ViewVC Help

View File | Revision Log | Show Annotations | Download File

/cvs/IO-AIO/AIO.pm

Comparing IO-AIO/AIO.pm (file contents):
Revision 1.24 by root, Sun Jul 31 18:45:48 2005 UTC vs.
Revision 1.288 by root, Tue Jul 31 22:27:49 2018 UTC

1	=head1 NAME	1	=head1 NAME
2		2
3	IO::AIO - Asynchronous Input/Output	3	IO::AIO - Asynchronous/Advanced Input/Output
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
		63	In addition to asynchronous I/O, this module also exports some rather
		64	arcane interfaces, such as C<madvise> or linux's C<splice> system call,
		65	which is why the C<A> in C<AIO> can also mean I<advanced>.
		66
52	Although the module will work with in the presence of other threads, it is	67	Although the module will work in the presence of other (Perl-) threads,
53	currently not reentrant, so use appropriate locking yourself, always call	68	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	69	yourself, always call C<poll_cb> from within the same thread, or never
55	C<aio_> functions) recursively.	70	call C<poll_cb> (or other C<aio_> functions) recursively.
		71
		72	=head2 EXAMPLE
		73
		74	This is a simple example that uses the EV module and loads
		75	F</etc/passwd> asynchronously:
		76
		77	use EV;
		78	use IO::AIO;
		79
		80	# register the IO::AIO callback with EV
		81	my $aio_w = EV::io IO::AIO::poll_fileno, EV::READ, \&IO::AIO::poll_cb;
		82
		83	# queue the request to open /etc/passwd
		84	aio_open "/etc/passwd", IO::AIO::O_RDONLY, 0, sub {
		85	my $fh = shift
		86	or die "error while opening: $!";
		87
		88	# stat'ing filehandles is generally non-blocking
		89	my $size = -s $fh;
		90
		91	# queue a request to read the file
		92	my $contents;
		93	aio_read $fh, 0, $size, $contents, 0, sub {
		94	$_[0] == $size
		95	or die "short read: $!";
		96
		97	close $fh;
		98
		99	# file contents now in $contents
		100	print $contents;
		101
		102	# exit event loop and program
		103	EV::break;
		104	};
		105	};
		106
		107	# possibly queue up other requests, or open GUI windows,
		108	# check for sockets etc. etc.
		109
		110	# process events as long as there are some:
		111	EV::run;
		112
		113	=head1 REQUEST ANATOMY AND LIFETIME
		114
		115	Every C<aio_*> function creates a request. which is a C data structure not
		116	directly visible to Perl.
		117
		118	If called in non-void context, every request function returns a Perl
		119	object representing the request. In void context, nothing is returned,
		120	which saves a bit of memory.
		121
		122	The perl object is a fairly standard ref-to-hash object. The hash contents
		123	are not used by IO::AIO so you are free to store anything you like in it.
		124
		125	During their existance, aio requests travel through the following states,
		126	in order:
		127
		128	=over 4
		129
		130	=item ready
		131
		132	Immediately after a request is created it is put into the ready state,
		133	waiting for a thread to execute it.
		134
		135	=item execute
		136
		137	A thread has accepted the request for processing and is currently
		138	executing it (e.g. blocking in read).
		139
		140	=item pending
		141
		142	The request has been executed and is waiting for result processing.
		143
		144	While request submission and execution is fully asynchronous, result
		145	processing is not and relies on the perl interpreter calling C<poll_cb>
		146	(or another function with the same effect).
		147
		148	=item result
		149
		150	The request results are processed synchronously by C<poll_cb>.
		151
		152	The C<poll_cb> function will process all outstanding aio requests by
		153	calling their callbacks, freeing memory associated with them and managing
		154	any groups they are contained in.
		155
		156	=item done
		157
		158	Request has reached the end of its lifetime and holds no resources anymore
		159	(except possibly for the Perl object, but its connection to the actual
		160	aio request is severed and calling its methods will either do nothing or
		161	result in a runtime error).
		162
		163	=back
56		164
57	=cut	165	=cut
58		166
59	package IO::AIO;	167	package IO::AIO;
60		168
61	no warnings;	169	use Carp ();
		170
		171	use common::sense;
62		172
63	use base 'Exporter';	173	use base 'Exporter';
64		174
65	use Fcntl ();
66
67	BEGIN {	175	BEGIN {
68	$VERSION = 1.1;	176	our $VERSION = 4.5;
69		177
70	@EXPORT = qw(aio_read aio_write aio_open aio_close aio_stat aio_lstat aio_unlink	178	our @AIO_REQ = qw(aio_sendfile aio_seek aio_read aio_write aio_open aio_close
71	aio_fsync aio_fdatasync aio_readahead);	179	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);	180	aio_scandir aio_symlink aio_readlink aio_realpath aio_fcntl aio_ioctl
		181	aio_sync aio_fsync aio_syncfs aio_fdatasync aio_sync_file_range
		182	aio_pathsync aio_readahead aio_fiemap aio_allocate
		183	aio_rename aio_rename2 aio_link aio_move aio_copy aio_group
		184	aio_nop aio_mknod aio_load aio_rmtree aio_mkdir aio_chown
		185	aio_chmod aio_utime aio_truncate
		186	aio_msync aio_mtouch aio_mlock aio_mlockall
		187	aio_statvfs
		188	aio_slurp
		189	aio_wd);
		190
		191	our @EXPORT = (@AIO_REQ, qw(aioreq_pri aioreq_nice));
		192	our @EXPORT_OK = qw(poll_fileno poll_cb poll_wait flush
		193	min_parallel max_parallel max_idle idle_timeout
		194	nreqs nready npending nthreads
		195	max_poll_time max_poll_reqs
		196	sendfile fadvise madvise
		197	mmap munmap mremap munlock munlockall);
		198
		199	push @AIO_REQ, qw(aio_busy); # not exported
		200
		201	@IO::AIO::GRP::ISA = 'IO::AIO::REQ';
73		202
74	require XSLoader;	203	require XSLoader;
75	XSLoader::load IO::AIO, $VERSION;	204	XSLoader::load ("IO::AIO", $VERSION);
76	}	205	}
77		206
78	=head1 FUNCTIONS	207	=head1 FUNCTIONS
79		208
80	=head2 AIO FUNCTIONS	209	=head2 QUICK OVERVIEW
		210
		211	This section simply lists the prototypes most of the functions for
		212	quick reference. See the following sections for function-by-function
		213	documentation.
		214
		215	aio_wd $pathname, $callback->($wd)
		216	aio_open $pathname, $flags, $mode, $callback->($fh)
		217	aio_close $fh, $callback->($status)
		218	aio_seek $fh,$offset,$whence, $callback->($offs)
		219	aio_read $fh,$offset,$length, $data,$dataoffset, $callback->($retval)
		220	aio_write $fh,$offset,$length, $data,$dataoffset, $callback->($retval)
		221	aio_sendfile $out_fh, $in_fh, $in_offset, $length, $callback->($retval)
		222	aio_readahead $fh,$offset,$length, $callback->($retval)
		223	aio_stat $fh_or_path, $callback->($status)
		224	aio_lstat $fh, $callback->($status)
		225	aio_statvfs $fh_or_path, $callback->($statvfs)
		226	aio_utime $fh_or_path, $atime, $mtime, $callback->($status)
		227	aio_chown $fh_or_path, $uid, $gid, $callback->($status)
		228	aio_chmod $fh_or_path, $mode, $callback->($status)
		229	aio_truncate $fh_or_path, $offset, $callback->($status)
		230	aio_allocate $fh, $mode, $offset, $len, $callback->($status)
		231	aio_fiemap $fh, $start, $length, $flags, $count, $cb->(\@extents)
		232	aio_unlink $pathname, $callback->($status)
		233	aio_mknod $pathname, $mode, $dev, $callback->($status)
		234	aio_link $srcpath, $dstpath, $callback->($status)
		235	aio_symlink $srcpath, $dstpath, $callback->($status)
		236	aio_readlink $pathname, $callback->($link)
		237	aio_realpath $pathname, $callback->($path)
		238	aio_rename $srcpath, $dstpath, $callback->($status)
		239	aio_rename2 $srcpath, $dstpath, $flags, $callback->($status)
		240	aio_mkdir $pathname, $mode, $callback->($status)
		241	aio_rmdir $pathname, $callback->($status)
		242	aio_readdir $pathname, $callback->($entries)
		243	aio_readdirx $pathname, $flags, $callback->($entries, $flags)
		244	IO::AIO::READDIR_DENTS IO::AIO::READDIR_DIRS_FIRST
		245	IO::AIO::READDIR_STAT_ORDER IO::AIO::READDIR_FOUND_UNKNOWN
		246	aio_scandir $pathname, $maxreq, $callback->($dirs, $nondirs)
		247	aio_load $pathname, $data, $callback->($status)
		248	aio_copy $srcpath, $dstpath, $callback->($status)
		249	aio_move $srcpath, $dstpath, $callback->($status)
		250	aio_rmtree $pathname, $callback->($status)
		251	aio_fcntl $fh, $cmd, $arg, $callback->($status)
		252	aio_ioctl $fh, $request, $buf, $callback->($status)
		253	aio_sync $callback->($status)
		254	aio_syncfs $fh, $callback->($status)
		255	aio_fsync $fh, $callback->($status)
		256	aio_fdatasync $fh, $callback->($status)
		257	aio_sync_file_range $fh, $offset, $nbytes, $flags, $callback->($status)
		258	aio_pathsync $pathname, $callback->($status)
		259	aio_msync $scalar, $offset = 0, $length = undef, flags = MS_SYNC, $callback->($status)
		260	aio_mtouch $scalar, $offset = 0, $length = undef, flags = 0, $callback->($status)
		261	aio_mlock $scalar, $offset = 0, $length = undef, $callback->($status)
		262	aio_mlockall $flags, $callback->($status)
		263	aio_group $callback->(...)
		264	aio_nop $callback->()
		265
		266	$prev_pri = aioreq_pri [$pri]
		267	aioreq_nice $pri_adjust
		268
		269	IO::AIO::poll_wait
		270	IO::AIO::poll_cb
		271	IO::AIO::poll
		272	IO::AIO::flush
		273	IO::AIO::max_poll_reqs $nreqs
		274	IO::AIO::max_poll_time $seconds
		275	IO::AIO::min_parallel $nthreads
		276	IO::AIO::max_parallel $nthreads
		277	IO::AIO::max_idle $nthreads
		278	IO::AIO::idle_timeout $seconds
		279	IO::AIO::max_outstanding $maxreqs
		280	IO::AIO::nreqs
		281	IO::AIO::nready
		282	IO::AIO::npending
		283	$nfd = IO::AIO::get_fdlimit [EXPERIMENTAL]
		284	IO::AIO::min_fdlimit $nfd [EXPERIMENTAL]
		285
		286	IO::AIO::sendfile $ofh, $ifh, $offset, $count
		287	IO::AIO::fadvise $fh, $offset, $len, $advice
		288	IO::AIO::mmap $scalar, $length, $prot, $flags[, $fh[, $offset]]
		289	IO::AIO::munmap $scalar
		290	IO::AIO::mremap $scalar, $new_length, $flags[, $new_address]
		291	IO::AIO::madvise $scalar, $offset, $length, $advice
		292	IO::AIO::mprotect $scalar, $offset, $length, $protect
		293	IO::AIO::munlock $scalar, $offset = 0, $length = undef
		294	IO::AIO::munlockall
		295
		296	=head2 API NOTES
81		297
82	All the C<aio_*> calls are more or less thin wrappers around the syscall	298	All the C<aio_*> calls are more or less thin wrappers around the syscall
83	with the same name (sans C<aio_>). The arguments are similar or identical,	299	with the same name (sans C<aio_>). The arguments are similar or identical,
84	and they all accept an additional (and optional) C<$callback> argument	300	and they all accept an additional (and optional) C<$callback> argument
85	which must be a code reference. This code reference will get called with	301	which must be a code reference. This code reference will be called after
86	the syscall return code (e.g. most syscalls return C<-1> on error, unlike	302	the syscall has been executed in an asynchronous fashion. The results
87	perl, which usually delivers "false") as it's sole argument when the given	303	of the request will be passed as arguments to the callback (and, if an
88	syscall has been executed asynchronously.	304	error occured, in C<$!>) - for most requests the syscall return code (e.g.
		305	most syscalls return C<-1> on error, unlike perl, which usually delivers
		306	"false").
		307
		308	Some requests (such as C<aio_readdir>) pass the actual results and
		309	communicate failures by passing C<undef>.
89		310
90	All functions expecting a filehandle keep a copy of the filehandle	311	All functions expecting a filehandle keep a copy of the filehandle
91	internally until the request has finished.	312	internally until the request has finished.
92		313
		314	All functions return request objects of type L<IO::AIO::REQ> that allow
		315	further manipulation of those requests while they are in-flight.
		316
93	The filenames you pass to these routines I<must> be absolute. The reason	317	The pathnames you pass to these routines I<should> be absolute. The
94	for this is that at the time the request is being executed, the current	318	reason for this is that at the time the request is being executed, the
95	working directory could have changed. Alternatively, you can make sure	319	current working directory could have changed. Alternatively, you can
96	that you never change the current working directory.	320	make sure that you never change the current working directory anywhere
		321	in the program and then use relative paths. You can also take advantage
		322	of IO::AIOs working directory abstraction, that lets you specify paths
		323	relative to some previously-opened "working directory object" - see the
		324	description of the C<IO::AIO::WD> class later in this document.
		325
		326	To encode pathnames as octets, either make sure you either: a) always pass
		327	in filenames you got from outside (command line, readdir etc.) without
		328	tinkering, b) are in your native filesystem encoding, c) use the Encode
		329	module and encode your pathnames to the locale (or other) encoding in
		330	effect in the user environment, d) use Glib::filename_from_unicode on
		331	unicode filenames or e) use something else to ensure your scalar has the
		332	correct contents.
		333
		334	This works, btw. independent of the internal UTF-8 bit, which IO::AIO
		335	handles correctly whether it is set or not.
		336
		337	=head2 AIO REQUEST FUNCTIONS
97		338
98	=over 4	339	=over 4
99		340
		341	=item $prev_pri = aioreq_pri [$pri]
		342
		343	Returns the priority value that would be used for the next request and, if
		344	C<$pri> is given, sets the priority for the next aio request.
		345
		346	The default priority is C<0>, the minimum and maximum priorities are C<-4>
		347	and C<4>, respectively. Requests with higher priority will be serviced
		348	first.
