ViewVC Help

View File | Revision Log | Show Annotations | Download File

/cvs/IO-AIO/AIO.pm

Comparing IO-AIO/AIO.pm (file contents):
Revision 1.70 by root, Tue Oct 24 03:40:38 2006 UTC vs.
Revision 1.199 by root, Wed Jun 29 12:46:36 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
…		…
25	$req->cancel; # cancel request if still in queue	26	$req->cancel; # cancel request if still in queue
26		27
27	my $grp = aio_group sub { print "all stats done\n" };	28	my $grp = aio_group sub { print "all stats done\n" };
28	add $grp aio_stat "..." for ...;	29	add $grp aio_stat "..." for ...;
29		30
30	# AnyEvent integration
31	open my $fh, "<&=" . IO::AIO::poll_fileno or die "$!";
32	my $w = AnyEvent->io (fh => $fh, poll => 'r', cb => sub { IO::AIO::poll_cb });
33
34	# Event integration
35	Event->io (fd => IO::AIO::poll_fileno,
36	poll => 'r',
37	cb => \&IO::AIO::poll_cb);
38
39	# Glib/Gtk2 integration
40	add_watch Glib::IO IO::AIO::poll_fileno,
41	in => sub { IO::AIO::poll_cb; 1 };
42
43	# Tk integration
44	Tk::Event::IO->fileevent (IO::AIO::poll_fileno, "",
45	readable => \&IO::AIO::poll_cb);
46
47	# Danga::Socket integration
48	Danga::Socket->AddOtherFds (IO::AIO::poll_fileno =>
49	\&IO::AIO::poll_cb);
50
51	=head1 DESCRIPTION	31	=head1 DESCRIPTION
52		32
53	This module implements asynchronous I/O using whatever means your	33	This module implements asynchronous I/O using whatever means your
54	operating system supports.	34	operating system supports. It is implemented as an interface to C<libeio>
		35	(L<http://software.schmorp.de/pkg/libeio.html>).
55		36
		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
56	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
57	and signal their completion. You don't need thread support in perl, and	54	requests and signal their completion. You don't need thread support
58	the threads created by this module will not be visible to perl. In the	55	in perl, and the threads created by this module will not be visible
59	future, this module might make use of the native aio functions available	56	to perl. In the future, this module might make use of the native aio
60	on many operating systems. However, they are often not well-supported	57	functions available on many operating systems. However, they are often
61	(Linux doesn't allow them on normal files currently, for example),	58	not well-supported or restricted (GNU/Linux doesn't allow them on normal
62	and they would only support aio_read and aio_write, so the remaining	59	files currently, for example), and they would only support aio_read and
63	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.
64		62
65	Although the module will work with in the presence of other threads,	63	Although the module will work in the presence of other (Perl-) threads,
66	it is currently not reentrant in any way, so use appropriate locking	64	it is currently not reentrant in any way, so use appropriate locking
67	yourself, always call C<poll_cb> from within the same thread, or never	65	yourself, always call C<poll_cb> from within the same thread, or never
68	call C<poll_cb> (or other C<aio_> functions) recursively.	66	call C<poll_cb> (or other C<aio_> functions) recursively.
69		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
		161
70	=cut	162	=cut
71		163
72	package IO::AIO;	164	package IO::AIO;
73		165
74	no warnings;	166	use Carp ();
75	use strict 'vars';	167
		168	use common::sense;
76		169
77	use base 'Exporter';	170	use base 'Exporter';
78		171
79	BEGIN {	172	BEGIN {
80	our $VERSION = '2.0';	173	our $VERSION = '3.92';
81		174
82	our @AIO_REQ = 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
83	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
84	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
85	aio_group aio_nop);	183	aio_statvfs);
		184
86	our @EXPORT = (@AIO_REQ, qw(aioreq_pri aioreq_nice));	185	our @EXPORT = (@AIO_REQ, qw(aioreq_pri aioreq_nice));
87	our @EXPORT_OK = qw(poll_fileno poll_cb poll_wait flush	186	our @EXPORT_OK = qw(poll_fileno poll_cb poll_wait flush
88	min_parallel max_parallel max_outstanding nreqs);	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
89		194
90	@IO::AIO::GRP::ISA = 'IO::AIO::REQ';	195	@IO::AIO::GRP::ISA = 'IO::AIO::REQ';
91		196
92	require XSLoader;	197	require XSLoader;
93	XSLoader::load ("IO::AIO", $VERSION);	198	XSLoader::load ("IO::AIO", $VERSION);
94	}	199	}
95		200
96	=head1 FUNCTIONS	201	=head1 FUNCTIONS
97		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, $length = undef
		274	IO::AIO::munlockall
		275
98	=head2 AIO FUNCTIONS	276	=head2 AIO REQUEST FUNCTIONS
99		277
100	All the C<aio_*> calls are more or less thin wrappers around the syscall	278	All the C<aio_*> calls are more or less thin wrappers around the syscall
101	with the same name (sans C<aio_>). The arguments are similar or identical,	279	with the same name (sans C<aio_>). The arguments are similar or identical,
102	and they all accept an additional (and optional) C<$callback> argument	280	and they all accept an additional (and optional) C<$callback> argument
103	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
104	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
105	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
106	syscall has been executed asynchronously.	284	syscall has been executed asynchronously.
107		285
108	All functions expecting a filehandle keep a copy of the filehandle	286	All functions expecting a filehandle keep a copy of the filehandle
109	internally until the request has finished.	287	internally until the request has finished.
110		288
111	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
112	manipulation of those requests while they are in-flight.	290	further manipulation of those requests while they are in-flight.
113		291
114	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
115	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
116	request is being executed, the current working directory could have	294	request is being executed, the current working directory could have
117	changed. Alternatively, you can make sure that you never change the	295	changed. Alternatively, you can make sure that you never change the
118	current working directory.	296	current working directory anywhere in the program and then use relative
		297	paths.
119		298
120	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
121	always pass in filenames you got from outside (command line, readdir	300	in filenames you got from outside (command line, readdir etc.) without
122	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
123	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
124	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)
125	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.
126		308
127	=over 4	309	=over 4
128		310
129	=item aioreq_pri $pri	311	=item $prev_pri = aioreq_pri [$pri]
130		312
131	Sets the priority for the next aio request. The default priority	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
132	is C<0>, the minimum and maximum priorities are C<-4> and C<4>,	316	The default priority is C<0>, the minimum and maximum priorities are C<-4>
133	respectively. Requests with higher priority will be serviced first.	317	and C<4>, respectively. Requests with higher priority will be serviced
		318	first.
134		319
135	The priority will be reset to C<0> after each call to one of the C<aio_>	320	The priority will be reset to C<0> after each call to one of the C<aio_*>
136	functions.	321	functions.
137		322
138	Example: open a file with low priority, then read something from it with	323	Example: open a file with low priority, then read something from it with
139	higher priority so the read request is serviced before other low priority	324	higher priority so the read request is serviced before other low priority
140	open requests (potentially spamming the cache):	325	open requests (potentially spamming the cache):
…		…
147	aio_read $_[0], ..., sub {	332	aio_read $_[0], ..., sub {
148	...	333	...
149	};	334	};
150	};	335	};
151		336
		337
152	=item aioreq_nice $pri_adjust	338	=item aioreq_nice $pri_adjust
153		339
154	Similar to C<aioreq_pri>, but subtracts the given value from the current	340	Similar to C<aioreq_pri>, but subtracts the given value from the current
155	priority, so effects are cumulative.	341	priority, so the effect is cumulative.
		342
156		343
157	=item aio_open $pathname, $flags, $mode, $callback->($fh)	344	=item aio_open $pathname, $flags, $mode, $callback->($fh)
158		345
159	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
160	created filehandle for the file.	347	created filehandle for the file.
…		…
166	list. They are the same as used by C<sysopen>.	353	list. They are the same as used by C<sysopen>.
167		354
168	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
169	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>,
170	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,
171	and C<0666> or C<0777> if you do).	358	and C<0666> or C<0777> if you do). Note that the C<$mode> will be modified
		359	by the umask in effect then the request is being executed, so better never
		360	change the umask.
172		361
173	Example:	362	Example:
174		363
175	aio_open "/etc/passwd", O_RDONLY, 0, sub {	364	aio_open "/etc/passwd", IO::AIO::O_RDONLY, 0, sub {
176	if ($_[0]) {	365	if ($_[0]) {
177	print "open successful, fh is $_[0]\n";	366	print "open successful, fh is $_[0]\n";
178	...	367	...
179	} else {	368	} else {
180	die "open failed: $!\n";	369	die "open failed: $!\n";
181	}	370	}
182	};	371	};
183		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
184	=item aio_close $fh, $callback->($status)	383	=item aio_close $fh, $callback->($status)
185		384
186	Asynchronously close a file and call the callback with the result	385	Asynchronously close a file and call the callback with the result
187	code. I<WARNING:> although accepted, you should not pass in a perl	386	code.
188	filehandle here, as perl will likely close the file descriptor another
189	time when the filehandle is destroyed. Normally, you can safely call perls
190	C<close> or just let filehandles go out of scope.
191		387
192	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
193	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
194		399
195	=item aio_read $fh,$offset,$length, $data,$dataoffset, $callback->($retval)	400	=item aio_read $fh,$offset,$length, $data,$dataoffset, $callback->($retval)
196		401
197	=item aio_write $fh,$offset,$length, $data,$dataoffset, $callback->($retval)	402	=item aio_write $fh,$offset,$length, $data,$dataoffset, $callback->($retval)
198		403
199	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
200	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<$dataoffset>
201	callback without the actual number of bytes read (or -1 on error, just	406	and calls the callback without the actual number of bytes read (or -1 on
202	like the syscall).	407	error, just like the syscall).
		408
		409	C<aio_read> will, like C<sysread>, shrink or grow the C<$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<$data>.
203		421
204	The C<$data> scalar I<MUST NOT> be modified in any way while the request	422	The C<$data> scalar I<MUST NOT> be modified in any way while the request
205	is outstanding. Modifying it can result in segfaults or WW3 (if the	423	is outstanding. Modifying it can result in segfaults or World War III (if
206	necessary/optional hardware is installed).	424	the necessary/optional hardware is installed).
