ViewVC Help

View File | Revision Log | Show Annotations | Download File

/cvs/IO-AIO/AIO.pm

Comparing IO-AIO/AIO.pm (file contents):
Revision 1.61 by root, Sun Oct 22 13:52:11 2006 UTC vs.
Revision 1.195 by root, Fri May 27 19:56:31 2011 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
…		…
18	};	19	};
19		20
20	# version 2+ has request and group objects	21	# version 2+ has request and group objects
21	use IO::AIO 2;	22	use IO::AIO 2;
22		23
		24	aioreq_pri 4; # give next request a very high priority
23	my $req = aio_unlink "/tmp/file", sub { };	25	my $req = aio_unlink "/tmp/file", sub { };
24	$req->cancel; # cancel request if still in queue	26	$req->cancel; # cancel request if still in queue
25		27
26	my $grp = aio_group sub { print "all stats done\n" };	28	my $grp = aio_group sub { print "all stats done\n" };
27	add $grp aio_stat "..." for ...;	29	add $grp aio_stat "..." for ...;
28		30
29	# AnyEvent integration
30	open my $fh, "<&=" . IO::AIO::poll_fileno or die "$!";
31	my $w = AnyEvent->io (fh => $fh, poll => 'r', cb => sub { IO::AIO::poll_cb });
32
33	# Event integration
34	Event->io (fd => IO::AIO::poll_fileno,
35	poll => 'r',
36	cb => \&IO::AIO::poll_cb);
37
38	# Glib/Gtk2 integration
39	add_watch Glib::IO IO::AIO::poll_fileno,
40	in => sub { IO::AIO::poll_cb; 1 };
41
42	# Tk integration
43	Tk::Event::IO->fileevent (IO::AIO::poll_fileno, "",
44	readable => \&IO::AIO::poll_cb);
45
46	# Danga::Socket integration
47	Danga::Socket->AddOtherFds (IO::AIO::poll_fileno =>
48	\&IO::AIO::poll_cb);
49
50	=head1 DESCRIPTION	31	=head1 DESCRIPTION
51		32
52	This module implements asynchronous I/O using whatever means your	33	This module implements asynchronous I/O using whatever means your
53	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>).
54		36
		37	Asynchronous means that operations that can normally block your program
		38	(e.g. reading from disk) will be done asynchronously: the operation
		39	will still block, but you can do something else in the meantime. This
		40	is extremely useful for programs that need to stay interactive even
		41	when doing heavy I/O (GUI programs, high performance network servers
		42	etc.), but can also be used to easily do operations in parallel that are
		43	normally done sequentially, e.g. stat'ing many files, which is much faster
		44	on a RAID volume or over NFS when you do a number of stat operations
		45	concurrently.
		46
		47	While most of this works on all types of file descriptors (for
		48	example sockets), using these functions on file descriptors that
		49	support nonblocking operation (again, sockets, pipes etc.) is
		50	very inefficient. Use an event loop for that (such as the L<EV>
		51	module): IO::AIO will naturally fit into such an event loop itself.
		52
55	Currently, a number of threads are started that execute your read/writes	53	In this version, a number of threads are started that execute your
56	and signal their completion. You don't need thread support in your libc or	54	requests and signal their completion. You don't need thread support
57	perl, and the threads created by this module will not be visible to the	55	in perl, and the threads created by this module will not be visible
58	pthreads library. In the future, this module might make use of the native	56	to perl. In the future, this module might make use of the native aio
59	aio functions available on many operating systems. However, they are often	57	functions available on many operating systems. However, they are often
60	not well-supported (Linux doesn't allow them on normal files currently,	58	not well-supported or restricted (GNU/Linux doesn't allow them on normal
61	for example), and they would only support aio_read and aio_write, so the	59	files currently, for example), and they would only support aio_read and
62	remaining functionality would have to be implemented using threads anyway.	60	aio_write, so the remaining functionality would have to be implemented
		61	using threads anyway.
63		62
64	Although the module will work with in the presence of other threads, it is	63	Although the module will work in the presence of other (Perl-) threads,
65	currently not reentrant, so use appropriate locking yourself, always call	64	it is currently not reentrant in any way, so use appropriate locking
66	C<poll_cb> from within the same thread, or never call C<poll_cb> (or other	65	yourself, always call C<poll_cb> from within the same thread, or never
67	C<aio_> functions) recursively.	66	call C<poll_cb> (or other C<aio_> functions) recursively.
		67
		68	=head2 EXAMPLE
		69
		70	This is a simple example that uses the EV module and loads
		71	F</etc/passwd> asynchronously:
		72
		73	use Fcntl;
		74	use EV;
		75	use IO::AIO;
		76
		77	# register the IO::AIO callback with EV
		78	my $aio_w = EV::io IO::AIO::poll_fileno, EV::READ, \&IO::AIO::poll_cb;
		79
		80	# queue the request to open /etc/passwd
		81	aio_open "/etc/passwd", IO::AIO::O_RDONLY, 0, sub {
		82	my $fh = shift
		83	or die "error while opening: $!";
		84
		85	# stat'ing filehandles is generally non-blocking
		86	my $size = -s $fh;
		87
		88	# queue a request to read the file
		89	my $contents;
		90	aio_read $fh, 0, $size, $contents, 0, sub {
		91	$_[0] == $size
		92	or die "short read: $!";
		93
		94	close $fh;
		95
		96	# file contents now in $contents
		97	print $contents;
		98
		99	# exit event loop and program
		100	EV::unloop;
		101	};
		102	};
		103
		104	# possibly queue up other requests, or open GUI windows,
		105	# check for sockets etc. etc.
		106
		107	# process events as long as there are some:
		108	EV::loop;
		109
		110	=head1 REQUEST ANATOMY AND LIFETIME
		111
		112	Every C<aio_*> function creates a request. which is a C data structure not
		113	directly visible to Perl.
		114
		115	If called in non-void context, every request function returns a Perl
		116	object representing the request. In void context, nothing is returned,
		117	which saves a bit of memory.
		118
		119	The perl object is a fairly standard ref-to-hash object. The hash contents
		120	are not used by IO::AIO so you are free to store anything you like in it.
		121
		122	During their existance, aio requests travel through the following states,
		123	in order:
		124
		125	=over 4
		126
		127	=item ready
		128
		129	Immediately after a request is created it is put into the ready state,
		130	waiting for a thread to execute it.
		131
		132	=item execute
		133
		134	A thread has accepted the request for processing and is currently
		135	executing it (e.g. blocking in read).
		136
		137	=item pending
		138
		139	The request has been executed and is waiting for result processing.
		140
		141	While request submission and execution is fully asynchronous, result
		142	processing is not and relies on the perl interpreter calling C<poll_cb>
		143	(or another function with the same effect).
		144
		145	=item result
		146
		147	The request results are processed synchronously by C<poll_cb>.
		148
		149	The C<poll_cb> function will process all outstanding aio requests by
		150	calling their callbacks, freeing memory associated with them and managing
		151	any groups they are contained in.
		152
		153	=item done
		154
		155	Request has reached the end of its lifetime and holds no resources anymore
		156	(except possibly for the Perl object, but its connection to the actual
		157	aio request is severed and calling its methods will either do nothing or
		158	result in a runtime error).
		159
		160	=back
68		161
69	=cut	162	=cut
70		163
71	package IO::AIO;	164	package IO::AIO;
72		165
73	no warnings;	166	use Carp ();
74	use strict 'vars';	167
		168	use common::sense;
75		169
76	use base 'Exporter';	170	use base 'Exporter';
77		171
78	BEGIN {	172	BEGIN {
79	our $VERSION = '2.0';	173	our $VERSION = '3.9';
80		174
81	our @EXPORT = qw(aio_sendfile aio_read aio_write aio_open aio_close aio_stat	175	our @AIO_REQ = qw(aio_sendfile aio_read aio_write aio_open aio_close
82	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
83	aio_fsync aio_fdatasync aio_readahead aio_rename aio_link aio_move	177	aio_scandir aio_symlink aio_readlink aio_sync aio_fsync
		178	aio_fdatasync aio_sync_file_range aio_pathsync aio_readahead
		179	aio_rename aio_link aio_move aio_copy aio_group
		180	aio_nop aio_mknod aio_load aio_rmtree aio_mkdir aio_chown
		181	aio_chmod aio_utime aio_truncate
		182	aio_msync aio_mtouch aio_mlock aio_mlockall
84	aio_group);	183	aio_statvfs);
85	our @EXPORT_OK = qw(poll_fileno poll_cb min_parallel max_parallel max_outstanding nreqs);	184
		185	our @EXPORT = (@AIO_REQ, qw(aioreq_pri aioreq_nice));
		186	our @EXPORT_OK = qw(poll_fileno poll_cb poll_wait flush
		187	min_parallel max_parallel max_idle idle_timeout
		188	nreqs nready npending nthreads
		189	max_poll_time max_poll_reqs
		190	sendfile fadvise madvise
		191	mmap munmap munlock munlockall);
		192
		193	push @AIO_REQ, qw(aio_busy); # not exported
86		194
87	@IO::AIO::GRP::ISA = 'IO::AIO::REQ';	195	@IO::AIO::GRP::ISA = 'IO::AIO::REQ';
88		196
89	require XSLoader;	197	require XSLoader;
90	XSLoader::load ("IO::AIO", $VERSION);	198	XSLoader::load ("IO::AIO", $VERSION);
91	}	199	}
92		200
93	=head1 FUNCTIONS	201	=head1 FUNCTIONS
94		202
		203	=head2 QUICK OVERVIEW
		204
		205	This section simply lists the prototypes of the most important functions
		206	for quick reference. See the following sections for function-by-function
		207	documentation.
		208
		209	aio_open $pathname, $flags, $mode, $callback->($fh)
		210	aio_close $fh, $callback->($status)
		211	aio_read $fh,$offset,$length, $data,$dataoffset, $callback->($retval)
		212	aio_write $fh,$offset,$length, $data,$dataoffset, $callback->($retval)
		213	aio_sendfile $out_fh, $in_fh, $in_offset, $length, $callback->($retval)
		214	aio_readahead $fh,$offset,$length, $callback->($retval)
		215	aio_stat $fh_or_path, $callback->($status)
		216	aio_lstat $fh, $callback->($status)
		217	aio_statvfs $fh_or_path, $callback->($statvfs)
		218	aio_utime $fh_or_path, $atime, $mtime, $callback->($status)
		219	aio_chown $fh_or_path, $uid, $gid, $callback->($status)
		220	aio_truncate $fh_or_path, $offset, $callback->($status)
		221	aio_chmod $fh_or_path, $mode, $callback->($status)
		222	aio_unlink $pathname, $callback->($status)
		223	aio_mknod $path, $mode, $dev, $callback->($status)
		224	aio_link $srcpath, $dstpath, $callback->($status)
		225	aio_symlink $srcpath, $dstpath, $callback->($status)
		226	aio_readlink $path, $callback->($link)
		227	aio_rename $srcpath, $dstpath, $callback->($status)
		228	aio_mkdir $pathname, $mode, $callback->($status)
		229	aio_rmdir $pathname, $callback->($status)
		230	aio_readdir $pathname, $callback->($entries)
		231	aio_readdirx $pathname, $flags, $callback->($entries, $flags)
		232	IO::AIO::READDIR_DENTS IO::AIO::READDIR_DIRS_FIRST
		233	IO::AIO::READDIR_STAT_ORDER IO::AIO::READDIR_FOUND_UNKNOWN
		234	aio_load $path, $data, $callback->($status)
		235	aio_copy $srcpath, $dstpath, $callback->($status)
		236	aio_move $srcpath, $dstpath, $callback->($status)
		237	aio_scandir $path, $maxreq, $callback->($dirs, $nondirs)
		238	aio_rmtree $path, $callback->($status)
		239	aio_sync $callback->($status)
		240	aio_fsync $fh, $callback->($status)
		241	aio_fdatasync $fh, $callback->($status)
		242	aio_sync_file_range $fh, $offset, $nbytes, $flags, $callback->($status)
		243	aio_pathsync $path, $callback->($status)
		244	aio_msync $scalar, $offset = 0, $length = undef, flags = 0, $callback->($status)
		245	aio_mtouch $scalar, $offset = 0, $length = undef, flags = 0, $callback->($status)
		246	aio_mlock $scalar, $offset = 0, $length = undef, $callback->($status)
		247	aio_mlockall $flags, $callback->($status)
		248	aio_group $callback->(...)