		349
		350	The priority will be reset to C<0> after each call to one of the C<aio_*>
		351	functions.
		352
		353	Example: open a file with low priority, then read something from it with
		354	higher priority so the read request is serviced before other low priority
		355	open requests (potentially spamming the cache):
		356
		357	aioreq_pri -3;
		358	aio_open ..., sub {
		359	return unless $_[0];
		360
		361	aioreq_pri -2;
		362	aio_read $_[0], ..., sub {
		363	...
		364	};
		365	};
		366
		367
		368	=item aioreq_nice $pri_adjust
		369
		370	Similar to C<aioreq_pri>, but subtracts the given value from the current
		371	priority, so the effect is cumulative.
		372
		373
100	=item aio_open $pathname, $flags, $mode, $callback	374	=item aio_open $pathname, $flags, $mode, $callback->($fh)
101		375
102	Asynchronously open or create a file and call the callback with a newly	376	Asynchronously open or create a file and call the callback with a newly
103	created filehandle for the file.	377	created filehandle for the file (or C<undef> in case of an error).
104		378
105	The pathname passed to C<aio_open> must be absolute. See API NOTES, above,	379	The pathname passed to C<aio_open> must be absolute. See API NOTES, above,
106	for an explanation.	380	for an explanation.
107		381
108	The C<$flags> argument is a bitmask. See the C<Fcntl> module for a	382	The C<$flags> argument is a bitmask. See the C<Fcntl> module for a
109	list. They are the same as used by C<sysopen>.	383	list. They are the same as used by C<sysopen>.
110		384
111	Likewise, C<$mode> specifies the mode of the newly created file, if it	385	Likewise, C<$mode> specifies the mode of the newly created file, if it
112	didn't exist and C<O_CREAT> has been given, just like perl's C<sysopen>,	386	didn't exist and C<O_CREAT> has been given, just like perl's C<sysopen>,
113	except that it is mandatory (i.e. use C<0> if you don't create new files,	387	except that it is mandatory (i.e. use C<0> if you don't create new files,
114	and C<0666> or C<0777> if you do).	388	and C<0666> or C<0777> if you do). Note that the C<$mode> will be modified
		389	by the umask in effect then the request is being executed, so better never
		390	change the umask.
115		391
116	Example:	392	Example:
117		393
118	aio_open "/etc/passwd", O_RDONLY, 0, sub {	394	aio_open "/etc/passwd", IO::AIO::O_RDONLY, 0, sub {
119	if ($_[0]) {	395	if ($_[0]) {
120	print "open successful, fh is $_[0]\n";	396	print "open successful, fh is $_[0]\n";
121	...	397	...
122	} else {	398	} else {
123	die "open failed: $!\n";	399	die "open failed: $!\n";
124	}	400	}
125	};	401	};
126		402
		403	In addition to all the common open modes/flags (C<O_RDONLY>, C<O_WRONLY>,
		404	C<O_RDWR>, C<O_CREAT>, C<O_TRUNC>, C<O_EXCL> and C<O_APPEND>), the
		405	following POSIX and non-POSIX constants are available (missing ones on
		406	your system are, as usual, C<0>):
		407
		408	C<O_ASYNC>, C<O_DIRECT>, C<O_NOATIME>, C<O_CLOEXEC>, C<O_NOCTTY>, C<O_NOFOLLOW>,
		409	C<O_NONBLOCK>, C<O_EXEC>, C<O_SEARCH>, C<O_DIRECTORY>, C<O_DSYNC>,
		410	C<O_RSYNC>, C<O_SYNC>, C<O_PATH>, C<O_TMPFILE>, C<O_TTY_INIT> and C<O_ACCMODE>.
		411
		412
127	=item aio_close $fh, $callback	413	=item aio_close $fh, $callback->($status)
128		414
129	Asynchronously close a file and call the callback with the result	415	Asynchronously close a file and call the callback with the result
130	code. I<WARNING:> although accepted, you should not pass in a perl	416	code.
131	filehandle here, as perl will likely close the file descriptor another
132	time when the filehandle is destroyed. Normally, you can safely call perls
133	C<close> or just let filehandles go out of scope.
134		417
135	This is supposed to be a bug in the API, so that might change. It's	418	Unfortunately, you can't do this to perl. Perl I<insists> very strongly on
136	therefore best to avoid this function.	419	closing the file descriptor associated with the filehandle itself.
137		420
		421	Therefore, C<aio_close> will not close the filehandle - instead it will
		422	use dup2 to overwrite the file descriptor with the write-end of a pipe
		423	(the pipe fd will be created on demand and will be cached).
		424
		425	Or in other words: the file descriptor will be closed, but it will not be
		426	free for reuse until the perl filehandle is closed.
		427
		428	=cut
		429
		430	=item aio_seek $fh, $offset, $whence, $callback->($offs)
		431
		432	Seeks the filehandle to the new C<$offset>, similarly to perl's
		433	C<sysseek>. The C<$whence> can use the traditional values (C<0> for
		434	C<IO::AIO::SEEK_SET>, C<1> for C<IO::AIO::SEEK_CUR> or C<2> for
		435	C<IO::AIO::SEEK_END>).
		436
		437	The resulting absolute offset will be passed to the callback, or C<-1> in
		438	case of an error.
		439
		440	In theory, the C<$whence> constants could be different than the
		441	corresponding values from L<Fcntl>, but perl guarantees they are the same,
		442	so don't panic.
		443
		444	As a GNU/Linux (and maybe Solaris) extension, also the constants
		445	C<IO::AIO::SEEK_DATA> and C<IO::AIO::SEEK_HOLE> are available, if they
		446	could be found. No guarantees about suitability for use in C<aio_seek> or
		447	Perl's C<sysseek> can be made though, although I would naively assume they
		448	"just work".
		449
138	=item aio_read $fh,$offset,$length, $data,$dataoffset,$callback	450	=item aio_read $fh,$offset,$length, $data,$dataoffset, $callback->($retval)
139		451
140	=item aio_write $fh,$offset,$length, $data,$dataoffset,$callback	452	=item aio_write $fh,$offset,$length, $data,$dataoffset, $callback->($retval)
141		453
142	Reads or writes C<length> bytes from the specified C<fh> and C<offset>	454	Reads or writes C<$length> bytes from or to the specified C<$fh> and
143	into the scalar given by C<data> and offset C<dataoffset> and calls the	455	C<$offset> into the scalar given by C<$data> and offset C<$dataoffset> and
144	callback without the actual number of bytes read (or -1 on error, just	456	calls the callback with the actual number of bytes transferred (or -1 on
145	like the syscall).	457	error, just like the syscall).
		458
		459	C<aio_read> will, like C<sysread>, shrink or grow the C<$data> scalar to
		460	offset plus the actual number of bytes read.
		461
		462	If C<$offset> is undefined, then the current file descriptor offset will
		463	be used (and updated), otherwise the file descriptor offset will not be
		464	changed by these calls.
		465
		466	If C<$length> is undefined in C<aio_write>, use the remaining length of
		467	C<$data>.
		468
		469	If C<$dataoffset> is less than zero, it will be counted from the end of
		470	C<$data>.
		471
		472	The C<$data> scalar I<MUST NOT> be modified in any way while the request
		473	is outstanding. Modifying it can result in segfaults or World War III (if
		474	the necessary/optional hardware is installed).
146		475
147	Example: Read 15 bytes at offset 7 into scalar C<$buffer>, starting at	476	Example: Read 15 bytes at offset 7 into scalar C<$buffer>, starting at
148	offset C<0> within the scalar:	477	offset C<0> within the scalar:
149		478
150	aio_read $fh, 7, 15, $buffer, 0, sub {	479	aio_read $fh, 7, 15, $buffer, 0, sub {
151	$_[0] > 0 or die "read error: $!";	480	$_[0] > 0 or die "read error: $!";
152	print "read $_[0] bytes: <$buffer>\n";	481	print "read $_[0] bytes: <$buffer>\n";
153	};	482	};
154		483
		484
		485	=item aio_sendfile $out_fh, $in_fh, $in_offset, $length, $callback->($retval)
		486
		487	Tries to copy C<$length> bytes from C<$in_fh> to C<$out_fh>. It starts
		488	reading at byte offset C<$in_offset>, and starts writing at the current
		489	file offset of C<$out_fh>. Because of that, it is not safe to issue more
		490	than one C<aio_sendfile> per C<$out_fh>, as they will interfere with each
		491	other. The same C<$in_fh> works fine though, as this function does not
		492	move or use the file offset of C<$in_fh>.
		493
		494	Please note that C<aio_sendfile> can read more bytes from C<$in_fh> than
		495	are written, and there is no way to find out how many more bytes have been
		496	read from C<aio_sendfile> alone, as C<aio_sendfile> only provides the
		497	number of bytes written to C<$out_fh>. Only if the result value equals
		498	C<$length> one can assume that C<$length> bytes have been read.
		499
		500	Unlike with other C<aio_> functions, it makes a lot of sense to use
		501	C<aio_sendfile> on non-blocking sockets, as long as one end (typically
		502	the C<$in_fh>) is a file - the file I/O will then be asynchronous, while
		503	the socket I/O will be non-blocking. Note, however, that you can run
		504	into a trap where C<aio_sendfile> reads some data with readahead, then
		505	fails to write all data, and when the socket is ready the next time, the
		506	data in the cache is already lost, forcing C<aio_sendfile> to again hit
		507	the disk. Explicit C<aio_read> + C<aio_write> let's you better control
		508	resource usage.
		509
		510	This call tries to make use of a native C<sendfile>-like syscall to
		511	provide zero-copy operation. For this to work, C<$out_fh> should refer to
		512	a socket, and C<$in_fh> should refer to an mmap'able file.
		513
		514	If a native sendfile cannot be found or it fails with C<ENOSYS>,
		515	C<EINVAL>, C<ENOTSUP>, C<EOPNOTSUPP>, C<EAFNOSUPPORT>, C<EPROTOTYPE> or
		516	C<ENOTSOCK>, it will be emulated, so you can call C<aio_sendfile> on any
		517	type of filehandle regardless of the limitations of the operating system.
		518
		519	As native sendfile syscalls (as practically any non-POSIX interface hacked
		520	together in a hurry to improve benchmark numbers) tend to be rather buggy
		521	on many systems, this implementation tries to work around some known bugs
		522	in Linux and FreeBSD kernels (probably others, too), but that might fail,
		523	so you really really should check the return value of C<aio_sendfile> -
		524	fewer bytes than expected might have been transferred.
		525
		526
155	=item aio_readahead $fh,$offset,$length, $callback	527	=item aio_readahead $fh,$offset,$length, $callback->($retval)
156
157	Asynchronously reads the specified byte range into the page cache, using
158	the C<readahead> syscall. If that syscall doesn't exist (likely if your OS
159	isn't Linux) the status will be C<-1> and C<$!> is set to C<ENOSYS>.
160		528
161	C<aio_readahead> populates the page cache with data from a file so that	529	C<aio_readahead> populates the page cache with data from a file so that
162	subsequent reads from that file will not block on disk I/O. The C<$offset>	530	subsequent reads from that file will not block on disk I/O. The C<$offset>
163	argument specifies the starting point from which data is to be read and	531	argument specifies the starting point from which data is to be read and
164	C<$length> specifies the number of bytes to be read. I/O is performed in	532	C<$length> specifies the number of bytes to be read. I/O is performed in
165	whole pages, so that offset is effectively rounded down to a page boundary	533	whole pages, so that offset is effectively rounded down to a page boundary
166	and bytes are read up to the next page boundary greater than or equal to	534	and bytes are read up to the next page boundary greater than or equal to
167	(off-set+length). C<aio_readahead> does not read beyond the end of the	535	(off-set+length). C<aio_readahead> does not read beyond the end of the
168	file. The current file offset of the file is left unchanged.	536	file. The current file offset of the file is left unchanged.
169		537
		538	If that syscall doesn't exist (likely if your kernel isn't Linux) it will
		539	be emulated by simply reading the data, which would have a similar effect.
		540
		541
170	=item aio_stat $fh_or_path, $callback	542	=item aio_stat $fh_or_path, $callback->($status)
171		543
172	=item aio_lstat $fh, $callback	544	=item aio_lstat $fh, $callback->($status)
173		545
174	Works like perl's C<stat> or C<lstat> in void context. The callback will	546	Works like perl's C<stat> or C<lstat> in void context. The callback will
175	be called after the stat and the results will be available using C<stat _>	547	be called after the stat and the results will be available using C<stat _>
176	or C<-s _> etc...	548	or C<-s _> etc...
177		549
…		…
179	for an explanation.	551	for an explanation.
180		552
181	Currently, the stats are always 64-bit-stats, i.e. instead of returning an	553	Currently, the stats are always 64-bit-stats, i.e. instead of returning an
182	error when stat'ing a large file, the results will be silently truncated	554	error when stat'ing a large file, the results will be silently truncated
183	unless perl itself is compiled with large file support.	555	unless perl itself is compiled with large file support.
		556
		557	To help interpret the mode and dev/rdev stat values, IO::AIO offers the
		558	following constants and functions (if not implemented, the constants will
		559	be C<0> and the functions will either C<croak> or fall back on traditional
		560	behaviour).
		561
		562	C<S_IFMT>, C<S_IFIFO>, C<S_IFCHR>, C<S_IFBLK>, C<S_IFLNK>, C<S_IFREG>,
		563	C<S_IFDIR>, C<S_IFWHT>, C<S_IFSOCK>, C<IO::AIO::major $dev_t>,
		564	C<IO::AIO::minor $dev_t>, C<IO::AIO::makedev $major, $minor>.
184		565
185	Example: Print the length of F</etc/passwd>:	566	Example: Print the length of F</etc/passwd>:
186		567
187	aio_stat "/etc/passwd", sub {	568	aio_stat "/etc/passwd", sub {
188	$_[0] and die "stat failed: $!";	569	$_[0] and die "stat failed: $!";
189	print "size is ", -s _, "\n";	570	print "size is ", -s _, "\n";
190	};	571	};
191		572
		573
		574	=item aio_statvfs $fh_or_path, $callback->($statvfs)
		575
		576	Works like the POSIX C<statvfs> or C<fstatvfs> syscalls, depending on
		577	whether a file handle or path was passed.
