ViewVC Help

View File | Revision Log | Show Annotations | Download File

/cvs/IO-AIO/AIO.pm

Comparing IO-AIO/AIO.pm (file contents):
Revision 1.83 by root, Fri Oct 27 20:11:58 2006 UTC vs.
Revision 1.223 by root, Fri Apr 6 22:51:02 2012 UTC

…		…
4		4
5	=head1 SYNOPSIS	5	=head1 SYNOPSIS
6		6
7	use IO::AIO;	7	use IO::AIO;
8		8
9	aio_open "/etc/passwd", O_RDONLY, 0, sub {	9	aio_open "/etc/passwd", IO::AIO::O_RDONLY, 0, sub {
10	my ($fh) = @_;	10	my $fh = shift
		11	or die "/etc/passwd: $!";
11	...	12	...
12	};	13	};
13		14
14	aio_unlink "/tmp/file", sub { };	15	aio_unlink "/tmp/file", sub { };
15		16
…		…
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>).
		36
		37	Asynchronous means that operations that can normally block your program
		38	(e.g. reading from disk) will be done asynchronously: the operation
		39	will still block, but you can do something else in the meantime. This
		40	is extremely useful for programs that need to stay interactive even
		41	when doing heavy I/O (GUI programs, high performance network servers
		42	etc.), but can also be used to easily do operations in parallel that are
		43	normally done sequentially, e.g. stat'ing many files, which is much faster
		44	on a RAID volume or over NFS when you do a number of stat operations
		45	concurrently.
		46
		47	While most of this works on all types of file descriptors (for
		48	example sockets), using these functions on file descriptors that
		49	support nonblocking operation (again, sockets, pipes etc.) is
		50	very inefficient. Use an event loop for that (such as the L<EV>
		51	module): IO::AIO will naturally fit into such an event loop itself.
55		52
56	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
57	requests and signal their completion. You don't need thread support	54	requests and signal their completion. You don't need thread support
58	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
59	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
60	functions available on many operating systems. However, they are often	57	functions available on many operating systems. However, they are often
61	not well-supported or restricted (Linux doesn't allow them on normal	58	not well-supported or restricted (GNU/Linux doesn't allow them on normal
62	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
63	aio_write, so the remaining functionality would have to be implemented	60	aio_write, so the remaining functionality would have to be implemented
64	using threads anyway.	61	using threads anyway.
65		62
66	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,
67	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
68	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
69	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 Fcntl;
		74	use EV;
		75	use IO::AIO;
		76
		77	# register the IO::AIO callback with EV
		78	my $aio_w = EV::io IO::AIO::poll_fileno, EV::READ, \&IO::AIO::poll_cb;
		79
		80	# queue the request to open /etc/passwd
		81	aio_open "/etc/passwd", IO::AIO::O_RDONLY, 0, sub {
		82	my $fh = shift
		83	or die "error while opening: $!";
		84
		85	# stat'ing filehandles is generally non-blocking
		86	my $size = -s $fh;
		87
		88	# queue a request to read the file
		89	my $contents;
		90	aio_read $fh, 0, $size, $contents, 0, sub {
		91	$_[0] == $size
		92	or die "short read: $!";
		93
		94	close $fh;
		95
		96	# file contents now in $contents
		97	print $contents;
		98
		99	# exit event loop and program
		100	EV::unloop;
		101	};
		102	};
		103
		104	# possibly queue up other requests, or open GUI windows,
		105	# check for sockets etc. etc.
		106
		107	# process events as long as there are some:
		108	EV::loop;
70		109
71	=head1 REQUEST ANATOMY AND LIFETIME	110	=head1 REQUEST ANATOMY AND LIFETIME
72		111
73	Every C<aio_*> function creates a request. which is a C data structure not	112	Every C<aio_*> function creates a request. which is a C data structure not
74	directly visible to Perl.	113	directly visible to Perl.
…		…
116	Request has reached the end of its lifetime and holds no resources anymore	155	Request has reached the end of its lifetime and holds no resources anymore
117	(except possibly for the Perl object, but its connection to the actual	156	(except possibly for the Perl object, but its connection to the actual
118	aio request is severed and calling its methods will either do nothing or	157	aio request is severed and calling its methods will either do nothing or
119	result in a runtime error).	158	result in a runtime error).
120		159
		160	=back
		161
121	=cut	162	=cut
122		163
123	package IO::AIO;	164	package IO::AIO;
124		165
125	no warnings;	166	use Carp ();
126	use strict 'vars';	167
		168	use common::sense;
127		169
128	use base 'Exporter';	170	use base 'Exporter';
129		171
130	BEGIN {	172	BEGIN {
131	our $VERSION = '2.0';	173	our $VERSION = '4.13';
132		174
133	our @AIO_REQ = qw(aio_sendfile aio_read aio_write aio_open aio_close aio_stat	175	our @AIO_REQ = qw(aio_sendfile aio_seek aio_read aio_write aio_open aio_close
134	aio_lstat aio_unlink aio_rmdir aio_readdir aio_scandir aio_symlink	176	aio_stat aio_lstat aio_unlink aio_rmdir aio_readdir aio_readdirx
135	aio_fsync aio_fdatasync aio_readahead aio_rename aio_link aio_move	177	aio_scandir aio_symlink aio_readlink aio_realpath aio_sync
136	aio_copy aio_group aio_nop aio_mknod);	178	aio_fsync aio_syncfs aio_fdatasync aio_sync_file_range aio_fallocate
		179	aio_pathsync aio_readahead aio_fiemap
		180	aio_rename aio_link aio_move aio_copy aio_group
		181	aio_nop aio_mknod aio_load aio_rmtree aio_mkdir aio_chown
		182	aio_chmod aio_utime aio_truncate
		183	aio_msync aio_mtouch aio_mlock aio_mlockall
		184	aio_statvfs
		185	aio_wd);
		186
137	our @EXPORT = (@AIO_REQ, qw(aioreq_pri aioreq_nice));	187	our @EXPORT = (@AIO_REQ, qw(aioreq_pri aioreq_nice));
138	our @EXPORT_OK = qw(poll_fileno poll_cb poll_wait flush	188	our @EXPORT_OK = qw(poll_fileno poll_cb poll_wait flush
139	min_parallel max_parallel nreqs nready npending);	189	min_parallel max_parallel max_idle idle_timeout
		190	nreqs nready npending nthreads
		191	max_poll_time max_poll_reqs
		192	sendfile fadvise madvise
		193	mmap munmap munlock munlockall);
		194
		195	push @AIO_REQ, qw(aio_busy); # not exported
140		196
141	@IO::AIO::GRP::ISA = 'IO::AIO::REQ';	197	@IO::AIO::GRP::ISA = 'IO::AIO::REQ';
142		198
143	require XSLoader;	199	require XSLoader;
144	XSLoader::load ("IO::AIO", $VERSION);	200	XSLoader::load ("IO::AIO", $VERSION);
145	}	201	}
146		202
147	=head1 FUNCTIONS	203	=head1 FUNCTIONS
148		204
149	=head2 AIO FUNCTIONS	205	=head2 QUICK OVERVIEW
		206
		207	This section simply lists the prototypes of the most important functions
		208	for quick reference. See the following sections for function-by-function
		209	documentation.
		210
		211	aio_wd $pathname, $callback->($wd)
		212	aio_open $pathname, $flags, $mode, $callback->($fh)
		213	aio_close $fh, $callback->($status)
		214	aio_seek $fh,$offset,$whence, $callback->($offs)
		215	aio_read $fh,$offset,$length, $data,$dataoffset, $callback->($retval)
		216	aio_write $fh,$offset,$length, $data,$dataoffset, $callback->($retval)
		217	aio_sendfile $out_fh, $in_fh, $in_offset, $length, $callback->($retval)
		218	aio_readahead $fh,$offset,$length, $callback->($retval)
		219	aio_stat $fh_or_path, $callback->($status)
		220	aio_lstat $fh, $callback->($status)
		221	aio_statvfs $fh_or_path, $callback->($statvfs)
		222	aio_utime $fh_or_path, $atime, $mtime, $callback->($status)
		223	aio_chown $fh_or_path, $uid, $gid, $callback->($status)
		224	aio_chmod $fh_or_path, $mode, $callback->($status)
		225	aio_truncate $fh_or_path, $offset, $callback->($status)
		226	aio_unlink $pathname, $callback->($status)
		227	aio_mknod $pathname, $mode, $dev, $callback->($status)
		228	aio_link $srcpath, $dstpath, $callback->($status)
		229	aio_symlink $srcpath, $dstpath, $callback->($status)
		230	aio_readlink $pathname, $callback->($link)
		231	aio_realpath $pathname, $callback->($link)
		232	aio_rename $srcpath, $dstpath, $callback->($status)
		233	aio_mkdir $pathname, $mode, $callback->($status)
		234	aio_rmdir $pathname, $callback->($status)
		235	aio_readdir $pathname, $callback->($entries)
		236	aio_readdirx $pathname, $flags, $callback->($entries, $flags)
		237	IO::AIO::READDIR_DENTS IO::AIO::READDIR_DIRS_FIRST
		238	IO::AIO::READDIR_STAT_ORDER IO::AIO::READDIR_FOUND_UNKNOWN
		239	aio_scandir $pathname, $maxreq, $callback->($dirs, $nondirs)
		240	aio_load $pathname, $data, $callback->($status)
		241	aio_copy $srcpath, $dstpath, $callback->($status)
		242	aio_move $srcpath, $dstpath, $callback->($status)
		243	aio_rmtree $pathname, $callback->($status)
		244	aio_sync $callback->($status)
		245	aio_syncfs $fh, $callback->($status)
		246	aio_fsync $fh, $callback->($status)
		247	aio_fdatasync $fh, $callback->($status)
		248	aio_sync_file_range $fh, $offset, $nbytes, $flags, $callback->($status)
		249	aio_pathsync $pathname, $callback->($status)
		250	aio_msync $scalar, $offset = 0, $length = undef, flags = 0, $callback->($status)
		251	aio_mtouch $scalar, $offset = 0, $length = undef, flags = 0, $callback->($status)
		252	aio_mlock $scalar, $offset = 0, $length = undef, $callback->($status)
		253	aio_mlockall $flags, $callback->($status)
		254	aio_group $callback->(...)
		255	aio_nop $callback->()
		256
		257	$prev_pri = aioreq_pri [$pri]
		258	aioreq_nice $pri_adjust
		259
		260	IO::AIO::poll_wait
		261	IO::AIO::poll_cb
		262	IO::AIO::poll
		263	IO::AIO::flush
		264	IO::AIO::max_poll_reqs $nreqs
		265	IO::AIO::max_poll_time $seconds
		266	IO::AIO::min_parallel $nthreads
		267	IO::AIO::max_parallel $nthreads
		268	IO::AIO::max_idle $nthreads
		269	IO::AIO::idle_timeout $seconds
		270	IO::AIO::max_outstanding $maxreqs
		271	IO::AIO::nreqs
		272	IO::AIO::nready
		273	IO::AIO::npending
		274
		275	IO::AIO::sendfile $ofh, $ifh, $offset, $count
		276	IO::AIO::fadvise $fh, $offset, $len, $advice
		277	IO::AIO::madvise $scalar, $offset, $length, $advice
		278	IO::AIO::mprotect $scalar, $offset, $length, $protect
		279	IO::AIO::munlock $scalar, $offset = 0, $length = undef
		280	IO::AIO::munlockall
		281
		282	=head2 API NOTES
150		283
151	All the C<aio_*> calls are more or less thin wrappers around the syscall	284	All the C<aio_*> calls are more or less thin wrappers around the syscall
152	with the same name (sans C<aio_>). The arguments are similar or identical,	285	with the same name (sans C<aio_>). The arguments are similar or identical,
153	and they all accept an additional (and optional) C<$callback> argument	286	and they all accept an additional (and optional) C<$callback> argument
154	which must be a code reference. This code reference will get called with	287	which must be a code reference. This code reference will be called after
155	the syscall return code (e.g. most syscalls return C<-1> on error, unlike	288	the syscall has been executed in an asynchronous fashion. The results
156	perl, which usually delivers "false") as it's sole argument when the given	289	of the request will be passed as arguments to the callback (and, if an
157	syscall has been executed asynchronously.	290	error occured, in C<$!>) - for most requests the syscall return code (e.g.
		291	most syscalls return C<-1> on error, unlike perl, which usually delivers
		292	"false").
		293
		294	Some requests (such as C<aio_readdir>) pass the actual results and
		295	communicate failures by passing C<undef>.
