ViewVC Help

View File | Revision Log | Show Annotations | Download File

/cvs/IO-AIO/AIO.pm

Comparing IO-AIO/AIO.pm (file contents):
Revision 1.47 by root, Thu Dec 29 15:44:13 2005 UTC vs.
Revision 1.231 by root, Fri Jul 27 19:03:18 2012 UTC

…		…
4		4
5	=head1 SYNOPSIS	5	=head1 SYNOPSIS
6		6
7	use IO::AIO;	7	use IO::AIO;
8		8
9	aio_open "/etc/passwd", O_RDONLY, 0, sub {	9	aio_open "/etc/passwd", IO::AIO::O_RDONLY, 0, sub {
10	my ($fh) = @_;	10	my $fh = shift
		11	or die "/etc/passwd: $!";
11	...	12	...
12	};	13	};
13		14
14	aio_unlink "/tmp/file", sub { };	15	aio_unlink "/tmp/file", sub { };
15		16
16	aio_read $fh, 30000, 1024, $buffer, 0, sub {	17	aio_read $fh, 30000, 1024, $buffer, 0, sub {
17	$_[0] > 0 or die "read error: $!";	18	$_[0] > 0 or die "read error: $!";
18	};	19	};
19		20
20	# AnyEvent	21	# version 2+ has request and group objects
21	open my $fh, "<&=" . IO::AIO::poll_fileno or die "$!";	22	use IO::AIO 2;
22	my $w = AnyEvent->io (fh => $fh, poll => 'r', cb => sub { IO::AIO::poll_cb });
23		23
24	# Event	24	aioreq_pri 4; # give next request a very high priority
25	Event->io (fd => IO::AIO::poll_fileno,	25	my $req = aio_unlink "/tmp/file", sub { };
26	poll => 'r',	26	$req->cancel; # cancel request if still in queue
27	cb => \&IO::AIO::poll_cb);
28		27
29	# Glib/Gtk2	28	my $grp = aio_group sub { print "all stats done\n" };
30	add_watch Glib::IO IO::AIO::poll_fileno,	29	add $grp aio_stat "..." for ...;
31	in => sub { IO::AIO::poll_cb; 1 };
32
33	# Tk
34	Tk::Event::IO->fileevent (IO::AIO::poll_fileno, "",
35	readable => \&IO::AIO::poll_cb);
36
37	# Danga::Socket
38	Danga::Socket->AddOtherFds (IO::AIO::poll_fileno =>
39	\&IO::AIO::poll_cb);
40
41		30
42	=head1 DESCRIPTION	31	=head1 DESCRIPTION
43		32
44	This module implements asynchronous I/O using whatever means your	33	This module implements asynchronous I/O using whatever means your
45	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>).
46		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
47	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
48	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
49	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
50	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
51	aio functions available on many operating systems. However, they are often	57	functions available on many operating systems. However, they are often
52	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
53	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
54	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.
55		62
56	Although the module will work with in the presence of other threads, it is	63	Although the module will work in the presence of other (Perl-) threads,
57	currently not reentrant, so use appropriate locking yourself, always call	64	it is currently not reentrant in any way, so use appropriate locking
58	C<poll_cb> from within the same thread, or never call C<poll_cb> (or other	65	yourself, always call C<poll_cb> from within the same thread, or never
59	C<aio_> functions) recursively.	66	call C<poll_cb> (or other C<aio_> functions) recursively.
		67
		68	=head2 EXAMPLE
		69
		70	This is a simple example that uses the EV module and loads
		71	F</etc/passwd> asynchronously:
		72
		73	use Fcntl;
		74	use EV;
		75	use IO::AIO;
		76
		77	# register the IO::AIO callback with EV
		78	my $aio_w = EV::io IO::AIO::poll_fileno, EV::READ, \&IO::AIO::poll_cb;
		79
		80	# queue the request to open /etc/passwd
		81	aio_open "/etc/passwd", IO::AIO::O_RDONLY, 0, sub {
		82	my $fh = shift
		83	or die "error while opening: $!";
		84
		85	# stat'ing filehandles is generally non-blocking
		86	my $size = -s $fh;
		87
		88	# queue a request to read the file
		89	my $contents;
		90	aio_read $fh, 0, $size, $contents, 0, sub {
		91	$_[0] == $size
		92	or die "short read: $!";
		93
		94	close $fh;
		95
		96	# file contents now in $contents
		97	print $contents;
		98
		99	# exit event loop and program
		100	EV::unloop;
		101	};
		102	};
		103
		104	# possibly queue up other requests, or open GUI windows,
		105	# check for sockets etc. etc.
		106
		107	# process events as long as there are some:
		108	EV::loop;
		109
		110	=head1 REQUEST ANATOMY AND LIFETIME
		111
		112	Every C<aio_*> function creates a request. which is a C data structure not
		113	directly visible to Perl.
		114
		115	If called in non-void context, every request function returns a Perl
		116	object representing the request. In void context, nothing is returned,
		117	which saves a bit of memory.
		118
		119	The perl object is a fairly standard ref-to-hash object. The hash contents
		120	are not used by IO::AIO so you are free to store anything you like in it.
		121
		122	During their existance, aio requests travel through the following states,
		123	in order:
		124
		125	=over 4
		126
		127	=item ready
		128
		129	Immediately after a request is created it is put into the ready state,
		130	waiting for a thread to execute it.
		131
		132	=item execute
		133
		134	A thread has accepted the request for processing and is currently
		135	executing it (e.g. blocking in read).
		136
		137	=item pending
		138
		139	The request has been executed and is waiting for result processing.
		140
		141	While request submission and execution is fully asynchronous, result
		142	processing is not and relies on the perl interpreter calling C<poll_cb>
		143	(or another function with the same effect).
		144
		145	=item result
		146
		147	The request results are processed synchronously by C<poll_cb>.
		148
		149	The C<poll_cb> function will process all outstanding aio requests by
		150	calling their callbacks, freeing memory associated with them and managing
		151	any groups they are contained in.
		152
		153	=item done
		154
		155	Request has reached the end of its lifetime and holds no resources anymore
		156	(except possibly for the Perl object, but its connection to the actual
		157	aio request is severed and calling its methods will either do nothing or
		158	result in a runtime error).
		159
		160	=back
60		161
61	=cut	162	=cut
62		163
63	package IO::AIO;	164	package IO::AIO;
64		165
65	no warnings;	166	use Carp ();
		167
		168	use common::sense;
66		169
67	use base 'Exporter';	170	use base 'Exporter';
68		171
69	use Fcntl ();
70
71	BEGIN {	172	BEGIN {
72	$VERSION = '1.72';	173	our $VERSION = '4.15';
73		174
74	@EXPORT = qw(aio_sendfile aio_read aio_write aio_open aio_close aio_stat	175	our @AIO_REQ = qw(aio_sendfile aio_seek aio_read aio_write aio_open aio_close
75	aio_lstat aio_unlink aio_rmdir aio_readdir aio_scandir aio_symlink	176	aio_stat aio_lstat aio_unlink aio_rmdir aio_readdir aio_readdirx
76	aio_fsync aio_fdatasync aio_readahead);	177	aio_scandir aio_symlink aio_readlink aio_realpath aio_sync
77	@EXPORT_OK = qw(poll_fileno poll_cb min_parallel max_parallel	178	aio_fsync aio_syncfs aio_fdatasync aio_sync_file_range aio_fallocate
78	max_outstanding nreqs);	179	aio_pathsync aio_readahead aio_fiemap
		180	aio_rename aio_link aio_move aio_copy aio_group
		181	aio_nop aio_mknod aio_load aio_rmtree aio_mkdir aio_chown
		182	aio_chmod aio_utime aio_truncate
		183	aio_msync aio_mtouch aio_mlock aio_mlockall
		184	aio_statvfs
		185	aio_wd);
		186
		187	our @EXPORT = (@AIO_REQ, qw(aioreq_pri aioreq_nice));
		188	our @EXPORT_OK = qw(poll_fileno poll_cb poll_wait flush
		189	min_parallel max_parallel max_idle idle_timeout
		190	nreqs nready npending nthreads
		191	max_poll_time max_poll_reqs
		192	sendfile fadvise madvise
		193	mmap munmap munlock munlockall);
		194
		195	push @AIO_REQ, qw(aio_busy); # not exported
		196
		197	@IO::AIO::GRP::ISA = 'IO::AIO::REQ';
79		198
80	require XSLoader;	199	require XSLoader;
81	XSLoader::load IO::AIO, $VERSION;	200	XSLoader::load ("IO::AIO", $VERSION);
82	}	201	}
83		202
84	=head1 FUNCTIONS	203	=head1 FUNCTIONS
85		204
86	=head2 AIO FUNCTIONS	205	=head2 QUICK OVERVIEW
		206
		207	This section simply lists the prototypes most of the functions for
		208	quick reference. See the following sections for function-by-function
		209	documentation.
		210
		211	aio_wd $pathname, $callback->($wd)
		212	aio_open $pathname, $flags, $mode, $callback->($fh)
		213	aio_close $fh, $callback->($status)
		214	aio_seek $fh,$offset,$whence, $callback->($offs)
		215	aio_read $fh,$offset,$length, $data,$dataoffset, $callback->($retval)
		216	aio_write $fh,$offset,$length, $data,$dataoffset, $callback->($retval)
		217	aio_sendfile $out_fh, $in_fh, $in_offset, $length, $callback->($retval)
		218	aio_readahead $fh,$offset,$length, $callback->($retval)
		219	aio_stat $fh_or_path, $callback->($status)
		220	aio_lstat $fh, $callback->($status)
		221	aio_statvfs $fh_or_path, $callback->($statvfs)
		222	aio_utime $fh_or_path, $atime, $mtime, $callback->($status)
		223	aio_chown $fh_or_path, $uid, $gid, $callback->($status)
		224	aio_chmod $fh_or_path, $mode, $callback->($status)
		225	aio_truncate $fh_or_path, $offset, $callback->($status)
		226	aio_allocate $fh, $mode, $offset, $len, $callback->($status)
		227	aio_fiemap $fh, $start, $length, $flags, $count, $cb->(\@extents)
		228	aio_unlink $pathname, $callback->($status)
		229	aio_mknod $pathname, $mode, $dev, $callback->($status)
		230	aio_link $srcpath, $dstpath, $callback->($status)
		231	aio_symlink $srcpath, $dstpath, $callback->($status)
		232	aio_readlink $pathname, $callback->($link)
		233	aio_realpath $pathname, $callback->($link)
		234	aio_rename $srcpath, $dstpath, $callback->($status)
		235	aio_mkdir $pathname, $mode, $callback->($status)
		236	aio_rmdir $pathname, $callback->($status)
		237	aio_readdir $pathname, $callback->($entries)
		238	aio_readdirx $pathname, $flags, $callback->($entries, $flags)
		239	IO::AIO::READDIR_DENTS IO::AIO::READDIR_DIRS_FIRST
		240	IO::AIO::READDIR_STAT_ORDER IO::AIO::READDIR_FOUND_UNKNOWN
		241	aio_scandir $pathname, $maxreq, $callback->($dirs, $nondirs)
		242	aio_load $pathname, $data, $callback->($status)
		243	aio_copy $srcpath, $dstpath, $callback->($status)
		244	aio_move $srcpath, $dstpath, $callback->($status)
		245	aio_rmtree $pathname, $callback->($status)
		246	aio_sync $callback->($status)
		247	aio_syncfs $fh, $callback->($status)
		248	aio_fsync $fh, $callback->($status)
		249	aio_fdatasync $fh, $callback->($status)
		250	aio_sync_file_range $fh, $offset, $nbytes, $flags, $callback->($status)
		251	aio_pathsync $pathname, $callback->($status)
		252	aio_msync $scalar, $offset = 0, $length = undef, flags = 0, $callback->($status)
		253	aio_mtouch $scalar, $offset = 0, $length = undef, flags = 0, $callback->($status)
		254	aio_mlock $scalar, $offset = 0, $length = undef, $callback->($status)
		255	aio_mlockall $flags, $callback->($status)
		256	aio_group $callback->(...)