		249	aio_nop $callback->()
		250
		251	$prev_pri = aioreq_pri [$pri]
		252	aioreq_nice $pri_adjust
		253
		254	IO::AIO::poll_wait
		255	IO::AIO::poll_cb
		256	IO::AIO::poll
		257	IO::AIO::flush
		258	IO::AIO::max_poll_reqs $nreqs
		259	IO::AIO::max_poll_time $seconds
		260	IO::AIO::min_parallel $nthreads
		261	IO::AIO::max_parallel $nthreads
		262	IO::AIO::max_idle $nthreads
		263	IO::AIO::idle_timeout $seconds
		264	IO::AIO::max_outstanding $maxreqs
		265	IO::AIO::nreqs
		266	IO::AIO::nready
		267	IO::AIO::npending
		268
		269	IO::AIO::sendfile $ofh, $ifh, $offset, $count
		270	IO::AIO::fadvise $fh, $offset, $len, $advice
		271	IO::AIO::madvise $scalar, $offset, $length, $advice
		272	IO::AIO::mprotect $scalar, $offset, $length, $protect
		273	IO::AIO::munlock $scalar, $offset = 0, $length = undef
		274	IO::AIO::munlockall
		275
95	=head2 AIO FUNCTIONS	276	=head2 AIO REQUEST FUNCTIONS
96		277
97	All the C<aio_*> calls are more or less thin wrappers around the syscall	278	All the C<aio_*> calls are more or less thin wrappers around the syscall
98	with the same name (sans C<aio_>). The arguments are similar or identical,	279	with the same name (sans C<aio_>). The arguments are similar or identical,
99	and they all accept an additional (and optional) C<$callback> argument	280	and they all accept an additional (and optional) C<$callback> argument
100	which must be a code reference. This code reference will get called with	281	which must be a code reference. This code reference will get called with
101	the syscall return code (e.g. most syscalls return C<-1> on error, unlike	282	the syscall return code (e.g. most syscalls return C<-1> on error, unlike
102	perl, which usually delivers "false") as it's sole argument when the given	283	perl, which usually delivers "false") as its sole argument after the given
103	syscall has been executed asynchronously.	284	syscall has been executed asynchronously.
104		285
105	All functions expecting a filehandle keep a copy of the filehandle	286	All functions expecting a filehandle keep a copy of the filehandle
106	internally until the request has finished.	287	internally until the request has finished.
107		288
108	All requests return objects of type L<IO::AIO::REQ> that allow further	289	All functions return request objects of type L<IO::AIO::REQ> that allow
109	manipulation of those requests while they are in-flight.	290	further manipulation of those requests while they are in-flight.
110		291
111	The pathnames you pass to these routines I<must> be absolute and	292	The pathnames you pass to these routines I<must> be absolute and
112	encoded in byte form. The reason for the former is that at the time the	293	encoded as octets. The reason for the former is that at the time the
113	request is being executed, the current working directory could have	294	request is being executed, the current working directory could have
114	changed. Alternatively, you can make sure that you never change the	295	changed. Alternatively, you can make sure that you never change the
115	current working directory.	296	current working directory anywhere in the program and then use relative
		297	paths.
116		298
117	To encode pathnames to byte form, either make sure you either: a)	299	To encode pathnames as octets, either make sure you either: a) always pass
118	always pass in filenames you got from outside (command line, readdir	300	in filenames you got from outside (command line, readdir etc.) without
119	etc.), b) are ASCII or ISO 8859-1, c) use the Encode module and encode	301	tinkering, b) are ASCII or ISO 8859-1, c) use the Encode module and encode
120	your pathnames to the locale (or other) encoding in effect in the user	302	your pathnames to the locale (or other) encoding in effect in the user
121	environment, d) use Glib::filename_from_unicode on unicode filenames or e)	303	environment, d) use Glib::filename_from_unicode on unicode filenames or e)
122	use something else.	304	use something else to ensure your scalar has the correct contents.
		305
		306	This works, btw. independent of the internal UTF-8 bit, which IO::AIO
		307	handles correctly whether it is set or not.
123		308
124	=over 4	309	=over 4
		310
		311	=item $prev_pri = aioreq_pri [$pri]
		312
		313	Returns the priority value that would be used for the next request and, if
		314	C<$pri> is given, sets the priority for the next aio request.
		315
		316	The default priority is C<0>, the minimum and maximum priorities are C<-4>
		317	and C<4>, respectively. Requests with higher priority will be serviced
		318	first.
		319
		320	The priority will be reset to C<0> after each call to one of the C<aio_*>
		321	functions.
		322
		323	Example: open a file with low priority, then read something from it with
		324	higher priority so the read request is serviced before other low priority
		325	open requests (potentially spamming the cache):
		326
		327	aioreq_pri -3;
		328	aio_open ..., sub {
		329	return unless $_[0];
		330
		331	aioreq_pri -2;
		332	aio_read $_[0], ..., sub {
		333	...
		334	};
		335	};
		336
		337
		338	=item aioreq_nice $pri_adjust
		339
		340	Similar to C<aioreq_pri>, but subtracts the given value from the current
		341	priority, so the effect is cumulative.
		342
125		343
126	=item aio_open $pathname, $flags, $mode, $callback->($fh)	344	=item aio_open $pathname, $flags, $mode, $callback->($fh)
127		345
128	Asynchronously open or create a file and call the callback with a newly	346	Asynchronously open or create a file and call the callback with a newly
129	created filehandle for the file.	347	created filehandle for the file.
…		…
135	list. They are the same as used by C<sysopen>.	353	list. They are the same as used by C<sysopen>.
136		354
137	Likewise, C<$mode> specifies the mode of the newly created file, if it	355	Likewise, C<$mode> specifies the mode of the newly created file, if it
138	didn't exist and C<O_CREAT> has been given, just like perl's C<sysopen>,	356	didn't exist and C<O_CREAT> has been given, just like perl's C<sysopen>,
139	except that it is mandatory (i.e. use C<0> if you don't create new files,	357	except that it is mandatory (i.e. use C<0> if you don't create new files,
140	and C<0666> or C<0777> if you do).	358	and C<0666> or C<0777> if you do). Note that the C<$mode> will be modified
		359	by the umask in effect then the request is being executed, so better never
		360	change the umask.
141		361
142	Example:	362	Example:
143		363
144	aio_open "/etc/passwd", O_RDONLY, 0, sub {	364	aio_open "/etc/passwd", IO::AIO::O_RDONLY, 0, sub {
145	if ($_[0]) {	365	if ($_[0]) {
146	print "open successful, fh is $_[0]\n";	366	print "open successful, fh is $_[0]\n";
147	...	367	...
148	} else {	368	} else {
149	die "open failed: $!\n";	369	die "open failed: $!\n";
150	}	370	}
151	};	371	};
152		372
		373	In addition to all the common open modes/flags (C<O_RDONLY>, C<O_WRONLY>,
		374	C<O_RDWR>, C<O_CREAT>, C<O_TRUNC>, C<O_EXCL> and C<O_APPEND>), the
		375	following POSIX and non-POSIX constants are available (missing ones on
		376	your system are, as usual, C<0>):
		377
		378	C<O_ASYNC>, C<O_DIRECT>, C<O_NOATIME>, C<O_CLOEXEC>, C<O_NOCTTY>, C<O_NOFOLLOW>,
		379	C<O_NONBLOCK>, C<O_EXEC>, C<O_SEARCH>, C<O_DIRECTORY>, C<O_DSYNC>,
		380	C<O_RSYNC>, C<O_SYNC> and C<O_TTY_INIT>.
		381
		382
153	=item aio_close $fh, $callback->($status)	383	=item aio_close $fh, $callback->($status)
154		384
155	Asynchronously close a file and call the callback with the result	385	Asynchronously close a file and call the callback with the result
156	code. I<WARNING:> although accepted, you should not pass in a perl	386	code.
157	filehandle here, as perl will likely close the file descriptor another
158	time when the filehandle is destroyed. Normally, you can safely call perls
159	C<close> or just let filehandles go out of scope.
160		387
161	This is supposed to be a bug in the API, so that might change. It's	388	Unfortunately, you can't do this to perl. Perl I<insists> very strongly on
162	therefore best to avoid this function.	389	closing the file descriptor associated with the filehandle itself.
		390
		391	Therefore, C<aio_close> will not close the filehandle - instead it will
		392	use dup2 to overwrite the file descriptor with the write-end of a pipe
		393	(the pipe fd will be created on demand and will be cached).
		394
		395	Or in other words: the file descriptor will be closed, but it will not be
		396	free for reuse until the perl filehandle is closed.
		397
		398	=cut
163		399
164	=item aio_read $fh,$offset,$length, $data,$dataoffset, $callback->($retval)	400	=item aio_read $fh,$offset,$length, $data,$dataoffset, $callback->($retval)
165		401
166	=item aio_write $fh,$offset,$length, $data,$dataoffset, $callback->($retval)	402	=item aio_write $fh,$offset,$length, $data,$dataoffset, $callback->($retval)
167		403
168	Reads or writes C<length> bytes from the specified C<fh> and C<offset>	404	Reads or writes C<$length> bytes from or to the specified C<$fh> and
169	into the scalar given by C<data> and offset C<dataoffset> and calls the	405	C<$offset> into the scalar given by C<$data> and offset C<$dataoffset>
170	callback without the actual number of bytes read (or -1 on error, just	406	and calls the callback without the actual number of bytes read (or -1 on
171	like the syscall).	407	error, just like the syscall).
		408
		409	C<aio_read> will, like C<sysread>, shrink or grow the C<$data> scalar to
		410	offset plus the actual number of bytes read.
		411
		412	If C<$offset> is undefined, then the current file descriptor offset will
		413	be used (and updated), otherwise the file descriptor offset will not be
		414	changed by these calls.
		415
		416	If C<$length> is undefined in C<aio_write>, use the remaining length of
		417	C<$data>.
		418
		419	If C<$dataoffset> is less than zero, it will be counted from the end of
		420	C<$data>.
172		421
173	The C<$data> scalar I<MUST NOT> be modified in any way while the request	422	The C<$data> scalar I<MUST NOT> be modified in any way while the request
174	is outstanding. Modifying it can result in segfaults or WW3 (if the	423	is outstanding. Modifying it can result in segfaults or World War III (if
175	necessary/optional hardware is installed).	424	the necessary/optional hardware is installed).
176		425
177	Example: Read 15 bytes at offset 7 into scalar C<$buffer>, starting at	426	Example: Read 15 bytes at offset 7 into scalar C<$buffer>, starting at
178	offset C<0> within the scalar:	427	offset C<0> within the scalar:
179		428
180	aio_read $fh, 7, 15, $buffer, 0, sub {	429	aio_read $fh, 7, 15, $buffer, 0, sub {
181	$_[0] > 0 or die "read error: $!";	430	$_[0] > 0 or die "read error: $!";
182	print "read $_[0] bytes: <$buffer>\n";	431	print "read $_[0] bytes: <$buffer>\n";
183	};	432	};
184		433
185	=item aio_move $srcpath, $dstpath, $callback->($status)
186
187	[EXPERIMENTAL due to internal aio_group use]
188
189	Try to move the I<file> (directories not supported as either source or
190	destination) from C<$srcpath> to C<$dstpath> and call the callback with
191	the C<0> (error) or C<-1> ok.
192
193	This is a composite request that tries to rename(2) the file first. If
194	rename files with C<EXDEV>, it creates the destination file with mode 0200
195	and copies the contents of the source file into it using C<aio_sendfile>,
196	followed by restoring atime, mtime, access mode and uid/gid, in that
197	order, and unlinking the C<$srcpath>.
198
199	If an error occurs, the partial destination file will be unlinked, if
200	possible, except when setting atime, mtime, access mode and uid/gid, where
201	errors are being ignored.