158		296
159	All functions expecting a filehandle keep a copy of the filehandle	297	All functions expecting a filehandle keep a copy of the filehandle
160	internally until the request has finished.	298	internally until the request has finished.
161		299
162	All requests return objects of type L<IO::AIO::REQ> that allow further	300	All functions return request objects of type L<IO::AIO::REQ> that allow
163	manipulation of those requests while they are in-flight.	301	further manipulation of those requests while they are in-flight.
164		302
165	The pathnames you pass to these routines I<must> be absolute and	303	The pathnames you pass to these routines I<should> be absolute. The
166	encoded in byte form. The reason for the former is that at the time the	304	reason for this is that at the time the request is being executed, the
167	request is being executed, the current working directory could have	305	current working directory could have changed. Alternatively, you can
168	changed. Alternatively, you can make sure that you never change the	306	make sure that you never change the current working directory anywhere
169	current working directory.	307	in the program and then use relative paths. You can also take advantage
		308	of IO::AIOs working directory abstraction, that lets you specify paths
		309	relative to some previously-opened "working directory object" - see the
		310	description of the C<IO::AIO::WD> class later in this document.
170		311
171	To encode pathnames to byte form, either make sure you either: a)	312	To encode pathnames as octets, either make sure you either: a) always pass
172	always pass in filenames you got from outside (command line, readdir	313	in filenames you got from outside (command line, readdir etc.) without
173	etc.), b) are ASCII or ISO 8859-1, c) use the Encode module and encode	314	tinkering, b) are in your native filesystem encoding, c) use the Encode
174	your pathnames to the locale (or other) encoding in effect in the user	315	module and encode your pathnames to the locale (or other) encoding in
175	environment, d) use Glib::filename_from_unicode on unicode filenames or e)	316	effect in the user environment, d) use Glib::filename_from_unicode on
176	use something else.	317	unicode filenames or e) use something else to ensure your scalar has the
		318	correct contents.
		319
		320	This works, btw. independent of the internal UTF-8 bit, which IO::AIO
		321	handles correctly whether it is set or not.
		322
		323	=head2 AIO REQUEST FUNCTIONS
177		324
178	=over 4	325	=over 4
179		326
180	=item $prev_pri = aioreq_pri [$pri]	327	=item $prev_pri = aioreq_pri [$pri]
181		328
…		…
201	aio_read $_[0], ..., sub {	348	aio_read $_[0], ..., sub {
202	...	349	...
203	};	350	};
204	};	351	};
205		352
		353
206	=item aioreq_nice $pri_adjust	354	=item aioreq_nice $pri_adjust
207		355
208	Similar to C<aioreq_pri>, but subtracts the given value from the current	356	Similar to C<aioreq_pri>, but subtracts the given value from the current
209	priority, so effects are cumulative.	357	priority, so the effect is cumulative.
		358
210		359
211	=item aio_open $pathname, $flags, $mode, $callback->($fh)	360	=item aio_open $pathname, $flags, $mode, $callback->($fh)
212		361
213	Asynchronously open or create a file and call the callback with a newly	362	Asynchronously open or create a file and call the callback with a newly
214	created filehandle for the file.	363	created filehandle for the file.
…		…
220	list. They are the same as used by C<sysopen>.	369	list. They are the same as used by C<sysopen>.
221		370
222	Likewise, C<$mode> specifies the mode of the newly created file, if it	371	Likewise, C<$mode> specifies the mode of the newly created file, if it
223	didn't exist and C<O_CREAT> has been given, just like perl's C<sysopen>,	372	didn't exist and C<O_CREAT> has been given, just like perl's C<sysopen>,
224	except that it is mandatory (i.e. use C<0> if you don't create new files,	373	except that it is mandatory (i.e. use C<0> if you don't create new files,
225	and C<0666> or C<0777> if you do).	374	and C<0666> or C<0777> if you do). Note that the C<$mode> will be modified
		375	by the umask in effect then the request is being executed, so better never
		376	change the umask.
226		377
227	Example:	378	Example:
228		379
229	aio_open "/etc/passwd", O_RDONLY, 0, sub {	380	aio_open "/etc/passwd", IO::AIO::O_RDONLY, 0, sub {
230	if ($_[0]) {	381	if ($_[0]) {
231	print "open successful, fh is $_[0]\n";	382	print "open successful, fh is $_[0]\n";
232	...	383	...
233	} else {	384	} else {
234	die "open failed: $!\n";	385	die "open failed: $!\n";
235	}	386	}
236	};	387	};
237		388
		389	In addition to all the common open modes/flags (C<O_RDONLY>, C<O_WRONLY>,
		390	C<O_RDWR>, C<O_CREAT>, C<O_TRUNC>, C<O_EXCL> and C<O_APPEND>), the
		391	following POSIX and non-POSIX constants are available (missing ones on
		392	your system are, as usual, C<0>):
		393
		394	C<O_ASYNC>, C<O_DIRECT>, C<O_NOATIME>, C<O_CLOEXEC>, C<O_NOCTTY>, C<O_NOFOLLOW>,
		395	C<O_NONBLOCK>, C<O_EXEC>, C<O_SEARCH>, C<O_DIRECTORY>, C<O_DSYNC>,
		396	C<O_RSYNC>, C<O_SYNC> and C<O_TTY_INIT>.
		397
		398
238	=item aio_close $fh, $callback->($status)	399	=item aio_close $fh, $callback->($status)
239		400
240	Asynchronously close a file and call the callback with the result	401	Asynchronously close a file and call the callback with the result
241	code. I<WARNING:> although accepted, you should not pass in a perl	402	code.
242	filehandle here, as perl will likely close the file descriptor another
243	time when the filehandle is destroyed. Normally, you can safely call perls
244	C<close> or just let filehandles go out of scope.
245		403
246	This is supposed to be a bug in the API, so that might change. It's	404	Unfortunately, you can't do this to perl. Perl I<insists> very strongly on
247	therefore best to avoid this function.	405	closing the file descriptor associated with the filehandle itself.
		406
		407	Therefore, C<aio_close> will not close the filehandle - instead it will
		408	use dup2 to overwrite the file descriptor with the write-end of a pipe
		409	(the pipe fd will be created on demand and will be cached).
		410
		411	Or in other words: the file descriptor will be closed, but it will not be
		412	free for reuse until the perl filehandle is closed.
		413
		414	=cut
		415
		416	=item aio_seek $fh, $offset, $whence, $callback->($offs)
		417
		418	Seeks the filehandle to the new C<$offset>, similarly to perl's
		419	C<sysseek>. The C<$whence> can use the traditional values (C<0> for
		420	C<IO::AIO::SEEK_SET>, C<1> for C<IO::AIO::SEEK_CUR> or C<2> for
		421	C<IO::AIO::SEEK_END>).
		422
		423	The resulting absolute offset will be passed to the callback, or C<-1> in
		424	case of an error.
		425
		426	In theory, the C<$whence> constants could be different than the
		427	corresponding values from L<Fcntl>, but perl guarantees they are the same,
		428	so don't panic.
248		429
249	=item aio_read $fh,$offset,$length, $data,$dataoffset, $callback->($retval)	430	=item aio_read $fh,$offset,$length, $data,$dataoffset, $callback->($retval)
250		431
251	=item aio_write $fh,$offset,$length, $data,$dataoffset, $callback->($retval)	432	=item aio_write $fh,$offset,$length, $data,$dataoffset, $callback->($retval)
252		433
253	Reads or writes C<length> bytes from the specified C<fh> and C<offset>	434	Reads or writes C<$length> bytes from or to the specified C<$fh> and
254	into the scalar given by C<data> and offset C<dataoffset> and calls the	435	C<$offset> into the scalar given by C<$data> and offset C<$dataoffset>
255	callback without the actual number of bytes read (or -1 on error, just	436	and calls the callback without the actual number of bytes read (or -1 on
256	like the syscall).	437	error, just like the syscall).
		438
		439	C<aio_read> will, like C<sysread>, shrink or grow the C<$data> scalar to
		440	offset plus the actual number of bytes read.
		441
		442	If C<$offset> is undefined, then the current file descriptor offset will
		443	be used (and updated), otherwise the file descriptor offset will not be
		444	changed by these calls.
		445
		446	If C<$length> is undefined in C<aio_write>, use the remaining length of
		447	C<$data>.
		448
		449	If C<$dataoffset> is less than zero, it will be counted from the end of
		450	C<$data>.
257		451
258	The C<$data> scalar I<MUST NOT> be modified in any way while the request	452	The C<$data> scalar I<MUST NOT> be modified in any way while the request
259	is outstanding. Modifying it can result in segfaults or WW3 (if the	453	is outstanding. Modifying it can result in segfaults or World War III (if
260	necessary/optional hardware is installed).	454	the necessary/optional hardware is installed).
261		455
262	Example: Read 15 bytes at offset 7 into scalar C<$buffer>, starting at	456	Example: Read 15 bytes at offset 7 into scalar C<$buffer>, starting at
263	offset C<0> within the scalar:	457	offset C<0> within the scalar:
264		458
265	aio_read $fh, 7, 15, $buffer, 0, sub {	459	aio_read $fh, 7, 15, $buffer, 0, sub {
266	$_[0] > 0 or die "read error: $!";	460	$_[0] > 0 or die "read error: $!";
267	print "read $_[0] bytes: <$buffer>\n";	461	print "read $_[0] bytes: <$buffer>\n";
268	};	462	};
269		463
		464
270	=item aio_sendfile $out_fh, $in_fh, $in_offset, $length, $callback->($retval)	465	=item aio_sendfile $out_fh, $in_fh, $in_offset, $length, $callback->($retval)
271		466
272	Tries to copy C<$length> bytes from C<$in_fh> to C<$out_fh>. It starts	467	Tries to copy C<$length> bytes from C<$in_fh> to C<$out_fh>. It starts
273	reading at byte offset C<$in_offset>, and starts writing at the current	468	reading at byte offset C<$in_offset>, and starts writing at the current
274	file offset of C<$out_fh>. Because of that, it is not safe to issue more	469	file offset of C<$out_fh>. Because of that, it is not safe to issue more
275	than one C<aio_sendfile> per C<$out_fh>, as they will interfere with each	470	than one C<aio_sendfile> per C<$out_fh>, as they will interfere with each
276	other.	471	other. The same C<$in_fh> works fine though, as this function does not
		472	move or use the file offset of C<$in_fh>.
277		473
		474	Please note that C<aio_sendfile> can read more bytes from C<$in_fh> than
		475	are written, and there is no way to find out how many more bytes have been
		476	read from C<aio_sendfile> alone, as C<aio_sendfile> only provides the
		477	number of bytes written to C<$out_fh>. Only if the result value equals
		478	C<$length> one can assume that C<$length> bytes have been read.
		479
		480	Unlike with other C<aio_> functions, it makes a lot of sense to use
		481	C<aio_sendfile> on non-blocking sockets, as long as one end (typically
		482	the C<$in_fh>) is a file - the file I/O will then be asynchronous, while
		483	the socket I/O will be non-blocking. Note, however, that you can run
		484	into a trap where C<aio_sendfile> reads some data with readahead, then
		485	fails to write all data, and when the socket is ready the next time, the
		486	data in the cache is already lost, forcing C<aio_sendfile> to again hit
		487	the disk. Explicit C<aio_read> + C<aio_write> let's you better control
		488	resource usage.
		489
278	This call tries to make use of a native C<sendfile> syscall to provide	490	This call tries to make use of a native C<sendfile>-like syscall to
279	zero-copy operation. For this to work, C<$out_fh> should refer to a	491	provide zero-copy operation. For this to work, C<$out_fh> should refer to
280	socket, and C<$in_fh> should refer to mmap'able file.	492	a socket, and C<$in_fh> should refer to an mmap'able file.
281		493
282	If the native sendfile call fails or is not implemented, it will be	494	If a native sendfile cannot be found or it fails with C<ENOSYS>,
283	emulated, so you can call C<aio_sendfile> on any type of filehandle	495	C<EINVAL>, C<ENOTSUP>, C<EOPNOTSUPP>, C<EAFNOSUPPORT>, C<EPROTOTYPE> or
		496	C<ENOTSOCK>, it will be emulated, so you can call C<aio_sendfile> on any
284	regardless of the limitations of the operating system.	497	type of filehandle regardless of the limitations of the operating system.
