ViewVC Help

View File | Revision Log | Show Annotations | Download File

/cvs/IO-AIO/README

Comparing IO-AIO/README (file contents):
Revision 1.18 by root, Thu Oct 26 16:28:33 2006 UTC vs.
Revision 1.43 by root, Sun Jan 10 23:44:02 2010 UTC

…		…
3		3
4	SYNOPSIS	4	SYNOPSIS
5	use IO::AIO;	5	use IO::AIO;
6		6
7	aio_open "/etc/passwd", O_RDONLY, 0, sub {	7	aio_open "/etc/passwd", O_RDONLY, 0, sub {
8	my ($fh) = @_;	8	my $fh = shift
		9	or die "/etc/passwd: $!";
9	...	10	...
10	};	11	};
11		12
12	aio_unlink "/tmp/file", sub { };	13	aio_unlink "/tmp/file", sub { };
13		14
…		…
23	$req->cancel; # cancel request if still in queue	24	$req->cancel; # cancel request if still in queue
24		25
25	my $grp = aio_group sub { print "all stats done\n" };	26	my $grp = aio_group sub { print "all stats done\n" };
26	add $grp aio_stat "..." for ...;	27	add $grp aio_stat "..." for ...;
27		28
28	# AnyEvent integration
29	open my $fh, "<&=" . IO::AIO::poll_fileno or die "$!";
30	my $w = AnyEvent->io (fh => $fh, poll => 'r', cb => sub { IO::AIO::poll_cb });
31
32	# Event integration
33	Event->io (fd => IO::AIO::poll_fileno,
34	poll => 'r',
35	cb => \&IO::AIO::poll_cb);
36
37	# Glib/Gtk2 integration
38	add_watch Glib::IO IO::AIO::poll_fileno,
39	in => sub { IO::AIO::poll_cb; 1 };
40
41	# Tk integration
42	Tk::Event::IO->fileevent (IO::AIO::poll_fileno, "",
43	readable => \&IO::AIO::poll_cb);
44
45	# Danga::Socket integration
46	Danga::Socket->AddOtherFds (IO::AIO::poll_fileno =>
47	\&IO::AIO::poll_cb);
48
49	DESCRIPTION	29	DESCRIPTION
50	This module implements asynchronous I/O using whatever means your	30	This module implements asynchronous I/O using whatever means your
51	operating system supports.	31	operating system supports. It is implemented as an interface to "libeio"
		32	(<http://software.schmorp.de/pkg/libeio.html>).
		33
		34	Asynchronous means that operations that can normally block your program
		35	(e.g. reading from disk) will be done asynchronously: the operation will
		36	still block, but you can do something else in the meantime. This is
		37	extremely useful for programs that need to stay interactive even when
		38	doing heavy I/O (GUI programs, high performance network servers etc.),
		39	but can also be used to easily do operations in parallel that are
		40	normally done sequentially, e.g. stat'ing many files, which is much
		41	faster on a RAID volume or over NFS when you do a number of stat
		42	operations concurrently.
		43
		44	While most of this works on all types of file descriptors (for example
		45	sockets), using these functions on file descriptors that support
		46	nonblocking operation (again, sockets, pipes etc.) is very inefficient.
		47	Use an event loop for that (such as the EV module): IO::AIO will
		48	naturally fit into such an event loop itself.
52		49
53	In this version, a number of threads are started that execute your	50	In this version, a number of threads are started that execute your
54	requests and signal their completion. You don't need thread support in	51	requests and signal their completion. You don't need thread support in
55	perl, and the threads created by this module will not be visible to	52	perl, and the threads created by this module will not be visible to
56	perl. In the future, this module might make use of the native aio	53	perl. In the future, this module might make use of the native aio
57	functions available on many operating systems. However, they are often	54	functions available on many operating systems. However, they are often
58	not well-supported or restricted (Linux doesn't allow them on normal	55	not well-supported or restricted (GNU/Linux doesn't allow them on normal
59	files currently, for example), and they would only support aio_read and	56	files currently, for example), and they would only support aio_read and
60	aio_write, so the remaining functionality would have to be implemented	57	aio_write, so the remaining functionality would have to be implemented
61	using threads anyway.	58	using threads anyway.
62		59
63	Although the module will work with in the presence of other (Perl-)	60	Although the module will work in the presence of other (Perl-) threads,
64	threads, it is currently not reentrant in any way, so use appropriate	61	it is currently not reentrant in any way, so use appropriate locking
65	locking yourself, always call "poll_cb" from within the same thread, or	62	yourself, always call "poll_cb" from within the same thread, or never
66	never call "poll_cb" (or other "aio_" functions) recursively.	63	call "poll_cb" (or other "aio_" functions) recursively.
		64
		65	EXAMPLE
		66	This is a simple example that uses the EV module and loads /etc/passwd
		67	asynchronously:
		68
		69	use Fcntl;
		70	use EV;
		71	use IO::AIO;
		72
		73	# register the IO::AIO callback with EV
		74	my $aio_w = EV::io IO::AIO::poll_fileno, EV::READ, \&IO::AIO::poll_cb;
		75
		76	# queue the request to open /etc/passwd
		77	aio_open "/etc/passwd", O_RDONLY, 0, sub {
		78	my $fh = shift
		79	or die "error while opening: $!";
		80
		81	# stat'ing filehandles is generally non-blocking
		82	my $size = -s $fh;
		83
		84	# queue a request to read the file
		85	my $contents;
		86	aio_read $fh, 0, $size, $contents, 0, sub {
		87	$_[0] == $size
		88	or die "short read: $!";
		89
		90	close $fh;
		91
		92	# file contents now in $contents
		93	print $contents;
		94
		95	# exit event loop and program
		96	EV::unloop;
		97	};
		98	};
		99
		100	# possibly queue up other requests, or open GUI windows,
		101	# check for sockets etc. etc.
		102
		103	# process events as long as there are some:
		104	EV::loop;
67		105
68	REQUEST ANATOMY AND LIFETIME	106	REQUEST ANATOMY AND LIFETIME
69	Every "aio_*" function creates a request. which is a C data structure	107	Every "aio_*" function creates a request. which is a C data structure
70	not directly visible to Perl.	108	not directly visible to Perl.
71		109
…		…
107	anymore (except possibly for the Perl object, but its connection to	145	anymore (except possibly for the Perl object, but its connection to
108	the actual aio request is severed and calling its methods will	146	the actual aio request is severed and calling its methods will
109	either do nothing or result in a runtime error).	147	either do nothing or result in a runtime error).
110		148
111	FUNCTIONS	149	FUNCTIONS
112	AIO FUNCTIONS	150	QUICK OVERVIEW
113	All the "aio_*" calls are more or less thin wrappers around the	151	This section simply lists the prototypes of the most important functions
114	syscall with the same name (sans "aio_"). The arguments are similar	152	for quick reference. See the following sections for function-by-function
115	or identical, and they all accept an additional (and optional)	153	documentation.
116	$callback argument which must be a code reference. This code
117	reference will get called with the syscall return code (e.g. most
118	syscalls return -1 on error, unlike perl, which usually delivers
119	"false") as it's sole argument when the given syscall has been
120	executed asynchronously.
121		154
122	All functions expecting a filehandle keep a copy of the filehandle
123	internally until the request has finished.
124
125	All requests return objects of type IO::AIO::REQ that allow further
126	manipulation of those requests while they are in-flight.
127
128	The pathnames you pass to these routines *must* be absolute and
129	encoded in byte form. The reason for the former is that at the time
130	the request is being executed, the current working directory could
131	have changed. Alternatively, you can make sure that you never change
132	the current working directory.
133
134	To encode pathnames to byte form, either make sure you either: a)
135	always pass in filenames you got from outside (command line, readdir
136	etc.), b) are ASCII or ISO 8859-1, c) use the Encode module and
137	encode your pathnames to the locale (or other) encoding in effect in
138	the user environment, d) use Glib::filename_from_unicode on unicode
139	filenames or e) use something else.
140
141	$prev_pri = aioreq_pri [$pri]
142	Returns the priority value that would be used for the next
143	request and, if $pri is given, sets the priority for the next
144	aio request.
145
146	The default priority is 0, the minimum and maximum priorities
147	are -4 and 4, respectively. Requests with higher priority will
148	be serviced first.
149
150	The priority will be reset to 0 after each call to one of the
151	"aio_*" functions.
152
153	Example: open a file with low priority, then read something from
154	it with higher priority so the read request is serviced before
155	other low priority open requests (potentially spamming the
156	cache):
157
158	aioreq_pri -3;
159	aio_open ..., sub {
160	return unless $_[0];
161
162	aioreq_pri -2;
163	aio_read $_[0], ..., sub {
164	...
165	};
166	};
167
168	aioreq_nice $pri_adjust
169	Similar to "aioreq_pri", but subtracts the given value from the
170	current priority, so effects are cumulative.
171
172	aio_open $pathname, $flags, $mode, $callback->($fh)	155	aio_open $pathname, $flags, $mode, $callback->($fh)
173	Asynchronously open or create a file and call the callback with
174	a newly created filehandle for the file.
175
176	The pathname passed to "aio_open" must be absolute. See API
177	NOTES, above, for an explanation.
178
179	The $flags argument is a bitmask. See the "Fcntl" module for a
180	list. They are the same as used by "sysopen".
181
182	Likewise, $mode specifies the mode of the newly created file, if
183	it didn't exist and "O_CREAT" has been given, just like perl's
184	"sysopen", except that it is mandatory (i.e. use 0 if you don't
185	create new files, and 0666 or 0777 if you do).
186
187	Example:
188
189	aio_open "/etc/passwd", O_RDONLY, 0, sub {
190	if ($_[0]) {
191	print "open successful, fh is $_[0]\n";
192	...
