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

Comparing IO-AIO/AIO.pm (file contents):
Revision 1.68 by root, Tue Oct 24 03:17:39 2006 UTC vs.
Revision 1.107 by root, Fri Jun 1 05:52:59 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
…		…
51	=head1 DESCRIPTION	52	=head1 DESCRIPTION
52		53
53	This module implements asynchronous I/O using whatever means your	54	This module implements asynchronous I/O using whatever means your
54	operating system supports.	55	operating system supports.
55		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 example
		68	sockets), using these functions on file descriptors that support
		69	nonblocking operation (again, sockets, pipes etc.) is very inefficient or
		70	might not work (aio_read fails on sockets/pipes/fifos). Use an event loop
		71	for that (such as the L<Event\|Event> module): IO::AIO will naturally fit
		72	into such an event loop itself.
		73
56	Currently, a number of threads are started that execute your read/writes	74	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	75	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	76	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	77	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	78	functions available on many operating systems. However, they are often
61	(Linux doesn't allow them on normal files currently, for example),	79	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	80	files currently, for example), and they would only support aio_read and
63	functionality would have to be implemented using threads anyway.	81	aio_write, so the remaining functionality would have to be implemented
		82	using threads anyway.
64		83
65	Although the module will work with in the presence of other threads,	84	Although the module will work with in the presence of other (Perl-)
66	it is currently not reentrant in any way, so use appropriate locking	85	threads, it is currently not reentrant in any way, so use appropriate
67	yourself, always call C<poll_cb> from within the same thread, or never	86	locking yourself, always call C<poll_cb> from within the same thread, or
68	call C<poll_cb> (or other C<aio_> functions) recursively.	87	never call C<poll_cb> (or other C<aio_> functions) recursively.
		88
		89	=head2 EXAMPLE
		90
		91	This is a simple example that uses the Event module and loads
		92	F</etc/passwd> asynchronously:
		93
		94	use Fcntl;
		95	use Event;
		96	use IO::AIO;
		97
		98	# register the IO::AIO callback with Event
		99	Event->io (fd => IO::AIO::poll_fileno,
		100	poll => 'r',
		101	cb => \&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	Event::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	Event::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
69		184
70	=cut	185	=cut
71		186
72	package IO::AIO;	187	package IO::AIO;
73		188
…		…
75	use strict 'vars';	190	use strict 'vars';
76		191
77	use base 'Exporter';	192	use base 'Exporter';
78		193
79	BEGIN {	194	BEGIN {
80	our $VERSION = '2.0';	195	our $VERSION = '2.4';
81		196
82	our @AIO_REQ = qw(aio_sendfile aio_read aio_write aio_open aio_close aio_stat	197	our @AIO_REQ = qw(aio_sendfile aio_read aio_write aio_open aio_close aio_stat
83	aio_lstat aio_unlink aio_rmdir aio_readdir aio_scandir aio_symlink	198	aio_lstat aio_unlink aio_rmdir aio_readdir aio_scandir aio_symlink
84	aio_fsync aio_fdatasync aio_readahead aio_rename aio_link aio_move	199	aio_readlink aio_fsync aio_fdatasync aio_readahead aio_rename aio_link
85	aio_group aio_nop);	200	aio_move aio_copy aio_group aio_nop aio_mknod aio_load aio_rmtree aio_mkdir
		201	aio_chown aio_chmod aio_utime);
86	our @EXPORT = (@AIO_REQ, qw(aioreq_pri));	202	our @EXPORT = (@AIO_REQ, qw(aioreq_pri aioreq_nice aio_block));
87	our @EXPORT_OK = qw(poll_fileno poll_cb poll_wait flush	203	our @EXPORT_OK = qw(poll_fileno poll_cb poll_wait flush
88	min_parallel max_parallel max_outstanding nreqs);	204	min_parallel max_parallel max_idle
		205	nreqs nready npending nthreads
		206	max_poll_time max_poll_reqs);
89		207
90	@IO::AIO::GRP::ISA = 'IO::AIO::REQ';	208	@IO::AIO::GRP::ISA = 'IO::AIO::REQ';
91		209
92	require XSLoader;	210	require XSLoader;
93	XSLoader::load ("IO::AIO", $VERSION);	211	XSLoader::load ("IO::AIO", $VERSION);
94	}	212	}
95		213
96	=head1 FUNCTIONS	214	=head1 FUNCTIONS
97		215
98	=head2 AIO FUNCTIONS	216	=head2 AIO REQUEST FUNCTIONS
99		217
100	All the C<aio_*> calls are more or less thin wrappers around the syscall	218	All the C<aio_*> calls are more or less thin wrappers around the syscall
101	with the same name (sans C<aio_>). The arguments are similar or identical,	219	with the same name (sans C<aio_>). The arguments are similar or identical,
102	and they all accept an additional (and optional) C<$callback> argument	220	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	221	which must be a code reference. This code reference will get called with
…		…
106	syscall has been executed asynchronously.	224	syscall has been executed asynchronously.
107		225
108	All functions expecting a filehandle keep a copy of the filehandle	226	All functions expecting a filehandle keep a copy of the filehandle
109	internally until the request has finished.	227	internally until the request has finished.
110		228
111	All requests return objects of type L<IO::AIO::REQ> that allow further	229	All functions return request objects of type L<IO::AIO::REQ> that allow
112	manipulation of those requests while they are in-flight.	230	further manipulation of those requests while they are in-flight.
113		231
114	The pathnames you pass to these routines I<must> be absolute and	232	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	233	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	234	request is being executed, the current working directory could have
117	changed. Alternatively, you can make sure that you never change the	235	changed. Alternatively, you can make sure that you never change the
118	current working directory.	236	current working directory anywhere in the program and then use relative
		237	paths.
119		238
120	To encode pathnames to byte form, either make sure you either: a)	239	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	240	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	241	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	242	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)	243	environment, d) use Glib::filename_from_unicode on unicode filenames or e)
125	use something else.	244	use something else to ensure your scalar has the correct contents.
		245
		246	This works, btw. independent of the internal UTF-8 bit, which IO::AIO
		247	handles correctly wether it is set or not.
126		248
127	=over 4	249	=over 4
128		250
129	=item aioreq_pri $pri	251	=item $prev_pri = aioreq_pri [$pri]
130		252
131	Sets the priority for the next aio request. The default priority	253	Returns the priority value that would be used for the next request and, if
		254	C<$pri> is given, sets the priority for the next aio request.
		255
132	is C<0>, the minimum and maximum priorities are C<-4> and C<4>,	256	The default priority is C<0>, the minimum and maximum priorities are C<-4>
133	respectively. Requests with higher priority will be serviced first.	257	and C<4>, respectively. Requests with higher priority will be serviced
		258	first.
