[ViewVC] Diff of: cvs/IO-AIO/README

Comparing IO-AIO/README (file contents):
Revision 1.34 by root, Thu Nov 20 09:01:40 2008 UTC vs.
Revision 1.49 by root, Mon Jul 18 03:09:06 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 = shift	8	my $fh = shift
9	or die "/etc/passwd: $!";	9	or die "/etc/passwd: $!";
10	...	10	...
11	};	11	};
12		12
…		…
24	$req->cancel; # cancel request if still in queue	24	$req->cancel; # cancel request if still in queue
25		25
26	my $grp = aio_group sub { print "all stats done\n" };	26	my $grp = aio_group sub { print "all stats done\n" };
27	add $grp aio_stat "..." for ...;	27	add $grp aio_stat "..." for ...;
28		28
29	# AnyEvent integration (EV, Event, Glib, Tk, POE, urxvt, pureperl...)
30	use AnyEvent::AIO;
31
32	# EV integration
33	my $w = EV::io IO::AIO::poll_fileno, EV::READ, \&IO::AIO::poll_cb;
34
35	# Event integration
36	Event->io (fd => IO::AIO::poll_fileno,
37	poll => 'r',
38	cb => \&IO::AIO::poll_cb);
39
40	# Glib/Gtk2 integration
41	add_watch Glib::IO IO::AIO::poll_fileno,
42	in => sub { IO::AIO::poll_cb; 1 };
43
44	# Tk integration
45	Tk::Event::IO->fileevent (IO::AIO::poll_fileno, "",
46	readable => \&IO::AIO::poll_cb);
47
48	# Danga::Socket integration
49	Danga::Socket->AddOtherFds (IO::AIO::poll_fileno =>
50	\&IO::AIO::poll_cb);
51
52	DESCRIPTION	29	DESCRIPTION
53	This module implements asynchronous I/O using whatever means your	30	This module implements asynchronous I/O using whatever means your
54	operating system supports.	31	operating system supports. It is implemented as an interface to "libeio"
		32	(<http://software.schmorp.de/pkg/libeio.html>).
55		33
56	Asynchronous means that operations that can normally block your program	34	Asynchronous means that operations that can normally block your program
57	(e.g. reading from disk) will be done asynchronously: the operation will	35	(e.g. reading from disk) will be done asynchronously: the operation will
58	still block, but you can do something else in the meantime. This is	36	still block, but you can do something else in the meantime. This is
59	extremely useful for programs that need to stay interactive even when	37	extremely useful for programs that need to stay interactive even when
…		…
64	operations concurrently.	42	operations concurrently.
65		43
66	While most of this works on all types of file descriptors (for example	44	While most of this works on all types of file descriptors (for example
67	sockets), using these functions on file descriptors that support	45	sockets), using these functions on file descriptors that support
68	nonblocking operation (again, sockets, pipes etc.) is very inefficient.	46	nonblocking operation (again, sockets, pipes etc.) is very inefficient.
69	Use an event loop for that (such as the Event module): IO::AIO will	47	Use an event loop for that (such as the EV module): IO::AIO will
70	naturally fit into such an event loop itself.	48	naturally fit into such an event loop itself.
71		49
72	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
73	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
74	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
…		…
83	it is currently not reentrant in any way, so use appropriate locking	61	it is currently not reentrant in any way, so use appropriate locking
84	yourself, always call "poll_cb" from within the same thread, or never	62	yourself, always call "poll_cb" from within the same thread, or never
85	call "poll_cb" (or other "aio_" functions) recursively.	63	call "poll_cb" (or other "aio_" functions) recursively.
86		64
87	EXAMPLE	65	EXAMPLE
88	This is a simple example that uses the Event module and loads	66	This is a simple example that uses the EV module and loads /etc/passwd
89	/etc/passwd asynchronously:	67	asynchronously:
90		68
91	use Fcntl;	69	use Fcntl;
92	use Event;	70	use EV;
93	use IO::AIO;	71	use IO::AIO;
94		72
95	# register the IO::AIO callback with Event	73	# register the IO::AIO callback with EV
96	Event->io (fd => IO::AIO::poll_fileno,	74	my $aio_w = EV::io IO::AIO::poll_fileno, EV::READ, \&IO::AIO::poll_cb;
97	poll => 'r',
98	cb => \&IO::AIO::poll_cb);
99		75
100	# queue the request to open /etc/passwd	76	# queue the request to open /etc/passwd
101	aio_open "/etc/passwd", O_RDONLY, 0, sub {	77	aio_open "/etc/passwd", IO::AIO::O_RDONLY, 0, sub {
102	my $fh = shift	78	my $fh = shift
103	or die "error while opening: $!";	79	or die "error while opening: $!";
104		80
105	# stat'ing filehandles is generally non-blocking	81	# stat'ing filehandles is generally non-blocking
106	my $size = -s $fh;	82	my $size = -s $fh;
…		…
115		91
116	# file contents now in $contents	92	# file contents now in $contents
117	print $contents;	93	print $contents;
118		94
119	# exit event loop and program	95	# exit event loop and program
120	Event::unloop;	96	EV::unloop;
121	};	97	};
122	};	98	};
123		99
124	# possibly queue up other requests, or open GUI windows,	100	# possibly queue up other requests, or open GUI windows,
125	# check for sockets etc. etc.	101	# check for sockets etc. etc.
126		102
127	# process events as long as there are some:	103	# process events as long as there are some:
128	Event::loop;	104	EV::loop;
129		105
130	REQUEST ANATOMY AND LIFETIME	106	REQUEST ANATOMY AND LIFETIME
131	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
132	not directly visible to Perl.	108	not directly visible to Perl.
133		109
…		…
169	anymore (except possibly for the Perl object, but its connection to	145	anymore (except possibly for the Perl object, but its connection to
170	the actual aio request is severed and calling its methods will	146	the actual aio request is severed and calling its methods will
171	either do nothing or result in a runtime error).	147	either do nothing or result in a runtime error).