207		425
208	Example: Read 15 bytes at offset 7 into scalar C<$buffer>, starting at	426	Example: Read 15 bytes at offset 7 into scalar C<$buffer>, starting at
209	offset C<0> within the scalar:	427	offset C<0> within the scalar:
210		428
211	aio_read $fh, 7, 15, $buffer, 0, sub {	429	aio_read $fh, 7, 15, $buffer, 0, sub {
212	$_[0] > 0 or die "read error: $!";	430	$_[0] > 0 or die "read error: $!";
213	print "read $_[0] bytes: <$buffer>\n";	431	print "read $_[0] bytes: <$buffer>\n";
214	};	432	};
215		433
216	=item aio_move $srcpath, $dstpath, $callback->($status)
217
218	[EXPERIMENTAL due to internal aio_group use]
219
220	Try to move the I<file> (directories not supported as either source or
221	destination) from C<$srcpath> to C<$dstpath> and call the callback with
222	the C<0> (error) or C<-1> ok.
223
224	This is a composite request that tries to rename(2) the file first. If
225	rename files with C<EXDEV>, it creates the destination file with mode 0200
226	and copies the contents of the source file into it using C<aio_sendfile>,
227	followed by restoring atime, mtime, access mode and uid/gid, in that
228	order, and unlinking the C<$srcpath>.
229
230	If an error occurs, the partial destination file will be unlinked, if
231	possible, except when setting atime, mtime, access mode and uid/gid, where
232	errors are being ignored.
233
234	=cut
235
236	sub aio_move($$$) {
237	my ($src, $dst, $cb) = @_;
238
239	my $grp = aio_group $cb;
240
241	add $grp aio_rename $src, $dst, sub {
242	if ($_[0] && $! == EXDEV) {
243	add $grp aio_open $src, O_RDONLY, 0, sub {
244	if (my $src_fh = $_[0]) {
245	my @stat = stat $src_fh;
246
247	add $grp aio_open $dst, O_WRONLY, 0200, sub {
248	if (my $dst_fh = $_[0]) {
249	add $grp aio_sendfile $dst_fh, $src_fh, 0, $stat[7], sub {
250	close $src_fh;
251
252	if ($_[0] == $stat[7]) {
253	utime $stat[8], $stat[9], $dst;
254	chmod $stat[2] & 07777, $dst_fh;
255	chown $stat[4], $stat[5], $dst_fh;
256	close $dst_fh;
257
258	add $grp aio_unlink $src, sub {
259	$grp->result ($_[0]);
260	};
261	} else {
262	my $errno = $!;
263	add $grp aio_unlink $dst, sub {
264	$! = $errno;
265	$grp->result (-1);
266	};
267	}
268	};
269	} else {
270	$grp->result (-1);
271	}
272	},
273
274	} else {
275	$grp->result (-1);
276	}
277	};
278	} else {
279	$grp->result ($_[0]);
280	}
281	};
282
283	$grp
284	}
285		434
286	=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)
287		436
288	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
289	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
290	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
291	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
292	other.	441	other. The same C<$in_fh> works fine though, as this function does not
		442	move or use the file offset of C<$in_fh>.
293		443
		444	Please note that C<aio_sendfile> can read more bytes from C<$in_fh> than
		445	are written, and there is no way to find out how many more bytes have been
		446	read from C<aio_sendfile> alone, as C<aio_sendfile> only provides the
		447	number of bytes written to C<$out_fh>. Only if the result value equals
		448	C<$length> one can assume that C<$length> bytes have been read.
		449
		450	Unlike with other C<aio_> functions, it makes a lot of sense to use
		451	C<aio_sendfile> on non-blocking sockets, as long as one end (typically
		452	the C<$in_fh>) is a file - the file I/O will then be asynchronous, while
		453	the socket I/O will be non-blocking. Note, however, that you can run
		454	into a trap where C<aio_sendfile> reads some data with readahead, then
		455	fails to write all data, and when the socket is ready the next time, the
		456	data in the cache is already lost, forcing C<aio_sendfile> to again hit
		457	the disk. Explicit C<aio_read> + C<aio_write> let's you better control
		458	resource usage.
		459
294	This call tries to make use of a native C<sendfile> syscall to provide	460	This call tries to make use of a native C<sendfile>-like syscall to
295	zero-copy operation. For this to work, C<$out_fh> should refer to a	461	provide zero-copy operation. For this to work, C<$out_fh> should refer to
296	socket, and C<$in_fh> should refer to mmap'able file.	462	a socket, and C<$in_fh> should refer to an mmap'able file.
297		463
298	If the native sendfile call fails or is not implemented, it will be	464	If a native sendfile cannot be found or it fails with C<ENOSYS>,
299	emulated, so you can call C<aio_sendfile> on any type of filehandle	465	C<EINVAL>, C<ENOTSUP>, C<EOPNOTSUPP>, C<EAFNOSUPPORT>, C<EPROTOTYPE> or
		466	C<ENOTSOCK>, it will be emulated, so you can call C<aio_sendfile> on any
300	regardless of the limitations of the operating system.	467	type of filehandle regardless of the limitations of the operating system.
301		468
302	Please note, however, that C<aio_sendfile> can read more bytes from	469	As native sendfile syscalls (as practically any non-POSIX interface hacked
303	C<$in_fh> than are written, and there is no way to find out how many	470	together in a hurry to improve benchmark numbers) tend to be rather buggy
304	bytes have been read from C<aio_sendfile> alone, as C<aio_sendfile> only	471	on many systems, this implementation tries to work around some known bugs
305	provides the number of bytes written to C<$out_fh>. Only if the result	472	in Linux and FreeBSD kernels (probably others, too), but that might fail,
306	value equals C<$length> one can assume that C<$length> bytes have been	473	so you really really should check the return value of C<aio_sendfile> -
307	read.	474	fewre bytes than expected might have been transferred.
		475
308		476
309	=item aio_readahead $fh,$offset,$length, $callback->($retval)	477	=item aio_readahead $fh,$offset,$length, $callback->($retval)
310		478
311	C<aio_readahead> populates the page cache with data from a file so that	479	C<aio_readahead> populates the page cache with data from a file so that
312	subsequent reads from that file will not block on disk I/O. The C<$offset>	480	subsequent reads from that file will not block on disk I/O. The C<$offset>
…		…
318	file. The current file offset of the file is left unchanged.	486	file. The current file offset of the file is left unchanged.
319		487
320	If that syscall doesn't exist (likely if your OS isn't Linux) it will be	488	If that syscall doesn't exist (likely if your OS isn't Linux) it will be
321	emulated by simply reading the data, which would have a similar effect.	489	emulated by simply reading the data, which would have a similar effect.
322		490
		491
323	=item aio_stat $fh_or_path, $callback->($status)	492	=item aio_stat $fh_or_path, $callback->($status)
324		493
325	=item aio_lstat $fh, $callback->($status)	494	=item aio_lstat $fh, $callback->($status)
326		495
327	Works like perl's C<stat> or C<lstat> in void context. The callback will	496	Works like perl's C<stat> or C<lstat> in void context. The callback will
…		…
332	for an explanation.	501	for an explanation.
333		502
334	Currently, the stats are always 64-bit-stats, i.e. instead of returning an	503	Currently, the stats are always 64-bit-stats, i.e. instead of returning an
335	error when stat'ing a large file, the results will be silently truncated	504	error when stat'ing a large file, the results will be silently truncated
336	unless perl itself is compiled with large file support.	505	unless perl itself is compiled with large file support.
		506
		507	To help interpret the mode and dev/rdev stat values, IO::AIO offers the
		508	following constants and functions (if not implemented, the constants will
		509	be C<0> and the functions will either C<croak> or fall back on traditional
		510	behaviour).
		511
		512	C<S_IFMT>, C<S_IFIFO>, C<S_IFCHR>, C<S_IFBLK>, C<S_IFLNK>, C<S_IFREG>,
		513	C<S_IFDIR>, C<S_IFWHT>, C<S_IFSOCK>, C<IO::AIO::major $dev_t>,
		514	C<IO::AIO::minor $dev_t>, C<IO::AIO::makedev $major, $minor>.
337		515
338	Example: Print the length of F</etc/passwd>:	516	Example: Print the length of F</etc/passwd>:
339		517
340	aio_stat "/etc/passwd", sub {	518	aio_stat "/etc/passwd", sub {
341	$_[0] and die "stat failed: $!";	519	$_[0] and die "stat failed: $!";
342	print "size is ", -s _, "\n";	520	print "size is ", -s _, "\n";
343	};	521	};
344		522
		523
		524	=item aio_statvfs $fh_or_path, $callback->($statvfs)
		525
		526	Works like the POSIX C<statvfs> or C<fstatvfs> syscalls, depending on
		527	whether a file handle or path was passed.
		528
		529	On success, the callback is passed a hash reference with the following
		530	members: C<bsize>, C<frsize>, C<blocks>, C<bfree>, C<bavail>, C<files>,
		531	C<ffree>, C<favail>, C<fsid>, C<flag> and C<namemax>. On failure, C<undef>
		532	is passed.
		533
		534	The following POSIX IO::AIO::ST_* constants are defined: C<ST_RDONLY> and
		535	C<ST_NOSUID>.
		536
		537	The following non-POSIX IO::AIO::ST_* flag masks are defined to
		538	their correct value when available, or to C<0> on systems that do
		539	not support them: C<ST_NODEV>, C<ST_NOEXEC>, C<ST_SYNCHRONOUS>,
		540	C<ST_MANDLOCK>, C<ST_WRITE>, C<ST_APPEND>, C<ST_IMMUTABLE>, C<ST_NOATIME>,
		541	C<ST_NODIRATIME> and C<ST_RELATIME>.
		542
		543	Example: stat C</wd> and dump out the data if successful.
		544
		545	aio_statvfs "/wd", sub {
		546	my $f = $_[0]
		547	or die "statvfs: $!";
		548
		549	use Data::Dumper;
		550	say Dumper $f;
		551	};
		552
		553	# result:
		554	{
		555	bsize => 1024,
		556	bfree => 4333064312,
		557	blocks => 10253828096,
		558	files => 2050765568,
		559	flag => 4096,
		560	favail => 2042092649,
		561	bavail => 4333064312,
		562	ffree => 2042092649,
		563	namemax => 255,
		564	frsize => 1024,
		565	fsid => 1810
		566	}
		567
		568
		569	=item aio_utime $fh_or_path, $atime, $mtime, $callback->($status)
		570
		571	Works like perl's C<utime> function (including the special case of $atime
		572	and $mtime being undef). Fractional times are supported if the underlying
		573	syscalls support them.