202
203	=cut
204
205	sub aio_move($$$) {
206	my ($src, $dst, $cb) = @_;
207
208	my $grp = aio_group $cb;
209
210	add $grp aio_rename $src, $dst, sub {
211	if ($_[0] && $! == EXDEV) {
212	add $grp aio_open $src, O_RDONLY, 0, sub {
213	if (my $src_fh = $_[0]) {
214	my @stat = stat $src_fh;
215
216	add $grp aio_open $dst, O_WRONLY, 0200, sub {
217	if (my $dst_fh = $_[0]) {
218	add $grp aio_sendfile $dst_fh, $src_fh, 0, $stat[7], sub {
219	close $src_fh;
220
221	if ($_[0] == $stat[7]) {
222	utime $stat[8], $stat[9], $dst;
223	chmod $stat[2] & 07777, $dst_fh;
224	chown $stat[4], $stat[5], $dst_fh;
225	close $dst_fh;
226
227	add $grp aio_unlink $src, sub {
228	$grp->result ($_[0]);
229	};
230	} else {
231	my $errno = $!;
232	add $grp aio_unlink $dst, sub {
233	$! = $errno;
234	$grp->result (-1);
235	};
236	}
237	};
238	} else {
239	$grp->result (-1);
240	}
241	},
242
243	} else {
244	$grp->result (-1);
245	}
246	};
247	} else {
248	$grp->result ($_[0]);
249	}
250	};
251
252	$grp
253	}
254		434
255	=item aio_sendfile $out_fh, $in_fh, $in_offset, $length, $callback->($retval)	435	=item aio_sendfile $out_fh, $in_fh, $in_offset, $length, $callback->($retval)
256		436
257	Tries to copy C<$length> bytes from C<$in_fh> to C<$out_fh>. It starts	437	Tries to copy C<$length> bytes from C<$in_fh> to C<$out_fh>. It starts
258	reading at byte offset C<$in_offset>, and starts writing at the current	438	reading at byte offset C<$in_offset>, and starts writing at the current
259	file offset of C<$out_fh>. Because of that, it is not safe to issue more	439	file offset of C<$out_fh>. Because of that, it is not safe to issue more
260	than one C<aio_sendfile> per C<$out_fh>, as they will interfere with each	440	than one C<aio_sendfile> per C<$out_fh>, as they will interfere with each
261	other.	441	other.
262		442
		443	Please note that C<aio_sendfile> can read more bytes from C<$in_fh> than
		444	are written, and there is no way to find out how many bytes have been read
		445	from C<aio_sendfile> alone, as C<aio_sendfile> only provides the number of
		446	bytes written to C<$out_fh>. Only if the result value equals C<$length>
		447	one can assume that C<$length> bytes have been read.
		448
		449	Unlike with other C<aio_> functions, it makes a lot of sense to use
		450	C<aio_sendfile> on non-blocking sockets, as long as one end (typically
		451	the C<$in_fh>) is a file - the file I/O will then be asynchronous, while
		452	the socket I/O will be non-blocking. Note, however, that you can run into
		453	a trap where C<aio_sendfile> reads some data with readahead, then fails
		454	to write all data, and when the socket is ready the next time, the data
		455	in the cache is already lost, forcing C<aio_sendfile> to again hit the
		456	disk. Explicit C<aio_read> + C<aio_write> let's you control resource usage
		457	much better.
		458
263	This call tries to make use of a native C<sendfile> syscall to provide	459	This call tries to make use of a native C<sendfile> syscall to provide
264	zero-copy operation. For this to work, C<$out_fh> should refer to a	460	zero-copy operation. For this to work, C<$out_fh> should refer to a
265	socket, and C<$in_fh> should refer to mmap'able file.	461	socket, and C<$in_fh> should refer to an mmap'able file.
266		462
267	If the native sendfile call fails or is not implemented, it will be	463	If a native sendfile cannot be found or it fails with C<ENOSYS>,
		464	C<ENOTSUP>, C<EOPNOTSUPP>, C<EAFNOSUPPORT>, C<EPROTOTYPE> or C<ENOTSOCK>,
268	emulated, so you can call C<aio_sendfile> on any type of filehandle	465	it will be emulated, so you can call C<aio_sendfile> on any type of
269	regardless of the limitations of the operating system.	466	filehandle regardless of the limitations of the operating system.
270		467
271	Please note, however, that C<aio_sendfile> can read more bytes from
272	C<$in_fh> than are written, and there is no way to find out how many
273	bytes have been read from C<aio_sendfile> alone, as C<aio_sendfile> only
274	provides the number of bytes written to C<$out_fh>. Only if the result
275	value equals C<$length> one can assume that C<$length> bytes have been
276	read.
277		468
278	=item aio_readahead $fh,$offset,$length, $callback->($retval)	469	=item aio_readahead $fh,$offset,$length, $callback->($retval)
279		470
280	C<aio_readahead> populates the page cache with data from a file so that	471	C<aio_readahead> populates the page cache with data from a file so that
281	subsequent reads from that file will not block on disk I/O. The C<$offset>	472	subsequent reads from that file will not block on disk I/O. The C<$offset>
…		…
287	file. The current file offset of the file is left unchanged.	478	file. The current file offset of the file is left unchanged.
288		479
289	If that syscall doesn't exist (likely if your OS isn't Linux) it will be	480	If that syscall doesn't exist (likely if your OS isn't Linux) it will be
290	emulated by simply reading the data, which would have a similar effect.	481	emulated by simply reading the data, which would have a similar effect.
291		482
		483
292	=item aio_stat $fh_or_path, $callback->($status)	484	=item aio_stat $fh_or_path, $callback->($status)
293		485
294	=item aio_lstat $fh, $callback->($status)	486	=item aio_lstat $fh, $callback->($status)
295		487
296	Works like perl's C<stat> or C<lstat> in void context. The callback will	488	Works like perl's C<stat> or C<lstat> in void context. The callback will
…		…
301	for an explanation.	493	for an explanation.
302		494
303	Currently, the stats are always 64-bit-stats, i.e. instead of returning an	495	Currently, the stats are always 64-bit-stats, i.e. instead of returning an
304	error when stat'ing a large file, the results will be silently truncated	496	error when stat'ing a large file, the results will be silently truncated
305	unless perl itself is compiled with large file support.	497	unless perl itself is compiled with large file support.
		498
		499	To help interpret the mode and dev/rdev stat values, IO::AIO offers the
		500	following constants and functions (if not implemented, the constants will
		501	be C<0> and the functions will either C<croak> or fall back on traditional
		502	behaviour).
		503
		504	C<S_IFMT>, C<S_IFIFO>, C<S_IFCHR>, C<S_IFBLK>, C<S_IFLNK>, C<S_IFREG>,
		505	C<S_IFDIR>, C<S_IFWHT>, C<S_IFSOCK>, C<IO::AIO::major $dev_t>,
		506	C<IO::AIO::minor $dev_t>, C<IO::AIO::makedev $major, $minor>.
306		507
307	Example: Print the length of F</etc/passwd>:	508	Example: Print the length of F</etc/passwd>:
308		509
309	aio_stat "/etc/passwd", sub {	510	aio_stat "/etc/passwd", sub {
310	$_[0] and die "stat failed: $!";	511	$_[0] and die "stat failed: $!";
311	print "size is ", -s _, "\n";	512	print "size is ", -s _, "\n";
312	};	513	};
313		514
		515
		516	=item aio_statvfs $fh_or_path, $callback->($statvfs)
		517
		518	Works like the POSIX C<statvfs> or C<fstatvfs> syscalls, depending on
		519	whether a file handle or path was passed.
		520
		521	On success, the callback is passed a hash reference with the following
		522	members: C<bsize>, C<frsize>, C<blocks>, C<bfree>, C<bavail>, C<files>,
		523	C<ffree>, C<favail>, C<fsid>, C<flag> and C<namemax>. On failure, C<undef>
		524	is passed.
		525
		526	The following POSIX IO::AIO::ST_* constants are defined: C<ST_RDONLY> and
		527	C<ST_NOSUID>.
		528
		529	The following non-POSIX IO::AIO::ST_* flag masks are defined to
		530	their correct value when available, or to C<0> on systems that do
		531	not support them: C<ST_NODEV>, C<ST_NOEXEC>, C<ST_SYNCHRONOUS>,
		532	C<ST_MANDLOCK>, C<ST_WRITE>, C<ST_APPEND>, C<ST_IMMUTABLE>, C<ST_NOATIME>,
		533	C<ST_NODIRATIME> and C<ST_RELATIME>.
		534
		535	Example: stat C</wd> and dump out the data if successful.
		536
		537	aio_statvfs "/wd", sub {
		538	my $f = $_[0]
		539	or die "statvfs: $!";
		540
		541	use Data::Dumper;
		542	say Dumper $f;
		543	};
		544
		545	# result:
		546	{
		547	bsize => 1024,
		548	bfree => 4333064312,
		549	blocks => 10253828096,
		550	files => 2050765568,
		551	flag => 4096,
		552	favail => 2042092649,
		553	bavail => 4333064312,
		554	ffree => 2042092649,
		555	namemax => 255,
		556	frsize => 1024,
		557	fsid => 1810
		558	}
		559
		560
		561	=item aio_utime $fh_or_path, $atime, $mtime, $callback->($status)
		562
		563	Works like perl's C<utime> function (including the special case of $atime
		564	and $mtime being undef). Fractional times are supported if the underlying
		565	syscalls support them.
		566
		567	When called with a pathname, uses utimes(2) if available, otherwise
		568	utime(2). If called on a file descriptor, uses futimes(2) if available,
		569	otherwise returns ENOSYS, so this is not portable.
		570
		571	Examples:
		572
		573	# set atime and mtime to current time (basically touch(1)):
		574	aio_utime "path", undef, undef;
		575	# set atime to current time and mtime to beginning of the epoch:
		576	aio_utime "path", time, undef; # undef==0
		577
		578
		579	=item aio_chown $fh_or_path, $uid, $gid, $callback->($status)
		580
		581	Works like perl's C<chown> function, except that C<undef> for either $uid
		582	or $gid is being interpreted as "do not change" (but -1 can also be used).
		583
		584	Examples:
		585
		586	# same as "chown root path" in the shell:
		587	aio_chown "path", 0, -1;
		588	# same as above:
		589	aio_chown "path", 0, undef;
		590
		591
		592	=item aio_truncate $fh_or_path, $offset, $callback->($status)
		593
		594	Works like truncate(2) or ftruncate(2).
		595
		596
		597	=item aio_chmod $fh_or_path, $mode, $callback->($status)
		598
		599	Works like perl's C<chmod> function.
		600
		601
314	=item aio_unlink $pathname, $callback->($status)	602	=item aio_unlink $pathname, $callback->($status)
315		603
316	Asynchronously unlink (delete) a file and call the callback with the	604	Asynchronously unlink (delete) a file and call the callback with the
317	result code.	605	result code.
318		606
		607
		608	=item aio_mknod $path, $mode, $dev, $callback->($status)
		609
		610	[EXPERIMENTAL]
		611
		612	Asynchronously create a device node (or fifo). See mknod(2).
		613
		614	The only (POSIX-) portable way of calling this function is:
		615
		616	aio_mknod $path, IO::AIO::S_IFIFO | $mode, 0, sub { ...
		617
		618	See C<aio_stat> for info about some potentially helpful extra constants
		619	and functions.
		620
319	=item aio_link $srcpath, $dstpath, $callback->($status)	621	=item aio_link $srcpath, $dstpath, $callback->($status)
320		622
321	Asynchronously create a new link to the existing object at C<$srcpath> at	623	Asynchronously create a new link to the existing object at C<$srcpath> at
322	the path C<$dstpath> and call the callback with the result code.	624	the path C<$dstpath> and call the callback with the result code.
323		625
		626
324	=item aio_symlink $srcpath, $dstpath, $callback->($status)	627	=item aio_symlink $srcpath, $dstpath, $callback->($status)
325		628
326	Asynchronously create a new symbolic link to the existing object at C<$srcpath> at	629	Asynchronously create a new symbolic link to the existing object at C<$srcpath> at
327	the path C<$dstpath> and call the callback with the result code.	630	the path C<$dstpath> and call the callback with the result code.
