ViewVC Help

View File | Revision Log | Show Annotations | Download File

/cvs/IO-AIO/AIO.pm

Comparing IO-AIO/AIO.pm (file contents):
Revision 1.85 by root, Sat Oct 28 01:40:30 2006 UTC vs.
Revision 1.248 by root, Sun Jul 27 22:10:53 2014 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
…		…
25	$req->cancel; # cancel request if still in queue	26	$req->cancel; # cancel request if still in queue
26		27
27	my $grp = aio_group sub { print "all stats done\n" };	28	my $grp = aio_group sub { print "all stats done\n" };
28	add $grp aio_stat "..." for ...;	29	add $grp aio_stat "..." for ...;
29		30
30	# AnyEvent integration
31	open my $fh, "<&=" . IO::AIO::poll_fileno or die "$!";
32	my $w = AnyEvent->io (fh => $fh, poll => 'r', cb => sub { IO::AIO::poll_cb });
33
34	# Event integration
35	Event->io (fd => IO::AIO::poll_fileno,
36	poll => 'r',
37	cb => \&IO::AIO::poll_cb);
38
39	# Glib/Gtk2 integration
40	add_watch Glib::IO IO::AIO::poll_fileno,
41	in => sub { IO::AIO::poll_cb; 1 };
42
43	# Tk integration
44	Tk::Event::IO->fileevent (IO::AIO::poll_fileno, "",
45	readable => \&IO::AIO::poll_cb);
46
47	# Danga::Socket integration
48	Danga::Socket->AddOtherFds (IO::AIO::poll_fileno =>
49	\&IO::AIO::poll_cb);
50
51	=head1 DESCRIPTION	31	=head1 DESCRIPTION
52		32
53	This module implements asynchronous I/O using whatever means your	33	This module implements asynchronous I/O using whatever means your
54	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>).
55		36
56	Asynchronous means that operations that can normally block your program	37	Asynchronous means that operations that can normally block your program
57	(e.g. reading from disk) will be done asynchronously: the operation	38	(e.g. reading from disk) will be done asynchronously: the operation
58	will still block, but you can do something else in the meantime. This	39	will still block, but you can do something else in the meantime. This
59	is extremely useful for programs that need to stay interactive even	40	is extremely useful for programs that need to stay interactive even
…		…
61	etc.), but can also be used to easily do operations in parallel that are	42	etc.), but can also be used to easily do operations in parallel that are
62	normally done sequentially, e.g. stat'ing many files, which is much faster	43	normally done sequentially, e.g. stat'ing many files, which is much faster
63	on a RAID volume or over NFS when you do a number of stat operations	44	on a RAID volume or over NFS when you do a number of stat operations
64	concurrently.	45	concurrently.
65		46
66	While this works on all types of file descriptors (for example sockets),	47	While most of this works on all types of file descriptors (for
67	using these functions on file descriptors that support nonblocking	48	example sockets), using these functions on file descriptors that
68	operation (again, sockets, pipes etc.) is very inefficient. Use an event	49	support nonblocking operation (again, sockets, pipes etc.) is
69	loop for that (such as the L<Event|Event> module): IO::AIO will naturally	50	very inefficient. Use an event loop for that (such as the L<EV>
70	fit into such an event loop itself.	51	module): IO::AIO will naturally fit into such an event loop itself.
71		52
72	In this version, a number of threads are started that execute your	53	In this version, a number of threads are started that execute your
73	requests and signal their completion. You don't need thread support	54	requests and signal their completion. You don't need thread support
74	in perl, and the threads created by this module will not be visible	55	in perl, and the threads created by this module will not be visible
75	to perl. In the future, this module might make use of the native aio	56	to perl. In the future, this module might make use of the native aio
…		…
77	not well-supported or restricted (GNU/Linux doesn't allow them on normal	58	not well-supported or restricted (GNU/Linux doesn't allow them on normal
78	files currently, for example), and they would only support aio_read and	59	files currently, for example), and they would only support aio_read and
79	aio_write, so the remaining functionality would have to be implemented	60	aio_write, so the remaining functionality would have to be implemented
80	using threads anyway.	61	using threads anyway.
81		62
82	Although the module will work with in the presence of other (Perl-)	63	Although the module will work in the presence of other (Perl-) threads,
83	threads, it is currently not reentrant in any way, so use appropriate	64	it is currently not reentrant in any way, so use appropriate locking
84	locking yourself, always call C<poll_cb> from within the same thread, or	65	yourself, always call C<poll_cb> from within the same thread, or never
85	never call C<poll_cb> (or other 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 EV;
		74	use IO::AIO;
		75
		76	# register the IO::AIO callback with EV
		77	my $aio_w = EV::io IO::AIO::poll_fileno, EV::READ, \&IO::AIO::poll_cb;
		78
		79	# queue the request to open /etc/passwd
		80	aio_open "/etc/passwd", IO::AIO::O_RDONLY, 0, sub {
		81	my $fh = shift
		82	or die "error while opening: $!";
		83
		84	# stat'ing filehandles is generally non-blocking
		85	my $size = -s $fh;
		86
		87	# queue a request to read the file
		88	my $contents;
		89	aio_read $fh, 0, $size, $contents, 0, sub {
		90	$_[0] == $size
		91	or die "short read: $!";
		92
		93	close $fh;
		94
		95	# file contents now in $contents
		96	print $contents;
		97
		98	# exit event loop and program
		99	EV::unloop;
		100	};
		101	};
		102
		103	# possibly queue up other requests, or open GUI windows,
		104	# check for sockets etc. etc.
		105
		106	# process events as long as there are some:
		107	EV::loop;
86		108
87	=head1 REQUEST ANATOMY AND LIFETIME	109	=head1 REQUEST ANATOMY AND LIFETIME
88		110
89	Every C<aio_*> function creates a request. which is a C data structure not	111	Every C<aio_*> function creates a request. which is a C data structure not
90	directly visible to Perl.	112	directly visible to Perl.
…		…
132	Request has reached the end of its lifetime and holds no resources anymore	154	Request has reached the end of its lifetime and holds no resources anymore
133	(except possibly for the Perl object, but its connection to the actual	155	(except possibly for the Perl object, but its connection to the actual
134	aio request is severed and calling its methods will either do nothing or	156	aio request is severed and calling its methods will either do nothing or
135	result in a runtime error).	157	result in a runtime error).
136		158
		159	=back
		160
137	=cut	161	=cut
138		162
139	package IO::AIO;	163	package IO::AIO;
140		164
141	no warnings;	165	use Carp ();
142	use strict 'vars';	166
		167	use common::sense;
143		168
144	use base 'Exporter';	169	use base 'Exporter';
145		170
146	BEGIN {	171	BEGIN {
147	our $VERSION = '2.0';	172	our $VERSION = 4.31;
148		173
149	our @AIO_REQ = qw(aio_sendfile aio_read aio_write aio_open aio_close aio_stat	174	our @AIO_REQ = qw(aio_sendfile aio_seek aio_read aio_write aio_open aio_close
150	aio_lstat aio_unlink aio_rmdir aio_readdir aio_scandir aio_symlink	175	aio_stat aio_lstat aio_unlink aio_rmdir aio_readdir aio_readdirx
151	aio_fsync aio_fdatasync aio_readahead aio_rename aio_link aio_move	176	aio_scandir aio_symlink aio_readlink aio_realpath aio_sync
152	aio_copy aio_group aio_nop aio_mknod);	177	aio_fsync aio_syncfs aio_fdatasync aio_sync_file_range aio_allocate
		178	aio_pathsync aio_readahead aio_fiemap
		179	aio_rename aio_link aio_move aio_copy aio_group
		180	aio_nop aio_mknod aio_load aio_rmtree aio_mkdir aio_chown
		181	aio_chmod aio_utime aio_truncate
		182	aio_msync aio_mtouch aio_mlock aio_mlockall
		183	aio_statvfs
		184	aio_wd);
		185
153	our @EXPORT = (@AIO_REQ, qw(aioreq_pri aioreq_nice));	186	our @EXPORT = (@AIO_REQ, qw(aioreq_pri aioreq_nice));
154	our @EXPORT_OK = qw(poll_fileno poll_cb poll_wait flush	187	our @EXPORT_OK = qw(poll_fileno poll_cb poll_wait flush
155	min_parallel max_parallel nreqs nready npending);	188	min_parallel max_parallel max_idle idle_timeout
		189	nreqs nready npending nthreads
		190	max_poll_time max_poll_reqs
		191	sendfile fadvise madvise
		192	mmap munmap munlock munlockall);
		193
		194	push @AIO_REQ, qw(aio_busy); # not exported
156		195
157	@IO::AIO::GRP::ISA = 'IO::AIO::REQ';	196	@IO::AIO::GRP::ISA = 'IO::AIO::REQ';
158		197
159	require XSLoader;	198	require XSLoader;
160	XSLoader::load ("IO::AIO", $VERSION);	199	XSLoader::load ("IO::AIO", $VERSION);
161	}	200	}
162		201
163	=head1 FUNCTIONS	202	=head1 FUNCTIONS
164		203
165	=head2 AIO FUNCTIONS	204	=head2 QUICK OVERVIEW
		205
		206	This section simply lists the prototypes most of the functions for
		207	quick reference. See the following sections for function-by-function
		208	documentation.
		209
		210	aio_wd $pathname, $callback->($wd)
		211	aio_open $pathname, $flags, $mode, $callback->($fh)
		212	aio_close $fh, $callback->($status)
		213	aio_seek $fh,$offset,$whence, $callback->($offs)
		214	aio_read $fh,$offset,$length, $data,$dataoffset, $callback->($retval)
		215	aio_write $fh,$offset,$length, $data,$dataoffset, $callback->($retval)
		216	aio_sendfile $out_fh, $in_fh, $in_offset, $length, $callback->($retval)
		217	aio_readahead $fh,$offset,$length, $callback->($retval)
		218	aio_stat $fh_or_path, $callback->($status)
		219	aio_lstat $fh, $callback->($status)
		220	aio_statvfs $fh_or_path, $callback->($statvfs)
		221	aio_utime $fh_or_path, $atime, $mtime, $callback->($status)
		222	aio_chown $fh_or_path, $uid, $gid, $callback->($status)
		223	aio_chmod $fh_or_path, $mode, $callback->($status)
		224	aio_truncate $fh_or_path, $offset, $callback->($status)
		225	aio_allocate $fh, $mode, $offset, $len, $callback->($status)
		226	aio_fiemap $fh, $start, $length, $flags, $count, $cb->(\@extents)
		227	aio_unlink $pathname, $callback->($status)
		228	aio_mknod $pathname, $mode, $dev, $callback->($status)
		229	aio_link $srcpath, $dstpath, $callback->($status)
		230	aio_symlink $srcpath, $dstpath, $callback->($status)
		231	aio_readlink $pathname, $callback->($link)
		232	aio_realpath $pathname, $callback->($link)
		233	aio_rename $srcpath, $dstpath, $callback->($status)
		234	aio_mkdir $pathname, $mode, $callback->($status)
		235	aio_rmdir $pathname, $callback->($status)
		236	aio_readdir $pathname, $callback->($entries)
		237	aio_readdirx $pathname, $flags, $callback->($entries, $flags)
		238	IO::AIO::READDIR_DENTS IO::AIO::READDIR_DIRS_FIRST
		239	IO::AIO::READDIR_STAT_ORDER IO::AIO::READDIR_FOUND_UNKNOWN
		240	aio_scandir $pathname, $maxreq, $callback->($dirs, $nondirs)
		241	aio_load $pathname, $data, $callback->($status)
		242	aio_copy $srcpath, $dstpath, $callback->($status)
		243	aio_move $srcpath, $dstpath, $callback->($status)
		244	aio_rmtree $pathname, $callback->($status)
		245	aio_sync $callback->($status)
		246	aio_syncfs $fh, $callback->($status)
		247	aio_fsync $fh, $callback->($status)
		248	aio_fdatasync $fh, $callback->($status)
		249	aio_sync_file_range $fh, $offset, $nbytes, $flags, $callback->($status)
		250	aio_pathsync $pathname, $callback->($status)
		251	aio_msync $scalar, $offset = 0, $length = undef, flags = 0, $callback->($status)
		252	aio_mtouch $scalar, $offset = 0, $length = undef, flags = 0, $callback->($status)
		253	aio_mlock $scalar, $offset = 0, $length = undef, $callback->($status)
		254	aio_mlockall $flags, $callback->($status)
		255	aio_group $callback->(...)
		256	aio_nop $callback->()
		257
		258	$prev_pri = aioreq_pri [$pri]
		259	aioreq_nice $pri_adjust
		260
		261	IO::AIO::poll_wait
		262	IO::AIO::poll_cb
		263	IO::AIO::poll
		264	IO::AIO::flush
		265	IO::AIO::max_poll_reqs $nreqs
		266	IO::AIO::max_poll_time $seconds
		267	IO::AIO::min_parallel $nthreads
		268	IO::AIO::max_parallel $nthreads
		269	IO::AIO::max_idle $nthreads
		270	IO::AIO::idle_timeout $seconds
		271	IO::AIO::max_outstanding $maxreqs
		272	IO::AIO::nreqs
		273	IO::AIO::nready
		274	IO::AIO::npending
		275
		276	IO::AIO::sendfile $ofh, $ifh, $offset, $count
		277	IO::AIO::fadvise $fh, $offset, $len, $advice
		278	IO::AIO::mmap $scalar, $length, $prot, $flags[, $fh[, $offset]]
		279	IO::AIO::munmap $scalar
		280	IO::AIO::madvise $scalar, $offset, $length, $advice
		281	IO::AIO::mprotect $scalar, $offset, $length, $protect
		282	IO::AIO::munlock $scalar, $offset = 0, $length = undef
		283	IO::AIO::munlockall
		284
		285	=head2 API NOTES
166		286
167	All the C<aio_*> calls are more or less thin wrappers around the syscall	287	All the C<aio_*> calls are more or less thin wrappers around the syscall
168	with the same name (sans C<aio_>). The arguments are similar or identical,	288	with the same name (sans C<aio_>). The arguments are similar or identical,
169	and they all accept an additional (and optional) C<$callback> argument	289	and they all accept an additional (and optional) C<$callback> argument
170	which must be a code reference. This code reference will get called with	290	which must be a code reference. This code reference will be called after
171	the syscall return code (e.g. most syscalls return C<-1> on error, unlike	291	the syscall has been executed in an asynchronous fashion. The results
172	perl, which usually delivers "false") as it's sole argument when the given	292	of the request will be passed as arguments to the callback (and, if an
173	syscall has been executed asynchronously.	293	error occured, in C<$!>) - for most requests the syscall return code (e.g.