		578
		579	On success, the callback is passed a hash reference with the following
		580	members: C<bsize>, C<frsize>, C<blocks>, C<bfree>, C<bavail>, C<files>,
		581	C<ffree>, C<favail>, C<fsid>, C<flag> and C<namemax>. On failure, C<undef>
		582	is passed.
		583
		584	The following POSIX IO::AIO::ST_* constants are defined: C<ST_RDONLY> and
		585	C<ST_NOSUID>.
		586
		587	The following non-POSIX IO::AIO::ST_* flag masks are defined to
		588	their correct value when available, or to C<0> on systems that do
		589	not support them: C<ST_NODEV>, C<ST_NOEXEC>, C<ST_SYNCHRONOUS>,
		590	C<ST_MANDLOCK>, C<ST_WRITE>, C<ST_APPEND>, C<ST_IMMUTABLE>, C<ST_NOATIME>,
		591	C<ST_NODIRATIME> and C<ST_RELATIME>.
		592
		593	Example: stat C</wd> and dump out the data if successful.
		594
		595	aio_statvfs "/wd", sub {
		596	my $f = $_[0]
		597	or die "statvfs: $!";
		598
		599	use Data::Dumper;
		600	say Dumper $f;
		601	};
		602
		603	# result:
		604	{
		605	bsize => 1024,
		606	bfree => 4333064312,
		607	blocks => 10253828096,
		608	files => 2050765568,
		609	flag => 4096,
		610	favail => 2042092649,
		611	bavail => 4333064312,
		612	ffree => 2042092649,
		613	namemax => 255,
		614	frsize => 1024,
		615	fsid => 1810
		616	}
		617
		618	=item aio_utime $fh_or_path, $atime, $mtime, $callback->($status)
		619
		620	Works like perl's C<utime> function (including the special case of $atime
		621	and $mtime being undef). Fractional times are supported if the underlying
		622	syscalls support them.
		623
		624	When called with a pathname, uses utimes(2) if available, otherwise
		625	utime(2). If called on a file descriptor, uses futimes(2) if available,
		626	otherwise returns ENOSYS, so this is not portable.
		627
		628	Examples:
		629
		630	# set atime and mtime to current time (basically touch(1)):
		631	aio_utime "path", undef, undef;
		632	# set atime to current time and mtime to beginning of the epoch:
		633	aio_utime "path", time, undef; # undef==0
		634
		635
		636	=item aio_chown $fh_or_path, $uid, $gid, $callback->($status)
		637
		638	Works like perl's C<chown> function, except that C<undef> for either $uid
		639	or $gid is being interpreted as "do not change" (but -1 can also be used).
		640
		641	Examples:
		642
		643	# same as "chown root path" in the shell:
		644	aio_chown "path", 0, -1;
		645	# same as above:
		646	aio_chown "path", 0, undef;
		647
		648
		649	=item aio_truncate $fh_or_path, $offset, $callback->($status)
		650
		651	Works like truncate(2) or ftruncate(2).
		652
		653
		654	=item aio_allocate $fh, $mode, $offset, $len, $callback->($status)
		655
		656	Allocates or frees disk space according to the C<$mode> argument. See the
		657	linux C<fallocate> documentation for details.
		658
		659	C<$mode> is usually C<0> or C<IO::AIO::FALLOC_FL_KEEP_SIZE> to allocate
		660	space, or C<IO::AIO::FALLOC_FL_PUNCH_HOLE | IO::AIO::FALLOC_FL_KEEP_SIZE>,
		661	to deallocate a file range.
		662
		663	IO::AIO also supports C<FALLOC_FL_COLLAPSE_RANGE>, to remove a range
		664	(without leaving a hole), C<FALLOC_FL_ZERO_RANGE>, to zero a range,
		665	C<FALLOC_FL_INSERT_RANGE> to insert a range and C<FALLOC_FL_UNSHARE_RANGE>
		666	to unshare shared blocks (see your L<fallocate(2)> manpage).
		667
		668	The file system block size used by C<fallocate> is presumably the
		669	C<f_bsize> returned by C<statvfs>, but different filesystems and filetypes
		670	can dictate other limitations.
		671
		672	If C<fallocate> isn't available or cannot be emulated (currently no
		673	emulation will be attempted), passes C<-1> and sets C<$!> to C<ENOSYS>.
		674
		675
		676	=item aio_chmod $fh_or_path, $mode, $callback->($status)
		677
		678	Works like perl's C<chmod> function.
		679
		680
192	=item aio_unlink $pathname, $callback	681	=item aio_unlink $pathname, $callback->($status)
193		682
194	Asynchronously unlink (delete) a file and call the callback with the	683	Asynchronously unlink (delete) a file and call the callback with the
195	result code.	684	result code.
196		685
		686
		687	=item aio_mknod $pathname, $mode, $dev, $callback->($status)
		688
		689	[EXPERIMENTAL]
		690
		691	Asynchronously create a device node (or fifo). See mknod(2).
		692
		693	The only (POSIX-) portable way of calling this function is:
		694
		695	aio_mknod $pathname, IO::AIO::S_IFIFO | $mode, 0, sub { ...
		696
		697	See C<aio_stat> for info about some potentially helpful extra constants
		698	and functions.
		699
		700	=item aio_link $srcpath, $dstpath, $callback->($status)
		701
		702	Asynchronously create a new link to the existing object at C<$srcpath> at
		703	the path C<$dstpath> and call the callback with the result code.
		704
		705
		706	=item aio_symlink $srcpath, $dstpath, $callback->($status)
		707
		708	Asynchronously create a new symbolic link to the existing object at C<$srcpath> at
		709	the path C<$dstpath> and call the callback with the result code.
		710
		711
		712	=item aio_readlink $pathname, $callback->($link)
		713
		714	Asynchronously read the symlink specified by C<$path> and pass it to
		715	the callback. If an error occurs, nothing or undef gets passed to the
		716	callback.
		717
		718
		719	=item aio_realpath $pathname, $callback->($path)
		720
		721	Asynchronously make the path absolute and resolve any symlinks in
		722	C<$path>. The resulting path only consists of directories (same as
		723	L<Cwd::realpath>).
		724
		725	This request can be used to get the absolute path of the current working
		726	directory by passing it a path of F<.> (a single dot).
		727
		728
		729	=item aio_rename $srcpath, $dstpath, $callback->($status)
		730
		731	Asynchronously rename the object at C<$srcpath> to C<$dstpath>, just as
		732	rename(2) and call the callback with the result code.
		733
		734	On systems that support the AIO::WD working directory abstraction
		735	natively, the case C<[$wd, "."]> as C<$srcpath> is specialcased - instead
		736	of failing, C<rename> is called on the absolute path of C<$wd>.
		737
		738
		739	=item aio_rename2 $srcpath, $dstpath, $flags, $callback->($status)
		740
		741	Basically a version of C<aio_rename> with an additional C<$flags>
		742	argument. Calling this with C<$flags=0> is the same as calling
		743	C<aio_rename>.
		744
		745	Non-zero flags are currently only supported on GNU/Linux systems that
		746	support renameat2. Other systems fail with C<ENOSYS> in this case.
		747
		748	The following constants are available (missing ones are, as usual C<0>),
		749	see renameat2(2) for details:
		750
		751	C<IO::AIO::RENAME_NOREPLACE>, C<IO::AIO::RENAME_EXCHANGE>
		752	and C<IO::AIO::RENAME_WHITEOUT>.
		753
		754
		755	=item aio_mkdir $pathname, $mode, $callback->($status)
		756
		757	Asynchronously mkdir (create) a directory and call the callback with
		758	the result code. C<$mode> will be modified by the umask at the time the
		759	request is executed, so do not change your umask.
		760
		761
		762	=item aio_rmdir $pathname, $callback->($status)
		763
		764	Asynchronously rmdir (delete) a directory and call the callback with the
		765	result code.
		766
		767	On systems that support the AIO::WD working directory abstraction
		768	natively, the case C<[$wd, "."]> is specialcased - instead of failing,
		769	C<rmdir> is called on the absolute path of C<$wd>.
		770
		771
		772	=item aio_readdir $pathname, $callback->($entries)
		773
		774	Unlike the POSIX call of the same name, C<aio_readdir> reads an entire
		775	directory (i.e. opendir + readdir + closedir). The entries will not be
		776	sorted, and will B<NOT> include the C<.> and C<..> entries.
		777
		778	The callback is passed a single argument which is either C<undef> or an
		779	array-ref with the filenames.
		780
		781
		782	=item aio_readdirx $pathname, $flags, $callback->($entries, $flags)
		783
		784	Quite similar to C<aio_readdir>, but the C<$flags> argument allows one to
		785	tune behaviour and output format. In case of an error, C<$entries> will be
		786	C<undef>.
		787
		788	The flags are a combination of the following constants, ORed together (the
		789	flags will also be passed to the callback, possibly modified):
		790
		791	=over 4
		792
		793	=item IO::AIO::READDIR_DENTS
		794
		795	Normally the callback gets an arrayref consisting of names only (as
		796	with C<aio_readdir>). If this flag is set, then the callback gets an
		797	arrayref with C<[$name, $type, $inode]> arrayrefs, each describing a
		798	single directory entry in more detail:
		799
		800	C<$name> is the name of the entry.
		801
		802	C<$type> is one of the C<IO::AIO::DT_xxx> constants:
		803
		804	C<IO::AIO::DT_UNKNOWN>, C<IO::AIO::DT_FIFO>, C<IO::AIO::DT_CHR>, C<IO::AIO::DT_DIR>,
		805	C<IO::AIO::DT_BLK>, C<IO::AIO::DT_REG>, C<IO::AIO::DT_LNK>, C<IO::AIO::DT_SOCK>,
		806	C<IO::AIO::DT_WHT>.
		807
		808	C<IO::AIO::DT_UNKNOWN> means just that: readdir does not know. If you need
		809	to know, you have to run stat yourself. Also, for speed/memory reasons,
		810	the C<$type> scalars are read-only: you must not modify them.
		811
		812	C<$inode> is the inode number (which might not be exact on systems with 64
		813	bit inode numbers and 32 bit perls). This field has unspecified content on
		814	systems that do not deliver the inode information.
		815
		816	=item IO::AIO::READDIR_DIRS_FIRST
		817
		818	When this flag is set, then the names will be returned in an order where
		819	likely directories come first, in optimal stat order. This is useful when
		820	you need to quickly find directories, or you want to find all directories
		821	while avoiding to stat() each entry.
		822
		823	If the system returns type information in readdir, then this is used
		824	to find directories directly. Otherwise, likely directories are names
		825	beginning with ".", or otherwise names with no dots, of which names with
		826	short names are tried first.
		827
		828	=item IO::AIO::READDIR_STAT_ORDER
		829
		830	When this flag is set, then the names will be returned in an order
		831	suitable for stat()'ing each one. That is, when you plan to stat() most or
		832	all files in the given directory, then the returned order will likely be
		833	faster.
		834
		835	If both this flag and C<IO::AIO::READDIR_DIRS_FIRST> are specified,
		836	then the likely dirs come first, resulting in a less optimal stat order
		837	for stat'ing all entries, but likely a more optimal order for finding
		838	subdirectories.
		839
		840	=item IO::AIO::READDIR_FOUND_UNKNOWN
		841
		842	This flag should not be set when calling C<aio_readdirx>. Instead, it
		843	is being set by C<aio_readdirx>, when any of the C<$type>'s found were
		844	C<IO::AIO::DT_UNKNOWN>. The absence of this flag therefore indicates that all
		845	C<$type>'s are known, which can be used to speed up some algorithms.
		846
		847	=back
		848
		849
		850	=item aio_slurp $pathname, $offset, $length, $data, $callback->($status)
		851
		852	Opens, reads and closes the given file. The data is put into C<$data>,
		853	which is resized as required.
		854
		855	If C<$offset> is negative, then it is counted from the end of the file.
		856
		857	If C<$length> is zero, then the remaining length of the file is
		858	used. Also, in this case, the same limitations to modifying C<$data> apply
		859	as when IO::AIO::mmap is used, i.e. it must only be modified in-place
		860	with C<substr>. If the size of the file is known, specifying a non-zero
		861	C<$length> results in a performance advantage.
		862
		863	This request is similar to the older C<aio_load> request, but since it is
		864	a single request, it might be more efficient to use.
		865
		866	Example: load F</etc/passwd> into C<$passwd>.
		867
		868	my $passwd;
		869	aio_slurp "/etc/passwd", 0, 0, $passwd, sub {
		870	$_[0] >= 0
		871	or die "/etc/passwd: $!\n";
		872
		873	printf "/etc/passwd is %d bytes long, and contains:\n", length $passwd;
		874	print $passwd;
		875	};
		876	IO::AIO::flush;
		877
		878
		879	=item aio_load $pathname, $data, $callback->($status)
		880
		881	This is a composite request that tries to fully load the given file into
		882	memory. Status is the same as with aio_read.
		883
		884	Using C<aio_slurp> might be more efficient, as it is a single request.
		885
		886	=cut
		887
		888	sub aio_load($$;$) {
		889	my ($path, undef, $cb) = @_;
		890	my $data = \$_[1];
		891
		892	my $pri = aioreq_pri;
		893	my $grp = aio_group $cb;
		894
		895	aioreq_pri $pri;
		896	add $grp aio_open $path, O_RDONLY, 0, sub {
		897	my $fh = shift
		898	or return $grp->result (-1);
		899
		900	aioreq_pri $pri;
		901	add $grp aio_read $fh, 0, (-s $fh), $$data, 0, sub {
		902	$grp->result ($_[0]);
		903	};
		904	};
		905
		906	$grp
		907	}
		908
		909	=item aio_copy $srcpath, $dstpath, $callback->($status)
		910
		911	Try to copy the I<file> (directories not supported as either source or
		912	destination) from C<$srcpath> to C<$dstpath> and call the callback with
		913	a status of C<0> (ok) or C<-1> (error, see C<$!>).
		914
		915	Existing destination files will be truncated.
		916
		917	This is a composite request that creates the destination file with
		918	mode 0200 and copies the contents of the source file into it using
		919	C<aio_sendfile>, followed by restoring atime, mtime, access mode and
		920	uid/gid, in that order.
		921
		922	If an error occurs, the partial destination file will be unlinked, if
		923	possible, except when setting atime, mtime, access mode and uid/gid, where
		924	errors are being ignored.