328		631
		632
		633	=item aio_readlink $path, $callback->($link)
		634
		635	Asynchronously read the symlink specified by C<$path> and pass it to
		636	the callback. If an error occurs, nothing or undef gets passed to the
		637	callback.
		638
		639
329	=item aio_rename $srcpath, $dstpath, $callback->($status)	640	=item aio_rename $srcpath, $dstpath, $callback->($status)
330		641
331	Asynchronously rename the object at C<$srcpath> to C<$dstpath>, just as	642	Asynchronously rename the object at C<$srcpath> to C<$dstpath>, just as
332	rename(2) and call the callback with the result code.	643	rename(2) and call the callback with the result code.
333		644
		645
		646	=item aio_mkdir $pathname, $mode, $callback->($status)
		647
		648	Asynchronously mkdir (create) a directory and call the callback with
		649	the result code. C<$mode> will be modified by the umask at the time the
		650	request is executed, so do not change your umask.
		651
		652
334	=item aio_rmdir $pathname, $callback->($status)	653	=item aio_rmdir $pathname, $callback->($status)
335		654
336	Asynchronously rmdir (delete) a directory and call the callback with the	655	Asynchronously rmdir (delete) a directory and call the callback with the
337	result code.	656	result code.
		657
338		658
339	=item aio_readdir $pathname, $callback->($entries)	659	=item aio_readdir $pathname, $callback->($entries)
340		660
341	Unlike the POSIX call of the same name, C<aio_readdir> reads an entire	661	Unlike the POSIX call of the same name, C<aio_readdir> reads an entire
342	directory (i.e. opendir + readdir + closedir). The entries will not be	662	directory (i.e. opendir + readdir + closedir). The entries will not be
343	sorted, and will B<NOT> include the C<.> and C<..> entries.	663	sorted, and will B<NOT> include the C<.> and C<..> entries.
344		664
345	The callback a single argument which is either C<undef> or an array-ref	665	The callback is passed a single argument which is either C<undef> or an
346	with the filenames.	666	array-ref with the filenames.
		667
		668
		669	=item aio_readdirx $pathname, $flags, $callback->($entries, $flags)
		670
		671	Quite similar to C<aio_readdir>, but the C<$flags> argument allows to tune
		672	behaviour and output format. In case of an error, C<$entries> will be
		673	C<undef>.
		674
		675	The flags are a combination of the following constants, ORed together (the
		676	flags will also be passed to the callback, possibly modified):
		677
		678	=over 4
		679
		680	=item IO::AIO::READDIR_DENTS
		681
		682	When this flag is off, then the callback gets an arrayref consisting of
		683	names only (as with C<aio_readdir>), otherwise it gets an arrayref with
		684	C<[$name, $type, $inode]> arrayrefs, each describing a single directory
		685	entry in more detail.
		686
		687	C<$name> is the name of the entry.
		688
		689	C<$type> is one of the C<IO::AIO::DT_xxx> constants:
		690
		691	C<IO::AIO::DT_UNKNOWN>, C<IO::AIO::DT_FIFO>, C<IO::AIO::DT_CHR>, C<IO::AIO::DT_DIR>,
		692	C<IO::AIO::DT_BLK>, C<IO::AIO::DT_REG>, C<IO::AIO::DT_LNK>, C<IO::AIO::DT_SOCK>,
		693	C<IO::AIO::DT_WHT>.
		694
		695	C<IO::AIO::DT_UNKNOWN> means just that: readdir does not know. If you need to
		696	know, you have to run stat yourself. Also, for speed reasons, the C<$type>
		697	scalars are read-only: you can not modify them.
		698
		699	C<$inode> is the inode number (which might not be exact on systems with 64
		700	bit inode numbers and 32 bit perls). This field has unspecified content on
		701	systems that do not deliver the inode information.
		702
		703	=item IO::AIO::READDIR_DIRS_FIRST
		704
		705	When this flag is set, then the names will be returned in an order where
		706	likely directories come first, in optimal stat order. This is useful when
		707	you need to quickly find directories, or you want to find all directories
		708	while avoiding to stat() each entry.
		709
		710	If the system returns type information in readdir, then this is used
		711	to find directories directly. Otherwise, likely directories are names
		712	beginning with ".", or otherwise names with no dots, of which names with
		713	short names are tried first.
		714
		715	=item IO::AIO::READDIR_STAT_ORDER
		716
		717	When this flag is set, then the names will be returned in an order
		718	suitable for stat()'ing each one. That is, when you plan to stat()
		719	all files in the given directory, then the returned order will likely
		720	be fastest.
		721
		722	If both this flag and C<IO::AIO::READDIR_DIRS_FIRST> are specified, then
		723	the likely dirs come first, resulting in a less optimal stat order.
		724
		725	=item IO::AIO::READDIR_FOUND_UNKNOWN
		726
		727	This flag should not be set when calling C<aio_readdirx>. Instead, it
		728	is being set by C<aio_readdirx>, when any of the C<$type>'s found were
		729	C<IO::AIO::DT_UNKNOWN>. The absense of this flag therefore indicates that all
		730	C<$type>'s are known, which can be used to speed up some algorithms.
		731
		732	=back
		733
		734
		735	=item aio_load $path, $data, $callback->($status)
		736
		737	This is a composite request that tries to fully load the given file into
		738	memory. Status is the same as with aio_read.
		739
		740	=cut
		741
		742	sub aio_load($$;$) {
		743	my ($path, undef, $cb) = @_;
		744	my $data = \$_[1];
		745
		746	my $pri = aioreq_pri;
		747	my $grp = aio_group $cb;
		748
		749	aioreq_pri $pri;
		750	add $grp aio_open $path, O_RDONLY, 0, sub {
		751	my $fh = shift
		752	or return $grp->result (-1);
		753
		754	aioreq_pri $pri;
		755	add $grp aio_read $fh, 0, (-s $fh), $$data, 0, sub {
		756	$grp->result ($_[0]);
		757	};
		758	};
		759
		760	$grp
		761	}
		762
		763	=item aio_copy $srcpath, $dstpath, $callback->($status)
		764
		765	Try to copy the I<file> (directories not supported as either source or
		766	destination) from C<$srcpath> to C<$dstpath> and call the callback with
		767	a status of C<0> (ok) or C<-1> (error, see C<$!>).
		768
		769	This is a composite request that creates the destination file with
		770	mode 0200 and copies the contents of the source file into it using
		771	C<aio_sendfile>, followed by restoring atime, mtime, access mode and
		772	uid/gid, in that order.
		773
		774	If an error occurs, the partial destination file will be unlinked, if
		775	possible, except when setting atime, mtime, access mode and uid/gid, where
		776	errors are being ignored.
		777
		778	=cut
		779
		780	sub aio_copy($$;$) {
		781	my ($src, $dst, $cb) = @_;
		782
		783	my $pri = aioreq_pri;
		784	my $grp = aio_group $cb;
		785
		786	aioreq_pri $pri;
		787	add $grp aio_open $src, O_RDONLY, 0, sub {
		788	if (my $src_fh = $_[0]) {
		789	my @stat = stat $src_fh; # hmm, might block over nfs?
		790
		791	aioreq_pri $pri;
		792	add $grp aio_open $dst, O_CREAT | O_WRONLY | O_TRUNC, 0200, sub {
		793	if (my $dst_fh = $_[0]) {
		794	aioreq_pri $pri;
		795	add $grp aio_sendfile $dst_fh, $src_fh, 0, $stat[7], sub {
		796	if ($_[0] == $stat[7]) {
		797	$grp->result (0);
		798	close $src_fh;
		799
		800	my $ch = sub {
		801	aioreq_pri $pri;
		802	add $grp aio_chmod $dst_fh, $stat[2] & 07777, sub {
		803	aioreq_pri $pri;
		804	add $grp aio_chown $dst_fh, $stat[4], $stat[5], sub {
		805	aioreq_pri $pri;
		806	add $grp aio_close $dst_fh;
		807	}
		808	};
		809	};
		810
		811	aioreq_pri $pri;
		812	add $grp aio_utime $dst_fh, $stat[8], $stat[9], sub {
		813	if ($_[0] < 0 && $! == ENOSYS) {
		814	aioreq_pri $pri;
		815	add $grp aio_utime $dst, $stat[8], $stat[9], $ch;
		816	} else {
		817	$ch->();
		818	}
		819	};
		820	} else {
		821	$grp->result (-1);
		822	close $src_fh;
		823	close $dst_fh;
		824
		825	aioreq $pri;
		826	add $grp aio_unlink $dst;
		827	}
		828	};
		829	} else {
		830	$grp->result (-1);
		831	}
		832	},
		833
		834	} else {
		835	$grp->result (-1);
		836	}
		837	};
		838
		839	$grp
		840	}
		841
		842	=item aio_move $srcpath, $dstpath, $callback->($status)
		843
		844	Try to move the I<file> (directories not supported as either source or
		845	destination) from C<$srcpath> to C<$dstpath> and call the callback with
		846	a status of C<0> (ok) or C<-1> (error, see C<$!>).
		847
		848	This is a composite request that tries to rename(2) the file first; if
		849	rename fails with C<EXDEV>, it copies the file with C<aio_copy> and, if
		850	that is successful, unlinks the C<$srcpath>.
		851
		852	=cut
		853
		854	sub aio_move($$;$) {
		855	my ($src, $dst, $cb) = @_;
		856
		857	my $pri = aioreq_pri;
		858	my $grp = aio_group $cb;
		859
		860	aioreq_pri $pri;
		861	add $grp aio_rename $src, $dst, sub {
		862	if ($_[0] && $! == EXDEV) {
		863	aioreq_pri $pri;
		864	add $grp aio_copy $src, $dst, sub {
		865	$grp->result ($_[0]);
		866
		867	if (!$_[0]) {
		868	aioreq_pri $pri;
		869	add $grp aio_unlink $src;
		870	}
		871	};
		872	} else {
		873	$grp->result ($_[0]);
		874	}
		875	};
		876
		877	$grp
		878	}
347		879
348	=item aio_scandir $path, $maxreq, $callback->($dirs, $nondirs)	880	=item aio_scandir $path, $maxreq, $callback->($dirs, $nondirs)
349		881
350	[EXPERIMENTAL due to internal aio_group use]
351
352	Scans a directory (similar to C<aio_readdir>) but additionally tries to	882	Scans a directory (similar to C<aio_readdir>) but additionally tries to
353	separate the entries of directory C<$path> into two sets of names, ones	883	efficiently separate the entries of directory C<$path> into two sets of
354	you can recurse into (directories or links to them), and ones you cannot	884	names, directories you can recurse into (directories), and ones you cannot
355	recurse into (everything else).	885	recurse into (everything else, including symlinks to directories).
356		886
357	C<aio_scandir> is a composite request that creates of many sub requests_	887	C<aio_scandir> is a composite request that creates of many sub requests_
358	C<$maxreq> specifies the maximum number of outstanding aio requests that	888	C<$maxreq> specifies the maximum number of outstanding aio requests that
359	this function generates. If it is C<< <= 0 >>, then a suitable default	889	this function generates. If it is C<< <= 0 >>, then a suitable default
360	will be chosen (currently 6).	890	will be chosen (currently 4).
361		891
362	On error, the callback is called without arguments, otherwise it receives	892	On error, the callback is called without arguments, otherwise it receives
363	two array-refs with path-relative entry names.	893	two array-refs with path-relative entry names.
364		894
365	Example:	895	Example:
…		…
372		902
373	Implementation notes.	903	Implementation notes.
374		904
375	The C<aio_readdir> cannot be avoided, but C<stat()>'ing every entry can.	905	The C<aio_readdir> cannot be avoided, but C<stat()>'ing every entry can.
376		906
		907	If readdir returns file type information, then this is used directly to
		908	find directories.
		909
377	After reading the directory, the modification time, size etc. of the	910	Otherwise, after reading the directory, the modification time, size etc.
378	directory before and after the readdir is checked, and if they match (and	911	of the directory before and after the readdir is checked, and if they
379	isn't the current time), the link count will be used to decide how many	912	match (and isn't the current time), the link count will be used to decide
380	entries are directories (if >= 2). Otherwise, no knowledge of the number	913	how many entries are directories (if >= 2). Otherwise, no knowledge of the
381	of subdirectories will be assumed.	914	number of subdirectories will be assumed.