193	} else {
194	die "open failed: $!\n";
195	}
196	};
197
198	aio_close $fh, $callback->($status)	156	aio_close $fh, $callback->($status)
199	Asynchronously close a file and call the callback with the
200	result code. *WARNING:* although accepted, you should not pass
201	in a perl filehandle here, as perl will likely close the file
202	descriptor another time when the filehandle is destroyed.
203	Normally, you can safely call perls "close" or just let
204	filehandles go out of scope.
205
206	This is supposed to be a bug in the API, so that might change.
207	It's therefore best to avoid this function.
208
209	aio_read $fh,$offset,$length, $data,$dataoffset,	157	aio_read $fh,$offset,$length, $data,$dataoffset, $callback->($retval)
210	$callback->($retval)
211	aio_write $fh,$offset,$length, $data,$dataoffset,	158	aio_write $fh,$offset,$length, $data,$dataoffset, $callback->($retval)
212	$callback->($retval)
213	Reads or writes "length" bytes from the specified "fh" and
214	"offset" into the scalar given by "data" and offset "dataoffset"
215	and calls the callback without the actual number of bytes read
216	(or -1 on error, just like the syscall).
217
218	The $data scalar *MUST NOT* be modified in any way while the
219	request is outstanding. Modifying it can result in segfaults or
220	WW3 (if the necessary/optional hardware is installed).
221
222	Example: Read 15 bytes at offset 7 into scalar $buffer, starting
223	at offset 0 within the scalar:
224
225	aio_read $fh, 7, 15, $buffer, 0, sub {
226	$_[0] > 0 or die "read error: $!";
227	print "read $_[0] bytes: <$buffer>\n";
228	};
229
230	aio_move $srcpath, $dstpath, $callback->($status)
231	Try to move the *file* (directories not supported as either
232	source or destination) from $srcpath to $dstpath and call the
233	callback with the 0 (error) or -1 ok.
234
235	This is a composite request that tries to rename(2) the file
236	first. If rename files with "EXDEV", it creates the destination
237	file with mode 0200 and copies the contents of the source file
238	into it using "aio_sendfile", followed by restoring atime,
239	mtime, access mode and uid/gid, in that order, and unlinking the
240	$srcpath.
241
242	If an error occurs, the partial destination file will be
243	unlinked, if possible, except when setting atime, mtime, access
244	mode and uid/gid, where errors are being ignored.
245
246	aio_sendfile $out_fh, $in_fh, $in_offset, $length,	159	aio_sendfile $out_fh, $in_fh, $in_offset, $length, $callback->($retval)
247	$callback->($retval)
248	Tries to copy $length bytes from $in_fh to $out_fh. It starts
249	reading at byte offset $in_offset, and starts writing at the
250	current file offset of $out_fh. Because of that, it is not safe
251	to issue more than one "aio_sendfile" per $out_fh, as they will
252	interfere with each other.
253
254	This call tries to make use of a native "sendfile" syscall to
255	provide zero-copy operation. For this to work, $out_fh should
256	refer to a socket, and $in_fh should refer to mmap'able file.
257
258	If the native sendfile call fails or is not implemented, it will
259	be emulated, so you can call "aio_sendfile" on any type of
260	filehandle regardless of the limitations of the operating
261	system.
262
263	Please note, however, that "aio_sendfile" can read more bytes
264	from $in_fh than are written, and there is no way to find out
265	how many bytes have been read from "aio_sendfile" alone, as
266	"aio_sendfile" only provides the number of bytes written to
267	$out_fh. Only if the result value equals $length one can assume
268	that $length bytes have been read.
269
270	aio_readahead $fh,$offset,$length, $callback->($retval)	160	aio_readahead $fh,$offset,$length, $callback->($retval)
271	"aio_readahead" populates the page cache with data from a file
272	so that subsequent reads from that file will not block on disk
273	I/O. The $offset argument specifies the starting point from
274	which data is to be read and $length specifies the number of
275	bytes to be read. I/O is performed in whole pages, so that
276	offset is effectively rounded down to a page boundary and bytes
277	are read up to the next page boundary greater than or equal to
278	(off-set+length). "aio_readahead" does not read beyond the end
279	of the file. The current file offset of the file is left
280	unchanged.
281
282	If that syscall doesn't exist (likely if your OS isn't Linux) it
283	will be emulated by simply reading the data, which would have a
284	similar effect.
285
286	aio_stat $fh_or_path, $callback->($status)	161	aio_stat $fh_or_path, $callback->($status)
287	aio_lstat $fh, $callback->($status)	162	aio_lstat $fh, $callback->($status)
288	Works like perl's "stat" or "lstat" in void context. The	163	aio_statvfs $fh_or_path, $callback->($statvfs)
289	callback will be called after the stat and the results will be	164	aio_utime $fh_or_path, $atime, $mtime, $callback->($status)
290	available using "stat _" or "-s _" etc...	165	aio_chown $fh_or_path, $uid, $gid, $callback->($status)
291		166	aio_truncate $fh_or_path, $offset, $callback->($status)
292	The pathname passed to "aio_stat" must be absolute. See API	167	aio_chmod $fh_or_path, $mode, $callback->($status)
293	NOTES, above, for an explanation.
294
295	Currently, the stats are always 64-bit-stats, i.e. instead of
296	returning an error when stat'ing a large file, the results will
297	be silently truncated unless perl itself is compiled with large
298	file support.
299
300	Example: Print the length of /etc/passwd:
301
302	aio_stat "/etc/passwd", sub {
303	$_[0] and die "stat failed: $!";
304	print "size is ", -s _, "\n";
305	};
306
307	aio_unlink $pathname, $callback->($status)	168	aio_unlink $pathname, $callback->($status)
308	Asynchronously unlink (delete) a file and call the callback with	169	aio_mknod $path, $mode, $dev, $callback->($status)
309	the result code.
310
311	aio_link $srcpath, $dstpath, $callback->($status)	170	aio_link $srcpath, $dstpath, $callback->($status)
312	Asynchronously create a new link to the existing object at
313	$srcpath at the path $dstpath and call the callback with the
314	result code.
315
316	aio_symlink $srcpath, $dstpath, $callback->($status)	171	aio_symlink $srcpath, $dstpath, $callback->($status)
317	Asynchronously create a new symbolic link to the existing object	172	aio_readlink $path, $callback->($link)
318	at $srcpath at the path $dstpath and call the callback with the
319	result code.
320
321	aio_rename $srcpath, $dstpath, $callback->($status)	173	aio_rename $srcpath, $dstpath, $callback->($status)
322	Asynchronously rename the object at $srcpath to $dstpath, just	174	aio_mkdir $pathname, $mode, $callback->($status)
323	as rename(2) and call the callback with the result code.
324
325	aio_rmdir $pathname, $callback->($status)	175	aio_rmdir $pathname, $callback->($status)
326	Asynchronously rmdir (delete) a directory and call the callback
327	with the result code.
328
329	aio_readdir $pathname, $callback->($entries)	176	aio_readdir $pathname, $callback->($entries)
330	Unlike the POSIX call of the same name, "aio_readdir" reads an	177	aio_readdirx $pathname, $flags, $callback->($entries, $flags)
331	entire directory (i.e. opendir + readdir + closedir). The	178	IO::AIO::READDIR_DENTS IO::AIO::READDIR_DIRS_FIRST
332	entries will not be sorted, and will NOT include the "." and	179	IO::AIO::READDIR_STAT_ORDER IO::AIO::READDIR_FOUND_UNKNOWN
333	".." entries.	180	aio_load $path, $data, $callback->($status)
334		181	aio_copy $srcpath, $dstpath, $callback->($status)
335	The callback a single argument which is either "undef" or an	182	aio_move $srcpath, $dstpath, $callback->($status)
336	array-ref with the filenames.
337
338	aio_scandir $path, $maxreq, $callback->($dirs, $nondirs)	183	aio_scandir $path, $maxreq, $callback->($dirs, $nondirs)
339	Scans a directory (similar to "aio_readdir") but additionally	184	aio_rmtree $path, $callback->($status)
340	tries to efficiently separate the entries of directory $path	185	aio_sync $callback->($status)
341	into two sets of names, directories you can recurse into
342	(directories), and ones you cannot recurse into (everything
343	else, including symlinks to directories).
344
345	"aio_scandir" is a composite request that creates of many sub
346	requests_ $maxreq specifies the maximum number of outstanding
347	aio requests that this function generates. If it is "<= 0", then
348	a suitable default will be chosen (currently 6).
349
350	On error, the callback is called without arguments, otherwise it
351	receives two array-refs with path-relative entry names.
352
353	Example:
354
355	aio_scandir $dir, 0, sub {
356	my ($dirs, $nondirs) = @_;
357	print "real directories: @$dirs\n";
358	print "everything else: @$nondirs\n";
359	};
360
361	Implementation notes.
362
363	The "aio_readdir" cannot be avoided, but "stat()"'ing every
364	entry can.
365
366	After reading the directory, the modification time, size etc. of
367	the directory before and after the readdir is checked, and if
368	they match (and isn't the current time), the link count will be
369	used to decide how many entries are directories (if >= 2).
370	Otherwise, no knowledge of the number of subdirectories will be
371	assumed.
372
373	Then entries will be sorted into likely directories (everything
374	without a non-initial dot currently) and likely non-directories
375	(everything else). Then every entry plus an appended "/." will
376	be "stat"'ed, likely directories first. If that succeeds, it
377	assumes that the entry is a directory or a symlink to directory
378	(which will be checked seperately). This is often faster than
379	stat'ing the entry itself because filesystems might detect the
380	type of the entry without reading the inode data (e.g. ext2fs
381	filetype feature).
382
383	If the known number of directories (link count - 2) has been
384	reached, the rest of the entries is assumed to be
385	non-directories.
386
387	This only works with certainty on POSIX (= UNIX) filesystems,
388	which fortunately are the vast majority of filesystems around.
389
390	It will also likely work on non-POSIX filesystems with reduced
391	efficiency as those tend to return 0 or 1 as link counts, which
392	disables the directory counting heuristic.