285		498
286	Please note, however, that C<aio_sendfile> can read more bytes from	499	As native sendfile syscalls (as practically any non-POSIX interface hacked
287	C<$in_fh> than are written, and there is no way to find out how many	500	together in a hurry to improve benchmark numbers) tend to be rather buggy
288	bytes have been read from C<aio_sendfile> alone, as C<aio_sendfile> only	501	on many systems, this implementation tries to work around some known bugs
289	provides the number of bytes written to C<$out_fh>. Only if the result	502	in Linux and FreeBSD kernels (probably others, too), but that might fail,
290	value equals C<$length> one can assume that C<$length> bytes have been	503	so you really really should check the return value of C<aio_sendfile> -
291	read.	504	fewre bytes than expected might have been transferred.
		505
292		506
293	=item aio_readahead $fh,$offset,$length, $callback->($retval)	507	=item aio_readahead $fh,$offset,$length, $callback->($retval)
294		508
295	C<aio_readahead> populates the page cache with data from a file so that	509	C<aio_readahead> populates the page cache with data from a file so that
296	subsequent reads from that file will not block on disk I/O. The C<$offset>	510	subsequent reads from that file will not block on disk I/O. The C<$offset>
…		…
302	file. The current file offset of the file is left unchanged.	516	file. The current file offset of the file is left unchanged.
303		517
304	If that syscall doesn't exist (likely if your OS isn't Linux) it will be	518	If that syscall doesn't exist (likely if your OS isn't Linux) it will be
305	emulated by simply reading the data, which would have a similar effect.	519	emulated by simply reading the data, which would have a similar effect.
306		520
		521
307	=item aio_stat $fh_or_path, $callback->($status)	522	=item aio_stat $fh_or_path, $callback->($status)
308		523
309	=item aio_lstat $fh, $callback->($status)	524	=item aio_lstat $fh, $callback->($status)
310		525
311	Works like perl's C<stat> or C<lstat> in void context. The callback will	526	Works like perl's C<stat> or C<lstat> in void context. The callback will
…		…
316	for an explanation.	531	for an explanation.
317		532
318	Currently, the stats are always 64-bit-stats, i.e. instead of returning an	533	Currently, the stats are always 64-bit-stats, i.e. instead of returning an
319	error when stat'ing a large file, the results will be silently truncated	534	error when stat'ing a large file, the results will be silently truncated
320	unless perl itself is compiled with large file support.	535	unless perl itself is compiled with large file support.
		536
		537	To help interpret the mode and dev/rdev stat values, IO::AIO offers the
		538	following constants and functions (if not implemented, the constants will
		539	be C<0> and the functions will either C<croak> or fall back on traditional
		540	behaviour).
		541
		542	C<S_IFMT>, C<S_IFIFO>, C<S_IFCHR>, C<S_IFBLK>, C<S_IFLNK>, C<S_IFREG>,
		543	C<S_IFDIR>, C<S_IFWHT>, C<S_IFSOCK>, C<IO::AIO::major $dev_t>,
		544	C<IO::AIO::minor $dev_t>, C<IO::AIO::makedev $major, $minor>.
321		545
322	Example: Print the length of F</etc/passwd>:	546	Example: Print the length of F</etc/passwd>:
323		547
324	aio_stat "/etc/passwd", sub {	548	aio_stat "/etc/passwd", sub {
325	$_[0] and die "stat failed: $!";	549	$_[0] and die "stat failed: $!";
326	print "size is ", -s _, "\n";	550	print "size is ", -s _, "\n";
327	};	551	};
328		552
		553
		554	=item aio_statvfs $fh_or_path, $callback->($statvfs)
		555
		556	Works like the POSIX C<statvfs> or C<fstatvfs> syscalls, depending on
		557	whether a file handle or path was passed.
		558
		559	On success, the callback is passed a hash reference with the following
		560	members: C<bsize>, C<frsize>, C<blocks>, C<bfree>, C<bavail>, C<files>,
		561	C<ffree>, C<favail>, C<fsid>, C<flag> and C<namemax>. On failure, C<undef>
		562	is passed.
		563
		564	The following POSIX IO::AIO::ST_* constants are defined: C<ST_RDONLY> and
		565	C<ST_NOSUID>.
		566
		567	The following non-POSIX IO::AIO::ST_* flag masks are defined to
		568	their correct value when available, or to C<0> on systems that do
		569	not support them: C<ST_NODEV>, C<ST_NOEXEC>, C<ST_SYNCHRONOUS>,
		570	C<ST_MANDLOCK>, C<ST_WRITE>, C<ST_APPEND>, C<ST_IMMUTABLE>, C<ST_NOATIME>,
		571	C<ST_NODIRATIME> and C<ST_RELATIME>.
		572
		573	Example: stat C</wd> and dump out the data if successful.
		574
		575	aio_statvfs "/wd", sub {
		576	my $f = $_[0]
		577	or die "statvfs: $!";
		578
		579	use Data::Dumper;
		580	say Dumper $f;
		581	};
		582
		583	# result:
		584	{
		585	bsize => 1024,
		586	bfree => 4333064312,
		587	blocks => 10253828096,
		588	files => 2050765568,
		589	flag => 4096,
		590	favail => 2042092649,
		591	bavail => 4333064312,
		592	ffree => 2042092649,
		593	namemax => 255,
		594	frsize => 1024,
		595	fsid => 1810
		596	}
		597
		598
		599	=item aio_utime $fh_or_path, $atime, $mtime, $callback->($status)
		600
		601	Works like perl's C<utime> function (including the special case of $atime
		602	and $mtime being undef). Fractional times are supported if the underlying
		603	syscalls support them.
		604
		605	When called with a pathname, uses utimes(2) if available, otherwise
		606	utime(2). If called on a file descriptor, uses futimes(2) if available,
		607	otherwise returns ENOSYS, so this is not portable.
		608
		609	Examples:
		610
		611	# set atime and mtime to current time (basically touch(1)):
		612	aio_utime "path", undef, undef;
		613	# set atime to current time and mtime to beginning of the epoch:
		614	aio_utime "path", time, undef; # undef==0
		615
		616
		617	=item aio_chown $fh_or_path, $uid, $gid, $callback->($status)
		618
		619	Works like perl's C<chown> function, except that C<undef> for either $uid
		620	or $gid is being interpreted as "do not change" (but -1 can also be used).
		621
		622	Examples:
		623
		624	# same as "chown root path" in the shell:
		625	aio_chown "path", 0, -1;
		626	# same as above:
		627	aio_chown "path", 0, undef;
		628
		629
		630	=item aio_truncate $fh_or_path, $offset, $callback->($status)
		631
		632	Works like truncate(2) or ftruncate(2).
		633
		634
		635	=item aio_chmod $fh_or_path, $mode, $callback->($status)
		636
		637	Works like perl's C<chmod> function.
		638
		639
329	=item aio_unlink $pathname, $callback->($status)	640	=item aio_unlink $pathname, $callback->($status)
330		641
331	Asynchronously unlink (delete) a file and call the callback with the	642	Asynchronously unlink (delete) a file and call the callback with the
332	result code.	643	result code.
333		644
		645
334	=item aio_mknod $path, $mode, $dev, $callback->($status)	646	=item aio_mknod $pathname, $mode, $dev, $callback->($status)
		647
		648	[EXPERIMENTAL]
335		649
336	Asynchronously create a device node (or fifo). See mknod(2).	650	Asynchronously create a device node (or fifo). See mknod(2).
337		651
338	The only portable (POSIX) way of calling this function is:	652	The only (POSIX-) portable way of calling this function is:
339		653
340	aio_mknod $path, IO::AIO::S_IFIFO | $mode, 0, sub { ...	654	aio_mknod $pathname, IO::AIO::S_IFIFO | $mode, 0, sub { ...
		655
		656	See C<aio_stat> for info about some potentially helpful extra constants
		657	and functions.
341		658
342	=item aio_link $srcpath, $dstpath, $callback->($status)	659	=item aio_link $srcpath, $dstpath, $callback->($status)
343		660
344	Asynchronously create a new link to the existing object at C<$srcpath> at	661	Asynchronously create a new link to the existing object at C<$srcpath> at
345	the path C<$dstpath> and call the callback with the result code.	662	the path C<$dstpath> and call the callback with the result code.
346		663
		664
347	=item aio_symlink $srcpath, $dstpath, $callback->($status)	665	=item aio_symlink $srcpath, $dstpath, $callback->($status)
348		666
349	Asynchronously create a new symbolic link to the existing object at C<$srcpath> at	667	Asynchronously create a new symbolic link to the existing object at C<$srcpath> at
350	the path C<$dstpath> and call the callback with the result code.	668	the path C<$dstpath> and call the callback with the result code.
351		669
		670
		671	=item aio_readlink $pathname, $callback->($link)
		672
		673	Asynchronously read the symlink specified by C<$path> and pass it to
		674	the callback. If an error occurs, nothing or undef gets passed to the
		675	callback.
		676
		677
		678	=item aio_realpath $pathname, $callback->($path)
		679
		680	Asynchronously make the path absolute and resolve any symlinks in
		681	C<$path>. The resulting path only consists of directories (Same as
		682	L<Cwd::realpath>).
		683
		684	This request can be used to get the absolute path of the current working
		685	directory by passing it a path of F<.> (a single dot).
		686
		687
352	=item aio_rename $srcpath, $dstpath, $callback->($status)	688	=item aio_rename $srcpath, $dstpath, $callback->($status)
353		689
354	Asynchronously rename the object at C<$srcpath> to C<$dstpath>, just as	690	Asynchronously rename the object at C<$srcpath> to C<$dstpath>, just as
355	rename(2) and call the callback with the result code.	691	rename(2) and call the callback with the result code.
356		692
		693
		694	=item aio_mkdir $pathname, $mode, $callback->($status)
		695
		696	Asynchronously mkdir (create) a directory and call the callback with
		697	the result code. C<$mode> will be modified by the umask at the time the
		698	request is executed, so do not change your umask.
		699
		700
357	=item aio_rmdir $pathname, $callback->($status)	701	=item aio_rmdir $pathname, $callback->($status)
358		702
359	Asynchronously rmdir (delete) a directory and call the callback with the	703	Asynchronously rmdir (delete) a directory and call the callback with the
360	result code.	704	result code.
		705
361		706
362	=item aio_readdir $pathname, $callback->($entries)	707	=item aio_readdir $pathname, $callback->($entries)
363		708
364	Unlike the POSIX call of the same name, C<aio_readdir> reads an entire	709	Unlike the POSIX call of the same name, C<aio_readdir> reads an entire
365	directory (i.e. opendir + readdir + closedir). The entries will not be	710	directory (i.e. opendir + readdir + closedir). The entries will not be
366	sorted, and will B<NOT> include the C<.> and C<..> entries.	711	sorted, and will B<NOT> include the C<.> and C<..> entries.
367		712
368	The callback a single argument which is either C<undef> or an array-ref	713	The callback is passed a single argument which is either C<undef> or an
369	with the filenames.	714	array-ref with the filenames.
		715
		716
		717	=item aio_readdirx $pathname, $flags, $callback->($entries, $flags)
		718
		719	Quite similar to C<aio_readdir>, but the C<$flags> argument allows one to
		720	tune behaviour and output format. In case of an error, C<$entries> will be
		721	C<undef>.
		722
		723	The flags are a combination of the following constants, ORed together (the
		724	flags will also be passed to the callback, possibly modified):
		725
		726	=over 4
		727
		728	=item IO::AIO::READDIR_DENTS
		729
		730	When this flag is off, then the callback gets an arrayref consisting of
		731	names only (as with C<aio_readdir>), otherwise it gets an arrayref with
		732	C<[$name, $type, $inode]> arrayrefs, each describing a single directory
		733	entry in more detail.
		734
		735	C<$name> is the name of the entry.
		736
		737	C<$type> is one of the C<IO::AIO::DT_xxx> constants:
		738
		739	C<IO::AIO::DT_UNKNOWN>, C<IO::AIO::DT_FIFO>, C<IO::AIO::DT_CHR>, C<IO::AIO::DT_DIR>,
		740	C<IO::AIO::DT_BLK>, C<IO::AIO::DT_REG>, C<IO::AIO::DT_LNK>, C<IO::AIO::DT_SOCK>,
		741	C<IO::AIO::DT_WHT>.
		742
		743	C<IO::AIO::DT_UNKNOWN> means just that: readdir does not know. If you need to
		744	know, you have to run stat yourself. Also, for speed reasons, the C<$type>
		745	scalars are read-only: you can not modify them.
		746
		747	C<$inode> is the inode number (which might not be exact on systems with 64
		748	bit inode numbers and 32 bit perls). This field has unspecified content on
		749	systems that do not deliver the inode information.
		750
		751	=item IO::AIO::READDIR_DIRS_FIRST
		752
		753	When this flag is set, then the names will be returned in an order where
		754	likely directories come first, in optimal stat order. This is useful when
		755	you need to quickly find directories, or you want to find all directories
		756	while avoiding to stat() each entry.