382		915
383	Then entries will be sorted into likely directories (everything without	916	Then entries will be sorted into likely directories a non-initial dot
384	a non-initial dot currently) and likely non-directories (everything	917	currently) and likely non-directories (see C<aio_readdirx>). Then every
385	else). Then every entry plus an appended C</.> will be C<stat>'ed,	918	entry plus an appended C</.> will be C<stat>'ed, likely directories first,
386	likely directories first. If that succeeds, it assumes that the entry	919	in order of their inode numbers. If that succeeds, it assumes that the
387	is a directory or a symlink to directory (which will be checked	920	entry is a directory or a symlink to directory (which will be checked
388	seperately). This is often faster than stat'ing the entry itself because	921	seperately). This is often faster than stat'ing the entry itself because
389	filesystems might detect the type of the entry without reading the inode	922	filesystems might detect the type of the entry without reading the inode
390	data (e.g. ext2fs filetype feature).	923	data (e.g. ext2fs filetype feature), even on systems that cannot return
		924	the filetype information on readdir.
391		925
392	If the known number of directories (link count - 2) has been reached, the	926	If the known number of directories (link count - 2) has been reached, the
393	rest of the entries is assumed to be non-directories.	927	rest of the entries is assumed to be non-directories.
394		928
395	This only works with certainty on POSIX (= UNIX) filesystems, which	929	This only works with certainty on POSIX (= UNIX) filesystems, which
…		…
399	as those tend to return 0 or 1 as link counts, which disables the	933	as those tend to return 0 or 1 as link counts, which disables the
400	directory counting heuristic.	934	directory counting heuristic.
401		935
402	=cut	936	=cut
403		937
404	sub aio_scandir($$$) {	938	sub aio_scandir($$;$) {
405	my ($path, $maxreq, $cb) = @_;	939	my ($path, $maxreq, $cb) = @_;
406		940
		941	my $pri = aioreq_pri;
		942
407	my $grp = aio_group $cb;	943	my $grp = aio_group $cb;
408		944
409	$maxreq = 6 if $maxreq <= 0;	945	$maxreq = 4 if $maxreq <= 0;
410		946
411	# stat once	947	# stat once
		948	aioreq_pri $pri;
412	add $grp aio_stat $path, sub {	949	add $grp aio_stat $path, sub {
413	return $grp->result () if $_[0];	950	return $grp->result () if $_[0];
414	my $now = time;	951	my $now = time;
415	my $hash1 = join ":", (stat _)[0,1,3,7,9];	952	my $hash1 = join ":", (stat _)[0,1,3,7,9];
416		953
417	# read the directory entries	954	# read the directory entries
		955	aioreq_pri $pri;
418	add $grp aio_readdir $path, sub {	956	add $grp aio_readdirx $path, READDIR_DIRS_FIRST, sub {
419	my $entries = shift	957	my $entries = shift
420	or return $grp->result ();	958	or return $grp->result ();
421		959
422	# stat the dir another time	960	# stat the dir another time
		961	aioreq_pri $pri;
423	add $grp aio_stat $path, sub {	962	add $grp aio_stat $path, sub {
424	my $hash2 = join ":", (stat _)[0,1,3,7,9];	963	my $hash2 = join ":", (stat _)[0,1,3,7,9];
425		964
426	my $ndirs;	965	my $ndirs;
427		966
428	# take the slow route if anything looks fishy	967	# take the slow route if anything looks fishy
429	if ($hash1 ne $hash2 or (stat _)[9] == $now) {	968	if ($hash1 ne $hash2 or (stat _)[9] == $now) {
430	$ndirs = -1;	969	$ndirs = -1;
431	} else {	970	} else {
432	# if nlink == 2, we are finished	971	# if nlink == 2, we are finished
433	# on non-posix-fs's, we rely on nlink < 2	972	# for non-posix-fs's, we rely on nlink < 2
434	$ndirs = (stat _)[3] - 2	973	$ndirs = (stat _)[3] - 2
435	or return $grp->result ([], $entries);	974	or return $grp->result ([], $entries);
436	}	975	}
437		976
438	# sort into likely dirs and likely nondirs
439	# dirs == files without ".", short entries first
440	$entries = [map $_->[0],
441	sort { $b->[1] cmp $a->[1] }
442	map [$_, sprintf "%s%04d", (/.\./ ? "1" : "0"), length],
443	@$entries];
444
445	my (@dirs, @nondirs);	977	my (@dirs, @nondirs);
446		978
447	my ($statcb, $schedcb);
448	my $nreq = 0;
449
450	my $statgrp = add $grp aio_group;	979	my $statgrp = add $grp aio_group sub {
		980	$grp->result (\@dirs, \@nondirs);
		981	};
451		982
452	$schedcb = sub {	983	limit $statgrp $maxreq;
453	if (@$entries) {	984	feed $statgrp sub {
454	if ($nreq < $maxreq) {	985	return unless @$entries;
455	my $ent = pop @$entries;	986	my $entry = shift @$entries;
		987
		988	aioreq_pri $pri;
		989	add $statgrp aio_stat "$path/$entry/.", sub {
		990	if ($_[0] < 0) {
		991	push @nondirs, $entry;
		992	} else {
		993	# need to check for real directory
		994	aioreq_pri $pri;
		995	add $statgrp aio_lstat "$path/$entry", sub {
		996	if (-d _) {
		997	push @dirs, $entry;
		998
		999	unless (--$ndirs) {
		1000	push @nondirs, @$entries;
		1001	feed $statgrp;
		1002	}
		1003	} else {
		1004	push @nondirs, $entry;
		1005	}
456	$nreq++;	1006	}
457	add $statgrp aio_stat "$path/$ent/.", sub { $statcb->($_[0], $ent) };
458	}	1007	}
459	} elsif (!$nreq) {
460	# finished
461	$statgrp->cancel;
462	undef $statcb;
463	undef $schedcb;
464	$grp->result (\@dirs, \@nondirs);
465	}	1008	};
466	};	1009	};
467	$statcb = sub {
468	my ($status, $entry) = @_;
469
470	if ($status < 0) {
471	$nreq--;
472	push @nondirs, $entry;
473	&$schedcb;
474	} else {
475	# need to check for real directory
476	add $grp aio_lstat "$path/$entry", sub {
477	$nreq--;
478
479	if (-d _) {
480	push @dirs, $entry;
481
482	if (!--$ndirs) {
483	push @nondirs, @$entries;
484	$entries = [];
485	}
486	} else {
487	push @nondirs, $entry;
488	}
489
490	&$schedcb;
491	}
492	}
493	};
494
495	&$schedcb while @$entries && $nreq < $maxreq;
496	};	1010	};
497	};	1011	};
498	};	1012	};
499		1013
500	$grp	1014	$grp
501	}	1015	}
502		1016
		1017	=item aio_rmtree $path, $callback->($status)
		1018
		1019	Delete a directory tree starting (and including) C<$path>, return the
		1020	status of the final C<rmdir> only. This is a composite request that
		1021	uses C<aio_scandir> to recurse into and rmdir directories, and unlink
		1022	everything else.
		1023
		1024	=cut
		1025
		1026	sub aio_rmtree;
		1027	sub aio_rmtree($;$) {
		1028	my ($path, $cb) = @_;
		1029
		1030	my $pri = aioreq_pri;
		1031	my $grp = aio_group $cb;
		1032
		1033	aioreq_pri $pri;
		1034	add $grp aio_scandir $path, 0, sub {
		1035	my ($dirs, $nondirs) = @_;
		1036
		1037	my $dirgrp = aio_group sub {
		1038	add $grp aio_rmdir $path, sub {
		1039	$grp->result ($_[0]);
		1040	};
		1041	};
		1042
		1043	(aioreq_pri $pri), add $dirgrp aio_rmtree "$path/$_" for @$dirs;
		1044	(aioreq_pri $pri), add $dirgrp aio_unlink "$path/$_" for @$nondirs;
		1045
		1046	add $grp $dirgrp;
		1047	};
		1048
		1049	$grp
		1050	}
		1051
		1052	=item aio_sync $callback->($status)
		1053
		1054	Asynchronously call sync and call the callback when finished.
		1055
503	=item aio_fsync $fh, $callback->($status)	1056	=item aio_fsync $fh, $callback->($status)
504		1057
505	Asynchronously call fsync on the given filehandle and call the callback	1058	Asynchronously call fsync on the given filehandle and call the callback
506	with the fsync result code.	1059	with the fsync result code.
507		1060
…		…
511	callback with the fdatasync result code.	1064	callback with the fdatasync result code.
512		1065
513	If this call isn't available because your OS lacks it or it couldn't be	1066	If this call isn't available because your OS lacks it or it couldn't be
514	detected, it will be emulated by calling C<fsync> instead.	1067	detected, it will be emulated by calling C<fsync> instead.
515		1068
		1069	=item aio_sync_file_range $fh, $offset, $nbytes, $flags, $callback->($status)
		1070
		1071	Sync the data portion of the file specified by C<$offset> and C<$length>
		1072	to disk (but NOT the metadata), by calling the Linux-specific
		1073	sync_file_range call. If sync_file_range is not available or it returns
		1074	ENOSYS, then fdatasync or fsync is being substituted.
		1075
		1076	C<$flags> can be a combination of C<IO::AIO::SYNC_FILE_RANGE_WAIT_BEFORE>,
		1077	C<IO::AIO::SYNC_FILE_RANGE_WRITE> and
		1078	C<IO::AIO::SYNC_FILE_RANGE_WAIT_AFTER>: refer to the sync_file_range
		1079	manpage for details.
		1080
		1081	=item aio_pathsync $path, $callback->($status)
		1082
		1083	This request tries to open, fsync and close the given path. This is a
		1084	composite request intended to sync directories after directory operations
		1085	(E.g. rename). This might not work on all operating systems or have any
		1086	specific effect, but usually it makes sure that directory changes get
		1087	written to disc. It works for anything that can be opened for read-only,
		1088	not just directories.
		1089
		1090	Future versions of this function might fall back to other methods when
		1091	C<fsync> on the directory fails (such as calling C<sync>).
		1092
		1093	Passes C<0> when everything went ok, and C<-1> on error.
		1094
		1095	=cut
		1096
		1097	sub aio_pathsync($;$) {
		1098	my ($path, $cb) = @_;
		1099
		1100	my $pri = aioreq_pri;
		1101	my $grp = aio_group $cb;
		1102
		1103	aioreq_pri $pri;
		1104	add $grp aio_open $path, O_RDONLY, 0, sub {
		1105	my ($fh) = @_;
		1106	if ($fh) {
		1107	aioreq_pri $pri;
		1108	add $grp aio_fsync $fh, sub {
		1109	$grp->result ($_[0]);
		1110
		1111	aioreq_pri $pri;
		1112	add $grp aio_close $fh;
		1113	};
		1114	} else {
		1115	$grp->result (-1);
		1116	}
		1117	};
		1118
		1119	$grp
		1120	}
		1121
		1122	=item aio_msync $scalar, $offset = 0, $length = undef, flags = 0, $callback->($status)
		1123
		1124	This is a rather advanced IO::AIO call, which only works on mmap(2)ed
		1125	scalars (see the C<IO::AIO::mmap> function, although it also works on data
		1126	scalars managed by the L<Sys::Mmap> or L<Mmap> modules, note that the
		1127	scalar must only be modified in-place while an aio operation is pending on
		1128	it).
		1129
		1130	It calls the C<msync> function of your OS, if available, with the memory
		1131	area starting at C<$offset> in the string and ending C<$length> bytes
		1132	later. If C<$length> is negative, counts from the end, and if C<$length>
		1133	is C<undef>, then it goes till the end of the string. The flags can be
		1134	a combination of C<IO::AIO::MS_ASYNC>, C<IO::AIO::MS_INVALIDATE> and
		1135	C<IO::AIO::MS_SYNC>.
		1136
		1137	=item aio_mtouch $scalar, $offset = 0, $length = undef, flags = 0, $callback->($status)
		1138
		1139	This is a rather advanced IO::AIO call, which works best on mmap(2)ed
		1140	scalars.
