unlink "...");
		1246
		1247	add $grp aio_stat "...", sub {
		1248	$_[0] or return $grp->result ("error");
		1249
		1250	# add another request dynamically, if first succeeded
		1251	add $grp aio_open "...", sub {
		1252	$grp->result ("ok");
		1253	};
		1254	};
		1255
		1256	This makes it very easy to create composite requests (see the source of
		1257	C<aio_move> for an application) that work and feel like simple requests.
		1258
		1259	=over 4
		1260
		1261	=item * The IO::AIO::GRP objects will be cleaned up during calls to
		1262	C<IO::AIO::poll_cb>, just like any other request.
		1263
		1264	=item * They can be canceled like any other request. Canceling will cancel not
		1265	only the request itself, but also all requests it contains.
		1266
		1267	=item * They can also can also be added to other IO::AIO::GRP objects.
		1268
		1269	=item * You must not add requests to a group from within the group callback (or
		1270	any later time).
		1271
		1272	=back
		1273
		1274	Their lifetime, simplified, looks like this: when they are empty, they
		1275	will finish very quickly. If they contain only requests that are in the
		1276	C<done> state, they will also finish. Otherwise they will continue to
		1277	exist.
		1278
		1279	That means after creating a group you have some time to add requests
		1280	(precisely before the callback has been invoked, which is only done within
		1281	the C<poll_cb>). And in the callbacks of those requests, you can add
		1282	further requests to the group. And only when all those requests have
		1283	finished will the the group itself finish.
		1284
		1285	=over 4
		1286
		1287	=item add $grp ...
		1288
		1289	=item $grp->add (...)
		1290
		1291	Add one or more requests to the group. Any type of L<IO::AIO::REQ> can
		1292	be added, including other groups, as long as you do not create circular
		1293	dependencies.
		1294
		1295	Returns all its arguments.
		1296
		1297	=item $grp->cancel_subs
		1298
		1299	Cancel all subrequests and clears any feeder, but not the group request
		1300	itself. Useful when you queued a lot of events but got a result early.
		1301
		1302	The group request will finish normally (you cannot add requests to the
		1303	group).
		1304
		1305	=item $grp->result (...)
		1306
		1307	Set the result value(s) that will be passed to the group callback when all
		1308	subrequests have finished and set the groups errno to the current value
		1309	of errno (just like calling C<errno> without an error number). By default,
		1310	no argument will be passed and errno is zero.
		1311
		1312	=item $grp->errno ([$errno])
		1313
		1314	Sets the group errno value to C<$errno>, or the current value of errno
		1315	when the argument is missing.
		1316
		1317	Every aio request has an associated errno value that is restored when
		1318	the callback is invoked. This method lets you change this value from its
		1319	default (0).
		1320
		1321	Calling C<result> will also set errno, so make sure you either set C<$!>
		1322	before the call to C<result>, or call c<errno> after it.
		1323
		1324	=item feed $grp $callback->($grp)
		1325
		1326	Sets a feeder/generator on this group: every group can have an attached
		1327	generator that generates requests if idle. The idea behind this is that,
		1328	although you could just queue as many requests as you want in a group,
		1329	this might starve other requests for a potentially long time. For example,
		1330	C<aio_scandir> might generate hundreds of thousands C<aio_stat> requests,
		1331	delaying any later requests for a long time.
		1332
		1333	To avoid this, and allow incremental generation of requests, you can
		1334	instead a group and set a feeder on it that generates those requests. The
		1335	feed callback will be called whenever there are few enough (see C<limit>,
		1336	below) requests active in the group itself and is expected to queue more
		1337	requests.
		1338
		1339	The feed callback can queue as many requests as it likes (i.e. C<add> does
		1340	not impose any limits).
		1341
		1342	If the feed does not queue more requests when called, it will be
		1343	automatically removed from the group.
