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 1.156 by root, Tue Jun 16 23:41:59 2009 UTC

…		…
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", 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
		31	# AnyEvent integration (EV, Event, Glib, Tk, POE, urxvt, pureperl...)
		32	use AnyEvent::AIO;
		33
30	# AnyEvent integration	34	# EV integration
31	open my $fh, "<&=" . IO::AIO::poll_fileno or die "$!";	35	my $aio_w = EV::io IO::AIO::poll_fileno, EV::READ, \&IO::AIO::poll_cb;
32	my $w = AnyEvent->io (fh => $fh, poll => 'r', cb => sub { IO::AIO::poll_cb });
33		36
34	# Event integration	37	# Event integration
35	Event->io (fd => IO::AIO::poll_fileno,	38	Event->io (fd => IO::AIO::poll_fileno,
36	poll => 'r',	39	poll => 'r',
37	cb => \&IO::AIO::poll_cb);	40	cb => \&IO::AIO::poll_cb);
…		…
49	\&IO::AIO::poll_cb);	52	\&IO::AIO::poll_cb);
50		53
51	=head1 DESCRIPTION	54	=head1 DESCRIPTION
52		55
53	This module implements asynchronous I/O using whatever means your	56	This module implements asynchronous I/O using whatever means your
54	operating system supports.	57	operating system supports. It is implemented as an interface to C<libeio>
		58	(L<http://software.schmorp.de/pkg/libeio.html>).
55		59
		60	Asynchronous means that operations that can normally block your program
		61	(e.g. reading from disk) will be done asynchronously: the operation
		62	will still block, but you can do something else in the meantime. This
		63	is extremely useful for programs that need to stay interactive even
		64	when doing heavy I/O (GUI programs, high performance network servers
		65	etc.), but can also be used to easily do operations in parallel that are
		66	normally done sequentially, e.g. stat'ing many files, which is much faster
		67	on a RAID volume or over NFS when you do a number of stat operations
		68	concurrently.
		69
		70	While most of this works on all types of file descriptors (for
		71	example sockets), using these functions on file descriptors that
		72	support nonblocking operation (again, sockets, pipes etc.) is
		73	very inefficient. Use an event loop for that (such as the L<EV>
		74	module): IO::AIO will naturally fit into such an event loop itself.
		75
56	Currently, a number of threads are started that execute your read/writes	76	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	77	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	78	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	79	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	80	functions available on many operating systems. However, they are often
61	(Linux doesn't allow them on normal files currently, for example),	81	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	82	files currently, for example), and they would only support aio_read and
63	functionality would have to be implemented using threads anyway.	83	aio_write, so the remaining functionality would have to be implemented
		84	using threads anyway.
64		85
65	Although the module will work with in the presence of other threads,	86	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	87	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	88	yourself, always call C<poll_cb> from within the same thread, or never
68	call C<poll_cb> (or other C<aio_> functions) recursively.	89	call C<poll_cb> (or other C<aio_> functions) recursively.
69		90
		91	=head2 EXAMPLE
		92
		93	This is a simple example that uses the EV module and loads
		94	F</etc/passwd> asynchronously:
		95
		96	use Fcntl;
		97	use EV;
		98	use IO::AIO;
		99
		100	# register the IO::AIO callback with EV
		101	my $aio_w = EV::io IO::AIO::poll_fileno, EV::READ, \&IO::AIO::poll_cb;
		102
		103	# queue the request to open /etc/passwd
		104	aio_open "/etc/passwd", O_RDONLY, 0, sub {
		105	my $fh = shift
		106	or die "error while opening: $!";
		107
		108	# stat'ing filehandles is generally non-blocking
		109	my $size = -s $fh;
		110
		111	# queue a request to read the file
		112	my $contents;
		113	aio_read $fh, 0, $size, $contents, 0, sub {
		114	$_[0] == $size
		115	or die "short read: $!";
		116
		117	close $fh;
		118
		119	# file contents now in $contents
		120	print $contents;
		121
		122	# exit event loop and program
		123	EV::unloop;
		124	};
		125	};
		126
		127	# possibly queue up other requests, or open GUI windows,
		128	# check for sockets etc. etc.
		129
		130	# process events as long as there are some:
		131	EV::loop;
		132
		133	=head1 REQUEST ANATOMY AND LIFETIME
		134
		135	Every C<aio_*> function creates a request. which is a C data structure not
		136	directly visible to Perl.
		137
		138	If called in non-void context, every request function returns a Perl
		139	object representing the request. In void context, nothing is returned,
		140	which saves a bit of memory.
		141
		142	The perl object is a fairly standard ref-to-hash object. The hash contents
		143	are not used by IO::AIO so you are free to store anything you like in it.
		144
		145	During their existance, aio requests travel through the following states,
		146	in order:
		147
		148	=over 4
		149
		150	=item ready
		151
		152	Immediately after a request is created it is put into the ready state,
		153	waiting for a thread to execute it.
		154
		155	=item execute
		156
		157	A thread has accepted the request for processing and is currently
		158	executing it (e.g. blocking in read).
		159
		160	=item pending
		161
		162	The request has been executed and is waiting for result processing.
		163
		164	While request submission and execution is fully asynchronous, result
		165	processing is not and relies on the perl interpreter calling C<poll_cb>
		166	(or another function with the same effect).
		167
		168	=item result
		169
		170	The request results are processed synchronously by C<poll_cb>.
		171
		172	The C<poll_cb> function will process all outstanding aio requests by
		173	calling their callbacks, freeing memory associated with them and managing
		174	any groups they are contained in.
		175
		176	=item done
		177
		178	Request has reached the end of its lifetime and holds no resources anymore
		179	(except possibly for the Perl object, but its connection to the actual
		180	aio request is severed and calling its methods will either do nothing or
		181	result in a runtime error).
		182
		183	=back
		184
70	=cut	185	=cut
71		186
72	package IO::AIO;	187	package IO::AIO;
		188
		189	use Carp ();
73		190
74	no warnings;	191	no warnings;
75	use strict 'vars';	192	use strict 'vars';
76		193
77	use base 'Exporter';	194	use base 'Exporter';
78		195
79	BEGIN {	196	BEGIN {
80	our $VERSION = '2.0';	197	our $VERSION = '3.23';
81		198
82	our @AIO_REQ = qw(aio_sendfile aio_read aio_write aio_open aio_close aio_stat	199	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	200	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	201	aio_scandir aio_symlink aio_readlink aio_sync aio_fsync
85	aio_group aio_nop);	202	aio_fdatasync aio_sync_file_range aio_pathsync aio_readahead
		203	aio_rename aio_link aio_move aio_copy aio_group
		204	aio_nop aio_mknod aio_load aio_rmtree aio_mkdir aio_chown
		205	aio_chmod aio_utime aio_truncate);
		206
86	our @EXPORT = (@AIO_REQ, qw(aioreq_pri aioreq_nice));	207	our @EXPORT = (@AIO_REQ, qw(aioreq_pri aioreq_nice));
87	our @EXPORT_OK = qw(poll_fileno poll_cb poll_wait flush	208	our @EXPORT_OK = qw(poll_fileno poll_cb poll_wait flush
88	min_parallel max_parallel max_outstanding nreqs);	209	min_parallel max_parallel max_idle
		210	nreqs nready npending nthreads
		211	max_poll_time max_poll_reqs);
		212
		213	push @AIO_REQ, qw(aio_busy); # not exported
89		214
90	@IO::AIO::GRP::ISA = 'IO::AIO::REQ';	215	@IO::AIO::GRP::ISA = 'IO::AIO::REQ';
91		216
92	require XSLoader;	217	require XSLoader;
93	XSLoader::load ("IO::AIO", $VERSION);	218	XSLoader::load ("IO::AIO", $VERSION);
94	}	219	}
95		220
96	=head1 FUNCTIONS	221	=head1 FUNCTIONS
97		222
98	=head2 AIO FUNCTIONS	223	=head2 AIO REQUEST FUNCTIONS
99		224
100	All the C<aio_*> calls are more or less thin wrappers around the syscall	225	All the C<aio_*> calls are more or less thin wrappers around the syscall
101	with the same name (sans C<aio_>). The arguments are similar or identical,	226	with the same name (sans C<aio_>). The arguments are similar or identical,
102	and they all accept an additional (and optional) C<$callback> argument	227	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	228	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	229	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	230	perl, which usually delivers "false") as its sole argument after the given
106	syscall has been executed asynchronously.	231	syscall has been executed asynchronously.