		757
		758	If the system returns type information in readdir, then this is used
		759	to find directories directly. Otherwise, likely directories are names
		760	beginning with ".", or otherwise names with no dots, of which names with
		761	short names are tried first.
		762
		763	=item IO::AIO::READDIR_STAT_ORDER
		764
		765	When this flag is set, then the names will be returned in an order
		766	suitable for stat()'ing each one. That is, when you plan to stat()
		767	all files in the given directory, then the returned order will likely
		768	be fastest.
		769
		770	If both this flag and C<IO::AIO::READDIR_DIRS_FIRST> are specified, then
		771	the likely dirs come first, resulting in a less optimal stat order.
		772
		773	=item IO::AIO::READDIR_FOUND_UNKNOWN
		774
		775	This flag should not be set when calling C<aio_readdirx>. Instead, it
		776	is being set by C<aio_readdirx>, when any of the C<$type>'s found were
		777	C<IO::AIO::DT_UNKNOWN>. The absence of this flag therefore indicates that all
		778	C<$type>'s are known, which can be used to speed up some algorithms.
		779
		780	=back
		781
		782
		783	=item aio_load $pathname, $data, $callback->($status)
		784
		785	This is a composite request that tries to fully load the given file into
		786	memory. Status is the same as with aio_read.
		787
		788	=cut
		789
		790	sub aio_load($$;$) {
		791	my ($path, undef, $cb) = @_;
		792	my $data = \$_[1];
		793
		794	my $pri = aioreq_pri;
		795	my $grp = aio_group $cb;
		796
		797	aioreq_pri $pri;
		798	add $grp aio_open $path, O_RDONLY, 0, sub {
		799	my $fh = shift
		800	or return $grp->result (-1);
		801
		802	aioreq_pri $pri;
		803	add $grp aio_read $fh, 0, (-s $fh), $$data, 0, sub {
		804	$grp->result ($_[0]);
		805	};
		806	};
		807
		808	$grp
		809	}
370		810
371	=item aio_copy $srcpath, $dstpath, $callback->($status)	811	=item aio_copy $srcpath, $dstpath, $callback->($status)
372		812
373	Try to copy the I<file> (directories not supported as either source or	813	Try to copy the I<file> (directories not supported as either source or
374	destination) from C<$srcpath> to C<$dstpath> and call the callback with	814	destination) from C<$srcpath> to C<$dstpath> and call the callback with
375	the C<0> (error) or C<-1> ok.	815	a status of C<0> (ok) or C<-1> (error, see C<$!>).
376		816
377	This is a composite request that it creates the destination file with	817	This is a composite request that creates the destination file with
378	mode 0200 and copies the contents of the source file into it using	818	mode 0200 and copies the contents of the source file into it using
379	C<aio_sendfile>, followed by restoring atime, mtime, access mode and	819	C<aio_sendfile>, followed by restoring atime, mtime, access mode and
380	uid/gid, in that order.	820	uid/gid, in that order.
381		821
382	If an error occurs, the partial destination file will be unlinked, if	822	If an error occurs, the partial destination file will be unlinked, if
…		…
392	my $grp = aio_group $cb;	832	my $grp = aio_group $cb;
393		833
394	aioreq_pri $pri;	834	aioreq_pri $pri;
395	add $grp aio_open $src, O_RDONLY, 0, sub {	835	add $grp aio_open $src, O_RDONLY, 0, sub {
396	if (my $src_fh = $_[0]) {	836	if (my $src_fh = $_[0]) {
397	my @stat = stat $src_fh;	837	my @stat = stat $src_fh; # hmm, might block over nfs?
398		838
399	aioreq_pri $pri;	839	aioreq_pri $pri;
400	add $grp aio_open $dst, O_CREAT | O_WRONLY | O_TRUNC, 0200, sub {	840	add $grp aio_open $dst, O_CREAT | O_WRONLY | O_TRUNC, 0200, sub {
401	if (my $dst_fh = $_[0]) {	841	if (my $dst_fh = $_[0]) {
402	aioreq_pri $pri;	842	aioreq_pri $pri;
403	add $grp aio_sendfile $dst_fh, $src_fh, 0, $stat[7], sub {	843	add $grp aio_sendfile $dst_fh, $src_fh, 0, $stat[7], sub {
404	if ($_[0] == $stat[7]) {	844	if ($_[0] == $stat[7]) {
405	$grp->result (0);	845	$grp->result (0);
406	close $src_fh;	846	close $src_fh;
407		847
408	# those should not normally block. should. should.	848	my $ch = sub {
		849	aioreq_pri $pri;
		850	add $grp aio_chmod $dst_fh, $stat[2] & 07777, sub {
		851	aioreq_pri $pri;
		852	add $grp aio_chown $dst_fh, $stat[4], $stat[5], sub {
		853	aioreq_pri $pri;
		854	add $grp aio_close $dst_fh;
		855	}
		856	};
		857	};
		858
		859	aioreq_pri $pri;
		860	add $grp aio_utime $dst_fh, $stat[8], $stat[9], sub {
		861	if ($_[0] < 0 && $! == ENOSYS) {
		862	aioreq_pri $pri;
409	utime $stat[8], $stat[9], $dst;	863	add $grp aio_utime $dst, $stat[8], $stat[9], $ch;
410	chmod $stat[2] & 07777, $dst_fh;	864	} else {
411	chown $stat[4], $stat[5], $dst_fh;	865	$ch->();
		866	}
412	close $dst_fh;	867	};
413	} else {	868	} else {
414	$grp->result (-1);	869	$grp->result (-1);
415	close $src_fh;	870	close $src_fh;
416	close $dst_fh;	871	close $dst_fh;
417		872
…		…
434		889
435	=item aio_move $srcpath, $dstpath, $callback->($status)	890	=item aio_move $srcpath, $dstpath, $callback->($status)
436		891
437	Try to move the I<file> (directories not supported as either source or	892	Try to move the I<file> (directories not supported as either source or
438	destination) from C<$srcpath> to C<$dstpath> and call the callback with	893	destination) from C<$srcpath> to C<$dstpath> and call the callback with
439	the C<0> (error) or C<-1> ok.	894	a status of C<0> (ok) or C<-1> (error, see C<$!>).
440		895
441	This is a composite request that tries to rename(2) the file first. If	896	This is a composite request that tries to rename(2) the file first; if
442	rename files with C<EXDEV>, it copies the file with C<aio_copy> and, if	897	rename fails with C<EXDEV>, it copies the file with C<aio_copy> and, if
443	that is successful, unlinking the C<$srcpath>.	898	that is successful, unlinks the C<$srcpath>.
444		899
445	=cut	900	=cut
446		901
447	sub aio_move($$;$) {	902	sub aio_move($$;$) {
448	my ($src, $dst, $cb) = @_;	903	my ($src, $dst, $cb) = @_;
…		…
455	if ($_[0] && $! == EXDEV) {	910	if ($_[0] && $! == EXDEV) {
456	aioreq_pri $pri;	911	aioreq_pri $pri;
457	add $grp aio_copy $src, $dst, sub {	912	add $grp aio_copy $src, $dst, sub {
458	$grp->result ($_[0]);	913	$grp->result ($_[0]);
459		914
460	if (!$_[0]) {	915	unless ($_[0]) {
461	aioreq_pri $pri;	916	aioreq_pri $pri;
462	add $grp aio_unlink $src;	917	add $grp aio_unlink $src;
463	}	918	}
464	};	919	};
465	} else {	920	} else {
…		…
468	};	923	};
469		924
470	$grp	925	$grp
471	}	926	}
472		927
473	=item aio_scandir $path, $maxreq, $callback->($dirs, $nondirs)	928	=item aio_scandir $pathname, $maxreq, $callback->($dirs, $nondirs)
474		929
475	Scans a directory (similar to C<aio_readdir>) but additionally tries to	930	Scans a directory (similar to C<aio_readdir>) but additionally tries to
476	efficiently separate the entries of directory C<$path> into two sets of	931	efficiently separate the entries of directory C<$path> into two sets of
477	names, directories you can recurse into (directories), and ones you cannot	932	names, directories you can recurse into (directories), and ones you cannot
478	recurse into (everything else, including symlinks to directories).	933	recurse into (everything else, including symlinks to directories).
…		…
495		950
496	Implementation notes.	951	Implementation notes.
497		952
498	The C<aio_readdir> cannot be avoided, but C<stat()>'ing every entry can.	953	The C<aio_readdir> cannot be avoided, but C<stat()>'ing every entry can.
499		954
		955	If readdir returns file type information, then this is used directly to
		956	find directories.
		957
500	After reading the directory, the modification time, size etc. of the	958	Otherwise, after reading the directory, the modification time, size etc.
501	directory before and after the readdir is checked, and if they match (and	959	of the directory before and after the readdir is checked, and if they
502	isn't the current time), the link count will be used to decide how many	960	match (and isn't the current time), the link count will be used to decide
503	entries are directories (if >= 2). Otherwise, no knowledge of the number	961	how many entries are directories (if >= 2). Otherwise, no knowledge of the
504	of subdirectories will be assumed.	962	number of subdirectories will be assumed.
505		963
506	Then entries will be sorted into likely directories (everything without	964	Then entries will be sorted into likely directories a non-initial dot
507	a non-initial dot currently) and likely non-directories (everything	965	currently) and likely non-directories (see C<aio_readdirx>). Then every
508	else). Then every entry plus an appended C</.> will be C<stat>'ed,	966	entry plus an appended C</.> will be C<stat>'ed, likely directories first,
509	likely directories first. If that succeeds, it assumes that the entry	967	in order of their inode numbers. If that succeeds, it assumes that the
510	is a directory or a symlink to directory (which will be checked	968	entry is a directory or a symlink to directory (which will be checked
511	seperately). This is often faster than stat'ing the entry itself because	969	separately). This is often faster than stat'ing the entry itself because
512	filesystems might detect the type of the entry without reading the inode	970	filesystems might detect the type of the entry without reading the inode
513	data (e.g. ext2fs filetype feature).	971	data (e.g. ext2fs filetype feature), even on systems that cannot return
		972	the filetype information on readdir.
514		973
515	If the known number of directories (link count - 2) has been reached, the	974	If the known number of directories (link count - 2) has been reached, the
516	rest of the entries is assumed to be non-directories.	975	rest of the entries is assumed to be non-directories.
517		976
518	This only works with certainty on POSIX (= UNIX) filesystems, which	977	This only works with certainty on POSIX (= UNIX) filesystems, which
…		…
522	as those tend to return 0 or 1 as link counts, which disables the	981	as those tend to return 0 or 1 as link counts, which disables the
523	directory counting heuristic.	982	directory counting heuristic.
