ViewVC Help

View File | Revision Log | Show Annotations | Download File

/cvs/IO-AIO/README

Comparing IO-AIO/README (file contents):
Revision 1.17 by root, Mon Jun 26 14:53:55 2006 UTC vs.
Revision 1.46 by root, Sun Mar 27 10:26:08 2011 UTC

…		…
2	IO::AIO - Asynchronous Input/Output	2	IO::AIO - Asynchronous Input/Output
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", IO::AIO::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
14	aio_read $fh, 30000, 1024, $buffer, 0, sub {	15	aio_read $fh, 30000, 1024, $buffer, 0, sub {
15	$_[0] > 0 or die "read error: $!";	16	$_[0] > 0 or die "read error: $!";
16	};	17	};
17		18
18	# AnyEvent	19	# version 2+ has request and group objects
19	open my $fh, "<&=" . IO::AIO::poll_fileno or die "$!";	20	use IO::AIO 2;
20	my $w = AnyEvent->io (fh => $fh, poll => 'r', cb => sub { IO::AIO::poll_cb });
21		21
22	# Event	22	aioreq_pri 4; # give next request a very high priority
23	Event->io (fd => IO::AIO::poll_fileno,	23	my $req = aio_unlink "/tmp/file", sub { };
24	poll => 'r',	24	$req->cancel; # cancel request if still in queue
25	cb => \&IO::AIO::poll_cb);
26		25
27	# Glib/Gtk2	26	my $grp = aio_group sub { print "all stats done\n" };
28	add_watch Glib::IO IO::AIO::poll_fileno,	27	add $grp aio_stat "..." for ...;
29	in => sub { IO::AIO::poll_cb; 1 };
30
31	# Tk
32	Tk::Event::IO->fileevent (IO::AIO::poll_fileno, "",
33	readable => \&IO::AIO::poll_cb);
34
35	# Danga::Socket
36	Danga::Socket->AddOtherFds (IO::AIO::poll_fileno =>
37	\&IO::AIO::poll_cb);
38		28
39	DESCRIPTION	29	DESCRIPTION
40	This module implements asynchronous I/O using whatever means your	30	This module implements asynchronous I/O using whatever means your
41	operating system supports.	31	operating system supports. It is implemented as an interface to "libeio"
		32	(<http://software.schmorp.de/pkg/libeio.html>).
42		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.
		49
43	Currently, a number of threads are started that execute your read/writes	50	In this version, a number of threads are started that execute your
44	and signal their completion. You don't need thread support in your libc	51	requests and signal their completion. You don't need thread support in
45	or 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
46	the pthreads library. In the future, this module might make use of the	53	perl. In the future, this module might make use of the native aio
47	native aio functions available on many operating systems. However, they	54	functions available on many operating systems. However, they are often
48	are often not well-supported (Linux doesn't allow them on normal files	55	not well-supported or restricted (GNU/Linux doesn't allow them on normal
49	currently, for example), and they would only support aio_read and	56	files currently, for example), and they would only support aio_read and
50	aio_write, so the remaining functionality would have to be implemented	57	aio_write, so the remaining functionality would have to be implemented
51	using threads anyway.	58	using threads anyway.
52		59
53	Although the module will work with in the presence of other threads, it	60	Although the module will work in the presence of other (Perl-) threads,
54	is currently not reentrant, so use appropriate locking yourself, always	61	it is currently not reentrant in any way, so use appropriate locking
55	call "poll_cb" from within the same thread, or never call "poll_cb" (or	62	yourself, always call "poll_cb" from within the same thread, or never
56	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", IO::AIO::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;
		105
		106	REQUEST ANATOMY AND LIFETIME
		107	Every "aio_*" function creates a request. which is a C data structure
		108	not directly visible to Perl.
		109
		110	If called in non-void context, every request function returns a Perl
		111	object representing the request. In void context, nothing is returned,
		112	which saves a bit of memory.
		113
		114	The perl object is a fairly standard ref-to-hash object. The hash
		115	contents are not used by IO::AIO so you are free to store anything you
		116	like in it.
		117
		118	During their existance, aio requests travel through the following
		119	states, in order:
		120
		121	ready
		122	Immediately after a request is created it is put into the ready
		123	state, waiting for a thread to execute it.
		124
		125	execute
		126	A thread has accepted the request for processing and is currently
		127	executing it (e.g. blocking in read).
		128
		129	pending
		130	The request has been executed and is waiting for result processing.
		131
		132	While request submission and execution is fully asynchronous, result
		133	processing is not and relies on the perl interpreter calling
		134	"poll_cb" (or another function with the same effect).
		135
		136	result
		137	The request results are processed synchronously by "poll_cb".
		138
		139	The "poll_cb" function will process all outstanding aio requests by
		140	calling their callbacks, freeing memory associated with them and
		141	managing any groups they are contained in.
		142
		143	done
		144	Request has reached the end of its lifetime and holds no resources
		145	anymore (except possibly for the Perl object, but its connection to
		146	the actual aio request is severed and calling its methods will
		147	either do nothing or result in a runtime error).
57		148
58	FUNCTIONS	149	FUNCTIONS
		150	QUICK OVERVIEW
		151	This section simply lists the prototypes of the most important functions
		152	for quick reference. See the following sections for function-by-function
		153	documentation.
		154
		155	aio_open $pathname, $flags, $mode, $callback->($fh)
		156	aio_close $fh, $callback->($status)
		157	aio_read $fh,$offset,$length, $data,$dataoffset, $callback->($retval)
		158	aio_write $fh,$offset,$length, $data,$dataoffset, $callback->($retval)
		159	aio_sendfile $out_fh, $in_fh, $in_offset, $length, $callback->($retval)
		160	aio_readahead $fh,$offset,$length, $callback->($retval)
		161	aio_stat $fh_or_path, $callback->($status)
		162	aio_lstat $fh, $callback->($status)
		163	aio_statvfs $fh_or_path, $callback->($statvfs)
		164	aio_utime $fh_or_path, $atime, $mtime, $callback->($status)
		165	aio_chown $fh_or_path, $uid, $gid, $callback->($status)
		166	aio_truncate $fh_or_path, $offset, $callback->($status)
		167	aio_chmod $fh_or_path, $mode, $callback->($status)
		168	aio_unlink $pathname, $callback->($status)
		169	aio_mknod $path, $mode, $dev, $callback->($status)
		170	aio_link $srcpath, $dstpath, $callback->($status)
		171	aio_symlink $srcpath, $dstpath, $callback->($status)
		172	aio_readlink $path, $callback->($link)
		173	aio_rename $srcpath, $dstpath, $callback->($status)
		174	aio_mkdir $pathname, $mode, $callback->($status)
		175	aio_rmdir $pathname, $callback->($status)
		176	aio_readdir $pathname, $callback->($entries)
		177	aio_readdirx $pathname, $flags, $callback->($entries, $flags)
		178	IO::AIO::READDIR_DENTS IO::AIO::READDIR_DIRS_FIRST
		179	IO::AIO::READDIR_STAT_ORDER IO::AIO::READDIR_FOUND_UNKNOWN
		180	aio_load $path, $data, $callback->($status)
		181	aio_copy $srcpath, $dstpath, $callback->($status)
		182	aio_move $srcpath, $dstpath, $callback->($status)
		183	aio_scandir $path, $maxreq, $callback->($dirs, $nondirs)
		184	aio_rmtree $path, $callback->($status)
		185	aio_sync $callback->($status)
		186	aio_fsync $fh, $callback->($status)
		187	aio_fdatasync $fh, $callback->($status)
		188	aio_sync_file_range $fh, $offset, $nbytes, $flags, $callback->($status)
		189	aio_pathsync $path, $callback->($status)
		190	aio_msync $scalar, $offset = 0, $length = undef, flags = 0, $callback->($status)
		191	aio_mtouch $scalar, $offset = 0, $length = undef, flags = 0, $callback->($status)
		192	aio_mlock $scalar, $offset = 0, $length = undef, $callback->($status)
		193	aio_mlockall $flags, $callback->($status)
		194	aio_group $callback->(...)