107		232
108	All functions expecting a filehandle keep a copy of the filehandle	233	All functions expecting a filehandle keep a copy of the filehandle
109	internally until the request has finished.	234	internally until the request has finished.
110		235
111	All requests return objects of type L<IO::AIO::REQ> that allow further	236	All functions return request objects of type L<IO::AIO::REQ> that allow
112	manipulation of those requests while they are in-flight.	237	further manipulation of those requests while they are in-flight.
113		238
114	The pathnames you pass to these routines I<must> be absolute and	239	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	240	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	241	request is being executed, the current working directory could have
117	changed. Alternatively, you can make sure that you never change the	242	changed. Alternatively, you can make sure that you never change the
118	current working directory.	243	current working directory anywhere in the program and then use relative
		244	paths.
119		245
120	To encode pathnames to byte form, either make sure you either: a)	246	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	247	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	248	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	249	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)	250	environment, d) use Glib::filename_from_unicode on unicode filenames or e)
125	use something else.	251	use something else to ensure your scalar has the correct contents.
		252
		253	This works, btw. independent of the internal UTF-8 bit, which IO::AIO
		254	handles correctly whether it is set or not.
126		255
127	=over 4	256	=over 4
128		257
129	=item aioreq_pri $pri	258	=item $prev_pri = aioreq_pri [$pri]
130		259
131	Sets the priority for the next aio request. The default priority	260	Returns the priority value that would be used for the next request and, if
		261	C<$pri> is given, sets the priority for the next aio request.
		262
132	is C<0>, the minimum and maximum priorities are C<-4> and C<4>,	263	The default priority is C<0>, the minimum and maximum priorities are C<-4>
133	respectively. Requests with higher priority will be serviced first.	264	and C<4>, respectively. Requests with higher priority will be serviced
		265	first.
134		266
135	The priority will be reset to C<0> after each call to one of the C<aio_>	267	The priority will be reset to C<0> after each call to one of the C<aio_*>
136	functions.	268	functions.
137		269
138	Example: open a file with low priority, then read something from it with	270	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	271	higher priority so the read request is serviced before other low priority
140	open requests (potentially spamming the cache):	272	open requests (potentially spamming the cache):
…		…
147	aio_read $_[0], ..., sub {	279	aio_read $_[0], ..., sub {
148	...	280	...
149	};	281	};
150	};	282	};
151		283
		284
152	=item aioreq_nice $pri_adjust	285	=item aioreq_nice $pri_adjust
153		286
154	Similar to C<aioreq_pri>, but subtracts the given value from the current	287	Similar to C<aioreq_pri>, but subtracts the given value from the current
155	priority, so effects are cumulative.	288	priority, so the effect is cumulative.
		289
156		290
157	=item aio_open $pathname, $flags, $mode, $callback->($fh)	291	=item aio_open $pathname, $flags, $mode, $callback->($fh)
158		292
159	Asynchronously open or create a file and call the callback with a newly	293	Asynchronously open or create a file and call the callback with a newly
160	created filehandle for the file.	294	created filehandle for the file.
…		…
166	list. They are the same as used by C<sysopen>.	300	list. They are the same as used by C<sysopen>.
167		301
168	Likewise, C<$mode> specifies the mode of the newly created file, if it	302	Likewise, C<$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>,	303	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,	304	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).	305	and C<0666> or C<0777> if you do). Note that the C<$mode> will be modified
		306	by the umask in effect then the request is being executed, so better never
		307	change the umask.
172		308
173	Example:	309	Example:
174		310
175	aio_open "/etc/passwd", O_RDONLY, 0, sub {	311	aio_open "/etc/passwd", O_RDONLY, 0, sub {
176	if ($_[0]) {	312	if ($_[0]) {
…		…
179	} else {	315	} else {
180	die "open failed: $!\n";	316	die "open failed: $!\n";
181	}	317	}
182	};	318	};
183		319
		320
184	=item aio_close $fh, $callback->($status)	321	=item aio_close $fh, $callback->($status)
185		322
186	Asynchronously close a file and call the callback with the result	323	Asynchronously close a file and call the callback with the result
187	code. I<WARNING:> although accepted, you should not pass in a perl	324	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		325
192	This is supposed to be a bug in the API, so that might change. It's	326	Unfortunately, you can't do this to perl. Perl I<insists> very strongly on
193	therefore best to avoid this function.	327	closing the file descriptor associated with the filehandle itself.
		328
		329	Therefore, C<aio_close> will not close the filehandle - instead it will
		330	use dup2 to overwrite the file descriptor with the write-end of a pipe
		331	(the pipe fd will be created on demand and will be cached).
		332
		333	Or in other words: the file descriptor will be closed, but it will not be
		334	free for reuse until the perl filehandle is closed.
		335
		336	=cut
194		337
195	=item aio_read $fh,$offset,$length, $data,$dataoffset, $callback->($retval)	338	=item aio_read $fh,$offset,$length, $data,$dataoffset, $callback->($retval)
196		339
197	=item aio_write $fh,$offset,$length, $data,$dataoffset, $callback->($retval)	340	=item aio_write $fh,$offset,$length, $data,$dataoffset, $callback->($retval)
198		341
199	Reads or writes C<length> bytes from the specified C<fh> and C<offset>	342	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	343	C<$offset> into the scalar given by C<$data> and offset C<$dataoffset>
201	callback without the actual number of bytes read (or -1 on error, just	344	and calls the callback without the actual number of bytes read (or -1 on
202	like the syscall).	345	error, just like the syscall).
		346
		347	C<aio_read> will, like C<sysread>, shrink or grow the C<$data> scalar to
		348	offset plus the actual number of bytes read.
		349
		350	If C<$offset> is undefined, then the current file descriptor offset will
		351	be used (and updated), otherwise the file descriptor offset will not be
		352	changed by these calls.
		353
		354	If C<$length> is undefined in C<aio_write>, use the remaining length of
		355	C<$data>.
		356
		357	If C<$dataoffset> is less than zero, it will be counted from the end of
		358	C<$data>.
203		359
204	The C<$data> scalar I<MUST NOT> be modified in any way while the request	360	The C<$data> scalar I<MUST NOT> be modified in any way while the request
205	is outstanding. Modifying it can result in segfaults or WW3 (if the	361	is outstanding. Modifying it can result in segfaults or World War III (if
206	necessary/optional hardware is installed).	362	the necessary/optional hardware is installed).
