[ViewVC] Diff of: cvs/IO-AIO/AIO.pm

Comparing IO-AIO/AIO.pm (file contents):
Revision 1.67 by root, Tue Oct 24 02:25:16 2006 UTC vs.
Revision 1.113 by root, Thu Sep 20 14:06:21 2007 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
…		…
18	};	19	};
19		20
20	# version 2+ has request and group objects	21	# version 2+ has request and group objects
21	use IO::AIO 2;	22	use IO::AIO 2;
22		23
		24	aioreq_pri 4; # give next request a very high priority
23	my $req = aio_unlink "/tmp/file", sub { };	25	my $req = aio_unlink "/tmp/file", sub { };
24	$req->cancel; # cancel request if still in queue	26	$req->cancel; # cancel request if still in queue
25		27
26	my $grp = aio_group sub { print "all stats done\n" };	28	my $grp = aio_group sub { print "all stats done\n" };
27	add $grp aio_stat "..." for ...;	29	add $grp aio_stat "..." for ...;
…		…
50	=head1 DESCRIPTION	52	=head1 DESCRIPTION
51		53
52	This module implements asynchronous I/O using whatever means your	54	This module implements asynchronous I/O using whatever means your
53	operating system supports.	55	operating system supports.
54		56
		57	Asynchronous means that operations that can normally block your program
		58	(e.g. reading from disk) will be done asynchronously: the operation
		59	will still block, but you can do something else in the meantime. This
		60	is extremely useful for programs that need to stay interactive even
		61	when doing heavy I/O (GUI programs, high performance network servers
		62	etc.), but can also be used to easily do operations in parallel that are
		63	normally done sequentially, e.g. stat'ing many files, which is much faster
		64	on a RAID volume or over NFS when you do a number of stat operations
		65	concurrently.
		66
		67	While most of this works on all types of file descriptors (for
		68	example sockets), using these functions on file descriptors that
		69	support nonblocking operation (again, sockets, pipes etc.) is very
		70	inefficient. Use an event loop for that (such as the L<Event\|Event>
		71	module): IO::AIO will naturally fit into such an event loop itself.
		72
55	Currently, a number of threads are started that execute your read/writes	73	In this version, a number of threads are started that execute your
56	and signal their completion. You don't need thread support in perl, and	74	requests and signal their completion. You don't need thread support
57	the threads created by this module will not be visible to perl. In the	75	in perl, and the threads created by this module will not be visible
58	future, this module might make use of the native aio functions available	76	to perl. In the future, this module might make use of the native aio
59	on many operating systems. However, they are often not well-supported	77	functions available on many operating systems. However, they are often
60	(Linux doesn't allow them on normal files currently, for example),	78	not well-supported or restricted (GNU/Linux doesn't allow them on normal
61	and they would only support aio_read and aio_write, so the remaining	79	files currently, for example), and they would only support aio_read and
62	functionality would have to be implemented using threads anyway.	80	aio_write, so the remaining functionality would have to be implemented
		81	using threads anyway.
63		82
64	Although the module will work with in the presence of other threads, it is	83	Although the module will work in the presence of other (Perl-) threads,
65	currently not reentrant, so use appropriate locking yourself, always call	84	it is currently not reentrant in any way, so use appropriate locking
66	C<poll_cb> from within the same thread, or never call C<poll_cb> (or other	85	yourself, always call C<poll_cb> from within the same thread, or never
67	C<aio_> functions) recursively.	86	call C<poll_cb> (or other C<aio_> functions) recursively.
		87
		88	=head2 EXAMPLE
		89
		90	This is a simple example that uses the Event module and loads
		91	F</etc/passwd> asynchronously:
		92
		93	use Fcntl;
		94	use Event;
		95	use IO::AIO;
		96
		97	# register the IO::AIO callback with Event
		98	Event->io (fd => IO::AIO::poll_fileno,
		99	poll => 'r',
		100	cb => \&IO::AIO::poll_cb);
		101
		102	# queue the request to open /etc/passwd
		103	aio_open "/etc/passwd", O_RDONLY, 0, sub {
		104	my $fh = shift
		105	or die "error while opening: $!";
		106
		107	# stat'ing filehandles is generally non-blocking
		108	my $size = -s $fh;
		109
		110	# queue a request to read the file
		111	my $contents;
		112	aio_read $fh, 0, $size, $contents, 0, sub {
		113	$_[0] == $size
		114	or die "short read: $!";
		115
		116	close $fh;
		117
		118	# file contents now in $contents
		119	print $contents;
		120
		121	# exit event loop and program
		122	Event::unloop;
		123	};
		124	};
		125
		126	# possibly queue up other requests, or open GUI windows,
		127	# check for sockets etc. etc.
		128
		129	# process events as long as there are some:
		130	Event::loop;
		131
		132	=head1 REQUEST ANATOMY AND LIFETIME
		133
		134	Every C<aio_*> function creates a request. which is a C data structure not
		135	directly visible to Perl.
		136
		137	If called in non-void context, every request function returns a Perl
		138	object representing the request. In void context, nothing is returned,
		139	which saves a bit of memory.
		140
		141	The perl object is a fairly standard ref-to-hash object. The hash contents
		142	are not used by IO::AIO so you are free to store anything you like in it.
		143
		144	During their existance, aio requests travel through the following states,
		145	in order:
		146
		147	=over 4
		148
		149	=item ready
		150
		151	Immediately after a request is created it is put into the ready state,
		152	waiting for a thread to execute it.
		153
		154	=item execute
		155
		156	A thread has accepted the request for processing and is currently
		157	executing it (e.g. blocking in read).
		158
		159	=item pending
		160
		161	The request has been executed and is waiting for result processing.
		162
		163	While request submission and execution is fully asynchronous, result
		164	processing is not and relies on the perl interpreter calling C<poll_cb>
		165	(or another function with the same effect).
		166
		167	=item result
		168
		169	The request results are processed synchronously by C<poll_cb>.
		170
		171	The C<poll_cb> function will process all outstanding aio requests by
		172	calling their callbacks, freeing memory associated with them and managing
		173	any groups they are contained in.
		174
		175	=item done
		176
		177	Request has reached the end of its lifetime and holds no resources anymore
		178	(except possibly for the Perl object, but its connection to the actual
		179	aio request is severed and calling its methods will either do nothing or
		180	result in a runtime error).
