[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 1.87 by root, Sun Oct 29 00:52:02 2006 UTC

…		…
18	};	18	};
19		19
20	# version 2+ has request and group objects	20	# version 2+ has request and group objects
21	use IO::AIO 2;	21	use IO::AIO 2;
22		22
		23	aioreq_pri 4; # give next request a very high priority
23	my $req = aio_unlink "/tmp/file", sub { };	24	my $req = aio_unlink "/tmp/file", sub { };
24	$req->cancel; # cancel request if still in queue	25	$req->cancel; # cancel request if still in queue
25		26
26	my $grp = aio_group sub { print "all stats done\n" };	27	my $grp = aio_group sub { print "all stats done\n" };
27	add $grp aio_stat "..." for ...;	28	add $grp aio_stat "..." for ...;
…		…
50	=head1 DESCRIPTION	51	=head1 DESCRIPTION
51		52
52	This module implements asynchronous I/O using whatever means your	53	This module implements asynchronous I/O using whatever means your
53	operating system supports.	54	operating system supports.
54		55
		56	Asynchronous means that operations that can normally block your program
		57	(e.g. reading from disk) will be done asynchronously: the operation
		58	will still block, but you can do something else in the meantime. This
		59	is extremely useful for programs that need to stay interactive even
		60	when doing heavy I/O (GUI programs, high performance network servers
		61	etc.), but can also be used to easily do operations in parallel that are
		62	normally done sequentially, e.g. stat'ing many files, which is much faster
		63	on a RAID volume or over NFS when you do a number of stat operations
		64	concurrently.
		65
		66	While this works on all types of file descriptors (for example sockets),
		67	using these functions on file descriptors that support nonblocking
		68	operation (again, sockets, pipes etc.) is very inefficient. Use an event
		69	loop for that (such as the L<Event\|Event> module): IO::AIO will naturally
		70	fit into such an event loop itself.
		71
55	Currently, a number of threads are started that execute your read/writes	72	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	73	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	74	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	75	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	76	functions available on many operating systems. However, they are often
60	(Linux doesn't allow them on normal files currently, for example),	77	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	78	files currently, for example), and they would only support aio_read and
62	functionality would have to be implemented using threads anyway.	79	aio_write, so the remaining functionality would have to be implemented
		80	using threads anyway.
63		81
64	Although the module will work with in the presence of other threads, it is	82	Although the module will work with in the presence of other (Perl-)
65	currently not reentrant, so use appropriate locking yourself, always call	83	threads, it is currently not reentrant in any way, so use appropriate
66	C<poll_cb> from within the same thread, or never call C<poll_cb> (or other	84	locking yourself, always call C<poll_cb> from within the same thread, or
67	C<aio_> functions) recursively.	85	never call C<poll_cb> (or other C<aio_> functions) recursively.
		86
		87	=head2 EXAMPLE
		88
		89	This is a simple example that uses the Event module and loads
		90	F</etc/passwd> asynchronously:
		91
		92	use Fcntl;
		93	use Event;
		94	use IO::AIO;
		95
		96	# register the IO::AIO callback with Event
		97	Event->io (fd => IO::AIO::poll_fileno,
		98	poll => 'r',
		99	cb => \&IO::AIO::poll_cb);
		100
		101	# queue the request to open /etc/passwd
		102	aio_open "/etc/passwd", O_RDONLY, 0, sub {
		103	my $fh = $_[0]
		104	or die "error while opening: $!";
		105
		106	# stat'ing filehandles is generally non-blocking
		107	my $size = -s $fh;
		108
		109	# queue a request to read the file
		110	my $contents;
		111	aio_read $fh, 0, $size, $contents, 0, sub {
		112	$_[0] == $size
		113	or die "short read: $!";
		114
		115	close $fh;
		116
		117	# file contents now in $contents
		118	print $contents;
		119
		120	# exit event loop and program
		121	Event::unloop;
		122	};
		123	};
		124
		125	# possibly queue up other requests, or open GUI windows,
		126	# check for sockets etc. etc.
		127
		128	# process events as long as there are some:
		129	Event::loop;
		130
		131	=head1 REQUEST ANATOMY AND LIFETIME
		132
		133	Every C<aio_*> function creates a request. which is a C data structure not
		134	directly visible to Perl.
		135
		136	If called in non-void context, every request function returns a Perl
		137	object representing the request. In void context, nothing is returned,
		138	which saves a bit of memory.
		139
		140	The perl object is a fairly standard ref-to-hash object. The hash contents
		141	are not used by IO::AIO so you are free to store anything you like in it.
		142
		143	During their existance, aio requests travel through the following states,
		144	in order:
		145
		146	=over 4
		147
		148	=item ready
		149
		150	Immediately after a request is created it is put into the ready state,
		151	waiting for a thread to execute it.