207		363
208	Example: Read 15 bytes at offset 7 into scalar C<$buffer>, starting at	364	Example: Read 15 bytes at offset 7 into scalar C<$buffer>, starting at
209	offset C<0> within the scalar:	365	offset C<0> within the scalar:
210		366
211	aio_read $fh, 7, 15, $buffer, 0, sub {	367	aio_read $fh, 7, 15, $buffer, 0, sub {
212	$_[0] > 0 or die "read error: $!";	368	$_[0] > 0 or die "read error: $!";
213	print "read $_[0] bytes: <$buffer>\n";	369	print "read $_[0] bytes: <$buffer>\n";
214	};	370	};
215		371
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		372
286	=item aio_sendfile $out_fh, $in_fh, $in_offset, $length, $callback->($retval)	373	=item aio_sendfile $out_fh, $in_fh, $in_offset, $length, $callback->($retval)
287		374
288	Tries to copy C<$length> bytes from C<$in_fh> to C<$out_fh>. It starts	375	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	376	reading at byte offset C<$in_offset>, and starts writing at the current
…		…
303	C<$in_fh> than are written, and there is no way to find out how many	390	C<$in_fh> than are written, and there is no way to find out how many
304	bytes have been read from C<aio_sendfile> alone, as C<aio_sendfile> only	391	bytes have been read from C<aio_sendfile> alone, as C<aio_sendfile> only
305	provides the number of bytes written to C<$out_fh>. Only if the result	392	provides the number of bytes written to C<$out_fh>. Only if the result
306	value equals C<$length> one can assume that C<$length> bytes have been	393	value equals C<$length> one can assume that C<$length> bytes have been
307	read.	394	read.
		395
308		396
309	=item aio_readahead $fh,$offset,$length, $callback->($retval)	397	=item aio_readahead $fh,$offset,$length, $callback->($retval)
310		398
311	C<aio_readahead> populates the page cache with data from a file so that	399	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>	400	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.	406	file. The current file offset of the file is left unchanged.
319		407
320	If that syscall doesn't exist (likely if your OS isn't Linux) it will be	408	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.	409	emulated by simply reading the data, which would have a similar effect.
322		410
		411
323	=item aio_stat $fh_or_path, $callback->($status)	412	=item aio_stat $fh_or_path, $callback->($status)
324		413
325	=item aio_lstat $fh, $callback->($status)	414	=item aio_lstat $fh, $callback->($status)
326		415
327	Works like perl's C<stat> or C<lstat> in void context. The callback will	416	Works like perl's C<stat> or C<lstat> in void context. The callback will
…		…
340	aio_stat "/etc/passwd", sub {	429	aio_stat "/etc/passwd", sub {
341	$_[0] and die "stat failed: $!";	430	$_[0] and die "stat failed: $!";
342	print "size is ", -s _, "\n";	431	print "size is ", -s _, "\n";
343	};	432	};
344		433
		434
		435	=item aio_utime $fh_or_path, $atime, $mtime, $callback->($status)
		436
		437	Works like perl's C<utime> function (including the special case of $atime
		438	and $mtime being undef). Fractional times are supported if the underlying
		439	syscalls support them.
		440
		441	When called with a pathname, uses utimes(2) if available, otherwise
		442	utime(2). If called on a file descriptor, uses futimes(2) if available,
		443	otherwise returns ENOSYS, so this is not portable.
		444
		445	Examples:
		446
		447	# set atime and mtime to current time (basically touch(1)):
		448	aio_utime "path", undef, undef;
		449	# set atime to current time and mtime to beginning of the epoch:
		450	aio_utime "path", time, undef; # undef==0
		451
		452
		453	=item aio_chown $fh_or_path, $uid, $gid, $callback->($status)
		454
		455	Works like perl's C<chown> function, except that C<undef> for either $uid
		456	or $gid is being interpreted as "do not change" (but -1 can also be used).
		457
		458	Examples:
		459
		460	# same as "chown root path" in the shell:
		461	aio_chown "path", 0, -1;
		462	# same as above:
		463	aio_chown "path", 0, undef;
		464
		465
		466	=item aio_truncate $fh_or_path, $offset, $callback->($status)
		467
		468	Works like truncate(2) or ftruncate(2).
		469
		470
		471	=item aio_chmod $fh_or_path, $mode, $callback->($status)
		472
		473	Works like perl's C<chmod> function.
		474
		475
345	=item aio_unlink $pathname, $callback->($status)	476	=item aio_unlink $pathname, $callback->($status)
346		477
347	Asynchronously unlink (delete) a file and call the callback with the	478	Asynchronously unlink (delete) a file and call the callback with the
348	result code.	479	result code.
349		480
		481
		482	=item aio_mknod $path, $mode, $dev, $callback->($status)
		483
		484	[EXPERIMENTAL]
		485
		486	Asynchronously create a device node (or fifo). See mknod(2).
		487
		488	The only (POSIX-) portable way of calling this function is:
		489
		490	aio_mknod $path, IO::AIO::S_IFIFO | $mode, 0, sub { ...
		491
		492
350	=item aio_link $srcpath, $dstpath, $callback->($status)	493	=item aio_link $srcpath, $dstpath, $callback->($status)
351		494
352	Asynchronously create a new link to the existing object at C<$srcpath> at	495	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.	496	the path C<$dstpath> and call the callback with the result code.
354		497
		498
355	=item aio_symlink $srcpath, $dstpath, $callback->($status)	499	=item aio_symlink $srcpath, $dstpath, $callback->($status)
356		500
357	Asynchronously create a new symbolic link to the existing object at C<$srcpath> at	501	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.	502	the path C<$dstpath> and call the callback with the result code.
359		503
		504
		505	=item aio_readlink $path, $callback->($link)
		506
		507	Asynchronously read the symlink specified by C<$path> and pass it to
		508	the callback. If an error occurs, nothing or undef gets passed to the
		509	callback.
		510
		511
360	=item aio_rename $srcpath, $dstpath, $callback->($status)	512	=item aio_rename $srcpath, $dstpath, $callback->($status)
361		513
362	Asynchronously rename the object at C<$srcpath> to C<$dstpath>, just as	514	Asynchronously rename the object at C<$srcpath> to C<$dstpath>, just as
363	rename(2) and call the callback with the result code.	515	rename(2) and call the callback with the result code.
364		516
		517
		518	=item aio_mkdir $pathname, $mode, $callback->($status)
		519
		520	Asynchronously mkdir (create) a directory and call the callback with
		521	the result code. C<$mode> will be modified by the umask at the time the
		522	request is executed, so do not change your umask.
		523
		524
365	=item aio_rmdir $pathname, $callback->($status)	525	=item aio_rmdir $pathname, $callback->($status)
366		526
367	Asynchronously rmdir (delete) a directory and call the callback with the	527	Asynchronously rmdir (delete) a directory and call the callback with the
368	result code.	528	result code.
		529
369		530
370	=item aio_readdir $pathname, $callback->($entries)	531	=item aio_readdir $pathname, $callback->($entries)
371		532
372	Unlike the POSIX call of the same name, C<aio_readdir> reads an entire	533	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	534	directory (i.e. opendir + readdir + closedir). The entries will not be
374	sorted, and will B<NOT> include the C<.> and C<..> entries.	535	sorted, and will B<NOT> include the C<.> and C<..> entries.
375		536
376	The callback a single argument which is either C<undef> or an array-ref	537	The callback is passed a single argument which is either C<undef> or an
377	with the filenames.	538	array-ref with the filenames.
		539
		540
		541	=item aio_readdirx $pathname, $flags, $callback->($entries, $flags)
		542
		543	Quite similar to C<aio_readdir>, but the C<$flags> argument allows to tune
		544	behaviour and output format. In case of an error, C<$entries> will be
		545	C<undef>.
		546
		547	The flags are a combination of the following constants, ORed together (the
		548	flags will also be passed to the callback, possibly modified):
		549
		550	=over 4
		551
		552	=item IO::AIO::READDIR_DENTS
		553
		554	When this flag is off, then the callback gets an arrayref with of names
		555	only (as with C<aio_readdir>), otherwise it gets an arrayref with
		556	C<[$name, $type, $inode]> arrayrefs, each describing a single directory
		557	entry in more detail.