		257	aio_nop $callback->()
		258
		259	$prev_pri = aioreq_pri [$pri]
		260	aioreq_nice $pri_adjust
		261
		262	IO::AIO::poll_wait
		263	IO::AIO::poll_cb
		264	IO::AIO::poll
		265	IO::AIO::flush
		266	IO::AIO::max_poll_reqs $nreqs
		267	IO::AIO::max_poll_time $seconds
		268	IO::AIO::min_parallel $nthreads
		269	IO::AIO::max_parallel $nthreads
		270	IO::AIO::max_idle $nthreads
		271	IO::AIO::idle_timeout $seconds
		272	IO::AIO::max_outstanding $maxreqs
		273	IO::AIO::nreqs
		274	IO::AIO::nready
		275	IO::AIO::npending
		276
		277	IO::AIO::sendfile $ofh, $ifh, $offset, $count
		278	IO::AIO::fadvise $fh, $offset, $len, $advice
		279	IO::AIO::mmap $scalar, $length, $prot, $flags[, $fh[, $offset]]
		280	IO::AIO::munmap $scalar
		281	IO::AIO::madvise $scalar, $offset, $length, $advice
		282	IO::AIO::mprotect $scalar, $offset, $length, $protect
		283	IO::AIO::munlock $scalar, $offset = 0, $length = undef
		284	IO::AIO::munlockall
		285
		286	=head2 API NOTES
87		287
88	All the C<aio_*> calls are more or less thin wrappers around the syscall	288	All the C<aio_*> calls are more or less thin wrappers around the syscall
89	with the same name (sans C<aio_>). The arguments are similar or identical,	289	with the same name (sans C<aio_>). The arguments are similar or identical,
90	and they all accept an additional (and optional) C<$callback> argument	290	and they all accept an additional (and optional) C<$callback> argument
91	which must be a code reference. This code reference will get called with	291	which must be a code reference. This code reference will be called after
92	the syscall return code (e.g. most syscalls return C<-1> on error, unlike	292	the syscall has been executed in an asynchronous fashion. The results
93	perl, which usually delivers "false") as it's sole argument when the given	293	of the request will be passed as arguments to the callback (and, if an
94	syscall has been executed asynchronously.	294	error occured, in C<$!>) - for most requests the syscall return code (e.g.
		295	most syscalls return C<-1> on error, unlike perl, which usually delivers
		296	"false").
		297
		298	Some requests (such as C<aio_readdir>) pass the actual results and
		299	communicate failures by passing C<undef>.
95		300
96	All functions expecting a filehandle keep a copy of the filehandle	301	All functions expecting a filehandle keep a copy of the filehandle
97	internally until the request has finished.	302	internally until the request has finished.
98		303
		304	All functions return request objects of type L<IO::AIO::REQ> that allow
		305	further manipulation of those requests while they are in-flight.
		306
99	The pathnames you pass to these routines I<must> be absolute and	307	The pathnames you pass to these routines I<should> be absolute. The
100	encoded in byte form. The reason for the former is that at the time the	308	reason for this is that at the time the request is being executed, the
101	request is being executed, the current working directory could have	309	current working directory could have changed. Alternatively, you can
102	changed. Alternatively, you can make sure that you never change the	310	make sure that you never change the current working directory anywhere
103	current working directory.	311	in the program and then use relative paths. You can also take advantage
		312	of IO::AIOs working directory abstraction, that lets you specify paths
		313	relative to some previously-opened "working directory object" - see the
		314	description of the C<IO::AIO::WD> class later in this document.
104		315
105	To encode pathnames to byte form, either make sure you either: a)	316	To encode pathnames as octets, either make sure you either: a) always pass
106	always pass in filenames you got from outside (command line, readdir	317	in filenames you got from outside (command line, readdir etc.) without
107	etc.), b) are ASCII or ISO 8859-1, c) use the Encode module and encode	318	tinkering, b) are in your native filesystem encoding, c) use the Encode
108	your pathnames to the locale (or other) encoding in effect in the user	319	module and encode your pathnames to the locale (or other) encoding in
109	environment, d) use Glib::filename_from_unicode on unicode filenames or e)	320	effect in the user environment, d) use Glib::filename_from_unicode on
110	use something else.	321	unicode filenames or e) use something else to ensure your scalar has the
		322	correct contents.
		323
		324	This works, btw. independent of the internal UTF-8 bit, which IO::AIO
		325	handles correctly whether it is set or not.
		326
		327	=head2 AIO REQUEST FUNCTIONS
111		328
112	=over 4	329	=over 4
		330
		331	=item $prev_pri = aioreq_pri [$pri]
		332
		333	Returns the priority value that would be used for the next request and, if
		334	C<$pri> is given, sets the priority for the next aio request.
		335
		336	The default priority is C<0>, the minimum and maximum priorities are C<-4>
		337	and C<4>, respectively. Requests with higher priority will be serviced
		338	first.
		339
		340	The priority will be reset to C<0> after each call to one of the C<aio_*>
		341	functions.
		342
		343	Example: open a file with low priority, then read something from it with
		344	higher priority so the read request is serviced before other low priority
		345	open requests (potentially spamming the cache):
		346
		347	aioreq_pri -3;
		348	aio_open ..., sub {
		349	return unless $_[0];
		350
		351	aioreq_pri -2;
		352	aio_read $_[0], ..., sub {
		353	...
		354	};
		355	};
		356
		357
		358	=item aioreq_nice $pri_adjust
		359
		360	Similar to C<aioreq_pri>, but subtracts the given value from the current
		361	priority, so the effect is cumulative.
		362
113		363
114	=item aio_open $pathname, $flags, $mode, $callback->($fh)	364	=item aio_open $pathname, $flags, $mode, $callback->($fh)
115		365
116	Asynchronously open or create a file and call the callback with a newly	366	Asynchronously open or create a file and call the callback with a newly
117	created filehandle for the file.	367	created filehandle for the file.
…		…
123	list. They are the same as used by C<sysopen>.	373	list. They are the same as used by C<sysopen>.
124		374
125	Likewise, C<$mode> specifies the mode of the newly created file, if it	375	Likewise, C<$mode> specifies the mode of the newly created file, if it
126	didn't exist and C<O_CREAT> has been given, just like perl's C<sysopen>,	376	didn't exist and C<O_CREAT> has been given, just like perl's C<sysopen>,
127	except that it is mandatory (i.e. use C<0> if you don't create new files,	377	except that it is mandatory (i.e. use C<0> if you don't create new files,
128	and C<0666> or C<0777> if you do).	378	and C<0666> or C<0777> if you do). Note that the C<$mode> will be modified
		379	by the umask in effect then the request is being executed, so better never
		380	change the umask.
129		381
130	Example:	382	Example:
131		383
132	aio_open "/etc/passwd", O_RDONLY, 0, sub {	384	aio_open "/etc/passwd", IO::AIO::O_RDONLY, 0, sub {
133	if ($_[0]) {	385	if ($_[0]) {
134	print "open successful, fh is $_[0]\n";	386	print "open successful, fh is $_[0]\n";
135	...	387	...
136	} else {	388	} else {
137	die "open failed: $!\n";	389	die "open failed: $!\n";
138	}	390	}
139	};	391	};
140		392
		393	In addition to all the common open modes/flags (C<O_RDONLY>, C<O_WRONLY>,
		394	C<O_RDWR>, C<O_CREAT>, C<O_TRUNC>, C<O_EXCL> and C<O_APPEND>), the
		395	following POSIX and non-POSIX constants are available (missing ones on
		396	your system are, as usual, C<0>):
		397
		398	C<O_ASYNC>, C<O_DIRECT>, C<O_NOATIME>, C<O_CLOEXEC>, C<O_NOCTTY>, C<O_NOFOLLOW>,
		399	C<O_NONBLOCK>, C<O_EXEC>, C<O_SEARCH>, C<O_DIRECTORY>, C<O_DSYNC>,
		400	C<O_RSYNC>, C<O_SYNC> and C<O_TTY_INIT>.
		401
		402
141	=item aio_close $fh, $callback->($status)	403	=item aio_close $fh, $callback->($status)
142		404
143	Asynchronously close a file and call the callback with the result	405	Asynchronously close a file and call the callback with the result
144	code. I<WARNING:> although accepted, you should not pass in a perl	406	code.
145	filehandle here, as perl will likely close the file descriptor another
146	time when the filehandle is destroyed. Normally, you can safely call perls
147	C<close> or just let filehandles go out of scope.
148		407
149	This is supposed to be a bug in the API, so that might change. It's	408	Unfortunately, you can't do this to perl. Perl I<insists> very strongly on
150	therefore best to avoid this function.	409	closing the file descriptor associated with the filehandle itself.
		410
		411	Therefore, C<aio_close> will not close the filehandle - instead it will
		412	use dup2 to overwrite the file descriptor with the write-end of a pipe
		413	(the pipe fd will be created on demand and will be cached).
		414
		415	Or in other words: the file descriptor will be closed, but it will not be
		416	free for reuse until the perl filehandle is closed.
		417
		418	=cut
		419
		420	=item aio_seek $fh, $offset, $whence, $callback->($offs)
		421
		422	Seeks the filehandle to the new C<$offset>, similarly to perl's
		423	C<sysseek>. The C<$whence> can use the traditional values (C<0> for
		424	C<IO::AIO::SEEK_SET>, C<1> for C<IO::AIO::SEEK_CUR> or C<2> for
		425	C<IO::AIO::SEEK_END>).
		426
		427	The resulting absolute offset will be passed to the callback, or C<-1> in
		428	case of an error.
		429
		430	In theory, the C<$whence> constants could be different than the
		431	corresponding values from L<Fcntl>, but perl guarantees they are the same,
		432	so don't panic.
		433
		434	As a GNU/Linux (and maybe Solaris) extension, also the constants
		435	C<IO::AIO::SEEK_DATA> and C<IO::AIO::SEEK_HOLE> are available, if they
		436	could be found. No guarantees about suitability for use in C<aio_seek> or
		437	Perl's C<sysseek> can be made though, although I would naively assume they
		438	"just work".
151		439
152	=item aio_read $fh,$offset,$length, $data,$dataoffset, $callback->($retval)	440	=item aio_read $fh,$offset,$length, $data,$dataoffset, $callback->($retval)
153		441
154	=item aio_write $fh,$offset,$length, $data,$dataoffset, $callback->($retval)	442	=item aio_write $fh,$offset,$length, $data,$dataoffset, $callback->($retval)
155		443
156	Reads or writes C<length> bytes from the specified C<fh> and C<offset>	444	Reads or writes C<$length> bytes from or to the specified C<$fh> and
157	into the scalar given by C<data> and offset C<dataoffset> and calls the	445	C<$offset> into the scalar given by C<$data> and offset C<$dataoffset>
158	callback without the actual number of bytes read (or -1 on error, just	446	and calls the callback without the actual number of bytes read (or -1 on
159	like the syscall).	447	error, just like the syscall).
		448
		449	C<aio_read> will, like C<sysread>, shrink or grow the C<$data> scalar to
		450	offset plus the actual number of bytes read.
		451
		452	If C<$offset> is undefined, then the current file descriptor offset will
		453	be used (and updated), otherwise the file descriptor offset will not be
		454	changed by these calls.
		455
		456	If C<$length> is undefined in C<aio_write>, use the remaining length of
		457	C<$data>.
		458
		459	If C<$dataoffset> is less than zero, it will be counted from the end of
		460	C<$data>.
160		461
161	The C<$data> scalar I<MUST NOT> be modified in any way while the request	462	The C<$data> scalar I<MUST NOT> be modified in any way while the request
162	is outstanding. Modifying it can result in segfaults or WW3 (if the	463	is outstanding. Modifying it can result in segfaults or World War III (if
163	necessary/optional hardware is installed).	464	the necessary/optional hardware is installed).
164		465
165	Example: Read 15 bytes at offset 7 into scalar C<$buffer>, starting at	466	Example: Read 15 bytes at offset 7 into scalar C<$buffer>, starting at
166	offset C<0> within the scalar:	467	offset C<0> within the scalar:
167		468
168	aio_read $fh, 7, 15, $buffer, 0, sub {	469	aio_read $fh, 7, 15, $buffer, 0, sub {
169	$_[0] > 0 or die "read error: $!";	470	$_[0] > 0 or die "read error: $!";
170	print "read $_[0] bytes: <$buffer>\n";	471	print "read $_[0] bytes: <$buffer>\n";
171	};	472	};
172		473
		474
173	=item aio_sendfile $out_fh, $in_fh, $in_offset, $length, $callback->($retval)	475	=item aio_sendfile $out_fh, $in_fh, $in_offset, $length, $callback->($retval)
174		476
175	Tries to copy C<$length> bytes from C<$in_fh> to C<$out_fh>. It starts	477	Tries to copy C<$length> bytes from C<$in_fh> to C<$out_fh>. It starts
176	reading at byte offset C<$in_offset>, and starts writing at the current	478	reading at byte offset C<$in_offset>, and starts writing at the current
177	file offset of C<$out_fh>. Because of that, it is not safe to issue more	479	file offset of C<$out_fh>. Because of that, it is not safe to issue more
178	than one C<aio_sendfile> per C<$out_fh>, as they will interfere with each	480	than one C<aio_sendfile> per C<$out_fh>, as they will interfere with each
179	other.	481	other. The same C<$in_fh> works fine though, as this function does not
		482	move or use the file offset of C<$in_fh>.