		152
		153	=item execute
		154
		155	A thread has accepted the request for processing and is currently
		156	executing it (e.g. blocking in read).
		157
		158	=item pending
		159
		160	The request has been executed and is waiting for result processing.
		161
		162	While request submission and execution is fully asynchronous, result
		163	processing is not and relies on the perl interpreter calling C<poll_cb>
		164	(or another function with the same effect).
		165
		166	=item result
		167
		168	The request results are processed synchronously by C<poll_cb>.
		169
		170	The C<poll_cb> function will process all outstanding aio requests by
		171	calling their callbacks, freeing memory associated with them and managing
		172	any groups they are contained in.
		173
		174	=item done
		175
		176	Request has reached the end of its lifetime and holds no resources anymore
		177	(except possibly for the Perl object, but its connection to the actual
		178	aio request is severed and calling its methods will either do nothing or
		179	result in a runtime error).
68		180
69	=cut	181	=cut
70		182
71	package IO::AIO;	183	package IO::AIO;
72		184
…		…
74	use strict 'vars';	186	use strict 'vars';
75		187
76	use base 'Exporter';	188	use base 'Exporter';
77		189
78	BEGIN {	190	BEGIN {
79	our $VERSION = '2.0';	191	our $VERSION = '2.1';
80		192
81	our @AIO_REQ = qw(aio_sendfile aio_read aio_write aio_open aio_close aio_stat	193	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	194	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	195	aio_fsync aio_fdatasync aio_readahead aio_rename aio_link aio_move
84	aio_group aio_nop);	196	aio_copy aio_group aio_nop aio_mknod);
85	our @EXPORT = (@AIO_REQ, qw(aioreq_pri));	197	our @EXPORT = (@AIO_REQ, qw(aioreq_pri aioreq_nice));
86	our @EXPORT_OK = qw(poll_fileno poll_cb poll_wait flush	198	our @EXPORT_OK = qw(poll_fileno poll_cb poll_wait flush
87	min_parallel max_parallel max_outstanding nreqs);	199	min_parallel max_parallel max_idle
		200	nreqs nready npending nthreads
		201	max_poll_time max_poll_reqs);
88		202
89	@IO::AIO::GRP::ISA = 'IO::AIO::REQ';	203	@IO::AIO::GRP::ISA = 'IO::AIO::REQ';
90		204
91	require XSLoader;	205	require XSLoader;
92	XSLoader::load ("IO::AIO", $VERSION);	206	XSLoader::load ("IO::AIO", $VERSION);
93	}	207	}
94		208
95	=head1 FUNCTIONS	209	=head1 FUNCTIONS
96		210
97	=head2 AIO FUNCTIONS	211	=head2 AIO REQUEST FUNCTIONS
98		212
99	All the C<aio_*> calls are more or less thin wrappers around the syscall	213	All the C<aio_*> calls are more or less thin wrappers around the syscall
100	with the same name (sans C<aio_>). The arguments are similar or identical,	214	with the same name (sans C<aio_>). The arguments are similar or identical,
101	and they all accept an additional (and optional) C<$callback> argument	215	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	216	which must be a code reference. This code reference will get called with
…		…
105	syscall has been executed asynchronously.	219	syscall has been executed asynchronously.
106		220
107	All functions expecting a filehandle keep a copy of the filehandle	221	All functions expecting a filehandle keep a copy of the filehandle
108	internally until the request has finished.	222	internally until the request has finished.
109		223
110	All requests return objects of type L<IO::AIO::REQ> that allow further	224	All functions return request objects of type L<IO::AIO::REQ> that allow
111	manipulation of those requests while they are in-flight.	225	further manipulation of those requests while they are in-flight.
112		226
113	The pathnames you pass to these routines I<must> be absolute and	227	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	228	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	229	request is being executed, the current working directory could have
116	changed. Alternatively, you can make sure that you never change the	230	changed. Alternatively, you can make sure that you never change the
117	current working directory.	231	current working directory anywhere in the program and then use relative
		232	paths.
118		233
119	To encode pathnames to byte form, either make sure you either: a)	234	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	235	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	236	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	237	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)	238	environment, d) use Glib::filename_from_unicode on unicode filenames or e)
124	use something else.	239	use something else to ensure your scalar has the correct contents.
		240
		241	This works, btw. independent of the internal UTF-8 bit, which IO::AIO
		242	handles correctly wether it is set or not.
125		243
126	=over 4	244	=over 4
		245
		246	=item $prev_pri = aioreq_pri [$pri]
		247
		248	Returns the priority value that would be used for the next request and, if
		249	C<$pri> is given, sets the priority for the next aio request.
		250
		251	The default priority is C<0>, the minimum and maximum priorities are C<-4>
		252	and C<4>, respectively. Requests with higher priority will be serviced
		253	first.
		254
		255	The priority will be reset to C<0> after each call to one of the C<aio_*>
		256	functions.