		925
		926	=cut
		927
		928	sub aio_copy($$;$) {
		929	my ($src, $dst, $cb) = @_;
		930
		931	my $pri = aioreq_pri;
		932	my $grp = aio_group $cb;
		933
		934	aioreq_pri $pri;
		935	add $grp aio_open $src, O_RDONLY, 0, sub {
		936	if (my $src_fh = $_[0]) {
		937	my @stat = stat $src_fh; # hmm, might block over nfs?
		938
		939	aioreq_pri $pri;
		940	add $grp aio_open $dst, O_CREAT | O_WRONLY | O_TRUNC, 0200, sub {
		941	if (my $dst_fh = $_[0]) {
		942	aioreq_pri $pri;
		943	add $grp aio_sendfile $dst_fh, $src_fh, 0, $stat[7], sub {
		944	if ($_[0] == $stat[7]) {
		945	$grp->result (0);
		946	close $src_fh;
		947
		948	my $ch = sub {
		949	aioreq_pri $pri;
		950	add $grp aio_chmod $dst_fh, $stat[2] & 07777, sub {
		951	aioreq_pri $pri;
		952	add $grp aio_chown $dst_fh, $stat[4], $stat[5], sub {
		953	aioreq_pri $pri;
		954	add $grp aio_close $dst_fh;
		955	}
		956	};
		957	};
		958
		959	aioreq_pri $pri;
		960	add $grp aio_utime $dst_fh, $stat[8], $stat[9], sub {
		961	if ($_[0] < 0 && $! == ENOSYS) {
		962	aioreq_pri $pri;
		963	add $grp aio_utime $dst, $stat[8], $stat[9], $ch;
		964	} else {
		965	$ch->();
		966	}
		967	};
		968	} else {
		969	$grp->result (-1);
		970	close $src_fh;
		971	close $dst_fh;
		972
		973	aioreq $pri;
		974	add $grp aio_unlink $dst;
		975	}
		976	};
		977	} else {
		978	$grp->result (-1);
		979	}
		980	},
		981
		982	} else {
		983	$grp->result (-1);
		984	}
		985	};
		986
		987	$grp
		988	}
		989
		990	=item aio_move $srcpath, $dstpath, $callback->($status)
		991
		992	Try to move the I<file> (directories not supported as either source or
		993	destination) from C<$srcpath> to C<$dstpath> and call the callback with
		994	a status of C<0> (ok) or C<-1> (error, see C<$!>).
		995
		996	This is a composite request that tries to rename(2) the file first; if
		997	rename fails with C<EXDEV>, it copies the file with C<aio_copy> and, if
		998	that is successful, unlinks the C<$srcpath>.
		999
		1000	=cut
		1001
		1002	sub aio_move($$;$) {
		1003	my ($src, $dst, $cb) = @_;
		1004
		1005	my $pri = aioreq_pri;
		1006	my $grp = aio_group $cb;
		1007
		1008	aioreq_pri $pri;
		1009	add $grp aio_rename $src, $dst, sub {
		1010	if ($_[0] && $! == EXDEV) {
		1011	aioreq_pri $pri;
		1012	add $grp aio_copy $src, $dst, sub {
		1013	$grp->result ($_[0]);
		1014
		1015	unless ($_[0]) {
		1016	aioreq_pri $pri;
		1017	add $grp aio_unlink $src;
		1018	}
		1019	};
		1020	} else {
		1021	$grp->result ($_[0]);
		1022	}
		1023	};
		1024
		1025	$grp
		1026	}
		1027
		1028	=item aio_scandir $pathname, $maxreq, $callback->($dirs, $nondirs)
		1029
		1030	Scans a directory (similar to C<aio_readdir>) but additionally tries to
		1031	efficiently separate the entries of directory C<$path> into two sets of
		1032	names, directories you can recurse into (directories), and ones you cannot
		1033	recurse into (everything else, including symlinks to directories).
		1034
		1035	C<aio_scandir> is a composite request that generates many sub requests.
		1036	C<$maxreq> specifies the maximum number of outstanding aio requests that
		1037	this function generates. If it is C<< <= 0 >>, then a suitable default
		1038	will be chosen (currently 4).
		1039
		1040	On error, the callback is called without arguments, otherwise it receives
		1041	two array-refs with path-relative entry names.
		1042
		1043	Example:
		1044
		1045	aio_scandir $dir, 0, sub {
		1046	my ($dirs, $nondirs) = @_;
		1047	print "real directories: @$dirs\n";
		1048	print "everything else: @$nondirs\n";
		1049	};
		1050
		1051	Implementation notes.
		1052
		1053	The C<aio_readdir> cannot be avoided, but C<stat()>'ing every entry can.
		1054
		1055	If readdir returns file type information, then this is used directly to
		1056	find directories.
		1057
		1058	Otherwise, after reading the directory, the modification time, size etc.
		1059	of the directory before and after the readdir is checked, and if they
		1060	match (and isn't the current time), the link count will be used to decide
		1061	how many entries are directories (if >= 2). Otherwise, no knowledge of the
		1062	number of subdirectories will be assumed.
		1063
		1064	Then entries will be sorted into likely directories a non-initial dot
		1065	currently) and likely non-directories (see C<aio_readdirx>). Then every
		1066	entry plus an appended C</.> will be C<stat>'ed, likely directories first,
		1067	in order of their inode numbers. If that succeeds, it assumes that the
		1068	entry is a directory or a symlink to directory (which will be checked
		1069	separately). This is often faster than stat'ing the entry itself because
		1070	filesystems might detect the type of the entry without reading the inode
		1071	data (e.g. ext2fs filetype feature), even on systems that cannot return
		1072	the filetype information on readdir.
		1073
		1074	If the known number of directories (link count - 2) has been reached, the
		1075	rest of the entries is assumed to be non-directories.
		1076
		1077	This only works with certainty on POSIX (= UNIX) filesystems, which
		1078	fortunately are the vast majority of filesystems around.
		1079
		1080	It will also likely work on non-POSIX filesystems with reduced efficiency
		1081	as those tend to return 0 or 1 as link counts, which disables the
		1082	directory counting heuristic.
		1083
		1084	=cut
		1085
		1086	sub aio_scandir($$;$) {
		1087	my ($path, $maxreq, $cb) = @_;
		1088
		1089	my $pri = aioreq_pri;
		1090
		1091	my $grp = aio_group $cb;
		1092
		1093	$maxreq = 4 if $maxreq <= 0;
		1094
		1095	# get a wd object
		1096	aioreq_pri $pri;
		1097	add $grp aio_wd $path, sub {
		1098	$_[0]
		1099	or return $grp->result ();
		1100
		1101	my $wd = [shift, "."];
		1102
		1103	# stat once
		1104	aioreq_pri $pri;
		1105	add $grp aio_stat $wd, sub {
		1106	return $grp->result () if $_[0];
		1107	my $now = time;
		1108	my $hash1 = join ":", (stat _)[0,1,3,7,9];
		1109
		1110	# read the directory entries
		1111	aioreq_pri $pri;
		1112	add $grp aio_readdirx $wd, READDIR_DIRS_FIRST, sub {
		1113	my $entries = shift
		1114	or return $grp->result ();
		1115
		1116	# stat the dir another time
		1117	aioreq_pri $pri;
		1118	add $grp aio_stat $wd, sub {
		1119	my $hash2 = join ":", (stat _)[0,1,3,7,9];
		1120
		1121	my $ndirs;
		1122
		1123	# take the slow route if anything looks fishy
		1124	if ($hash1 ne $hash2 or (stat _)[9] == $now) {
		1125	$ndirs = -1;
		1126	} else {
		1127	# if nlink == 2, we are finished
		1128	# for non-posix-fs's, we rely on nlink < 2
		1129	$ndirs = (stat _)[3] - 2
		1130	or return $grp->result ([], $entries);
		1131	}
		1132
		1133	my (@dirs, @nondirs);
		1134
		1135	my $statgrp = add $grp aio_group sub {
		1136	$grp->result (\@dirs, \@nondirs);
		1137	};
		1138
		1139	limit $statgrp $maxreq;
		1140	feed $statgrp sub {
		1141	return unless @$entries;
		1142	my $entry = shift @$entries;
		1143
		1144	aioreq_pri $pri;
		1145	$wd->[1] = "$entry/.";
		1146	add $statgrp aio_stat $wd, sub {
		1147	if ($_[0] < 0) {
		1148	push @nondirs, $entry;
		1149	} else {
		1150	# need to check for real directory
		1151	aioreq_pri $pri;
		1152	$wd->[1] = $entry;
		1153	add $statgrp aio_lstat $wd, sub {
		1154	if (-d _) {
		1155	push @dirs, $entry;
		1156
		1157	unless (--$ndirs) {
		1158	push @nondirs, @$entries;
		1159	feed $statgrp;
		1160	}
		1161	} else {
		1162	push @nondirs, $entry;
		1163	}
		1164	}
		1165	}
		1166	};
		1167	};
		1168	};
		1169	};
		1170	};
		1171	};
		1172
		1173	$grp
		1174	}
		1175
		1176	=item aio_rmtree $pathname, $callback->($status)
		1177
		1178	Delete a directory tree starting (and including) C<$path>, return the
		1179	status of the final C<rmdir> only. This is a composite request that
		1180	uses C<aio_scandir> to recurse into and rmdir directories, and unlink
		1181	everything else.
		1182
		1183	=cut
		1184
		1185	sub aio_rmtree;
		1186	sub aio_rmtree($;$) {
		1187	my ($path, $cb) = @_;
		1188
		1189	my $pri = aioreq_pri;
		1190	my $grp = aio_group $cb;
		1191
		1192	aioreq_pri $pri;
		1193	add $grp aio_scandir $path, 0, sub {
		1194	my ($dirs, $nondirs) = @_;
		1195
		1196	my $dirgrp = aio_group sub {
		1197	add $grp aio_rmdir $path, sub {
		1198	$grp->result ($_[0]);
		1199	};
		1200	};
		1201
		1202	(aioreq_pri $pri), add $dirgrp aio_rmtree "$path/$_" for @$dirs;
		1203	(aioreq_pri $pri), add $dirgrp aio_unlink "$path/$_" for @$nondirs;
		1204
		1205	add $grp $dirgrp;
		1206	};
		1207
		1208	$grp
		1209	}
		1210
		1211	=item aio_fcntl $fh, $cmd, $arg, $callback->($status)
		1212
		1213	=item aio_ioctl $fh, $request, $buf, $callback->($status)
		1214
		1215	These work just like the C<fcntl> and C<ioctl> built-in functions, except
		1216	they execute asynchronously and pass the return value to the callback.
		1217
		1218	Both calls can be used for a lot of things, some of which make more sense
		1219	to run asynchronously in their own thread, while some others make less
		1220	sense. For example, calls that block waiting for external events, such
		1221	as locking, will also lock down an I/O thread while it is waiting, which
		1222	can deadlock the whole I/O system. At the same time, there might be no
		1223	alternative to using a thread to wait.
		1224
		1225	So in general, you should only use these calls for things that do
		1226	(filesystem) I/O, not for things that wait for other events (network,
		1227	other processes), although if you are careful and know what you are doing,
		1228	you still can.
		1229
		1230	The following constants are available (missing ones are, as usual C<0>):
		1231
		1232	C<F_DUPFD_CLOEXEC>,
		1233
		1234	C<F_OFD_GETLK>, C<F_OFD_SETLK>, C<F_OFD_GETLKW>,
		1235
		1236	C<FIFREEZE>, C<FITHAW>, C<FITRIM>, C<FICLONE>, C<FICLONERANGE>, C<FIDEDUPERANGE>.
		1237
		1238	C<FS_IOC_GETFLAGS>, C<FS_IOC_SETFLAGS>, C<FS_IOC_GETVERSION>, C<FS_IOC_SETVERSION>,
		1239	C<FS_IOC_FIEMAP>.
		1240
		1241	C<FS_IOC_FSGETXATTR>, C<FS_IOC_FSSETXATTR>, C<FS_IOC_SET_ENCRYPTION_POLICY>,
		1242	C<FS_IOC_GET_ENCRYPTION_PWSALT>, C<FS_IOC_GET_ENCRYPTION_POLICY>, C<FS_KEY_DESCRIPTOR_SIZE>.
		1243
		1244	C<FS_SECRM_FL>, C<FS_UNRM_FL>, C<FS_COMPR_FL>, C<FS_SYNC_FL>, C<FS_IMMUTABLE_FL>,
		1245	C<FS_APPEND_FL>, C<FS_NODUMP_FL>, C<FS_NOATIME_FL>, C<FS_DIRTY_FL>,
		1246	C<FS_COMPRBLK_FL>, C<FS_NOCOMP_FL>, C<FS_ENCRYPT_FL>, C<FS_BTREE_FL>,
		1247	C<FS_INDEX_FL>, C<FS_JOURNAL_DATA_FL>, C<FS_NOTAIL_FL>, C<FS_DIRSYNC_FL>, C<FS_TOPDIR_FL>,
		1248	C<FS_FL_USER_MODIFIABLE>.
		1249
		1250	C<FS_XFLAG_REALTIME>, C<FS_XFLAG_PREALLOC>, C<FS_XFLAG_IMMUTABLE>, C<FS_XFLAG_APPEND>,
		1251	C<FS_XFLAG_SYNC>, C<FS_XFLAG_NOATIME>, C<FS_XFLAG_NODUMP>, C<FS_XFLAG_RTINHERIT>,
		1252	C<FS_XFLAG_PROJINHERIT>, C<FS_XFLAG_NOSYMLINKS>, C<FS_XFLAG_EXTSIZE>, C<FS_XFLAG_EXTSZINHERIT>,
		1253	C<FS_XFLAG_NODEFRAG>, C<FS_XFLAG_FILESTREAM>, C<FS_XFLAG_DAX>, C<FS_XFLAG_HASATTR>,
		1254
		1255	=item aio_sync $callback->($status)
		1256
		1257	Asynchronously call sync and call the callback when finished.
		1258
197	=item aio_fsync $fh, $callback	1259	=item aio_fsync $fh, $callback->($status)
198		1260
199	Asynchronously call fsync on the given filehandle and call the callback	1261	Asynchronously call fsync on the given filehandle and call the callback
200	with the fsync result code.	1262	with the fsync result code.