393
394	aio_fsync $fh, $callback->($status)	186	aio_fsync $fh, $callback->($status)
395	Asynchronously call fsync on the given filehandle and call the
396	callback with the fsync result code.
397
398	aio_fdatasync $fh, $callback->($status)	187	aio_fdatasync $fh, $callback->($status)
399	Asynchronously call fdatasync on the given filehandle and call	188	aio_sync_file_range $fh, $offset, $nbytes, $flags, $callback->($status)
400	the callback with the fdatasync result code.	189	aio_pathsync $path, $callback->($status)
401		190	aio_msync $scalar, $offset = 0, $length = undef, flags = 0, $callback->($status)
402	If this call isn't available because your OS lacks it or it	191	aio_mtouch $scalar, $offset = 0, $length = undef, flags = 0, $callback->($status)
403	couldn't be detected, it will be emulated by calling "fsync"
404	instead.
405
406	aio_group $callback->(...)	192	aio_group $callback->(...)
407	This is a very special aio request: Instead of doing something,
408	it is a container for other aio requests, which is useful if you
409	want to bundle many requests into a single, composite, request
410	with a definite callback and the ability to cancel the whole
411	request with its subrequests.
412
413	Returns an object of class IO::AIO::GRP. See its documentation
414	below for more info.
415
416	Example:
417
418	my $grp = aio_group sub {
419	print "all stats done\n";
420	};
421
422	add $grp
423	(aio_stat ...),
424	(aio_stat ...),
425	...;
426
427	aio_nop $callback->()	193	aio_nop $callback->()
428	This is a special request - it does nothing in itself and is
429	only used for side effects, such as when you want to add a dummy
430	request to a group so that finishing the requests in the group
431	depends on executing the given code.
432		194
433	While this request does nothing, it still goes through the	195	$prev_pri = aioreq_pri [$pri]
434	execution phase and still requires a worker thread. Thus, the	196	aioreq_nice $pri_adjust
435	callback will not be executed immediately but only after other
436	requests in the queue have entered their execution phase. This
437	can be used to measure request latency.
438		197
439	IO::AIO::aio_busy $fractional_seconds, $callback->() *NOT EXPORTED*	198	IO::AIO::poll_wait
440	Mainly used for debugging and benchmarking, this aio request	199	IO::AIO::poll_cb
441	puts one of the request workers to sleep for the given time.	200	IO::AIO::poll
		201	IO::AIO::flush
		202	IO::AIO::max_poll_reqs $nreqs
		203	IO::AIO::max_poll_time $seconds
		204	IO::AIO::min_parallel $nthreads
		205	IO::AIO::max_parallel $nthreads
		206	IO::AIO::max_idle $nthreads
		207	IO::AIO::max_outstanding $maxreqs
		208	IO::AIO::nreqs
		209	IO::AIO::nready
		210	IO::AIO::npending
442		211
443	While it is theoretically handy to have simple I/O scheduling	212	IO::AIO::sendfile $ofh, $ifh, $offset, $count
444	requests like sleep and file handle readable/writable, the	213	IO::AIO::fadvise $fh, $offset, $len, $advice
445	overhead this creates is immense (it blocks a thread for a long	214	IO::AIO::mlockall $flags
446	time) so do not use this function except to put your application	215	IO::AIO::munlockall
447	under artificial I/O pressure.
448		216
449	IO::AIO::REQ CLASS	217	AIO REQUEST FUNCTIONS
450	All non-aggregate "aio_*" functions return an object of this class	218	All the "aio_*" calls are more or less thin wrappers around the syscall
451	when called in non-void context.	219	with the same name (sans "aio_"). The arguments are similar or
		220	identical, and they all accept an additional (and optional) $callback
		221	argument which must be a code reference. This code reference will get
		222	called with the syscall return code (e.g. most syscalls return -1 on
		223	error, unlike perl, which usually delivers "false") as its sole argument
		224	after the given syscall has been executed asynchronously.
452		225
453	cancel $req	226	All functions expecting a filehandle keep a copy of the filehandle
454	Cancels the request, if possible. Has the effect of skipping	227	internally until the request has finished.
455	execution when entering the execute state and skipping calling
456	the callback when entering the the result state, but will leave
457	the request otherwise untouched. That means that requests that
458	currently execute will not be stopped and resources held by the
459	request will not be freed prematurely.
460		228
461	cb $req $callback->(...)	229	All functions return request objects of type IO::AIO::REQ that allow
462	Replace (or simply set) the callback registered to the request.	230	further manipulation of those requests while they are in-flight.
463		231
464	IO::AIO::GRP CLASS	232	The pathnames you pass to these routines *must* be absolute and encoded
465	This class is a subclass of IO::AIO::REQ, so all its methods apply	233	as octets. The reason for the former is that at the time the request is
466	to objects of this class, too.	234	being executed, the current working directory could have changed.
		235	Alternatively, you can make sure that you never change the current
		236	working directory anywhere in the program and then use relative paths.
467		237
468	A IO::AIO::GRP object is a special request that can contain multiple	238	To encode pathnames as octets, either make sure you either: a) always
469	other aio requests.	239	pass in filenames you got from outside (command line, readdir etc.)
		240	without tinkering, b) are ASCII or ISO 8859-1, c) use the Encode module
		241	and encode your pathnames to the locale (or other) encoding in effect in
		242	the user environment, d) use Glib::filename_from_unicode on unicode
		243	filenames or e) use something else to ensure your scalar has the correct
		244	contents.
470		245
471	You create one by calling the "aio_group" constructing function with	246	This works, btw. independent of the internal UTF-8 bit, which IO::AIO
472	a callback that will be called when all contained requests have	247	handles correctly whether it is set or not.
473	entered the "done" state:
474		248
475	my $grp = aio_group sub {	249	$prev_pri = aioreq_pri [$pri]
476	print "all requests are done\n";	250	Returns the priority value that would be used for the next request
477	};	251	and, if $pri is given, sets the priority for the next aio request.
478		252
479	You add requests by calling the "add" method with one or more	253	The default priority is 0, the minimum and maximum priorities are -4
480	"IO::AIO::REQ" objects:	254	and 4, respectively. Requests with higher priority will be serviced
		255	first.
481		256
482	$grp->add (aio_unlink "...");	257	The priority will be reset to 0 after each call to one of the
		258	"aio_*" functions.
483		259
484	add $grp aio_stat "...", sub {	260	Example: open a file with low priority, then read something from it
485	$_[0] or return $grp->result ("error");	261	with higher priority so the read request is serviced before other
		262	low priority open requests (potentially spamming the cache):
486		263
487	# add another request dynamically, if first succeeded	264	aioreq_pri -3;
488	add $grp aio_open "...", sub {	265	aio_open ..., sub {
489	$grp->result ("ok");	266	return unless $_[0];
		267
		268	aioreq_pri -2;
		269	aio_read $_[0], ..., sub {
		270	...
490	};	271	};
491	};	272	};
492		273
		274	aioreq_nice $pri_adjust
		275	Similar to "aioreq_pri", but subtracts the given value from the
		276	current priority, so the effect is cumulative.
		277
		278	aio_open $pathname, $flags, $mode, $callback->($fh)
		279	Asynchronously open or create a file and call the callback with a
		280	newly created filehandle for the file.
		281
		282	The pathname passed to "aio_open" must be absolute. See API NOTES,
		283	above, for an explanation.
		284
		285	The $flags argument is a bitmask. See the "Fcntl" module for a list.
		286	They are the same as used by "sysopen".
		287
		288	Likewise, $mode specifies the mode of the newly created file, if it
		289	didn't exist and "O_CREAT" has been given, just like perl's
		290	"sysopen", except that it is mandatory (i.e. use 0 if you don't
		291	create new files, and 0666 or 0777 if you do). Note that the $mode
		292	will be modified by the umask in effect then the request is being
		293	executed, so better never change the umask.
		294
		295	Example:
		296
		297	aio_open "/etc/passwd", O_RDONLY, 0, sub {
		298	if ($_[0]) {
		299	print "open successful, fh is $_[0]\n";
		300	...
		301	} else {
		302	die "open failed: $!\n";
		303	}
		304	};
		305
		306	aio_close $fh, $callback->($status)
		307	Asynchronously close a file and call the callback with the result
		308	code.
		309
		310	Unfortunately, you can't do this to perl. Perl *insists* very
		311	strongly on closing the file descriptor associated with the
		312	filehandle itself.
		313
		314	Therefore, "aio_close" will not close the filehandle - instead it
		315	will use dup2 to overwrite the file descriptor with the write-end of
		316	a pipe (the pipe fd will be created on demand and will be cached).
		317
		318	Or in other words: the file descriptor will be closed, but it will
		319	not be free for reuse until the perl filehandle is closed.
		320
		321	aio_read $fh,$offset,$length, $data,$dataoffset, $callback->($retval)
		322	aio_write $fh,$offset,$length, $data,$dataoffset, $callback->($retval)
		323	Reads or writes $length bytes from or to the specified $fh and
		324	$offset into the scalar given by $data and offset $dataoffset and
		325	calls the callback without the actual number of bytes read (or -1 on
		326	error, just like the syscall).
		327
		328	"aio_read" will, like "sysread", shrink or grow the $data scalar to
		329	offset plus the actual number of bytes read.
		330
		331	If $offset is undefined, then the current file descriptor offset
		332	will be used (and updated), otherwise the file descriptor offset
		333	will not be changed by these calls.
		334
		335	If $length is undefined in "aio_write", use the remaining length of
		336	$data.
		337
		338	If $dataoffset is less than zero, it will be counted from the end of
		339	$data.