		257
		258	Example: open a file with low priority, then read something from it with
		259	higher priority so the read request is serviced before other low priority
		260	open requests (potentially spamming the cache):
		261
		262	aioreq_pri -3;
		263	aio_open ..., sub {
		264	return unless $_[0];
		265
		266	aioreq_pri -2;
		267	aio_read $_[0], ..., sub {
		268	...
		269	};
		270	};
		271
		272	=item aioreq_nice $pri_adjust
		273
		274	Similar to C<aioreq_pri>, but subtracts the given value from the current
		275	priority, so the effect is cumulative.
127		276
128	=item aio_open $pathname, $flags, $mode, $callback->($fh)	277	=item aio_open $pathname, $flags, $mode, $callback->($fh)
129		278
130	Asynchronously open or create a file and call the callback with a newly	279	Asynchronously open or create a file and call the callback with a newly
131	created filehandle for the file.	280	created filehandle for the file.
…		…
181		330
182	aio_read $fh, 7, 15, $buffer, 0, sub {	331	aio_read $fh, 7, 15, $buffer, 0, sub {
183	$_[0] > 0 or die "read error: $!";	332	$_[0] > 0 or die "read error: $!";
184	print "read $_[0] bytes: <$buffer>\n";	333	print "read $_[0] bytes: <$buffer>\n";
185	};	334	};
186
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		335
257	=item aio_sendfile $out_fh, $in_fh, $in_offset, $length, $callback->($retval)	336	=item aio_sendfile $out_fh, $in_fh, $in_offset, $length, $callback->($retval)
258		337
259	Tries to copy C<$length> bytes from C<$in_fh> to C<$out_fh>. It starts	338	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	339	reading at byte offset C<$in_offset>, and starts writing at the current
…		…
316	=item aio_unlink $pathname, $callback->($status)	395	=item aio_unlink $pathname, $callback->($status)
317		396
318	Asynchronously unlink (delete) a file and call the callback with the	397	Asynchronously unlink (delete) a file and call the callback with the
319	result code.	398	result code.
320		399
		400	=item aio_mknod $path, $mode, $dev, $callback->($status)
		401
		402	[EXPERIMENTAL]
		403
		404	Asynchronously create a device node (or fifo). See mknod(2).
		405
		406	The only (POSIX-) portable way of calling this function is:
		407
		408	aio_mknod $path, IO::AIO::S_IFIFO \| $mode, 0, sub { ...
		409
321	=item aio_link $srcpath, $dstpath, $callback->($status)	410	=item aio_link $srcpath, $dstpath, $callback->($status)
322		411
323	Asynchronously create a new link to the existing object at C<$srcpath> at	412	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.	413	the path C<$dstpath> and call the callback with the result code.
325		414
…		…
345	sorted, and will B<NOT> include the C<.> and C<..> entries.	434	sorted, and will B<NOT> include the C<.> and C<..> entries.
346		435
347	The callback a single argument which is either C<undef> or an array-ref	436	The callback a single argument which is either C<undef> or an array-ref
348	with the filenames.	437	with the filenames.
349		438
		439	=item aio_copy $srcpath, $dstpath, $callback->($status)
		440
		441	Try to copy the I<file> (directories not supported as either source or
		442	destination) from C<$srcpath> to C<$dstpath> and call the callback with
		443	the C<0> (error) or C<-1> ok.
		444
		445	This is a composite request that it creates the destination file with
		446	mode 0200 and copies the contents of the source file into it using
		447	C<aio_sendfile>, followed by restoring atime, mtime, access mode and
		448	uid/gid, in that order.
		449
		450	If an error occurs, the partial destination file will be unlinked, if
		451	possible, except when setting atime, mtime, access mode and uid/gid, where
		452	errors are being ignored.
		453
		454	=cut
		455
		456	sub aio_copy($$;$) {
		457	my ($src, $dst, $cb) = @_;
		458
		459	my $pri = aioreq_pri;
		460	my $grp = aio_group $cb;
		461
		462	aioreq_pri $pri;
		463	add $grp aio_open $src, O_RDONLY, 0, sub {
		464	if (my $src_fh = $_[0]) {
		465	my @stat = stat $src_fh;
		466
		467	aioreq_pri $pri;
		468	add $grp aio_open $dst, O_CREAT \| O_WRONLY \| O_TRUNC, 0200, sub {
		469	if (my $dst_fh = $_[0]) {
		470	aioreq_pri $pri;
		471	add $grp aio_sendfile $dst_fh, $src_fh, 0, $stat[7], sub {
		472	if ($_[0] == $stat[7]) {
		473	$grp->result (0);
		474	close $src_fh;
		475
		476	# those should not normally block. should. should.