		195	aio_nop $callback->()
		196
		197	$prev_pri = aioreq_pri [$pri]
		198	aioreq_nice $pri_adjust
		199
		200	IO::AIO::poll_wait
		201	IO::AIO::poll_cb
		202	IO::AIO::poll
		203	IO::AIO::flush
		204	IO::AIO::max_poll_reqs $nreqs
		205	IO::AIO::max_poll_time $seconds
		206	IO::AIO::min_parallel $nthreads
		207	IO::AIO::max_parallel $nthreads
		208	IO::AIO::max_idle $nthreads
		209	IO::AIO::idle_timeout $seconds
		210	IO::AIO::max_outstanding $maxreqs
		211	IO::AIO::nreqs
		212	IO::AIO::nready
		213	IO::AIO::npending
		214
		215	IO::AIO::sendfile $ofh, $ifh, $offset, $count
		216	IO::AIO::fadvise $fh, $offset, $len, $advice
		217	IO::AIO::madvise $scalar, $offset, $length, $advice
		218	IO::AIO::mprotect $scalar, $offset, $length, $protect
		219	IO::AIO::munlock $scalar, $offset = 0, $length = undef
		220	IO::AIO::munlockall
		221
59	AIO FUNCTIONS	222	AIO REQUEST FUNCTIONS
60	All the "aio_*" calls are more or less thin wrappers around the syscall	223	All the "aio_*" calls are more or less thin wrappers around the syscall
61	with the same name (sans "aio_"). The arguments are similar or	224	with the same name (sans "aio_"). The arguments are similar or
62	identical, and they all accept an additional (and optional) $callback	225	identical, and they all accept an additional (and optional) $callback
63	argument which must be a code reference. This code reference will get	226	argument which must be a code reference. This code reference will get
64	called with the syscall return code (e.g. most syscalls return -1 on	227	called with the syscall return code (e.g. most syscalls return -1 on
65	error, unlike perl, which usually delivers "false") as it's sole	228	error, unlike perl, which usually delivers "false") as its sole argument
66	argument when the given syscall has been executed asynchronously.	229	after the given syscall has been executed asynchronously.
67		230
68	All functions expecting a filehandle keep a copy of the filehandle	231	All functions expecting a filehandle keep a copy of the filehandle
69	internally until the request has finished.	232	internally until the request has finished.
70		233
		234	All functions return request objects of type IO::AIO::REQ that allow
		235	further manipulation of those requests while they are in-flight.
		236
71	The pathnames you pass to these routines *must* be absolute and encoded	237	The pathnames you pass to these routines *must* be absolute and encoded
72	in byte form. The reason for the former is that at the time the request	238	as octets. The reason for the former is that at the time the request is
73	is being executed, the current working directory could have changed.	239	being executed, the current working directory could have changed.
74	Alternatively, you can make sure that you never change the current	240	Alternatively, you can make sure that you never change the current
75	working directory.	241	working directory anywhere in the program and then use relative paths.
76		242
77	To encode pathnames to byte form, either make sure you either: a) always	243	To encode pathnames as octets, either make sure you either: a) always
78	pass in filenames you got from outside (command line, readdir etc.), b)	244	pass in filenames you got from outside (command line, readdir etc.)
79	are ASCII or ISO 8859-1, c) use the Encode module and encode your	245	without tinkering, b) are ASCII or ISO 8859-1, c) use the Encode module
80	pathnames to the locale (or other) encoding in effect in the user	246	and encode your pathnames to the locale (or other) encoding in effect in
81	environment, d) use Glib::filename_from_unicode on unicode filenames or	247	the user environment, d) use Glib::filename_from_unicode on unicode
82	e) use something else.	248	filenames or e) use something else to ensure your scalar has the correct
		249	contents.
		250
		251	This works, btw. independent of the internal UTF-8 bit, which IO::AIO
		252	handles correctly whether it is set or not.
		253
		254	$prev_pri = aioreq_pri [$pri]
		255	Returns the priority value that would be used for the next request
		256	and, if $pri is given, sets the priority for the next aio request.
		257
		258	The default priority is 0, the minimum and maximum priorities are -4
		259	and 4, respectively. Requests with higher priority will be serviced
		260	first.
		261
		262	The priority will be reset to 0 after each call to one of the
		263	"aio_*" functions.
		264
		265	Example: open a file with low priority, then read something from it
		266	with higher priority so the read request is serviced before other
		267	low priority open requests (potentially spamming the cache):
		268
		269	aioreq_pri -3;
		270	aio_open ..., sub {
		271	return unless $_[0];
		272
		273	aioreq_pri -2;
		274	aio_read $_[0], ..., sub {
		275	...
		276	};
		277	};
		278
		279	aioreq_nice $pri_adjust
		280	Similar to "aioreq_pri", but subtracts the given value from the
		281	current priority, so the effect is cumulative.
83		282
84	aio_open $pathname, $flags, $mode, $callback->($fh)	283	aio_open $pathname, $flags, $mode, $callback->($fh)
85	Asynchronously open or create a file and call the callback with a	284	Asynchronously open or create a file and call the callback with a
86	newly created filehandle for the file.	285	newly created filehandle for the file.
87		286
…		…
92	They are the same as used by "sysopen".	291	They are the same as used by "sysopen".
93		292
94	Likewise, $mode specifies the mode of the newly created file, if it	293	Likewise, $mode specifies the mode of the newly created file, if it
95	didn't exist and "O_CREAT" has been given, just like perl's	294	didn't exist and "O_CREAT" has been given, just like perl's
96	"sysopen", except that it is mandatory (i.e. use 0 if you don't	295	"sysopen", except that it is mandatory (i.e. use 0 if you don't
97	create new files, and 0666 or 0777 if you do).	296	create new files, and 0666 or 0777 if you do). Note that the $mode
		297	will be modified by the umask in effect then the request is being
		298	executed, so better never change the umask.
98		299
99	Example:	300	Example:
100		301
101	aio_open "/etc/passwd", O_RDONLY, 0, sub {	302	aio_open "/etc/passwd", IO::AIO::O_RDONLY, 0, sub {
102	if ($_[0]) {	303	if ($_[0]) {
103	print "open successful, fh is $_[0]\n";	304	print "open successful, fh is $_[0]\n";
104	...	305	...
105	} else {	306	} else {
106	die "open failed: $!\n";	307	die "open failed: $!\n";
107	}	308	}
108	};	309	};
109		310
110	aio_close $fh, $callback->($status)	311	aio_close $fh, $callback->($status)
111	Asynchronously close a file and call the callback with the result	312	Asynchronously close a file and call the callback with the result
112	code. *WARNING:* although accepted, you should not pass in a perl	313	code.
113	filehandle here, as perl will likely close the file descriptor
114	another time when the filehandle is destroyed. Normally, you can
115	safely call perls "close" or just let filehandles go out of scope.
116		314
117	This is supposed to be a bug in the API, so that might change. It's	315	Unfortunately, you can't do this to perl. Perl *insists* very
118	therefore best to avoid this function.	316	strongly on closing the file descriptor associated with the
		317	filehandle itself.
		318
		319	Therefore, "aio_close" will not close the filehandle - instead it
		320	will use dup2 to overwrite the file descriptor with the write-end of
		321	a pipe (the pipe fd will be created on demand and will be cached).
		322
		323	Or in other words: the file descriptor will be closed, but it will
		324	not be free for reuse until the perl filehandle is closed.
119		325
120	aio_read $fh,$offset,$length, $data,$dataoffset, $callback->($retval)	326	aio_read $fh,$offset,$length, $data,$dataoffset, $callback->($retval)
121	aio_write $fh,$offset,$length, $data,$dataoffset, $callback->($retval)	327	aio_write $fh,$offset,$length, $data,$dataoffset, $callback->($retval)
122	Reads or writes "length" bytes from the specified "fh" and "offset"	328	Reads or writes $length bytes from or to the specified $fh and
123	into the scalar given by "data" and offset "dataoffset" and calls	329	$offset into the scalar given by $data and offset $dataoffset and
124	the callback without the actual number of bytes read (or -1 on	330	calls the callback without the actual number of bytes read (or -1 on
125	error, just like the syscall).	331	error, just like the syscall).
126		332
		333	"aio_read" will, like "sysread", shrink or grow the $data scalar to
		334	offset plus the actual number of bytes read.
		335
		336	If $offset is undefined, then the current file descriptor offset
		337	will be used (and updated), otherwise the file descriptor offset
		338	will not be changed by these calls.
		339
		340	If $length is undefined in "aio_write", use the remaining length of
		341	$data.
		342
		343	If $dataoffset is less than zero, it will be counted from the end of
		344	$data.
		345
127	The $data scalar *MUST NOT* be modified in any way while the request	346	The $data scalar *MUST NOT* be modified in any way while the request
128	is outstanding. Modifying it can result in segfaults or WW3 (if the	347	is outstanding. Modifying it can result in segfaults or World War
129	necessary/optional hardware is installed).	348	III (if the necessary/optional hardware is installed).