201		1263
202	=item aio_fdatasync $fh, $callback	1264	=item aio_fdatasync $fh, $callback->($status)
203		1265
204	Asynchronously call fdatasync on the given filehandle and call the	1266	Asynchronously call fdatasync on the given filehandle and call the
205	callback with the fdatasync result code. Might set C<$!> to C<ENOSYS> if	1267	callback with the fdatasync result code.
206	C<fdatasync> is not available.	1268
		1269	If this call isn't available because your OS lacks it or it couldn't be
		1270	detected, it will be emulated by calling C<fsync> instead.
		1271
		1272	=item aio_syncfs $fh, $callback->($status)
		1273
		1274	Asynchronously call the syncfs syscall to sync the filesystem associated
		1275	to the given filehandle and call the callback with the syncfs result
		1276	code. If syncfs is not available, calls sync(), but returns C<-1> and sets
		1277	errno to C<ENOSYS> nevertheless.
		1278
		1279	=item aio_sync_file_range $fh, $offset, $nbytes, $flags, $callback->($status)
		1280
		1281	Sync the data portion of the file specified by C<$offset> and C<$length>
		1282	to disk (but NOT the metadata), by calling the Linux-specific
		1283	sync_file_range call. If sync_file_range is not available or it returns
		1284	ENOSYS, then fdatasync or fsync is being substituted.
		1285
		1286	C<$flags> can be a combination of C<IO::AIO::SYNC_FILE_RANGE_WAIT_BEFORE>,
		1287	C<IO::AIO::SYNC_FILE_RANGE_WRITE> and
		1288	C<IO::AIO::SYNC_FILE_RANGE_WAIT_AFTER>: refer to the sync_file_range
		1289	manpage for details.
		1290
		1291	=item aio_pathsync $pathname, $callback->($status)
		1292
		1293	This request tries to open, fsync and close the given path. This is a
		1294	composite request intended to sync directories after directory operations
		1295	(E.g. rename). This might not work on all operating systems or have any
		1296	specific effect, but usually it makes sure that directory changes get
		1297	written to disc. It works for anything that can be opened for read-only,
		1298	not just directories.
		1299
		1300	Future versions of this function might fall back to other methods when
		1301	C<fsync> on the directory fails (such as calling C<sync>).
		1302
		1303	Passes C<0> when everything went ok, and C<-1> on error.
		1304
		1305	=cut
		1306
		1307	sub aio_pathsync($;$) {
		1308	my ($path, $cb) = @_;
		1309
		1310	my $pri = aioreq_pri;
		1311	my $grp = aio_group $cb;
		1312
		1313	aioreq_pri $pri;
		1314	add $grp aio_open $path, O_RDONLY, 0, sub {
		1315	my ($fh) = @_;
		1316	if ($fh) {
		1317	aioreq_pri $pri;
		1318	add $grp aio_fsync $fh, sub {
		1319	$grp->result ($_[0]);
		1320
		1321	aioreq_pri $pri;
		1322	add $grp aio_close $fh;
		1323	};
		1324	} else {
		1325	$grp->result (-1);
		1326	}
		1327	};
		1328
		1329	$grp
		1330	}
		1331
		1332	=item aio_msync $scalar, $offset = 0, $length = undef, flags = MS_SYNC, $callback->($status)
		1333
		1334	This is a rather advanced IO::AIO call, which only works on mmap(2)ed
		1335	scalars (see the C<IO::AIO::mmap> function, although it also works on data
		1336	scalars managed by the L<Sys::Mmap> or L<Mmap> modules, note that the
		1337	scalar must only be modified in-place while an aio operation is pending on
		1338	it).
		1339
		1340	It calls the C<msync> function of your OS, if available, with the memory
		1341	area starting at C<$offset> in the string and ending C<$length> bytes
		1342	later. If C<$length> is negative, counts from the end, and if C<$length>
		1343	is C<undef>, then it goes till the end of the string. The flags can be
		1344	either C<IO::AIO::MS_ASYNC> or C<IO::AIO::MS_SYNC>, plus an optional
		1345	C<IO::AIO::MS_INVALIDATE>.
		1346
		1347	=item aio_mtouch $scalar, $offset = 0, $length = undef, flags = 0, $callback->($status)
		1348
		1349	This is a rather advanced IO::AIO call, which works best on mmap(2)ed
		1350	scalars.
		1351
		1352	It touches (reads or writes) all memory pages in the specified
		1353	range inside the scalar. All caveats and parameters are the same
		1354	as for C<aio_msync>, above, except for flags, which must be either
		1355	C<0> (which reads all pages and ensures they are instantiated) or
		1356	C<IO::AIO::MT_MODIFY>, which modifies the memory pages (by reading and
		1357	writing an octet from it, which dirties the page).
		1358
		1359	=item aio_mlock $scalar, $offset = 0, $length = undef, $callback->($status)
		1360
		1361	This is a rather advanced IO::AIO call, which works best on mmap(2)ed
		1362	scalars.
		1363
		1364	It reads in all the pages of the underlying storage into memory (if any)
		1365	and locks them, so they are not getting swapped/paged out or removed.
		1366
		1367	If C<$length> is undefined, then the scalar will be locked till the end.
		1368
		1369	On systems that do not implement C<mlock>, this function returns C<-1>
		1370	and sets errno to C<ENOSYS>.
		1371
		1372	Note that the corresponding C<munlock> is synchronous and is
		1373	documented under L<MISCELLANEOUS FUNCTIONS>.
		1374
		1375	Example: open a file, mmap and mlock it - both will be undone when
		1376	C<$data> gets destroyed.
		1377
		1378	open my $fh, "<", $path or die "$path: $!";
		1379	my $data;
		1380	IO::AIO::mmap $data, -s $fh, IO::AIO::PROT_READ, IO::AIO::MAP_SHARED, $fh;
		1381	aio_mlock $data; # mlock in background
		1382
		1383	=item aio_mlockall $flags, $callback->($status)
		1384
		1385	Calls the C<mlockall> function with the given C<$flags> (a combination of
		1386	C<IO::AIO::MCL_CURRENT> and C<IO::AIO::MCL_FUTURE>).
		1387
		1388	On systems that do not implement C<mlockall>, this function returns C<-1>
		1389	and sets errno to C<ENOSYS>.
		1390
		1391	Note that the corresponding C<munlockall> is synchronous and is
		1392	documented under L<MISCELLANEOUS FUNCTIONS>.
		1393
		1394	Example: asynchronously lock all current and future pages into memory.
		1395
		1396	aio_mlockall IO::AIO::MCL_FUTURE;
		1397
		1398	=item aio_fiemap $fh, $start, $length, $flags, $count, $cb->(\@extents)
		1399
		1400	Queries the extents of the given file (by calling the Linux C<FIEMAP>
		1401	ioctl, see L<http://cvs.schmorp.de/IO-AIO/doc/fiemap.txt> for details). If
		1402	the ioctl is not available on your OS, then this request will fail with
		1403	C<ENOSYS>.
		1404
		1405	C<$start> is the starting offset to query extents for, C<$length> is the
		1406	size of the range to query - if it is C<undef>, then the whole file will
		1407	be queried.
		1408
		1409	C<$flags> is a combination of flags (C<IO::AIO::FIEMAP_FLAG_SYNC> or
		1410	C<IO::AIO::FIEMAP_FLAG_XATTR> - C<IO::AIO::FIEMAP_FLAGS_COMPAT> is also
		1411	exported), and is normally C<0> or C<IO::AIO::FIEMAP_FLAG_SYNC> to query
		1412	the data portion.
		1413
		1414	C<$count> is the maximum number of extent records to return. If it is
		1415	C<undef>, then IO::AIO queries all extents of the range. As a very special
		1416	case, if it is C<0>, then the callback receives the number of extents
		1417	instead of the extents themselves (which is unreliable, see below).
		1418
		1419	If an error occurs, the callback receives no arguments. The special
		1420	C<errno> value C<IO::AIO::EBADR> is available to test for flag errors.
		1421
		1422	Otherwise, the callback receives an array reference with extent
		1423	structures. Each extent structure is an array reference itself, with the
		1424	following members:
		1425
		1426	[$logical, $physical, $length, $flags]
		1427
		1428	Flags is any combination of the following flag values (typically either C<0>
		1429	or C<IO::AIO::FIEMAP_EXTENT_LAST> (1)):
		1430
		1431	C<IO::AIO::FIEMAP_EXTENT_LAST>, C<IO::AIO::FIEMAP_EXTENT_UNKNOWN>,
		1432	C<IO::AIO::FIEMAP_EXTENT_DELALLOC>, C<IO::AIO::FIEMAP_EXTENT_ENCODED>,
		1433	C<IO::AIO::FIEMAP_EXTENT_DATA_ENCRYPTED>, C<IO::AIO::FIEMAP_EXTENT_NOT_ALIGNED>,
		1434	C<IO::AIO::FIEMAP_EXTENT_DATA_INLINE>, C<IO::AIO::FIEMAP_EXTENT_DATA_TAIL>,
		1435	C<IO::AIO::FIEMAP_EXTENT_UNWRITTEN>, C<IO::AIO::FIEMAP_EXTENT_MERGED> or
		1436	C<IO::AIO::FIEMAP_EXTENT_SHARED>.
		1437
		1438	At the time of this writing (Linux 3.2), this request is unreliable unless
		1439	C<$count> is C<undef>, as the kernel has all sorts of bugs preventing
		1440	it to return all extents of a range for files with a large number of
		1441	extents. The code (only) works around all these issues if C<$count> is
		1442	C<undef>.
		1443
		1444	=item aio_group $callback->(...)
		1445
		1446	This is a very special aio request: Instead of doing something, it is a
		1447	container for other aio requests, which is useful if you want to bundle
		1448	many requests into a single, composite, request with a definite callback
		1449	and the ability to cancel the whole request with its subrequests.
		1450
		1451	Returns an object of class L<IO::AIO::GRP>. See its documentation below
		1452	for more info.
		1453
		1454	Example:
		1455
		1456	my $grp = aio_group sub {
		1457	print "all stats done\n";
		1458	};
		1459
		1460	add $grp
		1461	(aio_stat ...),
		1462	(aio_stat ...),
		1463	...;
		1464
		1465	=item aio_nop $callback->()
		1466
		1467	This is a special request - it does nothing in itself and is only used for
		1468	side effects, such as when you want to add a dummy request to a group so
		1469	that finishing the requests in the group depends on executing the given
		1470	code.
		1471
		1472	While this request does nothing, it still goes through the execution
		1473	phase and still requires a worker thread. Thus, the callback will not
		1474	be executed immediately but only after other requests in the queue have
		1475	entered their execution phase. This can be used to measure request
		1476	latency.
		1477
		1478	=item IO::AIO::aio_busy $fractional_seconds, $callback->() *NOT EXPORTED*
		1479
		1480	Mainly used for debugging and benchmarking, this aio request puts one of
		1481	the request workers to sleep for the given time.
		1482
		1483	While it is theoretically handy to have simple I/O scheduling requests
		1484	like sleep and file handle readable/writable, the overhead this creates is
		1485	immense (it blocks a thread for a long time) so do not use this function
		1486	except to put your application under artificial I/O pressure.
207		1487
208	=back	1488	=back
209		1489
		1490
		1491	=head2 IO::AIO::WD - multiple working directories
		1492
		1493	Your process only has one current working directory, which is used by all
		1494	threads. This makes it hard to use relative paths (some other component
		1495	could call C<chdir> at any time, and it is hard to control when the path
		1496	will be used by IO::AIO).
		1497
		1498	One solution for this is to always use absolute paths. This usually works,
		1499	but can be quite slow (the kernel has to walk the whole path on every
		1500	access), and can also be a hassle to implement.
		1501
		1502	Newer POSIX systems have a number of functions (openat, fdopendir,
		1503	futimensat and so on) that make it possible to specify working directories
		1504	per operation.
		1505
		1506	For portability, and because the clowns who "designed", or shall I write,
		1507	perpetrated this new interface were obviously half-drunk, this abstraction
		1508	cannot be perfect, though.
		1509
		1510	IO::AIO allows you to convert directory paths into a so-called IO::AIO::WD
		1511	object. This object stores the canonicalised, absolute version of the
		1512	path, and on systems that allow it, also a directory file descriptor.
		1513
		1514	Everywhere where a pathname is accepted by IO::AIO (e.g. in C<aio_stat>
		1515	or C<aio_unlink>), one can specify an array reference with an IO::AIO::WD
		1516	object and a pathname instead (or the IO::AIO::WD object alone, which
		1517	gets interpreted as C<[$wd, "."]>). If the pathname is absolute, the
		1518	IO::AIO::WD object is ignored, otherwise the pathname is resolved relative
		1519	to that IO::AIO::WD object.
		1520
		1521	For example, to get a wd object for F</etc> and then stat F<passwd>
		1522	inside, you would write:
		1523
		1524	aio_wd "/etc", sub {
		1525	my $etcdir = shift;
		1526
		1527	# although $etcdir can be undef on error, there is generally no reason
		1528	# to check for errors here, as aio_stat will fail with ENOENT
		1529	# when $etcdir is undef.
		1530
		1531	aio_stat [$etcdir, "passwd"], sub {
		1532	# yay
		1533	};
		1534	};
		1535
		1536	The fact that C<aio_wd> is a request and not a normal function shows that
		1537	creating an IO::AIO::WD object is itself a potentially blocking operation,
		1538	which is why it is done asynchronously.
		1539
		1540	To stat the directory obtained with C<aio_wd> above, one could write
		1541	either of the following three request calls:
		1542
		1543	aio_lstat "/etc" , sub { ... # pathname as normal string
		1544	aio_lstat [$wd, "."], sub { ... # "." relative to $wd (i.e. $wd itself)
		1545	aio_lstat $wd , sub { ... # shorthand for the previous
		1546
		1547	As with normal pathnames, IO::AIO keeps a copy of the working directory
		1548	object and the pathname string, so you could write the following without
		1549	causing any issues due to C<$path> getting reused:
		1550
		1551	my $path = [$wd, undef];
		1552
		1553	for my $name (qw(abc def ghi)) {
		1554	$path->[1] = $name;
		1555	aio_stat $path, sub {
		1556	# ...
		1557	};
		1558	}
		1559
		1560	There are some caveats: when directories get renamed (or deleted), the
		1561	pathname string doesn't change, so will point to the new directory (or
		1562	nowhere at all), while the directory fd, if available on the system,
		1563	will still point to the original directory. Most functions accepting a
		1564	pathname will use the directory fd on newer systems, and the string on
		1565	older systems. Some functions (such as C<aio_realpath>) will always rely on
		1566	the string form of the pathname.