180		483
		484	Please note that C<aio_sendfile> can read more bytes from C<$in_fh> than
		485	are written, and there is no way to find out how many more bytes have been
		486	read from C<aio_sendfile> alone, as C<aio_sendfile> only provides the
		487	number of bytes written to C<$out_fh>. Only if the result value equals
		488	C<$length> one can assume that C<$length> bytes have been read.
		489
		490	Unlike with other C<aio_> functions, it makes a lot of sense to use
		491	C<aio_sendfile> on non-blocking sockets, as long as one end (typically
		492	the C<$in_fh>) is a file - the file I/O will then be asynchronous, while
		493	the socket I/O will be non-blocking. Note, however, that you can run
		494	into a trap where C<aio_sendfile> reads some data with readahead, then
		495	fails to write all data, and when the socket is ready the next time, the
		496	data in the cache is already lost, forcing C<aio_sendfile> to again hit
		497	the disk. Explicit C<aio_read> + C<aio_write> let's you better control
		498	resource usage.
		499
181	This call tries to make use of a native C<sendfile> syscall to provide	500	This call tries to make use of a native C<sendfile>-like syscall to
182	zero-copy operation. For this to work, C<$out_fh> should refer to a	501	provide zero-copy operation. For this to work, C<$out_fh> should refer to
183	socket, and C<$in_fh> should refer to mmap'able file.	502	a socket, and C<$in_fh> should refer to an mmap'able file.
184		503
185	If the native sendfile call fails or is not implemented, it will be	504	If a native sendfile cannot be found or it fails with C<ENOSYS>,
186	emulated, so you can call C<aio_sendfile> on any type of filehandle	505	C<EINVAL>, C<ENOTSUP>, C<EOPNOTSUPP>, C<EAFNOSUPPORT>, C<EPROTOTYPE> or
		506	C<ENOTSOCK>, it will be emulated, so you can call C<aio_sendfile> on any
187	regardless of the limitations of the operating system.	507	type of filehandle regardless of the limitations of the operating system.
188		508
189	Please note, however, that C<aio_sendfile> can read more bytes from	509	As native sendfile syscalls (as practically any non-POSIX interface hacked
190	C<$in_fh> than are written, and there is no way to find out how many	510	together in a hurry to improve benchmark numbers) tend to be rather buggy
191	bytes have been read from C<aio_sendfile> alone, as C<aio_sendfile> only	511	on many systems, this implementation tries to work around some known bugs
192	provides the number of bytes written to C<$out_fh>. Only if the result	512	in Linux and FreeBSD kernels (probably others, too), but that might fail,
193	value equals C<$length> one can assume that C<$length> bytes have been	513	so you really really should check the return value of C<aio_sendfile> -
194	read.	514	fewre bytes than expected might have been transferred.
		515
195		516
196	=item aio_readahead $fh,$offset,$length, $callback->($retval)	517	=item aio_readahead $fh,$offset,$length, $callback->($retval)
197		518
198	C<aio_readahead> populates the page cache with data from a file so that	519	C<aio_readahead> populates the page cache with data from a file so that
199	subsequent reads from that file will not block on disk I/O. The C<$offset>	520	subsequent reads from that file will not block on disk I/O. The C<$offset>
…		…
205	file. The current file offset of the file is left unchanged.	526	file. The current file offset of the file is left unchanged.
206		527
207	If that syscall doesn't exist (likely if your OS isn't Linux) it will be	528	If that syscall doesn't exist (likely if your OS isn't Linux) it will be
208	emulated by simply reading the data, which would have a similar effect.	529	emulated by simply reading the data, which would have a similar effect.
209		530
		531
210	=item aio_stat $fh_or_path, $callback->($status)	532	=item aio_stat $fh_or_path, $callback->($status)
211		533
212	=item aio_lstat $fh, $callback->($status)	534	=item aio_lstat $fh, $callback->($status)
213		535
214	Works like perl's C<stat> or C<lstat> in void context. The callback will	536	Works like perl's C<stat> or C<lstat> in void context. The callback will
…		…
219	for an explanation.	541	for an explanation.
220		542
221	Currently, the stats are always 64-bit-stats, i.e. instead of returning an	543	Currently, the stats are always 64-bit-stats, i.e. instead of returning an
222	error when stat'ing a large file, the results will be silently truncated	544	error when stat'ing a large file, the results will be silently truncated
223	unless perl itself is compiled with large file support.	545	unless perl itself is compiled with large file support.
		546
		547	To help interpret the mode and dev/rdev stat values, IO::AIO offers the
		548	following constants and functions (if not implemented, the constants will
		549	be C<0> and the functions will either C<croak> or fall back on traditional
		550	behaviour).
		551
		552	C<S_IFMT>, C<S_IFIFO>, C<S_IFCHR>, C<S_IFBLK>, C<S_IFLNK>, C<S_IFREG>,
		553	C<S_IFDIR>, C<S_IFWHT>, C<S_IFSOCK>, C<IO::AIO::major $dev_t>,
		554	C<IO::AIO::minor $dev_t>, C<IO::AIO::makedev $major, $minor>.
224		555
225	Example: Print the length of F</etc/passwd>:	556	Example: Print the length of F</etc/passwd>:
226		557
227	aio_stat "/etc/passwd", sub {	558	aio_stat "/etc/passwd", sub {
228	$_[0] and die "stat failed: $!";	559	$_[0] and die "stat failed: $!";
229	print "size is ", -s _, "\n";	560	print "size is ", -s _, "\n";
230	};	561	};
231		562
		563
		564	=item aio_statvfs $fh_or_path, $callback->($statvfs)
		565
		566	Works like the POSIX C<statvfs> or C<fstatvfs> syscalls, depending on
		567	whether a file handle or path was passed.
		568
		569	On success, the callback is passed a hash reference with the following
		570	members: C<bsize>, C<frsize>, C<blocks>, C<bfree>, C<bavail>, C<files>,
		571	C<ffree>, C<favail>, C<fsid>, C<flag> and C<namemax>. On failure, C<undef>
		572	is passed.
		573
		574	The following POSIX IO::AIO::ST_* constants are defined: C<ST_RDONLY> and
		575	C<ST_NOSUID>.
		576
		577	The following non-POSIX IO::AIO::ST_* flag masks are defined to
		578	their correct value when available, or to C<0> on systems that do
		579	not support them: C<ST_NODEV>, C<ST_NOEXEC>, C<ST_SYNCHRONOUS>,
		580	C<ST_MANDLOCK>, C<ST_WRITE>, C<ST_APPEND>, C<ST_IMMUTABLE>, C<ST_NOATIME>,
		581	C<ST_NODIRATIME> and C<ST_RELATIME>.
		582
		583	Example: stat C</wd> and dump out the data if successful.
		584
		585	aio_statvfs "/wd", sub {
		586	my $f = $_[0]
		587	or die "statvfs: $!";
		588
		589	use Data::Dumper;
		590	say Dumper $f;
		591	};
		592
		593	# result:
		594	{
		595	bsize => 1024,
		596	bfree => 4333064312,
		597	blocks => 10253828096,
		598	files => 2050765568,
		599	flag => 4096,
		600	favail => 2042092649,
		601	bavail => 4333064312,
		602	ffree => 2042092649,
		603	namemax => 255,
		604	frsize => 1024,
		605	fsid => 1810
		606	}
		607
		608
		609	=item aio_utime $fh_or_path, $atime, $mtime, $callback->($status)
		610
		611	Works like perl's C<utime> function (including the special case of $atime
		612	and $mtime being undef). Fractional times are supported if the underlying
		613	syscalls support them.
		614
		615	When called with a pathname, uses utimes(2) if available, otherwise
		616	utime(2). If called on a file descriptor, uses futimes(2) if available,
		617	otherwise returns ENOSYS, so this is not portable.
		618
		619	Examples:
		620
		621	# set atime and mtime to current time (basically touch(1)):
		622	aio_utime "path", undef, undef;
		623	# set atime to current time and mtime to beginning of the epoch:
		624	aio_utime "path", time, undef; # undef==0
		625
		626
		627	=item aio_chown $fh_or_path, $uid, $gid, $callback->($status)
		628
		629	Works like perl's C<chown> function, except that C<undef> for either $uid
		630	or $gid is being interpreted as "do not change" (but -1 can also be used).
		631
		632	Examples:
		633
		634	# same as "chown root path" in the shell:
		635	aio_chown "path", 0, -1;
		636	# same as above:
		637	aio_chown "path", 0, undef;
		638
		639
		640	=item aio_truncate $fh_or_path, $offset, $callback->($status)
		641
		642	Works like truncate(2) or ftruncate(2).
		643
		644
		645	=item aio_allocate $fh, $mode, $offset, $len, $callback->($status)
		646
		647	Allocates or freed disk space according to the C<$mode> argument. See the
		648	linux C<fallocate> docuemntation for details.
		649
		650	C<$mode> can currently be C<0> or C<IO::AIO::FALLOC_FL_KEEP_SIZE>
		651	to allocate space, or C<IO::AIO::FALLOC_FL_PUNCH_HOLE |
		652	IO::AIO::FALLOC_FL_KEEP_SIZE>, to deallocate a file range.
		653
		654	The file system block size used by C<fallocate> is presumably the
		655	C<f_bsize> returned by C<statvfs>.
		656
		657	If C<fallocate> isn't available or cannot be emulated (currently no
		658	emulation will be attempted), passes C<-1> and sets C<$!> to C<ENOSYS>.
		659
		660
		661	=item aio_chmod $fh_or_path, $mode, $callback->($status)
		662
		663	Works like perl's C<chmod> function.
		664
		665
232	=item aio_unlink $pathname, $callback->($status)	666	=item aio_unlink $pathname, $callback->($status)
233		667
234	Asynchronously unlink (delete) a file and call the callback with the	668	Asynchronously unlink (delete) a file and call the callback with the
235	result code.	669	result code.
236		670
		671
		672	=item aio_mknod $pathname, $mode, $dev, $callback->($status)
		673
		674	[EXPERIMENTAL]
		675
		676	Asynchronously create a device node (or fifo). See mknod(2).
		677
		678	The only (POSIX-) portable way of calling this function is:
		679
		680	aio_mknod $pathname, IO::AIO::S_IFIFO | $mode, 0, sub { ...
		681
		682	See C<aio_stat> for info about some potentially helpful extra constants
		683	and functions.
		684
		685	=item aio_link $srcpath, $dstpath, $callback->($status)
		686
		687	Asynchronously create a new link to the existing object at C<$srcpath> at
		688	the path C<$dstpath> and call the callback with the result code.
		689
		690
		691	=item aio_symlink $srcpath, $dstpath, $callback->($status)
		692
		693	Asynchronously create a new symbolic link to the existing object at C<$srcpath> at
		694	the path C<$dstpath> and call the callback with the result code.
		695
		696
		697	=item aio_readlink $pathname, $callback->($link)
		698
		699	Asynchronously read the symlink specified by C<$path> and pass it to
		700	the callback. If an error occurs, nothing or undef gets passed to the
		701	callback.
		702
		703
		704	=item aio_realpath $pathname, $callback->($path)
		705
		706	Asynchronously make the path absolute and resolve any symlinks in
		707	C<$path>. The resulting path only consists of directories (Same as
		708	L<Cwd::realpath>).
		709
		710	This request can be used to get the absolute path of the current working
		711	directory by passing it a path of F<.> (a single dot).
		712
		713
		714	=item aio_rename $srcpath, $dstpath, $callback->($status)
		715
		716	Asynchronously rename the object at C<$srcpath> to C<$dstpath>, just as
		717	rename(2) and call the callback with the result code.
		718
		719
		720	=item aio_mkdir $pathname, $mode, $callback->($status)
		721
		722	Asynchronously mkdir (create) a directory and call the callback with
		723	the result code. C<$mode> will be modified by the umask at the time the
		724	request is executed, so do not change your umask.
		725
		726
237	=item aio_rmdir $pathname, $callback->($status)	727	=item aio_rmdir $pathname, $callback->($status)
238		728
239	Asynchronously rmdir (delete) a directory and call the callback with the	729	Asynchronously rmdir (delete) a directory and call the callback with the
240	result code.	730	result code.
		731
241		732
242	=item aio_readdir $pathname, $callback->($entries)	733	=item aio_readdir $pathname, $callback->($entries)
243		734
244	Unlike the POSIX call of the same name, C<aio_readdir> reads an entire	735	Unlike the POSIX call of the same name, C<aio_readdir> reads an entire
245	directory (i.e. opendir + readdir + closedir). The entries will not be	736	directory (i.e. opendir + readdir + closedir). The entries will not be
246	sorted, and will B<NOT> include the C<.> and C<..> entries.	737	sorted, and will B<NOT> include the C<.> and C<..> entries.
247		738
248	The callback a single argument which is either C<undef> or an array-ref	739	The callback is passed a single argument which is either C<undef> or an
249	with the filenames.	740	array-ref with the filenames.
250		741
		742
		743	=item aio_readdirx $pathname, $flags, $callback->($entries, $flags)
		744
		745	Quite similar to C<aio_readdir>, but the C<$flags> argument allows one to
		746	tune behaviour and output format. In case of an error, C<$entries> will be
		747	C<undef>.
		748
		749	The flags are a combination of the following constants, ORed together (the
		750	flags will also be passed to the callback, possibly modified):
		751
		752	=over 4
		753
		754	=item IO::AIO::READDIR_DENTS
		755
		756	When this flag is off, then the callback gets an arrayref consisting of
		757	names only (as with C<aio_readdir>), otherwise it gets an arrayref with
		758	C<[$name, $type, $inode]> arrayrefs, each describing a single directory
		759	entry in more detail.