172		148
173	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_realpath $path, $callback->($link)
		174	aio_rename $srcpath, $dstpath, $callback->($status)
		175	aio_mkdir $pathname, $mode, $callback->($status)
		176	aio_rmdir $pathname, $callback->($status)
		177	aio_readdir $pathname, $callback->($entries)
		178	aio_readdirx $pathname, $flags, $callback->($entries, $flags)
		179	IO::AIO::READDIR_DENTS IO::AIO::READDIR_DIRS_FIRST
		180	IO::AIO::READDIR_STAT_ORDER IO::AIO::READDIR_FOUND_UNKNOWN
		181	aio_load $path, $data, $callback->($status)
		182	aio_copy $srcpath, $dstpath, $callback->($status)
		183	aio_move $srcpath, $dstpath, $callback->($status)
		184	aio_scandir $path, $maxreq, $callback->($dirs, $nondirs)
		185	aio_rmtree $path, $callback->($status)
		186	aio_sync $callback->($status)
		187	aio_fsync $fh, $callback->($status)
		188	aio_fdatasync $fh, $callback->($status)
		189	aio_sync_file_range $fh, $offset, $nbytes, $flags, $callback->($status)
		190	aio_pathsync $path, $callback->($status)
		191	aio_msync $scalar, $offset = 0, $length = undef, flags = 0, $callback->($status)
		192	aio_mtouch $scalar, $offset = 0, $length = undef, flags = 0, $callback->($status)
		193	aio_mlock $scalar, $offset = 0, $length = undef, $callback->($status)
		194	aio_mlockall $flags, $callback->($status)
		195	aio_group $callback->(...)
		196	aio_nop $callback->()
		197
		198	$prev_pri = aioreq_pri [$pri]
		199	aioreq_nice $pri_adjust
		200
		201	IO::AIO::poll_wait
		202	IO::AIO::poll_cb
		203	IO::AIO::poll
		204	IO::AIO::flush
		205	IO::AIO::max_poll_reqs $nreqs
		206	IO::AIO::max_poll_time $seconds
		207	IO::AIO::min_parallel $nthreads
		208	IO::AIO::max_parallel $nthreads
		209	IO::AIO::max_idle $nthreads
		210	IO::AIO::idle_timeout $seconds
		211	IO::AIO::max_outstanding $maxreqs
		212	IO::AIO::nreqs
		213	IO::AIO::nready
		214	IO::AIO::npending
		215
		216	IO::AIO::sendfile $ofh, $ifh, $offset, $count
		217	IO::AIO::fadvise $fh, $offset, $len, $advice
		218	IO::AIO::madvise $scalar, $offset, $length, $advice
		219	IO::AIO::mprotect $scalar, $offset, $length, $protect
		220	IO::AIO::munlock $scalar, $offset = 0, $length = undef
		221	IO::AIO::munlockall
		222
174	AIO REQUEST FUNCTIONS	223	AIO REQUEST FUNCTIONS
175	All the "aio_*" calls are more or less thin wrappers around the syscall	224	All the "aio_*" calls are more or less thin wrappers around the syscall
176	with the same name (sans "aio_"). The arguments are similar or	225	with the same name (sans "aio_"). The arguments are similar or
177	identical, and they all accept an additional (and optional) $callback	226	identical, and they all accept an additional (and optional) $callback
178	argument which must be a code reference. This code reference will get	227	argument which must be a code reference. This code reference will get
…		…
249	will be modified by the umask in effect then the request is being	298	will be modified by the umask in effect then the request is being
250	executed, so better never change the umask.	299	executed, so better never change the umask.
251		300
252	Example:	301	Example:
253		302
254	aio_open "/etc/passwd", O_RDONLY, 0, sub {	303	aio_open "/etc/passwd", IO::AIO::O_RDONLY, 0, sub {
255	if ($_[0]) {	304	if ($_[0]) {
256	print "open successful, fh is $_[0]\n";	305	print "open successful, fh is $_[0]\n";
257	...	306	...
258	} else {	307	} else {
259	die "open failed: $!\n";	308	die "open failed: $!\n";
260	}	309	}
261	};	310	};
262		311
		312	In addition to all the common open modes/flags ("O_RDONLY",
		313	"O_WRONLY", "O_RDWR", "O_CREAT", "O_TRUNC", "O_EXCL" and
		314	"O_APPEND"), the following POSIX and non-POSIX constants are
		315	available (missing ones on your system are, as usual, 0):
		316
		317	"O_ASYNC", "O_DIRECT", "O_NOATIME", "O_CLOEXEC", "O_NOCTTY",
		318	"O_NOFOLLOW", "O_NONBLOCK", "O_EXEC", "O_SEARCH", "O_DIRECTORY",
		319	"O_DSYNC", "O_RSYNC", "O_SYNC" and "O_TTY_INIT".
		320
263	aio_close $fh, $callback->($status)	321	aio_close $fh, $callback->($status)
264	Asynchronously close a file and call the callback with the result	322	Asynchronously close a file and call the callback with the result
265	code.	323	code.
266		324
267	Unfortunately, you can't do this to perl. Perl insists very	325	Unfortunately, you can't do this to perl. Perl insists very
…		…
275	Or in other words: the file descriptor will be closed, but it will	333	Or in other words: the file descriptor will be closed, but it will
276	not be free for reuse until the perl filehandle is closed.	334	not be free for reuse until the perl filehandle is closed.
277		335
278	aio_read $fh,$offset,$length, $data,$dataoffset, $callback->($retval)	336	aio_read $fh,$offset,$length, $data,$dataoffset, $callback->($retval)
279	aio_write $fh,$offset,$length, $data,$dataoffset, $callback->($retval)	337	aio_write $fh,$offset,$length, $data,$dataoffset, $callback->($retval)
280	Reads or writes $length bytes from the specified $fh and $offset	338	Reads or writes $length bytes from or to the specified $fh and
281	into the scalar given by $data and offset $dataoffset and calls the	339	$offset into the scalar given by $data and offset $dataoffset and
282	callback without the actual number of bytes read (or -1 on error,	340	calls the callback without the actual number of bytes read (or -1 on
283	just like the syscall).	341	error, just like the syscall).
		342
		343	"aio_read" will, like "sysread", shrink or grow the $data scalar to
		344	offset plus the actual number of bytes read.
284		345
285	If $offset is undefined, then the current file descriptor offset	346	If $offset is undefined, then the current file descriptor offset
286	will be used (and updated), otherwise the file descriptor offset	347	will be used (and updated), otherwise the file descriptor offset
287	will not be changed by these calls.	348	will not be changed by these calls.