130		349
131	Example: Read 15 bytes at offset 7 into scalar $buffer, starting at	350	Example: Read 15 bytes at offset 7 into scalar $buffer, starting at
132	offset 0 within the scalar:	351	offset 0 within the scalar:
133		352
134	aio_read $fh, 7, 15, $buffer, 0, sub {	353	aio_read $fh, 7, 15, $buffer, 0, sub {
135	$_[0] > 0 or die "read error: $!";	354	$_[0] > 0 or die "read error: $!";
136	print "read $_[0] bytes: <$buffer>\n";	355	print "read $_[0] bytes: <$buffer>\n";
137	};	356	};
138
139	aio_move $srcpath, $dstpath, $callback->($status)
140	[EXPERIMENTAL]
141
142	Try to move the *file* (directories not supported as either source
143	or destination) from $srcpath to $dstpath and call the callback with
144	the 0 (error) or -1 ok.
145
146	This is a composite request that tries to rename(2) the file first.
147	If rename files with "EXDEV", it creates the destination file with
148	mode 0200 and copies the contents of the source file into it using
149	"aio_sendfile", followed by restoring atime, mtime, access mode and
150	uid/gid, in that order, and unlinking the $srcpath.
151
152	If an error occurs, the partial destination file will be unlinked,
153	if possible, except when setting atime, mtime, access mode and
154	uid/gid, where errors are being ignored.
155		357
156	aio_sendfile $out_fh, $in_fh, $in_offset, $length, $callback->($retval)	358	aio_sendfile $out_fh, $in_fh, $in_offset, $length, $callback->($retval)
157	Tries to copy $length bytes from $in_fh to $out_fh. It starts	359	Tries to copy $length bytes from $in_fh to $out_fh. It starts
158	reading at byte offset $in_offset, and starts writing at the current	360	reading at byte offset $in_offset, and starts writing at the current
159	file offset of $out_fh. Because of that, it is not safe to issue	361	file offset of $out_fh. Because of that, it is not safe to issue
160	more than one "aio_sendfile" per $out_fh, as they will interfere	362	more than one "aio_sendfile" per $out_fh, as they will interfere
161	with each other.	363	with each other.
162		364
		365	Please note that "aio_sendfile" can read more bytes from $in_fh than
		366	are written, and there is no way to find out how many bytes have
		367	been read from "aio_sendfile" alone, as "aio_sendfile" only provides
		368	the number of bytes written to $out_fh. Only if the result value
		369	equals $length one can assume that $length bytes have been read.
		370
		371	Unlike with other "aio_" functions, it makes a lot of sense to use
		372	"aio_sendfile" on non-blocking sockets, as long as one end
		373	(typically the $in_fh) is a file - the file I/O will then be
		374	asynchronous, while the socket I/O will be non-blocking. Note,
		375	however, that you can run into a trap where "aio_sendfile" reads
		376	some data with readahead, then fails to write all data, and when the
		377	socket is ready the next time, the data in the cache is already
		378	lost, forcing "aio_sendfile" to again hit the disk. Explicit
		379	"aio_read" + "aio_write" let's you control resource usage much
		380	better.
		381
163	This call tries to make use of a native "sendfile" syscall to	382	This call tries to make use of a native "sendfile" syscall to
164	provide zero-copy operation. For this to work, $out_fh should refer	383	provide zero-copy operation. For this to work, $out_fh should refer
165	to a socket, and $in_fh should refer to mmap'able file.	384	to a socket, and $in_fh should refer to an mmap'able file.
166		385
167	If the native sendfile call fails or is not implemented, it will be	386	If a native sendfile cannot be found or it fails with "ENOSYS",
		387	"ENOTSUP", "EOPNOTSUPP", "EAFNOSUPPORT", "EPROTOTYPE" or "ENOTSOCK",
168	emulated, so you can call "aio_sendfile" on any type of filehandle	388	it will be emulated, so you can call "aio_sendfile" on any type of
169	regardless of the limitations of the operating system.	389	filehandle regardless of the limitations of the operating system.
170
171	Please note, however, that "aio_sendfile" can read more bytes from
172	$in_fh than are written, and there is no way to find out how many
173	bytes have been read from "aio_sendfile" alone, as "aio_sendfile"
174	only provides the number of bytes written to $out_fh. Only if the
175	result value equals $length one can assume that $length bytes have
176	been read.
177		390
178	aio_readahead $fh,$offset,$length, $callback->($retval)	391	aio_readahead $fh,$offset,$length, $callback->($retval)
179	"aio_readahead" populates the page cache with data from a file so	392	"aio_readahead" populates the page cache with data from a file so
180	that subsequent reads from that file will not block on disk I/O. The	393	that subsequent reads from that file will not block on disk I/O. The
181	$offset argument specifies the starting point from which data is to	394	$offset argument specifies the starting point from which data is to
…		…
202	Currently, the stats are always 64-bit-stats, i.e. instead of	415	Currently, the stats are always 64-bit-stats, i.e. instead of
203	returning an error when stat'ing a large file, the results will be	416	returning an error when stat'ing a large file, the results will be
204	silently truncated unless perl itself is compiled with large file	417	silently truncated unless perl itself is compiled with large file
205	support.	418	support.
206		419
		420	To help interpret the mode and dev/rdev stat values, IO::AIO offers
		421	the following constants and functions (if not implemented, the
		422	constants will be 0 and the functions will either "croak" or fall
		423	back on traditional behaviour).
		424
		425	"S_IFMT", "S_IFIFO", "S_IFCHR", "S_IFBLK", "S_IFLNK", "S_IFREG",
		426	"S_IFDIR", "S_IFWHT", "S_IFSOCK", "IO::AIO::major $dev_t",
		427	"IO::AIO::minor $dev_t", "IO::AIO::makedev $major, $minor".
		428
207	Example: Print the length of /etc/passwd:	429	Example: Print the length of /etc/passwd:
208		430
209	aio_stat "/etc/passwd", sub {	431	aio_stat "/etc/passwd", sub {
210	$_[0] and die "stat failed: $!";	432	$_[0] and die "stat failed: $!";
211	print "size is ", -s _, "\n";	433	print "size is ", -s _, "\n";
212	};	434	};
213		435
		436	aio_statvfs $fh_or_path, $callback->($statvfs)
		437	Works like the POSIX "statvfs" or "fstatvfs" syscalls, depending on
		438	whether a file handle or path was passed.
		439
		440	On success, the callback is passed a hash reference with the
		441	following members: "bsize", "frsize", "blocks", "bfree", "bavail",
		442	"files", "ffree", "favail", "fsid", "flag" and "namemax". On
		443	failure, "undef" is passed.
		444
		445	The following POSIX IO::AIO::ST_* constants are defined: "ST_RDONLY"
		446	and "ST_NOSUID".
		447
		448	The following non-POSIX IO::AIO::ST_* flag masks are defined to
		449	their correct value when available, or to 0 on systems that do not
		450	support them: "ST_NODEV", "ST_NOEXEC", "ST_SYNCHRONOUS",
		451	"ST_MANDLOCK", "ST_WRITE", "ST_APPEND", "ST_IMMUTABLE",
		452	"ST_NOATIME", "ST_NODIRATIME" and "ST_RELATIME".
		453
		454	Example: stat "/wd" and dump out the data if successful.
		455
		456	aio_statvfs "/wd", sub {
		457	my $f = $_[0]
		458	or die "statvfs: $!";
		459
		460	use Data::Dumper;
		461	say Dumper $f;
		462	};
		463
		464	# result:
		465	{
		466	bsize => 1024,
		467	bfree => 4333064312,
		468	blocks => 10253828096,
		469	files => 2050765568,
		470	flag => 4096,
		471	favail => 2042092649,
		472	bavail => 4333064312,
		473	ffree => 2042092649,
		474	namemax => 255,
		475	frsize => 1024,
		476	fsid => 1810
		477	}
		478
		479	aio_utime $fh_or_path, $atime, $mtime, $callback->($status)
		480	Works like perl's "utime" function (including the special case of
		481	$atime and $mtime being undef). Fractional times are supported if
		482	the underlying syscalls support them.
		483
		484	When called with a pathname, uses utimes(2) if available, otherwise
		485	utime(2). If called on a file descriptor, uses futimes(2) if
		486	available, otherwise returns ENOSYS, so this is not portable.
		487
		488	Examples:
		489
		490	# set atime and mtime to current time (basically touch(1)):
		491	aio_utime "path", undef, undef;
		492	# set atime to current time and mtime to beginning of the epoch:
		493	aio_utime "path", time, undef; # undef==0
		494
		495	aio_chown $fh_or_path, $uid, $gid, $callback->($status)
		496	Works like perl's "chown" function, except that "undef" for either
		497	$uid or $gid is being interpreted as "do not change" (but -1 can
		498	also be used).