		760
		761	C<$name> is the name of the entry.
		762
		763	C<$type> is one of the C<IO::AIO::DT_xxx> constants:
		764
		765	C<IO::AIO::DT_UNKNOWN>, C<IO::AIO::DT_FIFO>, C<IO::AIO::DT_CHR>, C<IO::AIO::DT_DIR>,
		766	C<IO::AIO::DT_BLK>, C<IO::AIO::DT_REG>, C<IO::AIO::DT_LNK>, C<IO::AIO::DT_SOCK>,
		767	C<IO::AIO::DT_WHT>.
		768
		769	C<IO::AIO::DT_UNKNOWN> means just that: readdir does not know. If you need to
		770	know, you have to run stat yourself. Also, for speed reasons, the C<$type>
		771	scalars are read-only: you can not modify them.
		772
		773	C<$inode> is the inode number (which might not be exact on systems with 64
		774	bit inode numbers and 32 bit perls). This field has unspecified content on
		775	systems that do not deliver the inode information.
		776
		777	=item IO::AIO::READDIR_DIRS_FIRST
		778
		779	When this flag is set, then the names will be returned in an order where
		780	likely directories come first, in optimal stat order. This is useful when
		781	you need to quickly find directories, or you want to find all directories
		782	while avoiding to stat() each entry.
		783
		784	If the system returns type information in readdir, then this is used
		785	to find directories directly. Otherwise, likely directories are names
		786	beginning with ".", or otherwise names with no dots, of which names with
		787	short names are tried first.
		788
		789	=item IO::AIO::READDIR_STAT_ORDER
		790
		791	When this flag is set, then the names will be returned in an order
		792	suitable for stat()'ing each one. That is, when you plan to stat()
		793	all files in the given directory, then the returned order will likely
		794	be fastest.
		795
		796	If both this flag and C<IO::AIO::READDIR_DIRS_FIRST> are specified, then
		797	the likely dirs come first, resulting in a less optimal stat order.
		798
		799	=item IO::AIO::READDIR_FOUND_UNKNOWN
		800
		801	This flag should not be set when calling C<aio_readdirx>. Instead, it
		802	is being set by C<aio_readdirx>, when any of the C<$type>'s found were
		803	C<IO::AIO::DT_UNKNOWN>. The absence of this flag therefore indicates that all
		804	C<$type>'s are known, which can be used to speed up some algorithms.
		805
		806	=back
		807
		808
		809	=item aio_load $pathname, $data, $callback->($status)
		810
		811	This is a composite request that tries to fully load the given file into
		812	memory. Status is the same as with aio_read.
		813
		814	=cut
		815
		816	sub aio_load($$;$) {
		817	my ($path, undef, $cb) = @_;
		818	my $data = \$_[1];
		819
		820	my $pri = aioreq_pri;
		821	my $grp = aio_group $cb;
		822
		823	aioreq_pri $pri;
		824	add $grp aio_open $path, O_RDONLY, 0, sub {
		825	my $fh = shift
		826	or return $grp->result (-1);
		827
		828	aioreq_pri $pri;
		829	add $grp aio_read $fh, 0, (-s $fh), $$data, 0, sub {
		830	$grp->result ($_[0]);
		831	};
		832	};
		833
		834	$grp
		835	}
		836
		837	=item aio_copy $srcpath, $dstpath, $callback->($status)
		838
		839	Try to copy the I<file> (directories not supported as either source or
		840	destination) from C<$srcpath> to C<$dstpath> and call the callback with
		841	a status of C<0> (ok) or C<-1> (error, see C<$!>).
		842
		843	This is a composite request that creates the destination file with
		844	mode 0200 and copies the contents of the source file into it using
		845	C<aio_sendfile>, followed by restoring atime, mtime, access mode and
		846	uid/gid, in that order.
		847
		848	If an error occurs, the partial destination file will be unlinked, if
		849	possible, except when setting atime, mtime, access mode and uid/gid, where
		850	errors are being ignored.
		851
		852	=cut
		853
		854	sub aio_copy($$;$) {
		855	my ($src, $dst, $cb) = @_;
		856
		857	my $pri = aioreq_pri;
		858	my $grp = aio_group $cb;
		859
		860	aioreq_pri $pri;
		861	add $grp aio_open $src, O_RDONLY, 0, sub {
		862	if (my $src_fh = $_[0]) {
		863	my @stat = stat $src_fh; # hmm, might block over nfs?
		864
		865	aioreq_pri $pri;
		866	add $grp aio_open $dst, O_CREAT | O_WRONLY | O_TRUNC, 0200, sub {
		867	if (my $dst_fh = $_[0]) {
		868	aioreq_pri $pri;
		869	add $grp aio_sendfile $dst_fh, $src_fh, 0, $stat[7], sub {
		870	if ($_[0] == $stat[7]) {
		871	$grp->result (0);
		872	close $src_fh;
		873
		874	my $ch = sub {
		875	aioreq_pri $pri;
		876	add $grp aio_chmod $dst_fh, $stat[2] & 07777, sub {
		877	aioreq_pri $pri;
		878	add $grp aio_chown $dst_fh, $stat[4], $stat[5], sub {
		879	aioreq_pri $pri;
		880	add $grp aio_close $dst_fh;
		881	}
		882	};
		883	};
		884
		885	aioreq_pri $pri;
		886	add $grp aio_utime $dst_fh, $stat[8], $stat[9], sub {
		887	if ($_[0] < 0 && $! == ENOSYS) {
		888	aioreq_pri $pri;
		889	add $grp aio_utime $dst, $stat[8], $stat[9], $ch;
		890	} else {
		891	$ch->();
		892	}
		893	};
		894	} else {
		895	$grp->result (-1);
		896	close $src_fh;
		897	close $dst_fh;
		898
		899	aioreq $pri;
		900	add $grp aio_unlink $dst;
		901	}
		902	};
		903	} else {
		904	$grp->result (-1);
		905	}
		906	},
		907
		908	} else {
		909	$grp->result (-1);
		910	}
		911	};
		912
		913	$grp
		914	}
		915
		916	=item aio_move $srcpath, $dstpath, $callback->($status)
		917
		918	Try to move the I<file> (directories not supported as either source or
		919	destination) from C<$srcpath> to C<$dstpath> and call the callback with
		920	a status of C<0> (ok) or C<-1> (error, see C<$!>).
		921
		922	This is a composite request that tries to rename(2) the file first; if
		923	rename fails with C<EXDEV>, it copies the file with C<aio_copy> and, if
		924	that is successful, unlinks the C<$srcpath>.
		925
		926	=cut
		927
		928	sub aio_move($$;$) {
		929	my ($src, $dst, $cb) = @_;
		930
		931	my $pri = aioreq_pri;
		932	my $grp = aio_group $cb;
		933
		934	aioreq_pri $pri;
		935	add $grp aio_rename $src, $dst, sub {
		936	if ($_[0] && $! == EXDEV) {
		937	aioreq_pri $pri;
		938	add $grp aio_copy $src, $dst, sub {
		939	$grp->result ($_[0]);
		940
		941	unless ($_[0]) {
		942	aioreq_pri $pri;
		943	add $grp aio_unlink $src;
		944	}
		945	};
		946	} else {
		947	$grp->result ($_[0]);
		948	}
		949	};
		950
		951	$grp
		952	}
		953
251	=item aio_scandir $path, $maxreq, $callback->($dirs, $nondirs)	954	=item aio_scandir $pathname, $maxreq, $callback->($dirs, $nondirs)
252		955
253	Scans a directory (similar to C<aio_readdir>) and tries to separate the	956	Scans a directory (similar to C<aio_readdir>) but additionally tries to
254	entries of directory C<$path> into two sets of names, ones you can recurse	957	efficiently separate the entries of directory C<$path> into two sets of
255	into (directories), and ones you cannot recurse into (everything else).	958	names, directories you can recurse into (directories), and ones you cannot
		959	recurse into (everything else, including symlinks to directories).
256		960
257	C<aio_scandir> is a composite request that consists of many	961	C<aio_scandir> is a composite request that creates of many sub requests_
258	aio-primitives. C<$maxreq> specifies the maximum number of outstanding	962	C<$maxreq> specifies the maximum number of outstanding aio requests that
259	aio requests that this function generates. If it is C<< <= 0 >>, then a	963	this function generates. If it is C<< <= 0 >>, then a suitable default
260	suitable default will be chosen (currently 8).	964	will be chosen (currently 4).
261		965
262	On error, the callback is called without arguments, otherwise it receives	966	On error, the callback is called without arguments, otherwise it receives
263	two array-refs with path-relative entry names.	967	two array-refs with path-relative entry names.
264		968
265	Example:	969	Example:
…		…
272		976
273	Implementation notes.	977	Implementation notes.
274		978
275	The C<aio_readdir> cannot be avoided, but C<stat()>'ing every entry can.	979	The C<aio_readdir> cannot be avoided, but C<stat()>'ing every entry can.
276		980
		981	If readdir returns file type information, then this is used directly to
		982	find directories.
		983
277	After reading the directory, the modification time, size etc. of the	984	Otherwise, after reading the directory, the modification time, size etc.
278	directory before and after the readdir is checked, and if they match, the	985	of the directory before and after the readdir is checked, and if they
279	link count will be used to decide how many entries are directories (if	986	match (and isn't the current time), the link count will be used to decide
280	>= 2). Otherwise, no knowledge of the number of subdirectories will be	987	how many entries are directories (if >= 2). Otherwise, no knowledge of the
281	assumed.	988	number of subdirectories will be assumed.
282		989
283	Then entires will be sorted into likely directories (everything without a	990	Then entries will be sorted into likely directories a non-initial dot
284	non-initial dot) and likely non-directories (everything else). Then every	991	currently) and likely non-directories (see C<aio_readdirx>). Then every
285	entry + C</.> will be C<stat>'ed, likely directories first. This is often	992	entry plus an appended C</.> will be C<stat>'ed, likely directories first,
		993	in order of their inode numbers. If that succeeds, it assumes that the
		994	entry is a directory or a symlink to directory (which will be checked
		995	separately). This is often faster than stat'ing the entry itself because
286	faster because filesystems might detect the type of the entry without	996	filesystems might detect the type of the entry without reading the inode
287	reading the inode data (e.g. ext2s filetype feature). If that succeeds,	997	data (e.g. ext2fs filetype feature), even on systems that cannot return
288	it assumes that the entry is a directory or a symlink to directory (which	998	the filetype information on readdir.
289	will be checked seperately).
290		999
291	If the known number of directories has been reached, the rest of the	1000	If the known number of directories (link count - 2) has been reached, the
292	entries is assumed to be non-directories.	1001	rest of the entries is assumed to be non-directories.
		1002
		1003	This only works with certainty on POSIX (= UNIX) filesystems, which
		1004	fortunately are the vast majority of filesystems around.
		1005
		1006	It will also likely work on non-POSIX filesystems with reduced efficiency
		1007	as those tend to return 0 or 1 as link counts, which disables the
		1008	directory counting heuristic.