134		259
135	The priority will be reset to C<0> after each call to one of the C<aio_>	260	The priority will be reset to C<0> after each call to one of the C<aio_*>
136	functions.	261	functions.
		262
		263	Example: open a file with low priority, then read something from it with
		264	higher priority so the read request is serviced before other low priority
		265	open requests (potentially spamming the cache):
		266
		267	aioreq_pri -3;
		268	aio_open ..., sub {
		269	return unless $_[0];
		270
		271	aioreq_pri -2;
		272	aio_read $_[0], ..., sub {
		273	...
		274	};
		275	};
		276
		277
		278	=item aioreq_nice $pri_adjust
		279
		280	Similar to C<aioreq_pri>, but subtracts the given value from the current
		281	priority, so the effect is cumulative.
		282
137		283
138	=item aio_open $pathname, $flags, $mode, $callback->($fh)	284	=item aio_open $pathname, $flags, $mode, $callback->($fh)
139		285
140	Asynchronously open or create a file and call the callback with a newly	286	Asynchronously open or create a file and call the callback with a newly
141	created filehandle for the file.	287	created filehandle for the file.
…		…
147	list. They are the same as used by C<sysopen>.	293	list. They are the same as used by C<sysopen>.
148		294
149	Likewise, C<$mode> specifies the mode of the newly created file, if it	295	Likewise, C<$mode> specifies the mode of the newly created file, if it
150	didn't exist and C<O_CREAT> has been given, just like perl's C<sysopen>,	296	didn't exist and C<O_CREAT> has been given, just like perl's C<sysopen>,
151	except that it is mandatory (i.e. use C<0> if you don't create new files,	297	except that it is mandatory (i.e. use C<0> if you don't create new files,
152	and C<0666> or C<0777> if you do).	298	and C<0666> or C<0777> if you do). Note that the C<$mode> will be modified
		299	by the umask in effect then the request is being executed, so better never
		300	change the umask.
153		301
154	Example:	302	Example:
155		303
156	aio_open "/etc/passwd", O_RDONLY, 0, sub {	304	aio_open "/etc/passwd", O_RDONLY, 0, sub {
157	if ($_[0]) {	305	if ($_[0]) {
…		…
160	} else {	308	} else {
161	die "open failed: $!\n";	309	die "open failed: $!\n";
162	}	310	}
163	};	311	};
164		312
		313
165	=item aio_close $fh, $callback->($status)	314	=item aio_close $fh, $callback->($status)
166		315
167	Asynchronously close a file and call the callback with the result	316	Asynchronously close a file and call the callback with the result
168	code. I<WARNING:> although accepted, you should not pass in a perl	317	code. I<WARNING:> although accepted, you should not pass in a perl
169	filehandle here, as perl will likely close the file descriptor another	318	filehandle here, as perl will likely close the file descriptor another
…		…
171	C<close> or just let filehandles go out of scope.	320	C<close> or just let filehandles go out of scope.
172		321
173	This is supposed to be a bug in the API, so that might change. It's	322	This is supposed to be a bug in the API, so that might change. It's
174	therefore best to avoid this function.	323	therefore best to avoid this function.
175		324
		325
176	=item aio_read $fh,$offset,$length, $data,$dataoffset, $callback->($retval)	326	=item aio_read $fh,$offset,$length, $data,$dataoffset, $callback->($retval)
177		327
178	=item aio_write $fh,$offset,$length, $data,$dataoffset, $callback->($retval)	328	=item aio_write $fh,$offset,$length, $data,$dataoffset, $callback->($retval)
179		329
180	Reads or writes C<length> bytes from the specified C<fh> and C<offset>	330	Reads or writes C<length> bytes from the specified C<fh> and C<offset>
…		…
192	aio_read $fh, 7, 15, $buffer, 0, sub {	342	aio_read $fh, 7, 15, $buffer, 0, sub {
193	$_[0] > 0 or die "read error: $!";	343	$_[0] > 0 or die "read error: $!";
194	print "read $_[0] bytes: <$buffer>\n";	344	print "read $_[0] bytes: <$buffer>\n";
195	};	345	};
196		346
197	=item aio_move $srcpath, $dstpath, $callback->($status)
198
199	[EXPERIMENTAL due to internal aio_group use]
200
201	Try to move the I<file> (directories not supported as either source or
202	destination) from C<$srcpath> to C<$dstpath> and call the callback with
203	the C<0> (error) or C<-1> ok.
204
205	This is a composite request that tries to rename(2) the file first. If
206	rename files with C<EXDEV>, it creates the destination file with mode 0200
207	and copies the contents of the source file into it using C<aio_sendfile>,
208	followed by restoring atime, mtime, access mode and uid/gid, in that
209	order, and unlinking the C<$srcpath>.
210
211	If an error occurs, the partial destination file will be unlinked, if
212	possible, except when setting atime, mtime, access mode and uid/gid, where
213	errors are being ignored.
214
215	=cut
216
217	sub aio_move($$$) {
218	my ($src, $dst, $cb) = @_;
219
220	my $grp = aio_group $cb;
221
222	add $grp aio_rename $src, $dst, sub {
223	if ($_[0] && $! == EXDEV) {
224	add $grp aio_open $src, O_RDONLY, 0, sub {
225	if (my $src_fh = $_[0]) {
226	my @stat = stat $src_fh;
227
228	add $grp aio_open $dst, O_WRONLY, 0200, sub {
229	if (my $dst_fh = $_[0]) {
230	add $grp aio_sendfile $dst_fh, $src_fh, 0, $stat[7], sub {
231	close $src_fh;
232
233	if ($_[0] == $stat[7]) {
234	utime $stat[8], $stat[9], $dst;
235	chmod $stat[2] & 07777, $dst_fh;
236	chown $stat[4], $stat[5], $dst_fh;
237	close $dst_fh;
238
239	add $grp aio_unlink $src, sub {
240	$grp->result ($_[0]);
241	};
242	} else {
243	my $errno = $!;
244	add $grp aio_unlink $dst, sub {
245	$! = $errno;
246	$grp->result (-1);
247	};
248	}
249	};
250	} else {
251	$grp->result (-1);
252	}
253	},
254
255	} else {
256	$grp->result (-1);
257	}
258	};
259	} else {
260	$grp->result ($_[0]);
261	}
262	};
263
264	$grp
265	}
266		347
267	=item aio_sendfile $out_fh, $in_fh, $in_offset, $length, $callback->($retval)	348	=item aio_sendfile $out_fh, $in_fh, $in_offset, $length, $callback->($retval)
268		349
269	Tries to copy C<$length> bytes from C<$in_fh> to C<$out_fh>. It starts	350	Tries to copy C<$length> bytes from C<$in_fh> to C<$out_fh>. It starts
270	reading at byte offset C<$in_offset>, and starts writing at the current	351	reading at byte offset C<$in_offset>, and starts writing at the current
…		…
284	C<$in_fh> than are written, and there is no way to find out how many	365	C<$in_fh> than are written, and there is no way to find out how many
285	bytes have been read from C<aio_sendfile> alone, as C<aio_sendfile> only	366	bytes have been read from C<aio_sendfile> alone, as C<aio_sendfile> only
286	provides the number of bytes written to C<$out_fh>. Only if the result	367	provides the number of bytes written to C<$out_fh>. Only if the result
287	value equals C<$length> one can assume that C<$length> bytes have been	368	value equals C<$length> one can assume that C<$length> bytes have been
288	read.	369	read.