		477	utime $stat[8], $stat[9], $dst;
		478	chmod $stat[2] & 07777, $dst_fh;
		479	chown $stat[4], $stat[5], $dst_fh;
		480	close $dst_fh;
		481	} else {
		482	$grp->result (-1);
		483	close $src_fh;
		484	close $dst_fh;
		485
		486	aioreq $pri;
		487	add $grp aio_unlink $dst;
		488	}
		489	};
		490	} else {
		491	$grp->result (-1);
		492	}
		493	},
		494
		495	} else {
		496	$grp->result (-1);
		497	}
		498	};
		499
		500	$grp
		501	}
		502
		503	=item aio_move $srcpath, $dstpath, $callback->($status)
		504
		505	Try to move the I<file> (directories not supported as either source or
		506	destination) from C<$srcpath> to C<$dstpath> and call the callback with
		507	the C<0> (error) or C<-1> ok.
		508
		509	This is a composite request that tries to rename(2) the file first. If
		510	rename files with C<EXDEV>, it copies the file with C<aio_copy> and, if
		511	that is successful, unlinking the C<$srcpath>.
		512
		513	=cut
		514
		515	sub aio_move($$;$) {
		516	my ($src, $dst, $cb) = @_;
		517
		518	my $pri = aioreq_pri;
		519	my $grp = aio_group $cb;
		520
		521	aioreq_pri $pri;
		522	add $grp aio_rename $src, $dst, sub {
		523	if ($_[0] && $! == EXDEV) {
		524	aioreq_pri $pri;
		525	add $grp aio_copy $src, $dst, sub {
		526	$grp->result ($_[0]);
		527
		528	if (!$_[0]) {
		529	aioreq_pri $pri;
		530	add $grp aio_unlink $src;
		531	}
		532	};
		533	} else {
		534	$grp->result ($_[0]);
		535	}
		536	};
		537
		538	$grp
		539	}
		540
350	=item aio_scandir $path, $maxreq, $callback->($dirs, $nondirs)	541	=item aio_scandir $path, $maxreq, $callback->($dirs, $nondirs)
351		542
352	[EXPERIMENTAL due to internal aio_group use]
353
354	Scans a directory (similar to C<aio_readdir>) but additionally tries to	543	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	544	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	545	names, directories you can recurse into (directories), and ones you cannot
357	recurse into (everything else).	546	recurse into (everything else, including symlinks to directories).
358		547
359	C<aio_scandir> is a composite request that creates of many sub requests_	548	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	549	C<$maxreq> specifies the maximum number of outstanding aio requests that
361	this function generates. If it is C<< <= 0 >>, then a suitable default	550	this function generates. If it is C<< <= 0 >>, then a suitable default
362	will be chosen (currently 6).	551	will be chosen (currently 4).
363		552
364	On error, the callback is called without arguments, otherwise it receives	553	On error, the callback is called without arguments, otherwise it receives
365	two array-refs with path-relative entry names.	554	two array-refs with path-relative entry names.
366		555
367	Example:	556	Example:
…		…
404	=cut	593	=cut
405		594
406	sub aio_scandir($$$) {	595	sub aio_scandir($$$) {
407	my ($path, $maxreq, $cb) = @_;	596	my ($path, $maxreq, $cb) = @_;
408		597
		598	my $pri = aioreq_pri;
		599
409	my $grp = aio_group $cb;	600	my $grp = aio_group $cb;
410		601
411	$maxreq = 6 if $maxreq <= 0;	602	$maxreq = 4 if $maxreq <= 0;
412		603
413	# stat once	604	# stat once
		605	aioreq_pri $pri;
414	add $grp aio_stat $path, sub {	606	add $grp aio_stat $path, sub {
415	return $grp->result () if $_[0];	607	return $grp->result () if $_[0];
416	my $now = time;	608	my $now = time;
417	my $hash1 = join ":", (stat _)[0,1,3,7,9];	609	my $hash1 = join ":", (stat _)[0,1,3,7,9];
418		610
419	# read the directory entries	611	# read the directory entries
		612	aioreq_pri $pri;
420	add $grp aio_readdir $path, sub {	613	add $grp aio_readdir $path, sub {
421	my $entries = shift	614	my $entries = shift
422	or return $grp->result ();	615	or return $grp->result ();
423		616
424	# stat the dir another time	617	# stat the dir another time
		618	aioreq_pri $pri;
425	add $grp aio_stat $path, sub {	619	add $grp aio_stat $path, sub {
426	my $hash2 = join ":", (stat _)[0,1,3,7,9];	620	my $hash2 = join ":", (stat _)[0,1,3,7,9];
427		621
428	my $ndirs;	622	my $ndirs;
429		623
…		…
444	map [$_, sprintf "%s%04d", (/.\./ ? "1" : "0"), length],	638	map [$_, sprintf "%s%04d", (/.\./ ? "1" : "0"), length],
445	@$entries];	639	@$entries];
446		640
447	my (@dirs, @nondirs);	641	my (@dirs, @nondirs);
448		642
449	my ($statcb, $schedcb);
450	my $nreq = 0;
451
452	my $statgrp = add $grp aio_group;	643	my $statgrp = add $grp aio_group sub {
		644	$grp->result (\@dirs, \@nondirs);
		645	};
453		646
454	$schedcb = sub {	647	limit $statgrp $maxreq;
455	if (@$entries) {	648	feed $statgrp sub {
456	if ($nreq < $maxreq) {	649	return unless @$entries;
457	my $ent = pop @$entries;	650	my $entry = pop @$entries;
		651
		652	aioreq_pri $pri;
		653	add $statgrp aio_stat "$path/$entry/.", sub {
		654	if ($_[0] < 0) {
		655	push @nondirs, $entry;
		656	} else {
		657	# need to check for real directory