524		983
525	=cut	984	=cut
526		985
527	sub aio_scandir($$$) {	986	sub aio_scandir($$;$) {
528	my ($path, $maxreq, $cb) = @_;	987	my ($path, $maxreq, $cb) = @_;
529		988
530	my $pri = aioreq_pri;	989	my $pri = aioreq_pri;
531		990
532	my $grp = aio_group $cb;	991	my $grp = aio_group $cb;
533		992
534	$maxreq = 4 if $maxreq <= 0;	993	$maxreq = 4 if $maxreq <= 0;
535		994
536	# stat once	995	# get a wd object
537	aioreq_pri $pri;	996	aioreq_pri $pri;
538	add $grp aio_stat $path, sub {	997	add $grp aio_wd $path, sub {
		998	$_[0]
539	return $grp->result () if $_[0];	999	or return $grp->result ();
540	my $now = time;
541	my $hash1 = join ":", (stat _)[0,1,3,7,9];
542		1000
543	# read the directory entries	1001	my $wd = [shift, "."];
		1002
		1003	# stat once
544	aioreq_pri $pri;	1004	aioreq_pri $pri;
545	add $grp aio_readdir $path, sub {	1005	add $grp aio_stat $wd, sub {
546	my $entries = shift
547	or return $grp->result ();	1006	return $grp->result () if $_[0];
		1007	my $now = time;
		1008	my $hash1 = join ":", (stat _)[0,1,3,7,9];
548		1009
549	# stat the dir another time	1010	# read the directory entries
550	aioreq_pri $pri;	1011	aioreq_pri $pri;
		1012	add $grp aio_readdirx $wd, READDIR_DIRS_FIRST, sub {
		1013	my $entries = shift
		1014	or return $grp->result ();
		1015
		1016	# stat the dir another time
		1017	aioreq_pri $pri;
551	add $grp aio_stat $path, sub {	1018	add $grp aio_stat $wd, sub {
552	my $hash2 = join ":", (stat _)[0,1,3,7,9];	1019	my $hash2 = join ":", (stat _)[0,1,3,7,9];
553		1020
554	my $ndirs;	1021	my $ndirs;
555		1022
556	# take the slow route if anything looks fishy	1023	# take the slow route if anything looks fishy
557	if ($hash1 ne $hash2 or (stat _)[9] == $now) {	1024	if ($hash1 ne $hash2 or (stat _)[9] == $now) {
558	$ndirs = -1;	1025	$ndirs = -1;
559	} else {	1026	} else {
560	# if nlink == 2, we are finished	1027	# if nlink == 2, we are finished
561	# on non-posix-fs's, we rely on nlink < 2	1028	# for non-posix-fs's, we rely on nlink < 2
562	$ndirs = (stat _)[3] - 2	1029	$ndirs = (stat _)[3] - 2
563	or return $grp->result ([], $entries);	1030	or return $grp->result ([], $entries);
564	}	1031	}
565		1032
566	# sort into likely dirs and likely nondirs
567	# dirs == files without ".", short entries first
568	$entries = [map $_->[0],
569	sort { $b->[1] cmp $a->[1] }
570	map [$_, sprintf "%s%04d", (/.\./ ? "1" : "0"), length],
571	@$entries];
572
573	my (@dirs, @nondirs);	1033	my (@dirs, @nondirs);
574		1034
575	my $statgrp = add $grp aio_group sub {	1035	my $statgrp = add $grp aio_group sub {
576	$grp->result (\@dirs, \@nondirs);	1036	$grp->result (\@dirs, \@nondirs);
577	};	1037	};
578		1038
579	limit $statgrp $maxreq;	1039	limit $statgrp $maxreq;
580	feed $statgrp sub {	1040	feed $statgrp sub {
581	return unless @$entries;	1041	return unless @$entries;
582	my $entry = pop @$entries;	1042	my $entry = shift @$entries;
583		1043
584	aioreq_pri $pri;	1044	aioreq_pri $pri;
		1045	$wd->[1] = "$entry/.";
585	add $statgrp aio_stat "$path/$entry/.", sub {	1046	add $statgrp aio_stat $wd, sub {
586	if ($_[0] < 0) {	1047	if ($_[0] < 0) {
587	push @nondirs, $entry;	1048	push @nondirs, $entry;
588	} else {	1049	} else {
589	# need to check for real directory	1050	# need to check for real directory
590	aioreq_pri $pri;	1051	aioreq_pri $pri;
		1052	$wd->[1] = $entry;
591	add $statgrp aio_lstat "$path/$entry", sub {	1053	add $statgrp aio_lstat $wd, sub {
592	if (-d _) {	1054	if (-d _) {
593	push @dirs, $entry;	1055	push @dirs, $entry;
594		1056
595	unless (--$ndirs) {	1057	unless (--$ndirs) {
596	push @nondirs, @$entries;	1058	push @nondirs, @$entries;
597	feed $statgrp;	1059	feed $statgrp;
		1060	}
		1061	} else {
		1062	push @nondirs, $entry;
598	}	1063	}
599	} else {
600	push @nondirs, $entry;
601	}	1064	}
602	}	1065	}
603	}	1066	};
604	};	1067	};
605	};	1068	};
606	};	1069	};
607	};	1070	};
608	};	1071	};
609		1072
610	$grp	1073	$grp
611	}	1074	}
612		1075
		1076	=item aio_rmtree $pathname, $callback->($status)
		1077
		1078	Delete a directory tree starting (and including) C<$path>, return the
		1079	status of the final C<rmdir> only. This is a composite request that
		1080	uses C<aio_scandir> to recurse into and rmdir directories, and unlink
		1081	everything else.
		1082
		1083	=cut
		1084
		1085	sub aio_rmtree;
		1086	sub aio_rmtree($;$) {
		1087	my ($path, $cb) = @_;
		1088
		1089	my $pri = aioreq_pri;
		1090	my $grp = aio_group $cb;
		1091
		1092	aioreq_pri $pri;
		1093	add $grp aio_scandir $path, 0, sub {
		1094	my ($dirs, $nondirs) = @_;
		1095
		1096	my $dirgrp = aio_group sub {
		1097	add $grp aio_rmdir $path, sub {
		1098	$grp->result ($_[0]);
		1099	};
		1100	};
		1101
		1102	(aioreq_pri $pri), add $dirgrp aio_rmtree "$path/$_" for @$dirs;
		1103	(aioreq_pri $pri), add $dirgrp aio_unlink "$path/$_" for @$nondirs;
		1104
		1105	add $grp $dirgrp;
		1106	};
		1107
		1108	$grp
		1109	}
		1110
		1111	=item aio_sync $callback->($status)
		1112
		1113	Asynchronously call sync and call the callback when finished.
		1114
613	=item aio_fsync $fh, $callback->($status)	1115	=item aio_fsync $fh, $callback->($status)
614		1116
615	Asynchronously call fsync on the given filehandle and call the callback	1117	Asynchronously call fsync on the given filehandle and call the callback
616	with the fsync result code.	1118	with the fsync result code.
617		1119
…		…
620	Asynchronously call fdatasync on the given filehandle and call the	1122	Asynchronously call fdatasync on the given filehandle and call the
621	callback with the fdatasync result code.	1123	callback with the fdatasync result code.
622		1124
623	If this call isn't available because your OS lacks it or it couldn't be	1125	If this call isn't available because your OS lacks it or it couldn't be
624	detected, it will be emulated by calling C<fsync> instead.	1126	detected, it will be emulated by calling C<fsync> instead.
		1127
		1128	=item aio_syncfs $fh, $callback->($status)
		1129
		1130	Asynchronously call the syncfs syscall to sync the filesystem associated
		1131	to the given filehandle and call the callback with the syncfs result
		1132	code. If syncfs is not available, calls sync(), but returns C<-1> and sets
		1133	errno to C<ENOSYS> nevertheless.
		1134
		1135	=item aio_sync_file_range $fh, $offset, $nbytes, $flags, $callback->($status)
		1136
		1137	Sync the data portion of the file specified by C<$offset> and C<$length>
		1138	to disk (but NOT the metadata), by calling the Linux-specific
		1139	sync_file_range call. If sync_file_range is not available or it returns
		1140	ENOSYS, then fdatasync or fsync is being substituted.
		1141
		1142	C<$flags> can be a combination of C<IO::AIO::SYNC_FILE_RANGE_WAIT_BEFORE>,
		1143	C<IO::AIO::SYNC_FILE_RANGE_WRITE> and
		1144	C<IO::AIO::SYNC_FILE_RANGE_WAIT_AFTER>: refer to the sync_file_range
		1145	manpage for details.
		1146
		1147	=item aio_pathsync $pathname, $callback->($status)
		1148
		1149	This request tries to open, fsync and close the given path. This is a
		1150	composite request intended to sync directories after directory operations
		1151	(E.g. rename). This might not work on all operating systems or have any
		1152	specific effect, but usually it makes sure that directory changes get
		1153	written to disc. It works for anything that can be opened for read-only,
		1154	not just directories.
		1155
		1156	Future versions of this function might fall back to other methods when
		1157	C<fsync> on the directory fails (such as calling C<sync>).
		1158
		1159	Passes C<0> when everything went ok, and C<-1> on error.
		1160
		1161	=cut
		1162
		1163	sub aio_pathsync($;$) {
		1164	my ($path, $cb) = @_;
		1165
		1166	my $pri = aioreq_pri;
		1167	my $grp = aio_group $cb;
		1168
		1169	aioreq_pri $pri;
		1170	add $grp aio_open $path, O_RDONLY, 0, sub {
		1171	my ($fh) = @_;
		1172	if ($fh) {
		1173	aioreq_pri $pri;
		1174	add $grp aio_fsync $fh, sub {
		1175	$grp->result ($_[0]);
		1176
		1177	aioreq_pri $pri;
		1178	add $grp aio_close $fh;
		1179	};
		1180	} else {
		1181	$grp->result (-1);
		1182	}
		1183	};
		1184
		1185	$grp
		1186	}
		1187
		1188	=item aio_msync $scalar, $offset = 0, $length = undef, flags = 0, $callback->($status)
		1189
		1190	This is a rather advanced IO::AIO call, which only works on mmap(2)ed
		1191	scalars (see the C<IO::AIO::mmap> function, although it also works on data
		1192	scalars managed by the L<Sys::Mmap> or L<Mmap> modules, note that the
		1193	scalar must only be modified in-place while an aio operation is pending on
		1194	it).
		1195
		1196	It calls the C<msync> function of your OS, if available, with the memory
		1197	area starting at C<$offset> in the string and ending C<$length> bytes
		1198	later. If C<$length> is negative, counts from the end, and if C<$length>
		1199	is C<undef>, then it goes till the end of the string. The flags can be
		1200	a combination of C<IO::AIO::MS_ASYNC>, C<IO::AIO::MS_INVALIDATE> and
		1201	C<IO::AIO::MS_SYNC>.
		1202
		1203	=item aio_mtouch $scalar, $offset = 0, $length = undef, flags = 0, $callback->($status)
		1204
		1205	This is a rather advanced IO::AIO call, which works best on mmap(2)ed
		1206	scalars.
		1207
		1208	It touches (reads or writes) all memory pages in the specified
		1209	range inside the scalar. All caveats and parameters are the same
		1210	as for C<aio_msync>, above, except for flags, which must be either
		1211	C<0> (which reads all pages and ensures they are instantiated) or
		1212	C<IO::AIO::MT_MODIFY>, which modifies the memory page s(by reading and
		1213	writing an octet from it, which dirties the page).
		1214
		1215	=item aio_mlock $scalar, $offset = 0, $length = undef, $callback->($status)
		1216
		1217	This is a rather advanced IO::AIO call, which works best on mmap(2)ed
		1218	scalars.
		1219
		1220	It reads in all the pages of the underlying storage into memory (if any)
		1221	and locks them, so they are not getting swapped/paged out or removed.
		1222
		1223	If C<$length> is undefined, then the scalar will be locked till the end.
		1224
		1225	On systems that do not implement C<mlock>, this function returns C<-1>
		1226	and sets errno to C<ENOSYS>.
		1227
		1228	Note that the corresponding C<munlock> is synchronous and is
		1229	documented under L<MISCELLANEOUS FUNCTIONS>.
		1230
		1231	Example: open a file, mmap and mlock it - both will be undone when
		1232	C<$data> gets destroyed.
		1233
		1234	open my $fh, "<", $path or die "$path: $!";
		1235	my $data;
		1236	IO::AIO::mmap $data, -s $fh, IO::AIO::PROT_READ, IO::AIO::MAP_SHARED, $fh;
		1237	aio_mlock $data; # mlock in background
		1238
		1239	=item aio_mlockall $flags, $callback->($status)
		1240
		1241	Calls the C<mlockall> function with the given C<$flags> (a combination of
		1242	C<IO::AIO::MCL_CURRENT> and C<IO::AIO::MCL_FUTURE>).
		1243
		1244	On systems that do not implement C<mlockall>, this function returns C<-1>
		1245	and sets errno to C<ENOSYS>.
		1246
		1247	Note that the corresponding C<munlockall> is synchronous and is
		1248	documented under L<MISCELLANEOUS FUNCTIONS>.
		1249
		1250	Example: asynchronously lock all current and future pages into memory.
		1251
		1252	aio_mlockall IO::AIO::MCL_FUTURE;
		1253
		1254	=item aio_fiemap $fh, $start, $length, $flags, $count, $cb->(\@extents)
		1255
		1256	Queries the extents of the given file (by calling the Linux FIEMAP ioctl,
		1257	see L<http://cvs.schmorp.de/IO-AIO/doc/fiemap.txt> for details). If the
		1258	C<ioctl> is not available on your OS, then this rquiest will fail with
		1259	C<ENOSYS>.
		1260
		1261	C<$start> is the starting offset to query extents for, C<$length> is the
		1262	size of the range to query - if it is C<undef>, then the whole file will
		1263	be queried.
		1264
		1265	C<$flags> is a combination of flags (C<IO::AIO::FIEMAP_FLAG_SYNC> or
		1266	C<IO::AIO::FIEMAP_FLAG_XATTR> - C<IO::AIO::FIEMAP_FLAGS_COMPAT> is also
		1267	exported), and is normally C<0> or C<IO::AIO::FIEMAP_FLAG_SYNC> to query
		1268	the data portion.
		1269
		1270	C<$count> is the maximum number of extent records to return. If it is
		1271	C<undef>, then IO::AIO queries all extents of the file. As a very special
		1272	case, if it is C<0>, then the callback receives the number of extents
		1273	instead of the extents themselves.