		558
		559	C<$name> is the name of the entry.
		560
		561	C<$type> is one of the C<IO::AIO::DT_xxx> constants:
		562
		563	C<IO::AIO::DT_UNKNOWN>, C<IO::AIO::DT_FIFO>, C<IO::AIO::DT_CHR>, C<IO::AIO::DT_DIR>,
		564	C<IO::AIO::DT_BLK>, C<IO::AIO::DT_REG>, C<IO::AIO::DT_LNK>, C<IO::AIO::DT_SOCK>,
		565	C<IO::AIO::DT_WHT>.
		566
		567	C<IO::AIO::DT_UNKNOWN> means just that: readdir does not know. If you need to
		568	know, you have to run stat yourself. Also, for speed reasons, the C<$type>
		569	scalars are read-only: you can not modify them.
		570
		571	C<$inode> is the inode number (which might not be exact on systems with 64
		572	bit inode numbers and 32 bit perls). This field has unspecified content on
		573	systems that do not deliver the inode information.
		574
		575	=item IO::AIO::READDIR_DIRS_FIRST
		576
		577	When this flag is set, then the names will be returned in an order where
		578	likely directories come first. This is useful when you need to quickly
		579	find directories, or you want to find all directories while avoiding to
		580	stat() each entry.
		581
		582	If the system returns type information in readdir, then this is used
		583	to find directories directly. Otherwise, likely directories are files
		584	beginning with ".", or otherwise files with no dots, of which files with
		585	short names are tried first.
		586
		587	=item IO::AIO::READDIR_STAT_ORDER
		588
		589	When this flag is set, then the names will be returned in an order
		590	suitable for stat()'ing each one. That is, when you plan to stat()
		591	all files in the given directory, then the returned order will likely
		592	be fastest.
		593
		594	If both this flag and C<IO::AIO::READDIR_DIRS_FIRST> are specified, then
		595	the likely dirs come first, resulting in a less optimal stat order.
		596
		597	=item IO::AIO::READDIR_FOUND_UNKNOWN
		598
		599	This flag should not be set when calling C<aio_readdirx>. Instead, it
		600	is being set by C<aio_readdirx>, when any of the C<$type>'s found were
		601	C<IO::AIO::DT_UNKNOWN>. The absense of this flag therefore indicates that all
		602	C<$type>'s are known, which can be used to speed up some algorithms.
		603
		604	=back
		605
		606
		607	=item aio_load $path, $data, $callback->($status)
		608
		609	This is a composite request that tries to fully load the given file into
		610	memory. Status is the same as with aio_read.
		611
		612	=cut
		613
		614	sub aio_load($$;$) {
		615	my ($path, undef, $cb) = @_;
		616	my $data = \$_[1];
		617
		618	my $pri = aioreq_pri;
		619	my $grp = aio_group $cb;
		620
		621	aioreq_pri $pri;
		622	add $grp aio_open $path, O_RDONLY, 0, sub {
		623	my $fh = shift
		624	or return $grp->result (-1);
		625
		626	aioreq_pri $pri;
		627	add $grp aio_read $fh, 0, (-s $fh), $$data, 0, sub {
		628	$grp->result ($_[0]);
		629	};
		630	};
		631
		632	$grp
		633	}
		634
		635	=item aio_copy $srcpath, $dstpath, $callback->($status)
		636
		637	Try to copy the I<file> (directories not supported as either source or
		638	destination) from C<$srcpath> to C<$dstpath> and call the callback with
		639	the C<0> (error) or C<-1> ok.
		640
		641	This is a composite request that creates the destination file with
		642	mode 0200 and copies the contents of the source file into it using
		643	C<aio_sendfile>, followed by restoring atime, mtime, access mode and
		644	uid/gid, in that order.
		645
		646	If an error occurs, the partial destination file will be unlinked, if
		647	possible, except when setting atime, mtime, access mode and uid/gid, where
		648	errors are being ignored.
		649
		650	=cut
		651
		652	sub aio_copy($$;$) {
		653	my ($src, $dst, $cb) = @_;
		654
		655	my $pri = aioreq_pri;
		656	my $grp = aio_group $cb;
		657
		658	aioreq_pri $pri;
		659	add $grp aio_open $src, O_RDONLY, 0, sub {
		660	if (my $src_fh = $_[0]) {
		661	my @stat = stat $src_fh; # hmm, might bock over nfs?
		662
		663	aioreq_pri $pri;
		664	add $grp aio_open $dst, O_CREAT | O_WRONLY | O_TRUNC, 0200, sub {
		665	if (my $dst_fh = $_[0]) {
		666	aioreq_pri $pri;
		667	add $grp aio_sendfile $dst_fh, $src_fh, 0, $stat[7], sub {
		668	if ($_[0] == $stat[7]) {
		669	$grp->result (0);
		670	close $src_fh;
		671
		672	my $ch = sub {
		673	aioreq_pri $pri;
		674	add $grp aio_chmod $dst_fh, $stat[2] & 07777, sub {
		675	aioreq_pri $pri;
		676	add $grp aio_chown $dst_fh, $stat[4], $stat[5], sub {
		677	aioreq_pri $pri;
		678	add $grp aio_close $dst_fh;
		679	}
		680	};
		681	};
		682
		683	aioreq_pri $pri;
		684	add $grp aio_utime $dst_fh, $stat[8], $stat[9], sub {
		685	if ($_[0] < 0 && $! == ENOSYS) {
		686	aioreq_pri $pri;
		687	add $grp aio_utime $dst, $stat[8], $stat[9], $ch;
		688	} else {
		689	$ch->();
		690	}
		691	};
		692	} else {
		693	$grp->result (-1);
		694	close $src_fh;
		695	close $dst_fh;
		696
		697	aioreq $pri;
		698	add $grp aio_unlink $dst;
		699	}
		700	};
		701	} else {
		702	$grp->result (-1);
		703	}
		704	},
		705
		706	} else {
		707	$grp->result (-1);
		708	}
		709	};
		710
		711	$grp
		712	}
		713
		714	=item aio_move $srcpath, $dstpath, $callback->($status)
		715
		716	Try to move the I<file> (directories not supported as either source or
		717	destination) from C<$srcpath> to C<$dstpath> and call the callback with
		718	the C<0> (error) or C<-1> ok.
		719
		720	This is a composite request that tries to rename(2) the file first; if
		721	rename fails with C<EXDEV>, it copies the file with C<aio_copy> and, if
		722	that is successful, unlinks the C<$srcpath>.
		723
		724	=cut
		725
		726	sub aio_move($$;$) {
		727	my ($src, $dst, $cb) = @_;
		728
		729	my $pri = aioreq_pri;
		730	my $grp = aio_group $cb;
		731
		732	aioreq_pri $pri;
		733	add $grp aio_rename $src, $dst, sub {
		734	if ($_[0] && $! == EXDEV) {
		735	aioreq_pri $pri;
		736	add $grp aio_copy $src, $dst, sub {
		737	$grp->result ($_[0]);
		738
		739	if (!$_[0]) {
		740	aioreq_pri $pri;
		741	add $grp aio_unlink $src;
		742	}
		743	};
		744	} else {
		745	$grp->result ($_[0]);
		746	}
		747	};
		748
		749	$grp
		750	}
378		751
379	=item aio_scandir $path, $maxreq, $callback->($dirs, $nondirs)	752	=item aio_scandir $path, $maxreq, $callback->($dirs, $nondirs)
380		753
381	[EXPERIMENTAL due to internal aio_group use]
382
383	Scans a directory (similar to C<aio_readdir>) but additionally tries to	754	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	755	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	756	names, directories you can recurse into (directories), and ones you cannot
386	recurse into (everything else).	757	recurse into (everything else, including symlinks to directories).