		658	aioreq_pri $pri;
		659	add $statgrp aio_lstat "$path/$entry", sub {
		660	if (-d _) {
		661	push @dirs, $entry;
		662
		663	unless (--$ndirs) {
		664	push @nondirs, @$entries;
		665	feed $statgrp;
		666	}
		667	} else {
		668	push @nondirs, $entry;
		669	}
458	$nreq++;	670	}
459	add $statgrp aio_stat "$path/$ent/.", sub { $statcb->($_[0], $ent) };
460	}	671	}
461	} elsif (!$nreq) {
462	# finished
463	$statgrp->cancel;
464	undef $statcb;
465	undef $schedcb;
466	$grp->result (\@dirs, \@nondirs);
467	}	672	};
468	};	673	};
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	};	674	};
499	};	675	};
500	};	676	};
501		677
502	$grp	678	$grp
…		…
515	If this call isn't available because your OS lacks it or it couldn't be	691	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.	692	detected, it will be emulated by calling C<fsync> instead.
517		693
518	=item aio_group $callback->(...)	694	=item aio_group $callback->(...)
519		695
520	[EXPERIMENTAL]
521
522	This is a very special aio request: Instead of doing something, it is a	696	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	697	container for other aio requests, which is useful if you want to bundle
524	many requests into a single, composite, request.	698	many requests into a single, composite, request with a definite callback
		699	and the ability to cancel the whole request with its subrequests.
525		700
526	Returns an object of class L<IO::AIO::GRP>. See its documentation below	701	Returns an object of class L<IO::AIO::GRP>. See its documentation below
527	for more info.	702	for more info.
528		703
529	Example:	704	Example:
…		…
548	phase and still requires a worker thread. Thus, the callback will not	723	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	724	be executed immediately but only after other requests in the queue have
550	entered their execution phase. This can be used to measure request	725	entered their execution phase. This can be used to measure request
551	latency.	726	latency.
552		727
553	=item IO::AIO::aio_sleep $fractional_seconds, $callback->() NOT EXPORTED	728	=item IO::AIO::aio_busy $fractional_seconds, $callback->() NOT EXPORTED
554		729
555	Mainly used for debugging and benchmarking, this aio request puts one of	730	Mainly used for debugging and benchmarking, this aio request puts one of
556	the request workers to sleep for the given time.	731	the request workers to sleep for the given time.
557		732
558	While it is theoretically handy to have simple I/O scheduling requests	733	While it is theoretically handy to have simple I/O scheduling requests
559	like sleep and file handle readable/writable, the overhead this creates	734	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	735	immense (it blocks a thread for a long time) so do not use this function
561	under artificial I/O pressure.	736	except to put your application under artificial I/O pressure.
562		737
563	=back	738	=back
564		739
565	=head2 IO::AIO::REQ CLASS	740	=head2 IO::AIO::REQ CLASS
566		741
567	All non-aggregate C<aio_*> functions return an object of this class when	742	All non-aggregate C<aio_*> functions return an object of this class when
568	called in non-void context.	743	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		744
578	=over 4	745	=over 4
579		746
580	=item cancel $req	747	=item cancel $req
581		748
…		…
635	=item * They can also can also be added to other IO::AIO::GRP objects.	802	=item * They can also can also be added to other IO::AIO::GRP objects.
636		803
637	=item * You must not add requests to a group from within the group callback (or	804	=item * You must not add requests to a group from within the group callback (or
638	any later time).	805	any later time).
639		806
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	807	=back
645		808
646	Their lifetime, simplified, looks like this: when they are empty, they	809	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	810	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	811	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	826	be added, including other groups, as long as you do not create circular
664	dependencies.	827	dependencies.
665		828
666	Returns all its arguments.	829	Returns all its arguments.
667		830
		831	=item $grp->cancel_subs
		832
		833	Cancel all subrequests and clears any feeder, but not the group request
		834	itself. Useful when you queued a lot of events but got a result early.
		835
668	=item $grp->result (...)	836	=item $grp->result (...)
669		837
670	Set the result value(s) that will be passed to the group callback when all	838	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.	839	subrequests have finished and set thre groups errno to the current value
		840	of errno (just like calling C<errno> without an error number). By default,
		841	no argument will be passed and errno is zero.