		1274
		1275	If an error occurs, the callback receives no arguments. The special
		1276	C<errno> value C<IO::AIO::EBADR> is available to test for flag errors.
		1277
		1278	Otherwise, the callback receives an array reference with extent
		1279	structures. Each extent structure is an array reference itself, with the
		1280	following members:
		1281
		1282	[$logical, $physical, $length, $flags]
		1283
		1284	Flags is any combination of the following flag values (typically either C<0>
		1285	or C<IO::AIO::FIEMAP_EXTENT_LAST>):
		1286
		1287	C<IO::AIO::FIEMAP_EXTENT_LAST>, C<IO::AIO::FIEMAP_EXTENT_UNKNOWN>,
		1288	C<IO::AIO::FIEMAP_EXTENT_DELALLOC>, C<IO::AIO::FIEMAP_EXTENT_ENCODED>,
		1289	C<IO::AIO::FIEMAP_EXTENT_DATA_ENCRYPTED>, C<IO::AIO::FIEMAP_EXTENT_NOT_ALIGNED>,
		1290	C<IO::AIO::FIEMAP_EXTENT_DATA_INLINE>, C<IO::AIO::FIEMAP_EXTENT_DATA_TAIL>,
		1291	C<IO::AIO::FIEMAP_EXTENT_UNWRITTEN>, C<IO::AIO::FIEMAP_EXTENT_MERGED> or
		1292	C<IO::AIO::FIEMAP_EXTENT_SHARED>.
625		1293
626	=item aio_group $callback->(...)	1294	=item aio_group $callback->(...)
627		1295
628	This is a very special aio request: Instead of doing something, it is a	1296	This is a very special aio request: Instead of doing something, it is a
629	container for other aio requests, which is useful if you want to bundle	1297	container for other aio requests, which is useful if you want to bundle
…		…
667	immense (it blocks a thread for a long time) so do not use this function	1335	immense (it blocks a thread for a long time) so do not use this function
668	except to put your application under artificial I/O pressure.	1336	except to put your application under artificial I/O pressure.
669		1337
670	=back	1338	=back
671		1339
		1340
		1341	=head2 IO::AIO::WD - multiple working directories
		1342
		1343	Your process only has one current working directory, which is used by all
		1344	threads. This makes it hard to use relative paths (some other component
		1345	could call C<chdir> at any time, and it is hard to control when the path
		1346	will be used by IO::AIO).
		1347
		1348	One solution for this is to always use absolute paths. This usually works,
		1349	but can be quite slow (the kernel has to walk the whole path on every
		1350	access), and can also be a hassle to implement.
		1351
		1352	Newer POSIX systems have a number of functions (openat, fdopendir,
		1353	futimensat and so on) that make it possible to specify working directories
		1354	per operation.
		1355
		1356	For portability, and because the clowns who "designed", or shall I write,
		1357	perpetrated this new interface were obviously half-drunk, this abstraction
		1358	cannot be perfect, though.
		1359
		1360	IO::AIO allows you to convert directory paths into a so-called IO::AIO::WD
		1361	object. This object stores the canonicalised, absolute version of the
		1362	path, and on systems that allow it, also a directory file descriptor.
		1363
		1364	Everywhere where a pathname is accepted by IO::AIO (e.g. in C<aio_stat>
		1365	or C<aio_unlink>), one can specify an array reference with an IO::AIO::WD
		1366	object and a pathname instead (or the IO::AIO::WD object alone, which
		1367	gets interpreted as C<[$wd, "."]>). If the pathname is absolute, the
		1368	IO::AIO::WD object is ignored, otherwise the pathname is resolved relative
		1369	to that IO::AIO::WD object.
		1370
		1371	For example, to get a wd object for F</etc> and then stat F<passwd>
		1372	inside, you would write:
		1373
		1374	aio_wd "/etc", sub {
		1375	my $etcdir = shift;
		1376
		1377	# although $etcdir can be undef on error, there is generally no reason
		1378	# to check for errors here, as aio_stat will fail with ENOENT
		1379	# when $etcdir is undef.
		1380
		1381	aio_stat [$etcdir, "passwd"], sub {
		1382	# yay
		1383	};
		1384	};
		1385
		1386	That C<aio_wd> is a request and not a normal function shows that creating
		1387	an IO::AIO::WD object is itself a potentially blocking operation, which is
		1388	why it is done asynchronously.
		1389
		1390	To stat the directory obtained with C<aio_wd> above, one could write
		1391	either of the following three request calls:
		1392
		1393	aio_lstat "/etc" , sub { ... # pathname as normal string
		1394	aio_lstat [$wd, "."], sub { ... # "." relative to $wd (i.e. $wd itself)
		1395	aio_lstat $wd , sub { ... # shorthand for the previous
		1396
		1397	As with normal pathnames, IO::AIO keeps a copy of the working directory
		1398	object and the pathname string, so you could write the following without
		1399	causing any issues due to C<$path> getting reused:
		1400
		1401	my $path = [$wd, undef];
		1402
		1403	for my $name (qw(abc def ghi)) {
		1404	$path->[1] = $name;
		1405	aio_stat $path, sub {
		1406	# ...
		1407	};
		1408	}
		1409
		1410	There are some caveats: when directories get renamed (or deleted), the
		1411	pathname string doesn't change, so will point to the new directory (or
		1412	nowhere at all), while the directory fd, if available on the system,
		1413	will still point to the original directory. Most functions accepting a
		1414	pathname will use the directory fd on newer systems, and the string on
		1415	older systems. Some functions (such as realpath) will always rely on the
		1416	string form of the pathname.
		1417
		1418	So this fucntionality is mainly useful to get some protection against
		1419	C<chdir>, to easily get an absolute path out of a relative path for future
		1420	reference, and to speed up doing many operations in the same directory
		1421	(e.g. when stat'ing all files in a directory).
		1422
		1423	The following functions implement this working directory abstraction:
		1424
		1425	=over 4
		1426
		1427	=item aio_wd $pathname, $callback->($wd)
		1428
		1429	Asynchonously canonicalise the given pathname and convert it to an
		1430	IO::AIO::WD object representing it. If possible and supported on the
		1431	system, also open a directory fd to speed up pathname resolution relative
		1432	to this working directory.
		1433
		1434	If something goes wrong, then C<undef> is passwd to the callback instead
		1435	of a working directory object and C<$!> is set appropriately. Since
		1436	passing C<undef> as working directory component of a pathname fails the
		1437	request with C<ENOENT>, there is often no need for error checking in the
		1438	C<aio_wd> callback, as future requests using the value will fail in the
		1439	expected way.
		1440
		1441	If this call isn't available because your OS lacks it or it couldn't be
		1442	detected, it will be emulated by calling C<fsync> instead.
		1443
		1444	=item IO::AIO::CWD
		1445
		1446	This is a compiletime constant (object) that represents the process
		1447	current working directory.
		1448
		1449	Specifying this object as working directory object for a pathname is as
		1450	if the pathname would be specified directly, without a directory object,
		1451	e.g., these calls are functionally identical:
		1452
		1453	aio_stat "somefile", sub { ... };
		1454	aio_stat [IO::AIO::CWD, "somefile"], sub { ... };
		1455
		1456	=back
		1457
		1458
672	=head2 IO::AIO::REQ CLASS	1459	=head2 IO::AIO::REQ CLASS
673		1460
674	All non-aggregate C<aio_*> functions return an object of this class when	1461	All non-aggregate C<aio_*> functions return an object of this class when
675	called in non-void context.	1462	called in non-void context.
676		1463
…		…
679	=item cancel $req	1466	=item cancel $req
680		1467
681	Cancels the request, if possible. Has the effect of skipping execution	1468	Cancels the request, if possible. Has the effect of skipping execution
682	when entering the B<execute> state and skipping calling the callback when	1469	when entering the B<execute> state and skipping calling the callback when
683	entering the the B<result> state, but will leave the request otherwise	1470	entering the the B<result> state, but will leave the request otherwise
684	untouched. That means that requests that currently execute will not be	1471	untouched (with the exception of readdir). That means that requests that
685	stopped and resources held by the request will not be freed prematurely.	1472	currently execute will not be stopped and resources held by the request
		1473	will not be freed prematurely.
686		1474
687	=item cb $req $callback->(...)	1475	=item cb $req $callback->(...)
688		1476
689	Replace (or simply set) the callback registered to the request.	1477	Replace (or simply set) the callback registered to the request.
690		1478
…		…
741	Their lifetime, simplified, looks like this: when they are empty, they	1529	Their lifetime, simplified, looks like this: when they are empty, they
742	will finish very quickly. If they contain only requests that are in the	1530	will finish very quickly. If they contain only requests that are in the
743	C<done> state, they will also finish. Otherwise they will continue to	1531	C<done> state, they will also finish. Otherwise they will continue to
744	exist.	1532	exist.
745		1533
746	That means after creating a group you have some time to add requests. And	1534	That means after creating a group you have some time to add requests
747	in the callbacks of those requests, you can add further requests to the	1535	(precisely before the callback has been invoked, which is only done within
748	group. And only when all those requests have finished will the the group	1536	the C<poll_cb>). And in the callbacks of those requests, you can add
749	itself finish.	1537	further requests to the group. And only when all those requests have
		1538	finished will the the group itself finish.
750		1539
751	=over 4	1540	=over 4
752		1541
753	=item add $grp ...	1542	=item add $grp ...
754		1543
…		…
763	=item $grp->cancel_subs	1552	=item $grp->cancel_subs
764		1553
765	Cancel all subrequests and clears any feeder, but not the group request	1554	Cancel all subrequests and clears any feeder, but not the group request
766	itself. Useful when you queued a lot of events but got a result early.	1555	itself. Useful when you queued a lot of events but got a result early.
767		1556
		1557	The group request will finish normally (you cannot add requests to the
		1558	group).
		1559
768	=item $grp->result (...)	1560	=item $grp->result (...)
769		1561
770	Set the result value(s) that will be passed to the group callback when all	1562	Set the result value(s) that will be passed to the group callback when all
771	subrequests have finished and set thre groups errno to the current value	1563	subrequests have finished and set the groups errno to the current value
772	of errno (just like calling C<errno> without an error number). By default,	1564	of errno (just like calling C<errno> without an error number). By default,
773	no argument will be passed and errno is zero.	1565	no argument will be passed and errno is zero.
774		1566
775	=item $grp->errno ([$errno])	1567	=item $grp->errno ([$errno])
776		1568
…		…
787	=item feed $grp $callback->($grp)	1579	=item feed $grp $callback->($grp)
788		1580
789	Sets a feeder/generator on this group: every group can have an attached	1581	Sets a feeder/generator on this group: every group can have an attached
790	generator that generates requests if idle. The idea behind this is that,	1582	generator that generates requests if idle. The idea behind this is that,
791	although you could just queue as many requests as you want in a group,	1583	although you could just queue as many requests as you want in a group,
792	this might starve other requests for a potentially long time. For	1584	this might starve other requests for a potentially long time. For example,
793	example, C<aio_scandir> might generate hundreds of thousands C<aio_stat>	1585	C<aio_scandir> might generate hundreds of thousands of C<aio_stat>
794	requests, delaying any later requests for a long time.	1586	requests, delaying any later requests for a long time.
795		1587
796	To avoid this, and allow incremental generation of requests, you can	1588	To avoid this, and allow incremental generation of requests, you can
797	instead a group and set a feeder on it that generates those requests. The	1589	instead a group and set a feeder on it that generates those requests. The
798	feed callback will be called whenever there are few enough (see C<limit>,	1590	feed callback will be called whenever there are few enough (see C<limit>,
…		…
803	not impose any limits).	1595	not impose any limits).
804		1596
805	If the feed does not queue more requests when called, it will be	1597	If the feed does not queue more requests when called, it will be
806	automatically removed from the group.	1598	automatically removed from the group.
807		1599
808	If the feed limit is C<0>, it will be set to C<2> automatically.	1600	If the feed limit is C<0> when this method is called, it will be set to
		1601	C<2> automatically.
809		1602
810	Example:	1603	Example:
811		1604
812	# stat all files in @files, but only ever use four aio requests concurrently:	1605	# stat all files in @files, but only ever use four aio requests concurrently:
813		1606
…		…
825	Sets the feeder limit for the group: The feeder will be called whenever	1618	Sets the feeder limit for the group: The feeder will be called whenever
826	the group contains less than this many requests.	1619	the group contains less than this many requests.
827		1620
828	Setting the limit to C<0> will pause the feeding process.	1621	Setting the limit to C<0> will pause the feeding process.