387		758
388	C<aio_scandir> is a composite request that creates of many sub requests_	759	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	760	C<$maxreq> specifies the maximum number of outstanding aio requests that
390	this function generates. If it is C<< <= 0 >>, then a suitable default	761	this function generates. If it is C<< <= 0 >>, then a suitable default
391	will be chosen (currently 6).	762	will be chosen (currently 4).
392		763
393	On error, the callback is called without arguments, otherwise it receives	764	On error, the callback is called without arguments, otherwise it receives
394	two array-refs with path-relative entry names.	765	two array-refs with path-relative entry names.
395		766
396	Example:	767	Example:
…		…
403		774
404	Implementation notes.	775	Implementation notes.
405		776
406	The C<aio_readdir> cannot be avoided, but C<stat()>'ing every entry can.	777	The C<aio_readdir> cannot be avoided, but C<stat()>'ing every entry can.
407		778
		779	If readdir returns file type information, then this is used directly to
		780	find directories.
		781
408	After reading the directory, the modification time, size etc. of the	782	Otherwise, after reading the directory, the modification time, size etc.
409	directory before and after the readdir is checked, and if they match (and	783	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	784	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	785	how many entries are directories (if >= 2). Otherwise, no knowledge of the
412	of subdirectories will be assumed.	786	number of subdirectories will be assumed.
413		787
414	Then entries will be sorted into likely directories (everything without	788	Then entries will be sorted into likely directories a non-initial dot
415	a non-initial dot currently) and likely non-directories (everything	789	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,	790	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	791	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	792	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	793	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	794	filesystems might detect the type of the entry without reading the inode
421	data (e.g. ext2fs filetype feature).	795	data (e.g. ext2fs filetype feature), even on systems that cannot return
		796	the filetype information on readdir.
422		797
423	If the known number of directories (link count - 2) has been reached, the	798	If the known number of directories (link count - 2) has been reached, the
424	rest of the entries is assumed to be non-directories.	799	rest of the entries is assumed to be non-directories.
425		800
426	This only works with certainty on POSIX (= UNIX) filesystems, which	801	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	805	as those tend to return 0 or 1 as link counts, which disables the
431	directory counting heuristic.	806	directory counting heuristic.
432		807
433	=cut	808	=cut
434		809
435	sub aio_scandir($$$) {	810	sub aio_scandir($$;$) {
436	my ($path, $maxreq, $cb) = @_;	811	my ($path, $maxreq, $cb) = @_;
437		812
		813	my $pri = aioreq_pri;
		814
438	my $grp = aio_group $cb;	815	my $grp = aio_group $cb;
439		816
440	$maxreq = 6 if $maxreq <= 0;	817	$maxreq = 4 if $maxreq <= 0;
441		818
442	# stat once	819	# stat once
		820	aioreq_pri $pri;
443	add $grp aio_stat $path, sub {	821	add $grp aio_stat $path, sub {
444	return $grp->result () if $_[0];	822	return $grp->result () if $_[0];
445	my $now = time;	823	my $now = time;
446	my $hash1 = join ":", (stat _)[0,1,3,7,9];	824	my $hash1 = join ":", (stat _)[0,1,3,7,9];
447		825
448	# read the directory entries	826	# read the directory entries
		827	aioreq_pri $pri;
449	add $grp aio_readdir $path, sub {	828	add $grp aio_readdirx $path, READDIR_DIRS_FIRST, sub {
450	my $entries = shift	829	my $entries = shift
451	or return $grp->result ();	830	or return $grp->result ();
452		831
453	# stat the dir another time	832	# stat the dir another time
		833	aioreq_pri $pri;
454	add $grp aio_stat $path, sub {	834	add $grp aio_stat $path, sub {
455	my $hash2 = join ":", (stat _)[0,1,3,7,9];	835	my $hash2 = join ":", (stat _)[0,1,3,7,9];
456		836
457	my $ndirs;	837	my $ndirs;
458		838
459	# take the slow route if anything looks fishy	839	# take the slow route if anything looks fishy
460	if ($hash1 ne $hash2 or (stat _)[9] == $now) {	840	if ($hash1 ne $hash2 or (stat _)[9] == $now) {
461	$ndirs = -1;	841	$ndirs = -1;
462	} else {	842	} else {
463	# if nlink == 2, we are finished	843	# if nlink == 2, we are finished
464	# on non-posix-fs's, we rely on nlink < 2	844	# for non-posix-fs's, we rely on nlink < 2
465	$ndirs = (stat _)[3] - 2	845	$ndirs = (stat _)[3] - 2
466	or return $grp->result ([], $entries);	846	or return $grp->result ([], $entries);
467	}	847	}
468		848
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	@$entries];
475
476	my (@dirs, @nondirs);	849	my (@dirs, @nondirs);
477		850
478	my ($statcb, $schedcb);
479	my $nreq = 0;
480
481	my $statgrp = add $grp aio_group;	851	my $statgrp = add $grp aio_group sub {
		852	$grp->result (\@dirs, \@nondirs);
		853	};
482		854
483	$schedcb = sub {	855	limit $statgrp $maxreq;
484	if (@$entries) {	856	feed $statgrp sub {
485	if ($nreq < $maxreq) {	857	return unless @$entries;
486	my $ent = pop @$entries;	858	my $entry = shift @$entries;
		859
		860	aioreq_pri $pri;
		861	add $statgrp aio_stat "$path/$entry/.", sub {
		862	if ($_[0] < 0) {
		863	push @nondirs, $entry;
		864	} else {
		865	# need to check for real directory
		866	aioreq_pri $pri;
		867	add $statgrp aio_lstat "$path/$entry", sub {
		868	if (-d _) {
		869	push @dirs, $entry;
		870
		871	unless (--$ndirs) {
		872	push @nondirs, @$entries;
		873	feed $statgrp;
		874	}
		875	} else {
		876	push @nondirs, $entry;
		877	}
487	$nreq++;	878	}
488	add $statgrp aio_stat "$path/$ent/.", sub { $statcb->($_[0], $ent) };
489	}	879	}
490	} elsif (!$nreq) {
491	# finished
492	$statgrp->cancel;
493	undef $statcb;
494	undef $schedcb;
495	$grp->result (\@dirs, \@nondirs);
496	}	880	};
497	};	881	};
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	};	882	};
528	};	883	};
529	};	884	};
530		885
531	$grp	886	$grp
532	}	887	}
533		888
		889	=item aio_rmtree $path, $callback->($status)
		890
		891	Delete a directory tree starting (and including) C<$path>, return the
		892	status of the final C<rmdir> only. This is a composite request that
		893	uses C<aio_scandir> to recurse into and rmdir directories, and unlink
		894	everything else.
		895
		896	=cut
		897
		898	sub aio_rmtree;
		899	sub aio_rmtree($;$) {
		900	my ($path, $cb) = @_;
		901
		902	my $pri = aioreq_pri;
		903	my $grp = aio_group $cb;
		904
		905	aioreq_pri $pri;
		906	add $grp aio_scandir $path, 0, sub {
		907	my ($dirs, $nondirs) = @_;
		908
		909	my $dirgrp = aio_group sub {
		910	add $grp aio_rmdir $path, sub {
		911	$grp->result ($_[0]);
		912	};
		913	};
		914
		915	(aioreq_pri $pri), add $dirgrp aio_rmtree "$path/$_" for @$dirs;
		916	(aioreq_pri $pri), add $dirgrp aio_unlink "$path/$_" for @$nondirs;
		917
		918	add $grp $dirgrp;
		919	};
		920
		921	$grp
		922	}
		923
		924	=item aio_sync $callback->($status)
		925
		926	Asynchronously call sync and call the callback when finished.
		927
534	=item aio_fsync $fh, $callback->($status)	928	=item aio_fsync $fh, $callback->($status)
535		929
536	Asynchronously call fsync on the given filehandle and call the callback	930	Asynchronously call fsync on the given filehandle and call the callback
537	with the fsync result code.	931	with the fsync result code.