		499
		500	Examples:
		501
		502	# same as "chown root path" in the shell:
		503	aio_chown "path", 0, -1;
		504	# same as above:
		505	aio_chown "path", 0, undef;
		506
		507	aio_truncate $fh_or_path, $offset, $callback->($status)
		508	Works like truncate(2) or ftruncate(2).
		509
		510	aio_chmod $fh_or_path, $mode, $callback->($status)
		511	Works like perl's "chmod" function.
		512
214	aio_unlink $pathname, $callback->($status)	513	aio_unlink $pathname, $callback->($status)
215	Asynchronously unlink (delete) a file and call the callback with the	514	Asynchronously unlink (delete) a file and call the callback with the
216	result code.	515	result code.
		516
		517	aio_mknod $path, $mode, $dev, $callback->($status)
		518	[EXPERIMENTAL]
		519
		520	Asynchronously create a device node (or fifo). See mknod(2).
		521
		522	The only (POSIX-) portable way of calling this function is:
		523
		524	aio_mknod $path, IO::AIO::S_IFIFO | $mode, 0, sub { ...
		525
		526	See "aio_stat" for info about some potentially helpful extra
		527	constants and functions.
217		528
218	aio_link $srcpath, $dstpath, $callback->($status)	529	aio_link $srcpath, $dstpath, $callback->($status)
219	Asynchronously create a new link to the existing object at $srcpath	530	Asynchronously create a new link to the existing object at $srcpath
220	at the path $dstpath and call the callback with the result code.	531	at the path $dstpath and call the callback with the result code.
221		532
222	aio_symlink $srcpath, $dstpath, $callback->($status)	533	aio_symlink $srcpath, $dstpath, $callback->($status)
223	Asynchronously create a new symbolic link to the existing object at	534	Asynchronously create a new symbolic link to the existing object at
224	$srcpath at the path $dstpath and call the callback with the result	535	$srcpath at the path $dstpath and call the callback with the result
225	code.	536	code.
226		537
		538	aio_readlink $path, $callback->($link)
		539	Asynchronously read the symlink specified by $path and pass it to
		540	the callback. If an error occurs, nothing or undef gets passed to
		541	the callback.
		542
227	aio_rename $srcpath, $dstpath, $callback->($status)	543	aio_rename $srcpath, $dstpath, $callback->($status)
228	Asynchronously rename the object at $srcpath to $dstpath, just as	544	Asynchronously rename the object at $srcpath to $dstpath, just as
229	rename(2) and call the callback with the result code.	545	rename(2) and call the callback with the result code.
		546
		547	aio_mkdir $pathname, $mode, $callback->($status)
		548	Asynchronously mkdir (create) a directory and call the callback with
		549	the result code. $mode will be modified by the umask at the time the
		550	request is executed, so do not change your umask.
230		551
231	aio_rmdir $pathname, $callback->($status)	552	aio_rmdir $pathname, $callback->($status)
232	Asynchronously rmdir (delete) a directory and call the callback with	553	Asynchronously rmdir (delete) a directory and call the callback with
233	the result code.	554	the result code.
234		555
235	aio_readdir $pathname, $callback->($entries)	556	aio_readdir $pathname, $callback->($entries)
236	Unlike the POSIX call of the same name, "aio_readdir" reads an	557	Unlike the POSIX call of the same name, "aio_readdir" reads an
237	entire directory (i.e. opendir + readdir + closedir). The entries	558	entire directory (i.e. opendir + readdir + closedir). The entries
238	will not be sorted, and will NOT include the "." and ".." entries.	559	will not be sorted, and will NOT include the "." and ".." entries.
239		560
240	The callback a single argument which is either "undef" or an	561	The callback is passed a single argument which is either "undef" or
241	array-ref with the filenames.	562	an array-ref with the filenames.
		563
		564	aio_readdirx $pathname, $flags, $callback->($entries, $flags)
		565	Quite similar to "aio_readdir", but the $flags argument allows to
		566	tune behaviour and output format. In case of an error, $entries will
		567	be "undef".
		568
		569	The flags are a combination of the following constants, ORed
		570	together (the flags will also be passed to the callback, possibly
		571	modified):
		572
		573	IO::AIO::READDIR_DENTS
		574	When this flag is off, then the callback gets an arrayref with
		575	of names only (as with "aio_readdir"), otherwise it gets an
		576	arrayref with "[$name, $type, $inode]" arrayrefs, each
		577	describing a single directory entry in more detail.
		578
		579	$name is the name of the entry.
		580
		581	$type is one of the "IO::AIO::DT_xxx" constants:
		582
		583	"IO::AIO::DT_UNKNOWN", "IO::AIO::DT_FIFO", "IO::AIO::DT_CHR",
		584	"IO::AIO::DT_DIR", "IO::AIO::DT_BLK", "IO::AIO::DT_REG",
		585	"IO::AIO::DT_LNK", "IO::AIO::DT_SOCK", "IO::AIO::DT_WHT".
		586
		587	"IO::AIO::DT_UNKNOWN" means just that: readdir does not know. If
		588	you need to know, you have to run stat yourself. Also, for speed
		589	reasons, the $type scalars are read-only: you can not modify
		590	them.
		591
		592	$inode is the inode number (which might not be exact on systems
		593	with 64 bit inode numbers and 32 bit perls). This field has
		594	unspecified content on systems that do not deliver the inode
		595	information.
		596
		597	IO::AIO::READDIR_DIRS_FIRST
		598	When this flag is set, then the names will be returned in an
		599	order where likely directories come first. This is useful when
		600	you need to quickly find directories, or you want to find all
		601	directories while avoiding to stat() each entry.
		602
		603	If the system returns type information in readdir, then this is
		604	used to find directories directly. Otherwise, likely directories
		605	are files beginning with ".", or otherwise files with no dots,
		606	of which files with short names are tried first.
		607
		608	IO::AIO::READDIR_STAT_ORDER
		609	When this flag is set, then the names will be returned in an
		610	order suitable for stat()'ing each one. That is, when you plan
		611	to stat() all files in the given directory, then the returned
		612	order will likely be fastest.
		613
		614	If both this flag and "IO::AIO::READDIR_DIRS_FIRST" are
		615	specified, then the likely dirs come first, resulting in a less
		616	optimal stat order.
		617
		618	IO::AIO::READDIR_FOUND_UNKNOWN
		619	This flag should not be set when calling "aio_readdirx".
		620	Instead, it is being set by "aio_readdirx", when any of the
		621	$type's found were "IO::AIO::DT_UNKNOWN". The absense of this
		622	flag therefore indicates that all $type's are known, which can
		623	be used to speed up some algorithms.
		624
		625	aio_load $path, $data, $callback->($status)
		626	This is a composite request that tries to fully load the given file
		627	into memory. Status is the same as with aio_read.
		628
		629	aio_copy $srcpath, $dstpath, $callback->($status)
		630	Try to copy the *file* (directories not supported as either source
		631	or destination) from $srcpath to $dstpath and call the callback with
		632	a status of 0 (ok) or -1 (error, see $!).
		633
		634	This is a composite request that creates the destination file with
		635	mode 0200 and copies the contents of the source file into it using
		636	"aio_sendfile", followed by restoring atime, mtime, access mode and
		637	uid/gid, in that order.
		638
		639	If an error occurs, the partial destination file will be unlinked,
		640	if possible, except when setting atime, mtime, access mode and
		641	uid/gid, where errors are being ignored.
		642
		643	aio_move $srcpath, $dstpath, $callback->($status)
		644	Try to move the *file* (directories not supported as either source
		645	or destination) from $srcpath to $dstpath and call the callback with
		646	a status of 0 (ok) or -1 (error, see $!).
		647
		648	This is a composite request that tries to rename(2) the file first;
		649	if rename fails with "EXDEV", it copies the file with "aio_copy"
		650	and, if that is successful, unlinks the $srcpath.
242		651
243	aio_scandir $path, $maxreq, $callback->($dirs, $nondirs)	652	aio_scandir $path, $maxreq, $callback->($dirs, $nondirs)
244	Scans a directory (similar to "aio_readdir") and tries to separate	653	Scans a directory (similar to "aio_readdir") but additionally tries
245	the entries of directory $path into two sets of names, ones you can	654	to efficiently separate the entries of directory $path into two sets
246	recurse into (directories), and ones you cannot recurse into	655	of names, directories you can recurse into (directories), and ones
247	(everything else).	656	you cannot recurse into (everything else, including symlinks to
		657	directories).
248		658
249	"aio_scandir" is a composite request that consists of many	659	"aio_scandir" is a composite request that creates of many sub
250	aio-primitives. $maxreq specifies the maximum number of outstanding	660	requests_ $maxreq specifies the maximum number of outstanding aio
251	aio requests that this function generates. If it is "<= 0", then a	661	requests that this function generates. If it is "<= 0", then a
252	suitable default will be chosen (currently 8).	662	suitable default will be chosen (currently 4).