		1567
		1568	So this functionality is mainly useful to get some protection against
		1569	C<chdir>, to easily get an absolute path out of a relative path for future
		1570	reference, and to speed up doing many operations in the same directory
		1571	(e.g. when stat'ing all files in a directory).
		1572
		1573	The following functions implement this working directory abstraction:
		1574
		1575	=over 4
		1576
		1577	=item aio_wd $pathname, $callback->($wd)
		1578
		1579	Asynchonously canonicalise the given pathname and convert it to an
		1580	IO::AIO::WD object representing it. If possible and supported on the
		1581	system, also open a directory fd to speed up pathname resolution relative
		1582	to this working directory.
		1583
		1584	If something goes wrong, then C<undef> is passwd to the callback instead
		1585	of a working directory object and C<$!> is set appropriately. Since
		1586	passing C<undef> as working directory component of a pathname fails the
		1587	request with C<ENOENT>, there is often no need for error checking in the
		1588	C<aio_wd> callback, as future requests using the value will fail in the
		1589	expected way.
		1590
		1591	=item IO::AIO::CWD
		1592
		1593	This is a compiletime constant (object) that represents the process
		1594	current working directory.
		1595
		1596	Specifying this object as working directory object for a pathname is as if
		1597	the pathname would be specified directly, without a directory object. For
		1598	example, these calls are functionally identical:
		1599
		1600	aio_stat "somefile", sub { ... };
		1601	aio_stat [IO::AIO::CWD, "somefile"], sub { ... };
		1602
		1603	=back
		1604
		1605	To recover the path associated with an IO::AIO::WD object, you can use
		1606	C<aio_realpath>:
		1607
		1608	aio_realpath $wd, sub {
		1609	warn "path is $_[0]\n";
		1610	};
		1611
		1612	Currently, C<aio_statvfs> always, and C<aio_rename> and C<aio_rmdir>
		1613	sometimes, fall back to using an absolue path.
		1614
		1615	=head2 IO::AIO::REQ CLASS
		1616
		1617	All non-aggregate C<aio_*> functions return an object of this class when
		1618	called in non-void context.
		1619
		1620	=over 4
		1621
		1622	=item cancel $req
		1623
		1624	Cancels the request, if possible. Has the effect of skipping execution
		1625	when entering the B<execute> state and skipping calling the callback when
		1626	entering the the B<result> state, but will leave the request otherwise
		1627	untouched (with the exception of readdir). That means that requests that
		1628	currently execute will not be stopped and resources held by the request
		1629	will not be freed prematurely.
		1630
		1631	=item cb $req $callback->(...)
		1632
		1633	Replace (or simply set) the callback registered to the request.
		1634
		1635	=back
		1636
		1637	=head2 IO::AIO::GRP CLASS
		1638
		1639	This class is a subclass of L<IO::AIO::REQ>, so all its methods apply to
		1640	objects of this class, too.
		1641
		1642	A IO::AIO::GRP object is a special request that can contain multiple other
		1643	aio requests.
		1644
		1645	You create one by calling the C<aio_group> constructing function with a
		1646	callback that will be called when all contained requests have entered the
		1647	C<done> state:
		1648
		1649	my $grp = aio_group sub {
		1650	print "all requests are done\n";
		1651	};
		1652
		1653	You add requests by calling the C<add> method with one or more
		1654	C<IO::AIO::REQ> objects:
		1655
		1656	$grp->add (aio_unlink "...");
		1657
		1658	add $grp aio_stat "...", sub {
		1659	$_[0] or return $grp->result ("error");
		1660
		1661	# add another request dynamically, if first succeeded
		1662	add $grp aio_open "...", sub {
		1663	$grp->result ("ok");
		1664	};
		1665	};
		1666
		1667	This makes it very easy to create composite requests (see the source of
		1668	C<aio_move> for an application) that work and feel like simple requests.
		1669
		1670	=over 4
		1671
		1672	=item * The IO::AIO::GRP objects will be cleaned up during calls to
		1673	C<IO::AIO::poll_cb>, just like any other request.
		1674
		1675	=item * They can be canceled like any other request. Canceling will cancel not
		1676	only the request itself, but also all requests it contains.
		1677
		1678	=item * They can also can also be added to other IO::AIO::GRP objects.
		1679
		1680	=item * You must not add requests to a group from within the group callback (or
		1681	any later time).
		1682
		1683	=back
		1684
		1685	Their lifetime, simplified, looks like this: when they are empty, they
		1686	will finish very quickly. If they contain only requests that are in the
		1687	C<done> state, they will also finish. Otherwise they will continue to
		1688	exist.
		1689
		1690	That means after creating a group you have some time to add requests
		1691	(precisely before the callback has been invoked, which is only done within
		1692	the C<poll_cb>). And in the callbacks of those requests, you can add
		1693	further requests to the group. And only when all those requests have
		1694	finished will the the group itself finish.
		1695
		1696	=over 4
		1697
		1698	=item add $grp ...
		1699
		1700	=item $grp->add (...)
		1701
		1702	Add one or more requests to the group. Any type of L<IO::AIO::REQ> can
		1703	be added, including other groups, as long as you do not create circular
		1704	dependencies.
		1705
		1706	Returns all its arguments.
		1707
		1708	=item $grp->cancel_subs
		1709
		1710	Cancel all subrequests and clears any feeder, but not the group request
		1711	itself. Useful when you queued a lot of events but got a result early.
		1712
		1713	The group request will finish normally (you cannot add requests to the
		1714	group).
		1715
		1716	=item $grp->result (...)
		1717
		1718	Set the result value(s) that will be passed to the group callback when all
		1719	subrequests have finished and set the groups errno to the current value
		1720	of errno (just like calling C<errno> without an error number). By default,
		1721	no argument will be passed and errno is zero.
		1722
		1723	=item $grp->errno ([$errno])
		1724
		1725	Sets the group errno value to C<$errno>, or the current value of errno
		1726	when the argument is missing.
		1727
		1728	Every aio request has an associated errno value that is restored when
		1729	the callback is invoked. This method lets you change this value from its
		1730	default (0).
		1731
		1732	Calling C<result> will also set errno, so make sure you either set C<$!>
		1733	before the call to C<result>, or call c<errno> after it.
		1734
		1735	=item feed $grp $callback->($grp)
		1736
		1737	Sets a feeder/generator on this group: every group can have an attached
		1738	generator that generates requests if idle. The idea behind this is that,
		1739	although you could just queue as many requests as you want in a group,
		1740	this might starve other requests for a potentially long time. For example,
		1741	C<aio_scandir> might generate hundreds of thousands of C<aio_stat>
		1742	requests, delaying any later requests for a long time.
		1743
		1744	To avoid this, and allow incremental generation of requests, you can
		1745	instead a group and set a feeder on it that generates those requests. The
		1746	feed callback will be called whenever there are few enough (see C<limit>,
		1747	below) requests active in the group itself and is expected to queue more
		1748	requests.
		1749
		1750	The feed callback can queue as many requests as it likes (i.e. C<add> does
		1751	not impose any limits).
		1752
		1753	If the feed does not queue more requests when called, it will be
		1754	automatically removed from the group.
		1755
		1756	If the feed limit is C<0> when this method is called, it will be set to
		1757	C<2> automatically.
		1758
		1759	Example:
		1760
		1761	# stat all files in @files, but only ever use four aio requests concurrently:
		1762
		1763	my $grp = aio_group sub { print "finished\n" };
		1764	limit $grp 4;
		1765	feed $grp sub {
		1766	my $file = pop @files
		1767	or return;
		1768
		1769	add $grp aio_stat $file, sub { ... };
		1770	};
		1771
		1772	=item limit $grp $num
		1773
		1774	Sets the feeder limit for the group: The feeder will be called whenever
		1775	the group contains less than this many requests.
		1776
		1777	Setting the limit to C<0> will pause the feeding process.
		1778
		1779	The default value for the limit is C<0>, but note that setting a feeder
		1780	automatically bumps it up to C<2>.
		1781
		1782	=back
		1783
210	=head2 SUPPORT FUNCTIONS	1784	=head2 SUPPORT FUNCTIONS
211		1785
		1786	=head3 EVENT PROCESSING AND EVENT LOOP INTEGRATION
		1787
212	=over 4	1788	=over 4
213		1789
214	=item $fileno = IO::AIO::poll_fileno	1790	=item $fileno = IO::AIO::poll_fileno
215		1791
216	Return the I<request result pipe file descriptor>. This filehandle must be	1792	Return the I<request result pipe file descriptor>. This filehandle must be
217	polled for reading by some mechanism outside this module (e.g. Event or	1793	polled for reading by some mechanism outside this module (e.g. EV, Glib,
218	select, see below or the SYNOPSIS). If the pipe becomes readable you have	1794	select and so on, see below or the SYNOPSIS). If the pipe becomes readable
219	to call C<poll_cb> to check the results.	1795	you have to call C<poll_cb> to check the results.
220		1796
221	See C<poll_cb> for an example.	1797	See C<poll_cb> for an example.
222		1798
223	=item IO::AIO::poll_cb	1799	=item IO::AIO::poll_cb
224		1800
225	Process all outstanding events on the result pipe. You have to call this	1801	Process some requests that have reached the result phase (i.e. they have
226	regularly. Returns the number of events processed. Returns immediately	1802	been executed but the results are not yet reported). You have to call
227	when no events are outstanding.	1803	this "regularly" to finish outstanding requests.
		1804
		1805	Returns C<0> if all events could be processed (or there were no
		1806	events to process), or C<-1> if it returned earlier for whatever
		1807	reason. Returns immediately when no events are outstanding. The amount
		1808	of events processed depends on the settings of C<IO::AIO::max_poll_req>,
		1809	C<IO::AIO::max_poll_time> and C<IO::AIO::max_outstanding>.
		1810
		1811	If not all requests were processed for whatever reason, the poll file
		1812	descriptor will still be ready when C<poll_cb> returns, so normally you
		1813	don't have to do anything special to have it called later.
		1814
		1815	Apart from calling C<IO::AIO::poll_cb> when the event filehandle becomes
		1816	ready, it can be beneficial to call this function from loops which submit
		1817	a lot of requests, to make sure the results get processed when they become
		1818	available and not just when the loop is finished and the event loop takes
		1819	over again. This function returns very fast when there are no outstanding
		1820	requests.
228		1821
229	Example: Install an Event watcher that automatically calls	1822	Example: Install an Event watcher that automatically calls
230	IO::AIO::poll_cb with high priority:	1823	IO::AIO::poll_cb with high priority (more examples can be found in the
		1824	SYNOPSIS section, at the top of this document):
231		1825
232	Event->io (fd => IO::AIO::poll_fileno,	1826	Event->io (fd => IO::AIO::poll_fileno,
233	poll => 'r', async => 1,	1827	poll => 'r', async => 1,
234	cb => \&IO::AIO::poll_cb);	1828	cb => \&IO::AIO::poll_cb);
235		1829
236	=item IO::AIO::poll_wait	1830	=item IO::AIO::poll_wait
237		1831
238	Wait till the result filehandle becomes ready for reading (simply does a	1832	Wait until either at least one request is in the result phase or no
239	C<select> on the filehandle. This is useful if you want to synchronously wait	1833	requests are outstanding anymore.
240	for some requests to finish).	1834
		1835	This is useful if you want to synchronously wait for some requests to
		1836	become ready, without actually handling them.
241		1837
242	See C<nreqs> for an example.	1838	See C<nreqs> for an example.
243		1839
		1840	=item IO::AIO::poll
		1841
		1842	Waits until some requests have been handled.
		1843
		1844	Returns the number of requests processed, but is otherwise strictly
		1845	equivalent to:
		1846
		1847	IO::AIO::poll_wait, IO::AIO::poll_cb
		1848
244	=item IO::AIO::nreqs	1849	=item IO::AIO::flush
245		1850
246	Returns the number of requests currently outstanding (i.e. for which their	1851	Wait till all outstanding AIO requests have been handled.
247	callback has not been invoked yet).
248		1852
249	Example: wait till there are no outstanding requests anymore:	1853	Strictly equivalent to:
250		1854
251	IO::AIO::poll_wait, IO::AIO::poll_cb	1855	IO::AIO::poll_wait, IO::AIO::poll_cb
252	while IO::AIO::nreqs;	1856	while IO::AIO::nreqs;
253		1857
		1858	=item IO::AIO::max_poll_reqs $nreqs
		1859
		1860	=item IO::AIO::max_poll_time $seconds
		1861
		1862	These set the maximum number of requests (default C<0>, meaning infinity)
		1863	that are being processed by C<IO::AIO::poll_cb> in one call, respectively
		1864	the maximum amount of time (default C<0>, meaning infinity) spent in
		1865	C<IO::AIO::poll_cb> to process requests (more correctly the mininum amount
		1866	of time C<poll_cb> is allowed to use).
		1867
		1868	Setting C<max_poll_time> to a non-zero value creates an overhead of one
		1869	syscall per request processed, which is not normally a problem unless your
		1870	callbacks are really really fast or your OS is really really slow (I am
		1871	not mentioning Solaris here). Using C<max_poll_reqs> incurs no overhead.
		1872
		1873	Setting these is useful if you want to ensure some level of
		1874	interactiveness when perl is not fast enough to process all requests in
		1875	time.
		1876
		1877	For interactive programs, values such as C<0.01> to C<0.1> should be fine.
		1878
		1879	Example: Install an Event watcher that automatically calls
		1880	IO::AIO::poll_cb with low priority, to ensure that other parts of the
		1881	program get the CPU sometimes even under high AIO load.
		1882
		1883	# try not to spend much more than 0.1s in poll_cb
		1884	IO::AIO::max_poll_time 0.1;
		1885
		1886	# use a low priority so other tasks have priority
		1887	Event->io (fd => IO::AIO::poll_fileno,
		1888	poll => 'r', nice => 1,
		1889	cb => &IO::AIO::poll_cb);
		1890
		1891	=back
		1892
		1893	=head3 CONTROLLING THE NUMBER OF THREADS
		1894
		1895	=over
		1896
		1897	=item IO::AIO::min_parallel $nthreads
		1898
		1899	Set the minimum number of AIO threads to C<$nthreads>. The current
		1900	default is C<8>, which means eight asynchronous operations can execute
		1901	concurrently at any one time (the number of outstanding requests,
		1902	however, is unlimited).