288		349
…		…
307	aio_sendfile $out_fh, $in_fh, $in_offset, $length, $callback->($retval)	368	aio_sendfile $out_fh, $in_fh, $in_offset, $length, $callback->($retval)
308	Tries to copy $length bytes from $in_fh to $out_fh. It starts	369	Tries to copy $length bytes from $in_fh to $out_fh. It starts
309	reading at byte offset $in_offset, and starts writing at the current	370	reading at byte offset $in_offset, and starts writing at the current
310	file offset of $out_fh. Because of that, it is not safe to issue	371	file offset of $out_fh. Because of that, it is not safe to issue
311	more than one "aio_sendfile" per $out_fh, as they will interfere	372	more than one "aio_sendfile" per $out_fh, as they will interfere
312	with each other.	373	with each other. The same $in_fh works fine though, as this function
		374	does not move or use the file offset of $in_fh.
313		375
		376	Please note that "aio_sendfile" can read more bytes from $in_fh than
		377	are written, and there is no way to find out how many more bytes
		378	have been read from "aio_sendfile" alone, as "aio_sendfile" only
		379	provides the number of bytes written to $out_fh. Only if the result
		380	value equals $length one can assume that $length bytes have been
		381	read.
		382
		383	Unlike with other "aio_" functions, it makes a lot of sense to use
		384	"aio_sendfile" on non-blocking sockets, as long as one end
		385	(typically the $in_fh) is a file - the file I/O will then be
		386	asynchronous, while the socket I/O will be non-blocking. Note,
		387	however, that you can run into a trap where "aio_sendfile" reads
		388	some data with readahead, then fails to write all data, and when the
		389	socket is ready the next time, the data in the cache is already
		390	lost, forcing "aio_sendfile" to again hit the disk. Explicit
		391	"aio_read" + "aio_write" let's you better control resource usage.
		392
314	This call tries to make use of a native "sendfile" syscall to	393	This call tries to make use of a native "sendfile"-like syscall to
315	provide zero-copy operation. For this to work, $out_fh should refer	394	provide zero-copy operation. For this to work, $out_fh should refer
316	to a socket, and $in_fh should refer to mmap'able file.	395	to a socket, and $in_fh should refer to an mmap'able file.
317		396
318	If the native sendfile call fails or is not implemented, it will be	397	If a native sendfile cannot be found or it fails with "ENOSYS",
319	emulated, so you can call "aio_sendfile" on any type of filehandle	398	"EINVAL", "ENOTSUP", "EOPNOTSUPP", "EAFNOSUPPORT", "EPROTOTYPE" or
320	regardless of the limitations of the operating system.	399	"ENOTSOCK", it will be emulated, so you can call "aio_sendfile" on
		400	any type of filehandle regardless of the limitations of the
		401	operating system.
321		402
322	Please note, however, that "aio_sendfile" can read more bytes from	403	As native sendfile syscalls (as practically any non-POSIX interface
323	$in_fh than are written, and there is no way to find out how many	404	hacked together in a hurry to improve benchmark numbers) tend to be
324	bytes have been read from "aio_sendfile" alone, as "aio_sendfile"	405	rather buggy on many systems, this implementation tries to work
325	only provides the number of bytes written to $out_fh. Only if the	406	around some known bugs in Linux and FreeBSD kernels (probably
326	result value equals $length one can assume that $length bytes have	407	others, too), but that might fail, so you really really should check
327	been read.	408	the return value of "aio_sendfile" - fewre bytes than expected might
		409	have been transferred.
328		410
329	aio_readahead $fh,$offset,$length, $callback->($retval)	411	aio_readahead $fh,$offset,$length, $callback->($retval)
330	"aio_readahead" populates the page cache with data from a file so	412	"aio_readahead" populates the page cache with data from a file so
331	that subsequent reads from that file will not block on disk I/O. The	413	that subsequent reads from that file will not block on disk I/O. The
332	$offset argument specifies the starting point from which data is to	414	$offset argument specifies the starting point from which data is to
…		…
353	Currently, the stats are always 64-bit-stats, i.e. instead of	435	Currently, the stats are always 64-bit-stats, i.e. instead of
354	returning an error when stat'ing a large file, the results will be	436	returning an error when stat'ing a large file, the results will be
355	silently truncated unless perl itself is compiled with large file	437	silently truncated unless perl itself is compiled with large file
356	support.	438	support.
357		439
		440	To help interpret the mode and dev/rdev stat values, IO::AIO offers
		441	the following constants and functions (if not implemented, the
		442	constants will be 0 and the functions will either "croak" or fall
		443	back on traditional behaviour).
		444
		445	"S_IFMT", "S_IFIFO", "S_IFCHR", "S_IFBLK", "S_IFLNK", "S_IFREG",
		446	"S_IFDIR", "S_IFWHT", "S_IFSOCK", "IO::AIO::major $dev_t",
		447	"IO::AIO::minor $dev_t", "IO::AIO::makedev $major, $minor".
		448
358	Example: Print the length of /etc/passwd:	449	Example: Print the length of /etc/passwd:
359		450
360	aio_stat "/etc/passwd", sub {	451	aio_stat "/etc/passwd", sub {
361	$_[0] and die "stat failed: $!";	452	$_[0] and die "stat failed: $!";
362	print "size is ", -s _, "\n";	453	print "size is ", -s _, "\n";
363	};	454	};
364		455
		456	aio_statvfs $fh_or_path, $callback->($statvfs)
		457	Works like the POSIX "statvfs" or "fstatvfs" syscalls, depending on
		458	whether a file handle or path was passed.
		459
		460	On success, the callback is passed a hash reference with the
		461	following members: "bsize", "frsize", "blocks", "bfree", "bavail",
		462	"files", "ffree", "favail", "fsid", "flag" and "namemax". On
		463	failure, "undef" is passed.
		464
		465	The following POSIX IO::AIO::ST_* constants are defined: "ST_RDONLY"
		466	and "ST_NOSUID".
		467
		468	The following non-POSIX IO::AIO::ST_* flag masks are defined to
		469	their correct value when available, or to 0 on systems that do not
		470	support them: "ST_NODEV", "ST_NOEXEC", "ST_SYNCHRONOUS",
		471	"ST_MANDLOCK", "ST_WRITE", "ST_APPEND", "ST_IMMUTABLE",
		472	"ST_NOATIME", "ST_NODIRATIME" and "ST_RELATIME".