253		663
254	On error, the callback is called without arguments, otherwise it	664	On error, the callback is called without arguments, otherwise it
255	receives two array-refs with path-relative entry names.	665	receives two array-refs with path-relative entry names.
256		666
257	Example:	667	Example:
…		…
265	Implementation notes.	675	Implementation notes.
266		676
267	The "aio_readdir" cannot be avoided, but "stat()"'ing every entry	677	The "aio_readdir" cannot be avoided, but "stat()"'ing every entry
268	can.	678	can.
269		679
		680	If readdir returns file type information, then this is used directly
		681	to find directories.
		682
270	After reading the directory, the modification time, size etc. of the	683	Otherwise, after reading the directory, the modification time, size
271	directory before and after the readdir is checked, and if they	684	etc. of the directory before and after the readdir is checked, and
272	match, the link count will be used to decide how many entries are	685	if they match (and isn't the current time), the link count will be
273	directories (if >= 2). Otherwise, no knowledge of the number of	686	used to decide how many entries are directories (if >= 2).
274	subdirectories will be assumed.	687	Otherwise, no knowledge of the number of subdirectories will be
		688	assumed.
275		689
276	Then entires will be sorted into likely directories (everything	690	Then entries will be sorted into likely directories a non-initial
277	without a non-initial dot) and likely non-directories (everything	691	dot currently) and likely non-directories (see "aio_readdirx"). Then
278	else). Then every entry + "/." will be "stat"'ed, likely directories	692	every entry plus an appended "/." will be "stat"'ed, likely
279	first. This is often faster because filesystems might detect the	693	directories first, in order of their inode numbers. If that
		694	succeeds, it assumes that the entry is a directory or a symlink to
		695	directory (which will be checked seperately). This is often faster
		696	than stat'ing the entry itself because filesystems might detect the
280	type of the entry without reading the inode data (e.g. ext2fs	697	type of the entry without reading the inode data (e.g. ext2fs
281	filetype feature). If that succeeds, it assumes that the entry is a	698	filetype feature), even on systems that cannot return the filetype
282	directory or a symlink to directory (which will be checked	699	information on readdir.
283	seperately).
284		700
285	If the known number of directories has been reached, the rest of the	701	If the known number of directories (link count - 2) has been
286	entries is assumed to be non-directories.	702	reached, the rest of the entries is assumed to be non-directories.
		703
		704	This only works with certainty on POSIX (= UNIX) filesystems, which
		705	fortunately are the vast majority of filesystems around.
		706
		707	It will also likely work on non-POSIX filesystems with reduced
		708	efficiency as those tend to return 0 or 1 as link counts, which
		709	disables the directory counting heuristic.
		710
		711	aio_rmtree $path, $callback->($status)
		712	Delete a directory tree starting (and including) $path, return the
		713	status of the final "rmdir" only. This is a composite request that
		714	uses "aio_scandir" to recurse into and rmdir directories, and unlink
		715	everything else.
		716
		717	aio_sync $callback->($status)
		718	Asynchronously call sync and call the callback when finished.
287		719
288	aio_fsync $fh, $callback->($status)	720	aio_fsync $fh, $callback->($status)
289	Asynchronously call fsync on the given filehandle and call the	721	Asynchronously call fsync on the given filehandle and call the
290	callback with the fsync result code.	722	callback with the fsync result code.
291		723
…		…
294	callback with the fdatasync result code.	726	callback with the fdatasync result code.
295		727
296	If this call isn't available because your OS lacks it or it couldn't	728	If this call isn't available because your OS lacks it or it couldn't
297	be detected, it will be emulated by calling "fsync" instead.	729	be detected, it will be emulated by calling "fsync" instead.
298		730
		731	aio_sync_file_range $fh, $offset, $nbytes, $flags, $callback->($status)
		732	Sync the data portion of the file specified by $offset and $length
		733	to disk (but NOT the metadata), by calling the Linux-specific
		734	sync_file_range call. If sync_file_range is not available or it
		735	returns ENOSYS, then fdatasync or fsync is being substituted.
		736
		737	$flags can be a combination of
		738	"IO::AIO::SYNC_FILE_RANGE_WAIT_BEFORE",
		739	"IO::AIO::SYNC_FILE_RANGE_WRITE" and
		740	"IO::AIO::SYNC_FILE_RANGE_WAIT_AFTER": refer to the sync_file_range
		741	manpage for details.
		742
		743	aio_pathsync $path, $callback->($status)
		744	This request tries to open, fsync and close the given path. This is
		745	a composite request intended to sync directories after directory
		746	operations (E.g. rename). This might not work on all operating
		747	systems or have any specific effect, but usually it makes sure that
		748	directory changes get written to disc. It works for anything that
		749	can be opened for read-only, not just directories.
		750
		751	Future versions of this function might fall back to other methods
		752	when "fsync" on the directory fails (such as calling "sync").
		753
		754	Passes 0 when everything went ok, and -1 on error.
		755
		756	aio_msync $scalar, $offset = 0, $length = undef, flags = 0,
		757	$callback->($status)
		758	This is a rather advanced IO::AIO call, which only works on
		759	mmap(2)ed scalars (see the "IO::AIO::mmap" function, although it
		760	also works on data scalars managed by the Sys::Mmap or Mmap modules,
		761	note that the scalar must only be modified in-place while an aio
		762	operation is pending on it).
		763
		764	It calls the "msync" function of your OS, if available, with the
		765	memory area starting at $offset in the string and ending $length
		766	bytes later. If $length is negative, counts from the end, and if
		767	$length is "undef", then it goes till the end of the string. The
		768	flags can be a combination of "IO::AIO::MS_ASYNC",
		769	"IO::AIO::MS_INVALIDATE" and "IO::AIO::MS_SYNC".
		770
		771	aio_mtouch $scalar, $offset = 0, $length = undef, flags = 0,
		772	$callback->($status)
		773	This is a rather advanced IO::AIO call, which works best on
		774	mmap(2)ed scalars.
		775
		776	It touches (reads or writes) all memory pages in the specified range
		777	inside the scalar. All caveats and parameters are the same as for
		778	"aio_msync", above, except for flags, which must be either 0 (which
		779	reads all pages and ensures they are instantiated) or
		780	"IO::AIO::MT_MODIFY", which modifies the memory page s(by reading
		781	and writing an octet from it, which dirties the page).
		782
		783	aio_mlock $scalar, $offset = 0, $length = undef, $callback->($status)
		784	This is a rather advanced IO::AIO call, which works best on
		785	mmap(2)ed scalars.
		786
		787	It reads in all the pages of the underlying storage into memory (if
		788	any) and locks them, so they are not getting swapped/paged out or
		789	removed.
		790
		791	If $length is undefined, then the scalar will be locked till the
		792	end.
		793
		794	On systems that do not implement "mlock", this function returns -1
		795	and sets errno to "ENOSYS".
		796
		797	Note that the corresponding "munlock" is synchronous and is
		798	documented under "MISCELLANEOUS FUNCTIONS".
		799
		800	Example: open a file, mmap and mlock it - both will be undone when
		801	$data gets destroyed.
		802
		803	open my $fh, "<", $path or die "$path: $!";
		804	my $data;
		805	IO::AIO::mmap $data, -s $fh, IO::AIO::PROT_READ, IO::AIO::MAP_SHARED, $fh;
		806	aio_mlock $data; # mlock in background
		807
		808	aio_mlockall $flags, $callback->($status)
		809	Calls the "mlockall" function with the given $flags (a combination
		810	of "IO::AIO::MCL_CURRENT" and "IO::AIO::MCL_FUTURE").
		811
		812	On systems that do not implement "mlockall", this function returns
		813	-1 and sets errno to "ENOSYS".
		814
		815	Note that the corresponding "munlockall" is synchronous and is
		816	documented under "MISCELLANEOUS FUNCTIONS".
		817
		818	Example: asynchronously lock all current and future pages into
		819	memory.
		820
		821	aio_mlockall IO::AIO::MCL_FUTURE;
		822
		823	aio_group $callback->(...)
		824	This is a very special aio request: Instead of doing something, it
		825	is a container for other aio requests, which is useful if you want
		826	to bundle many requests into a single, composite, request with a
		827	definite callback and the ability to cancel the whole request with
		828	its subrequests.
		829
		830	Returns an object of class IO::AIO::GRP. See its documentation below
		831	for more info.