		1903
		1904	IO::AIO starts threads only on demand, when an AIO request is queued and
		1905	no free thread exists. Please note that queueing up a hundred requests can
		1906	create demand for a hundred threads, even if it turns out that everything
		1907	is in the cache and could have been processed faster by a single thread.
		1908
		1909	It is recommended to keep the number of threads relatively low, as some
		1910	Linux kernel versions will scale negatively with the number of threads
		1911	(higher parallelity => MUCH higher latency). With current Linux 2.6
		1912	versions, 4-32 threads should be fine.
		1913
		1914	Under most circumstances you don't need to call this function, as the
		1915	module selects a default that is suitable for low to moderate load.
		1916
		1917	=item IO::AIO::max_parallel $nthreads
		1918
		1919	Sets the maximum number of AIO threads to C<$nthreads>. If more than the
		1920	specified number of threads are currently running, this function kills
		1921	them. This function blocks until the limit is reached.
		1922
		1923	While C<$nthreads> are zero, aio requests get queued but not executed
		1924	until the number of threads has been increased again.
		1925
		1926	This module automatically runs C<max_parallel 0> at program end, to ensure
		1927	that all threads are killed and that there are no outstanding requests.
		1928
		1929	Under normal circumstances you don't need to call this function.
		1930
		1931	=item IO::AIO::max_idle $nthreads
		1932
		1933	Limit the number of threads (default: 4) that are allowed to idle
		1934	(i.e., threads that did not get a request to process within the idle
		1935	timeout (default: 10 seconds). That means if a thread becomes idle while
		1936	C<$nthreads> other threads are also idle, it will free its resources and
		1937	exit.
		1938
		1939	This is useful when you allow a large number of threads (e.g. 100 or 1000)
		1940	to allow for extremely high load situations, but want to free resources
		1941	under normal circumstances (1000 threads can easily consume 30MB of RAM).
		1942
		1943	The default is probably ok in most situations, especially if thread
		1944	creation is fast. If thread creation is very slow on your system you might
		1945	want to use larger values.
		1946
		1947	=item IO::AIO::idle_timeout $seconds
		1948
		1949	Sets the minimum idle timeout (default 10) after which worker threads are
		1950	allowed to exit. SEe C<IO::AIO::max_idle>.
		1951
		1952	=item IO::AIO::max_outstanding $maxreqs
		1953
		1954	Sets the maximum number of outstanding requests to C<$nreqs>. If
		1955	you do queue up more than this number of requests, the next call to
		1956	C<IO::AIO::poll_cb> (and other functions calling C<poll_cb>, such as
		1957	C<IO::AIO::flush> or C<IO::AIO::poll>) will block until the limit is no
		1958	longer exceeded.
		1959
		1960	In other words, this setting does not enforce a queue limit, but can be
		1961	used to make poll functions block if the limit is exceeded.
		1962
		1963	This is a very bad function to use in interactive programs because it
		1964	blocks, and a bad way to reduce concurrency because it is inexact: Better
		1965	use an C<aio_group> together with a feed callback.
		1966
		1967	Its main use is in scripts without an event loop - when you want to stat
		1968	a lot of files, you can write something like this:
		1969
		1970	IO::AIO::max_outstanding 32;
		1971
		1972	for my $path (...) {
		1973	aio_stat $path , ...;
		1974	IO::AIO::poll_cb;
		1975	}
		1976
		1977	IO::AIO::flush;
		1978
		1979	The call to C<poll_cb> inside the loop will normally return instantly, but
		1980	as soon as more thna C<32> reqeusts are in-flight, it will block until
		1981	some requests have been handled. This keeps the loop from pushing a large
		1982	number of C<aio_stat> requests onto the queue.
		1983
		1984	The default value for C<max_outstanding> is very large, so there is no
		1985	practical limit on the number of outstanding requests.
		1986
		1987	=back
		1988
		1989	=head3 STATISTICAL INFORMATION
		1990
		1991	=over
		1992
254	=item IO::AIO::flush	1993	=item IO::AIO::nreqs
255		1994
256	Wait till all outstanding AIO requests have been handled.	1995	Returns the number of requests currently in the ready, execute or pending
		1996	states (i.e. for which their callback has not been invoked yet).
257		1997
258	Strictly equivalent to:	1998	Example: wait till there are no outstanding requests anymore:
259		1999
260	IO::AIO::poll_wait, IO::AIO::poll_cb	2000	IO::AIO::poll_wait, IO::AIO::poll_cb
261	while IO::AIO::nreqs;	2001	while IO::AIO::nreqs;
262		2002
		2003	=item IO::AIO::nready
		2004
		2005	Returns the number of requests currently in the ready state (not yet
		2006	executed).
		2007
263	=item IO::AIO::poll	2008	=item IO::AIO::npending
264		2009
265	Waits until some requests have been handled.	2010	Returns the number of requests currently in the pending state (executed,
266		2011	but not yet processed by poll_cb).
267	Strictly equivalent to:
268
269	IO::AIO::poll_wait, IO::AIO::poll_cb
270	if IO::AIO::nreqs;
271
272	=item IO::AIO::min_parallel $nthreads
273
274	Set the minimum number of AIO threads to C<$nthreads>. The default is
275	C<1>, which means a single asynchronous operation can be done at one time
276	(the number of outstanding operations, however, is unlimited).
277
278	It is recommended to keep the number of threads low, as some Linux
279	kernel versions will scale negatively with the number of threads (higher
280	parallelity => MUCH higher latency). With current Linux 2.6 versions, 4-32
281	threads should be fine.
282
283	Under normal circumstances you don't need to call this function, as this
284	module automatically starts some threads (the exact number might change,
285	and is currently 4).
286
287	=item IO::AIO::max_parallel $nthreads
288
289	Sets the maximum number of AIO threads to C<$nthreads>. If more than
290	the specified number of threads are currently running, kill them. This
291	function blocks until the limit is reached.
292
293	This module automatically runs C<max_parallel 0> at program end, to ensure
294	that all threads are killed and that there are no outstanding requests.
295
296	Under normal circumstances you don't need to call this function.
297
298	=item $oldnreqs = IO::AIO::max_outstanding $nreqs
299
300	Sets the maximum number of outstanding requests to C<$nreqs>. If you
301	try to queue up more than this number of requests, the caller will block until
302	some requests have been handled.
303
304	The default is very large, so normally there is no practical limit. If you
305	queue up many requests in a loop it it often improves speed if you set
306	this to a relatively low number, such as C<100>.
307
308	Under normal circumstances you don't need to call this function.
309		2012
310	=back	2013	=back
311		2014
		2015	=head3 MISCELLANEOUS FUNCTIONS
		2016
		2017	IO::AIO implements some functions that are useful when you want to use
		2018	some "Advanced I/O" function not available to in Perl, without going the
		2019	"Asynchronous I/O" route. Many of these have an asynchronous C<aio_*>
		2020	counterpart.
		2021
		2022	=over 4
		2023
		2024	=item $numfd = IO::AIO::get_fdlimit
		2025
		2026	This function is I<EXPERIMENTAL> and subject to change.
		2027
		2028	Tries to find the current file descriptor limit and returns it, or
		2029	C<undef> and sets C<$!> in case of an error. The limit is one larger than
		2030	the highest valid file descriptor number.
		2031
		2032	=item IO::AIO::min_fdlimit [$numfd]
		2033
		2034	This function is I<EXPERIMENTAL> and subject to change.
		2035
		2036	Try to increase the current file descriptor limit(s) to at least C<$numfd>
		2037	by changing the soft or hard file descriptor resource limit. If C<$numfd>
		2038	is missing, it will try to set a very high limit, although this is not
		2039	recommended when you know the actual minimum that you require.
		2040
		2041	If the limit cannot be raised enough, the function makes a best-effort
		2042	attempt to increase the limit as much as possible, using various
		2043	tricks, while still failing. You can query the resulting limit using
		2044	C<IO::AIO::get_fdlimit>.
		2045
		2046	If an error occurs, returns C<undef> and sets C<$!>, otherwise returns
		2047	true.
		2048
		2049	=item IO::AIO::sendfile $ofh, $ifh, $offset, $count
		2050
		2051	Calls the C<eio_sendfile_sync> function, which is like C<aio_sendfile>,
		2052	but is blocking (this makes most sense if you know the input data is
		2053	likely cached already and the output filehandle is set to non-blocking
		2054	operations).
		2055
		2056	Returns the number of bytes copied, or C<-1> on error.
		2057
		2058	=item IO::AIO::fadvise $fh, $offset, $len, $advice
		2059
		2060	Simply calls the C<posix_fadvise> function (see its
		2061	manpage for details). The following advice constants are
		2062	available: C<IO::AIO::FADV_NORMAL>, C<IO::AIO::FADV_SEQUENTIAL>,
		2063	C<IO::AIO::FADV_RANDOM>, C<IO::AIO::FADV_NOREUSE>,
		2064	C<IO::AIO::FADV_WILLNEED>, C<IO::AIO::FADV_DONTNEED>.
		2065
		2066	On systems that do not implement C<posix_fadvise>, this function returns
		2067	ENOSYS, otherwise the return value of C<posix_fadvise>.
		2068
		2069	=item IO::AIO::madvise $scalar, $offset, $len, $advice
		2070
		2071	Simply calls the C<posix_madvise> function (see its
		2072	manpage for details). The following advice constants are
		2073	available: C<IO::AIO::MADV_NORMAL>, C<IO::AIO::MADV_SEQUENTIAL>,
		2074	C<IO::AIO::MADV_RANDOM>, C<IO::AIO::MADV_WILLNEED>,
		2075	C<IO::AIO::MADV_DONTNEED>.
		2076
		2077	If C<$offset> is negative, counts from the end. If C<$length> is negative,
		2078	the remaining length of the C<$scalar> is used. If possible, C<$length>
		2079	will be reduced to fit into the C<$scalar>.
		2080
		2081	On systems that do not implement C<posix_madvise>, this function returns
		2082	ENOSYS, otherwise the return value of C<posix_madvise>.
		2083
		2084	=item IO::AIO::mprotect $scalar, $offset, $len, $protect
		2085
		2086	Simply calls the C<mprotect> function on the preferably AIO::mmap'ed
		2087	$scalar (see its manpage for details). The following protect
		2088	constants are available: C<IO::AIO::PROT_NONE>, C<IO::AIO::PROT_READ>,
		2089	C<IO::AIO::PROT_WRITE>, C<IO::AIO::PROT_EXEC>.
		2090
		2091	If C<$offset> is negative, counts from the end. If C<$length> is negative,
		2092	the remaining length of the C<$scalar> is used. If possible, C<$length>
		2093	will be reduced to fit into the C<$scalar>.
		2094
		2095	On systems that do not implement C<mprotect>, this function returns
		2096	ENOSYS, otherwise the return value of C<mprotect>.
		2097
		2098	=item IO::AIO::mmap $scalar, $length, $prot, $flags, $fh[, $offset]
		2099
		2100	Memory-maps a file (or anonymous memory range) and attaches it to the
		2101	given C<$scalar>, which will act like a string scalar. Returns true on
		2102	success, and false otherwise.
		2103
		2104	The scalar must exist, but its contents do not matter - this means you
		2105	cannot use a nonexistant array or hash element. When in doubt, C<undef>
		2106	the scalar first.
		2107
		2108	The only operations allowed on the mmapped scalar are C<substr>/C<vec>,
		2109	which don't change the string length, and most read-only operations such
		2110	as copying it or searching it with regexes and so on.
		2111
		2112	Anything else is unsafe and will, at best, result in memory leaks.
		2113
		2114	The memory map associated with the C<$scalar> is automatically removed
		2115	when the C<$scalar> is undef'd or destroyed, or when the C<IO::AIO::mmap>
		2116	or C<IO::AIO::munmap> functions are called on it.
		2117
		2118	This calls the C<mmap>(2) function internally. See your system's manual
		2119	page for details on the C<$length>, C<$prot> and C<$flags> parameters.
		2120
		2121	The C<$length> must be larger than zero and smaller than the actual
		2122	filesize.
		2123
		2124	C<$prot> is a combination of C<IO::AIO::PROT_NONE>, C<IO::AIO::PROT_EXEC>,
		2125	C<IO::AIO::PROT_READ> and/or C<IO::AIO::PROT_WRITE>,
		2126
		2127	C<$flags> can be a combination of
		2128	C<IO::AIO::MAP_SHARED> or
		2129	C<IO::AIO::MAP_PRIVATE>,
		2130	or a number of system-specific flags (when not available, the are C<0>):
		2131	C<IO::AIO::MAP_ANONYMOUS> (which is set to C<MAP_ANON> if your system only provides this constant),
		2132	C<IO::AIO::MAP_LOCKED>,
		2133	C<IO::AIO::MAP_NORESERVE>,
		2134	C<IO::AIO::MAP_POPULATE>,
		2135	C<IO::AIO::MAP_NONBLOCK>,
		2136	C<IO::AIO::MAP_FIXED>,
		2137	C<IO::AIO::MAP_GROWSDOWN>,
		2138	C<IO::AIO::MAP_32BIT>,
		2139	C<IO::AIO::MAP_HUGETLB> or
		2140	C<IO::AIO::MAP_STACK>.
		2141
		2142	If C<$fh> is C<undef>, then a file descriptor of C<-1> is passed.
		2143
		2144	C<$offset> is the offset from the start of the file - it generally must be
		2145	a multiple of C<IO::AIO::PAGESIZE> and defaults to C<0>.
		2146
		2147	Example:
		2148
		2149	use Digest::MD5;
		2150	use IO::AIO;
		2151
		2152	open my $fh, "<verybigfile"
		2153	or die "$!";
		2154
		2155	IO::AIO::mmap my $data, -s $fh, IO::AIO::PROT_READ, IO::AIO::MAP_SHARED, $fh
		2156	or die "verybigfile: $!";
		2157
		2158	my $fast_md5 = md5 $data;
		2159
		2160	=item IO::AIO::munmap $scalar
		2161
		2162	Removes a previous mmap and undefines the C<$scalar>.
		2163
		2164	=item IO::AIO::mremap $scalar, $new_length, $flags = MREMAP_MAYMOVE[, $new_address = 0]
		2165
		2166	Calls the Linux-specific mremap(2) system call. The C<$scalar> must have
		2167	been mapped by C<IO::AIO::mmap>, and C<$flags> must currently either be
		2168	C<0> or C<IO::AIO::MREMAP_MAYMOVE>.
		2169
		2170	Returns true if successful, and false otherwise. If the underlying mmapped
		2171	region has changed address, then the true value has the numerical value
		2172	C<1>, otherwise it has the numerical value C<0>:
		2173
		2174	my $success = IO::AIO::mremap $mmapped, 8192, IO::AIO::MREMAP_MAYMOVE
		2175	or die "mremap: $!";
		2176
		2177	if ($success*1) {
		2178	warn "scalar has chanegd address in memory\n";
		2179	}
		2180
		2181	C<IO::AIO::MREMAP_FIXED> and the C<$new_address> argument are currently
		2182	implemented, but not supported and might go away in a future version.