		1344
		1345	If the feed limit is C<0> when this method is called, it will be set to
		1346	C<2> automatically.
		1347
		1348	Example:
		1349
		1350	# stat all files in @files, but only ever use four aio requests concurrently:
		1351
		1352	my $grp = aio_group sub { print "finished\n" };
		1353	limit $grp 4;
		1354	feed $grp sub {
		1355	my $file = pop @files
		1356	or return;
		1357
		1358	add $grp aio_stat $file, sub { ... };
		1359	};
		1360
		1361	=item limit $grp $num
		1362
		1363	Sets the feeder limit for the group: The feeder will be called whenever
		1364	the group contains less than this many requests.
		1365
		1366	Setting the limit to C<0> will pause the feeding process.
		1367
		1368	The default value for the limit is C<0>, but note that setting a feeder
		1369	automatically bumps it up to C<2>.
		1370
		1371	=back
		1372
224	=head2 SUPPORT FUNCTIONS	1373	=head2 SUPPORT FUNCTIONS
225		1374
		1375	=head3 EVENT PROCESSING AND EVENT LOOP INTEGRATION
		1376
226	=over 4	1377	=over 4
227		1378
228	=item $fileno = IO::AIO::poll_fileno	1379	=item $fileno = IO::AIO::poll_fileno
229		1380
230	Return the I<request result pipe file descriptor>. This filehandle must be	1381	Return the I<request result pipe file descriptor>. This filehandle must be
231	polled for reading by some mechanism outside this module (e.g. Event or	1382	polled for reading by some mechanism outside this module (e.g. EV, Glib,
232	select, see below or the SYNOPSIS). If the pipe becomes readable you have	1383	select and so on, see below or the SYNOPSIS). If the pipe becomes readable
233	to call C<poll_cb> to check the results.	1384	you have to call C<poll_cb> to check the results.
234		1385
235	See C<poll_cb> for an example.	1386	See C<poll_cb> for an example.
236		1387
237	=item IO::AIO::poll_cb	1388	=item IO::AIO::poll_cb
238		1389
239	Process all outstanding events on the result pipe. You have to call this	1390	Process some outstanding events on the result pipe. You have to call this
240	regularly. Returns the number of events processed. Returns immediately	1391	regularly. Returns C<0> if all events could be processed, or C<-1> if it
241	when no events are outstanding.	1392	returned earlier for whatever reason. Returns immediately when no events
		1393	are outstanding. The amount of events processed depends on the settings of
		1394	C<IO::AIO::max_poll_req> and C<IO::AIO::max_poll_time>.
		1395
		1396	If not all requests were processed for whatever reason, the filehandle
		1397	will still be ready when C<poll_cb> returns, so normally you don't have to
		1398	do anything special to have it called later.
242		1399
243	Example: Install an Event watcher that automatically calls	1400	Example: Install an Event watcher that automatically calls
244	IO::AIO::poll_cb with high priority:	1401	IO::AIO::poll_cb with high priority (more examples can be found in the
		1402	SYNOPSIS section, at the top of this document):
245		1403
246	Event->io (fd => IO::AIO::poll_fileno,	1404	Event->io (fd => IO::AIO::poll_fileno,
247	poll => 'r', async => 1,	1405	poll => 'r', async => 1,
248	cb => \&IO::AIO::poll_cb);	1406	cb => \&IO::AIO::poll_cb);
249		1407
250	=item IO::AIO::poll_wait	1408	=item IO::AIO::poll_wait
251		1409
		1410	If there are any outstanding requests and none of them in the result
252	Wait till the result filehandle becomes ready for reading (simply does a	1411	phase, wait till the result filehandle becomes ready for reading (simply
253	C<select> on the filehandle. This is useful if you want to synchronously wait	1412	does a C<select> on the filehandle. This is useful if you want to
254	for some requests to finish).	1413	synchronously wait for some requests to finish).