		370
289		371
290	=item aio_readahead $fh,$offset,$length, $callback->($retval)	372	=item aio_readahead $fh,$offset,$length, $callback->($retval)
291		373
292	C<aio_readahead> populates the page cache with data from a file so that	374	C<aio_readahead> populates the page cache with data from a file so that
293	subsequent reads from that file will not block on disk I/O. The C<$offset>	375	subsequent reads from that file will not block on disk I/O. The C<$offset>
…		…
299	file. The current file offset of the file is left unchanged.	381	file. The current file offset of the file is left unchanged.
300		382
301	If that syscall doesn't exist (likely if your OS isn't Linux) it will be	383	If that syscall doesn't exist (likely if your OS isn't Linux) it will be
302	emulated by simply reading the data, which would have a similar effect.	384	emulated by simply reading the data, which would have a similar effect.
303		385
		386
304	=item aio_stat $fh_or_path, $callback->($status)	387	=item aio_stat $fh_or_path, $callback->($status)
305		388
306	=item aio_lstat $fh, $callback->($status)	389	=item aio_lstat $fh, $callback->($status)
307		390
308	Works like perl's C<stat> or C<lstat> in void context. The callback will	391	Works like perl's C<stat> or C<lstat> in void context. The callback will
…		…
321	aio_stat "/etc/passwd", sub {	404	aio_stat "/etc/passwd", sub {
322	$_[0] and die "stat failed: $!";	405	$_[0] and die "stat failed: $!";
323	print "size is ", -s _, "\n";	406	print "size is ", -s _, "\n";
324	};	407	};
325		408
		409
		410	=item aio_utime $fh_or_path, $atime, $mtime, $callback->($status)
		411
		412	Works like perl's C<utime> function (including the special case of $atime
		413	and $mtime being undef). Fractional times are supported if the underlying
		414	syscalls support them.
		415
		416	When called with a pathname, uses utimes(2) if available, otherwise
		417	utime(2). If called on a file descriptor, uses futimes(2) if available,
		418	otherwise returns ENOSYS, so this is not portable.
		419
		420	Examples:
		421
		422	# set atime and mtime to current time (basically touch(1)):
		423	aio_utime "path", undef, undef;
		424	# set atime to current time and mtime to beginning of the epoch:
		425	aio_utime "path", time, undef; # undef==0
		426
		427
		428	=item aio_chown $fh_or_path, $uid, $gid, $callback->($status)
		429
		430	Works like perl's C<chown> function, except that C<undef> for either $uid
		431	or $gid is being interpreted as "do not change" (but -1 can also be used).
		432
		433	Examples:
		434
		435	# same as "chown root path" in the shell:
		436	aio_chown "path", 0, -1;
		437	# same as above:
		438	aio_chown "path", 0, undef;
		439
		440
		441	=item aio_chmod $fh_or_path, $mode, $callback->($status)
		442
		443	Works like perl's C<chmod> function.
		444
		445
326	=item aio_unlink $pathname, $callback->($status)	446	=item aio_unlink $pathname, $callback->($status)
327		447
328	Asynchronously unlink (delete) a file and call the callback with the	448	Asynchronously unlink (delete) a file and call the callback with the
329	result code.	449	result code.
330		450
		451
		452	=item aio_mknod $path, $mode, $dev, $callback->($status)
		453
		454	[EXPERIMENTAL]
		455
		456	Asynchronously create a device node (or fifo). See mknod(2).
		457
		458	The only (POSIX-) portable way of calling this function is:
		459
		460	aio_mknod $path, IO::AIO::S_IFIFO \| $mode, 0, sub { ...
		461
		462
331	=item aio_link $srcpath, $dstpath, $callback->($status)	463	=item aio_link $srcpath, $dstpath, $callback->($status)
332		464
333	Asynchronously create a new link to the existing object at C<$srcpath> at	465	Asynchronously create a new link to the existing object at C<$srcpath> at
334	the path C<$dstpath> and call the callback with the result code.	466	the path C<$dstpath> and call the callback with the result code.
335		467
		468
336	=item aio_symlink $srcpath, $dstpath, $callback->($status)	469	=item aio_symlink $srcpath, $dstpath, $callback->($status)
337		470
338	Asynchronously create a new symbolic link to the existing object at C<$srcpath> at	471	Asynchronously create a new symbolic link to the existing object at C<$srcpath> at
339	the path C<$dstpath> and call the callback with the result code.	472	the path C<$dstpath> and call the callback with the result code.
340		473
		474
		475	=item aio_readlink $path, $callback->($link)
		476
		477	Asynchronously read the symlink specified by C<$path> and pass it to
		478	the callback. If an error occurs, nothing or undef gets passed to the
		479	callback.
		480
		481
341	=item aio_rename $srcpath, $dstpath, $callback->($status)	482	=item aio_rename $srcpath, $dstpath, $callback->($status)
342		483
343	Asynchronously rename the object at C<$srcpath> to C<$dstpath>, just as	484	Asynchronously rename the object at C<$srcpath> to C<$dstpath>, just as
344	rename(2) and call the callback with the result code.	485	rename(2) and call the callback with the result code.
345		486
		487
		488	=item aio_mkdir $pathname, $mode, $callback->($status)
		489
		490	Asynchronously mkdir (create) a directory and call the callback with
		491	the result code. C<$mode> will be modified by the umask at the time the
		492	request is executed, so do not change your umask.
		493
		494
346	=item aio_rmdir $pathname, $callback->($status)	495	=item aio_rmdir $pathname, $callback->($status)
347		496
348	Asynchronously rmdir (delete) a directory and call the callback with the	497	Asynchronously rmdir (delete) a directory and call the callback with the
349	result code.	498	result code.