		1141
		1142	It touches (reads or writes) all memory pages in the specified
		1143	range inside the scalar. All caveats and parameters are the same
		1144	as for C<aio_msync>, above, except for flags, which must be either
		1145	C<0> (which reads all pages and ensures they are instantiated) or
		1146	C<IO::AIO::MT_MODIFY>, which modifies the memory page s(by reading and
		1147	writing an octet from it, which dirties the page).
		1148
		1149	=item aio_mlock $scalar, $offset = 0, $length = undef, $callback->($status)
		1150
		1151	This is a rather advanced IO::AIO call, which works best on mmap(2)ed
		1152	scalars.
		1153
		1154	It reads in all the pages of the underlying storage into memory (if any)
		1155	and locks them, so they are not getting swapped/paged out or removed.
		1156
		1157	If C<$length> is undefined, then the scalar will be locked till the end.
		1158
		1159	On systems that do not implement C<mlock>, this function returns C<-1>
		1160	and sets errno to C<ENOSYS>.
		1161
		1162	Note that the corresponding C<munlock> is synchronous and is
		1163	documented under L<MISCELLANEOUS FUNCTIONS>.
		1164
		1165	Example: open a file, mmap and mlock it - both will be undone when
		1166	C<$data> gets destroyed.
		1167
		1168	open my $fh, "<", $path or die "$path: $!";
		1169	my $data;
		1170	IO::AIO::mmap $data, -s $fh, IO::AIO::PROT_READ, IO::AIO::MAP_SHARED, $fh;
		1171	aio_mlock $data; # mlock in background
		1172
		1173	=item aio_mlockall $flags, $callback->($status)
		1174
		1175	Calls the C<mlockall> function with the given C<$flags> (a combination of
		1176	C<IO::AIO::MCL_CURRENT> and C<IO::AIO::MCL_FUTURE>).
		1177
		1178	On systems that do not implement C<mlockall>, this function returns C<-1>
		1179	and sets errno to C<ENOSYS>.
		1180
		1181	Note that the corresponding C<munlockall> is synchronous and is
		1182	documented under L<MISCELLANEOUS FUNCTIONS>.
		1183
		1184	Example: asynchronously lock all current and future pages into memory.
		1185
		1186	aio_mlockall IO::AIO::MCL_FUTURE;
		1187
516	=item aio_group $callback->(...)	1188	=item aio_group $callback->(...)
517
518	[EXPERIMENTAL]
519		1189
520	This is a very special aio request: Instead of doing something, it is a	1190	This is a very special aio request: Instead of doing something, it is a
521	container for other aio requests, which is useful if you want to bundle	1191	container for other aio requests, which is useful if you want to bundle
522	many requests into a single, composite, request.	1192	many requests into a single, composite, request with a definite callback
		1193	and the ability to cancel the whole request with its subrequests.
523		1194
524	Returns an object of class L<IO::AIO::GRP>. See its documentation below	1195	Returns an object of class L<IO::AIO::GRP>. See its documentation below
525	for more info.	1196	for more info.
526		1197
527	Example:	1198	Example:
…		…
533	add $grp	1204	add $grp
534	(aio_stat ...),	1205	(aio_stat ...),
535	(aio_stat ...),	1206	(aio_stat ...),
536	...;	1207	...;
537		1208
		1209	=item aio_nop $callback->()
		1210
		1211	This is a special request - it does nothing in itself and is only used for
		1212	side effects, such as when you want to add a dummy request to a group so
		1213	that finishing the requests in the group depends on executing the given
		1214	code.
		1215
		1216	While this request does nothing, it still goes through the execution
		1217	phase and still requires a worker thread. Thus, the callback will not
		1218	be executed immediately but only after other requests in the queue have
		1219	entered their execution phase. This can be used to measure request
		1220	latency.
		1221
538	=item IO::AIO::aio_sleep $fractional_seconds, $callback->() *NOT EXPORTED*	1222	=item IO::AIO::aio_busy $fractional_seconds, $callback->() *NOT EXPORTED*
539		1223
540	Mainly used for debugging and benchmarking, this aio request puts one of	1224	Mainly used for debugging and benchmarking, this aio request puts one of
541	the request workers to sleep for the given time.	1225	the request workers to sleep for the given time.
542		1226
543	While it is theoretically handy to have simple I/O scheduling requests	1227	While it is theoretically handy to have simple I/O scheduling requests
544	like sleep and file handle readable/writable, the overhead this creates	1228	like sleep and file handle readable/writable, the overhead this creates is
545	is immense, so do not use this function except to put your application	1229	immense (it blocks a thread for a long time) so do not use this function
546	under artificial I/O pressure.	1230	except to put your application under artificial I/O pressure.
547		1231
548	=back	1232	=back
549		1233
550	=head2 IO::AIO::REQ CLASS	1234	=head2 IO::AIO::REQ CLASS
551		1235
552	All non-aggregate C<aio_*> functions return an object of this class when	1236	All non-aggregate C<aio_*> functions return an object of this class when
553	called in non-void context.	1237	called in non-void context.
554		1238
555	A request always moves through the following five states in its lifetime,
556	in order: B<ready> (request has been created, but has not been executed
557	yet), B<execute> (request is currently being executed), B<pending>
558	(request has been executed but callback has not been called yet),
559	B<result> (results are being processed synchronously, includes calling the
560	callback) and B<done> (request has reached the end of its lifetime and
561	holds no resources anymore).
562
563	=over 4	1239	=over 4
564		1240
565	=item $req->cancel	1241	=item cancel $req
566		1242
567	Cancels the request, if possible. Has the effect of skipping execution	1243	Cancels the request, if possible. Has the effect of skipping execution
568	when entering the B<execute> state and skipping calling the callback when	1244	when entering the B<execute> state and skipping calling the callback when
569	entering the the B<result> state, but will leave the request otherwise	1245	entering the the B<result> state, but will leave the request otherwise
570	untouched. That means that requests that currently execute will not be	1246	untouched (with the exception of readdir). That means that requests that
571	stopped and resources held by the request will not be freed prematurely.	1247	currently execute will not be stopped and resources held by the request
		1248	will not be freed prematurely.
		1249
		1250	=item cb $req $callback->(...)
		1251
		1252	Replace (or simply set) the callback registered to the request.
572		1253
573	=back	1254	=back
574		1255
575	=head2 IO::AIO::GRP CLASS	1256	=head2 IO::AIO::GRP CLASS
576		1257
…		…
603	};	1284	};
604		1285
605	This makes it very easy to create composite requests (see the source of	1286	This makes it very easy to create composite requests (see the source of
606	C<aio_move> for an application) that work and feel like simple requests.	1287	C<aio_move> for an application) that work and feel like simple requests.
607		1288
		1289	=over 4
		1290
608	The IO::AIO::GRP objects will be cleaned up during calls to	1291	=item * The IO::AIO::GRP objects will be cleaned up during calls to
609	C<IO::AIO::poll_cb>, just like any other request.	1292	C<IO::AIO::poll_cb>, just like any other request.
610		1293
611	They can be canceled like any other request. Canceling will cancel not	1294	=item * They can be canceled like any other request. Canceling will cancel not
612	only the request itself, but also all requests it contains.	1295	only the request itself, but also all requests it contains.
613		1296
614	They can also can also be added to other IO::AIO::GRP objects.	1297	=item * They can also can also be added to other IO::AIO::GRP objects.
615		1298
616	You must not add requests to a group from within the group callback (or	1299	=item * You must not add requests to a group from within the group callback (or
617	any later time).	1300	any later time).
		1301
		1302	=back
618		1303
619	Their lifetime, simplified, looks like this: when they are empty, they	1304	Their lifetime, simplified, looks like this: when they are empty, they
620	will finish very quickly. If they contain only requests that are in the	1305	will finish very quickly. If they contain only requests that are in the
621	C<done> state, they will also finish. Otherwise they will continue to	1306	C<done> state, they will also finish. Otherwise they will continue to
622	exist.	1307	exist.
623		1308
624	That means after creating a group you have some time to add requests. And	1309	That means after creating a group you have some time to add requests
625	in the callbacks of those requests, you can add further requests to the	1310	(precisely before the callback has been invoked, which is only done within
626	group. And only when all those requests have finished will the the group	1311	the C<poll_cb>). And in the callbacks of those requests, you can add
627	itself finish.	1312	further requests to the group. And only when all those requests have
		1313	finished will the the group itself finish.
628		1314
629	=over 4	1315	=over 4
630		1316
		1317	=item add $grp ...
		1318
631	=item $grp->add (...)	1319	=item $grp->add (...)
632
633	=item add $grp ...
634		1320
635	Add one or more requests to the group. Any type of L<IO::AIO::REQ> can	1321	Add one or more requests to the group. Any type of L<IO::AIO::REQ> can
636	be added, including other groups, as long as you do not create circular	1322	be added, including other groups, as long as you do not create circular
637	dependencies.	1323	dependencies.
638		1324
639	Returns all its arguments.	1325	Returns all its arguments.
640		1326
		1327	=item $grp->cancel_subs
		1328
		1329	Cancel all subrequests and clears any feeder, but not the group request
		1330	itself. Useful when you queued a lot of events but got a result early.
		1331
		1332	The group request will finish normally (you cannot add requests to the
		1333	group).
		1334
641	=item $grp->result (...)	1335	=item $grp->result (...)
642		1336
643	Set the result value(s) that will be passed to the group callback when all	1337	Set the result value(s) that will be passed to the group callback when all
644	subrequests have finished. By default, no argument will be passed.	1338	subrequests have finished and set the groups errno to the current value
		1339	of errno (just like calling C<errno> without an error number). By default,
		1340	no argument will be passed and errno is zero.
645		1341
646	=item $grp->set_feeder ($callback->($grp))	1342	=item $grp->errno ([$errno])
647		1343
648	[VERY EXPERIMENTAL]	1344	Sets the group errno value to C<$errno>, or the current value of errno
		1345	when the argument is missing.
		1346
		1347	Every aio request has an associated errno value that is restored when
		1348	the callback is invoked. This method lets you change this value from its
		1349	default (0).
		1350
		1351	Calling C<result> will also set errno, so make sure you either set C<$!>
		1352	before the call to C<result>, or call c<errno> after it.
		1353
		1354	=item feed $grp $callback->($grp)
649		1355
650	Sets a feeder/generator on this group: every group can have an attached	1356	Sets a feeder/generator on this group: every group can have an attached
651	generator that generates requests if idle. The idea behind this is that,	1357	generator that generates requests if idle. The idea behind this is that,
652	although you could just queue as many requests as you want in a group,	1358	although you could just queue as many requests as you want in a group,
653	this might starve other requests for a potentially long time. For	1359	this might starve other requests for a potentially long time. For example,
654	example, C<aio_scandir> might generate hundreds of thousands C<aio_stat>	1360	C<aio_scandir> might generate hundreds of thousands C<aio_stat> requests,
655	requests, delaying any later requests for a long time.	1361	delaying any later requests for a long time.
656		1362
657	To avoid this, and allow incremental generation of requests, you can	1363	To avoid this, and allow incremental generation of requests, you can
658	instead a group and set a feeder on it that generates those requests. The	1364	instead a group and set a feeder on it that generates those requests. The
659	feeder will be called whenever there are few enough (see C<feeder_limit>,	1365	feed callback will be called whenever there are few enough (see C<limit>,
660	below) requests active in the group itself and is expected to queue more	1366	below) requests active in the group itself and is expected to queue more
661	requests.	1367	requests.
662		1368
663	The feeder can queue as many requests as it likes (i.e. C<add> does not	1369	The feed callback can queue as many requests as it likes (i.e. C<add> does
664	impose any limits).	1370	not impose any limits).
665		1371
666	If the feeder does not queue more requests when called, it will be	1372	If the feed does not queue more requests when called, it will be
667	automatically removed from the group.	1373	automatically removed from the group.