293		1009
294	=cut	1010	=cut
295		1011
296	sub aio_scandir($$$) {	1012	sub aio_scandir($$;$) {
297	my ($path, $maxreq, $cb) = @_;	1013	my ($path, $maxreq, $cb) = @_;
298		1014
		1015	my $pri = aioreq_pri;
		1016
		1017	my $grp = aio_group $cb;
		1018
299	$maxreq = 8 if $maxreq <= 0;	1019	$maxreq = 4 if $maxreq <= 0;
300		1020
		1021	# get a wd object
		1022	aioreq_pri $pri;
		1023	add $grp aio_wd $path, sub {
		1024	$_[0]
		1025	or return $grp->result ();
		1026
		1027	my $wd = [shift, "."];
		1028
301	# stat once	1029	# stat once
302	aio_stat $path, sub {	1030	aioreq_pri $pri;
		1031	add $grp aio_stat $wd, sub {
303	return $cb->() if $_[0];	1032	return $grp->result () if $_[0];
		1033	my $now = time;
304	my $hash1 = join ":", (stat _)[0,1,3,7,9];	1034	my $hash1 = join ":", (stat _)[0,1,3,7,9];
305		1035
306	# read the directory entries	1036	# read the directory entries
307	aio_readdir $path, sub {	1037	aioreq_pri $pri;
		1038	add $grp aio_readdirx $wd, READDIR_DIRS_FIRST, sub {
308	my $entries = shift	1039	my $entries = shift
309	or return $cb->();	1040	or return $grp->result ();
310		1041
311	# stat the dir another time	1042	# stat the dir another time
312	aio_stat $path, sub {	1043	aioreq_pri $pri;
		1044	add $grp aio_stat $wd, sub {
313	my $hash2 = join ":", (stat _)[0,1,3,7,9];	1045	my $hash2 = join ":", (stat _)[0,1,3,7,9];
314		1046
315	my $ndirs;	1047	my $ndirs;
316		1048
317	# take the slow route if anything looks fishy	1049	# take the slow route if anything looks fishy
318	if ($hash1 ne $hash2) {	1050	if ($hash1 ne $hash2 or (stat _)[9] == $now) {
319	$ndirs = -1;	1051	$ndirs = -1;
320	} else {	1052	} else {
321	# if nlink == 2, we are finished	1053	# if nlink == 2, we are finished
322	# on non-posix-fs's, we rely on nlink < 2	1054	# for non-posix-fs's, we rely on nlink < 2
323	$ndirs = (stat _)[3] - 2	1055	$ndirs = (stat _)[3] - 2
324	or return $cb->([], $entries);	1056	or return $grp->result ([], $entries);
325	}
326
327	# sort into likely dirs and likely nondirs
328	# dirs == files without ".", short entries first
329	$entries = [map $_->[0],
330	sort { $b->[1] cmp $a->[1] }
331	map [$_, sprintf "%s%04d", (/.\./ ? "1" : "0"), length],
332	@$entries];
333
334	my (@dirs, @nondirs);
335
336	my ($statcb, $schedcb);
337	my $nreq = 0;
338
339	$schedcb = sub {
340	if (@$entries) {
341	if ($nreq < $maxreq) {
342	my $ent = pop @$entries;
343	$nreq++;
344	aio_stat "$path/$ent/.", sub { $statcb->($_[0], $ent) };
345	}
346	} elsif (!$nreq) {
347	# finished
348	undef $statcb;
349	undef $schedcb;
350	$cb->(\@dirs, \@nondirs) if $cb;
351	undef $cb;
352	}	1057	}
		1058
		1059	my (@dirs, @nondirs);
		1060
		1061	my $statgrp = add $grp aio_group sub {
		1062	$grp->result (\@dirs, \@nondirs);
		1063	};
		1064
		1065	limit $statgrp $maxreq;
		1066	feed $statgrp sub {
		1067	return unless @$entries;
		1068	my $entry = shift @$entries;
		1069
		1070	aioreq_pri $pri;
		1071	$wd->[1] = "$entry/.";
		1072	add $statgrp aio_stat $wd, sub {
		1073	if ($_[0] < 0) {
		1074	push @nondirs, $entry;
		1075	} else {
		1076	# need to check for real directory
		1077	aioreq_pri $pri;
		1078	$wd->[1] = $entry;
		1079	add $statgrp aio_lstat $wd, sub {
		1080	if (-d _) {
		1081	push @dirs, $entry;
		1082
		1083	unless (--$ndirs) {
		1084	push @nondirs, @$entries;
		1085	feed $statgrp;
		1086	}
		1087	} else {
		1088	push @nondirs, $entry;
		1089	}
		1090	}
		1091	}
		1092	};
		1093	};
353	};	1094	};
354	$statcb = sub {
355	my ($status, $entry) = @_;
356
357	if ($status < 0) {
358	$nreq--;
359	push @nondirs, $entry;
360	&$schedcb;
361	} else {
362	# need to check for real directory
363	aio_lstat "$path/$entry", sub {
364	$nreq--;
365
366	if (-d _) {
367	push @dirs, $entry;
368
369	if (!--$ndirs) {
370	push @nondirs, @$entries;
371	$entries = [];
372	}
373	} else {
374	push @nondirs, $entry;
375	}
376
377	&$schedcb;
378	}
379	}
380	};
381
382	&$schedcb while @$entries && $nreq < $maxreq;
383	};	1095	};
384	};	1096	};
385	};	1097	};
		1098
		1099	$grp
386	}	1100	}
		1101
		1102	=item aio_rmtree $pathname, $callback->($status)
		1103
		1104	Delete a directory tree starting (and including) C<$path>, return the
		1105	status of the final C<rmdir> only. This is a composite request that
		1106	uses C<aio_scandir> to recurse into and rmdir directories, and unlink
		1107	everything else.
		1108
		1109	=cut
		1110
		1111	sub aio_rmtree;
		1112	sub aio_rmtree($;$) {
		1113	my ($path, $cb) = @_;
		1114
		1115	my $pri = aioreq_pri;
		1116	my $grp = aio_group $cb;
		1117
		1118	aioreq_pri $pri;
		1119	add $grp aio_scandir $path, 0, sub {
		1120	my ($dirs, $nondirs) = @_;
		1121
		1122	my $dirgrp = aio_group sub {
		1123	add $grp aio_rmdir $path, sub {
		1124	$grp->result ($_[0]);
		1125	};
		1126	};
		1127
		1128	(aioreq_pri $pri), add $dirgrp aio_rmtree "$path/$_" for @$dirs;
		1129	(aioreq_pri $pri), add $dirgrp aio_unlink "$path/$_" for @$nondirs;
		1130
		1131	add $grp $dirgrp;
		1132	};
		1133
		1134	$grp
		1135	}
		1136
		1137	=item aio_sync $callback->($status)
		1138
		1139	Asynchronously call sync and call the callback when finished.
387		1140
388	=item aio_fsync $fh, $callback->($status)	1141	=item aio_fsync $fh, $callback->($status)
389		1142
390	Asynchronously call fsync on the given filehandle and call the callback	1143	Asynchronously call fsync on the given filehandle and call the callback
391	with the fsync result code.	1144	with the fsync result code.
…		…
396	callback with the fdatasync result code.	1149	callback with the fdatasync result code.
397		1150
398	If this call isn't available because your OS lacks it or it couldn't be	1151	If this call isn't available because your OS lacks it or it couldn't be
399	detected, it will be emulated by calling C<fsync> instead.	1152	detected, it will be emulated by calling C<fsync> instead.
400		1153
		1154	=item aio_syncfs $fh, $callback->($status)
		1155
		1156	Asynchronously call the syncfs syscall to sync the filesystem associated
		1157	to the given filehandle and call the callback with the syncfs result
		1158	code. If syncfs is not available, calls sync(), but returns C<-1> and sets
		1159	errno to C<ENOSYS> nevertheless.
		1160
		1161	=item aio_sync_file_range $fh, $offset, $nbytes, $flags, $callback->($status)
		1162
		1163	Sync the data portion of the file specified by C<$offset> and C<$length>
		1164	to disk (but NOT the metadata), by calling the Linux-specific
		1165	sync_file_range call. If sync_file_range is not available or it returns
		1166	ENOSYS, then fdatasync or fsync is being substituted.
		1167
		1168	C<$flags> can be a combination of C<IO::AIO::SYNC_FILE_RANGE_WAIT_BEFORE>,
		1169	C<IO::AIO::SYNC_FILE_RANGE_WRITE> and
		1170	C<IO::AIO::SYNC_FILE_RANGE_WAIT_AFTER>: refer to the sync_file_range
		1171	manpage for details.
		1172
		1173	=item aio_pathsync $pathname, $callback->($status)
		1174
		1175	This request tries to open, fsync and close the given path. This is a
		1176	composite request intended to sync directories after directory operations
		1177	(E.g. rename). This might not work on all operating systems or have any
		1178	specific effect, but usually it makes sure that directory changes get
		1179	written to disc. It works for anything that can be opened for read-only,
		1180	not just directories.
		1181
		1182	Future versions of this function might fall back to other methods when
		1183	C<fsync> on the directory fails (such as calling C<sync>).
		1184
		1185	Passes C<0> when everything went ok, and C<-1> on error.
		1186
		1187	=cut
		1188
		1189	sub aio_pathsync($;$) {
		1190	my ($path, $cb) = @_;
		1191
		1192	my $pri = aioreq_pri;
		1193	my $grp = aio_group $cb;
		1194
		1195	aioreq_pri $pri;
		1196	add $grp aio_open $path, O_RDONLY, 0, sub {
		1197	my ($fh) = @_;
		1198	if ($fh) {
		1199	aioreq_pri $pri;
		1200	add $grp aio_fsync $fh, sub {
		1201	$grp->result ($_[0]);
		1202
		1203	aioreq_pri $pri;
		1204	add $grp aio_close $fh;
		1205	};
		1206	} else {
		1207	$grp->result (-1);
		1208	}
		1209	};
		1210
		1211	$grp
		1212	}
		1213
		1214	=item aio_msync $scalar, $offset = 0, $length = undef, flags = 0, $callback->($status)
		1215
		1216	This is a rather advanced IO::AIO call, which only works on mmap(2)ed
		1217	scalars (see the C<IO::AIO::mmap> function, although it also works on data
		1218	scalars managed by the L<Sys::Mmap> or L<Mmap> modules, note that the
		1219	scalar must only be modified in-place while an aio operation is pending on
		1220	it).
		1221
		1222	It calls the C<msync> function of your OS, if available, with the memory
		1223	area starting at C<$offset> in the string and ending C<$length> bytes
		1224	later. If C<$length> is negative, counts from the end, and if C<$length>
		1225	is C<undef>, then it goes till the end of the string. The flags can be
		1226	a combination of C<IO::AIO::MS_ASYNC>, C<IO::AIO::MS_INVALIDATE> and
		1227	C<IO::AIO::MS_SYNC>.
		1228
		1229	=item aio_mtouch $scalar, $offset = 0, $length = undef, flags = 0, $callback->($status)
		1230
		1231	This is a rather advanced IO::AIO call, which works best on mmap(2)ed
		1232	scalars.
		1233
		1234	It touches (reads or writes) all memory pages in the specified
		1235	range inside the scalar. All caveats and parameters are the same
		1236	as for C<aio_msync>, above, except for flags, which must be either
		1237	C<0> (which reads all pages and ensures they are instantiated) or
		1238	C<IO::AIO::MT_MODIFY>, which modifies the memory page s(by reading and
		1239	writing an octet from it, which dirties the page).
		1240
		1241	=item aio_mlock $scalar, $offset = 0, $length = undef, $callback->($status)
		1242
		1243	This is a rather advanced IO::AIO call, which works best on mmap(2)ed
		1244	scalars.
		1245
		1246	It reads in all the pages of the underlying storage into memory (if any)
		1247	and locks them, so they are not getting swapped/paged out or removed.
		1248
		1249	If C<$length> is undefined, then the scalar will be locked till the end.
		1250
		1251	On systems that do not implement C<mlock>, this function returns C<-1>
		1252	and sets errno to C<ENOSYS>.
		1253
		1254	Note that the corresponding C<munlock> is synchronous and is
		1255	documented under L<MISCELLANEOUS FUNCTIONS>.
		1256
		1257	Example: open a file, mmap and mlock it - both will be undone when
		1258	C<$data> gets destroyed.
		1259
		1260	open my $fh, "<", $path or die "$path: $!";
		1261	my $data;
		1262	IO::AIO::mmap $data, -s $fh, IO::AIO::PROT_READ, IO::AIO::MAP_SHARED, $fh;
		1263	aio_mlock $data; # mlock in background
		1264
		1265	=item aio_mlockall $flags, $callback->($status)
		1266
		1267	Calls the C<mlockall> function with the given C<$flags> (a combination of
		1268	C<IO::AIO::MCL_CURRENT> and C<IO::AIO::MCL_FUTURE>).
		1269
		1270	On systems that do not implement C<mlockall>, this function returns C<-1>
		1271	and sets errno to C<ENOSYS>.
		1272
		1273	Note that the corresponding C<munlockall> is synchronous and is
		1274	documented under L<MISCELLANEOUS FUNCTIONS>.
		1275
		1276	Example: asynchronously lock all current and future pages into memory.
		1277
		1278	aio_mlockall IO::AIO::MCL_FUTURE;
		1279
		1280	=item aio_fiemap $fh, $start, $length, $flags, $count, $cb->(\@extents)
		1281
		1282	Queries the extents of the given file (by calling the Linux FIEMAP ioctl,
		1283	see L<http://cvs.schmorp.de/IO-AIO/doc/fiemap.txt> for details). If the
		1284	C<ioctl> is not available on your OS, then this request will fail with
		1285	C<ENOSYS>.
		1286
		1287	C<$start> is the starting offset to query extents for, C<$length> is the
		1288	size of the range to query - if it is C<undef>, then the whole file will
		1289	be queried.
		1290
		1291	C<$flags> is a combination of flags (C<IO::AIO::FIEMAP_FLAG_SYNC> or
		1292	C<IO::AIO::FIEMAP_FLAG_XATTR> - C<IO::AIO::FIEMAP_FLAGS_COMPAT> is also
		1293	exported), and is normally C<0> or C<IO::AIO::FIEMAP_FLAG_SYNC> to query
		1294	the data portion.
		1295
		1296	C<$count> is the maximum number of extent records to return. If it is
		1297	C<undef>, then IO::AIO queries all extents of the file. As a very special
		1298	case, if it is C<0>, then the callback receives the number of extents
		1299	instead of the extents themselves.