829		1622
		1623	The default value for the limit is C<0>, but note that setting a feeder
		1624	automatically bumps it up to C<2>.
		1625
830	=back	1626	=back
831		1627
832	=head2 SUPPORT FUNCTIONS	1628	=head2 SUPPORT FUNCTIONS
833		1629
		1630	=head3 EVENT PROCESSING AND EVENT LOOP INTEGRATION
		1631
834	=over 4	1632	=over 4
835		1633
836	=item $fileno = IO::AIO::poll_fileno	1634	=item $fileno = IO::AIO::poll_fileno
837		1635
838	Return the I<request result pipe file descriptor>. This filehandle must be	1636	Return the I<request result pipe file descriptor>. This filehandle must be
839	polled for reading by some mechanism outside this module (e.g. Event or	1637	polled for reading by some mechanism outside this module (e.g. EV, Glib,
840	select, see below or the SYNOPSIS). If the pipe becomes readable you have	1638	select and so on, see below or the SYNOPSIS). If the pipe becomes readable
841	to call C<poll_cb> to check the results.	1639	you have to call C<poll_cb> to check the results.
842		1640
843	See C<poll_cb> for an example.	1641	See C<poll_cb> for an example.
844		1642
845	=item IO::AIO::poll_cb	1643	=item IO::AIO::poll_cb
846		1644
847	Process all outstanding events on the result pipe. You have to call this	1645	Process some outstanding events on the result pipe. You have to call
848	regularly. Returns the number of events processed. Returns immediately	1646	this regularly. Returns C<0> if all events could be processed (or there
849	when no events are outstanding.	1647	were no events to process), or C<-1> if it returned earlier for whatever
		1648	reason. Returns immediately when no events are outstanding. The amount of
		1649	events processed depends on the settings of C<IO::AIO::max_poll_req> and
		1650	C<IO::AIO::max_poll_time>.
850		1651
851	If not all requests were processed for whatever reason, the filehandle	1652	If not all requests were processed for whatever reason, the filehandle
852	will still be ready when C<poll_cb> returns.	1653	will still be ready when C<poll_cb> returns, so normally you don't have to
		1654	do anything special to have it called later.
		1655
		1656	Apart from calling C<IO::AIO::poll_cb> when the event filehandle becomes
		1657	ready, it can be beneficial to call this function from loops which submit
		1658	a lot of requests, to make sure the results get processed when they become
		1659	available and not just when the loop is finished and the event loop takes
		1660	over again. This function returns very fast when there are no outstanding
		1661	requests.
853		1662
854	Example: Install an Event watcher that automatically calls	1663	Example: Install an Event watcher that automatically calls
855	IO::AIO::poll_cb with high priority:	1664	IO::AIO::poll_cb with high priority (more examples can be found in the
		1665	SYNOPSIS section, at the top of this document):
856		1666
857	Event->io (fd => IO::AIO::poll_fileno,	1667	Event->io (fd => IO::AIO::poll_fileno,
858	poll => 'r', async => 1,	1668	poll => 'r', async => 1,
859	cb => \&IO::AIO::poll_cb);	1669	cb => \&IO::AIO::poll_cb);
860		1670
861	=item IO::AIO::poll_some $max_requests	1671	=item IO::AIO::poll_wait
862		1672
863	Similar to C<poll_cb>, but only processes up to C<$max_requests> requests	1673	If there are any outstanding requests and none of them in the result
864	at a time.	1674	phase, wait till the result filehandle becomes ready for reading (simply
		1675	does a C<select> on the filehandle. This is useful if you want to
		1676	synchronously wait for some requests to finish).
865		1677
866	Useful if you want to ensure some level of interactiveness when perl is	1678	See C<nreqs> for an example.
867	not fast enough to process all requests in time.	1679
		1680	=item IO::AIO::poll
		1681
		1682	Waits until some requests have been handled.
		1683
		1684	Returns the number of requests processed, but is otherwise strictly
		1685	equivalent to:
		1686
		1687	IO::AIO::poll_wait, IO::AIO::poll_cb
		1688
		1689	=item IO::AIO::flush
		1690
		1691	Wait till all outstanding AIO requests have been handled.
		1692
		1693	Strictly equivalent to:
		1694
		1695	IO::AIO::poll_wait, IO::AIO::poll_cb
		1696	while IO::AIO::nreqs;
		1697
		1698	=item IO::AIO::max_poll_reqs $nreqs
		1699
		1700	=item IO::AIO::max_poll_time $seconds
		1701
		1702	These set the maximum number of requests (default C<0>, meaning infinity)
		1703	that are being processed by C<IO::AIO::poll_cb> in one call, respectively
		1704	the maximum amount of time (default C<0>, meaning infinity) spent in
		1705	C<IO::AIO::poll_cb> to process requests (more correctly the mininum amount
		1706	of time C<poll_cb> is allowed to use).
		1707
		1708	Setting C<max_poll_time> to a non-zero value creates an overhead of one
		1709	syscall per request processed, which is not normally a problem unless your
		1710	callbacks are really really fast or your OS is really really slow (I am
		1711	not mentioning Solaris here). Using C<max_poll_reqs> incurs no overhead.
		1712
		1713	Setting these is useful if you want to ensure some level of
		1714	interactiveness when perl is not fast enough to process all requests in
		1715	time.
		1716
		1717	For interactive programs, values such as C<0.01> to C<0.1> should be fine.
868		1718
869	Example: Install an Event watcher that automatically calls	1719	Example: Install an Event watcher that automatically calls
870	IO::AIO::poll_some with low priority, to ensure that other parts of the	1720	IO::AIO::poll_cb with low priority, to ensure that other parts of the
871	program get the CPU sometimes even under high AIO load.	1721	program get the CPU sometimes even under high AIO load.
872		1722
		1723	# try not to spend much more than 0.1s in poll_cb
		1724	IO::AIO::max_poll_time 0.1;
		1725
		1726	# use a low priority so other tasks have priority
873	Event->io (fd => IO::AIO::poll_fileno,	1727	Event->io (fd => IO::AIO::poll_fileno,
874	poll => 'r', nice => 1,	1728	poll => 'r', nice => 1,
875	cb => sub { IO::AIO::poll_some 256 });	1729	cb => &IO::AIO::poll_cb);
876		1730
877	=item IO::AIO::poll_wait	1731	=back
878		1732
879	Wait till the result filehandle becomes ready for reading (simply does a	1733	=head3 CONTROLLING THE NUMBER OF THREADS
880	C<select> on the filehandle. This is useful if you want to synchronously wait
881	for some requests to finish).
882		1734
883	See C<nreqs> for an example.	1735	=over
884
885	=item IO::AIO::nreqs
886
887	Returns the number of requests currently in the ready, execute or pending
888	states (i.e. for which their callback has not been invoked yet).
889
890	Example: wait till there are no outstanding requests anymore:
891
892	IO::AIO::poll_wait, IO::AIO::poll_cb
893	while IO::AIO::nreqs;
894
895	=item IO::AIO::nready
896
897	Returns the number of requests currently in the ready state (not yet
898	executed).
899
900	=item IO::AIO::npending
901
902	Returns the number of requests currently in the pending state (executed,
903	but not yet processed by poll_cb).
904
905	=item IO::AIO::flush
906
907	Wait till all outstanding AIO requests have been handled.
908
909	Strictly equivalent to:
910
911	IO::AIO::poll_wait, IO::AIO::poll_cb
912	while IO::AIO::nreqs;
913
914	=item IO::AIO::poll
915
916	Waits until some requests have been handled.
917
918	Strictly equivalent to:
919
920	IO::AIO::poll_wait, IO::AIO::poll_cb
921	if IO::AIO::nreqs;
922		1736
923	=item IO::AIO::min_parallel $nthreads	1737	=item IO::AIO::min_parallel $nthreads
924		1738
925	Set the minimum number of AIO threads to C<$nthreads>. The current	1739	Set the minimum number of AIO threads to C<$nthreads>. The current
926	default is C<8>, which means eight asynchronous operations can execute	1740	default is C<8>, which means eight asynchronous operations can execute
927	concurrently at any one time (the number of outstanding requests,	1741	concurrently at any one time (the number of outstanding requests,
928	however, is unlimited).	1742	however, is unlimited).
929		1743
930	IO::AIO starts threads only on demand, when an AIO request is queued and	1744	IO::AIO starts threads only on demand, when an AIO request is queued and
931	no free thread exists.	1745	no free thread exists. Please note that queueing up a hundred requests can
		1746	create demand for a hundred threads, even if it turns out that everything
		1747	is in the cache and could have been processed faster by a single thread.
932		1748
933	It is recommended to keep the number of threads relatively low, as some	1749	It is recommended to keep the number of threads relatively low, as some
934	Linux kernel versions will scale negatively with the number of threads	1750	Linux kernel versions will scale negatively with the number of threads
935	(higher parallelity => MUCH higher latency). With current Linux 2.6	1751	(higher parallelity => MUCH higher latency). With current Linux 2.6
936	versions, 4-32 threads should be fine.	1752	versions, 4-32 threads should be fine.
…		…
950	This module automatically runs C<max_parallel 0> at program end, to ensure	1766	This module automatically runs C<max_parallel 0> at program end, to ensure
951	that all threads are killed and that there are no outstanding requests.	1767	that all threads are killed and that there are no outstanding requests.
952		1768
953	Under normal circumstances you don't need to call this function.	1769	Under normal circumstances you don't need to call this function.
954		1770
		1771	=item IO::AIO::max_idle $nthreads
		1772
		1773	Limit the number of threads (default: 4) that are allowed to idle
		1774	(i.e., threads that did not get a request to process within the idle
		1775	timeout (default: 10 seconds). That means if a thread becomes idle while
		1776	C<$nthreads> other threads are also idle, it will free its resources and
		1777	exit.
		1778
		1779	This is useful when you allow a large number of threads (e.g. 100 or 1000)
		1780	to allow for extremely high load situations, but want to free resources
		1781	under normal circumstances (1000 threads can easily consume 30MB of RAM).
		1782
		1783	The default is probably ok in most situations, especially if thread
		1784	creation is fast. If thread creation is very slow on your system you might
		1785	want to use larger values.
		1786
		1787	=item IO::AIO::idle_timeout $seconds
		1788
		1789	Sets the minimum idle timeout (default 10) after which worker threads are
		1790	allowed to exit. SEe C<IO::AIO::max_idle>.
		1791
955	=item $oldmaxreqs = IO::AIO::max_outstanding $maxreqs	1792	=item IO::AIO::max_outstanding $maxreqs
		1793
		1794	Sets the maximum number of outstanding requests to C<$nreqs>. If
		1795	you do queue up more than this number of requests, the next call to
		1796	C<IO::AIO::poll_cb> (and other functions calling C<poll_cb>, such as
		1797	C<IO::AIO::flush> or C<IO::AIO::poll>) will block until the limit is no
		1798	longer exceeded.
		1799
		1800	In other words, this setting does not enforce a queue limit, but can be
		1801	used to make poll functions block if the limit is exceeded.
956		1802
957	This is a very bad function to use in interactive programs because it	1803	This is a very bad function to use in interactive programs because it
958	blocks, and a bad way to reduce concurrency because it is inexact: Better	1804	blocks, and a bad way to reduce concurrency because it is inexact: Better
959	use an C<aio_group> together with a feed callback.	1805	use an C<aio_group> together with a feed callback.
960		1806
961	Sets the maximum number of outstanding requests to C<$nreqs>. If you	1807	It's main use is in scripts without an event loop - when you want to stat
962	to queue up more than this number of requests, the next call to the	1808	a lot of files, you can write somehting like this:
963	C<poll_cb> (and C<poll_some> and other functions calling C<poll_cb>)
964	function will block until the limit is no longer exceeded.
965		1809
966	The default value is very large, so there is no practical limit on the	1810	IO::AIO::max_outstanding 32;
967	number of outstanding requests.
968		1811
969	You can still queue as many requests as you want. Therefore,	1812	for my $path (...) {
970	C<max_oustsanding> is mainly useful in simple scripts (with low values) or	1813	aio_stat $path , ...;
971	as a stop gap to shield against fatal memory overflow (with large values).	1814	IO::AIO::poll_cb;
		1815	}
		1816
		1817	IO::AIO::flush;
		1818
		1819	The call to C<poll_cb> inside the loop will normally return instantly, but
		1820	as soon as more thna C<32> reqeusts are in-flight, it will block until
		1821	some requests have been handled. This keeps the loop from pushing a large
		1822	number of C<aio_stat> requests onto the queue.
		1823
		1824	The default value for C<max_outstanding> is very large, so there is no
		1825	practical limit on the number of outstanding requests.