		473
		474	Example: stat "/wd" and dump out the data if successful.
		475
		476	aio_statvfs "/wd", sub {
		477	my $f = $_[0]
		478	or die "statvfs: $!";
		479
		480	use Data::Dumper;
		481	say Dumper $f;
		482	};
		483
		484	# result:
		485	{
		486	bsize => 1024,
		487	bfree => 4333064312,
		488	blocks => 10253828096,
		489	files => 2050765568,
		490	flag => 4096,
		491	favail => 2042092649,
		492	bavail => 4333064312,
		493	ffree => 2042092649,
		494	namemax => 255,
		495	frsize => 1024,
		496	fsid => 1810
		497	}
		498
365	aio_utime $fh_or_path, $atime, $mtime, $callback->($status)	499	aio_utime $fh_or_path, $atime, $mtime, $callback->($status)
366	Works like perl's "utime" function (including the special case of	500	Works like perl's "utime" function (including the special case of
367	$atime and $mtime being undef). Fractional times are supported if	501	$atime and $mtime being undef). Fractional times are supported if
368	the underlying syscalls support them.	502	the underlying syscalls support them.
369		503
…		…
407		541
408	The only (POSIX-) portable way of calling this function is:	542	The only (POSIX-) portable way of calling this function is:
409		543
410	aio_mknod $path, IO::AIO::S_IFIFO \| $mode, 0, sub { ...	544	aio_mknod $path, IO::AIO::S_IFIFO \| $mode, 0, sub { ...
411		545
		546	See "aio_stat" for info about some potentially helpful extra
		547	constants and functions.
		548
412	aio_link $srcpath, $dstpath, $callback->($status)	549	aio_link $srcpath, $dstpath, $callback->($status)
413	Asynchronously create a new link to the existing object at $srcpath	550	Asynchronously create a new link to the existing object at $srcpath
414	at the path $dstpath and call the callback with the result code.	551	at the path $dstpath and call the callback with the result code.
415		552
416	aio_symlink $srcpath, $dstpath, $callback->($status)	553	aio_symlink $srcpath, $dstpath, $callback->($status)
…		…
421	aio_readlink $path, $callback->($link)	558	aio_readlink $path, $callback->($link)
422	Asynchronously read the symlink specified by $path and pass it to	559	Asynchronously read the symlink specified by $path and pass it to
423	the callback. If an error occurs, nothing or undef gets passed to	560	the callback. If an error occurs, nothing or undef gets passed to
424	the callback.	561	the callback.
425		562
		563	aio_realpath $path, $callback->($path)
		564	Asynchronously make the path absolute and resolve any symlinks in
		565	$path. The resulting path only consists of directories (Same as
		566	Cwd::realpath).
		567
		568	This request can be used to get the absolute path of the current
		569	working directory by passing it a path of . (a single dot).
		570
426	aio_rename $srcpath, $dstpath, $callback->($status)	571	aio_rename $srcpath, $dstpath, $callback->($status)
427	Asynchronously rename the object at $srcpath to $dstpath, just as	572	Asynchronously rename the object at $srcpath to $dstpath, just as
428	rename(2) and call the callback with the result code.	573	rename(2) and call the callback with the result code.
429		574
430	aio_mkdir $pathname, $mode, $callback->($status)	575	aio_mkdir $pathname, $mode, $callback->($status)
…		…
439	aio_readdir $pathname, $callback->($entries)	584	aio_readdir $pathname, $callback->($entries)
440	Unlike the POSIX call of the same name, "aio_readdir" reads an	585	Unlike the POSIX call of the same name, "aio_readdir" reads an
441	entire directory (i.e. opendir + readdir + closedir). The entries	586	entire directory (i.e. opendir + readdir + closedir). The entries
442	will not be sorted, and will NOT include the "." and ".." entries.	587	will not be sorted, and will NOT include the "." and ".." entries.
443		588
444	The callback a single argument which is either "undef" or an	589	The callback is passed a single argument which is either "undef" or
445	array-ref with the filenames.	590	an array-ref with the filenames.
		591
		592	aio_readdirx $pathname, $flags, $callback->($entries, $flags)
		593	Quite similar to "aio_readdir", but the $flags argument allows to
		594	tune behaviour and output format. In case of an error, $entries will
		595	be "undef".
		596
		597	The flags are a combination of the following constants, ORed
		598	together (the flags will also be passed to the callback, possibly
		599	modified):
		600
		601	IO::AIO::READDIR_DENTS
		602	When this flag is off, then the callback gets an arrayref
		603	consisting of names only (as with "aio_readdir"), otherwise it
		604	gets an arrayref with "[$name, $type, $inode]" arrayrefs, each
		605	describing a single directory entry in more detail.
		606
		607	$name is the name of the entry.
		608
		609	$type is one of the "IO::AIO::DT_xxx" constants:
		610
		611	"IO::AIO::DT_UNKNOWN", "IO::AIO::DT_FIFO", "IO::AIO::DT_CHR",
		612	"IO::AIO::DT_DIR", "IO::AIO::DT_BLK", "IO::AIO::DT_REG",
		613	"IO::AIO::DT_LNK", "IO::AIO::DT_SOCK", "IO::AIO::DT_WHT".
		614
		615	"IO::AIO::DT_UNKNOWN" means just that: readdir does not know. If
		616	you need to know, you have to run stat yourself. Also, for speed
		617	reasons, the $type scalars are read-only: you can not modify
		618	them.
		619
		620	$inode is the inode number (which might not be exact on systems
		621	with 64 bit inode numbers and 32 bit perls). This field has
		622	unspecified content on systems that do not deliver the inode
		623	information.
		624
		625	IO::AIO::READDIR_DIRS_FIRST
		626	When this flag is set, then the names will be returned in an
		627	order where likely directories come first, in optimal stat
		628	order. This is useful when you need to quickly find directories,
		629	or you want to find all directories while avoiding to stat()
		630	each entry.
		631
		632	If the system returns type information in readdir, then this is
		633	used to find directories directly. Otherwise, likely directories
		634	are names beginning with ".", or otherwise names with no dots,
		635	of which names with short names are tried first.
		636
		637	IO::AIO::READDIR_STAT_ORDER
		638	When this flag is set, then the names will be returned in an
		639	order suitable for stat()'ing each one. That is, when you plan
		640	to stat() all files in the given directory, then the returned
		641	order will likely be fastest.
		642
		643	If both this flag and "IO::AIO::READDIR_DIRS_FIRST" are
		644	specified, then the likely dirs come first, resulting in a less
		645	optimal stat order.