		181
		182	=back
68		183
69	=cut	184	=cut
70		185
71	package IO::AIO;	186	package IO::AIO;
72		187
…		…
74	use strict 'vars';	189	use strict 'vars';
75		190
76	use base 'Exporter';	191	use base 'Exporter';
77		192
78	BEGIN {	193	BEGIN {
79	our $VERSION = '2.0';	194	our $VERSION = '2.4';
80		195
81	our @AIO_REQ = qw(aio_sendfile aio_read aio_write aio_open aio_close aio_stat	196	our @AIO_REQ = qw(aio_sendfile aio_read aio_write aio_open aio_close aio_stat
82	aio_lstat aio_unlink aio_rmdir aio_readdir aio_scandir aio_symlink	197	aio_lstat aio_unlink aio_rmdir aio_readdir aio_scandir aio_symlink
83	aio_fsync aio_fdatasync aio_readahead aio_rename aio_link aio_move	198	aio_readlink aio_fsync aio_fdatasync aio_readahead aio_rename aio_link
84	aio_group aio_nop);	199	aio_move aio_copy aio_group aio_nop aio_mknod aio_load aio_rmtree aio_mkdir
		200	aio_chown aio_chmod aio_utime aio_truncate);
85	our @EXPORT = (@AIO_REQ, qw(aioreq_pri));	201	our @EXPORT = (@AIO_REQ, qw(aioreq_pri aioreq_nice aio_block));
86	our @EXPORT_OK = qw(poll_fileno poll_cb poll_wait flush	202	our @EXPORT_OK = qw(poll_fileno poll_cb poll_wait flush
87	min_parallel max_parallel max_outstanding nreqs);	203	min_parallel max_parallel max_idle
		204	nreqs nready npending nthreads
		205	max_poll_time max_poll_reqs);
88		206
89	@IO::AIO::GRP::ISA = 'IO::AIO::REQ';	207	@IO::AIO::GRP::ISA = 'IO::AIO::REQ';
90		208
91	require XSLoader;	209	require XSLoader;
92	XSLoader::load ("IO::AIO", $VERSION);	210	XSLoader::load ("IO::AIO", $VERSION);
93	}	211	}
94		212
95	=head1 FUNCTIONS	213	=head1 FUNCTIONS
96		214
97	=head2 AIO FUNCTIONS	215	=head2 AIO REQUEST FUNCTIONS
98		216
99	All the C<aio_*> calls are more or less thin wrappers around the syscall	217	All the C<aio_*> calls are more or less thin wrappers around the syscall
100	with the same name (sans C<aio_>). The arguments are similar or identical,	218	with the same name (sans C<aio_>). The arguments are similar or identical,
101	and they all accept an additional (and optional) C<$callback> argument	219	and they all accept an additional (and optional) C<$callback> argument
102	which must be a code reference. This code reference will get called with	220	which must be a code reference. This code reference will get called with
…		…
105	syscall has been executed asynchronously.	223	syscall has been executed asynchronously.
106		224
107	All functions expecting a filehandle keep a copy of the filehandle	225	All functions expecting a filehandle keep a copy of the filehandle
108	internally until the request has finished.	226	internally until the request has finished.
109		227
110	All requests return objects of type L<IO::AIO::REQ> that allow further	228	All functions return request objects of type L<IO::AIO::REQ> that allow
111	manipulation of those requests while they are in-flight.	229	further manipulation of those requests while they are in-flight.
112		230
113	The pathnames you pass to these routines I<must> be absolute and	231	The pathnames you pass to these routines I<must> be absolute and
114	encoded in byte form. The reason for the former is that at the time the	232	encoded as octets. The reason for the former is that at the time the
115	request is being executed, the current working directory could have	233	request is being executed, the current working directory could have
116	changed. Alternatively, you can make sure that you never change the	234	changed. Alternatively, you can make sure that you never change the
117	current working directory.	235	current working directory anywhere in the program and then use relative
		236	paths.
118		237
119	To encode pathnames to byte form, either make sure you either: a)	238	To encode pathnames as octets, either make sure you either: a) always pass
120	always pass in filenames you got from outside (command line, readdir	239	in filenames you got from outside (command line, readdir etc.) without
121	etc.), b) are ASCII or ISO 8859-1, c) use the Encode module and encode	240	tinkering, b) are ASCII or ISO 8859-1, c) use the Encode module and encode
122	your pathnames to the locale (or other) encoding in effect in the user	241	your pathnames to the locale (or other) encoding in effect in the user
123	environment, d) use Glib::filename_from_unicode on unicode filenames or e)	242	environment, d) use Glib::filename_from_unicode on unicode filenames or e)
124	use something else.	243	use something else to ensure your scalar has the correct contents.
		244
		245	This works, btw. independent of the internal UTF-8 bit, which IO::AIO
		246	handles correctly wether it is set or not.
125		247
126	=over 4	248	=over 4
		249
		250	=item $prev_pri = aioreq_pri [$pri]
		251
		252	Returns the priority value that would be used for the next request and, if
		253	C<$pri> is given, sets the priority for the next aio request.
		254
		255	The default priority is C<0>, the minimum and maximum priorities are C<-4>
		256	and C<4>, respectively. Requests with higher priority will be serviced
		257	first.
		258
		259	The priority will be reset to C<0> after each call to one of the C<aio_*>
		260	functions.
		261
		262	Example: open a file with low priority, then read something from it with
		263	higher priority so the read request is serviced before other low priority
		264	open requests (potentially spamming the cache):
		265
		266	aioreq_pri -3;
		267	aio_open ..., sub {
		268	return unless $_[0];
		269
		270	aioreq_pri -2;
		271	aio_read $_[0], ..., sub {
		272	...
		273	};
		274	};
		275
		276
		277	=item aioreq_nice $pri_adjust
		278
		279	Similar to C<aioreq_pri>, but subtracts the given value from the current
		280	priority, so the effect is cumulative.
		281
127		282
128	=item aio_open $pathname, $flags, $mode, $callback->($fh)	283	=item aio_open $pathname, $flags, $mode, $callback->($fh)
129		284
130	Asynchronously open or create a file and call the callback with a newly	285	Asynchronously open or create a file and call the callback with a newly
131	created filehandle for the file.	286	created filehandle for the file.
…		…
137	list. They are the same as used by C<sysopen>.	292	list. They are the same as used by C<sysopen>.
138		293
139	Likewise, C<$mode> specifies the mode of the newly created file, if it	294	Likewise, C<$mode> specifies the mode of the newly created file, if it
140	didn't exist and C<O_CREAT> has been given, just like perl's C<sysopen>,	295	didn't exist and C<O_CREAT> has been given, just like perl's C<sysopen>,
141	except that it is mandatory (i.e. use C<0> if you don't create new files,	296	except that it is mandatory (i.e. use C<0> if you don't create new files,
142	and C<0666> or C<0777> if you do).	297	and C<0666> or C<0777> if you do). Note that the C<$mode> will be modified
		298	by the umask in effect then the request is being executed, so better never
		299	change the umask.
143		300
144	Example:	301	Example:
145		302
146	aio_open "/etc/passwd", O_RDONLY, 0, sub {	303	aio_open "/etc/passwd", O_RDONLY, 0, sub {
147	if ($_[0]) {	304	if ($_[0]) {
…		…
150	} else {	307	} else {
151	die "open failed: $!\n";	308	die "open failed: $!\n";
152	}	309	}
153	};	310	};
154		311
		312
155	=item aio_close $fh, $callback->($status)	313	=item aio_close $fh, $callback->($status)
156		314
157	Asynchronously close a file and call the callback with the result	315	Asynchronously close a file and call the callback with the result
158	code. I<WARNING:> although accepted, you should not pass in a perl	316	code. I<WARNING:> although accepted, you should not pass in a perl
159	filehandle here, as perl will likely close the file descriptor another	317	filehandle here, as perl will likely close the file descriptor another
…		…
161	C<close> or just let filehandles go out of scope.	319	C<close> or just let filehandles go out of scope.
162		320
163	This is supposed to be a bug in the API, so that might change. It's	321	This is supposed to be a bug in the API, so that might change. It's
164	therefore best to avoid this function.	322	therefore best to avoid this function.
165		323
		324
166	=item aio_read $fh,$offset,$length, $data,$dataoffset, $callback->($retval)	325	=item aio_read $fh,$offset,$length, $data,$dataoffset, $callback->($retval)
167		326
168	=item aio_write $fh,$offset,$length, $data,$dataoffset, $callback->($retval)	327	=item aio_write $fh,$offset,$length, $data,$dataoffset, $callback->($retval)
169		328
170	Reads or writes C<length> bytes from the specified C<fh> and C<offset>	329	Reads or writes C<$length> bytes from the specified C<$fh> and C<$offset>
171	into the scalar given by C<data> and offset C<dataoffset> and calls the	330	into the scalar given by C<$data> and offset C<$dataoffset> and calls the
172	callback without the actual number of bytes read (or -1 on error, just	331	callback without the actual number of bytes read (or -1 on error, just
173	like the syscall).	332	like the syscall).
174		333
		334	If C<$offset> is undefined, then the current file descriptor offset will
		335	be used (and updated), otherwise the file descriptor offset will not be
		336	changed by these calls.