		499
350		500
351	=item aio_readdir $pathname, $callback->($entries)	501	=item aio_readdir $pathname, $callback->($entries)
352		502
353	Unlike the POSIX call of the same name, C<aio_readdir> reads an entire	503	Unlike the POSIX call of the same name, C<aio_readdir> reads an entire
354	directory (i.e. opendir + readdir + closedir). The entries will not be	504	directory (i.e. opendir + readdir + closedir). The entries will not be
355	sorted, and will B<NOT> include the C<.> and C<..> entries.	505	sorted, and will B<NOT> include the C<.> and C<..> entries.
356		506
357	The callback a single argument which is either C<undef> or an array-ref	507	The callback a single argument which is either C<undef> or an array-ref
358	with the filenames.	508	with the filenames.
359		509
		510
		511	=item aio_load $path, $data, $callback->($status)
		512
		513	This is a composite request that tries to fully load the given file into
		514	memory. Status is the same as with aio_read.
		515
		516	=cut
		517
		518	sub aio_load($$;$) {
		519	aio_block {
		520	my ($path, undef, $cb) = @_;
		521	my $data = \$_[1];
		522
		523	my $pri = aioreq_pri;
		524	my $grp = aio_group $cb;
		525
		526	aioreq_pri $pri;
		527	add $grp aio_open $path, O_RDONLY, 0, sub {
		528	my $fh = shift
		529	or return $grp->result (-1);
		530
		531	aioreq_pri $pri;
		532	add $grp aio_read $fh, 0, (-s $fh), $$data, 0, sub {
		533	$grp->result ($_[0]);
		534	};
		535	};
		536
		537	$grp
		538	}
		539	}
		540
		541	=item aio_copy $srcpath, $dstpath, $callback->($status)
		542
		543	Try to copy the I<file> (directories not supported as either source or
		544	destination) from C<$srcpath> to C<$dstpath> and call the callback with
		545	the C<0> (error) or C<-1> ok.
		546
		547	This is a composite request that it creates the destination file with
		548	mode 0200 and copies the contents of the source file into it using
		549	C<aio_sendfile>, followed by restoring atime, mtime, access mode and
		550	uid/gid, in that order.
		551
		552	If an error occurs, the partial destination file will be unlinked, if
		553	possible, except when setting atime, mtime, access mode and uid/gid, where
		554	errors are being ignored.
		555
		556	=cut
		557
		558	sub aio_copy($$;$) {
		559	aio_block {
		560	my ($src, $dst, $cb) = @_;
		561
		562	my $pri = aioreq_pri;
		563	my $grp = aio_group $cb;
		564
		565	aioreq_pri $pri;
		566	add $grp aio_open $src, O_RDONLY, 0, sub {
		567	if (my $src_fh = $_[0]) {
		568	my @stat = stat $src_fh;
		569
		570	aioreq_pri $pri;
		571	add $grp aio_open $dst, O_CREAT \| O_WRONLY \| O_TRUNC, 0200, sub {
		572	if (my $dst_fh = $_[0]) {
		573	aioreq_pri $pri;
		574	add $grp aio_sendfile $dst_fh, $src_fh, 0, $stat[7], sub {
		575	if ($_[0] == $stat[7]) {
		576	$grp->result (0);
		577	close $src_fh;
		578
		579	# those should not normally block. should. should.
		580	utime $stat[8], $stat[9], $dst;
		581	chmod $stat[2] & 07777, $dst_fh;
		582	chown $stat[4], $stat[5], $dst_fh;
		583	close $dst_fh;
		584	} else {
		585	$grp->result (-1);
		586	close $src_fh;
		587	close $dst_fh;
		588
		589	aioreq $pri;
		590	add $grp aio_unlink $dst;
		591	}
		592	};
		593	} else {
		594	$grp->result (-1);
		595	}
		596	},
		597
		598	} else {
		599	$grp->result (-1);
		600	}
		601	};
		602
		603	$grp
		604	}
		605	}
		606
		607	=item aio_move $srcpath, $dstpath, $callback->($status)
		608
		609	Try to move the I<file> (directories not supported as either source or
		610	destination) from C<$srcpath> to C<$dstpath> and call the callback with
		611	the C<0> (error) or C<-1> ok.
		612
		613	This is a composite request that tries to rename(2) the file first. If
		614	rename files with C<EXDEV>, it copies the file with C<aio_copy> and, if
		615	that is successful, unlinking the C<$srcpath>.
		616
		617	=cut
		618
		619	sub aio_move($$;$) {
		620	aio_block {
		621	my ($src, $dst, $cb) = @_;
		622
		623	my $pri = aioreq_pri;
		624	my $grp = aio_group $cb;
		625
		626	aioreq_pri $pri;
		627	add $grp aio_rename $src, $dst, sub {
		628	if ($_[0] && $! == EXDEV) {
		629	aioreq_pri $pri;
		630	add $grp aio_copy $src, $dst, sub {
		631	$grp->result ($_[0]);
		632
		633	if (!$_[0]) {
		634	aioreq_pri $pri;
		635	add $grp aio_unlink $src;
		636	}
		637	};
		638	} else {
		639	$grp->result ($_[0]);
		640	}
		641	};
		642
		643	$grp
		644	}
		645	}
		646
360	=item aio_scandir $path, $maxreq, $callback->($dirs, $nondirs)	647	=item aio_scandir $path, $maxreq, $callback->($dirs, $nondirs)
361		648
362	[EXPERIMENTAL due to internal aio_group use]
363
364	Scans a directory (similar to C<aio_readdir>) but additionally tries to	649	Scans a directory (similar to C<aio_readdir>) but additionally tries to
365	separate the entries of directory C<$path> into two sets of names, ones	650	efficiently separate the entries of directory C<$path> into two sets of
366	you can recurse into (directories or links to them), and ones you cannot	651	names, directories you can recurse into (directories), and ones you cannot
367	recurse into (everything else).	652	recurse into (everything else, including symlinks to directories).
368		653
369	C<aio_scandir> is a composite request that creates of many sub requests_	654	C<aio_scandir> is a composite request that creates of many sub requests_
370	C<$maxreq> specifies the maximum number of outstanding aio requests that	655	C<$maxreq> specifies the maximum number of outstanding aio requests that
371	this function generates. If it is C<< <= 0 >>, then a suitable default	656	this function generates. If it is C<< <= 0 >>, then a suitable default
372	will be chosen (currently 6).	657	will be chosen (currently 4).
373		658
374	On error, the callback is called without arguments, otherwise it receives	659	On error, the callback is called without arguments, otherwise it receives
375	two array-refs with path-relative entry names.	660	two array-refs with path-relative entry names.
376		661
377	Example:	662	Example:
…		…
411	as those tend to return 0 or 1 as link counts, which disables the	696	as those tend to return 0 or 1 as link counts, which disables the
412	directory counting heuristic.	697	directory counting heuristic.