255		1414
256	See C<nreqs> for an example.	1415	See C<nreqs> for an example.
257		1416
		1417	=item IO::AIO::poll
		1418
		1419	Waits until some requests have been handled.
		1420
		1421	Returns the number of requests processed, but is otherwise strictly
		1422	equivalent to:
		1423
		1424	IO::AIO::poll_wait, IO::AIO::poll_cb
		1425
258	=item IO::AIO::nreqs	1426	=item IO::AIO::flush
259		1427
260	Returns the number of requests currently outstanding (i.e. for which their	1428	Wait till all outstanding AIO requests have been handled.
261	callback has not been invoked yet).
262		1429
263	Example: wait till there are no outstanding requests anymore:	1430	Strictly equivalent to:
264		1431
265	IO::AIO::poll_wait, IO::AIO::poll_cb	1432	IO::AIO::poll_wait, IO::AIO::poll_cb
266	while IO::AIO::nreqs;	1433	while IO::AIO::nreqs;
267		1434
		1435	=item IO::AIO::max_poll_reqs $nreqs
		1436
		1437	=item IO::AIO::max_poll_time $seconds
		1438
		1439	These set the maximum number of requests (default C<0>, meaning infinity)
		1440	that are being processed by C<IO::AIO::poll_cb> in one call, respectively
		1441	the maximum amount of time (default C<0>, meaning infinity) spent in
		1442	C<IO::AIO::poll_cb> to process requests (more correctly the mininum amount
		1443	of time C<poll_cb> is allowed to use).
		1444
		1445	Setting C<max_poll_time> to a non-zero value creates an overhead of one
		1446	syscall per request processed, which is not normally a problem unless your
		1447	callbacks are really really fast or your OS is really really slow (I am
		1448	not mentioning Solaris here). Using C<max_poll_reqs> incurs no overhead.
		1449
		1450	Setting these is useful if you want to ensure some level of
		1451	interactiveness when perl is not fast enough to process all requests in
		1452	time.
		1453
		1454	For interactive programs, values such as C<0.01> to C<0.1> should be fine.
		1455
		1456	Example: Install an Event watcher that automatically calls
		1457	IO::AIO::poll_cb with low priority, to ensure that other parts of the
		1458	program get the CPU sometimes even under high AIO load.
		1459
		1460	# try not to spend much more than 0.1s in poll_cb
		1461	IO::AIO::max_poll_time 0.1;
		1462
		1463	# use a low priority so other tasks have priority
		1464	Event->io (fd => IO::AIO::poll_fileno,
		1465	poll => 'r', nice => 1,
		1466	cb => &IO::AIO::poll_cb);
		1467
		1468	=back
		1469
		1470	=head3 CONTROLLING THE NUMBER OF THREADS
		1471
		1472	=over
		1473
		1474	=item IO::AIO::min_parallel $nthreads
		1475
		1476	Set the minimum number of AIO threads to C<$nthreads>. The current
		1477	default is C<8>, which means eight asynchronous operations can execute
		1478	concurrently at any one time (the number of outstanding requests,
		1479	however, is unlimited).
		1480
		1481	IO::AIO starts threads only on demand, when an AIO request is queued and
		1482	no free thread exists. Please note that queueing up a hundred requests can
		1483	create demand for a hundred threads, even if it turns out that everything
		1484	is in the cache and could have been processed faster by a single thread.
		1485
		1486	It is recommended to keep the number of threads relatively low, as some
		1487	Linux kernel versions will scale negatively with the number of threads
		1488	(higher parallelity => MUCH higher latency). With current Linux 2.6
		1489	versions, 4-32 threads should be fine.
		1490
		1491	Under most circumstances you don't need to call this function, as the
		1492	module selects a default that is suitable for low to moderate load.
		1493
		1494	=item IO::AIO::max_parallel $nthreads
		1495
		1496	Sets the maximum number of AIO threads to C<$nthreads>. If more than the
		1497	specified number of threads are currently running, this function kills
		1498	them. This function blocks until the limit is reached.