		646
		647	IO::AIO::READDIR_FOUND_UNKNOWN
		648	This flag should not be set when calling "aio_readdirx".
		649	Instead, it is being set by "aio_readdirx", when any of the
		650	$type's found were "IO::AIO::DT_UNKNOWN". The absense of this
		651	flag therefore indicates that all $type's are known, which can
		652	be used to speed up some algorithms.
446		653
447	aio_load $path, $data, $callback->($status)	654	aio_load $path, $data, $callback->($status)
448	This is a composite request that tries to fully load the given file	655	This is a composite request that tries to fully load the given file
449	into memory. Status is the same as with aio_read.	656	into memory. Status is the same as with aio_read.
450		657
451	aio_copy $srcpath, $dstpath, $callback->($status)	658	aio_copy $srcpath, $dstpath, $callback->($status)
452	Try to copy the file (directories not supported as either source	659	Try to copy the file (directories not supported as either source
453	or destination) from $srcpath to $dstpath and call the callback with	660	or destination) from $srcpath to $dstpath and call the callback with
454	the 0 (error) or -1 ok.	661	a status of 0 (ok) or -1 (error, see $!).
455		662
456	This is a composite request that creates the destination file with	663	This is a composite request that creates the destination file with
457	mode 0200 and copies the contents of the source file into it using	664	mode 0200 and copies the contents of the source file into it using
458	"aio_sendfile", followed by restoring atime, mtime, access mode and	665	"aio_sendfile", followed by restoring atime, mtime, access mode and
459	uid/gid, in that order.	666	uid/gid, in that order.
…		…
463	uid/gid, where errors are being ignored.	670	uid/gid, where errors are being ignored.
464		671
465	aio_move $srcpath, $dstpath, $callback->($status)	672	aio_move $srcpath, $dstpath, $callback->($status)
466	Try to move the file (directories not supported as either source	673	Try to move the file (directories not supported as either source
467	or destination) from $srcpath to $dstpath and call the callback with	674	or destination) from $srcpath to $dstpath and call the callback with
468	the 0 (error) or -1 ok.	675	a status of 0 (ok) or -1 (error, see $!).
469		676
470	This is a composite request that tries to rename(2) the file first;	677	This is a composite request that tries to rename(2) the file first;
471	if rename fails with "EXDEV", it copies the file with "aio_copy"	678	if rename fails with "EXDEV", it copies the file with "aio_copy"
472	and, if that is successful, unlinks the $srcpath.	679	and, if that is successful, unlinks the $srcpath.
473		680
…		…
497	Implementation notes.	704	Implementation notes.
498		705
499	The "aio_readdir" cannot be avoided, but "stat()"'ing every entry	706	The "aio_readdir" cannot be avoided, but "stat()"'ing every entry
500	can.	707	can.
501		708
		709	If readdir returns file type information, then this is used directly
		710	to find directories.
		711
502	After reading the directory, the modification time, size etc. of the	712	Otherwise, after reading the directory, the modification time, size
503	directory before and after the readdir is checked, and if they match	713	etc. of the directory before and after the readdir is checked, and
504	(and isn't the current time), the link count will be used to decide	714	if they match (and isn't the current time), the link count will be
505	how many entries are directories (if >= 2). Otherwise, no knowledge	715	used to decide how many entries are directories (if >= 2).
506	of the number of subdirectories will be assumed.	716	Otherwise, no knowledge of the number of subdirectories will be
		717	assumed.
507		718
508	Then entries will be sorted into likely directories (everything	719	Then entries will be sorted into likely directories a non-initial
509	without a non-initial dot currently) and likely non-directories	720	dot currently) and likely non-directories (see "aio_readdirx"). Then
510	(everything else). Then every entry plus an appended "/." will be	721	every entry plus an appended "/." will be "stat"'ed, likely
511	"stat"'ed, likely directories first. If that succeeds, it assumes	722	directories first, in order of their inode numbers. If that
512	that the entry is a directory or a symlink to directory (which will	723	succeeds, it assumes that the entry is a directory or a symlink to
513	be checked seperately). This is often faster than stat'ing the entry	724	directory (which will be checked seperately). This is often faster
514	itself because filesystems might detect the type of the entry	725	than stat'ing the entry itself because filesystems might detect the
515	without reading the inode data (e.g. ext2fs filetype feature).	726	type of the entry without reading the inode data (e.g. ext2fs
		727	filetype feature), even on systems that cannot return the filetype
		728	information on readdir.
516		729
517	If the known number of directories (link count - 2) has been	730	If the known number of directories (link count - 2) has been
518	reached, the rest of the entries is assumed to be non-directories.	731	reached, the rest of the entries is assumed to be non-directories.
519		732
520	This only works with certainty on POSIX (= UNIX) filesystems, which	733	This only works with certainty on POSIX (= UNIX) filesystems, which
…		…
562	operations (E.g. rename). This might not work on all operating	775	operations (E.g. rename). This might not work on all operating
563	systems or have any specific effect, but usually it makes sure that	776	systems or have any specific effect, but usually it makes sure that
564	directory changes get written to disc. It works for anything that	777	directory changes get written to disc. It works for anything that
565	can be opened for read-only, not just directories.	778	can be opened for read-only, not just directories.
566		779
		780	Future versions of this function might fall back to other methods
		781	when "fsync" on the directory fails (such as calling "sync").
		782
567	Passes 0 when everything went ok, and -1 on error.	783	Passes 0 when everything went ok, and -1 on error.
		784
		785	aio_msync $scalar, $offset = 0, $length = undef, flags = 0,
		786	$callback->($status)
		787	This is a rather advanced IO::AIO call, which only works on
		788	mmap(2)ed scalars (see the "IO::AIO::mmap" function, although it
		789	also works on data scalars managed by the Sys::Mmap or Mmap modules,
		790	note that the scalar must only be modified in-place while an aio
		791	operation is pending on it).
		792
		793	It calls the "msync" function of your OS, if available, with the
		794	memory area starting at $offset in the string and ending $length
		795	bytes later. If $length is negative, counts from the end, and if
		796	$length is "undef", then it goes till the end of the string. The
		797	flags can be a combination of "IO::AIO::MS_ASYNC",
		798	"IO::AIO::MS_INVALIDATE" and "IO::AIO::MS_SYNC".
		799
		800	aio_mtouch $scalar, $offset = 0, $length = undef, flags = 0,
		801	$callback->($status)
		802	This is a rather advanced IO::AIO call, which works best on
		803	mmap(2)ed scalars.