538		932
…		…
542	callback with the fdatasync result code.	936	callback with the fdatasync result code.
543		937
544	If this call isn't available because your OS lacks it or it couldn't be	938	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.	939	detected, it will be emulated by calling C<fsync> instead.
546		940
		941	=item aio_sync_file_range $fh, $offset, $nbytes, $flags, $callback->($status)
		942
		943	Sync the data portion of the file specified by C<$offset> and C<$length>
		944	to disk (but NOT the metadata), by calling the Linux-specific
		945	sync_file_range call. If sync_file_range is not available or it returns
		946	ENOSYS, then fdatasync or fsync is being substituted.
		947
		948	C<$flags> can be a combination of C<IO::AIO::SYNC_FILE_RANGE_WAIT_BEFORE>,
		949	C<IO::AIO::SYNC_FILE_RANGE_WRITE> and
		950	C<IO::AIO::SYNC_FILE_RANGE_WAIT_AFTER>: refer to the sync_file_range
		951	manpage for details.
		952
		953	=item aio_pathsync $path, $callback->($status)
		954
		955	This request tries to open, fsync and close the given path. This is a
		956	composite request intended to sync directories after directory operations
		957	(E.g. rename). This might not work on all operating systems or have any
		958	specific effect, but usually it makes sure that directory changes get
		959	written to disc. It works for anything that can be opened for read-only,
		960	not just directories.
		961
		962	Passes C<0> when everything went ok, and C<-1> on error.
		963
		964	=cut
		965
		966	sub aio_pathsync($;$) {
		967	my ($path, $cb) = @_;
		968
		969	my $pri = aioreq_pri;
		970	my $grp = aio_group $cb;
		971
		972	aioreq_pri $pri;
		973	add $grp aio_open $path, O_RDONLY, 0, sub {
		974	my ($fh) = @_;
		975	if ($fh) {
		976	aioreq_pri $pri;
		977	add $grp aio_fsync $fh, sub {
		978	$grp->result ($_[0]);
		979
		980	aioreq_pri $pri;
		981	add $grp aio_close $fh;
		982	};
		983	} else {
		984	$grp->result (-1);
		985	}
		986	};
		987
		988	$grp
		989	}
		990
547	=item aio_group $callback->(...)	991	=item aio_group $callback->(...)
548
549	[EXPERIMENTAL]
550		992
551	This is a very special aio request: Instead of doing something, it is a	993	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	994	container for other aio requests, which is useful if you want to bundle
553	many requests into a single, composite, request.	995	many requests into a single, composite, request with a definite callback
		996	and the ability to cancel the whole request with its subrequests.
554		997
555	Returns an object of class L<IO::AIO::GRP>. See its documentation below	998	Returns an object of class L<IO::AIO::GRP>. See its documentation below
556	for more info.	999	for more info.
557		1000
558	Example:	1001	Example:
…		…
577	phase and still requires a worker thread. Thus, the callback will not	1020	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	1021	be executed immediately but only after other requests in the queue have
579	entered their execution phase. This can be used to measure request	1022	entered their execution phase. This can be used to measure request
580	latency.	1023	latency.
581		1024
582	=item IO::AIO::aio_sleep $fractional_seconds, $callback->() *NOT EXPORTED*	1025	=item IO::AIO::aio_busy $fractional_seconds, $callback->() *NOT EXPORTED*
583		1026
584	Mainly used for debugging and benchmarking, this aio request puts one of	1027	Mainly used for debugging and benchmarking, this aio request puts one of
585	the request workers to sleep for the given time.	1028	the request workers to sleep for the given time.
586		1029
587	While it is theoretically handy to have simple I/O scheduling requests	1030	While it is theoretically handy to have simple I/O scheduling requests
588	like sleep and file handle readable/writable, the overhead this creates	1031	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	1032	immense (it blocks a thread for a long time) so do not use this function
590	under artificial I/O pressure.	1033	except to put your application under artificial I/O pressure.
591		1034
592	=back	1035	=back
593		1036
594	=head2 IO::AIO::REQ CLASS	1037	=head2 IO::AIO::REQ CLASS
595		1038
596	All non-aggregate C<aio_*> functions return an object of this class when	1039	All non-aggregate C<aio_*> functions return an object of this class when
597	called in non-void context.	1040	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		1041
607	=over 4	1042	=over 4
608		1043
609	=item cancel $req	1044	=item cancel $req
610		1045
611	Cancels the request, if possible. Has the effect of skipping execution	1046	Cancels the request, if possible. Has the effect of skipping execution
612	when entering the B<execute> state and skipping calling the callback when	1047	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	1048	entering the the B<result> state, but will leave the request otherwise
614	untouched. That means that requests that currently execute will not be	1049	untouched (with the exception of readdir). That means that requests that
615	stopped and resources held by the request will not be freed prematurely.	1050	currently execute will not be stopped and resources held by the request
		1051	will not be freed prematurely.
616		1052
617	=item cb $req $callback->(...)	1053	=item cb $req $callback->(...)
618		1054
619	Replace (or simply set) the callback registered to the request.	1055	Replace (or simply set) the callback registered to the request.
620		1056
…		…
664	=item * They can also can also be added to other IO::AIO::GRP objects.	1100	=item * They can also can also be added to other IO::AIO::GRP objects.
665		1101
666	=item * You must not add requests to a group from within the group callback (or	1102	=item * You must not add requests to a group from within the group callback (or
667	any later time).	1103	any later time).
668		1104
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	1105	=back
674		1106
675	Their lifetime, simplified, looks like this: when they are empty, they	1107	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	1108	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	1109	C<done> state, they will also finish. Otherwise they will continue to
678	exist.	1110	exist.
679		1111
680	That means after creating a group you have some time to add requests. And	1112	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	1113	(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	1114	the C<poll_cb>). And in the callbacks of those requests, you can add
683	itself finish.	1115	further requests to the group. And only when all those requests have
		1116	finished will the the group itself finish.
684		1117
685	=over 4	1118	=over 4
686		1119
687	=item add $grp ...	1120	=item add $grp ...
688		1121
…		…
692	be added, including other groups, as long as you do not create circular	1125	be added, including other groups, as long as you do not create circular
693	dependencies.	1126	dependencies.
694		1127
695	Returns all its arguments.	1128	Returns all its arguments.
696		1129
		1130	=item $grp->cancel_subs
		1131
		1132	Cancel all subrequests and clears any feeder, but not the group request
		1133	itself. Useful when you queued a lot of events but got a result early.
		1134
697	=item $grp->result (...)	1135	=item $grp->result (...)
698		1136
699	Set the result value(s) that will be passed to the group callback when all	1137	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.	1138	subrequests have finished and set the groups errno to the current value
		1139	of errno (just like calling C<errno> without an error number). By default,
		1140	no argument will be passed and errno is zero.
		1141
		1142	=item $grp->errno ([$errno])
		1143
		1144	Sets the group errno value to C<$errno>, or the current value of errno
		1145	when the argument is missing.