		340
		341	The $data scalar *MUST NOT* be modified in any way while the request
		342	is outstanding. Modifying it can result in segfaults or World War
		343	III (if the necessary/optional hardware is installed).
		344
		345	Example: Read 15 bytes at offset 7 into scalar $buffer, starting at
		346	offset 0 within the scalar:
		347
		348	aio_read $fh, 7, 15, $buffer, 0, sub {
		349	$_[0] > 0 or die "read error: $!";
		350	print "read $_[0] bytes: <$buffer>\n";
		351	};
		352
		353	aio_sendfile $out_fh, $in_fh, $in_offset, $length, $callback->($retval)
		354	Tries to copy $length bytes from $in_fh to $out_fh. It starts
		355	reading at byte offset $in_offset, and starts writing at the current
		356	file offset of $out_fh. Because of that, it is not safe to issue
		357	more than one "aio_sendfile" per $out_fh, as they will interfere
		358	with each other.
		359
		360	This call tries to make use of a native "sendfile" syscall to
		361	provide zero-copy operation. For this to work, $out_fh should refer
		362	to a socket, and $in_fh should refer to an mmap'able file.
		363
		364	If a native sendfile cannot be found or it fails with "ENOSYS",
		365	"ENOTSUP", "EOPNOTSUPP", "EAFNOSUPPORT", "EPROTOTYPE" or "ENOTSOCK",
		366	it will be emulated, so you can call "aio_sendfile" on any type of
		367	filehandle regardless of the limitations of the operating system.
		368
		369	Please note, however, that "aio_sendfile" can read more bytes from
		370	$in_fh than are written, and there is no way to find out how many
		371	bytes have been read from "aio_sendfile" alone, as "aio_sendfile"
		372	only provides the number of bytes written to $out_fh. Only if the
		373	result value equals $length one can assume that $length bytes have
		374	been read.
		375
		376	aio_readahead $fh,$offset,$length, $callback->($retval)
		377	"aio_readahead" populates the page cache with data from a file so
		378	that subsequent reads from that file will not block on disk I/O. The
		379	$offset argument specifies the starting point from which data is to
		380	be read and $length specifies the number of bytes to be read. I/O is
		381	performed in whole pages, so that offset is effectively rounded down
		382	to a page boundary and bytes are read up to the next page boundary
		383	greater than or equal to (off-set+length). "aio_readahead" does not
		384	read beyond the end of the file. The current file offset of the file
		385	is left unchanged.
		386
		387	If that syscall doesn't exist (likely if your OS isn't Linux) it
		388	will be emulated by simply reading the data, which would have a
		389	similar effect.
		390
		391	aio_stat $fh_or_path, $callback->($status)
		392	aio_lstat $fh, $callback->($status)
		393	Works like perl's "stat" or "lstat" in void context. The callback
		394	will be called after the stat and the results will be available
		395	using "stat _" or "-s _" etc...
		396
		397	The pathname passed to "aio_stat" must be absolute. See API NOTES,
		398	above, for an explanation.
		399
		400	Currently, the stats are always 64-bit-stats, i.e. instead of
		401	returning an error when stat'ing a large file, the results will be
		402	silently truncated unless perl itself is compiled with large file
		403	support.
		404
		405	Example: Print the length of /etc/passwd:
		406
		407	aio_stat "/etc/passwd", sub {
		408	$_[0] and die "stat failed: $!";
		409	print "size is ", -s _, "\n";
		410	};
		411
		412	aio_statvfs $fh_or_path, $callback->($statvfs)
		413	Works like the POSIX "statvfs" or "fstatvfs" syscalls, depending on
		414	whether a file handle or path was passed.
		415
		416	On success, the callback is passed a hash reference with the
		417	following members: "bsize", "frsize", "blocks", "bfree", "bavail",
		418	"files", "ffree", "favail", "fsid", "flag" and "namemax". On
		419	failure, "undef" is passed.
		420
		421	The following POSIX IO::AIO::ST_* constants are defined: "ST_RDONLY"
		422	and "ST_NOSUID".
		423
		424	The following non-POSIX IO::AIO::ST_* flag masks are defined to
		425	their correct value when available, or to 0 on systems that do not
		426	support them: "ST_NODEV", "ST_NOEXEC", "ST_SYNCHRONOUS",
		427	"ST_MANDLOCK", "ST_WRITE", "ST_APPEND", "ST_IMMUTABLE",
		428	"ST_NOATIME", "ST_NODIRATIME" and "ST_RELATIME".
		429
		430	Example: stat "/wd" and dump out the data if successful.
		431
		432	aio_statvfs "/wd", sub {
		433	my $f = $_[0]
		434	or die "statvfs: $!";
		435
		436	use Data::Dumper;
		437	say Dumper $f;
		438	};
		439
		440	# result:
		441	{
		442	bsize => 1024,
		443	bfree => 4333064312,
		444	blocks => 10253828096,
		445	files => 2050765568,
		446	flag => 4096,
		447	favail => 2042092649,
		448	bavail => 4333064312,
		449	ffree => 2042092649,
		450	namemax => 255,
		451	frsize => 1024,
		452	fsid => 1810
		453	}
		454
		455	aio_utime $fh_or_path, $atime, $mtime, $callback->($status)
		456	Works like perl's "utime" function (including the special case of
		457	$atime and $mtime being undef). Fractional times are supported if
		458	the underlying syscalls support them.
		459
		460	When called with a pathname, uses utimes(2) if available, otherwise
		461	utime(2). If called on a file descriptor, uses futimes(2) if
		462	available, otherwise returns ENOSYS, so this is not portable.
		463
		464	Examples:
		465
		466	# set atime and mtime to current time (basically touch(1)):
		467	aio_utime "path", undef, undef;
		468	# set atime to current time and mtime to beginning of the epoch:
		469	aio_utime "path", time, undef; # undef==0
		470
		471	aio_chown $fh_or_path, $uid, $gid, $callback->($status)
		472	Works like perl's "chown" function, except that "undef" for either
		473	$uid or $gid is being interpreted as "do not change" (but -1 can
		474	also be used).
		475
		476	Examples:
		477
		478	# same as "chown root path" in the shell:
		479	aio_chown "path", 0, -1;
		480	# same as above:
		481	aio_chown "path", 0, undef;
		482
		483	aio_truncate $fh_or_path, $offset, $callback->($status)
		484	Works like truncate(2) or ftruncate(2).
		485
		486	aio_chmod $fh_or_path, $mode, $callback->($status)
		487	Works like perl's "chmod" function.
		488
		489	aio_unlink $pathname, $callback->($status)
		490	Asynchronously unlink (delete) a file and call the callback with the
		491	result code.
		492
		493	aio_mknod $path, $mode, $dev, $callback->($status)
		494	[EXPERIMENTAL]
		495
		496	Asynchronously create a device node (or fifo). See mknod(2).
		497
		498	The only (POSIX-) portable way of calling this function is:
		499
		500	aio_mknod $path, IO::AIO::S_IFIFO | $mode, 0, sub { ...
		501
		502	aio_link $srcpath, $dstpath, $callback->($status)
		503	Asynchronously create a new link to the existing object at $srcpath
		504	at the path $dstpath and call the callback with the result code.
		505
		506	aio_symlink $srcpath, $dstpath, $callback->($status)
		507	Asynchronously create a new symbolic link to the existing object at
		508	$srcpath at the path $dstpath and call the callback with the result
		509	code.
		510
		511	aio_readlink $path, $callback->($link)
		512	Asynchronously read the symlink specified by $path and pass it to
		513	the callback. If an error occurs, nothing or undef gets passed to
		514	the callback.
		515
		516	aio_rename $srcpath, $dstpath, $callback->($status)
		517	Asynchronously rename the object at $srcpath to $dstpath, just as
		518	rename(2) and call the callback with the result code.
		519
		520	aio_mkdir $pathname, $mode, $callback->($status)
		521	Asynchronously mkdir (create) a directory and call the callback with
		522	the result code. $mode will be modified by the umask at the time the
		523	request is executed, so do not change your umask.
		524
		525	aio_rmdir $pathname, $callback->($status)
		526	Asynchronously rmdir (delete) a directory and call the callback with
		527	the result code.
		528
		529	aio_readdir $pathname, $callback->($entries)
		530	Unlike the POSIX call of the same name, "aio_readdir" reads an
		531	entire directory (i.e. opendir + readdir + closedir). The entries
		532	will not be sorted, and will NOT include the "." and ".." entries.
		533
		534	The callback is passed a single argument which is either "undef" or
		535	an array-ref with the filenames.
		536
		537	aio_readdirx $pathname, $flags, $callback->($entries, $flags)
		538	Quite similar to "aio_readdir", but the $flags argument allows to
		539	tune behaviour and output format. In case of an error, $entries will
		540	be "undef".
		541
		542	The flags are a combination of the following constants, ORed
		543	together (the flags will also be passed to the callback, possibly
		544	modified):
		545
		546	IO::AIO::READDIR_DENTS
		547	When this flag is off, then the callback gets an arrayref with
		548	of names only (as with "aio_readdir"), otherwise it gets an
		549	arrayref with "[$name, $type, $inode]" arrayrefs, each
		550	describing a single directory entry in more detail.
		551
		552	$name is the name of the entry.
		553
		554	$type is one of the "IO::AIO::DT_xxx" constants:
		555
		556	"IO::AIO::DT_UNKNOWN", "IO::AIO::DT_FIFO", "IO::AIO::DT_CHR",
		557	"IO::AIO::DT_DIR", "IO::AIO::DT_BLK", "IO::AIO::DT_REG",
		558	"IO::AIO::DT_LNK", "IO::AIO::DT_SOCK", "IO::AIO::DT_WHT".
		559
		560	"IO::AIO::DT_UNKNOWN" means just that: readdir does not know. If
		561	you need to know, you have to run stat yourself. Also, for speed
		562	reasons, the $type scalars are read-only: you can not modify
		563	them.
		564
		565	$inode is the inode number (which might not be exact on systems
		566	with 64 bit inode numbers and 32 bit perls). This field has
		567	unspecified content on systems that do not deliver the inode
		568	information.
		569
		570	IO::AIO::READDIR_DIRS_FIRST
		571	When this flag is set, then the names will be returned in an
		572	order where likely directories come first. This is useful when
		573	you need to quickly find directories, or you want to find all
		574	directories while avoiding to stat() each entry.