		842
		843	=item $grp->errno ([$errno])
		844
		845	Sets the group errno value to C<$errno>, or the current value of errno
		846	when the argument is missing.
		847
		848	Every aio request has an associated errno value that is restored when
		849	the callback is invoked. This method lets you change this value from its
		850	default (0).
		851
		852	Calling C<result> will also set errno, so make sure you either set C<$!>
		853	before the call to C<result>, or call c<errno> after it.
672		854
673	=item feed $grp $callback->($grp)	855	=item feed $grp $callback->($grp)
674
675	[VERY EXPERIMENTAL]
676		856
677	Sets a feeder/generator on this group: every group can have an attached	857	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,	858	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,	859	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	860	this might starve other requests for a potentially long time. For
681	example, C<aio_scandir> might generate hundreds of thousands C<aio_stat>	861	example, C<aio_scandir> might generate hundreds of thousands C<aio_stat>
682	requests, delaying any later requests for a long time.	862	requests, delaying any later requests for a long time.
683		863
684	To avoid this, and allow incremental generation of requests, you can	864	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	865	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>,	866	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	867	below) requests active in the group itself and is expected to queue more
688	requests.	868	requests.
689		869
690	The feed can queue as many requests as it likes (i.e. C<add> does not	870	The feed callback can queue as many requests as it likes (i.e. C<add> does
691	impose any limits).	871	not impose any limits).
692		872
693	If the feed does not queue more requests when called, it will be	873	If the feed does not queue more requests when called, it will be
694	automatically removed from the group.	874	automatically removed from the group.
695		875
696	If the feed limit is C<0>, it will be set to C<2> automatically.	876	If the feed limit is C<0>, it will be set to C<2> automatically.
…		…
698	Example:	878	Example:
699		879
700	# stat all files in @files, but only ever use four aio requests concurrently:	880	# stat all files in @files, but only ever use four aio requests concurrently:
701		881
702	my $grp = aio_group sub { print "finished\n" };	882	my $grp = aio_group sub { print "finished\n" };
703	feed_limit $grp 4;	883	limit $grp 4;
704	feed $grp sub {	884	feed $grp sub {
705	my $file = pop @files	885	my $file = pop @files
706	or return;	886	or return;
707		887
708	add $grp aio_stat $file, sub { ... };	888	add $grp aio_stat $file, sub { ... };
709	};	889	};
710		890
711	=item feed_limit $grp $num	891	=item limit $grp $num
712		892
713	Sets the feeder limit for the group: The feeder will be called whenever	893	Sets the feeder limit for the group: The feeder will be called whenever
714	the group contains less than this many requests.	894	the group contains less than this many requests.
715		895
716	Setting the limit to C<0> will pause the feeding process.	896	Setting the limit to C<0> will pause the feeding process.
717		897
718	=back	898	=back
719		899
720	=head2 SUPPORT FUNCTIONS	900	=head2 SUPPORT FUNCTIONS
		901
		902	=head3 EVENT PROCESSING AND EVENT LOOP INTEGRATION
721		903
722	=over 4	904	=over 4
723		905
724	=item $fileno = IO::AIO::poll_fileno	906	=item $fileno = IO::AIO::poll_fileno
725		907
…		…
730		912
731	See C<poll_cb> for an example.	913	See C<poll_cb> for an example.
732		914
733	=item IO::AIO::poll_cb	915	=item IO::AIO::poll_cb
734		916
735	Process all outstanding events on the result pipe. You have to call this	917	Process some outstanding events on the result pipe. You have to call this
736	regularly. Returns the number of events processed. Returns immediately	918	regularly. Returns the number of events processed. Returns immediately
737	when no events are outstanding.	919	when no events are outstanding. The amount of events processed depends on
		920	the settings of C<IO::AIO::max_poll_req> and C<IO::AIO::max_poll_time>.
		921
		922	If not all requests were processed for whatever reason, the filehandle
		923	will still be ready when C<poll_cb> returns.
738		924
739	Example: Install an Event watcher that automatically calls	925	Example: Install an Event watcher that automatically calls
740	IO::AIO::poll_cb with high priority:	926	IO::AIO::poll_cb with high priority:
741		927
742	Event->io (fd => IO::AIO::poll_fileno,	928	Event->io (fd => IO::AIO::poll_fileno,
743	poll => 'r', async => 1,	929	poll => 'r', async => 1,
744	cb => \&IO::AIO::poll_cb);	930	cb => \&IO::AIO::poll_cb);
745		931
		932	=item IO::AIO::max_poll_reqs $nreqs
		933
		934	=item IO::AIO::max_poll_time $seconds
		935
		936	These set the maximum number of requests (default C<0>, meaning infinity)
		937	that are being processed by C<IO::AIO::poll_cb> in one call, respectively
		938	the maximum amount of time (default C<0>, meaning infinity) spent in
		939	C<IO::AIO::poll_cb> to process requests (more correctly the mininum amount
		940	of time C<poll_cb> is allowed to use).