		294	most syscalls return C<-1> on error, unlike perl, which usually delivers
		295	"false").
		296
		297	Some requests (such as C<aio_readdir>) pass the actual results and
		298	communicate failures by passing C<undef>.
174		299
175	All functions expecting a filehandle keep a copy of the filehandle	300	All functions expecting a filehandle keep a copy of the filehandle
176	internally until the request has finished.	301	internally until the request has finished.
177		302
178	All requests return objects of type L<IO::AIO::REQ> that allow further	303	All functions return request objects of type L<IO::AIO::REQ> that allow
179	manipulation of those requests while they are in-flight.	304	further manipulation of those requests while they are in-flight.
180		305
181	The pathnames you pass to these routines I<must> be absolute and	306	The pathnames you pass to these routines I<should> be absolute. The
182	encoded in byte form. The reason for the former is that at the time the	307	reason for this is that at the time the request is being executed, the
183	request is being executed, the current working directory could have	308	current working directory could have changed. Alternatively, you can
184	changed. Alternatively, you can make sure that you never change the	309	make sure that you never change the current working directory anywhere
185	current working directory.	310	in the program and then use relative paths. You can also take advantage
		311	of IO::AIOs working directory abstraction, that lets you specify paths
		312	relative to some previously-opened "working directory object" - see the
		313	description of the C<IO::AIO::WD> class later in this document.
186		314
187	To encode pathnames to byte form, either make sure you either: a)	315	To encode pathnames as octets, either make sure you either: a) always pass
188	always pass in filenames you got from outside (command line, readdir	316	in filenames you got from outside (command line, readdir etc.) without
189	etc.), b) are ASCII or ISO 8859-1, c) use the Encode module and encode	317	tinkering, b) are in your native filesystem encoding, c) use the Encode
190	your pathnames to the locale (or other) encoding in effect in the user	318	module and encode your pathnames to the locale (or other) encoding in
191	environment, d) use Glib::filename_from_unicode on unicode filenames or e)	319	effect in the user environment, d) use Glib::filename_from_unicode on
192	use something else.	320	unicode filenames or e) use something else to ensure your scalar has the
		321	correct contents.
		322
		323	This works, btw. independent of the internal UTF-8 bit, which IO::AIO
		324	handles correctly whether it is set or not.
		325
		326	=head2 AIO REQUEST FUNCTIONS
193		327
194	=over 4	328	=over 4
195		329
196	=item $prev_pri = aioreq_pri [$pri]	330	=item $prev_pri = aioreq_pri [$pri]
197		331
…		…
217	aio_read $_[0], ..., sub {	351	aio_read $_[0], ..., sub {
218	...	352	...
219	};	353	};
220	};	354	};
221		355
		356
222	=item aioreq_nice $pri_adjust	357	=item aioreq_nice $pri_adjust
223		358
224	Similar to C<aioreq_pri>, but subtracts the given value from the current	359	Similar to C<aioreq_pri>, but subtracts the given value from the current
225	priority, so effects are cumulative.	360	priority, so the effect is cumulative.
		361
226		362
227	=item aio_open $pathname, $flags, $mode, $callback->($fh)	363	=item aio_open $pathname, $flags, $mode, $callback->($fh)
228		364
229	Asynchronously open or create a file and call the callback with a newly	365	Asynchronously open or create a file and call the callback with a newly
230	created filehandle for the file.	366	created filehandle for the file (or C<undef> in case of an error).
231		367
232	The pathname passed to C<aio_open> must be absolute. See API NOTES, above,	368	The pathname passed to C<aio_open> must be absolute. See API NOTES, above,
233	for an explanation.	369	for an explanation.
234		370
235	The C<$flags> argument is a bitmask. See the C<Fcntl> module for a	371	The C<$flags> argument is a bitmask. See the C<Fcntl> module for a
236	list. They are the same as used by C<sysopen>.	372	list. They are the same as used by C<sysopen>.
237		373
238	Likewise, C<$mode> specifies the mode of the newly created file, if it	374	Likewise, C<$mode> specifies the mode of the newly created file, if it
239	didn't exist and C<O_CREAT> has been given, just like perl's C<sysopen>,	375	didn't exist and C<O_CREAT> has been given, just like perl's C<sysopen>,
240	except that it is mandatory (i.e. use C<0> if you don't create new files,	376	except that it is mandatory (i.e. use C<0> if you don't create new files,
241	and C<0666> or C<0777> if you do).	377	and C<0666> or C<0777> if you do). Note that the C<$mode> will be modified
		378	by the umask in effect then the request is being executed, so better never
		379	change the umask.
242		380
243	Example:	381	Example:
244		382
245	aio_open "/etc/passwd", O_RDONLY, 0, sub {	383	aio_open "/etc/passwd", IO::AIO::O_RDONLY, 0, sub {
246	if ($_[0]) {	384	if ($_[0]) {
247	print "open successful, fh is $_[0]\n";	385	print "open successful, fh is $_[0]\n";
248	...	386	...
249	} else {	387	} else {
250	die "open failed: $!\n";	388	die "open failed: $!\n";
251	}	389	}
252	};	390	};
253		391
		392	In addition to all the common open modes/flags (C<O_RDONLY>, C<O_WRONLY>,
		393	C<O_RDWR>, C<O_CREAT>, C<O_TRUNC>, C<O_EXCL> and C<O_APPEND>), the
		394	following POSIX and non-POSIX constants are available (missing ones on
		395	your system are, as usual, C<0>):
		396
		397	C<O_ASYNC>, C<O_DIRECT>, C<O_NOATIME>, C<O_CLOEXEC>, C<O_NOCTTY>, C<O_NOFOLLOW>,
		398	C<O_NONBLOCK>, C<O_EXEC>, C<O_SEARCH>, C<O_DIRECTORY>, C<O_DSYNC>,
		399	C<O_RSYNC>, C<O_SYNC> and C<O_TTY_INIT>.
		400
		401
254	=item aio_close $fh, $callback->($status)	402	=item aio_close $fh, $callback->($status)
255		403
256	Asynchronously close a file and call the callback with the result	404	Asynchronously close a file and call the callback with the result
257	code. I<WARNING:> although accepted, you should not pass in a perl	405	code.
258	filehandle here, as perl will likely close the file descriptor another
259	time when the filehandle is destroyed. Normally, you can safely call perls
260	C<close> or just let filehandles go out of scope.
261		406
262	This is supposed to be a bug in the API, so that might change. It's	407	Unfortunately, you can't do this to perl. Perl I<insists> very strongly on
263	therefore best to avoid this function.	408	closing the file descriptor associated with the filehandle itself.
		409
		410	Therefore, C<aio_close> will not close the filehandle - instead it will
		411	use dup2 to overwrite the file descriptor with the write-end of a pipe
		412	(the pipe fd will be created on demand and will be cached).
		413
		414	Or in other words: the file descriptor will be closed, but it will not be
		415	free for reuse until the perl filehandle is closed.
		416
		417	=cut
		418
		419	=item aio_seek $fh, $offset, $whence, $callback->($offs)
		420
		421	Seeks the filehandle to the new C<$offset>, similarly to perl's
		422	C<sysseek>. The C<$whence> can use the traditional values (C<0> for
		423	C<IO::AIO::SEEK_SET>, C<1> for C<IO::AIO::SEEK_CUR> or C<2> for
		424	C<IO::AIO::SEEK_END>).
		425
		426	The resulting absolute offset will be passed to the callback, or C<-1> in
		427	case of an error.
		428
		429	In theory, the C<$whence> constants could be different than the
		430	corresponding values from L<Fcntl>, but perl guarantees they are the same,
		431	so don't panic.
		432
		433	As a GNU/Linux (and maybe Solaris) extension, also the constants
		434	C<IO::AIO::SEEK_DATA> and C<IO::AIO::SEEK_HOLE> are available, if they
		435	could be found. No guarantees about suitability for use in C<aio_seek> or
		436	Perl's C<sysseek> can be made though, although I would naively assume they
		437	"just work".
264		438
265	=item aio_read $fh,$offset,$length, $data,$dataoffset, $callback->($retval)	439	=item aio_read $fh,$offset,$length, $data,$dataoffset, $callback->($retval)
266		440
267	=item aio_write $fh,$offset,$length, $data,$dataoffset, $callback->($retval)	441	=item aio_write $fh,$offset,$length, $data,$dataoffset, $callback->($retval)
268		442
269	Reads or writes C<length> bytes from the specified C<fh> and C<offset>	443	Reads or writes C<$length> bytes from or to the specified C<$fh> and
270	into the scalar given by C<data> and offset C<dataoffset> and calls the	444	C<$offset> into the scalar given by C<$data> and offset C<$dataoffset>
271	callback without the actual number of bytes read (or -1 on error, just	445	and calls the callback without the actual number of bytes read (or -1 on
272	like the syscall).	446	error, just like the syscall).
		447
		448	C<aio_read> will, like C<sysread>, shrink or grow the C<$data> scalar to
		449	offset plus the actual number of bytes read.
		450
		451	If C<$offset> is undefined, then the current file descriptor offset will
		452	be used (and updated), otherwise the file descriptor offset will not be
		453	changed by these calls.
		454
		455	If C<$length> is undefined in C<aio_write>, use the remaining length of
		456	C<$data>.
		457
		458	If C<$dataoffset> is less than zero, it will be counted from the end of
		459	C<$data>.
273		460
274	The C<$data> scalar I<MUST NOT> be modified in any way while the request	461	The C<$data> scalar I<MUST NOT> be modified in any way while the request
275	is outstanding. Modifying it can result in segfaults or WW3 (if the	462	is outstanding. Modifying it can result in segfaults or World War III (if
276	necessary/optional hardware is installed).	463	the necessary/optional hardware is installed).
277		464
278	Example: Read 15 bytes at offset 7 into scalar C<$buffer>, starting at	465	Example: Read 15 bytes at offset 7 into scalar C<$buffer>, starting at
279	offset C<0> within the scalar:	466	offset C<0> within the scalar:
280		467
281	aio_read $fh, 7, 15, $buffer, 0, sub {	468	aio_read $fh, 7, 15, $buffer, 0, sub {
282	$_[0] > 0 or die "read error: $!";	469	$_[0] > 0 or die "read error: $!";
283	print "read $_[0] bytes: <$buffer>\n";	470	print "read $_[0] bytes: <$buffer>\n";
284	};	471	};
285		472
		473
286	=item aio_sendfile $out_fh, $in_fh, $in_offset, $length, $callback->($retval)	474	=item aio_sendfile $out_fh, $in_fh, $in_offset, $length, $callback->($retval)
287		475
288	Tries to copy C<$length> bytes from C<$in_fh> to C<$out_fh>. It starts	476	Tries to copy C<$length> bytes from C<$in_fh> to C<$out_fh>. It starts
289	reading at byte offset C<$in_offset>, and starts writing at the current	477	reading at byte offset C<$in_offset>, and starts writing at the current
290	file offset of C<$out_fh>. Because of that, it is not safe to issue more	478	file offset of C<$out_fh>. Because of that, it is not safe to issue more
291	than one C<aio_sendfile> per C<$out_fh>, as they will interfere with each	479	than one C<aio_sendfile> per C<$out_fh>, as they will interfere with each
292	other.	480	other. The same C<$in_fh> works fine though, as this function does not
		481	move or use the file offset of C<$in_fh>.
293		482
		483	Please note that C<aio_sendfile> can read more bytes from C<$in_fh> than
		484	are written, and there is no way to find out how many more bytes have been
		485	read from C<aio_sendfile> alone, as C<aio_sendfile> only provides the
		486	number of bytes written to C<$out_fh>. Only if the result value equals
		487	C<$length> one can assume that C<$length> bytes have been read.
		488
		489	Unlike with other C<aio_> functions, it makes a lot of sense to use
		490	C<aio_sendfile> on non-blocking sockets, as long as one end (typically
		491	the C<$in_fh>) is a file - the file I/O will then be asynchronous, while
		492	the socket I/O will be non-blocking. Note, however, that you can run
		493	into a trap where C<aio_sendfile> reads some data with readahead, then
		494	fails to write all data, and when the socket is ready the next time, the
		495	data in the cache is already lost, forcing C<aio_sendfile> to again hit
		496	the disk. Explicit C<aio_read> + C<aio_write> let's you better control
		497	resource usage.
		498
294	This call tries to make use of a native C<sendfile> syscall to provide	499	This call tries to make use of a native C<sendfile>-like syscall to
295	zero-copy operation. For this to work, C<$out_fh> should refer to a	500	provide zero-copy operation. For this to work, C<$out_fh> should refer to
296	socket, and C<$in_fh> should refer to mmap'able file.	501	a socket, and C<$in_fh> should refer to an mmap'able file.
297		502
298	If the native sendfile call fails or is not implemented, it will be	503	If a native sendfile cannot be found or it fails with C<ENOSYS>,
299	emulated, so you can call C<aio_sendfile> on any type of filehandle	504	C<EINVAL>, C<ENOTSUP>, C<EOPNOTSUPP>, C<EAFNOSUPPORT>, C<EPROTOTYPE> or
		505	C<ENOTSOCK>, it will be emulated, so you can call C<aio_sendfile> on any
300	regardless of the limitations of the operating system.	506	type of filehandle regardless of the limitations of the operating system.