		804
		805	It touches (reads or writes) all memory pages in the specified range
		806	inside the scalar. All caveats and parameters are the same as for
		807	"aio_msync", above, except for flags, which must be either 0 (which
		808	reads all pages and ensures they are instantiated) or
		809	"IO::AIO::MT_MODIFY", which modifies the memory page s(by reading
		810	and writing an octet from it, which dirties the page).
		811
		812	aio_mlock $scalar, $offset = 0, $length = undef, $callback->($status)
		813	This is a rather advanced IO::AIO call, which works best on
		814	mmap(2)ed scalars.
		815
		816	It reads in all the pages of the underlying storage into memory (if
		817	any) and locks them, so they are not getting swapped/paged out or
		818	removed.
		819
		820	If $length is undefined, then the scalar will be locked till the
		821	end.
		822
		823	On systems that do not implement "mlock", this function returns -1
		824	and sets errno to "ENOSYS".
		825
		826	Note that the corresponding "munlock" is synchronous and is
		827	documented under "MISCELLANEOUS FUNCTIONS".
		828
		829	Example: open a file, mmap and mlock it - both will be undone when
		830	$data gets destroyed.
		831
		832	open my $fh, "<", $path or die "$path: $!";
		833	my $data;
		834	IO::AIO::mmap $data, -s $fh, IO::AIO::PROT_READ, IO::AIO::MAP_SHARED, $fh;
		835	aio_mlock $data; # mlock in background
		836
		837	aio_mlockall $flags, $callback->($status)
		838	Calls the "mlockall" function with the given $flags (a combination
		839	of "IO::AIO::MCL_CURRENT" and "IO::AIO::MCL_FUTURE").
		840
		841	On systems that do not implement "mlockall", this function returns
		842	-1 and sets errno to "ENOSYS".
		843
		844	Note that the corresponding "munlockall" is synchronous and is
		845	documented under "MISCELLANEOUS FUNCTIONS".
		846
		847	Example: asynchronously lock all current and future pages into
		848	memory.
		849
		850	aio_mlockall IO::AIO::MCL_FUTURE;
568		851
569	aio_group $callback->(...)	852	aio_group $callback->(...)
570	This is a very special aio request: Instead of doing something, it	853	This is a very special aio request: Instead of doing something, it
571	is a container for other aio requests, which is useful if you want	854	is a container for other aio requests, which is useful if you want
572	to bundle many requests into a single, composite, request with a	855	to bundle many requests into a single, composite, request with a
…		…
615		898
616	cancel $req	899	cancel $req
617	Cancels the request, if possible. Has the effect of skipping	900	Cancels the request, if possible. Has the effect of skipping
618	execution when entering the execute state and skipping calling the	901	execution when entering the execute state and skipping calling the
619	callback when entering the the result state, but will leave the	902	callback when entering the the result state, but will leave the
620	request otherwise untouched. That means that requests that currently	903	request otherwise untouched (with the exception of readdir). That
621	execute will not be stopped and resources held by the request will	904	means that requests that currently execute will not be stopped and
622	not be freed prematurely.	905	resources held by the request will not be freed prematurely.
623		906
624	cb $req $callback->(...)	907	cb $req $callback->(...)
625	Replace (or simply set) the callback registered to the request.	908	Replace (or simply set) the callback registered to the request.
626		909
627	IO::AIO::GRP CLASS	910	IO::AIO::GRP CLASS
…		…
688		971
689	$grp->cancel_subs	972	$grp->cancel_subs
690	Cancel all subrequests and clears any feeder, but not the group	973	Cancel all subrequests and clears any feeder, but not the group
691	request itself. Useful when you queued a lot of events but got a	974	request itself. Useful when you queued a lot of events but got a
692	result early.	975	result early.
		976
		977	The group request will finish normally (you cannot add requests to
		978	the group).
693		979
694	$grp->result (...)	980	$grp->result (...)
695	Set the result value(s) that will be passed to the group callback	981	Set the result value(s) that will be passed to the group callback
696	when all subrequests have finished and set the groups errno to the	982	when all subrequests have finished and set the groups errno to the
697	current value of errno (just like calling "errno" without an error	983	current value of errno (just like calling "errno" without an error
…		…
757	SUPPORT FUNCTIONS	1043	SUPPORT FUNCTIONS
758	EVENT PROCESSING AND EVENT LOOP INTEGRATION	1044	EVENT PROCESSING AND EVENT LOOP INTEGRATION
759	$fileno = IO::AIO::poll_fileno	1045	$fileno = IO::AIO::poll_fileno
760	Return the request result pipe file descriptor. This filehandle	1046	Return the request result pipe file descriptor. This filehandle
761	must be polled for reading by some mechanism outside this module	1047	must be polled for reading by some mechanism outside this module
762	(e.g. Event or select, see below or the SYNOPSIS). If the pipe	1048	(e.g. EV, Glib, select and so on, see below or the SYNOPSIS). If the
763	becomes readable you have to call "poll_cb" to check the results.	1049	pipe becomes readable you have to call "poll_cb" to check the
		1050	results.
764		1051
765	See "poll_cb" for an example.	1052	See "poll_cb" for an example.
766		1053
767	IO::AIO::poll_cb	1054	IO::AIO::poll_cb
768	Process some outstanding events on the result pipe. You have to call	1055	Process some outstanding events on the result pipe. You have to call
769	this regularly. Returns 0 if all events could be processed, or -1 if	1056	this regularly. Returns 0 if all events could be processed (or there
770	it returned earlier for whatever reason. Returns immediately when no	1057	were no events to process), or -1 if it returned earlier for
771	events are outstanding. The amount of events processed depends on	1058	whatever reason. Returns immediately when no events are outstanding.
772	the settings of "IO::AIO::max_poll_req" and	1059	The amount of events processed depends on the settings of
773	"IO::AIO::max_poll_time".	1060	"IO::AIO::max_poll_req" and "IO::AIO::max_poll_time".
774		1061
775	If not all requests were processed for whatever reason, the	1062	If not all requests were processed for whatever reason, the
776	filehandle will still be ready when "poll_cb" returns, so normally	1063	filehandle will still be ready when "poll_cb" returns, so normally
777	you don't have to do anything special to have it called later.	1064	you don't have to do anything special to have it called later.