		941
		942	Setting these is useful if you want to ensure some level of
		943	interactiveness when perl is not fast enough to process all requests in
		944	time.
		945
		946	For interactive programs, values such as C<0.01> to C<0.1> should be fine.
		947
		948	Example: Install an Event watcher that automatically calls
		949	IO::AIO::poll_some with low priority, to ensure that other parts of the
		950	program get the CPU sometimes even under high AIO load.
		951
		952	# try not to spend much more than 0.1s in poll_cb
		953	IO::AIO::max_poll_time 0.1;
		954
		955	# use a low priority so other tasks have priority
		956	Event->io (fd => IO::AIO::poll_fileno,
		957	poll => 'r', nice => 1,
		958	cb => &IO::AIO::poll_cb);
		959
746	=item IO::AIO::poll_wait	960	=item IO::AIO::poll_wait
747		961
748	Wait till the result filehandle becomes ready for reading (simply does a	962	Wait till the result filehandle becomes ready for reading (simply does a
749	C<select> on the filehandle. This is useful if you want to synchronously wait	963	C<select> on the filehandle. This is useful if you want to synchronously
750	for some requests to finish).	964	wait for some requests to finish).
751		965
752	See C<nreqs> for an example.	966	See C<nreqs> for an example.
753		967
		968	=item IO::AIO::poll
		969
		970	Waits until some requests have been handled.
		971
		972	Strictly equivalent to:
		973
		974	IO::AIO::poll_wait, IO::AIO::poll_cb
		975	if IO::AIO::nreqs;
		976
754	=item IO::AIO::nreqs	977	=item IO::AIO::flush
755		978
756	Returns the number of requests currently outstanding (i.e. for which their	979	Wait till all outstanding AIO requests have been handled.
757	callback has not been invoked yet).
758		980
759	Example: wait till there are no outstanding requests anymore:	981	Strictly equivalent to:
760		982
761	IO::AIO::poll_wait, IO::AIO::poll_cb	983	IO::AIO::poll_wait, IO::AIO::poll_cb
762	while IO::AIO::nreqs;	984	while IO::AIO::nreqs;
763		985
764	=item IO::AIO::flush	986	=head3 CONTROLLING THE NUMBER OF THREADS
765
766	Wait till all outstanding AIO requests have been handled.
767
768	Strictly equivalent to:
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		987
782	=item IO::AIO::min_parallel $nthreads	988	=item IO::AIO::min_parallel $nthreads
783		989
784	Set the minimum number of AIO threads to C<$nthreads>. The current	990	Set the minimum number of AIO threads to C<$nthreads>. The current
785	default is C<8>, which means eight asynchronous operations can execute	991	default is C<8>, which means eight asynchronous operations can execute
786	concurrently at any one time (the number of outstanding requests,	992	concurrently at any one time (the number of outstanding requests,
787	however, is unlimited).	993	however, is unlimited).
788		994
789	IO::AIO starts threads only on demand, when an AIO request is queued and	995	IO::AIO starts threads only on demand, when an AIO request is queued and
790	no free thread exists.	996	no free thread exists. Please note that queueing up a hundred requests can
		997	create demand for a hundred threads, even if it turns out that everything
		998	is in the cache and could have been processed faster by a single thread.
791		999
792	It is recommended to keep the number of threads relatively low, as some	1000	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	1001	Linux kernel versions will scale negatively with the number of threads
794	(higher parallelity => MUCH higher latency). With current Linux 2.6	1002	(higher parallelity => MUCH higher latency). With current Linux 2.6
795	versions, 4-32 threads should be fine.	1003	versions, 4-32 threads should be fine.
…		…
809	This module automatically runs C<max_parallel 0> at program end, to ensure	1017	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.	1018	that all threads are killed and that there are no outstanding requests.
811		1019
812	Under normal circumstances you don't need to call this function.	1020	Under normal circumstances you don't need to call this function.
813		1021
		1022	=item IO::AIO::max_idle $nthreads
		1023
		1024	Limit the number of threads (default: 4) that are allowed to idle (i.e.,
		1025	threads that did not get a request to process within 10 seconds). That
		1026	means if a thread becomes idle while C<$nthreads> other threads are also
		1027	idle, it will free its resources and exit.
		1028
		1029	This is useful when you allow a large number of threads (e.g. 100 or 1000)
		1030	to allow for extremely high load situations, but want to free resources
		1031	under normal circumstances (1000 threads can easily consume 30MB of RAM).
		1032
		1033	The default is probably ok in most situations, especially if thread
		1034	creation is fast. If thread creation is very slow on your system you might
		1035	want to use larger values.