301		507
302	Please note, however, that C<aio_sendfile> can read more bytes from	508	As native sendfile syscalls (as practically any non-POSIX interface hacked
303	C<$in_fh> than are written, and there is no way to find out how many	509	together in a hurry to improve benchmark numbers) tend to be rather buggy
304	bytes have been read from C<aio_sendfile> alone, as C<aio_sendfile> only	510	on many systems, this implementation tries to work around some known bugs
305	provides the number of bytes written to C<$out_fh>. Only if the result	511	in Linux and FreeBSD kernels (probably others, too), but that might fail,
306	value equals C<$length> one can assume that C<$length> bytes have been	512	so you really really should check the return value of C<aio_sendfile> -
307	read.	513	fewre bytes than expected might have been transferred.
		514
308		515
309	=item aio_readahead $fh,$offset,$length, $callback->($retval)	516	=item aio_readahead $fh,$offset,$length, $callback->($retval)
310		517
311	C<aio_readahead> populates the page cache with data from a file so that	518	C<aio_readahead> populates the page cache with data from a file so that
312	subsequent reads from that file will not block on disk I/O. The C<$offset>	519	subsequent reads from that file will not block on disk I/O. The C<$offset>
…		…
318	file. The current file offset of the file is left unchanged.	525	file. The current file offset of the file is left unchanged.
319		526
320	If that syscall doesn't exist (likely if your OS isn't Linux) it will be	527	If that syscall doesn't exist (likely if your OS isn't Linux) it will be
321	emulated by simply reading the data, which would have a similar effect.	528	emulated by simply reading the data, which would have a similar effect.
322		529
		530
323	=item aio_stat $fh_or_path, $callback->($status)	531	=item aio_stat $fh_or_path, $callback->($status)
324		532
325	=item aio_lstat $fh, $callback->($status)	533	=item aio_lstat $fh, $callback->($status)
326		534
327	Works like perl's C<stat> or C<lstat> in void context. The callback will	535	Works like perl's C<stat> or C<lstat> in void context. The callback will
…		…
332	for an explanation.	540	for an explanation.
333		541
334	Currently, the stats are always 64-bit-stats, i.e. instead of returning an	542	Currently, the stats are always 64-bit-stats, i.e. instead of returning an
335	error when stat'ing a large file, the results will be silently truncated	543	error when stat'ing a large file, the results will be silently truncated
336	unless perl itself is compiled with large file support.	544	unless perl itself is compiled with large file support.
		545
		546	To help interpret the mode and dev/rdev stat values, IO::AIO offers the
		547	following constants and functions (if not implemented, the constants will
		548	be C<0> and the functions will either C<croak> or fall back on traditional
		549	behaviour).
		550
		551	C<S_IFMT>, C<S_IFIFO>, C<S_IFCHR>, C<S_IFBLK>, C<S_IFLNK>, C<S_IFREG>,
		552	C<S_IFDIR>, C<S_IFWHT>, C<S_IFSOCK>, C<IO::AIO::major $dev_t>,
		553	C<IO::AIO::minor $dev_t>, C<IO::AIO::makedev $major, $minor>.
337		554
338	Example: Print the length of F</etc/passwd>:	555	Example: Print the length of F</etc/passwd>:
339		556
340	aio_stat "/etc/passwd", sub {	557	aio_stat "/etc/passwd", sub {
341	$_[0] and die "stat failed: $!";	558	$_[0] and die "stat failed: $!";
342	print "size is ", -s _, "\n";	559	print "size is ", -s _, "\n";
343	};	560	};
344		561
		562
		563	=item aio_statvfs $fh_or_path, $callback->($statvfs)
		564
		565	Works like the POSIX C<statvfs> or C<fstatvfs> syscalls, depending on
		566	whether a file handle or path was passed.
		567
		568	On success, the callback is passed a hash reference with the following
		569	members: C<bsize>, C<frsize>, C<blocks>, C<bfree>, C<bavail>, C<files>,
		570	C<ffree>, C<favail>, C<fsid>, C<flag> and C<namemax>. On failure, C<undef>
		571	is passed.
		572
		573	The following POSIX IO::AIO::ST_* constants are defined: C<ST_RDONLY> and
		574	C<ST_NOSUID>.
		575
		576	The following non-POSIX IO::AIO::ST_* flag masks are defined to
		577	their correct value when available, or to C<0> on systems that do
		578	not support them: C<ST_NODEV>, C<ST_NOEXEC>, C<ST_SYNCHRONOUS>,
		579	C<ST_MANDLOCK>, C<ST_WRITE>, C<ST_APPEND>, C<ST_IMMUTABLE>, C<ST_NOATIME>,
		580	C<ST_NODIRATIME> and C<ST_RELATIME>.
		581
		582	Example: stat C</wd> and dump out the data if successful.
		583
		584	aio_statvfs "/wd", sub {
		585	my $f = $_[0]
		586	or die "statvfs: $!";
		587
		588	use Data::Dumper;
		589	say Dumper $f;
		590	};
		591
		592	# result:
		593	{
		594	bsize => 1024,
		595	bfree => 4333064312,
		596	blocks => 10253828096,
		597	files => 2050765568,
		598	flag => 4096,
		599	favail => 2042092649,
		600	bavail => 4333064312,
		601	ffree => 2042092649,
		602	namemax => 255,
		603	frsize => 1024,
		604	fsid => 1810
		605	}
		606
		607	Here is a (likely partial - send me updates!) list of fsid values used by
		608	Linux - it is safe to hardcode these when C<$^O> is C<linux>:
		609
		610	0x0000adf5 adfs
		611	0x0000adff affs
		612	0x5346414f afs
		613	0x09041934 anon-inode filesystem
		614	0x00000187 autofs
		615	0x42465331 befs
		616	0x1badface bfs
		617	0x42494e4d binfmt_misc
		618	0x9123683e btrfs
		619	0x0027e0eb cgroupfs
		620	0xff534d42 cifs
		621	0x73757245 coda
		622	0x012ff7b7 coh
		623	0x28cd3d45 cramfs
		624	0x453dcd28 cramfs-wend (wrong endianness)
		625	0x64626720 debugfs
		626	0x00001373 devfs
		627	0x00001cd1 devpts
		628	0x0000f15f ecryptfs
		629	0x00414a53 efs
		630	0x0000137d ext
		631	0x0000ef53 ext2/ext3
		632	0x0000ef51 ext2
		633	0x00004006 fat
		634	0x65735546 fuseblk
		635	0x65735543 fusectl
		636	0x0bad1dea futexfs
		637	0x01161970 gfs2
		638	0x47504653 gpfs
		639	0x00004244 hfs
		640	0xf995e849 hpfs
		641	0x958458f6 hugetlbfs
		642	0x2bad1dea inotifyfs
		643	0x00009660 isofs
		644	0x000072b6 jffs2
		645	0x3153464a jfs
		646	0x6b414653 k-afs
		647	0x0bd00bd0 lustre
		648	0x0000137f minix
		649	0x0000138f minix 30 char names
		650	0x00002468 minix v2
		651	0x00002478 minix v2 30 char names
		652	0x00004d5a minix v3
		653	0x19800202 mqueue
		654	0x00004d44 msdos
		655	0x0000564c novell
		656	0x00006969 nfs
		657	0x6e667364 nfsd
		658	0x00003434 nilfs
		659	0x5346544e ntfs
		660	0x00009fa1 openprom
		661	0x7461636F ocfs2
		662	0x00009fa0 proc
		663	0x6165676c pstorefs
		664	0x0000002f qnx4
		665	0x858458f6 ramfs
		666	0x52654973 reiserfs
		667	0x00007275 romfs
		668	0x67596969 rpc_pipefs
		669	0x73636673 securityfs
		670	0xf97cff8c selinux
		671	0x0000517b smb
		672	0x534f434b sockfs
		673	0x73717368 squashfs
		674	0x62656572 sysfs
		675	0x012ff7b6 sysv2
		676	0x012ff7b5 sysv4
		677	0x01021994 tmpfs
		678	0x15013346 udf
		679	0x00011954 ufs
		680	0x54190100 ufs byteswapped
		681	0x00009fa2 usbdevfs
		682	0x01021997 v9fs
		683	0xa501fcf5 vxfs
		684	0xabba1974 xenfs
		685	0x012ff7b4 xenix
		686	0x58465342 xfs
		687	0x012fd16d xia
		688
		689	=item aio_utime $fh_or_path, $atime, $mtime, $callback->($status)
		690
		691	Works like perl's C<utime> function (including the special case of $atime
		692	and $mtime being undef). Fractional times are supported if the underlying
		693	syscalls support them.
		694
		695	When called with a pathname, uses utimes(2) if available, otherwise
		696	utime(2). If called on a file descriptor, uses futimes(2) if available,
		697	otherwise returns ENOSYS, so this is not portable.
		698
		699	Examples:
		700
		701	# set atime and mtime to current time (basically touch(1)):
		702	aio_utime "path", undef, undef;
		703	# set atime to current time and mtime to beginning of the epoch:
		704	aio_utime "path", time, undef; # undef==0
		705
		706
		707	=item aio_chown $fh_or_path, $uid, $gid, $callback->($status)
		708
		709	Works like perl's C<chown> function, except that C<undef> for either $uid
		710	or $gid is being interpreted as "do not change" (but -1 can also be used).
		711
		712	Examples:
		713
		714	# same as "chown root path" in the shell:
		715	aio_chown "path", 0, -1;
		716	# same as above:
		717	aio_chown "path", 0, undef;
		718
		719
		720	=item aio_truncate $fh_or_path, $offset, $callback->($status)
		721
		722	Works like truncate(2) or ftruncate(2).
		723
		724
		725	=item aio_allocate $fh, $mode, $offset, $len, $callback->($status)
		726
		727	Allocates or freed disk space according to the C<$mode> argument. See the
		728	linux C<fallocate> docuemntation for details.
		729
		730	C<$mode> can currently be C<0> or C<IO::AIO::FALLOC_FL_KEEP_SIZE>
		731	to allocate space, or C<IO::AIO::FALLOC_FL_PUNCH_HOLE |
		732	IO::AIO::FALLOC_FL_KEEP_SIZE>, to deallocate a file range.
		733
		734	The file system block size used by C<fallocate> is presumably the
		735	C<f_bsize> returned by C<statvfs>.
		736
		737	If C<fallocate> isn't available or cannot be emulated (currently no
		738	emulation will be attempted), passes C<-1> and sets C<$!> to C<ENOSYS>.
		739
		740
		741	=item aio_chmod $fh_or_path, $mode, $callback->($status)
		742
		743	Works like perl's C<chmod> function.
		744
		745
345	=item aio_unlink $pathname, $callback->($status)	746	=item aio_unlink $pathname, $callback->($status)
346		747
347	Asynchronously unlink (delete) a file and call the callback with the	748	Asynchronously unlink (delete) a file and call the callback with the
348	result code.	749	result code.
349		750
		751
350	=item aio_mknod $path, $mode, $dev, $callback->($status)	752	=item aio_mknod $pathname, $mode, $dev, $callback->($status)
		753
		754	[EXPERIMENTAL]
351		755
352	Asynchronously create a device node (or fifo). See mknod(2).	756	Asynchronously create a device node (or fifo). See mknod(2).
353		757
354	The only portable (POSIX) way of calling this function is:	758	The only (POSIX-) portable way of calling this function is:
355		759
356	aio_mknod $path, IO::AIO::S_IFIFO | $mode, 0, sub { ...	760	aio_mknod $pathname, IO::AIO::S_IFIFO | $mode, 0, sub { ...
		761
		762	See C<aio_stat> for info about some potentially helpful extra constants
		763	and functions.
357		764
358	=item aio_link $srcpath, $dstpath, $callback->($status)	765	=item aio_link $srcpath, $dstpath, $callback->($status)
359		766
360	Asynchronously create a new link to the existing object at C<$srcpath> at	767	Asynchronously create a new link to the existing object at C<$srcpath> at
361	the path C<$dstpath> and call the callback with the result code.	768	the path C<$dstpath> and call the callback with the result code.
362		769
		770
363	=item aio_symlink $srcpath, $dstpath, $callback->($status)	771	=item aio_symlink $srcpath, $dstpath, $callback->($status)
364		772
365	Asynchronously create a new symbolic link to the existing object at C<$srcpath> at	773	Asynchronously create a new symbolic link to the existing object at C<$srcpath> at
366	the path C<$dstpath> and call the callback with the result code.	774	the path C<$dstpath> and call the callback with the result code.
367		775
		776
		777	=item aio_readlink $pathname, $callback->($link)
		778
		779	Asynchronously read the symlink specified by C<$path> and pass it to
		780	the callback. If an error occurs, nothing or undef gets passed to the
		781	callback.
		782
		783
		784	=item aio_realpath $pathname, $callback->($path)
		785
		786	Asynchronously make the path absolute and resolve any symlinks in
		787	C<$path>. The resulting path only consists of directories (same as
		788	L<Cwd::realpath>).
		789
		790	This request can be used to get the absolute path of the current working
		791	directory by passing it a path of F<.> (a single dot).
		792
		793
368	=item aio_rename $srcpath, $dstpath, $callback->($status)	794	=item aio_rename $srcpath, $dstpath, $callback->($status)
369		795
370	Asynchronously rename the object at C<$srcpath> to C<$dstpath>, just as	796	Asynchronously rename the object at C<$srcpath> to C<$dstpath>, just as
371	rename(2) and call the callback with the result code.	797	rename(2) and call the callback with the result code.