		1499
		1500	While C<$nthreads> are zero, aio requests get queued but not executed
		1501	until the number of threads has been increased again.
		1502
		1503	This module automatically runs C<max_parallel 0> at program end, to ensure
		1504	that all threads are killed and that there are no outstanding requests.
		1505
		1506	Under normal circumstances you don't need to call this function.
		1507
		1508	=item IO::AIO::max_idle $nthreads
		1509
		1510	Limit the number of threads (default: 4) that are allowed to idle (i.e.,
		1511	threads that did not get a request to process within 10 seconds). That
		1512	means if a thread becomes idle while C<$nthreads> other threads are also
		1513	idle, it will free its resources and exit.
		1514
		1515	This is useful when you allow a large number of threads (e.g. 100 or 1000)
		1516	to allow for extremely high load situations, but want to free resources
		1517	under normal circumstances (1000 threads can easily consume 30MB of RAM).
		1518
		1519	The default is probably ok in most situations, especially if thread
		1520	creation is fast. If thread creation is very slow on your system you might
		1521	want to use larger values.
		1522
		1523	=item IO::AIO::max_outstanding $maxreqs
		1524
		1525	This is a very bad function to use in interactive programs because it
		1526	blocks, and a bad way to reduce concurrency because it is inexact: Better
		1527	use an C<aio_group> together with a feed callback.
		1528
		1529	Sets the maximum number of outstanding requests to C<$nreqs>. If you
		1530	do queue up more than this number of requests, the next call to the
		1531	C<poll_cb> (and C<poll_some> and other functions calling C<poll_cb>)
		1532	function will block until the limit is no longer exceeded.
		1533
		1534	The default value is very large, so there is no practical limit on the
		1535	number of outstanding requests.
		1536
		1537	You can still queue as many requests as you want. Therefore,
		1538	C<max_outstanding> is mainly useful in simple scripts (with low values) or
		1539	as a stop gap to shield against fatal memory overflow (with large values).
		1540
		1541	=back
		1542
		1543	=head3 STATISTICAL INFORMATION
		1544
		1545	=over
		1546
268	=item IO::AIO::flush	1547	=item IO::AIO::nreqs
269		1548
270	Wait till all outstanding AIO requests have been handled.	1549	Returns the number of requests currently in the ready, execute or pending
		1550	states (i.e. for which their callback has not been invoked yet).
271		1551
272	Strictly equivalent to:	1552	Example: wait till there are no outstanding requests anymore:
273		1553
274	IO::AIO::poll_wait, IO::AIO::poll_cb	1554	IO::AIO::poll_wait, IO::AIO::poll_cb
275	while IO::AIO::nreqs;	1555	while IO::AIO::nreqs;
276		1556
		1557	=item IO::AIO::nready
		1558
		1559	Returns the number of requests currently in the ready state (not yet
		1560	executed).
		1561
277	=item IO::AIO::poll	1562	=item IO::AIO::npending
278		1563
279	Waits until some requests have been handled.	1564	Returns the number of requests currently in the pending state (executed,
280		1565	but not yet processed by poll_cb).
281	Strictly equivalent to:
282
283	IO::AIO::poll_wait, IO::AIO::poll_cb
284	if IO::AIO::nreqs;
285
286	=item IO::AIO::min_parallel $nthreads
287
288	Set the minimum number of AIO threads to C<$nthreads>. The default is
289	C<1>, which means a single asynchronous operation can be done at one time
290	(the number of outstanding operations, however, is unlimited).
291
292	It is recommended to keep the number of threads low, as some Linux
293	kernel versions will scale negatively with the number of threads (higher
294	parallelity => MUCH higher latency). With current Linux 2.6 versions, 4-32
295	threads should be fine.
296
297	Under normal circumstances you don't need to call this function, as this
298	module automatically starts some threads (the exact number might change,
299	and is currently 4).