		832
		833	Example:
		834
		835	my $grp = aio_group sub {
		836	print "all stats done\n";
		837	};
		838
		839	add $grp
		840	(aio_stat ...),
		841	(aio_stat ...),
		842	...;
		843
		844	aio_nop $callback->()
		845	This is a special request - it does nothing in itself and is only
		846	used for side effects, such as when you want to add a dummy request
		847	to a group so that finishing the requests in the group depends on
		848	executing the given code.
		849
		850	While this request does nothing, it still goes through the execution
		851	phase and still requires a worker thread. Thus, the callback will
		852	not be executed immediately but only after other requests in the
		853	queue have entered their execution phase. This can be used to
		854	measure request latency.
		855
		856	IO::AIO::aio_busy $fractional_seconds, $callback->() *NOT EXPORTED*
		857	Mainly used for debugging and benchmarking, this aio request puts
		858	one of the request workers to sleep for the given time.
		859
		860	While it is theoretically handy to have simple I/O scheduling
		861	requests like sleep and file handle readable/writable, the overhead
		862	this creates is immense (it blocks a thread for a long time) so do
		863	not use this function except to put your application under
		864	artificial I/O pressure.
		865
		866	IO::AIO::REQ CLASS
		867	All non-aggregate "aio_*" functions return an object of this class when
		868	called in non-void context.
		869
		870	cancel $req
		871	Cancels the request, if possible. Has the effect of skipping
		872	execution when entering the execute state and skipping calling the
		873	callback when entering the the result state, but will leave the
		874	request otherwise untouched (with the exception of readdir). That
		875	means that requests that currently execute will not be stopped and
		876	resources held by the request will not be freed prematurely.
		877
		878	cb $req $callback->(...)
		879	Replace (or simply set) the callback registered to the request.
		880
		881	IO::AIO::GRP CLASS
		882	This class is a subclass of IO::AIO::REQ, so all its methods apply to
		883	objects of this class, too.
		884
		885	A IO::AIO::GRP object is a special request that can contain multiple
		886	other aio requests.
		887
		888	You create one by calling the "aio_group" constructing function with a
		889	callback that will be called when all contained requests have entered
		890	the "done" state:
		891
		892	my $grp = aio_group sub {
		893	print "all requests are done\n";
		894	};
		895
		896	You add requests by calling the "add" method with one or more
		897	"IO::AIO::REQ" objects:
		898
		899	$grp->add (aio_unlink "...");
		900
		901	add $grp aio_stat "...", sub {
		902	$_[0] or return $grp->result ("error");
		903
		904	# add another request dynamically, if first succeeded
		905	add $grp aio_open "...", sub {
		906	$grp->result ("ok");
		907	};
		908	};
		909
		910	This makes it very easy to create composite requests (see the source of
		911	"aio_move" for an application) that work and feel like simple requests.
		912
		913	* The IO::AIO::GRP objects will be cleaned up during calls to
		914	"IO::AIO::poll_cb", just like any other request.
		915
		916	* They can be canceled like any other request. Canceling will cancel
		917	not only the request itself, but also all requests it contains.
		918
		919	* They can also can also be added to other IO::AIO::GRP objects.
		920
		921	* You must not add requests to a group from within the group callback
		922	(or any later time).
		923
		924	Their lifetime, simplified, looks like this: when they are empty, they
		925	will finish very quickly. If they contain only requests that are in the
		926	"done" state, they will also finish. Otherwise they will continue to
		927	exist.
		928
		929	That means after creating a group you have some time to add requests
		930	(precisely before the callback has been invoked, which is only done
		931	within the "poll_cb"). And in the callbacks of those requests, you can
		932	add further requests to the group. And only when all those requests have
		933	finished will the the group itself finish.
		934
		935	add $grp ...
		936	$grp->add (...)
		937	Add one or more requests to the group. Any type of IO::AIO::REQ can
		938	be added, including other groups, as long as you do not create
		939	circular dependencies.
		940
		941	Returns all its arguments.
		942
		943	$grp->cancel_subs
		944	Cancel all subrequests and clears any feeder, but not the group
		945	request itself. Useful when you queued a lot of events but got a
		946	result early.
		947
		948	The group request will finish normally (you cannot add requests to
		949	the group).
		950
		951	$grp->result (...)
		952	Set the result value(s) that will be passed to the group callback
		953	when all subrequests have finished and set the groups errno to the
		954	current value of errno (just like calling "errno" without an error
		955	number). By default, no argument will be passed and errno is zero.
		956
		957	$grp->errno ([$errno])
		958	Sets the group errno value to $errno, or the current value of errno
		959	when the argument is missing.
		960
		961	Every aio request has an associated errno value that is restored
		962	when the callback is invoked. This method lets you change this value
		963	from its default (0).
		964
		965	Calling "result" will also set errno, so make sure you either set $!
		966	before the call to "result", or call c<errno> after it.
		967
		968	feed $grp $callback->($grp)
		969	Sets a feeder/generator on this group: every group can have an
		970	attached generator that generates requests if idle. The idea behind
		971	this is that, although you could just queue as many requests as you
		972	want in a group, this might starve other requests for a potentially
		973	long time. For example, "aio_scandir" might generate hundreds of
		974	thousands "aio_stat" requests, delaying any later requests for a
		975	long time.
		976
		977	To avoid this, and allow incremental generation of requests, you can
		978	instead a group and set a feeder on it that generates those
		979	requests. The feed callback will be called whenever there are few
		980	enough (see "limit", below) requests active in the group itself and
		981	is expected to queue more requests.
		982
		983	The feed callback can queue as many requests as it likes (i.e. "add"
		984	does not impose any limits).
		985
		986	If the feed does not queue more requests when called, it will be
		987	automatically removed from the group.
		988
		989	If the feed limit is 0 when this method is called, it will be set to
		990	2 automatically.
		991
		992	Example:
		993
		994	# stat all files in @files, but only ever use four aio requests concurrently:
		995
		996	my $grp = aio_group sub { print "finished\n" };
		997	limit $grp 4;
		998	feed $grp sub {
		999	my $file = pop @files
		1000	or return;
		1001
		1002	add $grp aio_stat $file, sub { ... };
		1003	};
		1004
		1005	limit $grp $num
		1006	Sets the feeder limit for the group: The feeder will be called
		1007	whenever the group contains less than this many requests.
		1008
		1009	Setting the limit to 0 will pause the feeding process.
		1010
		1011	The default value for the limit is 0, but note that setting a feeder
		1012	automatically bumps it up to 2.
		1013
299	SUPPORT FUNCTIONS	1014	SUPPORT FUNCTIONS
		1015	EVENT PROCESSING AND EVENT LOOP INTEGRATION
300	$fileno = IO::AIO::poll_fileno	1016	$fileno = IO::AIO::poll_fileno
301	Return the *request result pipe file descriptor*. This filehandle	1017	Return the *request result pipe file descriptor*. This filehandle
302	must be polled for reading by some mechanism outside this module	1018	must be polled for reading by some mechanism outside this module
303	(e.g. Event or select, see below or the SYNOPSIS). If the pipe	1019	(e.g. EV, Glib, select and so on, see below or the SYNOPSIS). If the
304	becomes readable you have to call "poll_cb" to check the results.	1020	pipe becomes readable you have to call "poll_cb" to check the
		1021	results.
305		1022
306	See "poll_cb" for an example.	1023	See "poll_cb" for an example.
307		1024
308	IO::AIO::poll_cb	1025	IO::AIO::poll_cb
309	Process all outstanding events on the result pipe. You have to call	1026	Process some outstanding events on the result pipe. You have to call
310	this regularly. Returns the number of events processed. Returns	1027	this regularly. Returns 0 if all events could be processed, or -1 if
311	immediately when no events are outstanding.	1028	it returned earlier for whatever reason. Returns immediately when no
		1029	events are outstanding. The amount of events processed depends on
		1030	the settings of "IO::AIO::max_poll_req" and
		1031	"IO::AIO::max_poll_time".
		1032
		1033	If not all requests were processed for whatever reason, the
		1034	filehandle will still be ready when "poll_cb" returns, so normally
		1035	you don't have to do anything special to have it called later.
312		1036
313	Example: Install an Event watcher that automatically calls	1037	Example: Install an Event watcher that automatically calls
314	IO::AIO::poll_cb with high priority:	1038	IO::AIO::poll_cb with high priority (more examples can be found in
		1039	the SYNOPSIS section, at the top of this document):
315		1040
316	Event->io (fd => IO::AIO::poll_fileno,	1041	Event->io (fd => IO::AIO::poll_fileno,
317	poll => 'r', async => 1,	1042	poll => 'r', async => 1,
318	cb => \&IO::AIO::poll_cb);	1043	cb => \&IO::AIO::poll_cb);
319		1044
320	IO::AIO::poll_wait	1045	IO::AIO::poll_wait
		1046	If there are any outstanding requests and none of them in the result
321	Wait till the result filehandle becomes ready for reading (simply	1047	phase, wait till the result filehandle becomes ready for reading
322	does a "select" on the filehandle. This is useful if you want to	1048	(simply does a "select" on the filehandle. This is useful if you
323	synchronously wait for some requests to finish).	1049	want to synchronously wait for some requests to finish).