668		1374
669	If the feeder limit is C<0>, it will be set to C<2> automatically.	1375	If the feed limit is C<0> when this method is called, it will be set to
		1376	C<2> automatically.
670		1377
671	Example:	1378	Example:
672		1379
673	# stat all files in @files, but only ever use four aio requests concurrently:	1380	# stat all files in @files, but only ever use four aio requests concurrently:
674		1381
675	my $grp = aio_group sub { print "finished\n" };	1382	my $grp = aio_group sub { print "finished\n" };
676	$grp->feeder_limit (4);	1383	limit $grp 4;
677	$grp->set_feeder (sub {	1384	feed $grp sub {
678	my $file = pop @files	1385	my $file = pop @files
679	or return;	1386	or return;
680		1387
681	add $grp aio_stat $file, sub { ... };	1388	add $grp aio_stat $file, sub { ... };
682	});	1389	};
683		1390
684	=item $grp->feeder_limit ($num)	1391	=item limit $grp $num
685		1392
686	Sets the feeder limit for the group: The feeder will be called whenever	1393	Sets the feeder limit for the group: The feeder will be called whenever
687	the group contains less than this many requests.	1394	the group contains less than this many requests.
688		1395
689	Setting the limit to C<0> will pause the feeding process.	1396	Setting the limit to C<0> will pause the feeding process.
690		1397
		1398	The default value for the limit is C<0>, but note that setting a feeder
		1399	automatically bumps it up to C<2>.
		1400
691	=back	1401	=back
692		1402
693	=head2 SUPPORT FUNCTIONS	1403	=head2 SUPPORT FUNCTIONS
694		1404
		1405	=head3 EVENT PROCESSING AND EVENT LOOP INTEGRATION
		1406
695	=over 4	1407	=over 4
696		1408
697	=item $fileno = IO::AIO::poll_fileno	1409	=item $fileno = IO::AIO::poll_fileno
698		1410
699	Return the I<request result pipe file descriptor>. This filehandle must be	1411	Return the I<request result pipe file descriptor>. This filehandle must be
700	polled for reading by some mechanism outside this module (e.g. Event or	1412	polled for reading by some mechanism outside this module (e.g. EV, Glib,
701	select, see below or the SYNOPSIS). If the pipe becomes readable you have	1413	select and so on, see below or the SYNOPSIS). If the pipe becomes readable
702	to call C<poll_cb> to check the results.	1414	you have to call C<poll_cb> to check the results.
703		1415
704	See C<poll_cb> for an example.	1416	See C<poll_cb> for an example.
705		1417
706	=item IO::AIO::poll_cb	1418	=item IO::AIO::poll_cb
707		1419
708	Process all outstanding events on the result pipe. You have to call this	1420	Process some outstanding events on the result pipe. You have to call
709	regularly. Returns the number of events processed. Returns immediately	1421	this regularly. Returns C<0> if all events could be processed (or there
710	when no events are outstanding.	1422	were no events to process), or C<-1> if it returned earlier for whatever
		1423	reason. Returns immediately when no events are outstanding. The amount of
		1424	events processed depends on the settings of C<IO::AIO::max_poll_req> and
		1425	C<IO::AIO::max_poll_time>.
		1426
		1427	If not all requests were processed for whatever reason, the filehandle
		1428	will still be ready when C<poll_cb> returns, so normally you don't have to
		1429	do anything special to have it called later.
		1430
		1431	Apart from calling C<IO::AIO::poll_cb> when the event filehandle becomes
		1432	ready, it can be beneficial to call this function from loops which submit
		1433	a lot of requests, to make sure the results get processed when they become
		1434	available and not just when the loop is finished and the event loop takes
		1435	over again. This function returns very fast when there are no outstanding
		1436	requests.
711		1437
712	Example: Install an Event watcher that automatically calls	1438	Example: Install an Event watcher that automatically calls
713	IO::AIO::poll_cb with high priority:	1439	IO::AIO::poll_cb with high priority (more examples can be found in the
		1440	SYNOPSIS section, at the top of this document):
714		1441
715	Event->io (fd => IO::AIO::poll_fileno,	1442	Event->io (fd => IO::AIO::poll_fileno,
716	poll => 'r', async => 1,	1443	poll => 'r', async => 1,
717	cb => \&IO::AIO::poll_cb);	1444	cb => \&IO::AIO::poll_cb);
718		1445
719	=item IO::AIO::poll_wait	1446	=item IO::AIO::poll_wait
720		1447
		1448	If there are any outstanding requests and none of them in the result
721	Wait till the result filehandle becomes ready for reading (simply does a	1449	phase, wait till the result filehandle becomes ready for reading (simply
722	C<select> on the filehandle. This is useful if you want to synchronously wait	1450	does a C<select> on the filehandle. This is useful if you want to
723	for some requests to finish).	1451	synchronously wait for some requests to finish).
724		1452
725	See C<nreqs> for an example.	1453	See C<nreqs> for an example.
726		1454
		1455	=item IO::AIO::poll
		1456
		1457	Waits until some requests have been handled.
		1458
		1459	Returns the number of requests processed, but is otherwise strictly
		1460	equivalent to:
		1461
		1462	IO::AIO::poll_wait, IO::AIO::poll_cb
		1463
727	=item IO::AIO::nreqs	1464	=item IO::AIO::flush
728		1465
729	Returns the number of requests currently outstanding (i.e. for which their	1466	Wait till all outstanding AIO requests have been handled.
730	callback has not been invoked yet).
731		1467
732	Example: wait till there are no outstanding requests anymore:	1468	Strictly equivalent to:
733		1469
734	IO::AIO::poll_wait, IO::AIO::poll_cb	1470	IO::AIO::poll_wait, IO::AIO::poll_cb
735	while IO::AIO::nreqs;	1471	while IO::AIO::nreqs;
736		1472
737	=item IO::AIO::flush	1473	=item IO::AIO::max_poll_reqs $nreqs
738		1474
739	Wait till all outstanding AIO requests have been handled.	1475	=item IO::AIO::max_poll_time $seconds
740		1476
741	Strictly equivalent to:	1477	These set the maximum number of requests (default C<0>, meaning infinity)
		1478	that are being processed by C<IO::AIO::poll_cb> in one call, respectively
		1479	the maximum amount of time (default C<0>, meaning infinity) spent in
		1480	C<IO::AIO::poll_cb> to process requests (more correctly the mininum amount
		1481	of time C<poll_cb> is allowed to use).
742		1482
743	IO::AIO::poll_wait, IO::AIO::poll_cb	1483	Setting C<max_poll_time> to a non-zero value creates an overhead of one
744	while IO::AIO::nreqs;	1484	syscall per request processed, which is not normally a problem unless your
		1485	callbacks are really really fast or your OS is really really slow (I am
		1486	not mentioning Solaris here). Using C<max_poll_reqs> incurs no overhead.
745		1487
746	=item IO::AIO::poll	1488	Setting these is useful if you want to ensure some level of
		1489	interactiveness when perl is not fast enough to process all requests in
		1490	time.
747		1491
748	Waits until some requests have been handled.	1492	For interactive programs, values such as C<0.01> to C<0.1> should be fine.
749		1493
750	Strictly equivalent to:	1494	Example: Install an Event watcher that automatically calls
		1495	IO::AIO::poll_cb with low priority, to ensure that other parts of the
		1496	program get the CPU sometimes even under high AIO load.
751		1497
752	IO::AIO::poll_wait, IO::AIO::poll_cb	1498	# try not to spend much more than 0.1s in poll_cb
753	if IO::AIO::nreqs;	1499	IO::AIO::max_poll_time 0.1;
		1500
		1501	# use a low priority so other tasks have priority
		1502	Event->io (fd => IO::AIO::poll_fileno,
		1503	poll => 'r', nice => 1,
		1504	cb => &IO::AIO::poll_cb);
		1505
		1506	=back
		1507
		1508	=head3 CONTROLLING THE NUMBER OF THREADS
		1509
		1510	=over
754		1511
755	=item IO::AIO::min_parallel $nthreads	1512	=item IO::AIO::min_parallel $nthreads
756		1513
757	Set the minimum number of AIO threads to C<$nthreads>. The current	1514	Set the minimum number of AIO threads to C<$nthreads>. The current
758	default is C<8>, which means eight asynchronous operations can execute	1515	default is C<8>, which means eight asynchronous operations can execute
759	concurrently at any one time (the number of outstanding requests,	1516	concurrently at any one time (the number of outstanding requests,
760	however, is unlimited).	1517	however, is unlimited).
761		1518
762	IO::AIO starts threads only on demand, when an AIO request is queued and	1519	IO::AIO starts threads only on demand, when an AIO request is queued and
763	no free thread exists.	1520	no free thread exists. Please note that queueing up a hundred requests can
		1521	create demand for a hundred threads, even if it turns out that everything
		1522	is in the cache and could have been processed faster by a single thread.
764		1523
765	It is recommended to keep the number of threads relatively low, as some	1524	It is recommended to keep the number of threads relatively low, as some
766	Linux kernel versions will scale negatively with the number of threads	1525	Linux kernel versions will scale negatively with the number of threads
767	(higher parallelity => MUCH higher latency). With current Linux 2.6	1526	(higher parallelity => MUCH higher latency). With current Linux 2.6
768	versions, 4-32 threads should be fine.	1527	versions, 4-32 threads should be fine.
…		…
782	This module automatically runs C<max_parallel 0> at program end, to ensure	1541	This module automatically runs C<max_parallel 0> at program end, to ensure
783	that all threads are killed and that there are no outstanding requests.	1542	that all threads are killed and that there are no outstanding requests.
784		1543
785	Under normal circumstances you don't need to call this function.	1544	Under normal circumstances you don't need to call this function.
786		1545
		1546	=item IO::AIO::max_idle $nthreads
		1547
		1548	Limit the number of threads (default: 4) that are allowed to idle
		1549	(i.e., threads that did not get a request to process within the idle
		1550	timeout (default: 10 seconds). That means if a thread becomes idle while
		1551	C<$nthreads> other threads are also idle, it will free its resources and
		1552	exit.
		1553
		1554	This is useful when you allow a large number of threads (e.g. 100 or 1000)
		1555	to allow for extremely high load situations, but want to free resources
		1556	under normal circumstances (1000 threads can easily consume 30MB of RAM).
		1557
		1558	The default is probably ok in most situations, especially if thread
		1559	creation is fast. If thread creation is very slow on your system you might
		1560	want to use larger values.
		1561
		1562	=item IO::AIO::idle_timeout $seconds
		1563
		1564	Sets the minimum idle timeout (default 10) after which worker threads are
		1565	allowed to exit. SEe C<IO::AIO::max_idle>.
		1566
787	=item $oldnreqs = IO::AIO::max_outstanding $nreqs	1567	=item IO::AIO::max_outstanding $maxreqs
788		1568
789	Sets the maximum number of outstanding requests to C<$nreqs>. If you	1569	Sets the maximum number of outstanding requests to C<$nreqs>. If
790	try to queue up more than this number of requests, the caller will block until	1570	you do queue up more than this number of requests, the next call to
791	some requests have been handled.	1571	C<IO::AIO::poll_cb> (and other functions calling C<poll_cb>, such as
		1572	C<IO::AIO::flush> or C<IO::AIO::poll>) will block until the limit is no
		1573	longer exceeded.
792		1574
793	The default is very large, so normally there is no practical limit. If you	1575	In other words, this setting does not enforce a queue limit, but can be
794	queue up many requests in a loop it often improves speed if you set	1576	used to make poll functions block if the limit is exceeded.
795	this to a relatively low number, such as C<100>.
796		1577
797	Under normal circumstances you don't need to call this function.	1578	This is a very bad function to use in interactive programs because it
		1579	blocks, and a bad way to reduce concurrency because it is inexact: Better
		1580	use an C<aio_group> together with a feed callback.
		1581
		1582	It's main use is in scripts without an event loop - when you want to stat
		1583	a lot of files, you can write somehting like this:
		1584
		1585	IO::AIO::max_outstanding 32;
		1586
		1587	for my $path (...) {
		1588	aio_stat $path , ...;
		1589	IO::AIO::poll_cb;
		1590	}
		1591
		1592	IO::AIO::flush;
		1593
		1594	The call to C<poll_cb> inside the loop will normally return instantly, but
		1595	as soon as more thna C<32> reqeusts are in-flight, it will block until
		1596	some requests have been handled. This keeps the loop from pushing a large
		1597	number of C<aio_stat> requests onto the queue.