		575
		576	If the system returns type information in readdir, then this is
		577	used to find directories directly. Otherwise, likely directories
		578	are files beginning with ".", or otherwise files with no dots,
		579	of which files with short names are tried first.
		580
		581	IO::AIO::READDIR_STAT_ORDER
		582	When this flag is set, then the names will be returned in an
		583	order suitable for stat()'ing each one. That is, when you plan
		584	to stat() all files in the given directory, then the returned
		585	order will likely be fastest.
		586
		587	If both this flag and "IO::AIO::READDIR_DIRS_FIRST" are
		588	specified, then the likely dirs come first, resulting in a less
		589	optimal stat order.
		590
		591	IO::AIO::READDIR_FOUND_UNKNOWN
		592	This flag should not be set when calling "aio_readdirx".
		593	Instead, it is being set by "aio_readdirx", when any of the
		594	$type's found were "IO::AIO::DT_UNKNOWN". The absense of this
		595	flag therefore indicates that all $type's are known, which can
		596	be used to speed up some algorithms.
		597
		598	aio_load $path, $data, $callback->($status)
		599	This is a composite request that tries to fully load the given file
		600	into memory. Status is the same as with aio_read.
		601
		602	aio_copy $srcpath, $dstpath, $callback->($status)
		603	Try to copy the *file* (directories not supported as either source
		604	or destination) from $srcpath to $dstpath and call the callback with
		605	a status of 0 (ok) or -1 (error, see $!).
		606
		607	This is a composite request that creates the destination file with
		608	mode 0200 and copies the contents of the source file into it using
		609	"aio_sendfile", followed by restoring atime, mtime, access mode and
		610	uid/gid, in that order.
		611
		612	If an error occurs, the partial destination file will be unlinked,
		613	if possible, except when setting atime, mtime, access mode and
		614	uid/gid, where errors are being ignored.
		615
		616	aio_move $srcpath, $dstpath, $callback->($status)
		617	Try to move the *file* (directories not supported as either source
		618	or destination) from $srcpath to $dstpath and call the callback with
		619	a status of 0 (ok) or -1 (error, see $!).
		620
		621	This is a composite request that tries to rename(2) the file first;
		622	if rename fails with "EXDEV", it copies the file with "aio_copy"
		623	and, if that is successful, unlinks the $srcpath.
		624
		625	aio_scandir $path, $maxreq, $callback->($dirs, $nondirs)
		626	Scans a directory (similar to "aio_readdir") but additionally tries
		627	to efficiently separate the entries of directory $path into two sets
		628	of names, directories you can recurse into (directories), and ones
		629	you cannot recurse into (everything else, including symlinks to
		630	directories).
		631
		632	"aio_scandir" is a composite request that creates of many sub
		633	requests_ $maxreq specifies the maximum number of outstanding aio
		634	requests that this function generates. If it is "<= 0", then a
		635	suitable default will be chosen (currently 4).
		636
		637	On error, the callback is called without arguments, otherwise it
		638	receives two array-refs with path-relative entry names.
		639
		640	Example:
		641
		642	aio_scandir $dir, 0, sub {
		643	my ($dirs, $nondirs) = @_;
		644	print "real directories: @$dirs\n";
		645	print "everything else: @$nondirs\n";
		646	};
		647
		648	Implementation notes.
		649
		650	The "aio_readdir" cannot be avoided, but "stat()"'ing every entry
		651	can.
		652
		653	If readdir returns file type information, then this is used directly
		654	to find directories.
		655
		656	Otherwise, after reading the directory, the modification time, size
		657	etc. of the directory before and after the readdir is checked, and
		658	if they match (and isn't the current time), the link count will be
		659	used to decide how many entries are directories (if >= 2).
		660	Otherwise, no knowledge of the number of subdirectories will be
		661	assumed.
		662
		663	Then entries will be sorted into likely directories a non-initial
		664	dot currently) and likely non-directories (see "aio_readdirx"). Then
		665	every entry plus an appended "/." will be "stat"'ed, likely
		666	directories first, in order of their inode numbers. If that
		667	succeeds, it assumes that the entry is a directory or a symlink to
		668	directory (which will be checked seperately). This is often faster
		669	than stat'ing the entry itself because filesystems might detect the
		670	type of the entry without reading the inode data (e.g. ext2fs
		671	filetype feature), even on systems that cannot return the filetype
		672	information on readdir.
		673
		674	If the known number of directories (link count - 2) has been
		675	reached, the rest of the entries is assumed to be non-directories.
		676
		677	This only works with certainty on POSIX (= UNIX) filesystems, which
		678	fortunately are the vast majority of filesystems around.
		679
		680	It will also likely work on non-POSIX filesystems with reduced
		681	efficiency as those tend to return 0 or 1 as link counts, which
		682	disables the directory counting heuristic.
		683
		684	aio_rmtree $path, $callback->($status)
		685	Delete a directory tree starting (and including) $path, return the
		686	status of the final "rmdir" only. This is a composite request that
		687	uses "aio_scandir" to recurse into and rmdir directories, and unlink
		688	everything else.
		689
		690	aio_sync $callback->($status)
		691	Asynchronously call sync and call the callback when finished.
		692
		693	aio_fsync $fh, $callback->($status)
		694	Asynchronously call fsync on the given filehandle and call the
		695	callback with the fsync result code.
		696
		697	aio_fdatasync $fh, $callback->($status)
		698	Asynchronously call fdatasync on the given filehandle and call the
		699	callback with the fdatasync result code.
		700
		701	If this call isn't available because your OS lacks it or it couldn't
		702	be detected, it will be emulated by calling "fsync" instead.
		703
		704	aio_sync_file_range $fh, $offset, $nbytes, $flags, $callback->($status)
		705	Sync the data portion of the file specified by $offset and $length
		706	to disk (but NOT the metadata), by calling the Linux-specific
		707	sync_file_range call. If sync_file_range is not available or it
		708	returns ENOSYS, then fdatasync or fsync is being substituted.
		709
		710	$flags can be a combination of
		711	"IO::AIO::SYNC_FILE_RANGE_WAIT_BEFORE",
		712	"IO::AIO::SYNC_FILE_RANGE_WRITE" and
		713	"IO::AIO::SYNC_FILE_RANGE_WAIT_AFTER": refer to the sync_file_range
		714	manpage for details.
		715
		716	aio_pathsync $path, $callback->($status)
		717	This request tries to open, fsync and close the given path. This is
		718	a composite request intended to sync directories after directory
		719	operations (E.g. rename). This might not work on all operating
		720	systems or have any specific effect, but usually it makes sure that
		721	directory changes get written to disc. It works for anything that
		722	can be opened for read-only, not just directories.
		723
		724	Future versions of this function might fall back to other methods
		725	when "fsync" on the directory fails (such as calling "sync").
		726
		727	Passes 0 when everything went ok, and -1 on error.
		728
		729	aio_msync $scalar, $offset = 0, $length = undef, flags = 0,
		730	$callback->($status)
		731	This is a rather advanced IO::AIO call, which only works on
		732	mmap(2)ed scalars (see the "IO::AIO::mmap" function, although it
		733	also works on data scalars managed by the Sys::Mmap or Mmap modules,
		734	note that the scalar must only be modified in-place while an aio
		735	operation is pending on it).
		736
		737	It calls the "msync" function of your OS, if available, with the
		738	memory area starting at $offset in the string and ending $length
		739	bytes later. If $length is negative, counts from the end, and if
		740	$length is "undef", then it goes till the end of the string. The
		741	flags can be a combination of "IO::AIO::MS_ASYNC",
		742	"IO::AIO::MS_INVALIDATE" and "IO::AIO::MS_SYNC".
		743
		744	aio_mtouch $scalar, $offset = 0, $length = undef, flags = 0,
		745	$callback->($status)
		746	This is a rather advanced IO::AIO call, which works best on
		747	mmap(2)ed scalars.
		748
		749	It touches (reads or writes) all memory pages in the specified range
		750	inside the scalar. All caveats and parameters are the same as for
		751	"aio_msync", above, except for flags, which must be either 0 (which
		752	reads all pages and ensures they are instantiated) or
		753	"IO::AIO::MT_MODIFY", which modifies the memory page s(by reading
		754	and writing an octet from it, which dirties the page).
		755
		756	aio_group $callback->(...)
		757	This is a very special aio request: Instead of doing something, it
		758	is a container for other aio requests, which is useful if you want
		759	to bundle many requests into a single, composite, request with a
		760	definite callback and the ability to cancel the whole request with
		761	its subrequests.
		762
		763	Returns an object of class IO::AIO::GRP. See its documentation below
		764	for more info.
		765
		766	Example:
		767
		768	my $grp = aio_group sub {
		769	print "all stats done\n";
		770	};
		771
		772	add $grp
		773	(aio_stat ...),
		774	(aio_stat ...),
		775	...;
		776
		777	aio_nop $callback->()
		778	This is a special request - it does nothing in itself and is only
		779	used for side effects, such as when you want to add a dummy request
		780	to a group so that finishing the requests in the group depends on
		781	executing the given code.
		782
		783	While this request does nothing, it still goes through the execution
		784	phase and still requires a worker thread. Thus, the callback will
		785	not be executed immediately but only after other requests in the
		786	queue have entered their execution phase. This can be used to
		787	measure request latency.
		788
		789	IO::AIO::aio_busy $fractional_seconds, $callback->() *NOT EXPORTED*
		790	Mainly used for debugging and benchmarking, this aio request puts
		791	one of the request workers to sleep for the given time.
		792
		793	While it is theoretically handy to have simple I/O scheduling
		794	requests like sleep and file handle readable/writable, the overhead
		795	this creates is immense (it blocks a thread for a long time) so do
		796	not use this function except to put your application under
		797	artificial I/O pressure.
		798
		799	IO::AIO::REQ CLASS
		800	All non-aggregate "aio_*" functions return an object of this class when
		801	called in non-void context.