324		1050
325	See "nreqs" for an example.	1051	See "nreqs" for an example.
326		1052
		1053	IO::AIO::poll
		1054	Waits until some requests have been handled.
		1055
		1056	Returns the number of requests processed, but is otherwise strictly
		1057	equivalent to:
		1058
		1059	IO::AIO::poll_wait, IO::AIO::poll_cb
		1060
327	IO::AIO::nreqs	1061	IO::AIO::flush
328	Returns the number of requests currently outstanding (i.e. for which	1062	Wait till all outstanding AIO requests have been handled.
329	their callback has not been invoked yet).
330		1063
331	Example: wait till there are no outstanding requests anymore:	1064	Strictly equivalent to:
332		1065
333	IO::AIO::poll_wait, IO::AIO::poll_cb	1066	IO::AIO::poll_wait, IO::AIO::poll_cb
334	while IO::AIO::nreqs;	1067	while IO::AIO::nreqs;
335		1068
336	IO::AIO::flush	1069	IO::AIO::max_poll_reqs $nreqs
337	Wait till all outstanding AIO requests have been handled.	1070	IO::AIO::max_poll_time $seconds
		1071	These set the maximum number of requests (default 0, meaning
		1072	infinity) that are being processed by "IO::AIO::poll_cb" in one
		1073	call, respectively the maximum amount of time (default 0, meaning
		1074	infinity) spent in "IO::AIO::poll_cb" to process requests (more
		1075	correctly the mininum amount of time "poll_cb" is allowed to use).
338		1076
339	Strictly equivalent to:	1077	Setting "max_poll_time" to a non-zero value creates an overhead of
		1078	one syscall per request processed, which is not normally a problem
		1079	unless your callbacks are really really fast or your OS is really
		1080	really slow (I am not mentioning Solaris here). Using
		1081	"max_poll_reqs" incurs no overhead.
340		1082
341	IO::AIO::poll_wait, IO::AIO::poll_cb	1083	Setting these is useful if you want to ensure some level of
342	while IO::AIO::nreqs;	1084	interactiveness when perl is not fast enough to process all requests
		1085	in time.
343		1086
344	IO::AIO::poll	1087	For interactive programs, values such as 0.01 to 0.1 should be fine.
345	Waits until some requests have been handled.
346		1088
347	Strictly equivalent to:	1089	Example: Install an Event watcher that automatically calls
		1090	IO::AIO::poll_cb with low priority, to ensure that other parts of
		1091	the program get the CPU sometimes even under high AIO load.
348		1092
349	IO::AIO::poll_wait, IO::AIO::poll_cb	1093	# try not to spend much more than 0.1s in poll_cb
350	if IO::AIO::nreqs;	1094	IO::AIO::max_poll_time 0.1;
351		1095
		1096	# use a low priority so other tasks have priority
		1097	Event->io (fd => IO::AIO::poll_fileno,
		1098	poll => 'r', nice => 1,
		1099	cb => &IO::AIO::poll_cb);
		1100
		1101	CONTROLLING THE NUMBER OF THREADS
352	IO::AIO::min_parallel $nthreads	1102	IO::AIO::min_parallel $nthreads
353	Set the minimum number of AIO threads to $nthreads. The current	1103	Set the minimum number of AIO threads to $nthreads. The current
354	default is 4, which means four asynchronous operations can be done	1104	default is 8, which means eight asynchronous operations can execute
355	at one time (the number of outstanding operations, however, is	1105	concurrently at any one time (the number of outstanding requests,
356	unlimited).	1106	however, is unlimited).
357		1107
358	IO::AIO starts threads only on demand, when an AIO request is queued	1108	IO::AIO starts threads only on demand, when an AIO request is queued
359	and no free thread exists.	1109	and no free thread exists. Please note that queueing up a hundred
		1110	requests can create demand for a hundred threads, even if it turns
		1111	out that everything is in the cache and could have been processed
		1112	faster by a single thread.
360		1113
361	It is recommended to keep the number of threads low, as some Linux	1114	It is recommended to keep the number of threads relatively low, as
362	kernel versions will scale negatively with the number of threads	1115	some Linux kernel versions will scale negatively with the number of
363	(higher parallelity => MUCH higher latency). With current Linux 2.6	1116	threads (higher parallelity => MUCH higher latency). With current
364	versions, 4-32 threads should be fine.	1117	Linux 2.6 versions, 4-32 threads should be fine.
365		1118
366	Under most circumstances you don't need to call this function, as	1119	Under most circumstances you don't need to call this function, as
367	the module selects a default that is suitable for low to moderate	1120	the module selects a default that is suitable for low to moderate
368	load.	1121	load.
369		1122
…		…
379	ensure that all threads are killed and that there are no outstanding	1132	ensure that all threads are killed and that there are no outstanding
380	requests.	1133	requests.
381		1134
382	Under normal circumstances you don't need to call this function.	1135	Under normal circumstances you don't need to call this function.
383		1136
		1137	IO::AIO::max_idle $nthreads
		1138	Limit the number of threads (default: 4) that are allowed to idle
		1139	(i.e., threads that did not get a request to process within the idle
		1140	timeout (default: 10 seconds). That means if a thread becomes idle
		1141	while $nthreads other threads are also idle, it will free its
		1142	resources and exit.
		1143
		1144	This is useful when you allow a large number of threads (e.g. 100 or
		1145	1000) to allow for extremely high load situations, but want to free
		1146	resources under normal circumstances (1000 threads can easily
		1147	consume 30MB of RAM).
		1148
		1149	The default is probably ok in most situations, especially if thread
		1150	creation is fast. If thread creation is very slow on your system you
		1151	might want to use larger values.
		1152
		1153	IO::AIO::idle_timeout $seconds
		1154	Sets the minimum idle timeout (default 10) after which worker
		1155	threads are allowed to exit. SEe "IO::AIO::max_idle".
		1156
384	$oldnreqs = IO::AIO::max_outstanding $nreqs	1157	IO::AIO::max_outstanding $maxreqs
		1158	This is a very bad function to use in interactive programs because
		1159	it blocks, and a bad way to reduce concurrency because it is
		1160	inexact: Better use an "aio_group" together with a feed callback.
		1161
385	Sets the maximum number of outstanding requests to $nreqs. If you	1162	Sets the maximum number of outstanding requests to $nreqs. If you do
386	try to queue up more than this number of requests, the caller will	1163	queue up more than this number of requests, the next call to the
387	block until some requests have been handled.	1164	"poll_cb" (and "poll_some" and other functions calling "poll_cb")
		1165	function will block until the limit is no longer exceeded.
388		1166
389	The default is very large, so normally there is no practical limit.	1167	The default value is very large, so there is no practical limit on
390	If you queue up many requests in a loop it often improves speed if	1168	the number of outstanding requests.
391	you set this to a relatively low number, such as 100.
392		1169
393	Under normal circumstances you don't need to call this function.	1170	You can still queue as many requests as you want. Therefore,
		1171	"max_outstanding" is mainly useful in simple scripts (with low
		1172	values) or as a stop gap to shield against fatal memory overflow
		1173	(with large values).
		1174
		1175	STATISTICAL INFORMATION
		1176	IO::AIO::nreqs
		1177	Returns the number of requests currently in the ready, execute or
		1178	pending states (i.e. for which their callback has not been invoked
		1179	yet).
		1180
		1181	Example: wait till there are no outstanding requests anymore:
		1182
		1183	IO::AIO::poll_wait, IO::AIO::poll_cb
		1184	while IO::AIO::nreqs;
		1185
		1186	IO::AIO::nready
		1187	Returns the number of requests currently in the ready state (not yet
		1188	executed).
		1189
		1190	IO::AIO::npending
		1191	Returns the number of requests currently in the pending state
		1192	(executed, but not yet processed by poll_cb).
		1193
		1194	MISCELLANEOUS FUNCTIONS
		1195	IO::AIO implements some functions that might be useful, but are not
		1196	asynchronous.
		1197
		1198	IO::AIO::sendfile $ofh, $ifh, $offset, $count
		1199	Calls the "eio_sendfile_sync" function, which is like
		1200	"aio_sendfile", but is blocking (this makes most sense if you know
		1201	the input data is likely cached already and the output filehandle is
		1202	set to non-blocking operations).