300
301	=item IO::AIO::max_parallel $nthreads
302
303	Sets the maximum number of AIO threads to C<$nthreads>. If more than
304	the specified number of threads are currently running, kill them. This
305	function blocks until the limit is reached.
306
307	This module automatically runs C<max_parallel 0> at program end, to ensure
308	that all threads are killed and that there are no outstanding requests.
309
310	Under normal circumstances you don't need to call this function.
311
312	=item $oldnreqs = IO::AIO::max_outstanding $nreqs
313
314	Sets the maximum number of outstanding requests to C<$nreqs>. If you
315	try to queue up more than this number of requests, the caller will block until
316	some requests have been handled.
317
318	The default is very large, so normally there is no practical limit. If you
319	queue up many requests in a loop it it often improves speed if you set
320	this to a relatively low number, such as C<100>.
321
322	Under normal circumstances you don't need to call this function.
323		1566
324	=back	1567	=back
325		1568
		1569	=head3 MISCELLANEOUS FUNCTIONS
		1570
		1571	IO::AIO implements some functions that might be useful, but are not
		1572	asynchronous.
		1573
		1574	=over 4
		1575
		1576	=item IO::AIO::sendfile $ofh, $ifh, $offset, $count
		1577
		1578	Calls the C<eio_sendfile_sync> function, which is like C<aio_sendfile>,
		1579	but is blocking (this makes most sense if you know the input data is
		1580	likely cached already and the output filehandle is set to non-blocking
		1581	operations).
		1582
		1583	Returns the number of bytes copied, or C<-1> on error.
		1584
		1585	=item IO::AIO::fadvise $fh, $offset, $len, $advice
		1586
		1587	Simply calls the C<posix_fadvise> function (see its
		1588	manpage for details). The following advice constants are
		1589	avaiable: C<IO::AIO::FADV_NORMAL>, C<IO::AIO::FADV_SEQUENTIAL>,
		1590	C<IO::AIO::FADV_RANDOM>, C<IO::AIO::FADV_NOREUSE>,
		1591	C<IO::AIO::FADV_WILLNEED>, C<IO::AIO::FADV_DONTNEED>.
		1592
		1593	On systems that do not implement C<posix_fadvise>, this function returns
		1594	ENOSYS, otherwise the return value of C<posix_fadvise>.
		1595
		1596	=item IO::AIO::madvise $scalar, $offset, $len, $advice
		1597
		1598	Simply calls the C<posix_madvise> function (see its
		1599	manpage for details). The following advice constants are
		1600	avaiable: C<IO::AIO::MADV_NORMAL>, C<IO::AIO::MADV_SEQUENTIAL>,
		1601	C<IO::AIO::MADV_RANDOM>, C<IO::AIO::MADV_WILLNEED>, C<IO::AIO::MADV_DONTNEED>.
		1602
		1603	On systems that do not implement C<posix_madvise>, this function returns
		1604	ENOSYS, otherwise the return value of C<posix_madvise>.
		1605
		1606	=item IO::AIO::mprotect $scalar, $offset, $len, $protect
		1607
		1608	Simply calls the C<mprotect> function on the preferably AIO::mmap'ed
		1609	$scalar (see its manpage for details). The following protect
		1610	constants are avaiable: C<IO::AIO::PROT_NONE>, C<IO::AIO::PROT_READ>,
		1611	C<IO::AIO::PROT_WRITE>, C<IO::AIO::PROT_EXEC>.
		1612
		1613	On systems that do not implement C<mprotect>, this function returns
		1614	ENOSYS, otherwise the return value of C<mprotect>.
		1615
		1616	=item IO::AIO::mmap $scalar, $length, $prot, $flags, $fh[, $offset]
		1617
		1618	Memory-maps a file (or anonymous memory range) and attaches it to the
		1619	given C<$scalar>, which will act like a string scalar.