972		1826
973	=back	1827	=back
974		1828
		1829	=head3 STATISTICAL INFORMATION
		1830
		1831	=over
		1832
		1833	=item IO::AIO::nreqs
		1834
		1835	Returns the number of requests currently in the ready, execute or pending
		1836	states (i.e. for which their callback has not been invoked yet).
		1837
		1838	Example: wait till there are no outstanding requests anymore:
		1839
		1840	IO::AIO::poll_wait, IO::AIO::poll_cb
		1841	while IO::AIO::nreqs;
		1842
		1843	=item IO::AIO::nready
		1844
		1845	Returns the number of requests currently in the ready state (not yet
		1846	executed).
		1847
		1848	=item IO::AIO::npending
		1849
		1850	Returns the number of requests currently in the pending state (executed,
		1851	but not yet processed by poll_cb).
		1852
		1853	=back
		1854
		1855	=head3 MISCELLANEOUS FUNCTIONS
		1856
		1857	IO::AIO implements some functions that might be useful, but are not
		1858	asynchronous.
		1859
		1860	=over 4
		1861
		1862	=item IO::AIO::sendfile $ofh, $ifh, $offset, $count
		1863
		1864	Calls the C<eio_sendfile_sync> function, which is like C<aio_sendfile>,
		1865	but is blocking (this makes most sense if you know the input data is
		1866	likely cached already and the output filehandle is set to non-blocking
		1867	operations).
		1868
		1869	Returns the number of bytes copied, or C<-1> on error.
		1870
		1871	=item IO::AIO::fadvise $fh, $offset, $len, $advice
		1872
		1873	Simply calls the C<posix_fadvise> function (see its
		1874	manpage for details). The following advice constants are
		1875	available: C<IO::AIO::FADV_NORMAL>, C<IO::AIO::FADV_SEQUENTIAL>,
		1876	C<IO::AIO::FADV_RANDOM>, C<IO::AIO::FADV_NOREUSE>,
		1877	C<IO::AIO::FADV_WILLNEED>, C<IO::AIO::FADV_DONTNEED>.
		1878
		1879	On systems that do not implement C<posix_fadvise>, this function returns
		1880	ENOSYS, otherwise the return value of C<posix_fadvise>.
		1881
		1882	=item IO::AIO::madvise $scalar, $offset, $len, $advice
		1883
		1884	Simply calls the C<posix_madvise> function (see its
		1885	manpage for details). The following advice constants are
		1886	available: C<IO::AIO::MADV_NORMAL>, C<IO::AIO::MADV_SEQUENTIAL>,
		1887	C<IO::AIO::MADV_RANDOM>, C<IO::AIO::MADV_WILLNEED>, C<IO::AIO::MADV_DONTNEED>.
		1888
		1889	On systems that do not implement C<posix_madvise>, this function returns
		1890	ENOSYS, otherwise the return value of C<posix_madvise>.
		1891
		1892	=item IO::AIO::mprotect $scalar, $offset, $len, $protect
		1893
		1894	Simply calls the C<mprotect> function on the preferably AIO::mmap'ed
		1895	$scalar (see its manpage for details). The following protect
		1896	constants are available: C<IO::AIO::PROT_NONE>, C<IO::AIO::PROT_READ>,
		1897	C<IO::AIO::PROT_WRITE>, C<IO::AIO::PROT_EXEC>.
		1898
		1899	On systems that do not implement C<mprotect>, this function returns
		1900	ENOSYS, otherwise the return value of C<mprotect>.
		1901
		1902	=item IO::AIO::mmap $scalar, $length, $prot, $flags, $fh[, $offset]
		1903
		1904	Memory-maps a file (or anonymous memory range) and attaches it to the
		1905	given C<$scalar>, which will act like a string scalar.
		1906
		1907	The only operations allowed on the scalar are C<substr>/C<vec> that don't
		1908	change the string length, and most read-only operations such as copying it
		1909	or searching it with regexes and so on.
		1910
		1911	Anything else is unsafe and will, at best, result in memory leaks.
		1912
		1913	The memory map associated with the C<$scalar> is automatically removed
		1914	when the C<$scalar> is destroyed, or when the C<IO::AIO::mmap> or
		1915	C<IO::AIO::munmap> functions are called.
		1916
		1917	This calls the C<mmap>(2) function internally. See your system's manual
		1918	page for details on the C<$length>, C<$prot> and C<$flags> parameters.
		1919
		1920	The C<$length> must be larger than zero and smaller than the actual
		1921	filesize.
		1922
		1923	C<$prot> is a combination of C<IO::AIO::PROT_NONE>, C<IO::AIO::PROT_EXEC>,
		1924	C<IO::AIO::PROT_READ> and/or C<IO::AIO::PROT_WRITE>,
		1925
		1926	C<$flags> can be a combination of C<IO::AIO::MAP_SHARED> or
		1927	C<IO::AIO::MAP_PRIVATE>, or a number of system-specific flags (when
		1928	not available, the are defined as 0): C<IO::AIO::MAP_ANONYMOUS>
		1929	(which is set to C<MAP_ANON> if your system only provides this
		1930	constant), C<IO::AIO::MAP_HUGETLB>, C<IO::AIO::MAP_LOCKED>,
		1931	C<IO::AIO::MAP_NORESERVE>, C<IO::AIO::MAP_POPULATE> or
		1932	C<IO::AIO::MAP_NONBLOCK>
		1933
		1934	If C<$fh> is C<undef>, then a file descriptor of C<-1> is passed.
		1935
		1936	C<$offset> is the offset from the start of the file - it generally must be
		1937	a multiple of C<IO::AIO::PAGESIZE> and defaults to C<0>.
		1938
		1939	Example:
		1940
		1941	use Digest::MD5;
		1942	use IO::AIO;
		1943
		1944	open my $fh, "<verybigfile"
		1945	or die "$!";
		1946
		1947	IO::AIO::mmap my $data, -s $fh, IO::AIO::PROT_READ, IO::AIO::MAP_SHARED, $fh
		1948	or die "verybigfile: $!";
		1949
		1950	my $fast_md5 = md5 $data;
		1951
		1952	=item IO::AIO::munmap $scalar
		1953
		1954	Removes a previous mmap and undefines the C<$scalar>.
		1955
		1956	=item IO::AIO::munlock $scalar, $offset = 0, $length = undef
		1957
		1958	Calls the C<munlock> function, undoing the effects of a previous
		1959	C<aio_mlock> call (see its description for details).
		1960
		1961	=item IO::AIO::munlockall
		1962
		1963	Calls the C<munlockall> function.
		1964
		1965	On systems that do not implement C<munlockall>, this function returns
		1966	ENOSYS, otherwise the return value of C<munlockall>.
		1967
		1968	=back
		1969
975	=cut	1970	=cut
976		1971
977	# support function to convert a fd into a perl filehandle
978	sub _fd2fh {
979	return undef if $_[0] < 0;
980
981	# try to generate nice filehandles
982	my $sym = "IO::AIO::fd#$_[0]";
983	local *$sym;
984
985	open *$sym, "+<&=$_[0]" # usually works under any unix
986	or open *$sym, "<&=$_[0]" # cygwin needs this
987	or open *$sym, ">&=$_[0]" # or this
988	or return undef;
989
990	*$sym
991	}
992
993	min_parallel 8;	1972	min_parallel 8;
994		1973
995	END {	1974	END { flush }
996	flush;
997	};
998		1975
999	1;	1976	1;
1000		1977
		1978	=head1 EVENT LOOP INTEGRATION
		1979
		1980	It is recommended to use L<AnyEvent::AIO> to integrate IO::AIO
		1981	automatically into many event loops:
		1982
		1983	# AnyEvent integration (EV, Event, Glib, Tk, POE, urxvt, pureperl...)
		1984	use AnyEvent::AIO;
		1985
		1986	You can also integrate IO::AIO manually into many event loops, here are
		1987	some examples of how to do this:
		1988
		1989	# EV integration
		1990	my $aio_w = EV::io IO::AIO::poll_fileno, EV::READ, \&IO::AIO::poll_cb;
		1991
		1992	# Event integration
		1993	Event->io (fd => IO::AIO::poll_fileno,
		1994	poll => 'r',
		1995	cb => \&IO::AIO::poll_cb);
		1996
		1997	# Glib/Gtk2 integration
		1998	add_watch Glib::IO IO::AIO::poll_fileno,
		1999	in => sub { IO::AIO::poll_cb; 1 };
		2000
		2001	# Tk integration
		2002	Tk::Event::IO->fileevent (IO::AIO::poll_fileno, "",
		2003	readable => \&IO::AIO::poll_cb);
		2004
		2005	# Danga::Socket integration
		2006	Danga::Socket->AddOtherFds (IO::AIO::poll_fileno =>
		2007	\&IO::AIO::poll_cb);
		2008
1001	=head2 FORK BEHAVIOUR	2009	=head2 FORK BEHAVIOUR
1002		2010
1003	This module should do "the right thing" when the process using it forks:	2011	Usage of pthreads in a program changes the semantics of fork
		2012	considerably. Specifically, only async-safe functions can be called after
		2013	fork. Perl doesn't know about this, so in general, you cannot call fork
		2014	with defined behaviour in perl if pthreads are involved. IO::AIO uses
		2015	pthreads, so this applies, but many other extensions and (for inexplicable
		2016	reasons) perl itself often is linked against pthreads, so this limitation
		2017	applies to quite a lot of perls.
1004		2018
1005	Before the fork, IO::AIO enters a quiescent state where no requests	2019	This module no longer tries to fight your OS, or POSIX. That means IO::AIO
1006	can be added in other threads and no results will be processed. After	2020	only works in the process that loaded it. Forking is fully supported, but
1007	the fork the parent simply leaves the quiescent state and continues	2021	using IO::AIO in the child is not.
1008	request/result processing, while the child frees the request/result queue
1009	(so that the requests started before the fork will only be handled in the
1010	parent). Threads will be started on demand until the limit set in the
1011	parent process has been reached again.
1012		2022
1013	In short: the parent will, after a short pause, continue as if fork had	2023	You might get around by not I<using> IO::AIO before (or after)
1014	not been called, while the child will act as if IO::AIO has not been used	2024	forking. You could also try to call the L<IO::AIO::reinit> function in the
1015	yet.	2025	child:
		2026
		2027	=over 4
		2028
		2029	=item IO::AIO::reinit
		2030
		2031	Abandons all current requests and I/O threads and simply reinitialises all
		2032	data structures. This is not an operation supported by any standards, but
		2033	happens to work on GNU/Linux and some newer BSD systems.
		2034
		2035	The only reasonable use for this function is to call it after forking, if
		2036	C<IO::AIO> was used in the parent. Calling it while IO::AIO is active in
		2037	the process will result in undefined behaviour. Calling it at any time
		2038	will also result in any undefined (by POSIX) behaviour.
		2039
		2040	=back
1016		2041
1017	=head2 MEMORY USAGE	2042	=head2 MEMORY USAGE
1018		2043
1019	Per-request usage:	2044	Per-request usage:
1020		2045
…		…
1022	bytes of memory. In addition, stat requests need a stat buffer (possibly	2047	bytes of memory. In addition, stat requests need a stat buffer (possibly
1023	a few hundred bytes), readdir requires a result buffer and so on. Perl	2048	a few hundred bytes), readdir requires a result buffer and so on. Perl
1024	scalars and other data passed into aio requests will also be locked and	2049	scalars and other data passed into aio requests will also be locked and
1025	will consume memory till the request has entered the done state.	2050	will consume memory till the request has entered the done state.
1026		2051
1027	This is now awfully much, so queuing lots of requests is not usually a	2052	This is not awfully much, so queuing lots of requests is not usually a
1028	problem.	2053	problem.
1029		2054
1030	Per-thread usage:	2055	Per-thread usage:
1031		2056
1032	In the execution phase, some aio requests require more memory for	2057	In the execution phase, some aio requests require more memory for
…		…
1037		2062
1038	Known bugs will be fixed in the next release.	2063	Known bugs will be fixed in the next release.
1039		2064
1040	=head1 SEE ALSO	2065	=head1 SEE ALSO
1041		2066
1042	L<Coro::AIO>.	2067	L<AnyEvent::AIO> for easy integration into event loops, L<Coro::AIO> for a
		2068	more natural syntax.
1043		2069
1044	=head1 AUTHOR	2070	=head1 AUTHOR
1045		2071
1046	Marc Lehmann <schmorp@schmorp.de>	2072	Marc Lehmann <schmorp@schmorp.de>
1047	http://home.schmorp.de/	2073	http://home.schmorp.de/

Diff Legend

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