		337
		338	If C<$length> is undefined in C<aio_write>, use the remaining length of C<$data>.
		339
		340	If C<$dataoffset> is less than zero, it will be counted from the end of
		341	C<$data>.
		342
175	The C<$data> scalar I<MUST NOT> be modified in any way while the request	343	The C<$data> scalar I<MUST NOT> be modified in any way while the request
176	is outstanding. Modifying it can result in segfaults or WW3 (if the	344	is outstanding. Modifying it can result in segfaults or World War III (if
177	necessary/optional hardware is installed).	345	the necessary/optional hardware is installed).
178		346
179	Example: Read 15 bytes at offset 7 into scalar C<$buffer>, starting at	347	Example: Read 15 bytes at offset 7 into scalar C<$buffer>, starting at
180	offset C<0> within the scalar:	348	offset C<0> within the scalar:
181		349
182	aio_read $fh, 7, 15, $buffer, 0, sub {	350	aio_read $fh, 7, 15, $buffer, 0, sub {
183	$_[0] > 0 or die "read error: $!";	351	$_[0] > 0 or die "read error: $!";
184	print "read $_[0] bytes: <$buffer>\n";	352	print "read $_[0] bytes: <$buffer>\n";
185	};	353	};
186		354
187	=item aio_move $srcpath, $dstpath, $callback->($status)
188
189	[EXPERIMENTAL due to internal aio_group use]
190
191	Try to move the I<file> (directories not supported as either source or
192	destination) from C<$srcpath> to C<$dstpath> and call the callback with
193	the C<0> (error) or C<-1> ok.
194
195	This is a composite request that tries to rename(2) the file first. If
196	rename files with C<EXDEV>, it creates the destination file with mode 0200
197	and copies the contents of the source file into it using C<aio_sendfile>,
198	followed by restoring atime, mtime, access mode and uid/gid, in that
199	order, and unlinking the C<$srcpath>.
200
201	If an error occurs, the partial destination file will be unlinked, if
202	possible, except when setting atime, mtime, access mode and uid/gid, where
203	errors are being ignored.
204
205	=cut
206
207	sub aio_move($$$) {
208	my ($src, $dst, $cb) = @_;
209
210	my $grp = aio_group $cb;
211
212	add $grp aio_rename $src, $dst, sub {
213	if ($_[0] && $! == EXDEV) {
214	add $grp aio_open $src, O_RDONLY, 0, sub {
215	if (my $src_fh = $_[0]) {
216	my @stat = stat $src_fh;
217
218	add $grp aio_open $dst, O_WRONLY, 0200, sub {
219	if (my $dst_fh = $_[0]) {
220	add $grp aio_sendfile $dst_fh, $src_fh, 0, $stat[7], sub {
221	close $src_fh;
222
223	if ($_[0] == $stat[7]) {
224	utime $stat[8], $stat[9], $dst;
225	chmod $stat[2] & 07777, $dst_fh;
226	chown $stat[4], $stat[5], $dst_fh;
227	close $dst_fh;
228
229	add $grp aio_unlink $src, sub {
230	$grp->result ($_[0]);
231	};
232	} else {
233	my $errno = $!;
234	add $grp aio_unlink $dst, sub {
235	$! = $errno;
236	$grp->result (-1);
237	};
238	}
239	};
240	} else {
241	$grp->result (-1);
242	}
243	},
244
245	} else {
246	$grp->result (-1);
247	}
248	};
249	} else {
250	$grp->result ($_[0]);
251	}
252	};
253
254	$grp
255	}
256		355
257	=item aio_sendfile $out_fh, $in_fh, $in_offset, $length, $callback->($retval)	356	=item aio_sendfile $out_fh, $in_fh, $in_offset, $length, $callback->($retval)
258		357
259	Tries to copy C<$length> bytes from C<$in_fh> to C<$out_fh>. It starts	358	Tries to copy C<$length> bytes from C<$in_fh> to C<$out_fh>. It starts
260	reading at byte offset C<$in_offset>, and starts writing at the current	359	reading at byte offset C<$in_offset>, and starts writing at the current
…		…
274	C<$in_fh> than are written, and there is no way to find out how many	373	C<$in_fh> than are written, and there is no way to find out how many
275	bytes have been read from C<aio_sendfile> alone, as C<aio_sendfile> only	374	bytes have been read from C<aio_sendfile> alone, as C<aio_sendfile> only
276	provides the number of bytes written to C<$out_fh>. Only if the result	375	provides the number of bytes written to C<$out_fh>. Only if the result
277	value equals C<$length> one can assume that C<$length> bytes have been	376	value equals C<$length> one can assume that C<$length> bytes have been
278	read.	377	read.
		378
279		379
280	=item aio_readahead $fh,$offset,$length, $callback->($retval)	380	=item aio_readahead $fh,$offset,$length, $callback->($retval)
281		381
282	C<aio_readahead> populates the page cache with data from a file so that	382	C<aio_readahead> populates the page cache with data from a file so that
283	subsequent reads from that file will not block on disk I/O. The C<$offset>	383	subsequent reads from that file will not block on disk I/O. The C<$offset>
…		…
289	file. The current file offset of the file is left unchanged.	389	file. The current file offset of the file is left unchanged.
290		390
291	If that syscall doesn't exist (likely if your OS isn't Linux) it will be	391	If that syscall doesn't exist (likely if your OS isn't Linux) it will be
292	emulated by simply reading the data, which would have a similar effect.	392	emulated by simply reading the data, which would have a similar effect.
293		393
		394
294	=item aio_stat $fh_or_path, $callback->($status)	395	=item aio_stat $fh_or_path, $callback->($status)
295		396
296	=item aio_lstat $fh, $callback->($status)	397	=item aio_lstat $fh, $callback->($status)
297		398
298	Works like perl's C<stat> or C<lstat> in void context. The callback will	399	Works like perl's C<stat> or C<lstat> in void context. The callback will
…		…
311	aio_stat "/etc/passwd", sub {	412	aio_stat "/etc/passwd", sub {
312	$_[0] and die "stat failed: $!";	413	$_[0] and die "stat failed: $!";
313	print "size is ", -s _, "\n";	414	print "size is ", -s _, "\n";
314	};	415	};
315		416
		417
		418	=item aio_utime $fh_or_path, $atime, $mtime, $callback->($status)
		419
		420	Works like perl's C<utime> function (including the special case of $atime
		421	and $mtime being undef). Fractional times are supported if the underlying
		422	syscalls support them.
		423
		424	When called with a pathname, uses utimes(2) if available, otherwise
		425	utime(2). If called on a file descriptor, uses futimes(2) if available,
		426	otherwise returns ENOSYS, so this is not portable.
		427
		428	Examples:
		429
		430	# set atime and mtime to current time (basically touch(1)):
		431	aio_utime "path", undef, undef;
		432	# set atime to current time and mtime to beginning of the epoch:
		433	aio_utime "path", time, undef; # undef==0
		434
		435
		436	=item aio_chown $fh_or_path, $uid, $gid, $callback->($status)
		437
		438	Works like perl's C<chown> function, except that C<undef> for either $uid
		439	or $gid is being interpreted as "do not change" (but -1 can also be used).
		440
		441	Examples:
		442
		443	# same as "chown root path" in the shell:
		444	aio_chown "path", 0, -1;
		445	# same as above:
		446	aio_chown "path", 0, undef;
		447
		448
		449	=item aio_truncate $fh_or_path, $offset, $callback->($status)
		450
		451	Works like truncate(2) or ftruncate(2).
		452
		453
		454	=item aio_chmod $fh_or_path, $mode, $callback->($status)
		455
		456	Works like perl's C<chmod> function.
		457
		458
316	=item aio_unlink $pathname, $callback->($status)	459	=item aio_unlink $pathname, $callback->($status)
317		460
318	Asynchronously unlink (delete) a file and call the callback with the	461	Asynchronously unlink (delete) a file and call the callback with the
319	result code.	462	result code.