413		698
414	=cut	699	=cut
415		700
416	sub aio_scandir($$$) {	701	sub aio_scandir($$;$) {
		702	aio_block {
417	my ($path, $maxreq, $cb) = @_;	703	my ($path, $maxreq, $cb) = @_;
418		704
		705	my $pri = aioreq_pri;
		706
419	my $grp = aio_group $cb;	707	my $grp = aio_group $cb;
420		708
421	$maxreq = 6 if $maxreq <= 0;	709	$maxreq = 4 if $maxreq <= 0;
422		710
423	# stat once	711	# stat once
		712	aioreq_pri $pri;
424	add $grp aio_stat $path, sub {	713	add $grp aio_stat $path, sub {
425	return $grp->result () if $_[0];	714	return $grp->result () if $_[0];
426	my $now = time;	715	my $now = time;
427	my $hash1 = join ":", (stat _)[0,1,3,7,9];	716	my $hash1 = join ":", (stat _)[0,1,3,7,9];
428		717
429	# read the directory entries	718	# read the directory entries
		719	aioreq_pri $pri;
430	add $grp aio_readdir $path, sub {	720	add $grp aio_readdir $path, sub {
431	my $entries = shift	721	my $entries = shift
432	or return $grp->result ();	722	or return $grp->result ();
433		723
434	# stat the dir another time	724	# stat the dir another time
		725	aioreq_pri $pri;
435	add $grp aio_stat $path, sub {	726	add $grp aio_stat $path, sub {
436	my $hash2 = join ":", (stat _)[0,1,3,7,9];	727	my $hash2 = join ":", (stat _)[0,1,3,7,9];
437		728
438	my $ndirs;	729	my $ndirs;
439		730
440	# take the slow route if anything looks fishy	731	# take the slow route if anything looks fishy
441	if ($hash1 ne $hash2 or (stat _)[9] == $now) {	732	if ($hash1 ne $hash2 or (stat _)[9] == $now) {
442	$ndirs = -1;	733	$ndirs = -1;
443	} else {	734	} else {
444	# if nlink == 2, we are finished	735	# if nlink == 2, we are finished
445	# on non-posix-fs's, we rely on nlink < 2	736	# on non-posix-fs's, we rely on nlink < 2
446	$ndirs = (stat _)[3] - 2	737	$ndirs = (stat _)[3] - 2
447	or return $grp->result ([], $entries);	738	or return $grp->result ([], $entries);
448	}	739	}
449		740
450	# sort into likely dirs and likely nondirs	741	# sort into likely dirs and likely nondirs
451	# dirs == files without ".", short entries first	742	# dirs == files without ".", short entries first
452	$entries = [map $_->[0],	743	$entries = [map $_->[0],
453	sort { $b->[1] cmp $a->[1] }	744	sort { $b->[1] cmp $a->[1] }
454	map [$_, sprintf "%s%04d", (/.\./ ? "1" : "0"), length],	745	map [$_, sprintf "%s%04d", (/.\./ ? "1" : "0"), length],
455	@$entries];	746	@$entries];
456		747
457	my (@dirs, @nondirs);	748	my (@dirs, @nondirs);
458		749
459	my ($statcb, $schedcb);
460	my $nreq = 0;
461
462	my $statgrp = add $grp aio_group;	750	my $statgrp = add $grp aio_group sub {
463
464	$schedcb = sub {
465	if (@$entries) {
466	if ($nreq < $maxreq) {
467	my $ent = pop @$entries;
468	$nreq++;
469	add $statgrp aio_stat "$path/$ent/.", sub { $statcb->($_[0], $ent) };
470	}
471	} elsif (!$nreq) {
472	# finished
473	$statgrp->cancel;
474	undef $statcb;
475	undef $schedcb;
476	$grp->result (\@dirs, \@nondirs);	751	$grp->result (\@dirs, \@nondirs);
477	}	752	};
		753
		754	limit $statgrp $maxreq;
		755	feed $statgrp sub {
		756	return unless @$entries;
		757	my $entry = pop @$entries;
		758
		759	aioreq_pri $pri;
		760	add $statgrp aio_stat "$path/$entry/.", sub {
		761	if ($_[0] < 0) {
		762	push @nondirs, $entry;
		763	} else {
		764	# need to check for real directory
		765	aioreq_pri $pri;
		766	add $statgrp aio_lstat "$path/$entry", sub {
		767	if (-d _) {
		768	push @dirs, $entry;
		769
		770	unless (--$ndirs) {
		771	push @nondirs, @$entries;
		772	feed $statgrp;
		773	}
		774	} else {
		775	push @nondirs, $entry;
		776	}
		777	}
		778	}
		779	};
		780	};
478	};	781	};
479	$statcb = sub {
480	my ($status, $entry) = @_;
481
482	if ($status < 0) {
483	$nreq--;
484	push @nondirs, $entry;
485	&$schedcb;
486	} else {
487	# need to check for real directory
488	add $grp aio_lstat "$path/$entry", sub {
489	$nreq--;
490
491	if (-d _) {
492	push @dirs, $entry;
493
494	if (!--$ndirs) {
495	push @nondirs, @$entries;
496	$entries = [];
497	}
498	} else {
499	push @nondirs, $entry;
500	}
501
502	&$schedcb;
503	}
504	}
505	};
506
507	&$schedcb while @$entries && $nreq < $maxreq;
508	};	782	};
509	};	783	};
		784
		785	$grp
510	};	786	}
		787	}
511		788
		789	=item aio_rmtree $path, $callback->($status)
		790
		791	Delete a directory tree starting (and including) C<$path>, return the
		792	status of the final C<rmdir> only. This is a composite request that
		793	uses C<aio_scandir> to recurse into and rmdir directories, and unlink
		794	everything else.
		795
		796	=cut
		797
		798	sub aio_rmtree;
		799	sub aio_rmtree($;$) {
		800	aio_block {
		801	my ($path, $cb) = @_;
		802
		803	my $pri = aioreq_pri;
		804	my $grp = aio_group $cb;
		805
		806	aioreq_pri $pri;
		807	add $grp aio_scandir $path, 0, sub {
		808	my ($dirs, $nondirs) = @_;
		809
		810	my $dirgrp = aio_group sub {
		811	add $grp aio_rmdir $path, sub {
		812	$grp->result ($_[0]);
		813	};
		814	};
		815
		816	(aioreq_pri $pri), add $dirgrp aio_rmtree "$path/$_" for @$dirs;
		817	(aioreq_pri $pri), add $dirgrp aio_unlink "$path/$_" for @$nondirs;
		818
		819	add $grp $dirgrp;
		820	};
		821
512	$grp	822	$grp
		823	}
513	}	824	}
514		825
515	=item aio_fsync $fh, $callback->($status)	826	=item aio_fsync $fh, $callback->($status)
516		827
517	Asynchronously call fsync on the given filehandle and call the callback	828	Asynchronously call fsync on the given filehandle and call the callback
…		…
525	If this call isn't available because your OS lacks it or it couldn't be	836	If this call isn't available because your OS lacks it or it couldn't be
526	detected, it will be emulated by calling C<fsync> instead.	837	detected, it will be emulated by calling C<fsync> instead.