372		798
		799	On systems that support the AIO::WD working directory abstraction
		800	natively, the case C<[$wd, "."]> as C<$srcpath> is specialcased - instead
		801	of failing, C<rename> is called on the absolute path of C<$wd>.
		802
		803
		804	=item aio_mkdir $pathname, $mode, $callback->($status)
		805
		806	Asynchronously mkdir (create) a directory and call the callback with
		807	the result code. C<$mode> will be modified by the umask at the time the
		808	request is executed, so do not change your umask.
		809
		810
373	=item aio_rmdir $pathname, $callback->($status)	811	=item aio_rmdir $pathname, $callback->($status)
374		812
375	Asynchronously rmdir (delete) a directory and call the callback with the	813	Asynchronously rmdir (delete) a directory and call the callback with the
376	result code.	814	result code.
		815
		816	On systems that support the AIO::WD working directory abstraction
		817	natively, the case C<[$wd, "."]> is specialcased - instead of failing,
		818	C<rmdir> is called on the absolute path of C<$wd>.
		819
377		820
378	=item aio_readdir $pathname, $callback->($entries)	821	=item aio_readdir $pathname, $callback->($entries)
379		822
380	Unlike the POSIX call of the same name, C<aio_readdir> reads an entire	823	Unlike the POSIX call of the same name, C<aio_readdir> reads an entire
381	directory (i.e. opendir + readdir + closedir). The entries will not be	824	directory (i.e. opendir + readdir + closedir). The entries will not be
382	sorted, and will B<NOT> include the C<.> and C<..> entries.	825	sorted, and will B<NOT> include the C<.> and C<..> entries.
383		826
384	The callback a single argument which is either C<undef> or an array-ref	827	The callback is passed a single argument which is either C<undef> or an
385	with the filenames.	828	array-ref with the filenames.
		829
		830
		831	=item aio_readdirx $pathname, $flags, $callback->($entries, $flags)
		832
		833	Quite similar to C<aio_readdir>, but the C<$flags> argument allows one to
		834	tune behaviour and output format. In case of an error, C<$entries> will be
		835	C<undef>.
		836
		837	The flags are a combination of the following constants, ORed together (the
		838	flags will also be passed to the callback, possibly modified):
		839
		840	=over 4
		841
		842	=item IO::AIO::READDIR_DENTS
		843
		844	When this flag is off, then the callback gets an arrayref consisting of
		845	names only (as with C<aio_readdir>), otherwise it gets an arrayref with
		846	C<[$name, $type, $inode]> arrayrefs, each describing a single directory
		847	entry in more detail.
		848
		849	C<$name> is the name of the entry.
		850
		851	C<$type> is one of the C<IO::AIO::DT_xxx> constants:
		852
		853	C<IO::AIO::DT_UNKNOWN>, C<IO::AIO::DT_FIFO>, C<IO::AIO::DT_CHR>, C<IO::AIO::DT_DIR>,
		854	C<IO::AIO::DT_BLK>, C<IO::AIO::DT_REG>, C<IO::AIO::DT_LNK>, C<IO::AIO::DT_SOCK>,
		855	C<IO::AIO::DT_WHT>.
		856
		857	C<IO::AIO::DT_UNKNOWN> means just that: readdir does not know. If you need to
		858	know, you have to run stat yourself. Also, for speed reasons, the C<$type>
		859	scalars are read-only: you can not modify them.
		860
		861	C<$inode> is the inode number (which might not be exact on systems with 64
		862	bit inode numbers and 32 bit perls). This field has unspecified content on
		863	systems that do not deliver the inode information.
		864
		865	=item IO::AIO::READDIR_DIRS_FIRST
		866
		867	When this flag is set, then the names will be returned in an order where
		868	likely directories come first, in optimal stat order. This is useful when
		869	you need to quickly find directories, or you want to find all directories
		870	while avoiding to stat() each entry.
		871
		872	If the system returns type information in readdir, then this is used
		873	to find directories directly. Otherwise, likely directories are names
		874	beginning with ".", or otherwise names with no dots, of which names with
		875	short names are tried first.
		876
		877	=item IO::AIO::READDIR_STAT_ORDER
		878
		879	When this flag is set, then the names will be returned in an order
		880	suitable for stat()'ing each one. That is, when you plan to stat()
		881	all files in the given directory, then the returned order will likely
		882	be fastest.
		883
		884	If both this flag and C<IO::AIO::READDIR_DIRS_FIRST> are specified, then
		885	the likely dirs come first, resulting in a less optimal stat order.
		886
		887	=item IO::AIO::READDIR_FOUND_UNKNOWN
		888
		889	This flag should not be set when calling C<aio_readdirx>. Instead, it
		890	is being set by C<aio_readdirx>, when any of the C<$type>'s found were
		891	C<IO::AIO::DT_UNKNOWN>. The absence of this flag therefore indicates that all
		892	C<$type>'s are known, which can be used to speed up some algorithms.
		893
		894	=back
		895
		896
		897	=item aio_load $pathname, $data, $callback->($status)
		898
		899	This is a composite request that tries to fully load the given file into
		900	memory. Status is the same as with aio_read.
		901
		902	=cut
		903
		904	sub aio_load($$;$) {
		905	my ($path, undef, $cb) = @_;
		906	my $data = \$_[1];
		907
		908	my $pri = aioreq_pri;
		909	my $grp = aio_group $cb;
		910
		911	aioreq_pri $pri;
		912	add $grp aio_open $path, O_RDONLY, 0, sub {
		913	my $fh = shift
		914	or return $grp->result (-1);
		915
		916	aioreq_pri $pri;
		917	add $grp aio_read $fh, 0, (-s $fh), $$data, 0, sub {
		918	$grp->result ($_[0]);
		919	};
		920	};
		921
		922	$grp
		923	}
386		924
387	=item aio_copy $srcpath, $dstpath, $callback->($status)	925	=item aio_copy $srcpath, $dstpath, $callback->($status)
388		926
389	Try to copy the I<file> (directories not supported as either source or	927	Try to copy the I<file> (directories not supported as either source or
390	destination) from C<$srcpath> to C<$dstpath> and call the callback with	928	destination) from C<$srcpath> to C<$dstpath> and call the callback with
391	the C<0> (error) or C<-1> ok.	929	a status of C<0> (ok) or C<-1> (error, see C<$!>).
392		930
393	This is a composite request that it creates the destination file with	931	This is a composite request that creates the destination file with
394	mode 0200 and copies the contents of the source file into it using	932	mode 0200 and copies the contents of the source file into it using
395	C<aio_sendfile>, followed by restoring atime, mtime, access mode and	933	C<aio_sendfile>, followed by restoring atime, mtime, access mode and
396	uid/gid, in that order.	934	uid/gid, in that order.
397		935
398	If an error occurs, the partial destination file will be unlinked, if	936	If an error occurs, the partial destination file will be unlinked, if
…		…
408	my $grp = aio_group $cb;	946	my $grp = aio_group $cb;
409		947
410	aioreq_pri $pri;	948	aioreq_pri $pri;
411	add $grp aio_open $src, O_RDONLY, 0, sub {	949	add $grp aio_open $src, O_RDONLY, 0, sub {
412	if (my $src_fh = $_[0]) {	950	if (my $src_fh = $_[0]) {
413	my @stat = stat $src_fh;	951	my @stat = stat $src_fh; # hmm, might block over nfs?
414		952
415	aioreq_pri $pri;	953	aioreq_pri $pri;
416	add $grp aio_open $dst, O_CREAT | O_WRONLY | O_TRUNC, 0200, sub {	954	add $grp aio_open $dst, O_CREAT | O_WRONLY | O_TRUNC, 0200, sub {
417	if (my $dst_fh = $_[0]) {	955	if (my $dst_fh = $_[0]) {
418	aioreq_pri $pri;	956	aioreq_pri $pri;
419	add $grp aio_sendfile $dst_fh, $src_fh, 0, $stat[7], sub {	957	add $grp aio_sendfile $dst_fh, $src_fh, 0, $stat[7], sub {
420	if ($_[0] == $stat[7]) {	958	if ($_[0] == $stat[7]) {
421	$grp->result (0);	959	$grp->result (0);
422	close $src_fh;	960	close $src_fh;
423		961
424	# those should not normally block. should. should.	962	my $ch = sub {
		963	aioreq_pri $pri;
		964	add $grp aio_chmod $dst_fh, $stat[2] & 07777, sub {
		965	aioreq_pri $pri;
		966	add $grp aio_chown $dst_fh, $stat[4], $stat[5], sub {
		967	aioreq_pri $pri;
		968	add $grp aio_close $dst_fh;
		969	}
		970	};
		971	};
		972
		973	aioreq_pri $pri;
		974	add $grp aio_utime $dst_fh, $stat[8], $stat[9], sub {
		975	if ($_[0] < 0 && $! == ENOSYS) {
		976	aioreq_pri $pri;
425	utime $stat[8], $stat[9], $dst;	977	add $grp aio_utime $dst, $stat[8], $stat[9], $ch;
426	chmod $stat[2] & 07777, $dst_fh;	978	} else {
427	chown $stat[4], $stat[5], $dst_fh;	979	$ch->();
		980	}
428	close $dst_fh;	981	};
429	} else {	982	} else {
430	$grp->result (-1);	983	$grp->result (-1);
431	close $src_fh;	984	close $src_fh;
432	close $dst_fh;	985	close $dst_fh;
433		986
…		…
450		1003
451	=item aio_move $srcpath, $dstpath, $callback->($status)	1004	=item aio_move $srcpath, $dstpath, $callback->($status)
452		1005
453	Try to move the I<file> (directories not supported as either source or	1006	Try to move the I<file> (directories not supported as either source or
454	destination) from C<$srcpath> to C<$dstpath> and call the callback with	1007	destination) from C<$srcpath> to C<$dstpath> and call the callback with
455	the C<0> (error) or C<-1> ok.	1008	a status of C<0> (ok) or C<-1> (error, see C<$!>).
456		1009
457	This is a composite request that tries to rename(2) the file first. If	1010	This is a composite request that tries to rename(2) the file first; if
458	rename files with C<EXDEV>, it copies the file with C<aio_copy> and, if	1011	rename fails with C<EXDEV>, it copies the file with C<aio_copy> and, if
459	that is successful, unlinking the C<$srcpath>.	1012	that is successful, unlinks the C<$srcpath>.
460		1013
461	=cut	1014	=cut
462		1015
463	sub aio_move($$;$) {	1016	sub aio_move($$;$) {
464	my ($src, $dst, $cb) = @_;	1017	my ($src, $dst, $cb) = @_;
…		…
471	if ($_[0] && $! == EXDEV) {	1024	if ($_[0] && $! == EXDEV) {
472	aioreq_pri $pri;	1025	aioreq_pri $pri;
473	add $grp aio_copy $src, $dst, sub {	1026	add $grp aio_copy $src, $dst, sub {
474	$grp->result ($_[0]);	1027	$grp->result ($_[0]);
475		1028
476	if (!$_[0]) {	1029	unless ($_[0]) {
477	aioreq_pri $pri;	1030	aioreq_pri $pri;
478	add $grp aio_unlink $src;	1031	add $grp aio_unlink $src;
479	}	1032	}
480	};	1033	};
481	} else {	1034	} else {
…		…
484	};	1037	};
485		1038
486	$grp	1039	$grp
487	}	1040	}
488		1041
489	=item aio_scandir $path, $maxreq, $callback->($dirs, $nondirs)	1042	=item aio_scandir $pathname, $maxreq, $callback->($dirs, $nondirs)
490		1043
491	Scans a directory (similar to C<aio_readdir>) but additionally tries to	1044	Scans a directory (similar to C<aio_readdir>) but additionally tries to
492	efficiently separate the entries of directory C<$path> into two sets of	1045	efficiently separate the entries of directory C<$path> into two sets of
493	names, directories you can recurse into (directories), and ones you cannot	1046	names, directories you can recurse into (directories), and ones you cannot
494	recurse into (everything else, including symlinks to directories).	1047	recurse into (everything else, including symlinks to directories).
…		…
511		1064
512	Implementation notes.	1065	Implementation notes.
513		1066
514	The C<aio_readdir> cannot be avoided, but C<stat()>'ing every entry can.	1067	The C<aio_readdir> cannot be avoided, but C<stat()>'ing every entry can.
515		1068
		1069	If readdir returns file type information, then this is used directly to
		1070	find directories.
		1071
516	After reading the directory, the modification time, size etc. of the	1072	Otherwise, after reading the directory, the modification time, size etc.
517	directory before and after the readdir is checked, and if they match (and	1073	of the directory before and after the readdir is checked, and if they
518	isn't the current time), the link count will be used to decide how many	1074	match (and isn't the current time), the link count will be used to decide
519	entries are directories (if >= 2). Otherwise, no knowledge of the number	1075	how many entries are directories (if >= 2). Otherwise, no knowledge of the
520	of subdirectories will be assumed.	1076	number of subdirectories will be assumed.
521		1077
522	Then entries will be sorted into likely directories (everything without	1078	Then entries will be sorted into likely directories a non-initial dot
523	a non-initial dot currently) and likely non-directories (everything	1079	currently) and likely non-directories (see C<aio_readdirx>). Then every
524	else). Then every entry plus an appended C</.> will be C<stat>'ed,	1080	entry plus an appended C</.> will be C<stat>'ed, likely directories first,
525	likely directories first. If that succeeds, it assumes that the entry	1081	in order of their inode numbers. If that succeeds, it assumes that the
526	is a directory or a symlink to directory (which will be checked	1082	entry is a directory or a symlink to directory (which will be checked
527	seperately). This is often faster than stat'ing the entry itself because	1083	separately). This is often faster than stat'ing the entry itself because
528	filesystems might detect the type of the entry without reading the inode	1084	filesystems might detect the type of the entry without reading the inode
529	data (e.g. ext2fs filetype feature).	1085	data (e.g. ext2fs filetype feature), even on systems that cannot return
		1086	the filetype information on readdir.