		1300
		1301	If an error occurs, the callback receives no arguments. The special
		1302	C<errno> value C<IO::AIO::EBADR> is available to test for flag errors.
		1303
		1304	Otherwise, the callback receives an array reference with extent
		1305	structures. Each extent structure is an array reference itself, with the
		1306	following members:
		1307
		1308	[$logical, $physical, $length, $flags]
		1309
		1310	Flags is any combination of the following flag values (typically either C<0>
		1311	or C<IO::AIO::FIEMAP_EXTENT_LAST> (1)):
		1312
		1313	C<IO::AIO::FIEMAP_EXTENT_LAST>, C<IO::AIO::FIEMAP_EXTENT_UNKNOWN>,
		1314	C<IO::AIO::FIEMAP_EXTENT_DELALLOC>, C<IO::AIO::FIEMAP_EXTENT_ENCODED>,
		1315	C<IO::AIO::FIEMAP_EXTENT_DATA_ENCRYPTED>, C<IO::AIO::FIEMAP_EXTENT_NOT_ALIGNED>,
		1316	C<IO::AIO::FIEMAP_EXTENT_DATA_INLINE>, C<IO::AIO::FIEMAP_EXTENT_DATA_TAIL>,
		1317	C<IO::AIO::FIEMAP_EXTENT_UNWRITTEN>, C<IO::AIO::FIEMAP_EXTENT_MERGED> or
		1318	C<IO::AIO::FIEMAP_EXTENT_SHARED>.
		1319
		1320	=item aio_group $callback->(...)
		1321
		1322	This is a very special aio request: Instead of doing something, it is a
		1323	container for other aio requests, which is useful if you want to bundle
		1324	many requests into a single, composite, request with a definite callback
		1325	and the ability to cancel the whole request with its subrequests.
		1326
		1327	Returns an object of class L<IO::AIO::GRP>. See its documentation below
		1328	for more info.
		1329
		1330	Example:
		1331
		1332	my $grp = aio_group sub {
		1333	print "all stats done\n";
		1334	};
		1335
		1336	add $grp
		1337	(aio_stat ...),
		1338	(aio_stat ...),
		1339	...;
		1340
		1341	=item aio_nop $callback->()
		1342
		1343	This is a special request - it does nothing in itself and is only used for
		1344	side effects, such as when you want to add a dummy request to a group so
		1345	that finishing the requests in the group depends on executing the given
		1346	code.
		1347
		1348	While this request does nothing, it still goes through the execution
		1349	phase and still requires a worker thread. Thus, the callback will not
		1350	be executed immediately but only after other requests in the queue have
		1351	entered their execution phase. This can be used to measure request
		1352	latency.
		1353
		1354	=item IO::AIO::aio_busy $fractional_seconds, $callback->() *NOT EXPORTED*
		1355
		1356	Mainly used for debugging and benchmarking, this aio request puts one of
		1357	the request workers to sleep for the given time.
		1358
		1359	While it is theoretically handy to have simple I/O scheduling requests
		1360	like sleep and file handle readable/writable, the overhead this creates is
		1361	immense (it blocks a thread for a long time) so do not use this function
		1362	except to put your application under artificial I/O pressure.
		1363
401	=back	1364	=back
402		1365
		1366
		1367	=head2 IO::AIO::WD - multiple working directories
		1368
		1369	Your process only has one current working directory, which is used by all
		1370	threads. This makes it hard to use relative paths (some other component
		1371	could call C<chdir> at any time, and it is hard to control when the path
		1372	will be used by IO::AIO).
		1373
		1374	One solution for this is to always use absolute paths. This usually works,
		1375	but can be quite slow (the kernel has to walk the whole path on every
		1376	access), and can also be a hassle to implement.
		1377
		1378	Newer POSIX systems have a number of functions (openat, fdopendir,
		1379	futimensat and so on) that make it possible to specify working directories
		1380	per operation.
		1381
		1382	For portability, and because the clowns who "designed", or shall I write,
		1383	perpetrated this new interface were obviously half-drunk, this abstraction
		1384	cannot be perfect, though.
		1385
		1386	IO::AIO allows you to convert directory paths into a so-called IO::AIO::WD
		1387	object. This object stores the canonicalised, absolute version of the
		1388	path, and on systems that allow it, also a directory file descriptor.
		1389
		1390	Everywhere where a pathname is accepted by IO::AIO (e.g. in C<aio_stat>
		1391	or C<aio_unlink>), one can specify an array reference with an IO::AIO::WD
		1392	object and a pathname instead (or the IO::AIO::WD object alone, which
		1393	gets interpreted as C<[$wd, "."]>). If the pathname is absolute, the
		1394	IO::AIO::WD object is ignored, otherwise the pathname is resolved relative
		1395	to that IO::AIO::WD object.
		1396
		1397	For example, to get a wd object for F</etc> and then stat F<passwd>
		1398	inside, you would write:
		1399
		1400	aio_wd "/etc", sub {
		1401	my $etcdir = shift;
		1402
		1403	# although $etcdir can be undef on error, there is generally no reason
		1404	# to check for errors here, as aio_stat will fail with ENOENT
		1405	# when $etcdir is undef.
		1406
		1407	aio_stat [$etcdir, "passwd"], sub {
		1408	# yay
		1409	};
		1410	};
		1411
		1412	That C<aio_wd> is a request and not a normal function shows that creating
		1413	an IO::AIO::WD object is itself a potentially blocking operation, which is
		1414	why it is done asynchronously.
		1415
		1416	To stat the directory obtained with C<aio_wd> above, one could write
		1417	either of the following three request calls:
		1418
		1419	aio_lstat "/etc" , sub { ... # pathname as normal string
		1420	aio_lstat [$wd, "."], sub { ... # "." relative to $wd (i.e. $wd itself)
		1421	aio_lstat $wd , sub { ... # shorthand for the previous
		1422
		1423	As with normal pathnames, IO::AIO keeps a copy of the working directory
		1424	object and the pathname string, so you could write the following without
		1425	causing any issues due to C<$path> getting reused:
		1426
		1427	my $path = [$wd, undef];
		1428
		1429	for my $name (qw(abc def ghi)) {
		1430	$path->[1] = $name;
		1431	aio_stat $path, sub {
		1432	# ...
		1433	};
		1434	}
		1435
		1436	There are some caveats: when directories get renamed (or deleted), the
		1437	pathname string doesn't change, so will point to the new directory (or
		1438	nowhere at all), while the directory fd, if available on the system,
		1439	will still point to the original directory. Most functions accepting a
		1440	pathname will use the directory fd on newer systems, and the string on
		1441	older systems. Some functions (such as realpath) will always rely on the
		1442	string form of the pathname.
		1443
		1444	So this fucntionality is mainly useful to get some protection against
		1445	C<chdir>, to easily get an absolute path out of a relative path for future
		1446	reference, and to speed up doing many operations in the same directory
		1447	(e.g. when stat'ing all files in a directory).
		1448
		1449	The following functions implement this working directory abstraction:
		1450
		1451	=over 4
		1452
		1453	=item aio_wd $pathname, $callback->($wd)
		1454
		1455	Asynchonously canonicalise the given pathname and convert it to an
		1456	IO::AIO::WD object representing it. If possible and supported on the
		1457	system, also open a directory fd to speed up pathname resolution relative
		1458	to this working directory.
		1459
		1460	If something goes wrong, then C<undef> is passwd to the callback instead
		1461	of a working directory object and C<$!> is set appropriately. Since
		1462	passing C<undef> as working directory component of a pathname fails the
		1463	request with C<ENOENT>, there is often no need for error checking in the
		1464	C<aio_wd> callback, as future requests using the value will fail in the
		1465	expected way.
		1466
		1467	If this call isn't available because your OS lacks it or it couldn't be
		1468	detected, it will be emulated by calling C<fsync> instead.
		1469
		1470	=item IO::AIO::CWD
		1471
		1472	This is a compiletime constant (object) that represents the process
		1473	current working directory.
		1474
		1475	Specifying this object as working directory object for a pathname is as
		1476	if the pathname would be specified directly, without a directory object,
		1477	e.g., these calls are functionally identical:
		1478
		1479	aio_stat "somefile", sub { ... };
		1480	aio_stat [IO::AIO::CWD, "somefile"], sub { ... };
		1481
		1482	=back
		1483
		1484
		1485	=head2 IO::AIO::REQ CLASS
		1486
		1487	All non-aggregate C<aio_*> functions return an object of this class when
		1488	called in non-void context.
		1489
		1490	=over 4
		1491
		1492	=item cancel $req
		1493
		1494	Cancels the request, if possible. Has the effect of skipping execution
		1495	when entering the B<execute> state and skipping calling the callback when
		1496	entering the the B<result> state, but will leave the request otherwise
		1497	untouched (with the exception of readdir). That means that requests that
		1498	currently execute will not be stopped and resources held by the request
		1499	will not be freed prematurely.
		1500
		1501	=item cb $req $callback->(...)
		1502
		1503	Replace (or simply set) the callback registered to the request.
		1504
		1505	=back
		1506
		1507	=head2 IO::AIO::GRP CLASS
		1508
		1509	This class is a subclass of L<IO::AIO::REQ>, so all its methods apply to
		1510	objects of this class, too.
		1511
		1512	A IO::AIO::GRP object is a special request that can contain multiple other
		1513	aio requests.
		1514
		1515	You create one by calling the C<aio_group> constructing function with a
		1516	callback that will be called when all contained requests have entered the
		1517	C<done> state:
		1518
		1519	my $grp = aio_group sub {
		1520	print "all requests are done\n";
		1521	};
		1522
		1523	You add requests by calling the C<add> method with one or more
		1524	C<IO::AIO::REQ> objects:
		1525
		1526	$grp->add (aio_unlink "...");
		1527
		1528	add $grp aio_stat "...", sub {
		1529	$_[0] or return $grp->result ("error");
		1530
		1531	# add another request dynamically, if first succeeded
		1532	add $grp aio_open "...", sub {
		1533	$grp->result ("ok");
		1534	};
		1535	};
		1536
		1537	This makes it very easy to create composite requests (see the source of
		1538	C<aio_move> for an application) that work and feel like simple requests.
		1539
		1540	=over 4
		1541
		1542	=item * The IO::AIO::GRP objects will be cleaned up during calls to
		1543	C<IO::AIO::poll_cb>, just like any other request.
		1544
		1545	=item * They can be canceled like any other request. Canceling will cancel not
		1546	only the request itself, but also all requests it contains.
		1547
		1548	=item * They can also can also be added to other IO::AIO::GRP objects.
		1549
		1550	=item * You must not add requests to a group from within the group callback (or
		1551	any later time).
		1552
		1553	=back
		1554
		1555	Their lifetime, simplified, looks like this: when they are empty, they
		1556	will finish very quickly. If they contain only requests that are in the
		1557	C<done> state, they will also finish. Otherwise they will continue to
		1558	exist.
		1559
		1560	That means after creating a group you have some time to add requests
		1561	(precisely before the callback has been invoked, which is only done within
		1562	the C<poll_cb>). And in the callbacks of those requests, you can add
		1563	further requests to the group. And only when all those requests have
		1564	finished will the the group itself finish.
		1565
		1566	=over 4
		1567
		1568	=item add $grp ...
		1569
		1570	=item $grp->add (...)
		1571
		1572	Add one or more requests to the group. Any type of L<IO::AIO::REQ> can
		1573	be added, including other groups, as long as you do not create circular
		1574	dependencies.
		1575
		1576	Returns all its arguments.
		1577
		1578	=item $grp->cancel_subs
		1579
		1580	Cancel all subrequests and clears any feeder, but not the group request
		1581	itself. Useful when you queued a lot of events but got a result early.
		1582
		1583	The group request will finish normally (you cannot add requests to the
		1584	group).
		1585
		1586	=item $grp->result (...)
		1587
		1588	Set the result value(s) that will be passed to the group callback when all
		1589	subrequests have finished and set the groups errno to the current value
		1590	of errno (just like calling C<errno> without an error number). By default,
		1591	no argument will be passed and errno is zero.
		1592
		1593	=item $grp->errno ([$errno])
		1594
		1595	Sets the group errno value to C<$errno>, or the current value of errno
		1596	when the argument is missing.
		1597
		1598	Every aio request has an associated errno value that is restored when
		1599	the callback is invoked. This method lets you change this value from its
		1600	default (0).
		1601
		1602	Calling C<result> will also set errno, so make sure you either set C<$!>
		1603	before the call to C<result>, or call c<errno> after it.
		1604
		1605	=item feed $grp $callback->($grp)
		1606
		1607	Sets a feeder/generator on this group: every group can have an attached
		1608	generator that generates requests if idle. The idea behind this is that,
		1609	although you could just queue as many requests as you want in a group,
		1610	this might starve other requests for a potentially long time. For example,
		1611	C<aio_scandir> might generate hundreds of thousands of C<aio_stat>
		1612	requests, delaying any later requests for a long time.