		2183
		2184	On systems where this call is not supported or is not emulated, this call
		2185	returns falls and sets C<$!> to C<ENOSYS>.
		2186
		2187	=item IO::AIO::munlock $scalar, $offset = 0, $length = undef
		2188
		2189	Calls the C<munlock> function, undoing the effects of a previous
		2190	C<aio_mlock> call (see its description for details).
		2191
		2192	=item IO::AIO::munlockall
		2193
		2194	Calls the C<munlockall> function.
		2195
		2196	On systems that do not implement C<munlockall>, this function returns
		2197	ENOSYS, otherwise the return value of C<munlockall>.
		2198
		2199	=item IO::AIO::splice $r_fh, $r_off, $w_fh, $w_off, $length, $flags
		2200
		2201	Calls the GNU/Linux C<splice(2)> syscall, if available. If C<$r_off> or
		2202	C<$w_off> are C<undef>, then C<NULL> is passed for these, otherwise they
		2203	should be the file offset.
		2204
		2205	C<$r_fh> and C<$w_fh> should not refer to the same file, as splice might
		2206	silently corrupt the data in this case.
		2207
		2208	The following symbol flag values are available: C<IO::AIO::SPLICE_F_MOVE>,
		2209	C<IO::AIO::SPLICE_F_NONBLOCK>, C<IO::AIO::SPLICE_F_MORE> and
		2210	C<IO::AIO::SPLICE_F_GIFT>.
		2211
		2212	See the C<splice(2)> manpage for details.
		2213
		2214	=item IO::AIO::tee $r_fh, $w_fh, $length, $flags
		2215
		2216	Calls the GNU/Linux C<tee(2)> syscall, see its manpage and the
		2217	description for C<IO::AIO::splice> above for details.
		2218
		2219	=item $actual_size = IO::AIO::pipesize $r_fh[, $new_size]
		2220
		2221	Attempts to query or change the pipe buffer size. Obviously works only
		2222	on pipes, and currently works only on GNU/Linux systems, and fails with
		2223	C<-1>/C<ENOSYS> everywhere else. If anybody knows how to influence pipe buffer
		2224	size on other systems, drop me a note.
		2225
		2226	=item ($rfh, $wfh) = IO::AIO::pipe2 [$flags]
		2227
		2228	This is a direct interface to the Linux L<pipe2(2)> system call. If
		2229	C<$flags> is missing or C<0>, then this should be the same as a call to
		2230	perl's built-in C<pipe> function and create a new pipe, and works on
		2231	systems that lack the pipe2 syscall. On win32, this case invokes C<_pipe
		2232	(..., 4096, O_BINARY)>.
		2233
		2234	If C<$flags> is non-zero, it tries to invoke the pipe2 system call with
		2235	the given flags (Linux 2.6.27, glibc 2.9).
		2236
		2237	On success, the read and write file handles are returned.
		2238
		2239	On error, nothing will be returned. If the pipe2 syscall is missing and
		2240	C<$flags> is non-zero, fails with C<ENOSYS>.
		2241
		2242	Please refer to L<pipe2(2)> for more info on the C<$flags>, but at the
		2243	time of this writing, C<IO::AIO::O_CLOEXEC>, C<IO::AIO::O_NONBLOCK> and
		2244	C<IO::AIO::O_DIRECT> (Linux 3.4, for packet-based pipes) were supported.
		2245
		2246	Example: create a pipe race-free w.r.t. threads and fork:
		2247
		2248	my ($rfh, $wfh) = IO::AIO::pipe2 IO::AIO::O_CLOEXEC
		2249	or die "pipe2: $!\n";
		2250
		2251	=item $fh = IO::AIO::eventfd [$initval, [$flags]]
		2252
		2253	This is a direct interface to the Linux L<eventfd(2)> system call. The
		2254	(unhelpful) defaults for C<$initval> and C<$flags> are C<0> for both.
		2255
		2256	On success, the new eventfd filehandle is returned, otherwise returns
		2257	C<undef>. If the eventfd syscall is missing, fails with C<ENOSYS>.
		2258
		2259	Please refer to L<eventfd(2)> for more info on this call.
		2260
		2261	The following symbol flag values are available: C<IO::AIO::EFD_CLOEXEC>,
		2262	C<IO::AIO::EFD_NONBLOCK> and C<IO::AIO::EFD_SEMAPHORE> (Linux 2.6.30).
		2263
		2264	Example: create a new eventfd filehandle:
		2265
		2266	$fh = IO::AIO::eventfd 0, IO::AIO::O_CLOEXEC
		2267	or die "eventfd: $!\n";
		2268
		2269	=item $fh = IO::AIO::timerfd_create $clockid[, $flags]
		2270
		2271	This is a direct interface to the Linux L<timerfd_create(2)> system call. The
		2272	(unhelpful) default for C<$flags> is C<0>.
		2273
		2274	On success, the new timerfd filehandle is returned, otherwise returns
		2275	C<undef>. If the eventfd syscall is missing, fails with C<ENOSYS>.
		2276
		2277	Please refer to L<timerfd_create(2)> for more info on this call.
		2278
		2279	The following C<$clockid> values are
		2280	available: C<IO::AIO::CLOCK_REALTIME>, C<IO::AIO::CLOCK_MONOTONIC>
		2281	C<IO::AIO::CLOCK_CLOCK_BOOTTIME> (Linux 3.15)
		2282	C<IO::AIO::CLOCK_CLOCK_REALTIME_ALARM> (Linux 3.11) and
		2283	C<IO::AIO::CLOCK_CLOCK_BOOTTIME_ALARM> (Linux 3.11).
		2284
		2285	The following C<$flags> values are available (Linux
		2286	2.6.27): C<IO::AIO::TFD_NONBLOCK> and C<IO::AIO::TFD_CLOEXEC>.
		2287
		2288	Example: create a new timerfd and set it to one-second repeated alarms,
		2289	then wait for two alarms:
		2290
		2291	my $fh = IO::AIO::timerfd_create IO::AIO::CLOCK_BOOTTIME, IO::AIO::TFD_CLOEXEC
		2292	or die "timerfd_create: $!\n";
		2293
		2294	defined IO::AIO::timerfd_settime $fh, 0, 1, 1
		2295	or die "timerfd_settime: $!\n";
		2296
		2297	for (1..2) {
		2298	8 == sysread $fh, my $buf, 8
		2299	or die "timerfd read failure\n";
		2300
		2301	printf "number of expirations (likely 1): %d\n",
		2302	unpack "Q", $buf;
		2303	}
		2304
		2305	=item ($cur_interval, $cur_value) = IO::AIO::timerfd_settime $fh, $flags, $new_interval, $nbw_value
		2306
		2307	This is a direct interface to the Linux L<timerfd_settime(2)> system
		2308	call. Please refer to its manpage for more info on this call.
		2309
		2310	The new itimerspec is specified using two (possibly fractional) second
		2311	values, C<$new_interval> and C<$new_value>).
		2312
		2313	On success, the current interval and value are returned (as per
		2314	C<timerfd_gettime>). On failure, the empty list is returned.
		2315
		2316	The following C<$flags> values are
		2317	available: C<IO::AIO::TFD_TIMER_ABSTIME> and
		2318	C<IO::AIO::TFD_TIMER_CANCEL_ON_SET>.
		2319
		2320	See C<IO::AIO::timerfd_create> for a full example.
		2321
		2322	=item ($cur_interval, $cur_value) = IO::AIO::timerfd_gettime $fh
		2323
		2324	This is a direct interface to the Linux L<timerfd_gettime(2)> system
		2325	call. Please refer to its manpage for more info on this call.
		2326
		2327	On success, returns the current values of interval and value for the given
		2328	timerfd (as potentially fractional second values). On failure, the empty
		2329	list is returned.
		2330
		2331	=back
		2332
312	=cut	2333	=cut
313		2334
314	# support function to convert a fd into a perl filehandle
315	sub _fd2fh {
316	return undef if $_[0] < 0;
317
318	# try to generate nice filehandles
319	my $sym = "IO::AIO::fd#$_[0]";
320	local *$sym;
321	open *$sym, "+<&=$_[0]"
322	or return undef;
323
324	*$sym
325	}
326
327	min_parallel 4;	2335	min_parallel 8;
328		2336
329	END {	2337	END { flush }
330	max_parallel 0;
331	}
332		2338
333	1;	2339	1;
334		2340
		2341	=head1 EVENT LOOP INTEGRATION
		2342
		2343	It is recommended to use L<AnyEvent::AIO> to integrate IO::AIO
		2344	automatically into many event loops:
		2345
		2346	# AnyEvent integration (EV, Event, Glib, Tk, POE, urxvt, pureperl...)
		2347	use AnyEvent::AIO;
		2348
		2349	You can also integrate IO::AIO manually into many event loops, here are
		2350	some examples of how to do this:
		2351
		2352	# EV integration
		2353	my $aio_w = EV::io IO::AIO::poll_fileno, EV::READ, \&IO::AIO::poll_cb;
		2354
		2355	# Event integration
		2356	Event->io (fd => IO::AIO::poll_fileno,
		2357	poll => 'r',
		2358	cb => \&IO::AIO::poll_cb);
		2359
		2360	# Glib/Gtk2 integration
		2361	add_watch Glib::IO IO::AIO::poll_fileno,
		2362	in => sub { IO::AIO::poll_cb; 1 };
		2363
		2364	# Tk integration
		2365	Tk::Event::IO->fileevent (IO::AIO::poll_fileno, "",
		2366	readable => \&IO::AIO::poll_cb);
		2367
		2368	# Danga::Socket integration
		2369	Danga::Socket->AddOtherFds (IO::AIO::poll_fileno =>
		2370	\&IO::AIO::poll_cb);
		2371
		2372	=head2 FORK BEHAVIOUR
		2373
		2374	Usage of pthreads in a program changes the semantics of fork
		2375	considerably. Specifically, only async-safe functions can be called after
		2376	fork. Perl doesn't know about this, so in general, you cannot call fork
		2377	with defined behaviour in perl if pthreads are involved. IO::AIO uses
		2378	pthreads, so this applies, but many other extensions and (for inexplicable
		2379	reasons) perl itself often is linked against pthreads, so this limitation
		2380	applies to quite a lot of perls.
		2381
		2382	This module no longer tries to fight your OS, or POSIX. That means IO::AIO
		2383	only works in the process that loaded it. Forking is fully supported, but
		2384	using IO::AIO in the child is not.
		2385
		2386	You might get around by not I<using> IO::AIO before (or after)
		2387	forking. You could also try to call the L<IO::AIO::reinit> function in the
		2388	child:
		2389
		2390	=over 4
		2391
		2392	=item IO::AIO::reinit
		2393
		2394	Abandons all current requests and I/O threads and simply reinitialises all
		2395	data structures. This is not an operation supported by any standards, but
		2396	happens to work on GNU/Linux and some newer BSD systems.
		2397
		2398	The only reasonable use for this function is to call it after forking, if
		2399	C<IO::AIO> was used in the parent. Calling it while IO::AIO is active in
		2400	the process will result in undefined behaviour. Calling it at any time
		2401	will also result in any undefined (by POSIX) behaviour.
		2402
		2403	=back
		2404
		2405	=head2 LINUX-SPECIFIC CALLS
		2406
		2407	When a call is documented as "linux-specific" then this means it
		2408	originated on GNU/Linux. C<IO::AIO> will usually try to autodetect the
		2409	availability and compatibility of such calls regardless of the platform
		2410	it is compiled on, so platforms such as FreeBSD which often implement
		2411	these calls will work. When in doubt, call them and see if they fail wth
		2412	C<ENOSYS>.
		2413
		2414	=head2 MEMORY USAGE
		2415
		2416	Per-request usage:
		2417
		2418	Each aio request uses - depending on your architecture - around 100-200
		2419	bytes of memory. In addition, stat requests need a stat buffer (possibly
		2420	a few hundred bytes), readdir requires a result buffer and so on. Perl
		2421	scalars and other data passed into aio requests will also be locked and
		2422	will consume memory till the request has entered the done state.
		2423
		2424	This is not awfully much, so queuing lots of requests is not usually a
		2425	problem.
		2426
		2427	Per-thread usage:
		2428
		2429	In the execution phase, some aio requests require more memory for
		2430	temporary buffers, and each thread requires a stack and other data
		2431	structures (usually around 16k-128k, depending on the OS).
		2432
		2433	=head1 KNOWN BUGS
		2434
		2435	Known bugs will be fixed in the next release :)
		2436
		2437	=head1 KNOWN ISSUES
		2438
		2439	Calls that try to "import" foreign memory areas (such as C<IO::AIO::mmap>
		2440	or C<IO::AIO::aio_slurp>) do not work with generic lvalues, such as
		2441	non-created hash slots or other scalars I didn't think of. It's best to
		2442	avoid such and either use scalar variables or making sure that the scalar
		2443	exists (e.g. by storing C<undef>) and isn't "funny" (e.g. tied).
		2444
		2445	I am not sure anything can be done about this, so this is considered a
		2446	known issue, rather than a bug.
		2447
335	=head1 SEE ALSO	2448	=head1 SEE ALSO
336		2449
337	L<Coro>, L<Linux::AIO>.	2450	L<AnyEvent::AIO> for easy integration into event loops, L<Coro::AIO> for a
		2451	more natural syntax.
338		2452
339	=head1 AUTHOR	2453	=head1 AUTHOR
340		2454
341	Marc Lehmann <schmorp@schmorp.de>	2455	Marc Lehmann <schmorp@schmorp.de>
342	http://home.schmorp.de/	2456	http://home.schmorp.de/

Diff Legend

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