530		1087
531	If the known number of directories (link count - 2) has been reached, the	1088	If the known number of directories (link count - 2) has been reached, the
532	rest of the entries is assumed to be non-directories.	1089	rest of the entries is assumed to be non-directories.
533		1090
534	This only works with certainty on POSIX (= UNIX) filesystems, which	1091	This only works with certainty on POSIX (= UNIX) filesystems, which
…		…
538	as those tend to return 0 or 1 as link counts, which disables the	1095	as those tend to return 0 or 1 as link counts, which disables the
539	directory counting heuristic.	1096	directory counting heuristic.
540		1097
541	=cut	1098	=cut
542		1099
543	sub aio_scandir($$$) {	1100	sub aio_scandir($$;$) {
544	my ($path, $maxreq, $cb) = @_;	1101	my ($path, $maxreq, $cb) = @_;
545		1102
546	my $pri = aioreq_pri;	1103	my $pri = aioreq_pri;
547		1104
548	my $grp = aio_group $cb;	1105	my $grp = aio_group $cb;
549		1106
550	$maxreq = 4 if $maxreq <= 0;	1107	$maxreq = 4 if $maxreq <= 0;
551		1108
552	# stat once	1109	# get a wd object
553	aioreq_pri $pri;	1110	aioreq_pri $pri;
554	add $grp aio_stat $path, sub {	1111	add $grp aio_wd $path, sub {
		1112	$_[0]
555	return $grp->result () if $_[0];	1113	or return $grp->result ();
556	my $now = time;
557	my $hash1 = join ":", (stat _)[0,1,3,7,9];
558		1114
559	# read the directory entries	1115	my $wd = [shift, "."];
		1116
		1117	# stat once
560	aioreq_pri $pri;	1118	aioreq_pri $pri;
561	add $grp aio_readdir $path, sub {	1119	add $grp aio_stat $wd, sub {
562	my $entries = shift
563	or return $grp->result ();	1120	return $grp->result () if $_[0];
		1121	my $now = time;
		1122	my $hash1 = join ":", (stat _)[0,1,3,7,9];
564		1123
565	# stat the dir another time	1124	# read the directory entries
566	aioreq_pri $pri;	1125	aioreq_pri $pri;
		1126	add $grp aio_readdirx $wd, READDIR_DIRS_FIRST, sub {
		1127	my $entries = shift
		1128	or return $grp->result ();
		1129
		1130	# stat the dir another time
		1131	aioreq_pri $pri;
567	add $grp aio_stat $path, sub {	1132	add $grp aio_stat $wd, sub {
568	my $hash2 = join ":", (stat _)[0,1,3,7,9];	1133	my $hash2 = join ":", (stat _)[0,1,3,7,9];
569		1134
570	my $ndirs;	1135	my $ndirs;
571		1136
572	# take the slow route if anything looks fishy	1137	# take the slow route if anything looks fishy
573	if ($hash1 ne $hash2 or (stat _)[9] == $now) {	1138	if ($hash1 ne $hash2 or (stat _)[9] == $now) {
574	$ndirs = -1;	1139	$ndirs = -1;
575	} else {	1140	} else {
576	# if nlink == 2, we are finished	1141	# if nlink == 2, we are finished
577	# on non-posix-fs's, we rely on nlink < 2	1142	# for non-posix-fs's, we rely on nlink < 2
578	$ndirs = (stat _)[3] - 2	1143	$ndirs = (stat _)[3] - 2
579	or return $grp->result ([], $entries);	1144	or return $grp->result ([], $entries);
580	}	1145	}
581		1146
582	# sort into likely dirs and likely nondirs
583	# dirs == files without ".", short entries first
584	$entries = [map $_->[0],
585	sort { $b->[1] cmp $a->[1] }
586	map [$_, sprintf "%s%04d", (/.\./ ? "1" : "0"), length],
587	@$entries];
588
589	my (@dirs, @nondirs);	1147	my (@dirs, @nondirs);
590		1148
591	my $statgrp = add $grp aio_group sub {	1149	my $statgrp = add $grp aio_group sub {
592	$grp->result (\@dirs, \@nondirs);	1150	$grp->result (\@dirs, \@nondirs);
593	};	1151	};
594		1152
595	limit $statgrp $maxreq;	1153	limit $statgrp $maxreq;
596	feed $statgrp sub {	1154	feed $statgrp sub {
597	return unless @$entries;	1155	return unless @$entries;
598	my $entry = pop @$entries;	1156	my $entry = shift @$entries;
599		1157
600	aioreq_pri $pri;	1158	aioreq_pri $pri;
		1159	$wd->[1] = "$entry/.";
601	add $statgrp aio_stat "$path/$entry/.", sub {	1160	add $statgrp aio_stat $wd, sub {
602	if ($_[0] < 0) {	1161	if ($_[0] < 0) {
603	push @nondirs, $entry;	1162	push @nondirs, $entry;
604	} else {	1163	} else {
605	# need to check for real directory	1164	# need to check for real directory
606	aioreq_pri $pri;	1165	aioreq_pri $pri;
		1166	$wd->[1] = $entry;
607	add $statgrp aio_lstat "$path/$entry", sub {	1167	add $statgrp aio_lstat $wd, sub {
608	if (-d _) {	1168	if (-d _) {
609	push @dirs, $entry;	1169	push @dirs, $entry;
610		1170
611	unless (--$ndirs) {	1171	unless (--$ndirs) {
612	push @nondirs, @$entries;	1172	push @nondirs, @$entries;
613	feed $statgrp;	1173	feed $statgrp;
		1174	}
		1175	} else {
		1176	push @nondirs, $entry;
614	}	1177	}
615	} else {
616	push @nondirs, $entry;
617	}	1178	}
618	}	1179	}
619	}	1180	};
620	};	1181	};
621	};	1182	};
622	};	1183	};
623	};	1184	};
624	};	1185	};
625		1186
626	$grp	1187	$grp
627	}	1188	}
628		1189
		1190	=item aio_rmtree $pathname, $callback->($status)
		1191
		1192	Delete a directory tree starting (and including) C<$path>, return the
		1193	status of the final C<rmdir> only. This is a composite request that
		1194	uses C<aio_scandir> to recurse into and rmdir directories, and unlink
		1195	everything else.
		1196
		1197	=cut
		1198
		1199	sub aio_rmtree;
		1200	sub aio_rmtree($;$) {
		1201	my ($path, $cb) = @_;
		1202
		1203	my $pri = aioreq_pri;
		1204	my $grp = aio_group $cb;
		1205
		1206	aioreq_pri $pri;
		1207	add $grp aio_scandir $path, 0, sub {
		1208	my ($dirs, $nondirs) = @_;
		1209
		1210	my $dirgrp = aio_group sub {
		1211	add $grp aio_rmdir $path, sub {
		1212	$grp->result ($_[0]);
		1213	};
		1214	};
		1215
		1216	(aioreq_pri $pri), add $dirgrp aio_rmtree "$path/$_" for @$dirs;
		1217	(aioreq_pri $pri), add $dirgrp aio_unlink "$path/$_" for @$nondirs;
		1218
		1219	add $grp $dirgrp;
		1220	};
		1221
		1222	$grp
		1223	}
		1224
		1225	=item aio_sync $callback->($status)
		1226
		1227	Asynchronously call sync and call the callback when finished.
		1228
629	=item aio_fsync $fh, $callback->($status)	1229	=item aio_fsync $fh, $callback->($status)
630		1230
631	Asynchronously call fsync on the given filehandle and call the callback	1231	Asynchronously call fsync on the given filehandle and call the callback
632	with the fsync result code.	1232	with the fsync result code.
633		1233
…		…
636	Asynchronously call fdatasync on the given filehandle and call the	1236	Asynchronously call fdatasync on the given filehandle and call the
637	callback with the fdatasync result code.	1237	callback with the fdatasync result code.
638		1238
639	If this call isn't available because your OS lacks it or it couldn't be	1239	If this call isn't available because your OS lacks it or it couldn't be
640	detected, it will be emulated by calling C<fsync> instead.	1240	detected, it will be emulated by calling C<fsync> instead.
		1241
		1242	=item aio_syncfs $fh, $callback->($status)
		1243
		1244	Asynchronously call the syncfs syscall to sync the filesystem associated
		1245	to the given filehandle and call the callback with the syncfs result
		1246	code. If syncfs is not available, calls sync(), but returns C<-1> and sets
		1247	errno to C<ENOSYS> nevertheless.
		1248
		1249	=item aio_sync_file_range $fh, $offset, $nbytes, $flags, $callback->($status)
		1250
		1251	Sync the data portion of the file specified by C<$offset> and C<$length>
		1252	to disk (but NOT the metadata), by calling the Linux-specific
		1253	sync_file_range call. If sync_file_range is not available or it returns
		1254	ENOSYS, then fdatasync or fsync is being substituted.
		1255
		1256	C<$flags> can be a combination of C<IO::AIO::SYNC_FILE_RANGE_WAIT_BEFORE>,
		1257	C<IO::AIO::SYNC_FILE_RANGE_WRITE> and
		1258	C<IO::AIO::SYNC_FILE_RANGE_WAIT_AFTER>: refer to the sync_file_range
		1259	manpage for details.
		1260
		1261	=item aio_pathsync $pathname, $callback->($status)
		1262
		1263	This request tries to open, fsync and close the given path. This is a
		1264	composite request intended to sync directories after directory operations
		1265	(E.g. rename). This might not work on all operating systems or have any
		1266	specific effect, but usually it makes sure that directory changes get
		1267	written to disc. It works for anything that can be opened for read-only,
		1268	not just directories.
		1269
		1270	Future versions of this function might fall back to other methods when
		1271	C<fsync> on the directory fails (such as calling C<sync>).
		1272
		1273	Passes C<0> when everything went ok, and C<-1> on error.
		1274
		1275	=cut
		1276
		1277	sub aio_pathsync($;$) {
		1278	my ($path, $cb) = @_;
		1279
		1280	my $pri = aioreq_pri;
		1281	my $grp = aio_group $cb;
		1282
		1283	aioreq_pri $pri;
		1284	add $grp aio_open $path, O_RDONLY, 0, sub {
		1285	my ($fh) = @_;
		1286	if ($fh) {
		1287	aioreq_pri $pri;
		1288	add $grp aio_fsync $fh, sub {
		1289	$grp->result ($_[0]);
		1290
		1291	aioreq_pri $pri;
		1292	add $grp aio_close $fh;
		1293	};
		1294	} else {
		1295	$grp->result (-1);
		1296	}
		1297	};
		1298
		1299	$grp
		1300	}
		1301
		1302	=item aio_msync $scalar, $offset = 0, $length = undef, flags = 0, $callback->($status)
		1303
		1304	This is a rather advanced IO::AIO call, which only works on mmap(2)ed
		1305	scalars (see the C<IO::AIO::mmap> function, although it also works on data
		1306	scalars managed by the L<Sys::Mmap> or L<Mmap> modules, note that the
		1307	scalar must only be modified in-place while an aio operation is pending on
		1308	it).
		1309
		1310	It calls the C<msync> function of your OS, if available, with the memory
		1311	area starting at C<$offset> in the string and ending C<$length> bytes
		1312	later. If C<$length> is negative, counts from the end, and if C<$length>
		1313	is C<undef>, then it goes till the end of the string. The flags can be
		1314	a combination of C<IO::AIO::MS_ASYNC>, C<IO::AIO::MS_INVALIDATE> and
		1315	C<IO::AIO::MS_SYNC>.
		1316
		1317	=item aio_mtouch $scalar, $offset = 0, $length = undef, flags = 0, $callback->($status)
		1318
		1319	This is a rather advanced IO::AIO call, which works best on mmap(2)ed
		1320	scalars.
		1321
		1322	It touches (reads or writes) all memory pages in the specified
		1323	range inside the scalar. All caveats and parameters are the same
		1324	as for C<aio_msync>, above, except for flags, which must be either
		1325	C<0> (which reads all pages and ensures they are instantiated) or
		1326	C<IO::AIO::MT_MODIFY>, which modifies the memory pages (by reading and
		1327	writing an octet from it, which dirties the page).
		1328
		1329	=item aio_mlock $scalar, $offset = 0, $length = undef, $callback->($status)
		1330
		1331	This is a rather advanced IO::AIO call, which works best on mmap(2)ed
		1332	scalars.
		1333
		1334	It reads in all the pages of the underlying storage into memory (if any)
		1335	and locks them, so they are not getting swapped/paged out or removed.
		1336
		1337	If C<$length> is undefined, then the scalar will be locked till the end.
		1338
		1339	On systems that do not implement C<mlock>, this function returns C<-1>
		1340	and sets errno to C<ENOSYS>.
		1341
		1342	Note that the corresponding C<munlock> is synchronous and is
		1343	documented under L<MISCELLANEOUS FUNCTIONS>.
		1344
		1345	Example: open a file, mmap and mlock it - both will be undone when
		1346	C<$data> gets destroyed.
		1347
		1348	open my $fh, "<", $path or die "$path: $!";
		1349	my $data;
		1350	IO::AIO::mmap $data, -s $fh, IO::AIO::PROT_READ, IO::AIO::MAP_SHARED, $fh;
		1351	aio_mlock $data; # mlock in background
		1352
		1353	=item aio_mlockall $flags, $callback->($status)
		1354
		1355	Calls the C<mlockall> function with the given C<$flags> (a combination of
		1356	C<IO::AIO::MCL_CURRENT> and C<IO::AIO::MCL_FUTURE>).
		1357
		1358	On systems that do not implement C<mlockall>, this function returns C<-1>
		1359	and sets errno to C<ENOSYS>.
		1360
		1361	Note that the corresponding C<munlockall> is synchronous and is
		1362	documented under L<MISCELLANEOUS FUNCTIONS>.