527		838
528	=item aio_group $callback->(...)	839	=item aio_group $callback->(...)
529		840
530	[EXPERIMENTAL]
531
532	This is a very special aio request: Instead of doing something, it is a	841	This is a very special aio request: Instead of doing something, it is a
533	container for other aio requests, which is useful if you want to bundle	842	container for other aio requests, which is useful if you want to bundle
534	many requests into a single, composite, request.	843	many requests into a single, composite, request with a definite callback
		844	and the ability to cancel the whole request with its subrequests.
535		845
536	Returns an object of class L<IO::AIO::GRP>. See its documentation below	846	Returns an object of class L<IO::AIO::GRP>. See its documentation below
537	for more info.	847	for more info.
538		848
539	Example:	849	Example:
…		…
558	phase and still requires a worker thread. Thus, the callback will not	868	phase and still requires a worker thread. Thus, the callback will not
559	be executed immediately but only after other requests in the queue have	869	be executed immediately but only after other requests in the queue have
560	entered their execution phase. This can be used to measure request	870	entered their execution phase. This can be used to measure request
561	latency.	871	latency.
562		872
563	=item IO::AIO::aio_sleep $fractional_seconds, $callback->() NOT EXPORTED	873	=item IO::AIO::aio_busy $fractional_seconds, $callback->() NOT EXPORTED
564		874
565	Mainly used for debugging and benchmarking, this aio request puts one of	875	Mainly used for debugging and benchmarking, this aio request puts one of
566	the request workers to sleep for the given time.	876	the request workers to sleep for the given time.
567		877
568	While it is theoretically handy to have simple I/O scheduling requests	878	While it is theoretically handy to have simple I/O scheduling requests
569	like sleep and file handle readable/writable, the overhead this creates	879	like sleep and file handle readable/writable, the overhead this creates is
570	is immense, so do not use this function except to put your application	880	immense (it blocks a thread for a long time) so do not use this function
571	under artificial I/O pressure.	881	except to put your application under artificial I/O pressure.
572		882
573	=back	883	=back
574		884
575	=head2 IO::AIO::REQ CLASS	885	=head2 IO::AIO::REQ CLASS
576		886
577	All non-aggregate C<aio_*> functions return an object of this class when	887	All non-aggregate C<aio_*> functions return an object of this class when
578	called in non-void context.	888	called in non-void context.
579
580	A request always moves through the following five states in its lifetime,
581	in order: B<ready> (request has been created, but has not been executed
582	yet), B<execute> (request is currently being executed), B<pending>
583	(request has been executed but callback has not been called yet),
584	B<result> (results are being processed synchronously, includes calling the
585	callback) and B<done> (request has reached the end of its lifetime and
586	holds no resources anymore).
587		889
588	=over 4	890	=over 4
589		891
590	=item cancel $req	892	=item cancel $req
591		893
…		…
645	=item * They can also can also be added to other IO::AIO::GRP objects.	947	=item * They can also can also be added to other IO::AIO::GRP objects.
646		948
647	=item * You must not add requests to a group from within the group callback (or	949	=item * You must not add requests to a group from within the group callback (or
648	any later time).	950	any later time).
649		951
650	=item * This does not harmonise well with C<max_outstanding>, so best do
651	not combine C<aio_group> with it. Groups and feeders are recommended for
652	this kind of concurrency-limiting.
653
654	=back	952	=back
655		953
656	Their lifetime, simplified, looks like this: when they are empty, they	954	Their lifetime, simplified, looks like this: when they are empty, they
657	will finish very quickly. If they contain only requests that are in the	955	will finish very quickly. If they contain only requests that are in the
658	C<done> state, they will also finish. Otherwise they will continue to	956	C<done> state, they will also finish. Otherwise they will continue to
…		…
673	be added, including other groups, as long as you do not create circular	971	be added, including other groups, as long as you do not create circular
674	dependencies.	972	dependencies.
675		973
676	Returns all its arguments.	974	Returns all its arguments.
677		975
		976	=item $grp->cancel_subs
		977
		978	Cancel all subrequests and clears any feeder, but not the group request
		979	itself. Useful when you queued a lot of events but got a result early.
		980
678	=item $grp->result (...)	981	=item $grp->result (...)
679		982
680	Set the result value(s) that will be passed to the group callback when all	983	Set the result value(s) that will be passed to the group callback when all
681	subrequests have finished. By default, no argument will be passed.	984	subrequests have finished and set thre groups errno to the current value
		985	of errno (just like calling C<errno> without an error number). By default,
		986	no argument will be passed and errno is zero.
		987
		988	=item $grp->errno ([$errno])
		989
		990	Sets the group errno value to C<$errno>, or the current value of errno
		991	when the argument is missing.
		992
		993	Every aio request has an associated errno value that is restored when
		994	the callback is invoked. This method lets you change this value from its
		995	default (0).
		996
		997	Calling C<result> will also set errno, so make sure you either set C<$!>
		998	before the call to C<result>, or call c<errno> after it.
682		999
683	=item feed $grp $callback->($grp)	1000	=item feed $grp $callback->($grp)
684
685	[VERY EXPERIMENTAL]
686		1001
687	Sets a feeder/generator on this group: every group can have an attached	1002	Sets a feeder/generator on this group: every group can have an attached
688	generator that generates requests if idle. The idea behind this is that,	1003	generator that generates requests if idle. The idea behind this is that,
689	although you could just queue as many requests as you want in a group,	1004	although you could just queue as many requests as you want in a group,
690	this might starve other requests for a potentially long time. For	1005	this might starve other requests for a potentially long time. For
…		…
727		1042
728	=back	1043	=back
729		1044
730	=head2 SUPPORT FUNCTIONS	1045	=head2 SUPPORT FUNCTIONS
731		1046
		1047	=head3 EVENT PROCESSING AND EVENT LOOP INTEGRATION
		1048
732	=over 4	1049	=over 4
733		1050
734	=item $fileno = IO::AIO::poll_fileno	1051	=item $fileno = IO::AIO::poll_fileno
735		1052
736	Return the I<request result pipe file descriptor>. This filehandle must be	1053	Return the I<request result pipe file descriptor>. This filehandle must be
…		…
740		1057
741	See C<poll_cb> for an example.	1058	See C<poll_cb> for an example.
742		1059
743	=item IO::AIO::poll_cb	1060	=item IO::AIO::poll_cb
744		1061
745	Process all outstanding events on the result pipe. You have to call this	1062	Process some outstanding events on the result pipe. You have to call this
746	regularly. Returns the number of events processed. Returns immediately	1063	regularly. Returns the number of events processed. Returns immediately
747	when no events are outstanding.	1064	when no events are outstanding. The amount of events processed depends on
		1065	the settings of C<IO::AIO::max_poll_req> and C<IO::AIO::max_poll_time>.