320		463
		464
		465	=item aio_mknod $path, $mode, $dev, $callback->($status)
		466
		467	[EXPERIMENTAL]
		468
		469	Asynchronously create a device node (or fifo). See mknod(2).
		470
		471	The only (POSIX-) portable way of calling this function is:
		472
		473	aio_mknod $path, IO::AIO::S_IFIFO \| $mode, 0, sub { ...
		474
		475
321	=item aio_link $srcpath, $dstpath, $callback->($status)	476	=item aio_link $srcpath, $dstpath, $callback->($status)
322		477
323	Asynchronously create a new link to the existing object at C<$srcpath> at	478	Asynchronously create a new link to the existing object at C<$srcpath> at
324	the path C<$dstpath> and call the callback with the result code.	479	the path C<$dstpath> and call the callback with the result code.
325		480
		481
326	=item aio_symlink $srcpath, $dstpath, $callback->($status)	482	=item aio_symlink $srcpath, $dstpath, $callback->($status)
327		483
328	Asynchronously create a new symbolic link to the existing object at C<$srcpath> at	484	Asynchronously create a new symbolic link to the existing object at C<$srcpath> at
329	the path C<$dstpath> and call the callback with the result code.	485	the path C<$dstpath> and call the callback with the result code.
330		486
		487
		488	=item aio_readlink $path, $callback->($link)
		489
		490	Asynchronously read the symlink specified by C<$path> and pass it to
		491	the callback. If an error occurs, nothing or undef gets passed to the
		492	callback.
		493
		494
331	=item aio_rename $srcpath, $dstpath, $callback->($status)	495	=item aio_rename $srcpath, $dstpath, $callback->($status)
332		496
333	Asynchronously rename the object at C<$srcpath> to C<$dstpath>, just as	497	Asynchronously rename the object at C<$srcpath> to C<$dstpath>, just as
334	rename(2) and call the callback with the result code.	498	rename(2) and call the callback with the result code.
335		499
		500
		501	=item aio_mkdir $pathname, $mode, $callback->($status)
		502
		503	Asynchronously mkdir (create) a directory and call the callback with
		504	the result code. C<$mode> will be modified by the umask at the time the
		505	request is executed, so do not change your umask.
		506
		507
336	=item aio_rmdir $pathname, $callback->($status)	508	=item aio_rmdir $pathname, $callback->($status)
337		509
338	Asynchronously rmdir (delete) a directory and call the callback with the	510	Asynchronously rmdir (delete) a directory and call the callback with the
339	result code.	511	result code.
		512
340		513
341	=item aio_readdir $pathname, $callback->($entries)	514	=item aio_readdir $pathname, $callback->($entries)
342		515
343	Unlike the POSIX call of the same name, C<aio_readdir> reads an entire	516	Unlike the POSIX call of the same name, C<aio_readdir> reads an entire
344	directory (i.e. opendir + readdir + closedir). The entries will not be	517	directory (i.e. opendir + readdir + closedir). The entries will not be
345	sorted, and will B<NOT> include the C<.> and C<..> entries.	518	sorted, and will B<NOT> include the C<.> and C<..> entries.
346		519
347	The callback a single argument which is either C<undef> or an array-ref	520	The callback a single argument which is either C<undef> or an array-ref
348	with the filenames.	521	with the filenames.
349		522
		523
		524	=item aio_load $path, $data, $callback->($status)
		525
		526	This is a composite request that tries to fully load the given file into
		527	memory. Status is the same as with aio_read.
		528
		529	=cut
		530
		531	sub aio_load($$;$) {
		532	aio_block {
		533	my ($path, undef, $cb) = @_;
		534	my $data = \$_[1];
		535
		536	my $pri = aioreq_pri;
		537	my $grp = aio_group $cb;
		538
		539	aioreq_pri $pri;
		540	add $grp aio_open $path, O_RDONLY, 0, sub {
		541	my $fh = shift
		542	or return $grp->result (-1);
		543
		544	aioreq_pri $pri;
		545	add $grp aio_read $fh, 0, (-s $fh), $$data, 0, sub {
		546	$grp->result ($_[0]);
		547	};
		548	};
		549
		550	$grp
		551	}
		552	}
		553
		554	=item aio_copy $srcpath, $dstpath, $callback->($status)
		555
		556	Try to copy the I<file> (directories not supported as either source or
		557	destination) from C<$srcpath> to C<$dstpath> and call the callback with
		558	the C<0> (error) or C<-1> ok.
		559
		560	This is a composite request that it creates the destination file with
		561	mode 0200 and copies the contents of the source file into it using
		562	C<aio_sendfile>, followed by restoring atime, mtime, access mode and
		563	uid/gid, in that order.
		564
		565	If an error occurs, the partial destination file will be unlinked, if
		566	possible, except when setting atime, mtime, access mode and uid/gid, where
		567	errors are being ignored.
		568
		569	=cut
		570
		571	sub aio_copy($$;$) {
		572	aio_block {
		573	my ($src, $dst, $cb) = @_;
		574
		575	my $pri = aioreq_pri;
		576	my $grp = aio_group $cb;
		577
		578	aioreq_pri $pri;
		579	add $grp aio_open $src, O_RDONLY, 0, sub {
		580	if (my $src_fh = $_[0]) {
		581	my @stat = stat $src_fh;
		582
		583	aioreq_pri $pri;
		584	add $grp aio_open $dst, O_CREAT \| O_WRONLY \| O_TRUNC, 0200, sub {
		585	if (my $dst_fh = $_[0]) {
		586	aioreq_pri $pri;
		587	add $grp aio_sendfile $dst_fh, $src_fh, 0, $stat[7], sub {
		588	if ($_[0] == $stat[7]) {
		589	$grp->result (0);
		590	close $src_fh;
		591
		592	# those should not normally block. should. should.
		593	utime $stat[8], $stat[9], $dst;
		594	chmod $stat[2] & 07777, $dst_fh;
		595	chown $stat[4], $stat[5], $dst_fh;
		596	close $dst_fh;
		597	} else {
		598	$grp->result (-1);
		599	close $src_fh;
		600	close $dst_fh;
		601
		602	aioreq $pri;
		603	add $grp aio_unlink $dst;
		604	}
		605	};
		606	} else {
		607	$grp->result (-1);
		608	}
		609	},
		610
		611	} else {
		612	$grp->result (-1);
		613	}
		614	};
		615
		616	$grp
		617	}
		618	}
		619
		620	=item aio_move $srcpath, $dstpath, $callback->($status)
		621
		622	Try to move the I<file> (directories not supported as either source or
		623	destination) from C<$srcpath> to C<$dstpath> and call the callback with
		624	the C<0> (error) or C<-1> ok.
		625
		626	This is a composite request that tries to rename(2) the file first. If
		627	rename files with C<EXDEV>, it copies the file with C<aio_copy> and, if
		628	that is successful, unlinking the C<$srcpath>.