778		1065
		1066	Apart from calling "IO::AIO::poll_cb" when the event filehandle
		1067	becomes ready, it can be beneficial to call this function from loops
		1068	which submit a lot of requests, to make sure the results get
		1069	processed when they become available and not just when the loop is
		1070	finished and the event loop takes over again. This function returns
		1071	very fast when there are no outstanding requests.
		1072
779	Example: Install an Event watcher that automatically calls	1073	Example: Install an Event watcher that automatically calls
780	IO::AIO::poll_cb with high priority:	1074	IO::AIO::poll_cb with high priority (more examples can be found in
		1075	the SYNOPSIS section, at the top of this document):
781		1076
782	Event->io (fd => IO::AIO::poll_fileno,	1077	Event->io (fd => IO::AIO::poll_fileno,
783	poll => 'r', async => 1,	1078	poll => 'r', async => 1,
784	cb => \&IO::AIO::poll_cb);	1079	cb => \&IO::AIO::poll_cb);
		1080
		1081	IO::AIO::poll_wait
		1082	If there are any outstanding requests and none of them in the result
		1083	phase, wait till the result filehandle becomes ready for reading
		1084	(simply does a "select" on the filehandle. This is useful if you
		1085	want to synchronously wait for some requests to finish).
		1086
		1087	See "nreqs" for an example.
		1088
		1089	IO::AIO::poll
		1090	Waits until some requests have been handled.
		1091
		1092	Returns the number of requests processed, but is otherwise strictly
		1093	equivalent to:
		1094
		1095	IO::AIO::poll_wait, IO::AIO::poll_cb
		1096
		1097	IO::AIO::flush
		1098	Wait till all outstanding AIO requests have been handled.
		1099
		1100	Strictly equivalent to:
		1101
		1102	IO::AIO::poll_wait, IO::AIO::poll_cb
		1103	while IO::AIO::nreqs;
785		1104
786	IO::AIO::max_poll_reqs $nreqs	1105	IO::AIO::max_poll_reqs $nreqs
787	IO::AIO::max_poll_time $seconds	1106	IO::AIO::max_poll_time $seconds
788	These set the maximum number of requests (default 0, meaning	1107	These set the maximum number of requests (default 0, meaning
789	infinity) that are being processed by "IO::AIO::poll_cb" in one	1108	infinity) that are being processed by "IO::AIO::poll_cb" in one
…		…
813	# use a low priority so other tasks have priority	1132	# use a low priority so other tasks have priority
814	Event->io (fd => IO::AIO::poll_fileno,	1133	Event->io (fd => IO::AIO::poll_fileno,
815	poll => 'r', nice => 1,	1134	poll => 'r', nice => 1,
816	cb => &IO::AIO::poll_cb);	1135	cb => &IO::AIO::poll_cb);
817		1136
818	IO::AIO::poll_wait
819	If there are any outstanding requests and none of them in the result
820	phase, wait till the result filehandle becomes ready for reading
821	(simply does a "select" on the filehandle. This is useful if you
822	want to synchronously wait for some requests to finish).
823
824	See "nreqs" for an example.
825
826	IO::AIO::poll
827	Waits until some requests have been handled.
828
829	Returns the number of requests processed, but is otherwise strictly
830	equivalent to:
831
832	IO::AIO::poll_wait, IO::AIO::poll_cb
833
834	IO::AIO::flush
835	Wait till all outstanding AIO requests have been handled.
836
837	Strictly equivalent to:
838
839	IO::AIO::poll_wait, IO::AIO::poll_cb
840	while IO::AIO::nreqs;
841
842	CONTROLLING THE NUMBER OF THREADS	1137	CONTROLLING THE NUMBER OF THREADS
843	IO::AIO::min_parallel $nthreads	1138	IO::AIO::min_parallel $nthreads
844	Set the minimum number of AIO threads to $nthreads. The current	1139	Set the minimum number of AIO threads to $nthreads. The current
845	default is 8, which means eight asynchronous operations can execute	1140	default is 8, which means eight asynchronous operations can execute
846	concurrently at any one time (the number of outstanding requests,	1141	concurrently at any one time (the number of outstanding requests,
…		…
875		1170
876	Under normal circumstances you don't need to call this function.	1171	Under normal circumstances you don't need to call this function.
877		1172
878	IO::AIO::max_idle $nthreads	1173	IO::AIO::max_idle $nthreads
879	Limit the number of threads (default: 4) that are allowed to idle	1174	Limit the number of threads (default: 4) that are allowed to idle
880	(i.e., threads that did not get a request to process within 10	1175	(i.e., threads that did not get a request to process within the idle
881	seconds). That means if a thread becomes idle while $nthreads other	1176	timeout (default: 10 seconds). That means if a thread becomes idle
882	threads are also idle, it will free its resources and exit.	1177	while $nthreads other threads are also idle, it will free its
		1178	resources and exit.
883		1179
884	This is useful when you allow a large number of threads (e.g. 100 or	1180	This is useful when you allow a large number of threads (e.g. 100 or
885	1000) to allow for extremely high load situations, but want to free	1181	1000) to allow for extremely high load situations, but want to free
886	resources under normal circumstances (1000 threads can easily	1182	resources under normal circumstances (1000 threads can easily
887	consume 30MB of RAM).	1183	consume 30MB of RAM).
888		1184
889	The default is probably ok in most situations, especially if thread	1185	The default is probably ok in most situations, especially if thread
890	creation is fast. If thread creation is very slow on your system you	1186	creation is fast. If thread creation is very slow on your system you
891	might want to use larger values.	1187	might want to use larger values.
892		1188
		1189	IO::AIO::idle_timeout $seconds
		1190	Sets the minimum idle timeout (default 10) after which worker
		1191	threads are allowed to exit. SEe "IO::AIO::max_idle".
		1192
893	IO::AIO::max_outstanding $maxreqs	1193	IO::AIO::max_outstanding $maxreqs
		1194	Sets the maximum number of outstanding requests to $nreqs. If you do
		1195	queue up more than this number of requests, the next call to
		1196	"IO::AIO::poll_cb" (and other functions calling "poll_cb", such as
		1197	"IO::AIO::flush" or "IO::AIO::poll") will block until the limit is
		1198	no longer exceeded.
		1199
		1200	In other words, this setting does not enforce a queue limit, but can
		1201	be used to make poll functions block if the limit is exceeded.