		574
		575	When called with a pathname, uses utimes(2) if available, otherwise
		576	utime(2). If called on a file descriptor, uses futimes(2) if available,
		577	otherwise returns ENOSYS, so this is not portable.
		578
		579	Examples:
		580
		581	# set atime and mtime to current time (basically touch(1)):
		582	aio_utime "path", undef, undef;
		583	# set atime to current time and mtime to beginning of the epoch:
		584	aio_utime "path", time, undef; # undef==0
		585
		586
		587	=item aio_chown $fh_or_path, $uid, $gid, $callback->($status)
		588
		589	Works like perl's C<chown> function, except that C<undef> for either $uid
		590	or $gid is being interpreted as "do not change" (but -1 can also be used).
		591
		592	Examples:
		593
		594	# same as "chown root path" in the shell:
		595	aio_chown "path", 0, -1;
		596	# same as above:
		597	aio_chown "path", 0, undef;
		598
		599
		600	=item aio_truncate $fh_or_path, $offset, $callback->($status)
		601
		602	Works like truncate(2) or ftruncate(2).
		603
		604
		605	=item aio_chmod $fh_or_path, $mode, $callback->($status)
		606
		607	Works like perl's C<chmod> function.
		608
		609
345	=item aio_unlink $pathname, $callback->($status)	610	=item aio_unlink $pathname, $callback->($status)
346		611
347	Asynchronously unlink (delete) a file and call the callback with the	612	Asynchronously unlink (delete) a file and call the callback with the
348	result code.	613	result code.
349		614
		615
		616	=item aio_mknod $path, $mode, $dev, $callback->($status)
		617
		618	[EXPERIMENTAL]
		619
		620	Asynchronously create a device node (or fifo). See mknod(2).
		621
		622	The only (POSIX-) portable way of calling this function is:
		623
		624	aio_mknod $path, IO::AIO::S_IFIFO | $mode, 0, sub { ...
		625
		626	See C<aio_stat> for info about some potentially helpful extra constants
		627	and functions.
		628
350	=item aio_link $srcpath, $dstpath, $callback->($status)	629	=item aio_link $srcpath, $dstpath, $callback->($status)
351		630
352	Asynchronously create a new link to the existing object at C<$srcpath> at	631	Asynchronously create a new link to the existing object at C<$srcpath> at
353	the path C<$dstpath> and call the callback with the result code.	632	the path C<$dstpath> and call the callback with the result code.
354		633
		634
355	=item aio_symlink $srcpath, $dstpath, $callback->($status)	635	=item aio_symlink $srcpath, $dstpath, $callback->($status)
356		636
357	Asynchronously create a new symbolic link to the existing object at C<$srcpath> at	637	Asynchronously create a new symbolic link to the existing object at C<$srcpath> at
358	the path C<$dstpath> and call the callback with the result code.	638	the path C<$dstpath> and call the callback with the result code.
359		639
		640
		641	=item aio_readlink $path, $callback->($link)
		642
		643	Asynchronously read the symlink specified by C<$path> and pass it to
		644	the callback. If an error occurs, nothing or undef gets passed to the
		645	callback.
		646
		647
360	=item aio_rename $srcpath, $dstpath, $callback->($status)	648	=item aio_rename $srcpath, $dstpath, $callback->($status)
361		649
362	Asynchronously rename the object at C<$srcpath> to C<$dstpath>, just as	650	Asynchronously rename the object at C<$srcpath> to C<$dstpath>, just as
363	rename(2) and call the callback with the result code.	651	rename(2) and call the callback with the result code.
364		652
		653
		654	=item aio_mkdir $pathname, $mode, $callback->($status)
		655
		656	Asynchronously mkdir (create) a directory and call the callback with
		657	the result code. C<$mode> will be modified by the umask at the time the
		658	request is executed, so do not change your umask.
		659
		660
365	=item aio_rmdir $pathname, $callback->($status)	661	=item aio_rmdir $pathname, $callback->($status)
366		662
367	Asynchronously rmdir (delete) a directory and call the callback with the	663	Asynchronously rmdir (delete) a directory and call the callback with the
368	result code.	664	result code.
		665
369		666
370	=item aio_readdir $pathname, $callback->($entries)	667	=item aio_readdir $pathname, $callback->($entries)
371		668
372	Unlike the POSIX call of the same name, C<aio_readdir> reads an entire	669	Unlike the POSIX call of the same name, C<aio_readdir> reads an entire
373	directory (i.e. opendir + readdir + closedir). The entries will not be	670	directory (i.e. opendir + readdir + closedir). The entries will not be
374	sorted, and will B<NOT> include the C<.> and C<..> entries.	671	sorted, and will B<NOT> include the C<.> and C<..> entries.
375		672
376	The callback a single argument which is either C<undef> or an array-ref	673	The callback is passed a single argument which is either C<undef> or an
377	with the filenames.	674	array-ref with the filenames.
		675
		676
		677	=item aio_readdirx $pathname, $flags, $callback->($entries, $flags)
		678
		679	Quite similar to C<aio_readdir>, but the C<$flags> argument allows to tune
		680	behaviour and output format. In case of an error, C<$entries> will be
		681	C<undef>.
		682
		683	The flags are a combination of the following constants, ORed together (the
		684	flags will also be passed to the callback, possibly modified):
		685
		686	=over 4
		687
		688	=item IO::AIO::READDIR_DENTS
		689
		690	When this flag is off, then the callback gets an arrayref consisting of
		691	names only (as with C<aio_readdir>), otherwise it gets an arrayref with
		692	C<[$name, $type, $inode]> arrayrefs, each describing a single directory
		693	entry in more detail.
		694
		695	C<$name> is the name of the entry.
		696
		697	C<$type> is one of the C<IO::AIO::DT_xxx> constants:
		698
		699	C<IO::AIO::DT_UNKNOWN>, C<IO::AIO::DT_FIFO>, C<IO::AIO::DT_CHR>, C<IO::AIO::DT_DIR>,
		700	C<IO::AIO::DT_BLK>, C<IO::AIO::DT_REG>, C<IO::AIO::DT_LNK>, C<IO::AIO::DT_SOCK>,
		701	C<IO::AIO::DT_WHT>.
		702
		703	C<IO::AIO::DT_UNKNOWN> means just that: readdir does not know. If you need to
		704	know, you have to run stat yourself. Also, for speed reasons, the C<$type>
		705	scalars are read-only: you can not modify them.
		706
		707	C<$inode> is the inode number (which might not be exact on systems with 64
		708	bit inode numbers and 32 bit perls). This field has unspecified content on
		709	systems that do not deliver the inode information.
		710
		711	=item IO::AIO::READDIR_DIRS_FIRST
		712
		713	When this flag is set, then the names will be returned in an order where
		714	likely directories come first, in optimal stat order. This is useful when
		715	you need to quickly find directories, or you want to find all directories
		716	while avoiding to stat() each entry.
		717
		718	If the system returns type information in readdir, then this is used
		719	to find directories directly. Otherwise, likely directories are names
		720	beginning with ".", or otherwise names with no dots, of which names with
		721	short names are tried first.
		722
		723	=item IO::AIO::READDIR_STAT_ORDER
		724
		725	When this flag is set, then the names will be returned in an order
		726	suitable for stat()'ing each one. That is, when you plan to stat()
		727	all files in the given directory, then the returned order will likely
		728	be fastest.
		729
		730	If both this flag and C<IO::AIO::READDIR_DIRS_FIRST> are specified, then
		731	the likely dirs come first, resulting in a less optimal stat order.
		732
		733	=item IO::AIO::READDIR_FOUND_UNKNOWN
		734
		735	This flag should not be set when calling C<aio_readdirx>. Instead, it
		736	is being set by C<aio_readdirx>, when any of the C<$type>'s found were
		737	C<IO::AIO::DT_UNKNOWN>. The absense of this flag therefore indicates that all
		738	C<$type>'s are known, which can be used to speed up some algorithms.
		739
		740	=back
		741
		742
		743	=item aio_load $path, $data, $callback->($status)
		744
		745	This is a composite request that tries to fully load the given file into
		746	memory. Status is the same as with aio_read.
		747
		748	=cut
		749
		750	sub aio_load($$;$) {
		751	my ($path, undef, $cb) = @_;
		752	my $data = \$_[1];
		753
		754	my $pri = aioreq_pri;
		755	my $grp = aio_group $cb;
		756
		757	aioreq_pri $pri;
		758	add $grp aio_open $path, O_RDONLY, 0, sub {
		759	my $fh = shift
		760	or return $grp->result (-1);
		761
		762	aioreq_pri $pri;
		763	add $grp aio_read $fh, 0, (-s $fh), $$data, 0, sub {
		764	$grp->result ($_[0]);
		765	};
		766	};
		767
		768	$grp
		769	}
		770
		771	=item aio_copy $srcpath, $dstpath, $callback->($status)
		772
		773	Try to copy the I<file> (directories not supported as either source or
		774	destination) from C<$srcpath> to C<$dstpath> and call the callback with
		775	a status of C<0> (ok) or C<-1> (error, see C<$!>).
		776
		777	This is a composite request that creates the destination file with
		778	mode 0200 and copies the contents of the source file into it using
		779	C<aio_sendfile>, followed by restoring atime, mtime, access mode and
		780	uid/gid, in that order.
		781
		782	If an error occurs, the partial destination file will be unlinked, if
		783	possible, except when setting atime, mtime, access mode and uid/gid, where
		784	errors are being ignored.
		785
		786	=cut
		787
		788	sub aio_copy($$;$) {
		789	my ($src, $dst, $cb) = @_;
		790
		791	my $pri = aioreq_pri;
		792	my $grp = aio_group $cb;
		793
		794	aioreq_pri $pri;
		795	add $grp aio_open $src, O_RDONLY, 0, sub {
		796	if (my $src_fh = $_[0]) {
		797	my @stat = stat $src_fh; # hmm, might block over nfs?