		802
		803	cancel $req
		804	Cancels the request, if possible. Has the effect of skipping
		805	execution when entering the execute state and skipping calling the
		806	callback when entering the the result state, but will leave the
		807	request otherwise untouched (with the exception of readdir). That
		808	means that requests that currently execute will not be stopped and
		809	resources held by the request will not be freed prematurely.
		810
		811	cb $req $callback->(...)
		812	Replace (or simply set) the callback registered to the request.
		813
		814	IO::AIO::GRP CLASS
		815	This class is a subclass of IO::AIO::REQ, so all its methods apply to
		816	objects of this class, too.
		817
		818	A IO::AIO::GRP object is a special request that can contain multiple
		819	other aio requests.
		820
		821	You create one by calling the "aio_group" constructing function with a
		822	callback that will be called when all contained requests have entered
		823	the "done" state:
		824
		825	my $grp = aio_group sub {
		826	print "all requests are done\n";
		827	};
		828
		829	You add requests by calling the "add" method with one or more
		830	"IO::AIO::REQ" objects:
		831
		832	$grp->add (aio_unlink "...");
		833
		834	add $grp aio_stat "...", sub {
		835	$_[0] or return $grp->result ("error");
		836
		837	# add another request dynamically, if first succeeded
		838	add $grp aio_open "...", sub {
		839	$grp->result ("ok");
		840	};
		841	};
		842
493	This makes it very easy to create composite requests (see the source	843	This makes it very easy to create composite requests (see the source of
494	of "aio_move" for an application) that work and feel like simple	844	"aio_move" for an application) that work and feel like simple requests.
		845
		846	* The IO::AIO::GRP objects will be cleaned up during calls to
		847	"IO::AIO::poll_cb", just like any other request.
		848
		849	* They can be canceled like any other request. Canceling will cancel
		850	not only the request itself, but also all requests it contains.
		851
		852	* They can also can also be added to other IO::AIO::GRP objects.
		853
		854	* You must not add requests to a group from within the group callback
		855	(or any later time).
		856
		857	Their lifetime, simplified, looks like this: when they are empty, they
		858	will finish very quickly. If they contain only requests that are in the
		859	"done" state, they will also finish. Otherwise they will continue to
		860	exist.
		861
		862	That means after creating a group you have some time to add requests
		863	(precisely before the callback has been invoked, which is only done
		864	within the "poll_cb"). And in the callbacks of those requests, you can
		865	add further requests to the group. And only when all those requests have
		866	finished will the the group itself finish.
		867
		868	add $grp ...
		869	$grp->add (...)
		870	Add one or more requests to the group. Any type of IO::AIO::REQ can
		871	be added, including other groups, as long as you do not create
		872	circular dependencies.
		873
		874	Returns all its arguments.
		875
		876	$grp->cancel_subs
		877	Cancel all subrequests and clears any feeder, but not the group
		878	request itself. Useful when you queued a lot of events but got a
		879	result early.
		880
		881	The group request will finish normally (you cannot add requests to
		882	the group).
		883
		884	$grp->result (...)
		885	Set the result value(s) that will be passed to the group callback
		886	when all subrequests have finished and set the groups errno to the
		887	current value of errno (just like calling "errno" without an error
		888	number). By default, no argument will be passed and errno is zero.
		889
		890	$grp->errno ([$errno])
		891	Sets the group errno value to $errno, or the current value of errno
		892	when the argument is missing.
		893
		894	Every aio request has an associated errno value that is restored
		895	when the callback is invoked. This method lets you change this value
		896	from its default (0).
		897
		898	Calling "result" will also set errno, so make sure you either set $!
		899	before the call to "result", or call c<errno> after it.
		900
		901	feed $grp $callback->($grp)
		902	Sets a feeder/generator on this group: every group can have an
		903	attached generator that generates requests if idle. The idea behind
		904	this is that, although you could just queue as many requests as you
		905	want in a group, this might starve other requests for a potentially
		906	long time. For example, "aio_scandir" might generate hundreds of
		907	thousands "aio_stat" requests, delaying any later requests for a
		908	long time.
		909
		910	To avoid this, and allow incremental generation of requests, you can
		911	instead a group and set a feeder on it that generates those
		912	requests. The feed callback will be called whenever there are few
		913	enough (see "limit", below) requests active in the group itself and
		914	is expected to queue more requests.
		915
		916	The feed callback can queue as many requests as it likes (i.e. "add"
		917	does not impose any limits).
		918
		919	If the feed does not queue more requests when called, it will be
		920	automatically removed from the group.
		921
		922	If the feed limit is 0 when this method is called, it will be set to
		923	2 automatically.
		924
		925	Example:
		926
		927	# stat all files in @files, but only ever use four aio requests concurrently:
		928
		929	my $grp = aio_group sub { print "finished\n" };
		930	limit $grp 4;
		931	feed $grp sub {
		932	my $file = pop @files
		933	or return;
		934
		935	add $grp aio_stat $file, sub { ... };
		936	};
		937
		938	limit $grp $num
		939	Sets the feeder limit for the group: The feeder will be called
		940	whenever the group contains less than this many requests.
		941
		942	Setting the limit to 0 will pause the feeding process.
		943
		944	The default value for the limit is 0, but note that setting a feeder
		945	automatically bumps it up to 2.
		946
		947	SUPPORT FUNCTIONS
		948	EVENT PROCESSING AND EVENT LOOP INTEGRATION
		949	$fileno = IO::AIO::poll_fileno
		950	Return the *request result pipe file descriptor*. This filehandle
		951	must be polled for reading by some mechanism outside this module
		952	(e.g. EV, Glib, select and so on, see below or the SYNOPSIS). If the
		953	pipe becomes readable you have to call "poll_cb" to check the
		954	results.
		955
		956	See "poll_cb" for an example.
		957
		958	IO::AIO::poll_cb
		959	Process some outstanding events on the result pipe. You have to call
		960	this regularly. Returns 0 if all events could be processed, or -1 if
		961	it returned earlier for whatever reason. Returns immediately when no
		962	events are outstanding. The amount of events processed depends on
		963	the settings of "IO::AIO::max_poll_req" and
		964	"IO::AIO::max_poll_time".
		965
		966	If not all requests were processed for whatever reason, the
		967	filehandle will still be ready when "poll_cb" returns, so normally
		968	you don't have to do anything special to have it called later.
		969
		970	Example: Install an Event watcher that automatically calls
		971	IO::AIO::poll_cb with high priority (more examples can be found in
		972	the SYNOPSIS section, at the top of this document):
		973
		974	Event->io (fd => IO::AIO::poll_fileno,
		975	poll => 'r', async => 1,
		976	cb => \&IO::AIO::poll_cb);
		977
		978	IO::AIO::poll_wait
		979	If there are any outstanding requests and none of them in the result
		980	phase, wait till the result filehandle becomes ready for reading
		981	(simply does a "select" on the filehandle. This is useful if you
		982	want to synchronously wait for some requests to finish).
		983
		984	See "nreqs" for an example.
		985
		986	IO::AIO::poll
		987	Waits until some requests have been handled.
		988
		989	Returns the number of requests processed, but is otherwise strictly
		990	equivalent to:
		991
		992	IO::AIO::poll_wait, IO::AIO::poll_cb
		993
		994	IO::AIO::flush
		995	Wait till all outstanding AIO requests have been handled.
		996
		997	Strictly equivalent to:
		998
		999	IO::AIO::poll_wait, IO::AIO::poll_cb
		1000	while IO::AIO::nreqs;
		1001
		1002	IO::AIO::max_poll_reqs $nreqs
		1003	IO::AIO::max_poll_time $seconds
		1004	These set the maximum number of requests (default 0, meaning
		1005	infinity) that are being processed by "IO::AIO::poll_cb" in one
		1006	call, respectively the maximum amount of time (default 0, meaning
		1007	infinity) spent in "IO::AIO::poll_cb" to process requests (more
		1008	correctly the mininum amount of time "poll_cb" is allowed to use).
		1009
		1010	Setting "max_poll_time" to a non-zero value creates an overhead of
		1011	one syscall per request processed, which is not normally a problem
		1012	unless your callbacks are really really fast or your OS is really
		1013	really slow (I am not mentioning Solaris here). Using
		1014	"max_poll_reqs" incurs no overhead.
		1015
		1016	Setting these is useful if you want to ensure some level of
		1017	interactiveness when perl is not fast enough to process all requests
		1018	in time.
		1019
		1020	For interactive programs, values such as 0.01 to 0.1 should be fine.
		1021
		1022	Example: Install an Event watcher that automatically calls
		1023	IO::AIO::poll_cb with low priority, to ensure that other parts of
		1024	the program get the CPU sometimes even under high AIO load.
		1025
		1026	# try not to spend much more than 0.1s in poll_cb
		1027	IO::AIO::max_poll_time 0.1;
		1028
		1029	# use a low priority so other tasks have priority
		1030	Event->io (fd => IO::AIO::poll_fileno,
		1031	poll => 'r', nice => 1,
		1032	cb => &IO::AIO::poll_cb);
		1033
		1034	CONTROLLING THE NUMBER OF THREADS
		1035	IO::AIO::min_parallel $nthreads
		1036	Set the minimum number of AIO threads to $nthreads. The current
		1037	default is 8, which means eight asynchronous operations can execute
		1038	concurrently at any one time (the number of outstanding requests,
		1039	however, is unlimited).
		1040
		1041	IO::AIO starts threads only on demand, when an AIO request is queued
		1042	and no free thread exists. Please note that queueing up a hundred
		1043	requests can create demand for a hundred threads, even if it turns
		1044	out that everything is in the cache and could have been processed
		1045	faster by a single thread.
		1046
		1047	It is recommended to keep the number of threads relatively low, as
		1048	some Linux kernel versions will scale negatively with the number of
		1049	threads (higher parallelity => MUCH higher latency). With current
		1050	Linux 2.6 versions, 4-32 threads should be fine.
		1051
		1052	Under most circumstances you don't need to call this function, as
		1053	the module selects a default that is suitable for low to moderate
		1054	load.