		629
		630	=cut
		631
		632	sub aio_move($$;$) {
		633	aio_block {
		634	my ($src, $dst, $cb) = @_;
		635
		636	my $pri = aioreq_pri;
		637	my $grp = aio_group $cb;
		638
		639	aioreq_pri $pri;
		640	add $grp aio_rename $src, $dst, sub {
		641	if ($_[0] && $! == EXDEV) {
		642	aioreq_pri $pri;
		643	add $grp aio_copy $src, $dst, sub {
		644	$grp->result ($_[0]);
		645
		646	if (!$_[0]) {
		647	aioreq_pri $pri;
		648	add $grp aio_unlink $src;
		649	}
		650	};
		651	} else {
		652	$grp->result ($_[0]);
		653	}
		654	};
		655
		656	$grp
		657	}
		658	}
		659
350	=item aio_scandir $path, $maxreq, $callback->($dirs, $nondirs)	660	=item aio_scandir $path, $maxreq, $callback->($dirs, $nondirs)
351		661
352	[EXPERIMENTAL due to internal aio_group use]
353
354	Scans a directory (similar to C<aio_readdir>) but additionally tries to	662	Scans a directory (similar to C<aio_readdir>) but additionally tries to
355	separate the entries of directory C<$path> into two sets of names, ones	663	efficiently separate the entries of directory C<$path> into two sets of
356	you can recurse into (directories or links to them), and ones you cannot	664	names, directories you can recurse into (directories), and ones you cannot
357	recurse into (everything else).	665	recurse into (everything else, including symlinks to directories).
358		666
359	C<aio_scandir> is a composite request that creates of many sub requests_	667	C<aio_scandir> is a composite request that creates of many sub requests_
360	C<$maxreq> specifies the maximum number of outstanding aio requests that	668	C<$maxreq> specifies the maximum number of outstanding aio requests that
361	this function generates. If it is C<< <= 0 >>, then a suitable default	669	this function generates. If it is C<< <= 0 >>, then a suitable default
362	will be chosen (currently 6).	670	will be chosen (currently 4).
363		671
364	On error, the callback is called without arguments, otherwise it receives	672	On error, the callback is called without arguments, otherwise it receives
365	two array-refs with path-relative entry names.	673	two array-refs with path-relative entry names.
366		674
367	Example:	675	Example:
…		…
401	as those tend to return 0 or 1 as link counts, which disables the	709	as those tend to return 0 or 1 as link counts, which disables the
402	directory counting heuristic.	710	directory counting heuristic.
403		711
404	=cut	712	=cut
405		713
406	sub aio_scandir($$$) {	714	sub aio_scandir($$;$) {
		715	aio_block {
407	my ($path, $maxreq, $cb) = @_;	716	my ($path, $maxreq, $cb) = @_;
408		717
		718	my $pri = aioreq_pri;
		719
409	my $grp = aio_group $cb;	720	my $grp = aio_group $cb;
410		721
411	$maxreq = 6 if $maxreq <= 0;	722	$maxreq = 4 if $maxreq <= 0;
412		723
413	# stat once	724	# stat once
		725	aioreq_pri $pri;
414	add $grp aio_stat $path, sub {	726	add $grp aio_stat $path, sub {
415	return $grp->result () if $_[0];	727	return $grp->result () if $_[0];
416	my $now = time;	728	my $now = time;
417	my $hash1 = join ":", (stat _)[0,1,3,7,9];	729	my $hash1 = join ":", (stat _)[0,1,3,7,9];
418		730
419	# read the directory entries	731	# read the directory entries
		732	aioreq_pri $pri;
420	add $grp aio_readdir $path, sub {	733	add $grp aio_readdir $path, sub {
421	my $entries = shift	734	my $entries = shift
422	or return $grp->result ();	735	or return $grp->result ();
423		736
424	# stat the dir another time	737	# stat the dir another time
		738	aioreq_pri $pri;
425	add $grp aio_stat $path, sub {	739	add $grp aio_stat $path, sub {
426	my $hash2 = join ":", (stat _)[0,1,3,7,9];	740	my $hash2 = join ":", (stat _)[0,1,3,7,9];
427		741
428	my $ndirs;	742	my $ndirs;
429		743
430	# take the slow route if anything looks fishy	744	# take the slow route if anything looks fishy
431	if ($hash1 ne $hash2 or (stat _)[9] == $now) {	745	if ($hash1 ne $hash2 or (stat _)[9] == $now) {
432	$ndirs = -1;	746	$ndirs = -1;
433	} else {	747	} else {
434	# if nlink == 2, we are finished	748	# if nlink == 2, we are finished
435	# on non-posix-fs's, we rely on nlink < 2	749	# on non-posix-fs's, we rely on nlink < 2
436	$ndirs = (stat _)[3] - 2	750	$ndirs = (stat _)[3] - 2
437	or return $grp->result ([], $entries);	751	or return $grp->result ([], $entries);
438	}	752	}
439		753
440	# sort into likely dirs and likely nondirs	754	# sort into likely dirs and likely nondirs
441	# dirs == files without ".", short entries first	755	# dirs == files without ".", short entries first
442	$entries = [map $_->[0],	756	$entries = [map $_->[0],
443	sort { $b->[1] cmp $a->[1] }	757	sort { $b->[1] cmp $a->[1] }
444	map [$_, sprintf "%s%04d", (/.\./ ? "1" : "0"), length],	758	map [$_, sprintf "%s%04d", (/.\./ ? "1" : "0"), length],
445	@$entries];	759	@$entries];
446		760
447	my (@dirs, @nondirs);	761	my (@dirs, @nondirs);
448		762
449	my ($statcb, $schedcb);
450	my $nreq = 0;
451
452	my $statgrp = add $grp aio_group;	763	my $statgrp = add $grp aio_group sub {
453
454	$schedcb = sub {
455	if (@$entries) {
456	if ($nreq < $maxreq) {
457	my $ent = pop @$entries;
458	$nreq++;
459	add $statgrp aio_stat "$path/$ent/.", sub { $statcb->($_[0], $ent) };
460	}
461	} elsif (!$nreq) {
462	# finished
463	$statgrp->cancel;
464	undef $statcb;
465	undef $schedcb;
466	$grp->result (\@dirs, \@nondirs);	764	$grp->result (\@dirs, \@nondirs);
467	}	765	};
		766
		767	limit $statgrp $maxreq;
		768	feed $statgrp sub {
		769	return unless @$entries;
		770	my $entry = pop @$entries;
		771
		772	aioreq_pri $pri;
		773	add $statgrp aio_stat "$path/$entry/.", sub {
		774	if ($_[0] < 0) {
		775	push @nondirs, $entry;
		776	} else {
		777	# need to check for real directory
		778	aioreq_pri $pri;
		779	add $statgrp aio_lstat "$path/$entry", sub {
		780	if (-d _) {
		781	push @dirs, $entry;
		782
		783	unless (--$ndirs) {
		784	push @nondirs, @$entries;
		785	feed $statgrp;
		786	}
		787	} else {
		788	push @nondirs, $entry;
		789	}
		790	}
		791	}
		792	};
		793	};
468	};	794	};
469	$statcb = sub {
470	my ($status, $entry) = @_;
471
472	if ($status < 0) {
473	$nreq--;
474	push @nondirs, $entry;
475	&$schedcb;
476	} else {
477	# need to check for real directory
478	add $grp aio_lstat "$path/$entry", sub {
479	$nreq--;
480
481	if (-d _) {
482	push @dirs, $entry;
483
484	if (!--$ndirs) {
485	push @nondirs, @$entries;
486	$entries = [];
487	}
488	} else {
489	push @nondirs, $entry;
490	}
491
492	&$schedcb;
493	}
494	}
495	};
496
497	&$schedcb while @$entries && $nreq < $maxreq;
498	};	795	};
499	};	796	};
		797
		798	$grp
500	};	799	}
		800	}
501		801
		802	=item aio_rmtree $path, $callback->($status)
		803
		804	Delete a directory tree starting (and including) C<$path>, return the
		805	status of the final C<rmdir> only. This is a composite request that
		806	uses C<aio_scandir> to recurse into and rmdir directories, and unlink
		807	everything else.