		798
		799	aioreq_pri $pri;
		800	add $grp aio_open $dst, O_CREAT | O_WRONLY | O_TRUNC, 0200, sub {
		801	if (my $dst_fh = $_[0]) {
		802	aioreq_pri $pri;
		803	add $grp aio_sendfile $dst_fh, $src_fh, 0, $stat[7], sub {
		804	if ($_[0] == $stat[7]) {
		805	$grp->result (0);
		806	close $src_fh;
		807
		808	my $ch = sub {
		809	aioreq_pri $pri;
		810	add $grp aio_chmod $dst_fh, $stat[2] & 07777, sub {
		811	aioreq_pri $pri;
		812	add $grp aio_chown $dst_fh, $stat[4], $stat[5], sub {
		813	aioreq_pri $pri;
		814	add $grp aio_close $dst_fh;
		815	}
		816	};
		817	};
		818
		819	aioreq_pri $pri;
		820	add $grp aio_utime $dst_fh, $stat[8], $stat[9], sub {
		821	if ($_[0] < 0 && $! == ENOSYS) {
		822	aioreq_pri $pri;
		823	add $grp aio_utime $dst, $stat[8], $stat[9], $ch;
		824	} else {
		825	$ch->();
		826	}
		827	};
		828	} else {
		829	$grp->result (-1);
		830	close $src_fh;
		831	close $dst_fh;
		832
		833	aioreq $pri;
		834	add $grp aio_unlink $dst;
		835	}
		836	};
		837	} else {
		838	$grp->result (-1);
		839	}
		840	},
		841
		842	} else {
		843	$grp->result (-1);
		844	}
		845	};
		846
		847	$grp
		848	}
		849
		850	=item aio_move $srcpath, $dstpath, $callback->($status)
		851
		852	Try to move the I<file> (directories not supported as either source or
		853	destination) from C<$srcpath> to C<$dstpath> and call the callback with
		854	a status of C<0> (ok) or C<-1> (error, see C<$!>).
		855
		856	This is a composite request that tries to rename(2) the file first; if
		857	rename fails with C<EXDEV>, it copies the file with C<aio_copy> and, if
		858	that is successful, unlinks the C<$srcpath>.
		859
		860	=cut
		861
		862	sub aio_move($$;$) {
		863	my ($src, $dst, $cb) = @_;
		864
		865	my $pri = aioreq_pri;
		866	my $grp = aio_group $cb;
		867
		868	aioreq_pri $pri;
		869	add $grp aio_rename $src, $dst, sub {
		870	if ($_[0] && $! == EXDEV) {
		871	aioreq_pri $pri;
		872	add $grp aio_copy $src, $dst, sub {
		873	$grp->result ($_[0]);
		874
		875	unless ($_[0]) {
		876	aioreq_pri $pri;
		877	add $grp aio_unlink $src;
		878	}
		879	};
		880	} else {
		881	$grp->result ($_[0]);
		882	}
		883	};
		884
		885	$grp
		886	}
378		887
379	=item aio_scandir $path, $maxreq, $callback->($dirs, $nondirs)	888	=item aio_scandir $path, $maxreq, $callback->($dirs, $nondirs)
380		889
381	[EXPERIMENTAL due to internal aio_group use]
382
383	Scans a directory (similar to C<aio_readdir>) but additionally tries to	890	Scans a directory (similar to C<aio_readdir>) but additionally tries to
384	separate the entries of directory C<$path> into two sets of names, ones	891	efficiently separate the entries of directory C<$path> into two sets of
385	you can recurse into (directories or links to them), and ones you cannot	892	names, directories you can recurse into (directories), and ones you cannot
386	recurse into (everything else).	893	recurse into (everything else, including symlinks to directories).
387		894
388	C<aio_scandir> is a composite request that creates of many sub requests_	895	C<aio_scandir> is a composite request that creates of many sub requests_
389	C<$maxreq> specifies the maximum number of outstanding aio requests that	896	C<$maxreq> specifies the maximum number of outstanding aio requests that
390	this function generates. If it is C<< <= 0 >>, then a suitable default	897	this function generates. If it is C<< <= 0 >>, then a suitable default
391	will be chosen (currently 6).	898	will be chosen (currently 4).
392		899
393	On error, the callback is called without arguments, otherwise it receives	900	On error, the callback is called without arguments, otherwise it receives
394	two array-refs with path-relative entry names.	901	two array-refs with path-relative entry names.
395		902
396	Example:	903	Example:
…		…
403		910
404	Implementation notes.	911	Implementation notes.
405		912
406	The C<aio_readdir> cannot be avoided, but C<stat()>'ing every entry can.	913	The C<aio_readdir> cannot be avoided, but C<stat()>'ing every entry can.
407		914
		915	If readdir returns file type information, then this is used directly to
		916	find directories.
		917
408	After reading the directory, the modification time, size etc. of the	918	Otherwise, after reading the directory, the modification time, size etc.
409	directory before and after the readdir is checked, and if they match (and	919	of the directory before and after the readdir is checked, and if they
410	isn't the current time), the link count will be used to decide how many	920	match (and isn't the current time), the link count will be used to decide
411	entries are directories (if >= 2). Otherwise, no knowledge of the number	921	how many entries are directories (if >= 2). Otherwise, no knowledge of the
412	of subdirectories will be assumed.	922	number of subdirectories will be assumed.
413		923
414	Then entries will be sorted into likely directories (everything without	924	Then entries will be sorted into likely directories a non-initial dot
415	a non-initial dot currently) and likely non-directories (everything	925	currently) and likely non-directories (see C<aio_readdirx>). Then every
416	else). Then every entry plus an appended C</.> will be C<stat>'ed,	926	entry plus an appended C</.> will be C<stat>'ed, likely directories first,
417	likely directories first. If that succeeds, it assumes that the entry	927	in order of their inode numbers. If that succeeds, it assumes that the
418	is a directory or a symlink to directory (which will be checked	928	entry is a directory or a symlink to directory (which will be checked
419	seperately). This is often faster than stat'ing the entry itself because	929	seperately). This is often faster than stat'ing the entry itself because
420	filesystems might detect the type of the entry without reading the inode	930	filesystems might detect the type of the entry without reading the inode
421	data (e.g. ext2fs filetype feature).	931	data (e.g. ext2fs filetype feature), even on systems that cannot return
		932	the filetype information on readdir.
422		933
423	If the known number of directories (link count - 2) has been reached, the	934	If the known number of directories (link count - 2) has been reached, the
424	rest of the entries is assumed to be non-directories.	935	rest of the entries is assumed to be non-directories.
425		936
426	This only works with certainty on POSIX (= UNIX) filesystems, which	937	This only works with certainty on POSIX (= UNIX) filesystems, which
…		…
430	as those tend to return 0 or 1 as link counts, which disables the	941	as those tend to return 0 or 1 as link counts, which disables the
431	directory counting heuristic.	942	directory counting heuristic.
432		943
433	=cut	944	=cut
434		945
435	sub aio_scandir($$$) {	946	sub aio_scandir($$;$) {
436	my ($path, $maxreq, $cb) = @_;	947	my ($path, $maxreq, $cb) = @_;
437		948
		949	my $pri = aioreq_pri;
		950
438	my $grp = aio_group $cb;	951	my $grp = aio_group $cb;
439		952
440	$maxreq = 6 if $maxreq <= 0;	953	$maxreq = 4 if $maxreq <= 0;
441		954
442	# stat once	955	# stat once
		956	aioreq_pri $pri;
443	add $grp aio_stat $path, sub {	957	add $grp aio_stat $path, sub {
444	return $grp->result () if $_[0];	958	return $grp->result () if $_[0];
445	my $now = time;	959	my $now = time;
446	my $hash1 = join ":", (stat _)[0,1,3,7,9];	960	my $hash1 = join ":", (stat _)[0,1,3,7,9];
447		961
448	# read the directory entries	962	# read the directory entries
		963	aioreq_pri $pri;
449	add $grp aio_readdir $path, sub {	964	add $grp aio_readdirx $path, READDIR_DIRS_FIRST, sub {
450	my $entries = shift	965	my $entries = shift
451	or return $grp->result ();	966	or return $grp->result ();
452		967
453	# stat the dir another time	968	# stat the dir another time
		969	aioreq_pri $pri;
454	add $grp aio_stat $path, sub {	970	add $grp aio_stat $path, sub {
455	my $hash2 = join ":", (stat _)[0,1,3,7,9];	971	my $hash2 = join ":", (stat _)[0,1,3,7,9];
456		972
457	my $ndirs;	973	my $ndirs;
458		974
459	# take the slow route if anything looks fishy	975	# take the slow route if anything looks fishy
460	if ($hash1 ne $hash2 or (stat _)[9] == $now) {	976	if ($hash1 ne $hash2 or (stat _)[9] == $now) {
461	$ndirs = -1;	977	$ndirs = -1;
462	} else {	978	} else {
463	# if nlink == 2, we are finished	979	# if nlink == 2, we are finished
464	# on non-posix-fs's, we rely on nlink < 2	980	# for non-posix-fs's, we rely on nlink < 2
465	$ndirs = (stat _)[3] - 2	981	$ndirs = (stat _)[3] - 2
466	or return $grp->result ([], $entries);	982	or return $grp->result ([], $entries);
467	}	983	}
468		984
469	# sort into likely dirs and likely nondirs
470	# dirs == files without ".", short entries first
471	$entries = [map $_->[0],
472	sort { $b->[1] cmp $a->[1] }
473	map [$_, sprintf "%s%04d", (/.\./ ? "1" : "0"), length],
474	@$entries];
475
476	my (@dirs, @nondirs);	985	my (@dirs, @nondirs);
477		986
478	my ($statcb, $schedcb);
479	my $nreq = 0;
480
481	my $statgrp = add $grp aio_group;	987	my $statgrp = add $grp aio_group sub {
		988	$grp->result (\@dirs, \@nondirs);
		989	};
482		990
483	$schedcb = sub {	991	limit $statgrp $maxreq;
484	if (@$entries) {	992	feed $statgrp sub {
485	if ($nreq < $maxreq) {	993	return unless @$entries;
486	my $ent = pop @$entries;	994	my $entry = shift @$entries;
		995
		996	aioreq_pri $pri;
		997	add $statgrp aio_stat "$path/$entry/.", sub {
		998	if ($_[0] < 0) {
		999	push @nondirs, $entry;
		1000	} else {
		1001	# need to check for real directory
		1002	aioreq_pri $pri;
		1003	add $statgrp aio_lstat "$path/$entry", sub {
		1004	if (-d _) {
		1005	push @dirs, $entry;
		1006
		1007	unless (--$ndirs) {
		1008	push @nondirs, @$entries;
		1009	feed $statgrp;
		1010	}
		1011	} else {
		1012	push @nondirs, $entry;
		1013	}
487	$nreq++;	1014	}
488	add $statgrp aio_stat "$path/$ent/.", sub { $statcb->($_[0], $ent) };
489	}	1015	}
490	} elsif (!$nreq) {
491	# finished
492	$statgrp->cancel;
493	undef $statcb;
494	undef $schedcb;
495	$grp->result (\@dirs, \@nondirs);
496	}	1016	};
497	};	1017	};
498	$statcb = sub {
499	my ($status, $entry) = @_;
500
501	if ($status < 0) {
502	$nreq--;
503	push @nondirs, $entry;
504	&$schedcb;
505	} else {
506	# need to check for real directory
507	add $grp aio_lstat "$path/$entry", sub {
508	$nreq--;
509
510	if (-d _) {
511	push @dirs, $entry;
512
513	if (!--$ndirs) {
514	push @nondirs, @$entries;
515	$entries = [];
516	}
517	} else {
518	push @nondirs, $entry;
519	}
520
521	&$schedcb;
522	}
523	}
524	};
525
526	&$schedcb while @$entries && $nreq < $maxreq;
527	};	1018	};
528	};	1019	};
529	};	1020	};
530		1021
531	$grp	1022	$grp
532	}	1023	}
533		1024
		1025	=item aio_rmtree $path, $callback->($status)
		1026
		1027	Delete a directory tree starting (and including) C<$path>, return the
		1028	status of the final C<rmdir> only. This is a composite request that
		1029	uses C<aio_scandir> to recurse into and rmdir directories, and unlink
		1030	everything else.