		1055
		1056	IO::AIO::max_parallel $nthreads
		1057	Sets the maximum number of AIO threads to $nthreads. If more than
		1058	the specified number of threads are currently running, this function
		1059	kills them. This function blocks until the limit is reached.
		1060
		1061	While $nthreads are zero, aio requests get queued but not executed
		1062	until the number of threads has been increased again.
		1063
		1064	This module automatically runs "max_parallel 0" at program end, to
		1065	ensure that all threads are killed and that there are no outstanding
495	requests.	1066	requests.
496		1067
497	* The IO::AIO::GRP objects will be cleaned up during calls to
498	"IO::AIO::poll_cb", just like any other request.
499	* They can be canceled like any other request. Canceling will cancel
500	not only the request itself, but also all requests it contains.
501	* They can also can also be added to other IO::AIO::GRP objects.
502	* You must not add requests to a group from within the group
503	callback (or any later time).
504
505	Their lifetime, simplified, looks like this: when they are empty,
506	they will finish very quickly. If they contain only requests that
507	are in the "done" state, they will also finish. Otherwise they will
508	continue to exist.
509
510	That means after creating a group you have some time to add
511	requests. And in the callbacks of those requests, you can add
512	further requests to the group. And only when all those requests have
513	finished will the the group itself finish.
514
515	add $grp ...
516	$grp->add (...)
517	Add one or more requests to the group. Any type of IO::AIO::REQ
518	can be added, including other groups, as long as you do not
519	create circular dependencies.
520
521	Returns all its arguments.
522
523	$grp->cancel_subs
524	Cancel all subrequests and clears any feeder, but not the group
525	request itself. Useful when you queued a lot of events but got a
526	result early.
527
528	$grp->result (...)
529	Set the result value(s) that will be passed to the group
530	callback when all subrequests have finished and set thre groups
531	errno to the current value of errno (just like calling "errno"
532	without an error number). By default, no argument will be passed
533	and errno is zero.
534
535	$grp->errno ([$errno])
536	Sets the group errno value to $errno, or the current value of
537	errno when the argument is missing.
538
539	Every aio request has an associated errno value that is restored
540	when the callback is invoked. This method lets you change this
541	value from its default (0).
542
543	Calling "result" will also set errno, so make sure you either
544	set $! before the call to "result", or call c<errno> after it.
545
546	feed $grp $callback->($grp)
547	Sets a feeder/generator on this group: every group can have an
548	attached generator that generates requests if idle. The idea
549	behind this is that, although you could just queue as many
550	requests as you want in a group, this might starve other
551	requests for a potentially long time. For example, "aio_scandir"
552	might generate hundreds of thousands "aio_stat" requests,
553	delaying any later requests for a long time.
554
555	To avoid this, and allow incremental generation of requests, you
556	can instead a group and set a feeder on it that generates those
557	requests. The feed callback will be called whenever there are
558	few enough (see "limit", below) requests active in the group
559	itself and is expected to queue more requests.
560
561	The feed callback can queue as many requests as it likes (i.e.
562	"add" does not impose any limits).
563
564	If the feed does not queue more requests when called, it will be
565	automatically removed from the group.
566
567	If the feed limit is 0, it will be set to 2 automatically.
568
569	Example:
570
571	# stat all files in @files, but only ever use four aio requests concurrently:
572
573	my $grp = aio_group sub { print "finished\n" };
574	limit $grp 4;
575	feed $grp sub {
576	my $file = pop @files
577	or return;
578
579	add $grp aio_stat $file, sub { ... };
580	};
581
582	limit $grp $num
583	Sets the feeder limit for the group: The feeder will be called
584	whenever the group contains less than this many requests.
585
586	Setting the limit to 0 will pause the feeding process.
587
588	SUPPORT FUNCTIONS
589	$fileno = IO::AIO::poll_fileno
590	Return the *request result pipe file descriptor*. This
591	filehandle must be polled for reading by some mechanism outside
592	this module (e.g. Event or select, see below or the SYNOPSIS).
593	If the pipe becomes readable you have to call "poll_cb" to check
594	the results.
595
596	See "poll_cb" for an example.
597
598	IO::AIO::poll_cb
599	Process all outstanding events on the result pipe. You have to
600	call this regularly. Returns the number of events processed.
601	Returns immediately when no events are outstanding.
602
603	If not all requests were processed for whatever reason, the
604	filehandle will still be ready when "poll_cb" returns.
605
606	Example: Install an Event watcher that automatically calls
607	IO::AIO::poll_cb with high priority:
608
609	Event->io (fd => IO::AIO::poll_fileno,
610	poll => 'r', async => 1,
611	cb => \&IO::AIO::poll_cb);
612
613	IO::AIO::poll_some $max_requests
614	Similar to "poll_cb", but only processes up to $max_requests
615	requests at a time.
616
617	Useful if you want to ensure some level of interactiveness when
618	perl is not fast enough to process all requests in time.
619
620	Example: Install an Event watcher that automatically calls
621	IO::AIO::poll_some with low priority, to ensure that other parts
622	of the program get the CPU sometimes even under high AIO load.
623
624	Event->io (fd => IO::AIO::poll_fileno,
625	poll => 'r', nice => 1,
626	cb => sub { IO::AIO::poll_some 256 });
627
628	IO::AIO::poll_wait
629	Wait till the result filehandle becomes ready for reading
630	(simply does a "select" on the filehandle. This is useful if you
631	want to synchronously wait for some requests to finish).
632
633	See "nreqs" for an example.
634
635	IO::AIO::nreqs
636	Returns the number of requests currently in the ready, execute
637	or pending states (i.e. for which their callback has not been
638	invoked yet).
639
640	Example: wait till there are no outstanding requests anymore:
641
642	IO::AIO::poll_wait, IO::AIO::poll_cb
643	while IO::AIO::nreqs;
644
645	IO::AIO::nready
646	Returns the number of requests currently in the ready state (not
647	yet executed).
648
649	IO::AIO::npending
650	Returns the number of requests currently in the pending state
651	(executed, but not yet processed by poll_cb).
652
653	IO::AIO::flush
654	Wait till all outstanding AIO requests have been handled.
655
656	Strictly equivalent to:
657
658	IO::AIO::poll_wait, IO::AIO::poll_cb
659	while IO::AIO::nreqs;
660
661	IO::AIO::poll
662	Waits until some requests have been handled.
663
664	Strictly equivalent to:
665
666	IO::AIO::poll_wait, IO::AIO::poll_cb
667	if IO::AIO::nreqs;
668
669	IO::AIO::min_parallel $nthreads
670	Set the minimum number of AIO threads to $nthreads. The current
671	default is 8, which means eight asynchronous operations can
672	execute concurrently at any one time (the number of outstanding
673	requests, however, is unlimited).
674
675	IO::AIO starts threads only on demand, when an AIO request is
676	queued and no free thread exists.
677
678	It is recommended to keep the number of threads relatively low,
679	as some Linux kernel versions will scale negatively with the
680	number of threads (higher parallelity => MUCH higher latency).
681	With current Linux 2.6 versions, 4-32 threads should be fine.
682
683	Under most circumstances you don't need to call this function,
684	as the module selects a default that is suitable for low to
685	moderate load.
686
687	IO::AIO::max_parallel $nthreads
688	Sets the maximum number of AIO threads to $nthreads. If more
689	than the specified number of threads are currently running, this
690	function kills them. This function blocks until the limit is
691	reached.
692
693	While $nthreads are zero, aio requests get queued but not
694	executed until the number of threads has been increased again.
695
696	This module automatically runs "max_parallel 0" at program end,
697	to ensure that all threads are killed and that there are no
698	outstanding requests.
699
700	Under normal circumstances you don't need to call this function.	1068	Under normal circumstances you don't need to call this function.
701		1069
		1070	IO::AIO::max_idle $nthreads
		1071	Limit the number of threads (default: 4) that are allowed to idle
		1072	(i.e., threads that did not get a request to process within 10
		1073	seconds). That means if a thread becomes idle while $nthreads other
		1074	threads are also idle, it will free its resources and exit.
		1075
		1076	This is useful when you allow a large number of threads (e.g. 100 or
		1077	1000) to allow for extremely high load situations, but want to free
		1078	resources under normal circumstances (1000 threads can easily
		1079	consume 30MB of RAM).
		1080
		1081	The default is probably ok in most situations, especially if thread
		1082	creation is fast. If thread creation is very slow on your system you
		1083	might want to use larger values.
		1084
702	$oldmaxreqs = IO::AIO::max_outstanding $maxreqs	1085	IO::AIO::max_outstanding $maxreqs
703	This is a very bad function to use in interactive programs	1086	This is a very bad function to use in interactive programs because
704	because it blocks, and a bad way to reduce concurrency because	1087	it blocks, and a bad way to reduce concurrency because it is
705	it is inexact: Better use an "aio_group" together with a feed	1088	inexact: Better use an "aio_group" together with a feed callback.
706	callback.
707		1089
708	Sets the maximum number of outstanding requests to $nreqs. If	1090	Sets the maximum number of outstanding requests to $nreqs. If you do
709	you to queue up more than this number of requests, the next call	1091	queue up more than this number of requests, the next call to the
710	to the "poll_cb" (and "poll_some" and other functions calling	1092	"poll_cb" (and "poll_some" and other functions calling "poll_cb")
711	"poll_cb") function will block until the limit is no longer	1093	function will block until the limit is no longer exceeded.
712	exceeded.
713		1094
714	The default value is very large, so there is no practical limit	1095	The default value is very large, so there is no practical limit on
715	on the number of outstanding requests.	1096	the number of outstanding requests.