		1620
		1621	The only operations allowed on the scalar are C<substr>/C<vec> that don't
		1622	change the string length, and most read-only operations such as copying it
		1623	or searching it with regexes and so on.
		1624
		1625	Anything else is unsafe and will, at best, result in memory leaks.
		1626
		1627	The memory map associated with the C<$scalar> is automatically removed
		1628	when the C<$scalar> is destroyed, or when the C<IO::AIO::mmap> or
		1629	C<IO::AIO::munmap> functions are called.
		1630
		1631	This calls the C<mmap>(2) function internally. See your system's manual
		1632	page for details on the C<$length>, C<$prot> and C<$flags> parameters.
		1633
		1634	The C<$length> must be larger than zero and smaller than the actual
		1635	filesize.
		1636
		1637	C<$prot> is a combination of C<IO::AIO::PROT_NONE>, C<IO::AIO::PROT_EXEC>,
		1638	C<IO::AIO::PROT_READ> and/or C<IO::AIO::PROT_WRITE>,
		1639
		1640	C<$flags> can be a combination of C<IO::AIO::MAP_SHARED> or
		1641	C<IO::AIO::MAP_PRIVATE>, or a number of system-specific flags (when
		1642	not available, the are defined as 0): C<IO::AIO::MAP_ANONYMOUS>
		1643	(which is set to C<MAP_ANON> if your system only provides this
		1644	constant), C<IO::AIO::MAP_HUGETLB>, C<IO::AIO::MAP_LOCKED>,
		1645	C<IO::AIO::MAP_NORESERVE>, C<IO::AIO::MAP_POPULATE> or
		1646	C<IO::AIO::MAP_NONBLOCK>
		1647
		1648	If C<$fh> is C<undef>, then a file descriptor of C<-1> is passed.
		1649
		1650	C<$offset> is the offset from the start of the file - it generally must be
		1651	a multiple of C<IO::AIO::PAGESIZE> and defaults to C<0>.
		1652
		1653	Example:
		1654
		1655	use Digest::MD5;
		1656	use IO::AIO;
		1657
		1658	open my $fh, "<verybigfile"
		1659	or die "$!";
		1660
		1661	IO::AIO::mmap my $data, -s $fh, IO::AIO::PROT_READ, IO::AIO::MAP_SHARED, $fh
		1662	or die "verybigfile: $!";
		1663
		1664	my $fast_md5 = md5 $data;
		1665
		1666	=item IO::AIO::munmap $scalar
		1667
		1668	Removes a previous mmap and undefines the C<$scalar>.
		1669
		1670	=item IO::AIO::munlock $scalar, $offset = 0, $length = undef
		1671
		1672	Calls the C<munlock> function, undoing the effects of a previous
		1673	C<aio_mlock> call (see its description for details).
		1674
		1675	=item IO::AIO::munlockall
		1676
		1677	Calls the C<munlockall> function.
		1678
		1679	On systems that do not implement C<munlockall>, this function returns
		1680	ENOSYS, otherwise the return value of C<munlockall>.
		1681
		1682	=back
		1683
326	=cut	1684	=cut
327		1685
328	# support function to convert a fd into a perl filehandle
329	sub _fd2fh {
330	return undef if $_[0] < 0;
331
332	# try to generate nice filehandles
333	my $sym = "IO::AIO::fd#$_[0]";
334	local *$sym;
335
336	open *$sym, "+<&=$_[0]" # usually works under any unix
337	or open *$sym, "<&=$_[0]" # cygwin needs this
338	or open *$sym, ">&=$_[0]" # or this
339	or return undef;
340
341	*$sym
342	}
343
344	min_parallel 4;	1686	min_parallel 8;
345		1687
346	END {	1688	END { flush }
347	max_parallel 0;
348	}
349		1689
350	1;	1690	1;
351		1691
		1692	=head1 EVENT LOOP INTEGRATION
		1693
		1694	It is recommended to use L<AnyEvent::AIO> to integrate IO::AIO
		1695	automatically into many event loops:
		1696
		1697	# AnyEvent integration (EV, Event, Glib, Tk, POE, urxvt, pureperl...)