		808
		809	=cut
		810
		811	sub aio_rmtree;
		812	sub aio_rmtree($;$) {
		813	aio_block {
		814	my ($path, $cb) = @_;
		815
		816	my $pri = aioreq_pri;
		817	my $grp = aio_group $cb;
		818
		819	aioreq_pri $pri;
		820	add $grp aio_scandir $path, 0, sub {
		821	my ($dirs, $nondirs) = @_;
		822
		823	my $dirgrp = aio_group sub {
		824	add $grp aio_rmdir $path, sub {
		825	$grp->result ($_[0]);
		826	};
		827	};
		828
		829	(aioreq_pri $pri), add $dirgrp aio_rmtree "$path/$_" for @$dirs;
		830	(aioreq_pri $pri), add $dirgrp aio_unlink "$path/$_" for @$nondirs;
		831
		832	add $grp $dirgrp;
		833	};
		834
502	$grp	835	$grp
		836	}
503	}	837	}
504		838
505	=item aio_fsync $fh, $callback->($status)	839	=item aio_fsync $fh, $callback->($status)
506		840
507	Asynchronously call fsync on the given filehandle and call the callback	841	Asynchronously call fsync on the given filehandle and call the callback
…		…
515	If this call isn't available because your OS lacks it or it couldn't be	849	If this call isn't available because your OS lacks it or it couldn't be
516	detected, it will be emulated by calling C<fsync> instead.	850	detected, it will be emulated by calling C<fsync> instead.
517		851
518	=item aio_group $callback->(...)	852	=item aio_group $callback->(...)
519		853
520	[EXPERIMENTAL]
521
522	This is a very special aio request: Instead of doing something, it is a	854	This is a very special aio request: Instead of doing something, it is a
523	container for other aio requests, which is useful if you want to bundle	855	container for other aio requests, which is useful if you want to bundle
524	many requests into a single, composite, request.	856	many requests into a single, composite, request with a definite callback
		857	and the ability to cancel the whole request with its subrequests.
525		858
526	Returns an object of class L<IO::AIO::GRP>. See its documentation below	859	Returns an object of class L<IO::AIO::GRP>. See its documentation below
527	for more info.	860	for more info.
528		861
529	Example:	862	Example:
…		…
548	phase and still requires a worker thread. Thus, the callback will not	881	phase and still requires a worker thread. Thus, the callback will not
549	be executed immediately but only after other requests in the queue have	882	be executed immediately but only after other requests in the queue have
550	entered their execution phase. This can be used to measure request	883	entered their execution phase. This can be used to measure request
551	latency.	884	latency.
552		885
553	=item IO::AIO::aio_sleep $fractional_seconds, $callback->() NOT EXPORTED	886	=item IO::AIO::aio_busy $fractional_seconds, $callback->() NOT EXPORTED
554		887
555	Mainly used for debugging and benchmarking, this aio request puts one of	888	Mainly used for debugging and benchmarking, this aio request puts one of
556	the request workers to sleep for the given time.	889	the request workers to sleep for the given time.
557		890
558	While it is theoretically handy to have simple I/O scheduling requests	891	While it is theoretically handy to have simple I/O scheduling requests
559	like sleep and file handle readable/writable, the overhead this creates	892	like sleep and file handle readable/writable, the overhead this creates is
560	is immense, so do not use this function except to put your application	893	immense (it blocks a thread for a long time) so do not use this function
561	under artificial I/O pressure.	894	except to put your application under artificial I/O pressure.
562		895
563	=back	896	=back
564		897
565	=head2 IO::AIO::REQ CLASS	898	=head2 IO::AIO::REQ CLASS
566		899
567	All non-aggregate C<aio_*> functions return an object of this class when	900	All non-aggregate C<aio_*> functions return an object of this class when
568	called in non-void context.	901	called in non-void context.
569
570	A request always moves through the following five states in its lifetime,
571	in order: B<ready> (request has been created, but has not been executed
572	yet), B<execute> (request is currently being executed), B<pending>
573	(request has been executed but callback has not been called yet),
574	B<result> (results are being processed synchronously, includes calling the
575	callback) and B<done> (request has reached the end of its lifetime and
576	holds no resources anymore).
577		902
578	=over 4	903	=over 4
579		904
580	=item cancel $req	905	=item cancel $req
581		906
…		…
635	=item * They can also can also be added to other IO::AIO::GRP objects.	960	=item * They can also can also be added to other IO::AIO::GRP objects.
636		961
637	=item * You must not add requests to a group from within the group callback (or	962	=item * You must not add requests to a group from within the group callback (or
638	any later time).	963	any later time).
639		964
640	=item * This does not harmonise well with C<max_outstanding>, so best do
641	not combine C<aio_group> with it. Groups and feeders are recommended for
642	this kind of concurrency-limiting.
643
644	=back	965	=back
645		966
646	Their lifetime, simplified, looks like this: when they are empty, they	967	Their lifetime, simplified, looks like this: when they are empty, they
647	will finish very quickly. If they contain only requests that are in the	968	will finish very quickly. If they contain only requests that are in the
648	C<done> state, they will also finish. Otherwise they will continue to	969	C<done> state, they will also finish. Otherwise they will continue to
…		…
663	be added, including other groups, as long as you do not create circular	984	be added, including other groups, as long as you do not create circular
664	dependencies.	985	dependencies.
665		986
666	Returns all its arguments.	987	Returns all its arguments.
667		988
		989	=item $grp->cancel_subs
		990
		991	Cancel all subrequests and clears any feeder, but not the group request
		992	itself. Useful when you queued a lot of events but got a result early.
		993
668	=item $grp->result (...)	994	=item $grp->result (...)
669		995
670	Set the result value(s) that will be passed to the group callback when all	996	Set the result value(s) that will be passed to the group callback when all
671	subrequests have finished. By default, no argument will be passed.	997	subrequests have finished and set thre groups errno to the current value
		998	of errno (just like calling C<errno> without an error number). By default,
		999	no argument will be passed and errno is zero.
		1000
		1001	=item $grp->errno ([$errno])
		1002
		1003	Sets the group errno value to C<$errno>, or the current value of errno
		1004	when the argument is missing.
		1005
		1006	Every aio request has an associated errno value that is restored when
		1007	the callback is invoked. This method lets you change this value from its
		1008	default (0).
		1009
		1010	Calling C<result> will also set errno, so make sure you either set C<$!>
		1011	before the call to C<result>, or call c<errno> after it.
672		1012
673	=item feed $grp $callback->($grp)	1013	=item feed $grp $callback->($grp)
674
675	[VERY EXPERIMENTAL]
676		1014
677	Sets a feeder/generator on this group: every group can have an attached	1015	Sets a feeder/generator on this group: every group can have an attached
678	generator that generates requests if idle. The idea behind this is that,	1016	generator that generates requests if idle. The idea behind this is that,
679	although you could just queue as many requests as you want in a group,	1017	although you could just queue as many requests as you want in a group,
680	this might starve other requests for a potentially long time. For	1018	this might starve other requests for a potentially long time. For
681	example, C<aio_scandir> might generate hundreds of thousands C<aio_stat>	1019	example, C<aio_scandir> might generate hundreds of thousands C<aio_stat>
682	requests, delaying any later requests for a long time.	1020	requests, delaying any later requests for a long time.
683		1021
684	To avoid this, and allow incremental generation of requests, you can	1022	To avoid this, and allow incremental generation of requests, you can
685	instead a group and set a feeder on it that generates those requests. The	1023	instead a group and set a feeder on it that generates those requests. The
686	feed callback will be called whenever there are few enough (see C<feed_limit>,	1024	feed callback will be called whenever there are few enough (see C<limit>,
687	below) requests active in the group itself and is expected to queue more	1025	below) requests active in the group itself and is expected to queue more
688	requests.	1026	requests.
689		1027
690	The feed can queue as many requests as it likes (i.e. C<add> does not	1028	The feed callback can queue as many requests as it likes (i.e. C<add> does
691	impose any limits).	1029	not impose any limits).