		1066
		1067	If not all requests were processed for whatever reason, the filehandle
		1068	will still be ready when C<poll_cb> returns.
748		1069
749	Example: Install an Event watcher that automatically calls	1070	Example: Install an Event watcher that automatically calls
750	IO::AIO::poll_cb with high priority:	1071	IO::AIO::poll_cb with high priority:
751		1072
752	Event->io (fd => IO::AIO::poll_fileno,	1073	Event->io (fd => IO::AIO::poll_fileno,
753	poll => 'r', async => 1,	1074	poll => 'r', async => 1,
754	cb => \&IO::AIO::poll_cb);	1075	cb => \&IO::AIO::poll_cb);
755		1076
		1077	=item IO::AIO::max_poll_reqs $nreqs
		1078
		1079	=item IO::AIO::max_poll_time $seconds
		1080
		1081	These set the maximum number of requests (default C<0>, meaning infinity)
		1082	that are being processed by C<IO::AIO::poll_cb> in one call, respectively
		1083	the maximum amount of time (default C<0>, meaning infinity) spent in
		1084	C<IO::AIO::poll_cb> to process requests (more correctly the mininum amount
		1085	of time C<poll_cb> is allowed to use).
		1086
		1087	Setting C<max_poll_time> to a non-zero value creates an overhead of one
		1088	syscall per request processed, which is not normally a problem unless your
		1089	callbacks are really really fast or your OS is really really slow (I am
		1090	not mentioning Solaris here). Using C<max_poll_reqs> incurs no overhead.
		1091
		1092	Setting these is useful if you want to ensure some level of
		1093	interactiveness when perl is not fast enough to process all requests in
		1094	time.
		1095
		1096	For interactive programs, values such as C<0.01> to C<0.1> should be fine.
		1097
		1098	Example: Install an Event watcher that automatically calls
		1099	IO::AIO::poll_cb with low priority, to ensure that other parts of the
		1100	program get the CPU sometimes even under high AIO load.
		1101
		1102	# try not to spend much more than 0.1s in poll_cb
		1103	IO::AIO::max_poll_time 0.1;
		1104
		1105	# use a low priority so other tasks have priority
		1106	Event->io (fd => IO::AIO::poll_fileno,
		1107	poll => 'r', nice => 1,
		1108	cb => &IO::AIO::poll_cb);
		1109
756	=item IO::AIO::poll_wait	1110	=item IO::AIO::poll_wait
757		1111
		1112	If there are any outstanding requests and none of them in the result
758	Wait till the result filehandle becomes ready for reading (simply does a	1113	phase, wait till the result filehandle becomes ready for reading (simply
759	C<select> on the filehandle. This is useful if you want to synchronously wait	1114	does a C<select> on the filehandle. This is useful if you want to
760	for some requests to finish).	1115	synchronously wait for some requests to finish).
761		1116
762	See C<nreqs> for an example.	1117	See C<nreqs> for an example.
763		1118
		1119	=item IO::AIO::poll
		1120
		1121	Waits until some requests have been handled.
		1122
		1123	Returns the number of requests processed, but is otherwise strictly
		1124	equivalent to:
		1125
		1126	IO::AIO::poll_wait, IO::AIO::poll_cb
		1127
764	=item IO::AIO::nreqs	1128	=item IO::AIO::flush
765		1129
766	Returns the number of requests currently outstanding (i.e. for which their	1130	Wait till all outstanding AIO requests have been handled.
767	callback has not been invoked yet).
768		1131
769	Example: wait till there are no outstanding requests anymore:	1132	Strictly equivalent to:
770		1133
771	IO::AIO::poll_wait, IO::AIO::poll_cb	1134	IO::AIO::poll_wait, IO::AIO::poll_cb
772	while IO::AIO::nreqs;	1135	while IO::AIO::nreqs;
773		1136
774	=item IO::AIO::flush	1137	=back
775		1138
776	Wait till all outstanding AIO requests have been handled.	1139	=head3 CONTROLLING THE NUMBER OF THREADS
777		1140
778	Strictly equivalent to:	1141	=over
779
780	IO::AIO::poll_wait, IO::AIO::poll_cb
781	while IO::AIO::nreqs;
782
783	=item IO::AIO::poll
784
785	Waits until some requests have been handled.
786
787	Strictly equivalent to:
788
789	IO::AIO::poll_wait, IO::AIO::poll_cb
790	if IO::AIO::nreqs;
791		1142
792	=item IO::AIO::min_parallel $nthreads	1143	=item IO::AIO::min_parallel $nthreads
793		1144
794	Set the minimum number of AIO threads to C<$nthreads>. The current	1145	Set the minimum number of AIO threads to C<$nthreads>. The current
795	default is C<8>, which means eight asynchronous operations can execute	1146	default is C<8>, which means eight asynchronous operations can execute
796	concurrently at any one time (the number of outstanding requests,	1147	concurrently at any one time (the number of outstanding requests,
797	however, is unlimited).	1148	however, is unlimited).
798		1149
799	IO::AIO starts threads only on demand, when an AIO request is queued and	1150	IO::AIO starts threads only on demand, when an AIO request is queued and
800	no free thread exists.	1151	no free thread exists. Please note that queueing up a hundred requests can
		1152	create demand for a hundred threads, even if it turns out that everything
		1153	is in the cache and could have been processed faster by a single thread.
801		1154
802	It is recommended to keep the number of threads relatively low, as some	1155	It is recommended to keep the number of threads relatively low, as some
803	Linux kernel versions will scale negatively with the number of threads	1156	Linux kernel versions will scale negatively with the number of threads
804	(higher parallelity => MUCH higher latency). With current Linux 2.6	1157	(higher parallelity => MUCH higher latency). With current Linux 2.6
805	versions, 4-32 threads should be fine.	1158	versions, 4-32 threads should be fine.
…		…
819	This module automatically runs C<max_parallel 0> at program end, to ensure	1172	This module automatically runs C<max_parallel 0> at program end, to ensure
820	that all threads are killed and that there are no outstanding requests.	1173	that all threads are killed and that there are no outstanding requests.
821		1174
822	Under normal circumstances you don't need to call this function.	1175	Under normal circumstances you don't need to call this function.