		1036
814	=item $oldnreqs = IO::AIO::max_outstanding $nreqs	1037	=item $oldmaxreqs = IO::AIO::max_outstanding $maxreqs
815		1038
816	[DEPRECATED]	1039	This is a very bad function to use in interactive programs because it
		1040	blocks, and a bad way to reduce concurrency because it is inexact: Better
		1041	use an C<aio_group> together with a feed callback.
817		1042
818	Sets the maximum number of outstanding requests to C<$nreqs>. If you	1043	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	1044	to queue up more than this number of requests, the next call to the
820	some requests have been handled.	1045	C<poll_cb> (and C<poll_some> and other functions calling C<poll_cb>)
		1046	function will block until the limit is no longer exceeded.
821		1047
822	The default is very large, so normally there is no practical limit. If you	1048	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	1049	number of outstanding requests.
824	this to a relatively low number, such as C<100>.
825		1050
826	This function does not work well together with C<aio_group>'s, and their	1051	You can still queue as many requests as you want. Therefore,
827	feeder interface is better suited to limiting concurrency, so do not use	1052	C<max_oustsanding> is mainly useful in simple scripts (with low values) or
828	this function.	1053	as a stop gap to shield against fatal memory overflow (with large values).
829		1054
830	Under normal circumstances you don't need to call this function.	1055	=head3 STATISTICAL INFORMATION
		1056
		1057	=item IO::AIO::nreqs
		1058
		1059	Returns the number of requests currently in the ready, execute or pending
		1060	states (i.e. for which their callback has not been invoked yet).
		1061
		1062	Example: wait till there are no outstanding requests anymore:
		1063
		1064	IO::AIO::poll_wait, IO::AIO::poll_cb
		1065	while IO::AIO::nreqs;
		1066
		1067	=item IO::AIO::nready
		1068
		1069	Returns the number of requests currently in the ready state (not yet
		1070	executed).
		1071
		1072	=item IO::AIO::npending
		1073
		1074	Returns the number of requests currently in the pending state (executed,
		1075	but not yet processed by poll_cb).
831		1076
832	=back	1077	=back
833		1078
834	=cut	1079	=cut
835		1080
…		…
850	}	1095	}
851		1096
852	min_parallel 8;	1097	min_parallel 8;
853		1098
854	END {	1099	END {
855	max_parallel 0;	1100	min_parallel 1;
856	}	1101	flush;
		1102	};
857		1103
858	1;	1104	1;
859		1105
860	=head2 FORK BEHAVIOUR	1106	=head2 FORK BEHAVIOUR
861		1107
862	This module should do "the right thing" when the process using it forks:	1108	This module should do "the right thing" when the process using it forks:
863		1109
864	Before the fork, IO::AIO enters a quiescent state where no requests	1110	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	1111	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	1112	the fork the parent simply leaves the quiescent state and continues
867	request/result processing, while the child clears the request/result	1113	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	1114	(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	1115	parent). Threads will be started on demand until the limit set in the
870	parent process has been reached again.	1116	parent process has been reached again.
871		1117
872	In short: the parent will, after a short pause, continue as if fork had	1118	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	1119	not been called, while the child will act as if IO::AIO has not been used
874	yet.	1120	yet.
875		1121
876	=head2 MEMORY USAGE	1122	=head2 MEMORY USAGE
877		1123
		1124	Per-request usage:
		1125
878	Each aio request uses - depending on your architecture - around 128 bytes	1126	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	1127	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	1128	a few hundred bytes), readdir requires a result buffer and so on. Perl
881	also be locked.	1129	scalars and other data passed into aio requests will also be locked and
		1130	will consume memory till the request has entered the done state.
882		1131
883	This is now awfully much, so queuing lots of requests is not usually a	1132	This is now awfully much, so queuing lots of requests is not usually a
884	problem.	1133	problem.
885		1134
886	Each thread needs a stack area which is usually around 16k, sometimes much	1135	Per-thread usage:
887	larger, depending on the OS.	1136
		1137	In the execution phase, some aio requests require more memory for
		1138	temporary buffers, and each thread requires a stack and other data
		1139	structures (usually around 16k-128k, depending on the OS).
		1140
		1141	=head1 KNOWN BUGS
		1142
		1143	Known bugs will be fixed in the next release.
888		1144
889	=head1 SEE ALSO	1145	=head1 SEE ALSO
890		1146
891	L<Coro>, L<Linux::AIO> (obsolete).	1147	L<Coro::AIO>.
892		1148
893	=head1 AUTHOR	1149	=head1 AUTHOR
894		1150
895	Marc Lehmann <schmorp@schmorp.de>	1151	Marc Lehmann <schmorp@schmorp.de>
896	http://home.schmorp.de/	1152	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 1.87 by root, Sun Oct 29 00:52:02 2006 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 1.87 by root, Sun Oct 29 00:52:02 2006 UTC