		1031
		1032	=cut
		1033
		1034	sub aio_rmtree;
		1035	sub aio_rmtree($;$) {
		1036	my ($path, $cb) = @_;
		1037
		1038	my $pri = aioreq_pri;
		1039	my $grp = aio_group $cb;
		1040
		1041	aioreq_pri $pri;
		1042	add $grp aio_scandir $path, 0, sub {
		1043	my ($dirs, $nondirs) = @_;
		1044
		1045	my $dirgrp = aio_group sub {
		1046	add $grp aio_rmdir $path, sub {
		1047	$grp->result ($_[0]);
		1048	};
		1049	};
		1050
		1051	(aioreq_pri $pri), add $dirgrp aio_rmtree "$path/$_" for @$dirs;
		1052	(aioreq_pri $pri), add $dirgrp aio_unlink "$path/$_" for @$nondirs;
		1053
		1054	add $grp $dirgrp;
		1055	};
		1056
		1057	$grp
		1058	}
		1059
		1060	=item aio_sync $callback->($status)
		1061
		1062	Asynchronously call sync and call the callback when finished.
		1063
534	=item aio_fsync $fh, $callback->($status)	1064	=item aio_fsync $fh, $callback->($status)
535		1065
536	Asynchronously call fsync on the given filehandle and call the callback	1066	Asynchronously call fsync on the given filehandle and call the callback
537	with the fsync result code.	1067	with the fsync result code.
538		1068
…		…
542	callback with the fdatasync result code.	1072	callback with the fdatasync result code.
543		1073
544	If this call isn't available because your OS lacks it or it couldn't be	1074	If this call isn't available because your OS lacks it or it couldn't be
545	detected, it will be emulated by calling C<fsync> instead.	1075	detected, it will be emulated by calling C<fsync> instead.
546		1076
		1077	=item aio_sync_file_range $fh, $offset, $nbytes, $flags, $callback->($status)
		1078
		1079	Sync the data portion of the file specified by C<$offset> and C<$length>
		1080	to disk (but NOT the metadata), by calling the Linux-specific
		1081	sync_file_range call. If sync_file_range is not available or it returns
		1082	ENOSYS, then fdatasync or fsync is being substituted.
		1083
		1084	C<$flags> can be a combination of C<IO::AIO::SYNC_FILE_RANGE_WAIT_BEFORE>,
		1085	C<IO::AIO::SYNC_FILE_RANGE_WRITE> and
		1086	C<IO::AIO::SYNC_FILE_RANGE_WAIT_AFTER>: refer to the sync_file_range
		1087	manpage for details.
		1088
		1089	=item aio_pathsync $path, $callback->($status)
		1090
		1091	This request tries to open, fsync and close the given path. This is a
		1092	composite request intended to sync directories after directory operations
		1093	(E.g. rename). This might not work on all operating systems or have any
		1094	specific effect, but usually it makes sure that directory changes get
		1095	written to disc. It works for anything that can be opened for read-only,
		1096	not just directories.
		1097
		1098	Future versions of this function might fall back to other methods when
		1099	C<fsync> on the directory fails (such as calling C<sync>).
		1100
		1101	Passes C<0> when everything went ok, and C<-1> on error.
		1102
		1103	=cut
		1104
		1105	sub aio_pathsync($;$) {
		1106	my ($path, $cb) = @_;
		1107
		1108	my $pri = aioreq_pri;
		1109	my $grp = aio_group $cb;
		1110
		1111	aioreq_pri $pri;
		1112	add $grp aio_open $path, O_RDONLY, 0, sub {
		1113	my ($fh) = @_;
		1114	if ($fh) {
		1115	aioreq_pri $pri;
		1116	add $grp aio_fsync $fh, sub {
		1117	$grp->result ($_[0]);
		1118
		1119	aioreq_pri $pri;
		1120	add $grp aio_close $fh;
		1121	};
		1122	} else {
		1123	$grp->result (-1);
		1124	}
		1125	};
		1126
		1127	$grp
		1128	}
		1129
		1130	=item aio_msync $scalar, $offset = 0, $length = undef, flags = 0, $callback->($status)
		1131
		1132	This is a rather advanced IO::AIO call, which only works on mmap(2)ed
		1133	scalars (see the C<IO::AIO::mmap> function, although it also works on data
		1134	scalars managed by the L<Sys::Mmap> or L<Mmap> modules, note that the
		1135	scalar must only be modified in-place while an aio operation is pending on
		1136	it).
		1137
		1138	It calls the C<msync> function of your OS, if available, with the memory
		1139	area starting at C<$offset> in the string and ending C<$length> bytes
		1140	later. If C<$length> is negative, counts from the end, and if C<$length>
		1141	is C<undef>, then it goes till the end of the string. The flags can be
		1142	a combination of C<IO::AIO::MS_ASYNC>, C<IO::AIO::MS_INVALIDATE> and
		1143	C<IO::AIO::MS_SYNC>.
		1144
		1145	=item aio_mtouch $scalar, $offset = 0, $length = undef, flags = 0, $callback->($status)
		1146
		1147	This is a rather advanced IO::AIO call, which works best on mmap(2)ed
		1148	scalars.
		1149
		1150	It touches (reads or writes) all memory pages in the specified
		1151	range inside the scalar. All caveats and parameters are the same
		1152	as for C<aio_msync>, above, except for flags, which must be either
		1153	C<0> (which reads all pages and ensures they are instantiated) or
		1154	C<IO::AIO::MT_MODIFY>, which modifies the memory page s(by reading and
		1155	writing an octet from it, which dirties the page).
		1156
		1157	=item aio_mlock $scalar, $offset = 0, $length = undef, $callback->($status)
		1158
		1159	This is a rather advanced IO::AIO call, which works best on mmap(2)ed
		1160	scalars.
		1161
		1162	It reads in all the pages of the underlying storage into memory (if any)
		1163	and locks them, so they are not getting swapped/paged out or removed.
		1164
		1165	If C<$length> is undefined, then the scalar will be locked till the end.
		1166
		1167	On systems that do not implement C<mlock>, this function returns C<-1>
		1168	and sets errno to C<ENOSYS>.
		1169
		1170	Note that the corresponding C<munlock> is synchronous and is
		1171	documented under L<MISCELLANEOUS FUNCTIONS>.
		1172
		1173	Example: open a file, mmap and mlock it - both will be undone when
		1174	C<$data> gets destroyed.
		1175
		1176	open my $fh, "<", $path or die "$path: $!";
		1177	my $data;
		1178	IO::AIO::mmap $data, -s $fh, IO::AIO::PROT_READ, IO::AIO::MAP_SHARED, $fh;
		1179	aio_mlock $data; # mlock in background
		1180
		1181	=item aio_mlockall $flags, $callback->($status)
		1182
		1183	Calls the C<mlockall> function with the given C<$flags> (a combination of
		1184	C<IO::AIO::MCL_CURRENT> and C<IO::AIO::MCL_FUTURE>).
		1185
		1186	On systems that do not implement C<mlockall>, this function returns C<-1>
		1187	and sets errno to C<ENOSYS>.
		1188
		1189	Note that the corresponding C<munlockall> is synchronous and is
		1190	documented under L<MISCELLANEOUS FUNCTIONS>.
		1191
		1192	Example: asynchronously lock all current and future pages into memory.
		1193
		1194	aio_mlockall IO::AIO::MCL_FUTURE;
		1195
547	=item aio_group $callback->(...)	1196	=item aio_group $callback->(...)
548
549	[EXPERIMENTAL]
550		1197
551	This is a very special aio request: Instead of doing something, it is a	1198	This is a very special aio request: Instead of doing something, it is a
552	container for other aio requests, which is useful if you want to bundle	1199	container for other aio requests, which is useful if you want to bundle
553	many requests into a single, composite, request.	1200	many requests into a single, composite, request with a definite callback
		1201	and the ability to cancel the whole request with its subrequests.
554		1202
555	Returns an object of class L<IO::AIO::GRP>. See its documentation below	1203	Returns an object of class L<IO::AIO::GRP>. See its documentation below
556	for more info.	1204	for more info.
557		1205
558	Example:	1206	Example:
…		…
577	phase and still requires a worker thread. Thus, the callback will not	1225	phase and still requires a worker thread. Thus, the callback will not
578	be executed immediately but only after other requests in the queue have	1226	be executed immediately but only after other requests in the queue have
579	entered their execution phase. This can be used to measure request	1227	entered their execution phase. This can be used to measure request
580	latency.	1228	latency.