		1146
		1147	Every aio request has an associated errno value that is restored when
		1148	the callback is invoked. This method lets you change this value from its
		1149	default (0).
		1150
		1151	Calling C<result> will also set errno, so make sure you either set C<$!>
		1152	before the call to C<result>, or call c<errno> after it.
701		1153
702	=item feed $grp $callback->($grp)	1154	=item feed $grp $callback->($grp)
703
704	[VERY EXPERIMENTAL]
705		1155
706	Sets a feeder/generator on this group: every group can have an attached	1156	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,	1157	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,	1158	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	1159	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>	1160	C<aio_scandir> might generate hundreds of thousands C<aio_stat> requests,
711	requests, delaying any later requests for a long time.	1161	delaying any later requests for a long time.
712		1162
713	To avoid this, and allow incremental generation of requests, you can	1163	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	1164	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>,	1165	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	1166	below) requests active in the group itself and is expected to queue more
…		…
720	not impose any limits).	1170	not impose any limits).
721		1171
722	If the feed does not queue more requests when called, it will be	1172	If the feed does not queue more requests when called, it will be
723	automatically removed from the group.	1173	automatically removed from the group.
724		1174
725	If the feed limit is C<0>, it will be set to C<2> automatically.	1175	If the feed limit is C<0> when this method is called, it will be set to
		1176	C<2> automatically.
726		1177
727	Example:	1178	Example:
728		1179
729	# stat all files in @files, but only ever use four aio requests concurrently:	1180	# stat all files in @files, but only ever use four aio requests concurrently:
730		1181
…		…
742	Sets the feeder limit for the group: The feeder will be called whenever	1193	Sets the feeder limit for the group: The feeder will be called whenever
743	the group contains less than this many requests.	1194	the group contains less than this many requests.
744		1195
745	Setting the limit to C<0> will pause the feeding process.	1196	Setting the limit to C<0> will pause the feeding process.
746		1197
		1198	The default value for the limit is C<0>, but note that setting a feeder
		1199	automatically bumps it up to C<2>.
		1200
747	=back	1201	=back
748		1202
749	=head2 SUPPORT FUNCTIONS	1203	=head2 SUPPORT FUNCTIONS
750		1204
		1205	=head3 EVENT PROCESSING AND EVENT LOOP INTEGRATION
		1206
751	=over 4	1207	=over 4
752		1208
753	=item $fileno = IO::AIO::poll_fileno	1209	=item $fileno = IO::AIO::poll_fileno
754		1210
755	Return the I<request result pipe file descriptor>. This filehandle must be	1211	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	1212	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	1213	select and so on, see below or the SYNOPSIS). If the pipe becomes readable
758	to call C<poll_cb> to check the results.	1214	you have to call C<poll_cb> to check the results.
759		1215
760	See C<poll_cb> for an example.	1216	See C<poll_cb> for an example.
761		1217
762	=item IO::AIO::poll_cb	1218	=item IO::AIO::poll_cb
763		1219
764	Process all outstanding events on the result pipe. You have to call this	1220	Process some outstanding events on the result pipe. You have to call this
765	regularly. Returns the number of events processed. Returns immediately	1221	regularly. Returns C<0> if all events could be processed, or C<-1> if it
766	when no events are outstanding.	1222	returned earlier for whatever reason. Returns immediately when no events
		1223	are outstanding. The amount of events processed depends on the settings of
		1224	C<IO::AIO::max_poll_req> and C<IO::AIO::max_poll_time>.
		1225
		1226	If not all requests were processed for whatever reason, the filehandle
		1227	will still be ready when C<poll_cb> returns, so normally you don't have to
		1228	do anything special to have it called later.
767		1229
768	Example: Install an Event watcher that automatically calls	1230	Example: Install an Event watcher that automatically calls
769	IO::AIO::poll_cb with high priority:	1231	IO::AIO::poll_cb with high priority (more examples can be found in the
		1232	SYNOPSIS section, at the top of this document):
770		1233
771	Event->io (fd => IO::AIO::poll_fileno,	1234	Event->io (fd => IO::AIO::poll_fileno,
772	poll => 'r', async => 1,	1235	poll => 'r', async => 1,
773	cb => \&IO::AIO::poll_cb);	1236	cb => \&IO::AIO::poll_cb);
774		1237
		1238	=item IO::AIO::max_poll_reqs $nreqs
		1239
		1240	=item IO::AIO::max_poll_time $seconds
		1241
		1242	These set the maximum number of requests (default C<0>, meaning infinity)
		1243	that are being processed by C<IO::AIO::poll_cb> in one call, respectively
		1244	the maximum amount of time (default C<0>, meaning infinity) spent in
		1245	C<IO::AIO::poll_cb> to process requests (more correctly the mininum amount
		1246	of time C<poll_cb> is allowed to use).
		1247
		1248	Setting C<max_poll_time> to a non-zero value creates an overhead of one
		1249	syscall per request processed, which is not normally a problem unless your
		1250	callbacks are really really fast or your OS is really really slow (I am
		1251	not mentioning Solaris here). Using C<max_poll_reqs> incurs no overhead.
		1252
		1253	Setting these is useful if you want to ensure some level of
		1254	interactiveness when perl is not fast enough to process all requests in
		1255	time.
		1256
		1257	For interactive programs, values such as C<0.01> to C<0.1> should be fine.
		1258
		1259	Example: Install an Event watcher that automatically calls
		1260	IO::AIO::poll_cb with low priority, to ensure that other parts of the
		1261	program get the CPU sometimes even under high AIO load.
		1262
		1263	# try not to spend much more than 0.1s in poll_cb
		1264	IO::AIO::max_poll_time 0.1;
		1265
		1266	# use a low priority so other tasks have priority
		1267	Event->io (fd => IO::AIO::poll_fileno,
		1268	poll => 'r', nice => 1,
		1269	cb => &IO::AIO::poll_cb);
		1270
775	=item IO::AIO::poll_wait	1271	=item IO::AIO::poll_wait
776		1272
		1273	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	1274	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	1275	does a C<select> on the filehandle. This is useful if you want to
779	for some requests to finish).	1276	synchronously wait for some requests to finish).
780		1277
781	See C<nreqs> for an example.	1278	See C<nreqs> for an example.
782		1279
		1280	=item IO::AIO::poll
		1281
		1282	Waits until some requests have been handled.
		1283
		1284	Returns the number of requests processed, but is otherwise strictly
		1285	equivalent to:
		1286
		1287	IO::AIO::poll_wait, IO::AIO::poll_cb
		1288
783	=item IO::AIO::nreqs	1289	=item IO::AIO::flush
784		1290
785	Returns the number of requests currently outstanding (i.e. for which their	1291	Wait till all outstanding AIO requests have been handled.
786	callback has not been invoked yet).
787		1292
788	Example: wait till there are no outstanding requests anymore:	1293	Strictly equivalent to:
789		1294
790	IO::AIO::poll_wait, IO::AIO::poll_cb	1295	IO::AIO::poll_wait, IO::AIO::poll_cb
791	while IO::AIO::nreqs;	1296	while IO::AIO::nreqs;
792		1297
793	=item IO::AIO::flush	1298	=back
794		1299
795	Wait till all outstanding AIO requests have been handled.	1300	=head3 CONTROLLING THE NUMBER OF THREADS
796		1301
797	Strictly equivalent to:	1302	=over
798
799	IO::AIO::poll_wait, IO::AIO::poll_cb
800	while IO::AIO::nreqs;
801
802	=item IO::AIO::poll
803
804	Waits until some requests have been handled.
805
806	Strictly equivalent to:
807
808	IO::AIO::poll_wait, IO::AIO::poll_cb
809	if IO::AIO::nreqs;
810		1303
811	=item IO::AIO::min_parallel $nthreads	1304	=item IO::AIO::min_parallel $nthreads
812		1305
813	Set the minimum number of AIO threads to C<$nthreads>. The current	1306	Set the minimum number of AIO threads to C<$nthreads>. The current
814	default is C<8>, which means eight asynchronous operations can execute	1307	default is C<8>, which means eight asynchronous operations can execute
815	concurrently at any one time (the number of outstanding requests,	1308	concurrently at any one time (the number of outstanding requests,
816	however, is unlimited).	1309	however, is unlimited).