		1203
		1204	Returns the number of bytes copied, or -1 on error.
		1205
		1206	IO::AIO::fadvise $fh, $offset, $len, $advice
		1207	Simply calls the "posix_fadvise" function (see its manpage for
		1208	details). The following advice constants are avaiable:
		1209	"IO::AIO::FADV_NORMAL", "IO::AIO::FADV_SEQUENTIAL",
		1210	"IO::AIO::FADV_RANDOM", "IO::AIO::FADV_NOREUSE",
		1211	"IO::AIO::FADV_WILLNEED", "IO::AIO::FADV_DONTNEED".
		1212
		1213	On systems that do not implement "posix_fadvise", this function
		1214	returns ENOSYS, otherwise the return value of "posix_fadvise".
		1215
		1216	IO::AIO::madvise $scalar, $offset, $len, $advice
		1217	Simply calls the "posix_madvise" function (see its manpage for
		1218	details). The following advice constants are avaiable:
		1219	"IO::AIO::MADV_NORMAL", "IO::AIO::MADV_SEQUENTIAL",
		1220	"IO::AIO::MADV_RANDOM", "IO::AIO::MADV_WILLNEED",
		1221	"IO::AIO::MADV_DONTNEED".
		1222
		1223	On systems that do not implement "posix_madvise", this function
		1224	returns ENOSYS, otherwise the return value of "posix_madvise".
		1225
		1226	IO::AIO::mprotect $scalar, $offset, $len, $protect
		1227	Simply calls the "mprotect" function on the preferably AIO::mmap'ed
		1228	$scalar (see its manpage for details). The following protect
		1229	constants are avaiable: "IO::AIO::PROT_NONE", "IO::AIO::PROT_READ",
		1230	"IO::AIO::PROT_WRITE", "IO::AIO::PROT_EXEC".
		1231
		1232	On systems that do not implement "mprotect", this function returns
		1233	ENOSYS, otherwise the return value of "mprotect".
		1234
		1235	IO::AIO::mmap $scalar, $length, $prot, $flags, $fh[, $offset]
		1236	Memory-maps a file (or anonymous memory range) and attaches it to
		1237	the given $scalar, which will act like a string scalar.
		1238
		1239	The only operations allowed on the scalar are "substr"/"vec" that
		1240	don't change the string length, and most read-only operations such
		1241	as copying it or searching it with regexes and so on.
		1242
		1243	Anything else is unsafe and will, at best, result in memory leaks.
		1244
		1245	The memory map associated with the $scalar is automatically removed
		1246	when the $scalar is destroyed, or when the "IO::AIO::mmap" or
		1247	"IO::AIO::munmap" functions are called.
		1248
		1249	This calls the "mmap"(2) function internally. See your system's
		1250	manual page for details on the $length, $prot and $flags parameters.
		1251
		1252	The $length must be larger than zero and smaller than the actual
		1253	filesize.
		1254
		1255	$prot is a combination of "IO::AIO::PROT_NONE",
		1256	"IO::AIO::PROT_EXEC", "IO::AIO::PROT_READ" and/or
		1257	"IO::AIO::PROT_WRITE",
		1258
		1259	$flags can be a combination of "IO::AIO::MAP_SHARED" or
		1260	"IO::AIO::MAP_PRIVATE", or a number of system-specific flags (when
		1261	not available, the are defined as 0): "IO::AIO::MAP_ANONYMOUS"
		1262	(which is set to "MAP_ANON" if your system only provides this
		1263	constant), "IO::AIO::MAP_HUGETLB", "IO::AIO::MAP_LOCKED",
		1264	"IO::AIO::MAP_NORESERVE", "IO::AIO::MAP_POPULATE" or
		1265	"IO::AIO::MAP_NONBLOCK"
		1266
		1267	If $fh is "undef", then a file descriptor of -1 is passed.
		1268
		1269	$offset is the offset from the start of the file - it generally must
		1270	be a multiple of "IO::AIO::PAGESIZE" and defaults to 0.
		1271
		1272	Example:
		1273
		1274	use Digest::MD5;
		1275	use IO::AIO;
		1276
		1277	open my $fh, "<verybigfile"
		1278	or die "$!";
		1279
		1280	IO::AIO::mmap my $data, -s $fh, IO::AIO::PROT_READ, IO::AIO::MAP_SHARED, $fh
		1281	or die "verybigfile: $!";
		1282
		1283	my $fast_md5 = md5 $data;
		1284
		1285	IO::AIO::munmap $scalar
		1286	Removes a previous mmap and undefines the $scalar.
		1287
		1288	IO::AIO::munlock $scalar, $offset = 0, $length = undef
		1289	Calls the "munlock" function, undoing the effects of a previous
		1290	"aio_mlock" call (see its description for details).
		1291
		1292	IO::AIO::munlockall
		1293	Calls the "munlockall" function.
		1294
		1295	On systems that do not implement "munlockall", this function returns
		1296	ENOSYS, otherwise the return value of "munlockall".
		1297
		1298	EVENT LOOP INTEGRATION
		1299	It is recommended to use AnyEvent::AIO to integrate IO::AIO
		1300	automatically into many event loops:
		1301
		1302	# AnyEvent integration (EV, Event, Glib, Tk, POE, urxvt, pureperl...)
		1303	use AnyEvent::AIO;
		1304
		1305	You can also integrate IO::AIO manually into many event loops, here are
		1306	some examples of how to do this:
		1307
		1308	# EV integration
		1309	my $aio_w = EV::io IO::AIO::poll_fileno, EV::READ, \&IO::AIO::poll_cb;
		1310
		1311	# Event integration
		1312	Event->io (fd => IO::AIO::poll_fileno,
		1313	poll => 'r',
		1314	cb => \&IO::AIO::poll_cb);
		1315
		1316	# Glib/Gtk2 integration
		1317	add_watch Glib::IO IO::AIO::poll_fileno,
		1318	in => sub { IO::AIO::poll_cb; 1 };
		1319
		1320	# Tk integration
		1321	Tk::Event::IO->fileevent (IO::AIO::poll_fileno, "",
		1322	readable => \&IO::AIO::poll_cb);
		1323
		1324	# Danga::Socket integration
		1325	Danga::Socket->AddOtherFds (IO::AIO::poll_fileno =>
		1326	\&IO::AIO::poll_cb);
394		1327
395	FORK BEHAVIOUR	1328	FORK BEHAVIOUR
		1329	This module should do "the right thing" when the process using it forks:
		1330
396	Before the fork, IO::AIO enters a quiescent state where no requests can	1331	Before the fork, IO::AIO enters a quiescent state where no requests can
397	be added in other threads and no results will be processed. After the	1332	be added in other threads and no results will be processed. After the
398	fork the parent simply leaves the quiescent state and continues	1333	fork the parent simply leaves the quiescent state and continues
399	request/result processing, while the child clears the request/result	1334	request/result processing, while the child frees the request/result
400	queue (so the requests started before the fork will only be handled in	1335	queue (so that the requests started before the fork will only be handled
401	the parent). Threats will be started on demand until the limit ste in	1336	in the parent). Threads will be started on demand until the limit set in
402	the parent process has been reached again.	1337	the parent process has been reached again.
403		1338
		1339	In short: the parent will, after a short pause, continue as if fork had
		1340	not been called, while the child will act as if IO::AIO has not been
		1341	used yet.
		1342
		1343	MEMORY USAGE
		1344	Per-request usage:
		1345
		1346	Each aio request uses - depending on your architecture - around 100-200
		1347	bytes of memory. In addition, stat requests need a stat buffer (possibly
		1348	a few hundred bytes), readdir requires a result buffer and so on. Perl
		1349	scalars and other data passed into aio requests will also be locked and
		1350	will consume memory till the request has entered the done state.
		1351
		1352	This is not awfully much, so queuing lots of requests is not usually a
		1353	problem.
		1354
		1355	Per-thread usage:
		1356
		1357	In the execution phase, some aio requests require more memory for
		1358	temporary buffers, and each thread requires a stack and other data
		1359	structures (usually around 16k-128k, depending on the OS).
		1360
		1361	KNOWN BUGS
		1362	Known bugs will be fixed in the next release.
		1363
404	SEE ALSO	1364	SEE ALSO
405	Coro, Linux::AIO.	1365	AnyEvent::AIO for easy integration into event loops, Coro::AIO for a
		1366	more natural syntax.
406		1367
407	AUTHOR	1368	AUTHOR
408	Marc Lehmann <schmorp@schmorp.de>	1369	Marc Lehmann <schmorp@schmorp.de>
409	http://home.schmorp.de/	1370	http://home.schmorp.de/
410		1371

Diff Legend

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