581		1229
582	=item IO::AIO::aio_sleep $fractional_seconds, $callback->() *NOT EXPORTED*	1230	=item IO::AIO::aio_busy $fractional_seconds, $callback->() *NOT EXPORTED*
583		1231
584	Mainly used for debugging and benchmarking, this aio request puts one of	1232	Mainly used for debugging and benchmarking, this aio request puts one of
585	the request workers to sleep for the given time.	1233	the request workers to sleep for the given time.
586		1234
587	While it is theoretically handy to have simple I/O scheduling requests	1235	While it is theoretically handy to have simple I/O scheduling requests
588	like sleep and file handle readable/writable, the overhead this creates	1236	like sleep and file handle readable/writable, the overhead this creates is
589	is immense, so do not use this function except to put your application	1237	immense (it blocks a thread for a long time) so do not use this function
590	under artificial I/O pressure.	1238	except to put your application under artificial I/O pressure.
591		1239
592	=back	1240	=back
593		1241
594	=head2 IO::AIO::REQ CLASS	1242	=head2 IO::AIO::REQ CLASS
595		1243
596	All non-aggregate C<aio_*> functions return an object of this class when	1244	All non-aggregate C<aio_*> functions return an object of this class when
597	called in non-void context.	1245	called in non-void context.
598
599	A request always moves through the following five states in its lifetime,
600	in order: B<ready> (request has been created, but has not been executed
601	yet), B<execute> (request is currently being executed), B<pending>
602	(request has been executed but callback has not been called yet),
603	B<result> (results are being processed synchronously, includes calling the
604	callback) and B<done> (request has reached the end of its lifetime and
605	holds no resources anymore).
606		1246
607	=over 4	1247	=over 4
608		1248
609	=item cancel $req	1249	=item cancel $req
610		1250
611	Cancels the request, if possible. Has the effect of skipping execution	1251	Cancels the request, if possible. Has the effect of skipping execution
612	when entering the B<execute> state and skipping calling the callback when	1252	when entering the B<execute> state and skipping calling the callback when
613	entering the the B<result> state, but will leave the request otherwise	1253	entering the the B<result> state, but will leave the request otherwise
614	untouched. That means that requests that currently execute will not be	1254	untouched (with the exception of readdir). That means that requests that
615	stopped and resources held by the request will not be freed prematurely.	1255	currently execute will not be stopped and resources held by the request
		1256	will not be freed prematurely.
616		1257
617	=item cb $req $callback->(...)	1258	=item cb $req $callback->(...)
618		1259
619	Replace (or simply set) the callback registered to the request.	1260	Replace (or simply set) the callback registered to the request.
620		1261
…		…
664	=item * They can also can also be added to other IO::AIO::GRP objects.	1305	=item * They can also can also be added to other IO::AIO::GRP objects.
665		1306
666	=item * You must not add requests to a group from within the group callback (or	1307	=item * You must not add requests to a group from within the group callback (or
667	any later time).	1308	any later time).
668		1309
669	=item * This does not harmonise well with C<max_outstanding>, so best do
670	not combine C<aio_group> with it. Groups and feeders are recommended for
671	this kind of concurrency-limiting.
672
673	=back	1310	=back
674		1311
675	Their lifetime, simplified, looks like this: when they are empty, they	1312	Their lifetime, simplified, looks like this: when they are empty, they
676	will finish very quickly. If they contain only requests that are in the	1313	will finish very quickly. If they contain only requests that are in the
677	C<done> state, they will also finish. Otherwise they will continue to	1314	C<done> state, they will also finish. Otherwise they will continue to
678	exist.	1315	exist.
679		1316
680	That means after creating a group you have some time to add requests. And	1317	That means after creating a group you have some time to add requests
681	in the callbacks of those requests, you can add further requests to the	1318	(precisely before the callback has been invoked, which is only done within
682	group. And only when all those requests have finished will the the group	1319	the C<poll_cb>). And in the callbacks of those requests, you can add
683	itself finish.	1320	further requests to the group. And only when all those requests have
		1321	finished will the the group itself finish.
684		1322
685	=over 4	1323	=over 4
686		1324
687	=item add $grp ...	1325	=item add $grp ...
688		1326
…		…
692	be added, including other groups, as long as you do not create circular	1330	be added, including other groups, as long as you do not create circular
693	dependencies.	1331	dependencies.
694		1332
695	Returns all its arguments.	1333	Returns all its arguments.
696		1334
		1335	=item $grp->cancel_subs
		1336
		1337	Cancel all subrequests and clears any feeder, but not the group request
		1338	itself. Useful when you queued a lot of events but got a result early.
		1339
		1340	The group request will finish normally (you cannot add requests to the
		1341	group).
		1342
697	=item $grp->result (...)	1343	=item $grp->result (...)
698		1344
699	Set the result value(s) that will be passed to the group callback when all	1345	Set the result value(s) that will be passed to the group callback when all
700	subrequests have finished. By default, no argument will be passed.	1346	subrequests have finished and set the groups errno to the current value
		1347	of errno (just like calling C<errno> without an error number). By default,
		1348	no argument will be passed and errno is zero.
		1349
		1350	=item $grp->errno ([$errno])
		1351
		1352	Sets the group errno value to C<$errno>, or the current value of errno
		1353	when the argument is missing.
		1354
		1355	Every aio request has an associated errno value that is restored when
		1356	the callback is invoked. This method lets you change this value from its
		1357	default (0).
		1358
		1359	Calling C<result> will also set errno, so make sure you either set C<$!>
		1360	before the call to C<result>, or call c<errno> after it.
701		1361
702	=item feed $grp $callback->($grp)	1362	=item feed $grp $callback->($grp)
703
704	[VERY EXPERIMENTAL]
705		1363
706	Sets a feeder/generator on this group: every group can have an attached	1364	Sets a feeder/generator on this group: every group can have an attached
707	generator that generates requests if idle. The idea behind this is that,	1365	generator that generates requests if idle. The idea behind this is that,
708	although you could just queue as many requests as you want in a group,	1366	although you could just queue as many requests as you want in a group,
709	this might starve other requests for a potentially long time. For	1367	this might starve other requests for a potentially long time. For example,
710	example, C<aio_scandir> might generate hundreds of thousands C<aio_stat>	1368	C<aio_scandir> might generate hundreds of thousands C<aio_stat> requests,
711	requests, delaying any later requests for a long time.	1369	delaying any later requests for a long time.
712		1370
713	To avoid this, and allow incremental generation of requests, you can	1371	To avoid this, and allow incremental generation of requests, you can
714	instead a group and set a feeder on it that generates those requests. The	1372	instead a group and set a feeder on it that generates those requests. The
715	feed callback will be called whenever there are few enough (see C<limit>,	1373	feed callback will be called whenever there are few enough (see C<limit>,
716	below) requests active in the group itself and is expected to queue more	1374	below) requests active in the group itself and is expected to queue more
…		…
720	not impose any limits).	1378	not impose any limits).
721		1379
722	If the feed does not queue more requests when called, it will be	1380	If the feed does not queue more requests when called, it will be
723	automatically removed from the group.	1381	automatically removed from the group.
724		1382
725	If the feed limit is C<0>, it will be set to C<2> automatically.	1383	If the feed limit is C<0> when this method is called, it will be set to
		1384	C<2> automatically.
726		1385
727	Example:	1386	Example:
728		1387
729	# stat all files in @files, but only ever use four aio requests concurrently:	1388	# stat all files in @files, but only ever use four aio requests concurrently:
730		1389
…		…
742	Sets the feeder limit for the group: The feeder will be called whenever	1401	Sets the feeder limit for the group: The feeder will be called whenever
743	the group contains less than this many requests.	1402	the group contains less than this many requests.
744		1403
745	Setting the limit to C<0> will pause the feeding process.	1404	Setting the limit to C<0> will pause the feeding process.
746		1405
		1406	The default value for the limit is C<0>, but note that setting a feeder
		1407	automatically bumps it up to C<2>.
		1408
747	=back	1409	=back
748		1410
749	=head2 SUPPORT FUNCTIONS	1411	=head2 SUPPORT FUNCTIONS
750		1412
		1413	=head3 EVENT PROCESSING AND EVENT LOOP INTEGRATION
		1414
751	=over 4	1415	=over 4
752		1416
753	=item $fileno = IO::AIO::poll_fileno	1417	=item $fileno = IO::AIO::poll_fileno
754		1418
755	Return the I<request result pipe file descriptor>. This filehandle must be	1419	Return the I<request result pipe file descriptor>. This filehandle must be
756	polled for reading by some mechanism outside this module (e.g. Event or	1420	polled for reading by some mechanism outside this module (e.g. EV, Glib,
757	select, see below or the SYNOPSIS). If the pipe becomes readable you have	1421	select and so on, see below or the SYNOPSIS). If the pipe becomes readable
758	to call C<poll_cb> to check the results.	1422	you have to call C<poll_cb> to check the results.
759		1423
760	See C<poll_cb> for an example.	1424	See C<poll_cb> for an example.
761		1425
762	=item IO::AIO::poll_cb	1426	=item IO::AIO::poll_cb
763		1427
764	Process all outstanding events on the result pipe. You have to call this	1428	Process some outstanding events on the result pipe. You have to call
765	regularly. Returns the number of events processed. Returns immediately	1429	this regularly. Returns C<0> if all events could be processed (or there
766	when no events are outstanding.	1430	were no events to process), or C<-1> if it returned earlier for whatever
		1431	reason. Returns immediately when no events are outstanding. The amount of
		1432	events processed depends on the settings of C<IO::AIO::max_poll_req> and
		1433	C<IO::AIO::max_poll_time>.
		1434
		1435	If not all requests were processed for whatever reason, the filehandle
		1436	will still be ready when C<poll_cb> returns, so normally you don't have to
		1437	do anything special to have it called later.
		1438
		1439	Apart from calling C<IO::AIO::poll_cb> when the event filehandle becomes
		1440	ready, it can be beneficial to call this function from loops which submit
		1441	a lot of requests, to make sure the results get processed when they become
		1442	available and not just when the loop is finished and the event loop takes
		1443	over again. This function returns very fast when there are no outstanding
		1444	requests.