823		1176
		1177	=item IO::AIO::max_idle $nthreads
		1178
		1179	Limit the number of threads (default: 4) that are allowed to idle (i.e.,
		1180	threads that did not get a request to process within 10 seconds). That
		1181	means if a thread becomes idle while C<$nthreads> other threads are also
		1182	idle, it will free its resources and exit.
		1183
		1184	This is useful when you allow a large number of threads (e.g. 100 or 1000)
		1185	to allow for extremely high load situations, but want to free resources
		1186	under normal circumstances (1000 threads can easily consume 30MB of RAM).
		1187
		1188	The default is probably ok in most situations, especially if thread
		1189	creation is fast. If thread creation is very slow on your system you might
		1190	want to use larger values.
		1191
824	=item $oldnreqs = IO::AIO::max_outstanding $nreqs	1192	=item $oldmaxreqs = IO::AIO::max_outstanding $maxreqs
825		1193
826	[DEPRECATED]	1194	This is a very bad function to use in interactive programs because it
		1195	blocks, and a bad way to reduce concurrency because it is inexact: Better
		1196	use an C<aio_group> together with a feed callback.
827		1197
828	Sets the maximum number of outstanding requests to C<$nreqs>. If you	1198	Sets the maximum number of outstanding requests to C<$nreqs>. If you
829	try to queue up more than this number of requests, the caller will block until	1199	to queue up more than this number of requests, the next call to the
830	some requests have been handled.	1200	C<poll_cb> (and C<poll_some> and other functions calling C<poll_cb>)
		1201	function will block until the limit is no longer exceeded.
831		1202
832	The default is very large, so normally there is no practical limit. If you	1203	The default value is very large, so there is no practical limit on the
833	queue up many requests in a loop it often improves speed if you set	1204	number of outstanding requests.
834	this to a relatively low number, such as C<100>.
835		1205
836	This function does not work well together with C<aio_group>'s, and their	1206	You can still queue as many requests as you want. Therefore,
837	feeder interface is better suited to limiting concurrency, so do not use	1207	C<max_oustsanding> is mainly useful in simple scripts (with low values) or
838	this function.	1208	as a stop gap to shield against fatal memory overflow (with large values).
839		1209
840	Under normal circumstances you don't need to call this function.	1210	=back
		1211
		1212	=head3 STATISTICAL INFORMATION
		1213
		1214	=over
		1215
		1216	=item IO::AIO::nreqs
		1217
		1218	Returns the number of requests currently in the ready, execute or pending
		1219	states (i.e. for which their callback has not been invoked yet).
		1220
		1221	Example: wait till there are no outstanding requests anymore:
		1222
		1223	IO::AIO::poll_wait, IO::AIO::poll_cb
		1224	while IO::AIO::nreqs;
		1225
		1226	=item IO::AIO::nready
		1227
		1228	Returns the number of requests currently in the ready state (not yet
		1229	executed).
		1230
		1231	=item IO::AIO::npending
		1232
		1233	Returns the number of requests currently in the pending state (executed,
		1234	but not yet processed by poll_cb).
841		1235
842	=back	1236	=back
843		1237
844	=cut	1238	=cut
845		1239
…		…
859	*$sym	1253	*$sym
860	}	1254	}
861		1255
862	min_parallel 8;	1256	min_parallel 8;
863		1257
864	END {	1258	END { flush }
865	max_parallel 0;
866	}
867		1259
868	1;	1260	1;
869		1261
870	=head2 FORK BEHAVIOUR	1262	=head2 FORK BEHAVIOUR
871		1263
872	This module should do "the right thing" when the process using it forks:	1264	This module should do "the right thing" when the process using it forks:
873		1265
874	Before the fork, IO::AIO enters a quiescent state where no requests	1266	Before the fork, IO::AIO enters a quiescent state where no requests
875	can be added in other threads and no results will be processed. After	1267	can be added in other threads and no results will be processed. After
876	the fork the parent simply leaves the quiescent state and continues	1268	the fork the parent simply leaves the quiescent state and continues
877	request/result processing, while the child clears the request/result	1269	request/result processing, while the child frees the request/result queue
878	queue (so the requests started before the fork will only be handled in	1270	(so that the requests started before the fork will only be handled in the
879	the parent). Threads will be started on demand until the limit ste in the	1271	parent). Threads will be started on demand until the limit set in the
880	parent process has been reached again.	1272	parent process has been reached again.
881		1273
882	In short: the parent will, after a short pause, continue as if fork had	1274	In short: the parent will, after a short pause, continue as if fork had
883	not been called, while the child will act as if IO::AIO has not been used	1275	not been called, while the child will act as if IO::AIO has not been used
884	yet.	1276	yet.
885		1277
886	=head2 MEMORY USAGE	1278	=head2 MEMORY USAGE
887		1279
		1280	Per-request usage:
		1281
888	Each aio request uses - depending on your architecture - around 128 bytes	1282	Each aio request uses - depending on your architecture - around 100-200
889	of memory. In addition, stat requests need a stat buffer (possibly a few	1283	bytes of memory. In addition, stat requests need a stat buffer (possibly
890	hundred bytes). Perl scalars and other data passed into aio requests will	1284	a few hundred bytes), readdir requires a result buffer and so on. Perl
891	also be locked.	1285	scalars and other data passed into aio requests will also be locked and
		1286	will consume memory till the request has entered the done state.
892		1287
893	This is now awfully much, so queuing lots of requests is not usually a	1288	This is now awfully much, so queuing lots of requests is not usually a
894	problem.	1289	problem.
895		1290
896	Each thread needs a stack area which is usually around 16k, sometimes much	1291	Per-thread usage:
897	larger, depending on the OS.	1292
		1293	In the execution phase, some aio requests require more memory for
		1294	temporary buffers, and each thread requires a stack and other data
		1295	structures (usually around 16k-128k, depending on the OS).
		1296
		1297	=head1 KNOWN BUGS
		1298
		1299	Known bugs will be fixed in the next release.
898		1300
899	=head1 SEE ALSO	1301	=head1 SEE ALSO
900		1302
901	L<Coro::AIO>.	1303	L<Coro::AIO>.
902		1304

Diff Legend

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

Comparing IO-AIO/AIO.pm (file contents): Revision 1.68 by root, Tue Oct 24 03:17:39 2006 UTC vs. Revision 1.107 by root, Fri Jun 1 05:52:59 2007 UTC

Diff Legend

Comparing IO-AIO/AIO.pm (file contents):
Revision 1.68 by root, Tue Oct 24 03:17:39 2006 UTC vs.
Revision 1.107 by root, Fri Jun 1 05:52:59 2007 UTC