		1698	use AnyEvent::AIO;
		1699
		1700	You can also integrate IO::AIO manually into many event loops, here are
		1701	some examples of how to do this:
		1702
		1703	# EV integration
		1704	my $aio_w = EV::io IO::AIO::poll_fileno, EV::READ, \&IO::AIO::poll_cb;
		1705
		1706	# Event integration
		1707	Event->io (fd => IO::AIO::poll_fileno,
		1708	poll => 'r',
		1709	cb => \&IO::AIO::poll_cb);
		1710
		1711	# Glib/Gtk2 integration
		1712	add_watch Glib::IO IO::AIO::poll_fileno,
		1713	in => sub { IO::AIO::poll_cb; 1 };
		1714
		1715	# Tk integration
		1716	Tk::Event::IO->fileevent (IO::AIO::poll_fileno, "",
		1717	readable => \&IO::AIO::poll_cb);
		1718
		1719	# Danga::Socket integration
		1720	Danga::Socket->AddOtherFds (IO::AIO::poll_fileno =>
		1721	\&IO::AIO::poll_cb);
		1722
352	=head2 FORK BEHAVIOUR	1723	=head2 FORK BEHAVIOUR
353		1724
354	Before the fork IO::AIO first handles all outstanding requests - if other	1725	This module should do "the right thing" when the process using it forks:
355	threads add requests during this period, this time is prolonged. It then	1726
356	enters a quiescent state where no requests can be added in other threads	1727	Before the fork, IO::AIO enters a quiescent state where no requests
357	and no results will be processed. After the fork the parent simply leaves	1728	can be added in other threads and no results will be processed. After
358	the quiescent state and continues request processing, while the child	1729	the fork the parent simply leaves the quiescent state and continues
359	starts the same number of threads as were in use by the parent.	1730	request/result processing, while the child frees the request/result queue
		1731	(so that the requests started before the fork will only be handled in the
		1732	parent). Threads will be started on demand until the limit set in the
		1733	parent process has been reached again.
		1734
		1735	In short: the parent will, after a short pause, continue as if fork had
		1736	not been called, while the child will act as if IO::AIO has not been used
		1737	yet.
		1738
		1739	=head2 MEMORY USAGE
		1740
		1741	Per-request usage:
		1742
		1743	Each aio request uses - depending on your architecture - around 100-200
		1744	bytes of memory. In addition, stat requests need a stat buffer (possibly
		1745	a few hundred bytes), readdir requires a result buffer and so on. Perl
		1746	scalars and other data passed into aio requests will also be locked and
		1747	will consume memory till the request has entered the done state.
		1748
		1749	This is not awfully much, so queuing lots of requests is not usually a
		1750	problem.
		1751
		1752	Per-thread usage:
		1753
		1754	In the execution phase, some aio requests require more memory for
		1755	temporary buffers, and each thread requires a stack and other data
		1756	structures (usually around 16k-128k, depending on the OS).
		1757
		1758	=head1 KNOWN BUGS
		1759
		1760	Known bugs will be fixed in the next release.
360		1761
361	=head1 SEE ALSO	1762	=head1 SEE ALSO
362		1763
363	L<Coro>, L<Linux::AIO>.	1764	L<AnyEvent::AIO> for easy integration into event loops, L<Coro::AIO> for a
		1765	more natural syntax.
364		1766
365	=head1 AUTHOR	1767	=head1 AUTHOR
366		1768
367	Marc Lehmann <schmorp@schmorp.de>	1769	Marc Lehmann <schmorp@schmorp.de>
368	http://home.schmorp.de/	1770	http://home.schmorp.de/

Diff Legend

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