692		1030
693	If the feed does not queue more requests when called, it will be	1031	If the feed does not queue more requests when called, it will be
694	automatically removed from the group.	1032	automatically removed from the group.
695		1033
696	If the feed limit is C<0>, it will be set to C<2> automatically.	1034	If the feed limit is C<0>, it will be set to C<2> automatically.
…		…
698	Example:	1036	Example:
699		1037
700	# stat all files in @files, but only ever use four aio requests concurrently:	1038	# stat all files in @files, but only ever use four aio requests concurrently:
701		1039
702	my $grp = aio_group sub { print "finished\n" };	1040	my $grp = aio_group sub { print "finished\n" };
703	feed_limit $grp 4;	1041	limit $grp 4;
704	feed $grp sub {	1042	feed $grp sub {
705	my $file = pop @files	1043	my $file = pop @files
706	or return;	1044	or return;
707		1045
708	add $grp aio_stat $file, sub { ... };	1046	add $grp aio_stat $file, sub { ... };
709	};	1047	};
710		1048
711	=item feed_limit $grp $num	1049	=item limit $grp $num
712		1050
713	Sets the feeder limit for the group: The feeder will be called whenever	1051	Sets the feeder limit for the group: The feeder will be called whenever
714	the group contains less than this many requests.	1052	the group contains less than this many requests.
715		1053
716	Setting the limit to C<0> will pause the feeding process.	1054	Setting the limit to C<0> will pause the feeding process.
717		1055
718	=back	1056	=back
719		1057
720	=head2 SUPPORT FUNCTIONS	1058	=head2 SUPPORT FUNCTIONS
		1059
		1060	=head3 EVENT PROCESSING AND EVENT LOOP INTEGRATION
721		1061
722	=over 4	1062	=over 4
723		1063
724	=item $fileno = IO::AIO::poll_fileno	1064	=item $fileno = IO::AIO::poll_fileno
725		1065
…		…
730		1070
731	See C<poll_cb> for an example.	1071	See C<poll_cb> for an example.
732		1072
733	=item IO::AIO::poll_cb	1073	=item IO::AIO::poll_cb
734		1074
735	Process all outstanding events on the result pipe. You have to call this	1075	Process some outstanding events on the result pipe. You have to call this
736	regularly. Returns the number of events processed. Returns immediately	1076	regularly. Returns the number of events processed. Returns immediately
737	when no events are outstanding.	1077	when no events are outstanding. The amount of events processed depends on
		1078	the settings of C<IO::AIO::max_poll_req> and C<IO::AIO::max_poll_time>.
		1079
		1080	If not all requests were processed for whatever reason, the filehandle
		1081	will still be ready when C<poll_cb> returns.
738		1082
739	Example: Install an Event watcher that automatically calls	1083	Example: Install an Event watcher that automatically calls
740	IO::AIO::poll_cb with high priority:	1084	IO::AIO::poll_cb with high priority:
741		1085
742	Event->io (fd => IO::AIO::poll_fileno,	1086	Event->io (fd => IO::AIO::poll_fileno,
743	poll => 'r', async => 1,	1087	poll => 'r', async => 1,
744	cb => \&IO::AIO::poll_cb);	1088	cb => \&IO::AIO::poll_cb);
745		1089
		1090	=item IO::AIO::max_poll_reqs $nreqs
		1091
		1092	=item IO::AIO::max_poll_time $seconds
		1093
		1094	These set the maximum number of requests (default C<0>, meaning infinity)
		1095	that are being processed by C<IO::AIO::poll_cb> in one call, respectively
		1096	the maximum amount of time (default C<0>, meaning infinity) spent in
		1097	C<IO::AIO::poll_cb> to process requests (more correctly the mininum amount
		1098	of time C<poll_cb> is allowed to use).
		1099
		1100	Setting C<max_poll_time> to a non-zero value creates an overhead of one
		1101	syscall per request processed, which is not normally a problem unless your
		1102	callbacks are really really fast or your OS is really really slow (I am
		1103	not mentioning Solaris here). Using C<max_poll_reqs> incurs no overhead.
		1104
		1105	Setting these is useful if you want to ensure some level of
		1106	interactiveness when perl is not fast enough to process all requests in
		1107	time.
		1108
		1109	For interactive programs, values such as C<0.01> to C<0.1> should be fine.
		1110
		1111	Example: Install an Event watcher that automatically calls
		1112	IO::AIO::poll_cb with low priority, to ensure that other parts of the
		1113	program get the CPU sometimes even under high AIO load.
		1114
		1115	# try not to spend much more than 0.1s in poll_cb
		1116	IO::AIO::max_poll_time 0.1;
		1117
		1118	# use a low priority so other tasks have priority
		1119	Event->io (fd => IO::AIO::poll_fileno,
		1120	poll => 'r', nice => 1,
		1121	cb => &IO::AIO::poll_cb);
		1122
746	=item IO::AIO::poll_wait	1123	=item IO::AIO::poll_wait
747		1124
		1125	If there are any outstanding requests and none of them in the result
748	Wait till the result filehandle becomes ready for reading (simply does a	1126	phase, wait till the result filehandle becomes ready for reading (simply
749	C<select> on the filehandle. This is useful if you want to synchronously wait	1127	does a C<select> on the filehandle. This is useful if you want to
750	for some requests to finish).	1128	synchronously wait for some requests to finish).
751		1129
752	See C<nreqs> for an example.	1130	See C<nreqs> for an example.
753		1131
		1132	=item IO::AIO::poll
		1133
		1134	Waits until some requests have been handled.
		1135
		1136	Returns the number of requests processed, but is otherwise strictly
		1137	equivalent to:
		1138
		1139	IO::AIO::poll_wait, IO::AIO::poll_cb
		1140
754	=item IO::AIO::nreqs	1141	=item IO::AIO::flush
755		1142
756	Returns the number of requests currently outstanding (i.e. for which their	1143	Wait till all outstanding AIO requests have been handled.
757	callback has not been invoked yet).
758		1144
759	Example: wait till there are no outstanding requests anymore:	1145	Strictly equivalent to:
760		1146
761	IO::AIO::poll_wait, IO::AIO::poll_cb	1147	IO::AIO::poll_wait, IO::AIO::poll_cb
762	while IO::AIO::nreqs;	1148	while IO::AIO::nreqs;
763		1149
764	=item IO::AIO::flush	1150	=back
765		1151
766	Wait till all outstanding AIO requests have been handled.	1152	=head3 CONTROLLING THE NUMBER OF THREADS
767		1153
768	Strictly equivalent to:	1154	=over
769
770	IO::AIO::poll_wait, IO::AIO::poll_cb
771	while IO::AIO::nreqs;
772
773	=item IO::AIO::poll
774
775	Waits until some requests have been handled.
776
777	Strictly equivalent to:
778
779	IO::AIO::poll_wait, IO::AIO::poll_cb
780	if IO::AIO::nreqs;
781		1155
782	=item IO::AIO::min_parallel $nthreads	1156	=item IO::AIO::min_parallel $nthreads
783		1157
784	Set the minimum number of AIO threads to C<$nthreads>. The current	1158	Set the minimum number of AIO threads to C<$nthreads>. The current
785	default is C<8>, which means eight asynchronous operations can execute	1159	default is C<8>, which means eight asynchronous operations can execute
786	concurrently at any one time (the number of outstanding requests,	1160	concurrently at any one time (the number of outstanding requests,
787	however, is unlimited).	1161	however, is unlimited).
788		1162
789	IO::AIO starts threads only on demand, when an AIO request is queued and	1163	IO::AIO starts threads only on demand, when an AIO request is queued and
790	no free thread exists.	1164	no free thread exists. Please note that queueing up a hundred requests can
		1165	create demand for a hundred threads, even if it turns out that everything
		1166	is in the cache and could have been processed faster by a single thread.