767		1445
768	Example: Install an Event watcher that automatically calls	1446	Example: Install an Event watcher that automatically calls
769	IO::AIO::poll_cb with high priority:	1447	IO::AIO::poll_cb with high priority (more examples can be found in the
		1448	SYNOPSIS section, at the top of this document):
770		1449
771	Event->io (fd => IO::AIO::poll_fileno,	1450	Event->io (fd => IO::AIO::poll_fileno,
772	poll => 'r', async => 1,	1451	poll => 'r', async => 1,
773	cb => \&IO::AIO::poll_cb);	1452	cb => \&IO::AIO::poll_cb);
774		1453
775	=item IO::AIO::poll_wait	1454	=item IO::AIO::poll_wait
776		1455
		1456	If there are any outstanding requests and none of them in the result
777	Wait till the result filehandle becomes ready for reading (simply does a	1457	phase, wait till the result filehandle becomes ready for reading (simply
778	C<select> on the filehandle. This is useful if you want to synchronously wait	1458	does a C<select> on the filehandle. This is useful if you want to
779	for some requests to finish).	1459	synchronously wait for some requests to finish).
780		1460
781	See C<nreqs> for an example.	1461	See C<nreqs> for an example.
782		1462
		1463	=item IO::AIO::poll
		1464
		1465	Waits until some requests have been handled.
		1466
		1467	Returns the number of requests processed, but is otherwise strictly
		1468	equivalent to:
		1469
		1470	IO::AIO::poll_wait, IO::AIO::poll_cb
		1471
783	=item IO::AIO::nreqs	1472	=item IO::AIO::flush
784		1473
785	Returns the number of requests currently outstanding (i.e. for which their	1474	Wait till all outstanding AIO requests have been handled.
786	callback has not been invoked yet).
787		1475
788	Example: wait till there are no outstanding requests anymore:	1476	Strictly equivalent to:
789		1477
790	IO::AIO::poll_wait, IO::AIO::poll_cb	1478	IO::AIO::poll_wait, IO::AIO::poll_cb
791	while IO::AIO::nreqs;	1479	while IO::AIO::nreqs;
792		1480
793	=item IO::AIO::flush	1481	=item IO::AIO::max_poll_reqs $nreqs
794		1482
795	Wait till all outstanding AIO requests have been handled.	1483	=item IO::AIO::max_poll_time $seconds
796		1484
797	Strictly equivalent to:	1485	These set the maximum number of requests (default C<0>, meaning infinity)
		1486	that are being processed by C<IO::AIO::poll_cb> in one call, respectively
		1487	the maximum amount of time (default C<0>, meaning infinity) spent in
		1488	C<IO::AIO::poll_cb> to process requests (more correctly the mininum amount
		1489	of time C<poll_cb> is allowed to use).
798		1490
799	IO::AIO::poll_wait, IO::AIO::poll_cb	1491	Setting C<max_poll_time> to a non-zero value creates an overhead of one
800	while IO::AIO::nreqs;	1492	syscall per request processed, which is not normally a problem unless your
		1493	callbacks are really really fast or your OS is really really slow (I am
		1494	not mentioning Solaris here). Using C<max_poll_reqs> incurs no overhead.
801		1495
802	=item IO::AIO::poll	1496	Setting these is useful if you want to ensure some level of
		1497	interactiveness when perl is not fast enough to process all requests in
		1498	time.
803		1499
804	Waits until some requests have been handled.	1500	For interactive programs, values such as C<0.01> to C<0.1> should be fine.
805		1501
806	Strictly equivalent to:	1502	Example: Install an Event watcher that automatically calls
		1503	IO::AIO::poll_cb with low priority, to ensure that other parts of the
		1504	program get the CPU sometimes even under high AIO load.
807		1505
808	IO::AIO::poll_wait, IO::AIO::poll_cb	1506	# try not to spend much more than 0.1s in poll_cb
809	if IO::AIO::nreqs;	1507	IO::AIO::max_poll_time 0.1;
		1508
		1509	# use a low priority so other tasks have priority
		1510	Event->io (fd => IO::AIO::poll_fileno,
		1511	poll => 'r', nice => 1,
		1512	cb => &IO::AIO::poll_cb);
		1513
		1514	=back
		1515
		1516	=head3 CONTROLLING THE NUMBER OF THREADS
		1517
		1518	=over
810		1519
811	=item IO::AIO::min_parallel $nthreads	1520	=item IO::AIO::min_parallel $nthreads
812		1521
813	Set the minimum number of AIO threads to C<$nthreads>. The current	1522	Set the minimum number of AIO threads to C<$nthreads>. The current
814	default is C<8>, which means eight asynchronous operations can execute	1523	default is C<8>, which means eight asynchronous operations can execute
815	concurrently at any one time (the number of outstanding requests,	1524	concurrently at any one time (the number of outstanding requests,
816	however, is unlimited).	1525	however, is unlimited).
817		1526
818	IO::AIO starts threads only on demand, when an AIO request is queued and	1527	IO::AIO starts threads only on demand, when an AIO request is queued and
819	no free thread exists.	1528	no free thread exists. Please note that queueing up a hundred requests can
		1529	create demand for a hundred threads, even if it turns out that everything
		1530	is in the cache and could have been processed faster by a single thread.
820		1531
821	It is recommended to keep the number of threads relatively low, as some	1532	It is recommended to keep the number of threads relatively low, as some
822	Linux kernel versions will scale negatively with the number of threads	1533	Linux kernel versions will scale negatively with the number of threads
823	(higher parallelity => MUCH higher latency). With current Linux 2.6	1534	(higher parallelity => MUCH higher latency). With current Linux 2.6
824	versions, 4-32 threads should be fine.	1535	versions, 4-32 threads should be fine.
…		…
838	This module automatically runs C<max_parallel 0> at program end, to ensure	1549	This module automatically runs C<max_parallel 0> at program end, to ensure
839	that all threads are killed and that there are no outstanding requests.	1550	that all threads are killed and that there are no outstanding requests.
840		1551
841	Under normal circumstances you don't need to call this function.	1552	Under normal circumstances you don't need to call this function.
842		1553
		1554	=item IO::AIO::max_idle $nthreads
		1555
		1556	Limit the number of threads (default: 4) that are allowed to idle
		1557	(i.e., threads that did not get a request to process within the idle
		1558	timeout (default: 10 seconds). That means if a thread becomes idle while
		1559	C<$nthreads> other threads are also idle, it will free its resources and
		1560	exit.
		1561
		1562	This is useful when you allow a large number of threads (e.g. 100 or 1000)
		1563	to allow for extremely high load situations, but want to free resources
		1564	under normal circumstances (1000 threads can easily consume 30MB of RAM).
		1565
		1566	The default is probably ok in most situations, especially if thread
		1567	creation is fast. If thread creation is very slow on your system you might
		1568	want to use larger values.
		1569
		1570	=item IO::AIO::idle_timeout $seconds
		1571
		1572	Sets the minimum idle timeout (default 10) after which worker threads are
		1573	allowed to exit. SEe C<IO::AIO::max_idle>.
		1574
843	=item $oldnreqs = IO::AIO::max_outstanding $nreqs	1575	=item IO::AIO::max_outstanding $maxreqs
844		1576
845	[DEPRECATED]
846
847	Sets the maximum number of outstanding requests to C<$nreqs>. If you	1577	Sets the maximum number of outstanding requests to C<$nreqs>. If
848	try to queue up more than this number of requests, the caller will block until	1578	you do queue up more than this number of requests, the next call to
849	some requests have been handled.	1579	C<IO::AIO::poll_cb> (and other functions calling C<poll_cb>, such as
		1580	C<IO::AIO::flush> or C<IO::AIO::poll>) will block until the limit is no
		1581	longer exceeded.
850		1582
851	The default is very large, so normally there is no practical limit. If you	1583	In other words, this setting does not enforce a queue limit, but can be
852	queue up many requests in a loop it often improves speed if you set	1584	used to make poll functions block if the limit is exceeded.
853	this to a relatively low number, such as C<100>.
854		1585
855	This function does not work well together with C<aio_group>'s, and their	1586	This is a very bad function to use in interactive programs because it
856	feeder interface is better suited to limiting concurrency, so do not use	1587	blocks, and a bad way to reduce concurrency because it is inexact: Better
857	this function.	1588	use an C<aio_group> together with a feed callback.
858		1589
859	Under normal circumstances you don't need to call this function.	1590	It's main use is in scripts without an event loop - when you want to stat
		1591	a lot of files, you can write somehting like this:
		1592
		1593	IO::AIO::max_outstanding 32;
		1594
		1595	for my $path (...) {
		1596	aio_stat $path , ...;
		1597	IO::AIO::poll_cb;
		1598	}
		1599
		1600	IO::AIO::flush;
		1601
		1602	The call to C<poll_cb> inside the loop will normally return instantly, but
		1603	as soon as more thna C<32> reqeusts are in-flight, it will block until
		1604	some requests have been handled. This keeps the loop from pushing a large
		1605	number of C<aio_stat> requests onto the queue.
		1606
		1607	The default value for C<max_outstanding> is very large, so there is no
		1608	practical limit on the number of outstanding requests.
860		1609
861	=back	1610	=back
862		1611
		1612	=head3 STATISTICAL INFORMATION
		1613
		1614	=over
		1615
		1616	=item IO::AIO::nreqs
		1617
		1618	Returns the number of requests currently in the ready, execute or pending
		1619	states (i.e. for which their callback has not been invoked yet).
		1620
		1621	Example: wait till there are no outstanding requests anymore:
		1622
		1623	IO::AIO::poll_wait, IO::AIO::poll_cb
		1624	while IO::AIO::nreqs;
		1625
		1626	=item IO::AIO::nready
		1627
		1628	Returns the number of requests currently in the ready state (not yet
		1629	executed).
		1630
		1631	=item IO::AIO::npending
		1632
		1633	Returns the number of requests currently in the pending state (executed,
		1634	but not yet processed by poll_cb).
		1635
		1636	=back
		1637
		1638	=head3 MISCELLANEOUS FUNCTIONS
		1639
		1640	IO::AIO implements some functions that might be useful, but are not
		1641	asynchronous.
		1642
		1643	=over 4
		1644
		1645	=item IO::AIO::sendfile $ofh, $ifh, $offset, $count
		1646
		1647	Calls the C<eio_sendfile_sync> function, which is like C<aio_sendfile>,
		1648	but is blocking (this makes most sense if you know the input data is
		1649	likely cached already and the output filehandle is set to non-blocking
		1650	operations).