		1363
		1364	Example: asynchronously lock all current and future pages into memory.
		1365
		1366	aio_mlockall IO::AIO::MCL_FUTURE;
		1367
		1368	=item aio_fiemap $fh, $start, $length, $flags, $count, $cb->(\@extents)
		1369
		1370	Queries the extents of the given file (by calling the Linux C<FIEMAP>
		1371	ioctl, see L<http://cvs.schmorp.de/IO-AIO/doc/fiemap.txt> for details). If
		1372	the ioctl is not available on your OS, then this request will fail with
		1373	C<ENOSYS>.
		1374
		1375	C<$start> is the starting offset to query extents for, C<$length> is the
		1376	size of the range to query - if it is C<undef>, then the whole file will
		1377	be queried.
		1378
		1379	C<$flags> is a combination of flags (C<IO::AIO::FIEMAP_FLAG_SYNC> or
		1380	C<IO::AIO::FIEMAP_FLAG_XATTR> - C<IO::AIO::FIEMAP_FLAGS_COMPAT> is also
		1381	exported), and is normally C<0> or C<IO::AIO::FIEMAP_FLAG_SYNC> to query
		1382	the data portion.
		1383
		1384	C<$count> is the maximum number of extent records to return. If it is
		1385	C<undef>, then IO::AIO queries all extents of the range. As a very special
		1386	case, if it is C<0>, then the callback receives the number of extents
		1387	instead of the extents themselves (which is unreliable, see below).
		1388
		1389	If an error occurs, the callback receives no arguments. The special
		1390	C<errno> value C<IO::AIO::EBADR> is available to test for flag errors.
		1391
		1392	Otherwise, the callback receives an array reference with extent
		1393	structures. Each extent structure is an array reference itself, with the
		1394	following members:
		1395
		1396	[$logical, $physical, $length, $flags]
		1397
		1398	Flags is any combination of the following flag values (typically either C<0>
		1399	or C<IO::AIO::FIEMAP_EXTENT_LAST> (1)):
		1400
		1401	C<IO::AIO::FIEMAP_EXTENT_LAST>, C<IO::AIO::FIEMAP_EXTENT_UNKNOWN>,
		1402	C<IO::AIO::FIEMAP_EXTENT_DELALLOC>, C<IO::AIO::FIEMAP_EXTENT_ENCODED>,
		1403	C<IO::AIO::FIEMAP_EXTENT_DATA_ENCRYPTED>, C<IO::AIO::FIEMAP_EXTENT_NOT_ALIGNED>,
		1404	C<IO::AIO::FIEMAP_EXTENT_DATA_INLINE>, C<IO::AIO::FIEMAP_EXTENT_DATA_TAIL>,
		1405	C<IO::AIO::FIEMAP_EXTENT_UNWRITTEN>, C<IO::AIO::FIEMAP_EXTENT_MERGED> or
		1406	C<IO::AIO::FIEMAP_EXTENT_SHARED>.
		1407
		1408	At the time of this writing (Linux 3.2), this requets is unreliable unless
		1409	C<$count> is C<undef>, as the kernel has all sorts of bugs preventing
		1410	it to return all extents of a range for files with large number of
		1411	extents. The code works around all these issues if C<$count> is undef.
641		1412
642	=item aio_group $callback->(...)	1413	=item aio_group $callback->(...)
643		1414
644	This is a very special aio request: Instead of doing something, it is a	1415	This is a very special aio request: Instead of doing something, it is a
645	container for other aio requests, which is useful if you want to bundle	1416	container for other aio requests, which is useful if you want to bundle
…		…
683	immense (it blocks a thread for a long time) so do not use this function	1454	immense (it blocks a thread for a long time) so do not use this function
684	except to put your application under artificial I/O pressure.	1455	except to put your application under artificial I/O pressure.
685		1456
686	=back	1457	=back
687		1458
		1459
		1460	=head2 IO::AIO::WD - multiple working directories
		1461
		1462	Your process only has one current working directory, which is used by all
		1463	threads. This makes it hard to use relative paths (some other component
		1464	could call C<chdir> at any time, and it is hard to control when the path
		1465	will be used by IO::AIO).
		1466
		1467	One solution for this is to always use absolute paths. This usually works,
		1468	but can be quite slow (the kernel has to walk the whole path on every
		1469	access), and can also be a hassle to implement.
		1470
		1471	Newer POSIX systems have a number of functions (openat, fdopendir,
		1472	futimensat and so on) that make it possible to specify working directories
		1473	per operation.
		1474
		1475	For portability, and because the clowns who "designed", or shall I write,
		1476	perpetrated this new interface were obviously half-drunk, this abstraction
		1477	cannot be perfect, though.
		1478
		1479	IO::AIO allows you to convert directory paths into a so-called IO::AIO::WD
		1480	object. This object stores the canonicalised, absolute version of the
		1481	path, and on systems that allow it, also a directory file descriptor.
		1482
		1483	Everywhere where a pathname is accepted by IO::AIO (e.g. in C<aio_stat>
		1484	or C<aio_unlink>), one can specify an array reference with an IO::AIO::WD
		1485	object and a pathname instead (or the IO::AIO::WD object alone, which
		1486	gets interpreted as C<[$wd, "."]>). If the pathname is absolute, the
		1487	IO::AIO::WD object is ignored, otherwise the pathname is resolved relative
		1488	to that IO::AIO::WD object.
		1489
		1490	For example, to get a wd object for F</etc> and then stat F<passwd>
		1491	inside, you would write:
		1492
		1493	aio_wd "/etc", sub {
		1494	my $etcdir = shift;
		1495
		1496	# although $etcdir can be undef on error, there is generally no reason
		1497	# to check for errors here, as aio_stat will fail with ENOENT
		1498	# when $etcdir is undef.
		1499
		1500	aio_stat [$etcdir, "passwd"], sub {
		1501	# yay
		1502	};
		1503	};
		1504
		1505	That C<aio_wd> is a request and not a normal function shows that creating
		1506	an IO::AIO::WD object is itself a potentially blocking operation, which is
		1507	why it is done asynchronously.
		1508
		1509	To stat the directory obtained with C<aio_wd> above, one could write
		1510	either of the following three request calls:
		1511
		1512	aio_lstat "/etc" , sub { ... # pathname as normal string
		1513	aio_lstat [$wd, "."], sub { ... # "." relative to $wd (i.e. $wd itself)
		1514	aio_lstat $wd , sub { ... # shorthand for the previous
		1515
		1516	As with normal pathnames, IO::AIO keeps a copy of the working directory
		1517	object and the pathname string, so you could write the following without
		1518	causing any issues due to C<$path> getting reused:
		1519
		1520	my $path = [$wd, undef];
		1521
		1522	for my $name (qw(abc def ghi)) {
		1523	$path->[1] = $name;
		1524	aio_stat $path, sub {
		1525	# ...
		1526	};
		1527	}
		1528
		1529	There are some caveats: when directories get renamed (or deleted), the
		1530	pathname string doesn't change, so will point to the new directory (or
		1531	nowhere at all), while the directory fd, if available on the system,
		1532	will still point to the original directory. Most functions accepting a
		1533	pathname will use the directory fd on newer systems, and the string on
		1534	older systems. Some functions (such as realpath) will always rely on the
		1535	string form of the pathname.
		1536
		1537	So this functionality is mainly useful to get some protection against
		1538	C<chdir>, to easily get an absolute path out of a relative path for future
		1539	reference, and to speed up doing many operations in the same directory
		1540	(e.g. when stat'ing all files in a directory).
		1541
		1542	The following functions implement this working directory abstraction:
		1543
		1544	=over 4
		1545
		1546	=item aio_wd $pathname, $callback->($wd)
		1547
		1548	Asynchonously canonicalise the given pathname and convert it to an
		1549	IO::AIO::WD object representing it. If possible and supported on the
		1550	system, also open a directory fd to speed up pathname resolution relative
		1551	to this working directory.
		1552
		1553	If something goes wrong, then C<undef> is passwd to the callback instead
		1554	of a working directory object and C<$!> is set appropriately. Since
		1555	passing C<undef> as working directory component of a pathname fails the
		1556	request with C<ENOENT>, there is often no need for error checking in the
		1557	C<aio_wd> callback, as future requests using the value will fail in the
		1558	expected way.
		1559
		1560	=item IO::AIO::CWD
		1561
		1562	This is a compiletime constant (object) that represents the process
		1563	current working directory.
		1564
		1565	Specifying this object as working directory object for a pathname is as if
		1566	the pathname would be specified directly, without a directory object. For
		1567	example, these calls are functionally identical:
		1568
		1569	aio_stat "somefile", sub { ... };
		1570	aio_stat [IO::AIO::CWD, "somefile"], sub { ... };
		1571
		1572	=back
		1573
		1574	To recover the path associated with an IO::AIO::WD object, you can use
		1575	C<aio_realpath>:
		1576
		1577	aio_realpath $wd, sub {
		1578	warn "path is $_[0]\n";
		1579	};
		1580
		1581	Currently, C<aio_statvfs> always, and C<aio_rename> and C<aio_rmdir>
		1582	sometimes, fall back to using an absolue path.
		1583
688	=head2 IO::AIO::REQ CLASS	1584	=head2 IO::AIO::REQ CLASS
689		1585
690	All non-aggregate C<aio_*> functions return an object of this class when	1586	All non-aggregate C<aio_*> functions return an object of this class when
691	called in non-void context.	1587	called in non-void context.
692		1588
…		…
695	=item cancel $req	1591	=item cancel $req
696		1592
697	Cancels the request, if possible. Has the effect of skipping execution	1593	Cancels the request, if possible. Has the effect of skipping execution
698	when entering the B<execute> state and skipping calling the callback when	1594	when entering the B<execute> state and skipping calling the callback when
699	entering the the B<result> state, but will leave the request otherwise	1595	entering the the B<result> state, but will leave the request otherwise
700	untouched. That means that requests that currently execute will not be	1596	untouched (with the exception of readdir). That means that requests that
701	stopped and resources held by the request will not be freed prematurely.	1597	currently execute will not be stopped and resources held by the request
		1598	will not be freed prematurely.
702		1599
703	=item cb $req $callback->(...)	1600	=item cb $req $callback->(...)
704		1601
705	Replace (or simply set) the callback registered to the request.	1602	Replace (or simply set) the callback registered to the request.
706		1603
…		…
757	Their lifetime, simplified, looks like this: when they are empty, they	1654	Their lifetime, simplified, looks like this: when they are empty, they
758	will finish very quickly. If they contain only requests that are in the	1655	will finish very quickly. If they contain only requests that are in the
759	C<done> state, they will also finish. Otherwise they will continue to	1656	C<done> state, they will also finish. Otherwise they will continue to
760	exist.	1657	exist.
761		1658
762	That means after creating a group you have some time to add requests. And	1659	That means after creating a group you have some time to add requests
763	in the callbacks of those requests, you can add further requests to the	1660	(precisely before the callback has been invoked, which is only done within
764	group. And only when all those requests have finished will the the group	1661	the C<poll_cb>). And in the callbacks of those requests, you can add
765	itself finish.	1662	further requests to the group. And only when all those requests have
		1663	finished will the the group itself finish.
766		1664
767	=over 4	1665	=over 4
768		1666
769	=item add $grp ...	1667	=item add $grp ...
770		1668
…		…
779	=item $grp->cancel_subs	1677	=item $grp->cancel_subs
780		1678
781	Cancel all subrequests and clears any feeder, but not the group request	1679	Cancel all subrequests and clears any feeder, but not the group request
782	itself. Useful when you queued a lot of events but got a result early.	1680	itself. Useful when you queued a lot of events but got a result early.
783		1681
		1682	The group request will finish normally (you cannot add requests to the
		1683	group).
		1684
784	=item $grp->result (...)	1685	=item $grp->result (...)
785		1686
786	Set the result value(s) that will be passed to the group callback when all	1687	Set the result value(s) that will be passed to the group callback when all
787	subrequests have finished and set thre groups errno to the current value	1688	subrequests have finished and set the groups errno to the current value
788	of errno (just like calling C<errno> without an error number). By default,	1689	of errno (just like calling C<errno> without an error number). By default,
789	no argument will be passed and errno is zero.	1690	no argument will be passed and errno is zero.
790		1691
791	=item $grp->errno ([$errno])	1692	=item $grp->errno ([$errno])
792		1693
…		…
803	=item feed $grp $callback->($grp)	1704	=item feed $grp $callback->($grp)
804		1705
805	Sets a feeder/generator on this group: every group can have an attached	1706	Sets a feeder/generator on this group: every group can have an attached
806	generator that generates requests if idle. The idea behind this is that,	1707	generator that generates requests if idle. The idea behind this is that,
807	although you could just queue as many requests as you want in a group,	1708	although you could just queue as many requests as you want in a group,
808	this might starve other requests for a potentially long time. For	1709	this might starve other requests for a potentially long time. For example,
809	example, C<aio_scandir> might generate hundreds of thousands C<aio_stat>	1710	C<aio_scandir> might generate hundreds of thousands of C<aio_stat>
810	requests, delaying any later requests for a long time.	1711	requests, delaying any later requests for a long time.
811		1712
812	To avoid this, and allow incremental generation of requests, you can	1713	To avoid this, and allow incremental generation of requests, you can
813	instead a group and set a feeder on it that generates those requests. The	1714	instead a group and set a feeder on it that generates those requests. The
814	feed callback will be called whenever there are few enough (see C<limit>,	1715	feed callback will be called whenever there are few enough (see C<limit>,
…		…
819	not impose any limits).	1720	not impose any limits).
820		1721
821	If the feed does not queue more requests when called, it will be	1722	If the feed does not queue more requests when called, it will be
822	automatically removed from the group.	1723	automatically removed from the group.