791		1167
792	It is recommended to keep the number of threads relatively low, as some	1168	It is recommended to keep the number of threads relatively low, as some
793	Linux kernel versions will scale negatively with the number of threads	1169	Linux kernel versions will scale negatively with the number of threads
794	(higher parallelity => MUCH higher latency). With current Linux 2.6	1170	(higher parallelity => MUCH higher latency). With current Linux 2.6
795	versions, 4-32 threads should be fine.	1171	versions, 4-32 threads should be fine.
…		…
809	This module automatically runs C<max_parallel 0> at program end, to ensure	1185	This module automatically runs C<max_parallel 0> at program end, to ensure
810	that all threads are killed and that there are no outstanding requests.	1186	that all threads are killed and that there are no outstanding requests.
811		1187
812	Under normal circumstances you don't need to call this function.	1188	Under normal circumstances you don't need to call this function.
813		1189
		1190	=item IO::AIO::max_idle $nthreads
		1191
		1192	Limit the number of threads (default: 4) that are allowed to idle (i.e.,
		1193	threads that did not get a request to process within 10 seconds). That
		1194	means if a thread becomes idle while C<$nthreads> other threads are also
		1195	idle, it will free its resources and exit.
		1196
		1197	This is useful when you allow a large number of threads (e.g. 100 or 1000)
		1198	to allow for extremely high load situations, but want to free resources
		1199	under normal circumstances (1000 threads can easily consume 30MB of RAM).
		1200
		1201	The default is probably ok in most situations, especially if thread
		1202	creation is fast. If thread creation is very slow on your system you might
		1203	want to use larger values.
		1204
814	=item $oldnreqs = IO::AIO::max_outstanding $nreqs	1205	=item $oldmaxreqs = IO::AIO::max_outstanding $maxreqs
815		1206
816	[DEPRECATED]	1207	This is a very bad function to use in interactive programs because it
		1208	blocks, and a bad way to reduce concurrency because it is inexact: Better
		1209	use an C<aio_group> together with a feed callback.
817		1210
818	Sets the maximum number of outstanding requests to C<$nreqs>. If you	1211	Sets the maximum number of outstanding requests to C<$nreqs>. If you
819	try to queue up more than this number of requests, the caller will block until	1212	do queue up more than this number of requests, the next call to the
820	some requests have been handled.	1213	C<poll_cb> (and C<poll_some> and other functions calling C<poll_cb>)
		1214	function will block until the limit is no longer exceeded.
821		1215
822	The default is very large, so normally there is no practical limit. If you	1216	The default value is very large, so there is no practical limit on the
823	queue up many requests in a loop it often improves speed if you set	1217	number of outstanding requests.
824	this to a relatively low number, such as C<100>.
825		1218
826	This function does not work well together with C<aio_group>'s, and their	1219	You can still queue as many requests as you want. Therefore,
827	feeder interface is better suited to limiting concurrency, so do not use	1220	C<max_oustsanding> is mainly useful in simple scripts (with low values) or
828	this function.	1221	as a stop gap to shield against fatal memory overflow (with large values).
829		1222
830	Under normal circumstances you don't need to call this function.	1223	=back
		1224
		1225	=head3 STATISTICAL INFORMATION
		1226
		1227	=over
		1228
		1229	=item IO::AIO::nreqs
		1230
		1231	Returns the number of requests currently in the ready, execute or pending
		1232	states (i.e. for which their callback has not been invoked yet).
		1233
		1234	Example: wait till there are no outstanding requests anymore:
		1235
		1236	IO::AIO::poll_wait, IO::AIO::poll_cb
		1237	while IO::AIO::nreqs;
		1238
		1239	=item IO::AIO::nready
		1240
		1241	Returns the number of requests currently in the ready state (not yet
		1242	executed).
		1243
		1244	=item IO::AIO::npending
		1245
		1246	Returns the number of requests currently in the pending state (executed,
		1247	but not yet processed by poll_cb).
831		1248
832	=back	1249	=back
833		1250
834	=cut	1251	=cut
835		1252
…		…
849	*$sym	1266	*$sym
850	}	1267	}
851		1268
852	min_parallel 8;	1269	min_parallel 8;
853		1270
854	END {	1271	END { flush }
855	max_parallel 0;
856	}
857		1272
858	1;	1273	1;
859		1274
860	=head2 FORK BEHAVIOUR	1275	=head2 FORK BEHAVIOUR
861		1276
862	This module should do "the right thing" when the process using it forks:	1277	This module should do "the right thing" when the process using it forks:
863		1278
864	Before the fork, IO::AIO enters a quiescent state where no requests	1279	Before the fork, IO::AIO enters a quiescent state where no requests
865	can be added in other threads and no results will be processed. After	1280	can be added in other threads and no results will be processed. After
866	the fork the parent simply leaves the quiescent state and continues	1281	the fork the parent simply leaves the quiescent state and continues
867	request/result processing, while the child clears the request/result	1282	request/result processing, while the child frees the request/result queue
868	queue (so the requests started before the fork will only be handled in	1283	(so that the requests started before the fork will only be handled in the
869	the parent). Threads will be started on demand until the limit ste in the	1284	parent). Threads will be started on demand until the limit set in the
870	parent process has been reached again.	1285	parent process has been reached again.
871		1286
872	In short: the parent will, after a short pause, continue as if fork had	1287	In short: the parent will, after a short pause, continue as if fork had
873	not been called, while the child will act as if IO::AIO has not been used	1288	not been called, while the child will act as if IO::AIO has not been used
874	yet.	1289	yet.
875		1290
876	=head2 MEMORY USAGE	1291	=head2 MEMORY USAGE
877		1292
		1293	Per-request usage:
		1294
878	Each aio request uses - depending on your architecture - around 128 bytes	1295	Each aio request uses - depending on your architecture - around 100-200
879	of memory. In addition, stat requests need a stat buffer (possibly a few	1296	bytes of memory. In addition, stat requests need a stat buffer (possibly
880	hundred bytes). Perl scalars and other data passed into aio requests will	1297	a few hundred bytes), readdir requires a result buffer and so on. Perl
881	also be locked.	1298	scalars and other data passed into aio requests will also be locked and
		1299	will consume memory till the request has entered the done state.
882		1300
883	This is now awfully much, so queuing lots of requests is not usually a	1301	This is not awfully much, so queuing lots of requests is not usually a
884	problem.	1302	problem.
885		1303
886	Each thread needs a stack area which is usually around 16k, sometimes much	1304	Per-thread usage:
887	larger, depending on the OS.	1305
		1306	In the execution phase, some aio requests require more memory for
		1307	temporary buffers, and each thread requires a stack and other data
		1308	structures (usually around 16k-128k, depending on the OS).
		1309
		1310	=head1 KNOWN BUGS
		1311
		1312	Known bugs will be fixed in the next release.
888		1313
889	=head1 SEE ALSO	1314	=head1 SEE ALSO
890		1315
891	L<Coro>, L<Linux::AIO> (obsolete).	1316	L<Coro::AIO>.
892		1317
893	=head1 AUTHOR	1318	=head1 AUTHOR
894		1319
895	Marc Lehmann <schmorp@schmorp.de>	1320	Marc Lehmann <schmorp@schmorp.de>
896	http://home.schmorp.de/	1321	http://home.schmorp.de/

Diff Legend

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

Comparing IO-AIO/AIO.pm (file contents): Revision 1.67 by root, Tue Oct 24 02:25:16 2006 UTC vs. Revision 1.113 by root, Thu Sep 20 14:06:21 2007 UTC

Diff Legend

Comparing IO-AIO/AIO.pm (file contents):
Revision 1.67 by root, Tue Oct 24 02:25:16 2006 UTC vs.
Revision 1.113 by root, Thu Sep 20 14:06:21 2007 UTC