		1651
		1652	Returns the number of bytes copied, or C<-1> on error.
		1653
		1654	=item IO::AIO::fadvise $fh, $offset, $len, $advice
		1655
		1656	Simply calls the C<posix_fadvise> function (see its
		1657	manpage for details). The following advice constants are
		1658	avaiable: C<IO::AIO::FADV_NORMAL>, C<IO::AIO::FADV_SEQUENTIAL>,
		1659	C<IO::AIO::FADV_RANDOM>, C<IO::AIO::FADV_NOREUSE>,
		1660	C<IO::AIO::FADV_WILLNEED>, C<IO::AIO::FADV_DONTNEED>.
		1661
		1662	On systems that do not implement C<posix_fadvise>, this function returns
		1663	ENOSYS, otherwise the return value of C<posix_fadvise>.
		1664
		1665	=item IO::AIO::madvise $scalar, $offset, $len, $advice
		1666
		1667	Simply calls the C<posix_madvise> function (see its
		1668	manpage for details). The following advice constants are
		1669	avaiable: C<IO::AIO::MADV_NORMAL>, C<IO::AIO::MADV_SEQUENTIAL>,
		1670	C<IO::AIO::MADV_RANDOM>, C<IO::AIO::MADV_WILLNEED>, C<IO::AIO::MADV_DONTNEED>.
		1671
		1672	On systems that do not implement C<posix_madvise>, this function returns
		1673	ENOSYS, otherwise the return value of C<posix_madvise>.
		1674
		1675	=item IO::AIO::mprotect $scalar, $offset, $len, $protect
		1676
		1677	Simply calls the C<mprotect> function on the preferably AIO::mmap'ed
		1678	$scalar (see its manpage for details). The following protect
		1679	constants are avaiable: C<IO::AIO::PROT_NONE>, C<IO::AIO::PROT_READ>,
		1680	C<IO::AIO::PROT_WRITE>, C<IO::AIO::PROT_EXEC>.
		1681
		1682	On systems that do not implement C<mprotect>, this function returns
		1683	ENOSYS, otherwise the return value of C<mprotect>.
		1684
		1685	=item IO::AIO::mmap $scalar, $length, $prot, $flags, $fh[, $offset]
		1686
		1687	Memory-maps a file (or anonymous memory range) and attaches it to the
		1688	given C<$scalar>, which will act like a string scalar.
		1689
		1690	The only operations allowed on the scalar are C<substr>/C<vec> that don't
		1691	change the string length, and most read-only operations such as copying it
		1692	or searching it with regexes and so on.
		1693
		1694	Anything else is unsafe and will, at best, result in memory leaks.
		1695
		1696	The memory map associated with the C<$scalar> is automatically removed
		1697	when the C<$scalar> is destroyed, or when the C<IO::AIO::mmap> or
		1698	C<IO::AIO::munmap> functions are called.
		1699
		1700	This calls the C<mmap>(2) function internally. See your system's manual
		1701	page for details on the C<$length>, C<$prot> and C<$flags> parameters.
		1702
		1703	The C<$length> must be larger than zero and smaller than the actual
		1704	filesize.
		1705
		1706	C<$prot> is a combination of C<IO::AIO::PROT_NONE>, C<IO::AIO::PROT_EXEC>,
		1707	C<IO::AIO::PROT_READ> and/or C<IO::AIO::PROT_WRITE>,
		1708
		1709	C<$flags> can be a combination of C<IO::AIO::MAP_SHARED> or
		1710	C<IO::AIO::MAP_PRIVATE>, or a number of system-specific flags (when
		1711	not available, the are defined as 0): C<IO::AIO::MAP_ANONYMOUS>
		1712	(which is set to C<MAP_ANON> if your system only provides this
		1713	constant), C<IO::AIO::MAP_HUGETLB>, C<IO::AIO::MAP_LOCKED>,
		1714	C<IO::AIO::MAP_NORESERVE>, C<IO::AIO::MAP_POPULATE> or
		1715	C<IO::AIO::MAP_NONBLOCK>
		1716
		1717	If C<$fh> is C<undef>, then a file descriptor of C<-1> is passed.
		1718
		1719	C<$offset> is the offset from the start of the file - it generally must be
		1720	a multiple of C<IO::AIO::PAGESIZE> and defaults to C<0>.
		1721
		1722	Example:
		1723
		1724	use Digest::MD5;
		1725	use IO::AIO;
		1726
		1727	open my $fh, "<verybigfile"
		1728	or die "$!";
		1729
		1730	IO::AIO::mmap my $data, -s $fh, IO::AIO::PROT_READ, IO::AIO::MAP_SHARED, $fh
		1731	or die "verybigfile: $!";
		1732
		1733	my $fast_md5 = md5 $data;
		1734
		1735	=item IO::AIO::munmap $scalar
		1736
		1737	Removes a previous mmap and undefines the C<$scalar>.
		1738
		1739	=item IO::AIO::munlock $scalar, $offset = 0, $length = undef
		1740
		1741	Calls the C<munlock> function, undoing the effects of a previous
		1742	C<aio_mlock> call (see its description for details).
		1743
		1744	=item IO::AIO::munlockall
		1745
		1746	Calls the C<munlockall> function.
		1747
		1748	On systems that do not implement C<munlockall>, this function returns
		1749	ENOSYS, otherwise the return value of C<munlockall>.
		1750
		1751	=back
		1752
863	=cut	1753	=cut
864		1754
865	# support function to convert a fd into a perl filehandle
866	sub _fd2fh {
867	return undef if $_[0] < 0;
868
869	# try to generate nice filehandles
870	my $sym = "IO::AIO::fd#$_[0]";
871	local *$sym;
872
873	open *$sym, "+<&=$_[0]" # usually works under any unix
874	or open *$sym, "<&=$_[0]" # cygwin needs this
875	or open *$sym, ">&=$_[0]" # or this
876	or return undef;
877
878	*$sym
879	}
880
881	min_parallel 8;	1755	min_parallel 8;
882		1756
883	END {	1757	END { flush }
884	max_parallel 0;
885	}
886		1758
887	1;	1759	1;
888		1760
		1761	=head1 EVENT LOOP INTEGRATION
		1762
		1763	It is recommended to use L<AnyEvent::AIO> to integrate IO::AIO
		1764	automatically into many event loops:
		1765
		1766	# AnyEvent integration (EV, Event, Glib, Tk, POE, urxvt, pureperl...)
		1767	use AnyEvent::AIO;
		1768
		1769	You can also integrate IO::AIO manually into many event loops, here are
		1770	some examples of how to do this:
		1771
		1772	# EV integration
		1773	my $aio_w = EV::io IO::AIO::poll_fileno, EV::READ, \&IO::AIO::poll_cb;
		1774
		1775	# Event integration
		1776	Event->io (fd => IO::AIO::poll_fileno,
		1777	poll => 'r',
		1778	cb => \&IO::AIO::poll_cb);
		1779
		1780	# Glib/Gtk2 integration
		1781	add_watch Glib::IO IO::AIO::poll_fileno,
		1782	in => sub { IO::AIO::poll_cb; 1 };
		1783
		1784	# Tk integration
		1785	Tk::Event::IO->fileevent (IO::AIO::poll_fileno, "",
		1786	readable => \&IO::AIO::poll_cb);
		1787
		1788	# Danga::Socket integration
		1789	Danga::Socket->AddOtherFds (IO::AIO::poll_fileno =>
		1790	\&IO::AIO::poll_cb);
		1791
889	=head2 FORK BEHAVIOUR	1792	=head2 FORK BEHAVIOUR
890		1793
891	This module should do "the right thing" when the process using it forks:	1794	Usage of pthreads in a program changes the semantics of fork
		1795	considerably. Specifically, only async-safe functions can be called after
		1796	fork. Perl doesn't know about this, so in general, you cannot call fork
		1797	with defined behaviour in perl. IO::AIO uses pthreads, so this applies,
		1798	but many other extensions and (for inexplicable reasons) perl itself often
		1799	is linked against pthreads, so this limitation applies.
892		1800
893	Before the fork, IO::AIO enters a quiescent state where no requests	1801	Some operating systems have extensions that allow safe use of fork, and
894	can be added in other threads and no results will be processed. After	1802	this module should do "the right thing" on those, and tries on others. At
895	the fork the parent simply leaves the quiescent state and continues	1803	the time of this writing (2011) only GNU/Linux supports these extensions
896	request/result processing, while the child clears the request/result	1804	to POSIX.
897	queue (so the requests started before the fork will only be handled in
898	the parent). Threads will be started on demand until the limit ste in the
899	parent process has been reached again.
900
901	In short: the parent will, after a short pause, continue as if fork had
902	not been called, while the child will act as if IO::AIO has not been used
903	yet.
904		1805
905	=head2 MEMORY USAGE	1806	=head2 MEMORY USAGE
906		1807
		1808	Per-request usage:
		1809
907	Each aio request uses - depending on your architecture - around 128 bytes	1810	Each aio request uses - depending on your architecture - around 100-200
908	of memory. In addition, stat requests need a stat buffer (possibly a few	1811	bytes of memory. In addition, stat requests need a stat buffer (possibly
909	hundred bytes). Perl scalars and other data passed into aio requests will	1812	a few hundred bytes), readdir requires a result buffer and so on. Perl
910	also be locked.	1813	scalars and other data passed into aio requests will also be locked and
		1814	will consume memory till the request has entered the done state.
911		1815
912	This is now awfully much, so queuing lots of requests is not usually a	1816	This is not awfully much, so queuing lots of requests is not usually a
913	problem.	1817	problem.
914		1818
915	Each thread needs a stack area which is usually around 16k, sometimes much	1819	Per-thread usage:
916	larger, depending on the OS.	1820
		1821	In the execution phase, some aio requests require more memory for
		1822	temporary buffers, and each thread requires a stack and other data
		1823	structures (usually around 16k-128k, depending on the OS).
		1824
		1825	=head1 KNOWN BUGS
		1826
		1827	Known bugs will be fixed in the next release.
917		1828
918	=head1 SEE ALSO	1829	=head1 SEE ALSO
919		1830
920	L<Coro::AIO>.	1831	L<AnyEvent::AIO> for easy integration into event loops, L<Coro::AIO> for a
		1832	more natural syntax.
921		1833
922	=head1 AUTHOR	1834	=head1 AUTHOR
923		1835
924	Marc Lehmann <schmorp@schmorp.de>	1836	Marc Lehmann <schmorp@schmorp.de>
925	http://home.schmorp.de/	1837	http://home.schmorp.de/

Diff Legend

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