823		1724
824	If the feed limit is C<0>, it will be set to C<2> automatically.	1725	If the feed limit is C<0> when this method is called, it will be set to
		1726	C<2> automatically.
825		1727
826	Example:	1728	Example:
827		1729
828	# stat all files in @files, but only ever use four aio requests concurrently:	1730	# stat all files in @files, but only ever use four aio requests concurrently:
829		1731
…		…
841	Sets the feeder limit for the group: The feeder will be called whenever	1743	Sets the feeder limit for the group: The feeder will be called whenever
842	the group contains less than this many requests.	1744	the group contains less than this many requests.
843		1745
844	Setting the limit to C<0> will pause the feeding process.	1746	Setting the limit to C<0> will pause the feeding process.
845		1747
		1748	The default value for the limit is C<0>, but note that setting a feeder
		1749	automatically bumps it up to C<2>.
		1750
846	=back	1751	=back
847		1752
848	=head2 SUPPORT FUNCTIONS	1753	=head2 SUPPORT FUNCTIONS
849		1754
		1755	=head3 EVENT PROCESSING AND EVENT LOOP INTEGRATION
		1756
850	=over 4	1757	=over 4
851		1758
852	=item $fileno = IO::AIO::poll_fileno	1759	=item $fileno = IO::AIO::poll_fileno
853		1760
854	Return the I<request result pipe file descriptor>. This filehandle must be	1761	Return the I<request result pipe file descriptor>. This filehandle must be
855	polled for reading by some mechanism outside this module (e.g. Event or	1762	polled for reading by some mechanism outside this module (e.g. EV, Glib,
856	select, see below or the SYNOPSIS). If the pipe becomes readable you have	1763	select and so on, see below or the SYNOPSIS). If the pipe becomes readable
857	to call C<poll_cb> to check the results.	1764	you have to call C<poll_cb> to check the results.
858		1765
859	See C<poll_cb> for an example.	1766	See C<poll_cb> for an example.
860		1767
861	=item IO::AIO::poll_cb	1768	=item IO::AIO::poll_cb
862		1769
863	Process all outstanding events on the result pipe. You have to call this	1770	Process some requests that have reached the result phase (i.e. they have
864	regularly. Returns the number of events processed. Returns immediately	1771	been executed but the results are not yet reported). You have to call
865	when no events are outstanding.	1772	this "regularly" to finish outstanding requests.
866		1773
		1774	Returns C<0> if all events could be processed (or there were no
		1775	events to process), or C<-1> if it returned earlier for whatever
		1776	reason. Returns immediately when no events are outstanding. The amount
		1777	of events processed depends on the settings of C<IO::AIO::max_poll_req>,
		1778	C<IO::AIO::max_poll_time> and C<IO::AIO::max_outstanding>.
		1779
867	If not all requests were processed for whatever reason, the filehandle	1780	If not all requests were processed for whatever reason, the poll file
868	will still be ready when C<poll_cb> returns.	1781	descriptor will still be ready when C<poll_cb> returns, so normally you
		1782	don't have to do anything special to have it called later.
		1783
		1784	Apart from calling C<IO::AIO::poll_cb> when the event filehandle becomes
		1785	ready, it can be beneficial to call this function from loops which submit
		1786	a lot of requests, to make sure the results get processed when they become
		1787	available and not just when the loop is finished and the event loop takes
		1788	over again. This function returns very fast when there are no outstanding
		1789	requests.
869		1790
870	Example: Install an Event watcher that automatically calls	1791	Example: Install an Event watcher that automatically calls
871	IO::AIO::poll_cb with high priority:	1792	IO::AIO::poll_cb with high priority (more examples can be found in the
		1793	SYNOPSIS section, at the top of this document):
872		1794
873	Event->io (fd => IO::AIO::poll_fileno,	1795	Event->io (fd => IO::AIO::poll_fileno,
874	poll => 'r', async => 1,	1796	poll => 'r', async => 1,
875	cb => \&IO::AIO::poll_cb);	1797	cb => \&IO::AIO::poll_cb);
876		1798
877	=item IO::AIO::poll_some $max_requests	1799	=item IO::AIO::poll_wait
878		1800
879	Similar to C<poll_cb>, but only processes up to C<$max_requests> requests	1801	Wait until either at least one request is in the result phase or no
880	at a time.	1802	requests are outstanding anymore.
881		1803
882	Useful if you want to ensure some level of interactiveness when perl is	1804	This is useful if you want to synchronously wait for some requests to
883	not fast enough to process all requests in time.	1805	become ready, without actually handling them.
		1806
		1807	See C<nreqs> for an example.
		1808
		1809	=item IO::AIO::poll
		1810
		1811	Waits until some requests have been handled.
		1812
		1813	Returns the number of requests processed, but is otherwise strictly
		1814	equivalent to:
		1815
		1816	IO::AIO::poll_wait, IO::AIO::poll_cb
		1817
		1818	=item IO::AIO::flush
		1819
		1820	Wait till all outstanding AIO requests have been handled.
		1821
		1822	Strictly equivalent to:
		1823
		1824	IO::AIO::poll_wait, IO::AIO::poll_cb
		1825	while IO::AIO::nreqs;
		1826
		1827	=item IO::AIO::max_poll_reqs $nreqs
		1828
		1829	=item IO::AIO::max_poll_time $seconds
		1830
		1831	These set the maximum number of requests (default C<0>, meaning infinity)
		1832	that are being processed by C<IO::AIO::poll_cb> in one call, respectively
		1833	the maximum amount of time (default C<0>, meaning infinity) spent in
		1834	C<IO::AIO::poll_cb> to process requests (more correctly the mininum amount
		1835	of time C<poll_cb> is allowed to use).
		1836
		1837	Setting C<max_poll_time> to a non-zero value creates an overhead of one
		1838	syscall per request processed, which is not normally a problem unless your
		1839	callbacks are really really fast or your OS is really really slow (I am
		1840	not mentioning Solaris here). Using C<max_poll_reqs> incurs no overhead.
		1841
		1842	Setting these is useful if you want to ensure some level of
		1843	interactiveness when perl is not fast enough to process all requests in
		1844	time.
		1845
		1846	For interactive programs, values such as C<0.01> to C<0.1> should be fine.
884		1847
885	Example: Install an Event watcher that automatically calls	1848	Example: Install an Event watcher that automatically calls
886	IO::AIO::poll_some with low priority, to ensure that other parts of the	1849	IO::AIO::poll_cb with low priority, to ensure that other parts of the
887	program get the CPU sometimes even under high AIO load.	1850	program get the CPU sometimes even under high AIO load.
888		1851
		1852	# try not to spend much more than 0.1s in poll_cb
		1853	IO::AIO::max_poll_time 0.1;
		1854
		1855	# use a low priority so other tasks have priority
889	Event->io (fd => IO::AIO::poll_fileno,	1856	Event->io (fd => IO::AIO::poll_fileno,
890	poll => 'r', nice => 1,	1857	poll => 'r', nice => 1,
891	cb => sub { IO::AIO::poll_some 256 });	1858	cb => &IO::AIO::poll_cb);
892		1859
893	=item IO::AIO::poll_wait	1860	=back
894		1861
895	Wait till the result filehandle becomes ready for reading (simply does a	1862	=head3 CONTROLLING THE NUMBER OF THREADS
896	C<select> on the filehandle. This is useful if you want to synchronously wait
897	for some requests to finish).
898		1863
899	See C<nreqs> for an example.	1864	=over
900
901	=item IO::AIO::nreqs
902
903	Returns the number of requests currently in the ready, execute or pending
904	states (i.e. for which their callback has not been invoked yet).
905
906	Example: wait till there are no outstanding requests anymore:
907
908	IO::AIO::poll_wait, IO::AIO::poll_cb
909	while IO::AIO::nreqs;
910
911	=item IO::AIO::nready
912
913	Returns the number of requests currently in the ready state (not yet
914	executed).
915
916	=item IO::AIO::npending
917
918	Returns the number of requests currently in the pending state (executed,
919	but not yet processed by poll_cb).
920
921	=item IO::AIO::flush
922
923	Wait till all outstanding AIO requests have been handled.
924
925	Strictly equivalent to:
926
927	IO::AIO::poll_wait, IO::AIO::poll_cb
928	while IO::AIO::nreqs;
929
930	=item IO::AIO::poll
931
932	Waits until some requests have been handled.
933
934	Strictly equivalent to:
935
936	IO::AIO::poll_wait, IO::AIO::poll_cb
937	if IO::AIO::nreqs;
938		1865
939	=item IO::AIO::min_parallel $nthreads	1866	=item IO::AIO::min_parallel $nthreads
940		1867
941	Set the minimum number of AIO threads to C<$nthreads>. The current	1868	Set the minimum number of AIO threads to C<$nthreads>. The current
942	default is C<8>, which means eight asynchronous operations can execute	1869	default is C<8>, which means eight asynchronous operations can execute
943	concurrently at any one time (the number of outstanding requests,	1870	concurrently at any one time (the number of outstanding requests,
944	however, is unlimited).	1871	however, is unlimited).
945		1872
946	IO::AIO starts threads only on demand, when an AIO request is queued and	1873	IO::AIO starts threads only on demand, when an AIO request is queued and
947	no free thread exists.	1874	no free thread exists. Please note that queueing up a hundred requests can
		1875	create demand for a hundred threads, even if it turns out that everything
		1876	is in the cache and could have been processed faster by a single thread.
948		1877
949	It is recommended to keep the number of threads relatively low, as some	1878	It is recommended to keep the number of threads relatively low, as some
950	Linux kernel versions will scale negatively with the number of threads	1879	Linux kernel versions will scale negatively with the number of threads
951	(higher parallelity => MUCH higher latency). With current Linux 2.6	1880	(higher parallelity => MUCH higher latency). With current Linux 2.6
952	versions, 4-32 threads should be fine.	1881	versions, 4-32 threads should be fine.
…		…
966	This module automatically runs C<max_parallel 0> at program end, to ensure	1895	This module automatically runs C<max_parallel 0> at program end, to ensure
967	that all threads are killed and that there are no outstanding requests.	1896	that all threads are killed and that there are no outstanding requests.
968		1897
969	Under normal circumstances you don't need to call this function.	1898	Under normal circumstances you don't need to call this function.
970		1899
		1900	=item IO::AIO::max_idle $nthreads
		1901
		1902	Limit the number of threads (default: 4) that are allowed to idle
		1903	(i.e., threads that did not get a request to process within the idle
		1904	timeout (default: 10 seconds). That means if a thread becomes idle while
		1905	C<$nthreads> other threads are also idle, it will free its resources and
		1906	exit.
		1907
		1908	This is useful when you allow a large number of threads (e.g. 100 or 1000)
		1909	to allow for extremely high load situations, but want to free resources
		1910	under normal circumstances (1000 threads can easily consume 30MB of RAM).
		1911
		1912	The default is probably ok in most situations, especially if thread
		1913	creation is fast. If thread creation is very slow on your system you might
		1914	want to use larger values.
		1915
		1916	=item IO::AIO::idle_timeout $seconds
		1917
		1918	Sets the minimum idle timeout (default 10) after which worker threads are
		1919	allowed to exit. SEe C<IO::AIO::max_idle>.
		1920
971	=item $oldmaxreqs = IO::AIO::max_outstanding $maxreqs	1921	=item IO::AIO::max_outstanding $maxreqs
		1922
		1923	Sets the maximum number of outstanding requests to C<$nreqs>. If
		1924	you do queue up more than this number of requests, the next call to
		1925	C<IO::AIO::poll_cb> (and other functions calling C<poll_cb>, such as
		1926	C<IO::AIO::flush> or C<IO::AIO::poll>) will block until the limit is no
		1927	longer exceeded.
		1928
		1929	In other words, this setting does not enforce a queue limit, but can be
		1930	used to make poll functions block if the limit is exceeded.
972		1931
973	This is a very bad function to use in interactive programs because it	1932	This is a very bad function to use in interactive programs because it
974	blocks, and a bad way to reduce concurrency because it is inexact: Better	1933	blocks, and a bad way to reduce concurrency because it is inexact: Better
975	use an C<aio_group> together with a feed callback.	1934	use an C<aio_group> together with a feed callback.
976		1935
977	Sets the maximum number of outstanding requests to C<$nreqs>. If you	1936	Its main use is in scripts without an event loop - when you want to stat
978	to queue up more than this number of requests, the next call to the	1937	a lot of files, you can write somehting like this:
979	C<poll_cb> (and C<poll_some> and other functions calling C<poll_cb>)
980	function will block until the limit is no longer exceeded.
981		1938
982	The default value is very large, so there is no practical limit on the	1939	IO::AIO::max_outstanding 32;
983	number of outstanding requests.
984		1940
985	You can still queue as many requests as you want. Therefore,	1941	for my $path (...) {
986	C<max_oustsanding> is mainly useful in simple scripts (with low values) or	1942	aio_stat $path , ...;
987	as a stop gap to shield against fatal memory overflow (with large values).	1943	IO::AIO::poll_cb;
		1944	}
		1945
		1946	IO::AIO::flush;
		1947
		1948	The call to C<poll_cb> inside the loop will normally return instantly, but
		1949	as soon as more thna C<32> reqeusts are in-flight, it will block until
		1950	some requests have been handled. This keeps the loop from pushing a large
		1951	number of C<aio_stat> requests onto the queue.
		1952
		1953	The default value for C<max_outstanding> is very large, so there is no
		1954	practical limit on the number of outstanding requests.
988		1955
989	=back	1956	=back
990		1957
		1958	=head3 STATISTICAL INFORMATION
		1959
		1960	=over
		1961
		1962	=item IO::AIO::nreqs
		1963
		1964	Returns the number of requests currently in the ready, execute or pending
		1965	states (i.e. for which their callback has not been invoked yet).
		1966
		1967	Example: wait till there are no outstanding requests anymore:
		1968
		1969	IO::AIO::poll_wait, IO::AIO::poll_cb
		1970	while IO::AIO::nreqs;
		1971
		1972	=item IO::AIO::nready
		1973
		1974	Returns the number of requests currently in the ready state (not yet
		1975	executed).