		1598
		1599	The default value for C<max_outstanding> is very large, so there is no
		1600	practical limit on the number of outstanding requests.
798		1601
799	=back	1602	=back
800		1603
		1604	=head3 STATISTICAL INFORMATION
		1605
		1606	=over
		1607
		1608	=item IO::AIO::nreqs
		1609
		1610	Returns the number of requests currently in the ready, execute or pending
		1611	states (i.e. for which their callback has not been invoked yet).
		1612
		1613	Example: wait till there are no outstanding requests anymore:
		1614
		1615	IO::AIO::poll_wait, IO::AIO::poll_cb
		1616	while IO::AIO::nreqs;
		1617
		1618	=item IO::AIO::nready
		1619
		1620	Returns the number of requests currently in the ready state (not yet
		1621	executed).
		1622
		1623	=item IO::AIO::npending
		1624
		1625	Returns the number of requests currently in the pending state (executed,
		1626	but not yet processed by poll_cb).
		1627
		1628	=back
		1629
		1630	=head3 MISCELLANEOUS FUNCTIONS
		1631
		1632	IO::AIO implements some functions that might be useful, but are not
		1633	asynchronous.
		1634
		1635	=over 4
		1636
		1637	=item IO::AIO::sendfile $ofh, $ifh, $offset, $count
		1638
		1639	Calls the C<eio_sendfile_sync> function, which is like C<aio_sendfile>,
		1640	but is blocking (this makes most sense if you know the input data is
		1641	likely cached already and the output filehandle is set to non-blocking
		1642	operations).
		1643
		1644	Returns the number of bytes copied, or C<-1> on error.
		1645
		1646	=item IO::AIO::fadvise $fh, $offset, $len, $advice
		1647
		1648	Simply calls the C<posix_fadvise> function (see its
		1649	manpage for details). The following advice constants are
		1650	avaiable: C<IO::AIO::FADV_NORMAL>, C<IO::AIO::FADV_SEQUENTIAL>,
		1651	C<IO::AIO::FADV_RANDOM>, C<IO::AIO::FADV_NOREUSE>,
		1652	C<IO::AIO::FADV_WILLNEED>, C<IO::AIO::FADV_DONTNEED>.
		1653
		1654	On systems that do not implement C<posix_fadvise>, this function returns
		1655	ENOSYS, otherwise the return value of C<posix_fadvise>.
		1656
		1657	=item IO::AIO::madvise $scalar, $offset, $len, $advice
		1658
		1659	Simply calls the C<posix_madvise> function (see its
		1660	manpage for details). The following advice constants are
		1661	avaiable: C<IO::AIO::MADV_NORMAL>, C<IO::AIO::MADV_SEQUENTIAL>,
		1662	C<IO::AIO::MADV_RANDOM>, C<IO::AIO::MADV_WILLNEED>, C<IO::AIO::MADV_DONTNEED>.
		1663
		1664	On systems that do not implement C<posix_madvise>, this function returns
		1665	ENOSYS, otherwise the return value of C<posix_madvise>.
		1666
		1667	=item IO::AIO::mprotect $scalar, $offset, $len, $protect
		1668
		1669	Simply calls the C<mprotect> function on the preferably AIO::mmap'ed
		1670	$scalar (see its manpage for details). The following protect
		1671	constants are avaiable: C<IO::AIO::PROT_NONE>, C<IO::AIO::PROT_READ>,
		1672	C<IO::AIO::PROT_WRITE>, C<IO::AIO::PROT_EXEC>.
		1673
		1674	On systems that do not implement C<mprotect>, this function returns
		1675	ENOSYS, otherwise the return value of C<mprotect>.
		1676
		1677	=item IO::AIO::mmap $scalar, $length, $prot, $flags, $fh[, $offset]
		1678
		1679	Memory-maps a file (or anonymous memory range) and attaches it to the
		1680	given C<$scalar>, which will act like a string scalar.
		1681
		1682	The only operations allowed on the scalar are C<substr>/C<vec> that don't
		1683	change the string length, and most read-only operations such as copying it
		1684	or searching it with regexes and so on.
		1685
		1686	Anything else is unsafe and will, at best, result in memory leaks.
		1687
		1688	The memory map associated with the C<$scalar> is automatically removed
		1689	when the C<$scalar> is destroyed, or when the C<IO::AIO::mmap> or
		1690	C<IO::AIO::munmap> functions are called.
		1691
		1692	This calls the C<mmap>(2) function internally. See your system's manual
		1693	page for details on the C<$length>, C<$prot> and C<$flags> parameters.
		1694
		1695	The C<$length> must be larger than zero and smaller than the actual
		1696	filesize.
		1697
		1698	C<$prot> is a combination of C<IO::AIO::PROT_NONE>, C<IO::AIO::PROT_EXEC>,
		1699	C<IO::AIO::PROT_READ> and/or C<IO::AIO::PROT_WRITE>,
		1700
		1701	C<$flags> can be a combination of C<IO::AIO::MAP_SHARED> or
		1702	C<IO::AIO::MAP_PRIVATE>, or a number of system-specific flags (when
		1703	not available, the are defined as 0): C<IO::AIO::MAP_ANONYMOUS>
		1704	(which is set to C<MAP_ANON> if your system only provides this
		1705	constant), C<IO::AIO::MAP_HUGETLB>, C<IO::AIO::MAP_LOCKED>,
		1706	C<IO::AIO::MAP_NORESERVE>, C<IO::AIO::MAP_POPULATE> or
		1707	C<IO::AIO::MAP_NONBLOCK>
		1708
		1709	If C<$fh> is C<undef>, then a file descriptor of C<-1> is passed.
		1710
		1711	C<$offset> is the offset from the start of the file - it generally must be
		1712	a multiple of C<IO::AIO::PAGESIZE> and defaults to C<0>.
		1713
		1714	Example:
		1715
		1716	use Digest::MD5;
		1717	use IO::AIO;
		1718
		1719	open my $fh, "<verybigfile"
		1720	or die "$!";
		1721
		1722	IO::AIO::mmap my $data, -s $fh, IO::AIO::PROT_READ, IO::AIO::MAP_SHARED, $fh
		1723	or die "verybigfile: $!";
		1724
		1725	my $fast_md5 = md5 $data;
		1726
		1727	=item IO::AIO::munmap $scalar
		1728
		1729	Removes a previous mmap and undefines the C<$scalar>.
		1730
		1731	=item IO::AIO::munlock $scalar, $offset = 0, $length = undef
		1732
		1733	Calls the C<munlock> function, undoing the effects of a previous
		1734	C<aio_mlock> call (see its description for details).
		1735
		1736	=item IO::AIO::munlockall
		1737
		1738	Calls the C<munlockall> function.
		1739
		1740	On systems that do not implement C<munlockall>, this function returns
		1741	ENOSYS, otherwise the return value of C<munlockall>.
		1742
		1743	=back
		1744
801	=cut	1745	=cut
802		1746
803	# support function to convert a fd into a perl filehandle
804	sub _fd2fh {
805	return undef if $_[0] < 0;
806
807	# try to generate nice filehandles
808	my $sym = "IO::AIO::fd#$_[0]";
809	local *$sym;
810
811	open *$sym, "+<&=$_[0]" # usually works under any unix
812	or open *$sym, "<&=$_[0]" # cygwin needs this
813	or open *$sym, ">&=$_[0]" # or this
814	or return undef;
815
816	*$sym
817	}
818
819	min_parallel 8;	1747	min_parallel 8;
820		1748
821	END {	1749	END { flush }
822	max_parallel 0;
823	}
824		1750
825	1;	1751	1;
		1752
		1753	=head1 EVENT LOOP INTEGRATION
		1754
		1755	It is recommended to use L<AnyEvent::AIO> to integrate IO::AIO
		1756	automatically into many event loops:
		1757
		1758	# AnyEvent integration (EV, Event, Glib, Tk, POE, urxvt, pureperl...)
		1759	use AnyEvent::AIO;
		1760
		1761	You can also integrate IO::AIO manually into many event loops, here are
		1762	some examples of how to do this:
		1763
		1764	# EV integration
		1765	my $aio_w = EV::io IO::AIO::poll_fileno, EV::READ, \&IO::AIO::poll_cb;
		1766
		1767	# Event integration
		1768	Event->io (fd => IO::AIO::poll_fileno,
		1769	poll => 'r',
		1770	cb => \&IO::AIO::poll_cb);
		1771
		1772	# Glib/Gtk2 integration
		1773	add_watch Glib::IO IO::AIO::poll_fileno,
		1774	in => sub { IO::AIO::poll_cb; 1 };
		1775
		1776	# Tk integration
		1777	Tk::Event::IO->fileevent (IO::AIO::poll_fileno, "",
		1778	readable => \&IO::AIO::poll_cb);
		1779
		1780	# Danga::Socket integration
		1781	Danga::Socket->AddOtherFds (IO::AIO::poll_fileno =>
		1782	\&IO::AIO::poll_cb);
826		1783
827	=head2 FORK BEHAVIOUR	1784	=head2 FORK BEHAVIOUR
828		1785
829	This module should do "the right thing" when the process using it forks:	1786	This module should do "the right thing" when the process using it forks:
830		1787
831	Before the fork, IO::AIO enters a quiescent state where no requests	1788	Before the fork, IO::AIO enters a quiescent state where no requests
832	can be added in other threads and no results will be processed. After	1789	can be added in other threads and no results will be processed. After
833	the fork the parent simply leaves the quiescent state and continues	1790	the fork the parent simply leaves the quiescent state and continues
834	request/result processing, while the child clears the request/result	1791	request/result processing, while the child frees the request/result queue
835	queue (so the requests started before the fork will only be handled in	1792	(so that the requests started before the fork will only be handled in the
836	the parent). Threads will be started on demand until the limit ste in the	1793	parent). Threads will be started on demand until the limit set in the
837	parent process has been reached again.	1794	parent process has been reached again.
838		1795
839	In short: the parent will, after a short pause, continue as if fork had	1796	In short: the parent will, after a short pause, continue as if fork had
840	not been called, while the child will act as if IO::AIO has not been used	1797	not been called, while the child will act as if IO::AIO has not been used
841	yet.	1798	yet.
842		1799
843	=head2 MEMORY USAGE	1800	=head2 MEMORY USAGE
844		1801
		1802	Per-request usage:
		1803
845	Each aio request uses - depending on your architecture - around 128 bytes	1804	Each aio request uses - depending on your architecture - around 100-200
846	of memory. In addition, stat requests need a stat buffer (possibly a few	1805	bytes of memory. In addition, stat requests need a stat buffer (possibly
847	hundred bytes). Perl scalars and other data passed into aio requests will	1806	a few hundred bytes), readdir requires a result buffer and so on. Perl
848	also be locked.	1807	scalars and other data passed into aio requests will also be locked and
		1808	will consume memory till the request has entered the done state.
849		1809
850	This is now awfully much, so queuing lots of requests is not usually a	1810	This is not awfully much, so queuing lots of requests is not usually a
851	problem.	1811	problem.
852		1812
853	Each thread needs a stack area which is usually around 16k, sometimes much	1813	Per-thread usage:
854	larger, depending on the OS.	1814
		1815	In the execution phase, some aio requests require more memory for
		1816	temporary buffers, and each thread requires a stack and other data
		1817	structures (usually around 16k-128k, depending on the OS).
		1818
		1819	=head1 KNOWN BUGS
		1820
		1821	Known bugs will be fixed in the next release.
855		1822
856	=head1 SEE ALSO	1823	=head1 SEE ALSO
857		1824
858	L<Coro>, L<Linux::AIO> (obsolete).	1825	L<AnyEvent::AIO> for easy integration into event loops, L<Coro::AIO> for a
		1826	more natural syntax.
859		1827
860	=head1 AUTHOR	1828	=head1 AUTHOR
861		1829
862	Marc Lehmann <schmorp@schmorp.de>	1830	Marc Lehmann <schmorp@schmorp.de>
863	http://home.schmorp.de/	1831	http://home.schmorp.de/

Diff Legend

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