716		1097
717	You can still queue as many requests as you want. Therefore,	1098	You can still queue as many requests as you want. Therefore,
718	"max_oustsanding" is mainly useful in simple scripts (with low	1099	"max_outstanding" is mainly useful in simple scripts (with low
719	values) or as a stop gap to shield against fatal memory overflow	1100	values) or as a stop gap to shield against fatal memory overflow
720	(with large values).	1101	(with large values).
		1102
		1103	STATISTICAL INFORMATION
		1104	IO::AIO::nreqs
		1105	Returns the number of requests currently in the ready, execute or
		1106	pending states (i.e. for which their callback has not been invoked
		1107	yet).
		1108
		1109	Example: wait till there are no outstanding requests anymore:
		1110
		1111	IO::AIO::poll_wait, IO::AIO::poll_cb
		1112	while IO::AIO::nreqs;
		1113
		1114	IO::AIO::nready
		1115	Returns the number of requests currently in the ready state (not yet
		1116	executed).
		1117
		1118	IO::AIO::npending
		1119	Returns the number of requests currently in the pending state
		1120	(executed, but not yet processed by poll_cb).
		1121
		1122	MISCELLANEOUS FUNCTIONS
		1123	IO::AIO implements some functions that might be useful, but are not
		1124	asynchronous.
		1125
		1126	IO::AIO::sendfile $ofh, $ifh, $offset, $count
		1127	Calls the "eio_sendfile_sync" function, which is like
		1128	"aio_sendfile", but is blocking (this makes most sense if you know
		1129	the input data is likely cached already and the output filehandle is
		1130	set to non-blocking operations).
		1131
		1132	Returns the number of bytes copied, or -1 on error.
		1133
		1134	IO::AIO::fadvise $fh, $offset, $len, $advice
		1135	Simply calls the "posix_fadvise" function (see it's manpage for
		1136	details). The following advice constants are avaiable:
		1137	"IO::AIO::FADV_NORMAL", "IO::AIO::FADV_SEQUENTIAL",
		1138	"IO::AIO::FADV_RANDOM", "IO::AIO::FADV_NOREUSE",
		1139	"IO::AIO::FADV_WILLNEED", "IO::AIO::FADV_DONTNEED".
		1140
		1141	On systems that do not implement "posix_fadvise", this function
		1142	returns ENOSYS, otherwise the return value of "posix_fadvise".
		1143
		1144	IO::AIO::mmap $scalar, $length, $prot, $flags, $fh[, $offset]
		1145	Memory-maps a file (or anonymous memory range) and attaches it to
		1146	the given $scalar, which will act like a string scalar.
		1147
		1148	The only operations allowed on the scalar are "substr"/"vec" that
		1149	don't change the string length, and most read-only operations such
		1150	as copying it or searching it with regexes and so on.
		1151
		1152	Anything else is unsafe and will, at best, result in memory leaks.
		1153
		1154	The memory map associated with the $scalar is automatically removed
		1155	when the $scalar is destroyed, or when the "IO::AIO::mmap" or
		1156	"IO::AIO::munmap" functions are called.
		1157
		1158	This calls the "mmap"(2) function internally. See your system's
		1159	manual page for details on the $length, $prot and $flags parameters.
		1160
		1161	The $length must be larger than zero and smaller than the actual
		1162	filesize.
		1163
		1164	$prot is a combination of "IO::AIO::PROT_NONE",
		1165	"IO::AIO::PROT_EXEC", "IO::AIO::PROT_READ" and/or
		1166	"IO::AIO::PROT_WRITE",
		1167
		1168	$flags can be a combination of "IO::AIO::MAP_SHARED" or
		1169	"IO::AIO::MAP_PRIVATE", or a number of system-specific flags (when
		1170	not available, the are defined as 0): "IO::AIO::MAP_ANONYMOUS"
		1171	(which is set to "MAP_ANON" if your system only provides this
		1172	constant), "IO::AIO::MAP_HUGETLB", "IO::AIO::MAP_LOCKED",
		1173	"IO::AIO::MAP_NORESERVE", "IO::AIO::MAP_POPULATE" or
		1174	"IO::AIO::MAP_NONBLOCK"
		1175
		1176	If $fh is "undef", then a file descriptor of -1 is passed.
		1177
		1178	$offset is the offset from the start of the file - it generally must
		1179	be a multiple of "IO::AIO::PAGESIZE" and defaults to 0.
		1180
		1181	Example:
		1182
		1183	use Digest::MD5;
		1184	use IO::AIO;
		1185
		1186	open my $fh, "<verybigfile"
		1187	or die "$!";
		1188
		1189	IO::AIO::mmap my $data, -s $fh, IO::AIO::PROT_READ, IO::AIO::MAP_SHARED, $fh
		1190	or die "verybigfile: $!";
		1191
		1192	my $fast_md5 = md5 $data;
		1193
		1194	IO::AIO::munmap $scalar
		1195	Removes a previous mmap and undefines the $scalar.
		1196
		1197	IO::AIO::mlockall $flags
		1198	Calls the "mlockall" function with the given $flags (a combination
		1199	of "IO::AIO::MCL_CURRENT" and "IO::AIO::MCL__FUTURE").
		1200
		1201	On systems that do not implement "mlockall", this function returns
		1202	ENOSYS, otherwise the return value of "mlockall".
		1203
		1204	IO::AIO::munlockall
		1205	Calls the "munlockall" function.
		1206
		1207	On systems that do not implement "munlockall", this function returns
		1208	ENOSYS, otherwise the return value of "munlockall".
		1209
		1210	EVENT LOOP INTEGRATION
		1211	It is recommended to use AnyEvent::AIO to integrate IO::AIO
		1212	automatically into many event loops:
		1213
		1214	# AnyEvent integration (EV, Event, Glib, Tk, POE, urxvt, pureperl...)
		1215	use AnyEvent::AIO;
		1216
		1217	You can also integrate IO::AIO manually into many event loops, here are
		1218	some examples of how to do this:
		1219
		1220	# EV integration
		1221	my $aio_w = EV::io IO::AIO::poll_fileno, EV::READ, \&IO::AIO::poll_cb;
		1222
		1223	# Event integration
		1224	Event->io (fd => IO::AIO::poll_fileno,
		1225	poll => 'r',
		1226	cb => \&IO::AIO::poll_cb);
		1227
		1228	# Glib/Gtk2 integration
		1229	add_watch Glib::IO IO::AIO::poll_fileno,
		1230	in => sub { IO::AIO::poll_cb; 1 };
		1231
		1232	# Tk integration
		1233	Tk::Event::IO->fileevent (IO::AIO::poll_fileno, "",
		1234	readable => \&IO::AIO::poll_cb);
		1235
		1236	# Danga::Socket integration
		1237	Danga::Socket->AddOtherFds (IO::AIO::poll_fileno =>
		1238	\&IO::AIO::poll_cb);
721		1239
722	FORK BEHAVIOUR	1240	FORK BEHAVIOUR
723	This module should do "the right thing" when the process using it	1241	This module should do "the right thing" when the process using it forks:
724	forks:
725		1242
726	Before the fork, IO::AIO enters a quiescent state where no requests	1243	Before the fork, IO::AIO enters a quiescent state where no requests can
727	can be added in other threads and no results will be processed.	1244	be added in other threads and no results will be processed. After the
728	After the fork the parent simply leaves the quiescent state and	1245	fork the parent simply leaves the quiescent state and continues
729	continues request/result processing, while the child frees the	1246	request/result processing, while the child frees the request/result
730	request/result queue (so that the requests started before the fork	1247	queue (so that the requests started before the fork will only be handled
731	will only be handled in the parent). Threads will be started on	1248	in the parent). Threads will be started on demand until the limit set in
732	demand until the limit set in the parent process has been reached	1249	the parent process has been reached again.
733	again.
734		1250
735	In short: the parent will, after a short pause, continue as if fork	1251	In short: the parent will, after a short pause, continue as if fork had
736	had not been called, while the child will act as if IO::AIO has not	1252	not been called, while the child will act as if IO::AIO has not been
737	been used yet.	1253	used yet.
738		1254
739	MEMORY USAGE	1255	MEMORY USAGE
740	Per-request usage:	1256	Per-request usage:
741		1257
742	Each aio request uses - depending on your architecture - around	1258	Each aio request uses - depending on your architecture - around 100-200
743	100-200 bytes of memory. In addition, stat requests need a stat	1259	bytes of memory. In addition, stat requests need a stat buffer (possibly
744	buffer (possibly a few hundred bytes), readdir requires a result	1260	a few hundred bytes), readdir requires a result buffer and so on. Perl
745	buffer and so on. Perl scalars and other data passed into aio	1261	scalars and other data passed into aio requests will also be locked and
746	requests will also be locked and will consume memory till the	1262	will consume memory till the request has entered the done state.
747	request has entered the done state.
748		1263
749	This is now awfully much, so queuing lots of requests is not usually	1264	This is not awfully much, so queuing lots of requests is not usually a
750	a problem.	1265	problem.
751		1266
752	Per-thread usage:	1267	Per-thread usage:
753		1268
754	In the execution phase, some aio requests require more memory for	1269	In the execution phase, some aio requests require more memory for
755	temporary buffers, and each thread requires a stack and other data	1270	temporary buffers, and each thread requires a stack and other data
756	structures (usually around 16k-128k, depending on the OS).	1271	structures (usually around 16k-128k, depending on the OS).
757		1272
758	KNOWN BUGS	1273	KNOWN BUGS
759	Known bugs will be fixed in the next release.	1274	Known bugs will be fixed in the next release.
760		1275
761	SEE ALSO	1276	SEE ALSO
762	Coro::AIO.	1277	AnyEvent::AIO for easy integration into event loops, Coro::AIO for a
		1278	more natural syntax.
763		1279
764	AUTHOR	1280	AUTHOR
765	Marc Lehmann <schmorp@schmorp.de>	1281	Marc Lehmann <schmorp@schmorp.de>
766	http://home.schmorp.de/	1282	http://home.schmorp.de/
767		1283

Diff Legend

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