		1976
		1977	=item IO::AIO::npending
		1978
		1979	Returns the number of requests currently in the pending state (executed,
		1980	but not yet processed by poll_cb).
		1981
		1982	=back
		1983
		1984	=head3 MISCELLANEOUS FUNCTIONS
		1985
		1986	IO::AIO implements some functions that are useful when you want to use
		1987	some "Advanced I/O" function not available to in Perl, without going the
		1988	"Asynchronous I/O" route. Many of these have an asynchronous C<aio_*>
		1989	counterpart.
		1990
		1991	=over 4
		1992
		1993	=item IO::AIO::sendfile $ofh, $ifh, $offset, $count
		1994
		1995	Calls the C<eio_sendfile_sync> function, which is like C<aio_sendfile>,
		1996	but is blocking (this makes most sense if you know the input data is
		1997	likely cached already and the output filehandle is set to non-blocking
		1998	operations).
		1999
		2000	Returns the number of bytes copied, or C<-1> on error.
		2001
		2002	=item IO::AIO::fadvise $fh, $offset, $len, $advice
		2003
		2004	Simply calls the C<posix_fadvise> function (see its
		2005	manpage for details). The following advice constants are
		2006	available: C<IO::AIO::FADV_NORMAL>, C<IO::AIO::FADV_SEQUENTIAL>,
		2007	C<IO::AIO::FADV_RANDOM>, C<IO::AIO::FADV_NOREUSE>,
		2008	C<IO::AIO::FADV_WILLNEED>, C<IO::AIO::FADV_DONTNEED>.
		2009
		2010	On systems that do not implement C<posix_fadvise>, this function returns
		2011	ENOSYS, otherwise the return value of C<posix_fadvise>.
		2012
		2013	=item IO::AIO::madvise $scalar, $offset, $len, $advice
		2014
		2015	Simply calls the C<posix_madvise> function (see its
		2016	manpage for details). The following advice constants are
		2017	available: C<IO::AIO::MADV_NORMAL>, C<IO::AIO::MADV_SEQUENTIAL>,
		2018	C<IO::AIO::MADV_RANDOM>, C<IO::AIO::MADV_WILLNEED>, C<IO::AIO::MADV_DONTNEED>.
		2019
		2020	On systems that do not implement C<posix_madvise>, this function returns
		2021	ENOSYS, otherwise the return value of C<posix_madvise>.
		2022
		2023	=item IO::AIO::mprotect $scalar, $offset, $len, $protect
		2024
		2025	Simply calls the C<mprotect> function on the preferably AIO::mmap'ed
		2026	$scalar (see its manpage for details). The following protect
		2027	constants are available: C<IO::AIO::PROT_NONE>, C<IO::AIO::PROT_READ>,
		2028	C<IO::AIO::PROT_WRITE>, C<IO::AIO::PROT_EXEC>.
		2029
		2030	On systems that do not implement C<mprotect>, this function returns
		2031	ENOSYS, otherwise the return value of C<mprotect>.
		2032
		2033	=item IO::AIO::mmap $scalar, $length, $prot, $flags, $fh[, $offset]
		2034
		2035	Memory-maps a file (or anonymous memory range) and attaches it to the
		2036	given C<$scalar>, which will act like a string scalar. Returns true on
		2037	success, and false otherwise.
		2038
		2039	The only operations allowed on the scalar are C<substr>/C<vec> that don't
		2040	change the string length, and most read-only operations such as copying it
		2041	or searching it with regexes and so on.
		2042
		2043	Anything else is unsafe and will, at best, result in memory leaks.
		2044
		2045	The memory map associated with the C<$scalar> is automatically removed
		2046	when the C<$scalar> is destroyed, or when the C<IO::AIO::mmap> or
		2047	C<IO::AIO::munmap> functions are called.
		2048
		2049	This calls the C<mmap>(2) function internally. See your system's manual
		2050	page for details on the C<$length>, C<$prot> and C<$flags> parameters.
		2051
		2052	The C<$length> must be larger than zero and smaller than the actual
		2053	filesize.
		2054
		2055	C<$prot> is a combination of C<IO::AIO::PROT_NONE>, C<IO::AIO::PROT_EXEC>,
		2056	C<IO::AIO::PROT_READ> and/or C<IO::AIO::PROT_WRITE>,
		2057
		2058	C<$flags> can be a combination of C<IO::AIO::MAP_SHARED> or
		2059	C<IO::AIO::MAP_PRIVATE>, or a number of system-specific flags (when
		2060	not available, the are defined as 0): C<IO::AIO::MAP_ANONYMOUS>
		2061	(which is set to C<MAP_ANON> if your system only provides this
		2062	constant), C<IO::AIO::MAP_HUGETLB>, C<IO::AIO::MAP_LOCKED>,
		2063	C<IO::AIO::MAP_NORESERVE>, C<IO::AIO::MAP_POPULATE> or
		2064	C<IO::AIO::MAP_NONBLOCK>
		2065
		2066	If C<$fh> is C<undef>, then a file descriptor of C<-1> is passed.
		2067
		2068	C<$offset> is the offset from the start of the file - it generally must be
		2069	a multiple of C<IO::AIO::PAGESIZE> and defaults to C<0>.
		2070
		2071	Example:
		2072
		2073	use Digest::MD5;
		2074	use IO::AIO;
		2075
		2076	open my $fh, "<verybigfile"
		2077	or die "$!";
		2078
		2079	IO::AIO::mmap my $data, -s $fh, IO::AIO::PROT_READ, IO::AIO::MAP_SHARED, $fh
		2080	or die "verybigfile: $!";
		2081
		2082	my $fast_md5 = md5 $data;
		2083
		2084	=item IO::AIO::munmap $scalar
		2085
		2086	Removes a previous mmap and undefines the C<$scalar>.
		2087
		2088	=item IO::AIO::munlock $scalar, $offset = 0, $length = undef
		2089
		2090	Calls the C<munlock> function, undoing the effects of a previous
		2091	C<aio_mlock> call (see its description for details).
		2092
		2093	=item IO::AIO::munlockall
		2094
		2095	Calls the C<munlockall> function.
		2096
		2097	On systems that do not implement C<munlockall>, this function returns
		2098	ENOSYS, otherwise the return value of C<munlockall>.
		2099
		2100	=item IO::AIO::splice $r_fh, $r_off, $w_fh, $w_off, $length, $flags
		2101
		2102	Calls the GNU/Linux C<splice(2)> syscall, if available. If C<$r_off> or
		2103	C<$w_off> are C<undef>, then C<NULL> is passed for these, otherwise they
		2104	should be the file offset.
		2105
		2106	C<$r_fh> and C<$w_fh> should not refer to the same file, as splice might
		2107	silently corrupt the data in this case.
		2108
		2109	The following symbol flag values are available: C<IO::AIO::SPLICE_F_MOVE>,
		2110	C<IO::AIO::SPLICE_F_NONBLOCK>, C<IO::AIO::SPLICE_F_MORE> and
		2111	C<IO::AIO::SPLICE_F_GIFT>.
		2112
		2113	See the C<splice(2)> manpage for details.
		2114
		2115	=item IO::AIO::tee $r_fh, $w_fh, $length, $flags
		2116
		2117	Calls the GNU/Linux C<tee(2)> syscall, see its manpage and the
		2118	description for C<IO::AIO::splice> above for details.
		2119
		2120	=item $actual_size = IO::AIO::pipesize $r_fh[, $new_size]
		2121
		2122	Attempts to query or change the pipe buffer size. Obviously works only
		2123	on pipes, and currently works only on GNU/Linux systems, and fails with
		2124	C<-1>/C<ENOSYS> everywhere else. If anybody knows how to influence pipe buffer
		2125	size on other systems, drop me a note.
		2126
		2127	=back
		2128
991	=cut	2129	=cut
992		2130
993	# support function to convert a fd into a perl filehandle
994	sub _fd2fh {
995	return undef if $_[0] < 0;
996
997	# try to generate nice filehandles
998	my $sym = "IO::AIO::fd#$_[0]";
999	local *$sym;
1000
1001	open *$sym, "+<&=$_[0]" # usually works under any unix
1002	or open *$sym, "<&=$_[0]" # cygwin needs this
1003	or open *$sym, ">&=$_[0]" # or this
1004	or return undef;
1005
1006	*$sym
1007	}
1008
1009	min_parallel 8;	2131	min_parallel 8;
1010		2132
1011	END {	2133	END { flush }
1012	min_parallel 1;
1013	flush;
1014	};
1015		2134
1016	1;	2135	1;
1017		2136
		2137	=head1 EVENT LOOP INTEGRATION
		2138
		2139	It is recommended to use L<AnyEvent::AIO> to integrate IO::AIO
		2140	automatically into many event loops:
		2141
		2142	# AnyEvent integration (EV, Event, Glib, Tk, POE, urxvt, pureperl...)
		2143	use AnyEvent::AIO;
		2144
		2145	You can also integrate IO::AIO manually into many event loops, here are
		2146	some examples of how to do this:
		2147
		2148	# EV integration
		2149	my $aio_w = EV::io IO::AIO::poll_fileno, EV::READ, \&IO::AIO::poll_cb;
		2150
		2151	# Event integration
		2152	Event->io (fd => IO::AIO::poll_fileno,
		2153	poll => 'r',
		2154	cb => \&IO::AIO::poll_cb);
		2155
		2156	# Glib/Gtk2 integration
		2157	add_watch Glib::IO IO::AIO::poll_fileno,
		2158	in => sub { IO::AIO::poll_cb; 1 };
		2159
		2160	# Tk integration
		2161	Tk::Event::IO->fileevent (IO::AIO::poll_fileno, "",
		2162	readable => \&IO::AIO::poll_cb);
		2163
		2164	# Danga::Socket integration
		2165	Danga::Socket->AddOtherFds (IO::AIO::poll_fileno =>
		2166	\&IO::AIO::poll_cb);
		2167
1018	=head2 FORK BEHAVIOUR	2168	=head2 FORK BEHAVIOUR
1019		2169
1020	This module should do "the right thing" when the process using it forks:	2170	Usage of pthreads in a program changes the semantics of fork
		2171	considerably. Specifically, only async-safe functions can be called after
		2172	fork. Perl doesn't know about this, so in general, you cannot call fork
		2173	with defined behaviour in perl if pthreads are involved. IO::AIO uses
		2174	pthreads, so this applies, but many other extensions and (for inexplicable
		2175	reasons) perl itself often is linked against pthreads, so this limitation
		2176	applies to quite a lot of perls.
1021		2177
1022	Before the fork, IO::AIO enters a quiescent state where no requests	2178	This module no longer tries to fight your OS, or POSIX. That means IO::AIO
1023	can be added in other threads and no results will be processed. After	2179	only works in the process that loaded it. Forking is fully supported, but
1024	the fork the parent simply leaves the quiescent state and continues	2180	using IO::AIO in the child is not.
1025	request/result processing, while the child frees the request/result queue
1026	(so that the requests started before the fork will only be handled in the
1027	parent). Threads will be started on demand until the limit set in the
1028	parent process has been reached again.
1029		2181
1030	In short: the parent will, after a short pause, continue as if fork had	2182	You might get around by not I<using> IO::AIO before (or after)
1031	not been called, while the child will act as if IO::AIO has not been used	2183	forking. You could also try to call the L<IO::AIO::reinit> function in the
1032	yet.	2184	child:
		2185
		2186	=over 4
		2187
		2188	=item IO::AIO::reinit
		2189
		2190	Abandons all current requests and I/O threads and simply reinitialises all
		2191	data structures. This is not an operation supported by any standards, but
		2192	happens to work on GNU/Linux and some newer BSD systems.
		2193
		2194	The only reasonable use for this function is to call it after forking, if
		2195	C<IO::AIO> was used in the parent. Calling it while IO::AIO is active in
		2196	the process will result in undefined behaviour. Calling it at any time
		2197	will also result in any undefined (by POSIX) behaviour.
		2198
		2199	=back
1033		2200
1034	=head2 MEMORY USAGE	2201	=head2 MEMORY USAGE
1035		2202
1036	Per-request usage:	2203	Per-request usage:
1037		2204
…		…
1039	bytes of memory. In addition, stat requests need a stat buffer (possibly	2206	bytes of memory. In addition, stat requests need a stat buffer (possibly
1040	a few hundred bytes), readdir requires a result buffer and so on. Perl	2207	a few hundred bytes), readdir requires a result buffer and so on. Perl
1041	scalars and other data passed into aio requests will also be locked and	2208	scalars and other data passed into aio requests will also be locked and
1042	will consume memory till the request has entered the done state.	2209	will consume memory till the request has entered the done state.
1043		2210
1044	This is now awfully much, so queuing lots of requests is not usually a	2211	This is not awfully much, so queuing lots of requests is not usually a
1045	problem.	2212	problem.
1046		2213
1047	Per-thread usage:	2214	Per-thread usage:
1048		2215
1049	In the execution phase, some aio requests require more memory for	2216	In the execution phase, some aio requests require more memory for
…		…
1054		2221
1055	Known bugs will be fixed in the next release.	2222	Known bugs will be fixed in the next release.
1056		2223
1057	=head1 SEE ALSO	2224	=head1 SEE ALSO
1058		2225
1059	L<Coro::AIO>.	2226	L<AnyEvent::AIO> for easy integration into event loops, L<Coro::AIO> for a
		2227	more natural syntax.
1060		2228
1061	=head1 AUTHOR	2229	=head1 AUTHOR
1062		2230
1063	Marc Lehmann <schmorp@schmorp.de>	2231	Marc Lehmann <schmorp@schmorp.de>
1064	http://home.schmorp.de/	2232	http://home.schmorp.de/

Diff Legend

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