817		1310
818	IO::AIO starts threads only on demand, when an AIO request is queued and	1311	IO::AIO starts threads only on demand, when an AIO request is queued and
819	no free thread exists.	1312	no free thread exists. Please note that queueing up a hundred requests can
		1313	create demand for a hundred threads, even if it turns out that everything
		1314	is in the cache and could have been processed faster by a single thread.
820		1315
821	It is recommended to keep the number of threads relatively low, as some	1316	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	1317	Linux kernel versions will scale negatively with the number of threads
823	(higher parallelity => MUCH higher latency). With current Linux 2.6	1318	(higher parallelity => MUCH higher latency). With current Linux 2.6
824	versions, 4-32 threads should be fine.	1319	versions, 4-32 threads should be fine.
…		…
838	This module automatically runs C<max_parallel 0> at program end, to ensure	1333	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.	1334	that all threads are killed and that there are no outstanding requests.
840		1335
841	Under normal circumstances you don't need to call this function.	1336	Under normal circumstances you don't need to call this function.
842		1337
		1338	=item IO::AIO::max_idle $nthreads
		1339
		1340	Limit the number of threads (default: 4) that are allowed to idle (i.e.,
		1341	threads that did not get a request to process within 10 seconds). That
		1342	means if a thread becomes idle while C<$nthreads> other threads are also
		1343	idle, it will free its resources and exit.
		1344
		1345	This is useful when you allow a large number of threads (e.g. 100 or 1000)
		1346	to allow for extremely high load situations, but want to free resources
		1347	under normal circumstances (1000 threads can easily consume 30MB of RAM).
		1348
		1349	The default is probably ok in most situations, especially if thread
		1350	creation is fast. If thread creation is very slow on your system you might
		1351	want to use larger values.
		1352
843	=item $oldnreqs = IO::AIO::max_outstanding $nreqs	1353	=item IO::AIO::max_outstanding $maxreqs
844		1354
845	[DEPRECATED]	1355	This is a very bad function to use in interactive programs because it
		1356	blocks, and a bad way to reduce concurrency because it is inexact: Better
		1357	use an C<aio_group> together with a feed callback.
846		1358
847	Sets the maximum number of outstanding requests to C<$nreqs>. If you	1359	Sets the maximum number of outstanding requests to C<$nreqs>. If you
848	try to queue up more than this number of requests, the caller will block until	1360	do queue up more than this number of requests, the next call to the
849	some requests have been handled.	1361	C<poll_cb> (and C<poll_some> and other functions calling C<poll_cb>)
		1362	function will block until the limit is no longer exceeded.
850		1363
851	The default is very large, so normally there is no practical limit. If you	1364	The default value is very large, so there is no practical limit on the
852	queue up many requests in a loop it often improves speed if you set	1365	number of outstanding requests.
853	this to a relatively low number, such as C<100>.
854		1366
855	This function does not work well together with C<aio_group>'s, and their	1367	You can still queue as many requests as you want. Therefore,
856	feeder interface is better suited to limiting concurrency, so do not use	1368	C<max_outstanding> is mainly useful in simple scripts (with low values) or
857	this function.	1369	as a stop gap to shield against fatal memory overflow (with large values).
858
859	Under normal circumstances you don't need to call this function.
860		1370
861	=back	1371	=back
862		1372
		1373	=head3 STATISTICAL INFORMATION
		1374
		1375	=over
		1376
		1377	=item IO::AIO::nreqs
		1378
		1379	Returns the number of requests currently in the ready, execute or pending
		1380	states (i.e. for which their callback has not been invoked yet).
		1381
		1382	Example: wait till there are no outstanding requests anymore:
		1383
		1384	IO::AIO::poll_wait, IO::AIO::poll_cb
		1385	while IO::AIO::nreqs;
		1386
		1387	=item IO::AIO::nready
		1388
		1389	Returns the number of requests currently in the ready state (not yet
		1390	executed).
		1391
		1392	=item IO::AIO::npending
		1393
		1394	Returns the number of requests currently in the pending state (executed,
		1395	but not yet processed by poll_cb).
		1396
		1397	=back
		1398
863	=cut	1399	=cut
864		1400
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;	1401	min_parallel 8;
882		1402
883	END {	1403	END { flush }
884	max_parallel 0;
885	}
886		1404
887	1;	1405	1;
888		1406
889	=head2 FORK BEHAVIOUR	1407	=head2 FORK BEHAVIOUR
890		1408
891	This module should do "the right thing" when the process using it forks:	1409	This module should do "the right thing" when the process using it forks:
892		1410
893	Before the fork, IO::AIO enters a quiescent state where no requests	1411	Before the fork, IO::AIO enters a quiescent state where no requests
894	can be added in other threads and no results will be processed. After	1412	can be added in other threads and no results will be processed. After
895	the fork the parent simply leaves the quiescent state and continues	1413	the fork the parent simply leaves the quiescent state and continues
896	request/result processing, while the child clears the request/result	1414	request/result processing, while the child frees the request/result queue
897	queue (so the requests started before the fork will only be handled in	1415	(so that the requests started before the fork will only be handled in the
898	the parent). Threads will be started on demand until the limit ste in the	1416	parent). Threads will be started on demand until the limit set in the
899	parent process has been reached again.	1417	parent process has been reached again.
900		1418
901	In short: the parent will, after a short pause, continue as if fork had	1419	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	1420	not been called, while the child will act as if IO::AIO has not been used
903	yet.	1421	yet.
904		1422
905	=head2 MEMORY USAGE	1423	=head2 MEMORY USAGE
906		1424
		1425	Per-request usage:
		1426
907	Each aio request uses - depending on your architecture - around 128 bytes	1427	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	1428	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	1429	a few hundred bytes), readdir requires a result buffer and so on. Perl
910	also be locked.	1430	scalars and other data passed into aio requests will also be locked and
		1431	will consume memory till the request has entered the done state.
911		1432
912	This is now awfully much, so queuing lots of requests is not usually a	1433	This is not awfully much, so queuing lots of requests is not usually a
913	problem.	1434	problem.
914		1435
915	Each thread needs a stack area which is usually around 16k, sometimes much	1436	Per-thread usage:
916	larger, depending on the OS.	1437
		1438	In the execution phase, some aio requests require more memory for
		1439	temporary buffers, and each thread requires a stack and other data
		1440	structures (usually around 16k-128k, depending on the OS).
		1441
		1442	=head1 KNOWN BUGS
		1443
		1444	Known bugs will be fixed in the next release.
917		1445
918	=head1 SEE ALSO	1446	=head1 SEE ALSO
919		1447
920	L<Coro::AIO>.	1448	L<AnyEvent::AIO> for easy integration into event loops, L<Coro::AIO> for a
		1449	more natural syntax.
921		1450
922	=head1 AUTHOR	1451	=head1 AUTHOR
923		1452
924	Marc Lehmann <schmorp@schmorp.de>	1453	Marc Lehmann <schmorp@schmorp.de>
925	http://home.schmorp.de/	1454	http://home.schmorp.de/

Diff Legend

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