		1613
		1614	To avoid this, and allow incremental generation of requests, you can
		1615	instead a group and set a feeder on it that generates those requests. The
		1616	feed callback will be called whenever there are few enough (see C<limit>,
		1617	below) requests active in the group itself and is expected to queue more
		1618	requests.
		1619
		1620	The feed callback can queue as many requests as it likes (i.e. C<add> does
		1621	not impose any limits).
		1622
		1623	If the feed does not queue more requests when called, it will be
		1624	automatically removed from the group.
		1625
		1626	If the feed limit is C<0> when this method is called, it will be set to
		1627	C<2> automatically.
		1628
		1629	Example:
		1630
		1631	# stat all files in @files, but only ever use four aio requests concurrently:
		1632
		1633	my $grp = aio_group sub { print "finished\n" };
		1634	limit $grp 4;
		1635	feed $grp sub {
		1636	my $file = pop @files
		1637	or return;
		1638
		1639	add $grp aio_stat $file, sub { ... };
		1640	};
		1641
		1642	=item limit $grp $num
		1643
		1644	Sets the feeder limit for the group: The feeder will be called whenever
		1645	the group contains less than this many requests.
		1646
		1647	Setting the limit to C<0> will pause the feeding process.
		1648
		1649	The default value for the limit is C<0>, but note that setting a feeder
		1650	automatically bumps it up to C<2>.
		1651
		1652	=back
		1653
403	=head2 SUPPORT FUNCTIONS	1654	=head2 SUPPORT FUNCTIONS
404		1655
		1656	=head3 EVENT PROCESSING AND EVENT LOOP INTEGRATION
		1657
405	=over 4	1658	=over 4
406		1659
407	=item $fileno = IO::AIO::poll_fileno	1660	=item $fileno = IO::AIO::poll_fileno
408		1661
409	Return the I<request result pipe file descriptor>. This filehandle must be	1662	Return the I<request result pipe file descriptor>. This filehandle must be
410	polled for reading by some mechanism outside this module (e.g. Event or	1663	polled for reading by some mechanism outside this module (e.g. EV, Glib,
411	select, see below or the SYNOPSIS). If the pipe becomes readable you have	1664	select and so on, see below or the SYNOPSIS). If the pipe becomes readable
412	to call C<poll_cb> to check the results.	1665	you have to call C<poll_cb> to check the results.
413		1666
414	See C<poll_cb> for an example.	1667	See C<poll_cb> for an example.
415		1668
416	=item IO::AIO::poll_cb	1669	=item IO::AIO::poll_cb
417		1670
418	Process all outstanding events on the result pipe. You have to call this	1671	Process some outstanding events on the result pipe. You have to call
419	regularly. Returns the number of events processed. Returns immediately	1672	this regularly. Returns C<0> if all events could be processed (or there
420	when no events are outstanding.	1673	were no events to process), or C<-1> if it returned earlier for whatever
		1674	reason. Returns immediately when no events are outstanding. The amount of
		1675	events processed depends on the settings of C<IO::AIO::max_poll_req> and
		1676	C<IO::AIO::max_poll_time>.
		1677
		1678	If not all requests were processed for whatever reason, the filehandle
		1679	will still be ready when C<poll_cb> returns, so normally you don't have to
		1680	do anything special to have it called later.
		1681
		1682	Apart from calling C<IO::AIO::poll_cb> when the event filehandle becomes
		1683	ready, it can be beneficial to call this function from loops which submit
		1684	a lot of requests, to make sure the results get processed when they become
		1685	available and not just when the loop is finished and the event loop takes
		1686	over again. This function returns very fast when there are no outstanding
		1687	requests.
421		1688
422	Example: Install an Event watcher that automatically calls	1689	Example: Install an Event watcher that automatically calls
423	IO::AIO::poll_cb with high priority:	1690	IO::AIO::poll_cb with high priority (more examples can be found in the
		1691	SYNOPSIS section, at the top of this document):
424		1692
425	Event->io (fd => IO::AIO::poll_fileno,	1693	Event->io (fd => IO::AIO::poll_fileno,
426	poll => 'r', async => 1,	1694	poll => 'r', async => 1,
427	cb => \&IO::AIO::poll_cb);	1695	cb => \&IO::AIO::poll_cb);
428		1696
429	=item IO::AIO::poll_wait	1697	=item IO::AIO::poll_wait
430		1698
		1699	If there are any outstanding requests and none of them in the result
431	Wait till the result filehandle becomes ready for reading (simply does a	1700	phase, wait till the result filehandle becomes ready for reading (simply
432	C<select> on the filehandle. This is useful if you want to synchronously wait	1701	does a C<select> on the filehandle. This is useful if you want to
433	for some requests to finish).	1702	synchronously wait for some requests to finish).
434		1703
435	See C<nreqs> for an example.	1704	See C<nreqs> for an example.
436		1705
		1706	=item IO::AIO::poll
		1707
		1708	Waits until some requests have been handled.
		1709
		1710	Returns the number of requests processed, but is otherwise strictly
		1711	equivalent to:
		1712
		1713	IO::AIO::poll_wait, IO::AIO::poll_cb
		1714
437	=item IO::AIO::nreqs	1715	=item IO::AIO::flush
438		1716
439	Returns the number of requests currently outstanding (i.e. for which their	1717	Wait till all outstanding AIO requests have been handled.
440	callback has not been invoked yet).
441		1718
442	Example: wait till there are no outstanding requests anymore:	1719	Strictly equivalent to:
443		1720
444	IO::AIO::poll_wait, IO::AIO::poll_cb	1721	IO::AIO::poll_wait, IO::AIO::poll_cb
445	while IO::AIO::nreqs;	1722	while IO::AIO::nreqs;
446		1723
447	=item IO::AIO::flush	1724	=item IO::AIO::max_poll_reqs $nreqs
448		1725
449	Wait till all outstanding AIO requests have been handled.	1726	=item IO::AIO::max_poll_time $seconds
450		1727
451	Strictly equivalent to:	1728	These set the maximum number of requests (default C<0>, meaning infinity)
		1729	that are being processed by C<IO::AIO::poll_cb> in one call, respectively
		1730	the maximum amount of time (default C<0>, meaning infinity) spent in
		1731	C<IO::AIO::poll_cb> to process requests (more correctly the mininum amount
		1732	of time C<poll_cb> is allowed to use).
452		1733
453	IO::AIO::poll_wait, IO::AIO::poll_cb	1734	Setting C<max_poll_time> to a non-zero value creates an overhead of one
454	while IO::AIO::nreqs;	1735	syscall per request processed, which is not normally a problem unless your
		1736	callbacks are really really fast or your OS is really really slow (I am
		1737	not mentioning Solaris here). Using C<max_poll_reqs> incurs no overhead.
455		1738
456	=item IO::AIO::poll	1739	Setting these is useful if you want to ensure some level of
		1740	interactiveness when perl is not fast enough to process all requests in
		1741	time.
457		1742
458	Waits until some requests have been handled.	1743	For interactive programs, values such as C<0.01> to C<0.1> should be fine.
459		1744
460	Strictly equivalent to:	1745	Example: Install an Event watcher that automatically calls
		1746	IO::AIO::poll_cb with low priority, to ensure that other parts of the
		1747	program get the CPU sometimes even under high AIO load.
461		1748
462	IO::AIO::poll_wait, IO::AIO::poll_cb	1749	# try not to spend much more than 0.1s in poll_cb
463	if IO::AIO::nreqs;	1750	IO::AIO::max_poll_time 0.1;
		1751
		1752	# use a low priority so other tasks have priority
		1753	Event->io (fd => IO::AIO::poll_fileno,
		1754	poll => 'r', nice => 1,
		1755	cb => &IO::AIO::poll_cb);
		1756
		1757	=back
		1758
		1759	=head3 CONTROLLING THE NUMBER OF THREADS
		1760
		1761	=over
464		1762
465	=item IO::AIO::min_parallel $nthreads	1763	=item IO::AIO::min_parallel $nthreads
466		1764
467	Set the minimum number of AIO threads to C<$nthreads>. The current default	1765	Set the minimum number of AIO threads to C<$nthreads>. The current
468	is C<4>, which means four asynchronous operations can be done at one time	1766	default is C<8>, which means eight asynchronous operations can execute
469	(the number of outstanding operations, however, is unlimited).	1767	concurrently at any one time (the number of outstanding requests,
		1768	however, is unlimited).
470		1769
471	IO::AIO starts threads only on demand, when an AIO request is queued and	1770	IO::AIO starts threads only on demand, when an AIO request is queued and
472	no free thread exists.	1771	no free thread exists. Please note that queueing up a hundred requests can
		1772	create demand for a hundred threads, even if it turns out that everything
		1773	is in the cache and could have been processed faster by a single thread.
473		1774
474	It is recommended to keep the number of threads low, as some Linux	1775	It is recommended to keep the number of threads relatively low, as some
475	kernel versions will scale negatively with the number of threads (higher	1776	Linux kernel versions will scale negatively with the number of threads
476	parallelity => MUCH higher latency). With current Linux 2.6 versions, 4-32	1777	(higher parallelity => MUCH higher latency). With current Linux 2.6
477	threads should be fine.	1778	versions, 4-32 threads should be fine.
478		1779
479	Under most circumstances you don't need to call this function, as the	1780	Under most circumstances you don't need to call this function, as the
480	module selects a default that is suitable for low to moderate load.	1781	module selects a default that is suitable for low to moderate load.
481		1782
482	=item IO::AIO::max_parallel $nthreads	1783	=item IO::AIO::max_parallel $nthreads
…		…
491	This module automatically runs C<max_parallel 0> at program end, to ensure	1792	This module automatically runs C<max_parallel 0> at program end, to ensure
492	that all threads are killed and that there are no outstanding requests.	1793	that all threads are killed and that there are no outstanding requests.
493		1794
494	Under normal circumstances you don't need to call this function.	1795	Under normal circumstances you don't need to call this function.
495		1796
		1797	=item IO::AIO::max_idle $nthreads
		1798
		1799	Limit the number of threads (default: 4) that are allowed to idle
		1800	(i.e., threads that did not get a request to process within the idle
		1801	timeout (default: 10 seconds). That means if a thread becomes idle while
		1802	C<$nthreads> other threads are also idle, it will free its resources and
		1803	exit.
		1804
		1805	This is useful when you allow a large number of threads (e.g. 100 or 1000)
		1806	to allow for extremely high load situations, but want to free resources
		1807	under normal circumstances (1000 threads can easily consume 30MB of RAM).
		1808
		1809	The default is probably ok in most situations, especially if thread
		1810	creation is fast. If thread creation is very slow on your system you might
		1811	want to use larger values.
		1812
		1813	=item IO::AIO::idle_timeout $seconds
		1814
		1815	Sets the minimum idle timeout (default 10) after which worker threads are
		1816	allowed to exit. SEe C<IO::AIO::max_idle>.
		1817
496	=item $oldnreqs = IO::AIO::max_outstanding $nreqs	1818	=item IO::AIO::max_outstanding $maxreqs
497		1819
498	Sets the maximum number of outstanding requests to C<$nreqs>. If you	1820	Sets the maximum number of outstanding requests to C<$nreqs>. If
499	try to queue up more than this number of requests, the caller will block until	1821	you do queue up more than this number of requests, the next call to
500	some requests have been handled.	1822	C<IO::AIO::poll_cb> (and other functions calling C<poll_cb>, such as
		1823	C<IO::AIO::flush> or C<IO::AIO::poll>) will block until the limit is no
		1824	longer exceeded.
501		1825
502	The default is very large, so normally there is no practical limit. If you	1826	In other words, this setting does not enforce a queue limit, but can be
503	queue up many requests in a loop it often improves speed if you set	1827	used to make poll functions block if the limit is exceeded.
504	this to a relatively low number, such as C<100>.
505		1828
506	Under normal circumstances you don't need to call this function.	1829	This is a very bad function to use in interactive programs because it
		1830	blocks, and a bad way to reduce concurrency because it is inexact: Better
		1831	use an C<aio_group> together with a feed callback.
		1832
		1833	It's main use is in scripts without an event loop - when you want to stat
		1834	a lot of files, you can write somehting like this:
		1835
		1836	IO::AIO::max_outstanding 32;
		1837
		1838	for my $path (...) {
		1839	aio_stat $path , ...;
		1840	IO::AIO::poll_cb;
		1841	}
		1842
		1843	IO::AIO::flush;
		1844
		1845	The call to C<poll_cb> inside the loop will normally return instantly, but
		1846	as soon as more thna C<32> reqeusts are in-flight, it will block until
		1847	some requests have been handled. This keeps the loop from pushing a large
		1848	number of C<aio_stat> requests onto the queue.
		1849
		1850	The default value for C<max_outstanding> is very large, so there is no
		1851	practical limit on the number of outstanding requests.
507		1852
508	=back	1853	=back
509		1854
		1855	=head3 STATISTICAL INFORMATION
		1856
		1857	=over
		1858
		1859	=item IO::AIO::nreqs
		1860
		1861	Returns the number of requests currently in the ready, execute or pending
		1862	states (i.e. for which their callback has not been invoked yet).
		1863
		1864	Example: wait till there are no outstanding requests anymore:
		1865
		1866	IO::AIO::poll_wait, IO::AIO::poll_cb
		1867	while IO::AIO::nreqs;
		1868
		1869	=item IO::AIO::nready
		1870
		1871	Returns the number of requests currently in the ready state (not yet
		1872	executed).
		1873
		1874	=item IO::AIO::npending
		1875
		1876	Returns the number of requests currently in the pending state (executed,
		1877	but not yet processed by poll_cb).
		1878
		1879	=back
		1880
		1881	=head3 MISCELLANEOUS FUNCTIONS
		1882
		1883	IO::AIO implements some functions that might be useful, but are not
		1884	asynchronous.
		1885
		1886	=over 4
		1887
		1888	=item IO::AIO::sendfile $ofh, $ifh, $offset, $count
		1889
		1890	Calls the C<eio_sendfile_sync> function, which is like C<aio_sendfile>,
		1891	but is blocking (this makes most sense if you know the input data is
		1892	likely cached already and the output filehandle is set to non-blocking
		1893	operations).
		1894
		1895	Returns the number of bytes copied, or C<-1> on error.
		1896
		1897	=item IO::AIO::fadvise $fh, $offset, $len, $advice
		1898
		1899	Simply calls the C<posix_fadvise> function (see its
		1900	manpage for details). The following advice constants are
		1901	available: C<IO::AIO::FADV_NORMAL>, C<IO::AIO::FADV_SEQUENTIAL>,
		1902	C<IO::AIO::FADV_RANDOM>, C<IO::AIO::FADV_NOREUSE>,
		1903	C<IO::AIO::FADV_WILLNEED>, C<IO::AIO::FADV_DONTNEED>.
		1904
		1905	On systems that do not implement C<posix_fadvise>, this function returns
		1906	ENOSYS, otherwise the return value of C<posix_fadvise>.
		1907
		1908	=item IO::AIO::madvise $scalar, $offset, $len, $advice
		1909
		1910	Simply calls the C<posix_madvise> function (see its
		1911	manpage for details). The following advice constants are
		1912	available: C<IO::AIO::MADV_NORMAL>, C<IO::AIO::MADV_SEQUENTIAL>,
		1913	C<IO::AIO::MADV_RANDOM>, C<IO::AIO::MADV_WILLNEED>, C<IO::AIO::MADV_DONTNEED>.
		1914
		1915	On systems that do not implement C<posix_madvise>, this function returns
		1916	ENOSYS, otherwise the return value of C<posix_madvise>.
		1917
		1918	=item IO::AIO::mprotect $scalar, $offset, $len, $protect
		1919
		1920	Simply calls the C<mprotect> function on the preferably AIO::mmap'ed
		1921	$scalar (see its manpage for details). The following protect
		1922	constants are available: C<IO::AIO::PROT_NONE>, C<IO::AIO::PROT_READ>,
		1923	C<IO::AIO::PROT_WRITE>, C<IO::AIO::PROT_EXEC>.
		1924
		1925	On systems that do not implement C<mprotect>, this function returns
		1926	ENOSYS, otherwise the return value of C<mprotect>.
		1927
		1928	=item IO::AIO::mmap $scalar, $length, $prot, $flags, $fh[, $offset]
		1929
		1930	Memory-maps a file (or anonymous memory range) and attaches it to the
		1931	given C<$scalar>, which will act like a string scalar. Returns true on
		1932	success, and false otherwise.
		1933
		1934	The only operations allowed on the scalar are C<substr>/C<vec> that don't
		1935	change the string length, and most read-only operations such as copying it
		1936	or searching it with regexes and so on.
		1937
		1938	Anything else is unsafe and will, at best, result in memory leaks.
		1939
		1940	The memory map associated with the C<$scalar> is automatically removed
		1941	when the C<$scalar> is destroyed, or when the C<IO::AIO::mmap> or
		1942	C<IO::AIO::munmap> functions are called.
		1943
		1944	This calls the C<mmap>(2) function internally. See your system's manual
		1945	page for details on the C<$length>, C<$prot> and C<$flags> parameters.
		1946
		1947	The C<$length> must be larger than zero and smaller than the actual
		1948	filesize.
		1949
		1950	C<$prot> is a combination of C<IO::AIO::PROT_NONE>, C<IO::AIO::PROT_EXEC>,
		1951	C<IO::AIO::PROT_READ> and/or C<IO::AIO::PROT_WRITE>,
		1952
		1953	C<$flags> can be a combination of C<IO::AIO::MAP_SHARED> or
		1954	C<IO::AIO::MAP_PRIVATE>, or a number of system-specific flags (when
		1955	not available, the are defined as 0): C<IO::AIO::MAP_ANONYMOUS>
		1956	(which is set to C<MAP_ANON> if your system only provides this
		1957	constant), C<IO::AIO::MAP_HUGETLB>, C<IO::AIO::MAP_LOCKED>,
		1958	C<IO::AIO::MAP_NORESERVE>, C<IO::AIO::MAP_POPULATE> or
		1959	C<IO::AIO::MAP_NONBLOCK>
		1960
		1961	If C<$fh> is C<undef>, then a file descriptor of C<-1> is passed.
		1962
		1963	C<$offset> is the offset from the start of the file - it generally must be
		1964	a multiple of C<IO::AIO::PAGESIZE> and defaults to C<0>.
		1965
		1966	Example:
		1967
		1968	use Digest::MD5;
		1969	use IO::AIO;
		1970
		1971	open my $fh, "<verybigfile"
		1972	or die "$!";
		1973
		1974	IO::AIO::mmap my $data, -s $fh, IO::AIO::PROT_READ, IO::AIO::MAP_SHARED, $fh
		1975	or die "verybigfile: $!";
		1976
		1977	my $fast_md5 = md5 $data;
		1978
		1979	=item IO::AIO::munmap $scalar
		1980
		1981	Removes a previous mmap and undefines the C<$scalar>.
		1982
		1983	=item IO::AIO::munlock $scalar, $offset = 0, $length = undef
		1984
		1985	Calls the C<munlock> function, undoing the effects of a previous
		1986	C<aio_mlock> call (see its description for details).
		1987
		1988	=item IO::AIO::munlockall
		1989
		1990	Calls the C<munlockall> function.
		1991
		1992	On systems that do not implement C<munlockall>, this function returns
		1993	ENOSYS, otherwise the return value of C<munlockall>.
		1994
		1995	=item IO::AIO::splice $r_fh, $r_off, $w_fh, $w_off, $length, $flags
		1996
		1997	Calls the GNU/Linux C<splice(2)> syscall, if available. If C<$r_off> or
		1998	C<$w_off> are C<undef>, then C<NULL> is passed for these, otherwise they
		1999	should be the file offset.
		2000
		2001	C<$r_fh> and C<$w_fh> should not refer to the same file, as splice might
		2002	silently corrupt the data in this case.
		2003
		2004	The following symbol flag values are available: C<IO::AIO::SPLICE_F_MOVE>,
		2005	C<IO::AIO::SPLICE_F_NONBLOCK>, C<IO::AIO::SPLICE_F_MORE> and
		2006	C<IO::AIO::SPLICE_F_GIFT>.
		2007
		2008	See the C<splice(2)> manpage for details.
		2009
		2010	=item IO::AIO::tee $r_fh, $w_fh, $length, $flags
		2011
		2012	Calls the GNU/Linux C<tee(2)> syscall, see it's manpage and the
		2013	description for C<IO::AIO::splice> above for details.
		2014
		2015	=back
		2016
510	=cut	2017	=cut
511		2018
512	# support function to convert a fd into a perl filehandle
513	sub _fd2fh {
514	return undef if $_[0] < 0;
515
516	# try to generate nice filehandles
517	my $sym = "IO::AIO::fd#$_[0]";
518	local *$sym;
519
520	open *$sym, "+<&=$_[0]" # usually works under any unix
521	or open *$sym, "<&=$_[0]" # cygwin needs this
522	or open *$sym, ">&=$_[0]" # or this
523	or return undef;
524
525	*$sym
526	}
527
528	min_parallel 4;	2019	min_parallel 8;
529		2020
530	END {	2021	END { flush }
531	max_parallel 0;
532	}
533		2022
534	1;	2023	1;
535		2024
		2025	=head1 EVENT LOOP INTEGRATION
		2026
		2027	It is recommended to use L<AnyEvent::AIO> to integrate IO::AIO
		2028	automatically into many event loops:
		2029
		2030	# AnyEvent integration (EV, Event, Glib, Tk, POE, urxvt, pureperl...)
		2031	use AnyEvent::AIO;
		2032
		2033	You can also integrate IO::AIO manually into many event loops, here are
		2034	some examples of how to do this:
		2035
		2036	# EV integration
		2037	my $aio_w = EV::io IO::AIO::poll_fileno, EV::READ, \&IO::AIO::poll_cb;
		2038
		2039	# Event integration
		2040	Event->io (fd => IO::AIO::poll_fileno,
		2041	poll => 'r',
		2042	cb => \&IO::AIO::poll_cb);
		2043
		2044	# Glib/Gtk2 integration
		2045	add_watch Glib::IO IO::AIO::poll_fileno,
		2046	in => sub { IO::AIO::poll_cb; 1 };
		2047
		2048	# Tk integration
		2049	Tk::Event::IO->fileevent (IO::AIO::poll_fileno, "",
		2050	readable => \&IO::AIO::poll_cb);
		2051
		2052	# Danga::Socket integration
		2053	Danga::Socket->AddOtherFds (IO::AIO::poll_fileno =>
		2054	\&IO::AIO::poll_cb);
		2055
536	=head2 FORK BEHAVIOUR	2056	=head2 FORK BEHAVIOUR
537		2057
538	Before the fork, IO::AIO enters a quiescent state where no requests	2058	Usage of pthreads in a program changes the semantics of fork
539	can be added in other threads and no results will be processed. After	2059	considerably. Specifically, only async-safe functions can be called after
540	the fork the parent simply leaves the quiescent state and continues	2060	fork. Perl doesn't know about this, so in general, you cannot call fork
541	request/result processing, while the child clears the request/result	2061	with defined behaviour in perl if pthreads are involved. IO::AIO uses
542	queue (so the requests started before the fork will only be handled in	2062	pthreads, so this applies, but many other extensions and (for inexplicable
543	the parent). Threats will be started on demand until the limit ste in the	2063	reasons) perl itself often is linked against pthreads, so this limitation
544	parent process has been reached again.	2064	applies to quite a lot of perls.
		2065
		2066	This module no longer tries to fight your OS, or POSIX. That means IO::AIO
		2067	only works in the process that loaded it. Forking is fully supported, but
		2068	using IO::AIO in the child is not.
		2069
		2070	You might get around by not I<using> IO::AIO before (or after)
		2071	forking. You could also try to call the L<IO::AIO::reinit> function in the
		2072	child:
		2073
		2074	=over 4
		2075
		2076	=item IO::AIO::reinit
		2077
		2078	Abandons all current requests and I/O threads and simply reinitialises all
		2079	data structures. This is not an operation supported by any standards, but
		2080	happens to work on GNU/Linux and some newer BSD systems.
		2081
		2082	The only reasonable use for this function is to call it after forking, if
		2083	C<IO::AIO> was used in the parent. Calling it while IO::AIO is active in
		2084	the process will result in undefined behaviour. Calling it at any time
		2085	will also result in any undefined (by POSIX) behaviour.
		2086
		2087	=back
		2088
		2089	=head2 MEMORY USAGE
		2090
		2091	Per-request usage:
		2092
		2093	Each aio request uses - depending on your architecture - around 100-200
		2094	bytes of memory. In addition, stat requests need a stat buffer (possibly
		2095	a few hundred bytes), readdir requires a result buffer and so on. Perl
		2096	scalars and other data passed into aio requests will also be locked and
		2097	will consume memory till the request has entered the done state.
		2098
		2099	This is not awfully much, so queuing lots of requests is not usually a
		2100	problem.
		2101
		2102	Per-thread usage:
		2103
		2104	In the execution phase, some aio requests require more memory for
		2105	temporary buffers, and each thread requires a stack and other data
		2106	structures (usually around 16k-128k, depending on the OS).
		2107
		2108	=head1 KNOWN BUGS
		2109
		2110	Known bugs will be fixed in the next release.
545		2111
546	=head1 SEE ALSO	2112	=head1 SEE ALSO
547		2113
548	L<Coro>, L<Linux::AIO>.	2114	L<AnyEvent::AIO> for easy integration into event loops, L<Coro::AIO> for a
		2115	more natural syntax.
549		2116
550	=head1 AUTHOR	2117	=head1 AUTHOR
551		2118
552	Marc Lehmann <schmorp@schmorp.de>	2119	Marc Lehmann <schmorp@schmorp.de>
553	http://home.schmorp.de/	2120	http://home.schmorp.de/

Diff Legend

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