		1202
894	This is a very bad function to use in interactive programs because	1203	This is a very bad function to use in interactive programs because
895	it blocks, and a bad way to reduce concurrency because it is	1204	it blocks, and a bad way to reduce concurrency because it is
896	inexact: Better use an "aio_group" together with a feed callback.	1205	inexact: Better use an "aio_group" together with a feed callback.
897		1206
898	Sets the maximum number of outstanding requests to $nreqs. If you do	1207	It's main use is in scripts without an event loop - when you want to
899	queue up more than this number of requests, the next call to the	1208	stat a lot of files, you can write somehting like this:
900	"poll_cb" (and "poll_some" and other functions calling "poll_cb")
901	function will block until the limit is no longer exceeded.
902		1209
903	The default value is very large, so there is no practical limit on	1210	IO::AIO::max_outstanding 32;
		1211
		1212	for my $path (...) {
		1213	aio_stat $path , ...;
		1214	IO::AIO::poll_cb;
		1215	}
		1216
		1217	IO::AIO::flush;
		1218
		1219	The call to "poll_cb" inside the loop will normally return
		1220	instantly, but as soon as more thna 32 reqeusts are in-flight, it
		1221	will block until some requests have been handled. This keeps the
		1222	loop from pushing a large number of "aio_stat" requests onto the
		1223	queue.
		1224
		1225	The default value for "max_outstanding" is very large, so there is
904	the number of outstanding requests.	1226	no practical limit on the number of outstanding requests.
905
906	You can still queue as many requests as you want. Therefore,
907	"max_outstanding" is mainly useful in simple scripts (with low
908	values) or as a stop gap to shield against fatal memory overflow
909	(with large values).
910		1227
911	STATISTICAL INFORMATION	1228	STATISTICAL INFORMATION
912	IO::AIO::nreqs	1229	IO::AIO::nreqs
913	Returns the number of requests currently in the ready, execute or	1230	Returns the number of requests currently in the ready, execute or
914	pending states (i.e. for which their callback has not been invoked	1231	pending states (i.e. for which their callback has not been invoked
…		…
925		1242
926	IO::AIO::npending	1243	IO::AIO::npending
927	Returns the number of requests currently in the pending state	1244	Returns the number of requests currently in the pending state
928	(executed, but not yet processed by poll_cb).	1245	(executed, but not yet processed by poll_cb).
929		1246
		1247	MISCELLANEOUS FUNCTIONS
		1248	IO::AIO implements some functions that might be useful, but are not
		1249	asynchronous.
		1250
		1251	IO::AIO::sendfile $ofh, $ifh, $offset, $count
		1252	Calls the "eio_sendfile_sync" function, which is like
		1253	"aio_sendfile", but is blocking (this makes most sense if you know
		1254	the input data is likely cached already and the output filehandle is
		1255	set to non-blocking operations).
		1256
		1257	Returns the number of bytes copied, or -1 on error.
		1258
		1259	IO::AIO::fadvise $fh, $offset, $len, $advice
		1260	Simply calls the "posix_fadvise" function (see its manpage for
		1261	details). The following advice constants are avaiable:
		1262	"IO::AIO::FADV_NORMAL", "IO::AIO::FADV_SEQUENTIAL",
		1263	"IO::AIO::FADV_RANDOM", "IO::AIO::FADV_NOREUSE",
		1264	"IO::AIO::FADV_WILLNEED", "IO::AIO::FADV_DONTNEED".
		1265
		1266	On systems that do not implement "posix_fadvise", this function
		1267	returns ENOSYS, otherwise the return value of "posix_fadvise".
		1268
		1269	IO::AIO::madvise $scalar, $offset, $len, $advice
		1270	Simply calls the "posix_madvise" function (see its manpage for
		1271	details). The following advice constants are avaiable:
		1272	"IO::AIO::MADV_NORMAL", "IO::AIO::MADV_SEQUENTIAL",
		1273	"IO::AIO::MADV_RANDOM", "IO::AIO::MADV_WILLNEED",
		1274	"IO::AIO::MADV_DONTNEED".
		1275
		1276	On systems that do not implement "posix_madvise", this function
		1277	returns ENOSYS, otherwise the return value of "posix_madvise".
		1278
		1279	IO::AIO::mprotect $scalar, $offset, $len, $protect
		1280	Simply calls the "mprotect" function on the preferably AIO::mmap'ed
		1281	$scalar (see its manpage for details). The following protect
		1282	constants are avaiable: "IO::AIO::PROT_NONE", "IO::AIO::PROT_READ",
		1283	"IO::AIO::PROT_WRITE", "IO::AIO::PROT_EXEC".
		1284
		1285	On systems that do not implement "mprotect", this function returns
		1286	ENOSYS, otherwise the return value of "mprotect".
		1287
		1288	IO::AIO::mmap $scalar, $length, $prot, $flags, $fh[, $offset]
		1289	Memory-maps a file (or anonymous memory range) and attaches it to
		1290	the given $scalar, which will act like a string scalar.
		1291
		1292	The only operations allowed on the scalar are "substr"/"vec" that
		1293	don't change the string length, and most read-only operations such
		1294	as copying it or searching it with regexes and so on.
		1295
		1296	Anything else is unsafe and will, at best, result in memory leaks.
		1297
		1298	The memory map associated with the $scalar is automatically removed
		1299	when the $scalar is destroyed, or when the "IO::AIO::mmap" or
		1300	"IO::AIO::munmap" functions are called.
		1301
		1302	This calls the "mmap"(2) function internally. See your system's
		1303	manual page for details on the $length, $prot and $flags parameters.
		1304
		1305	The $length must be larger than zero and smaller than the actual
		1306	filesize.
		1307
		1308	$prot is a combination of "IO::AIO::PROT_NONE",
		1309	"IO::AIO::PROT_EXEC", "IO::AIO::PROT_READ" and/or
		1310	"IO::AIO::PROT_WRITE",
		1311
		1312	$flags can be a combination of "IO::AIO::MAP_SHARED" or
		1313	"IO::AIO::MAP_PRIVATE", or a number of system-specific flags (when
		1314	not available, the are defined as 0): "IO::AIO::MAP_ANONYMOUS"
		1315	(which is set to "MAP_ANON" if your system only provides this
		1316	constant), "IO::AIO::MAP_HUGETLB", "IO::AIO::MAP_LOCKED",
		1317	"IO::AIO::MAP_NORESERVE", "IO::AIO::MAP_POPULATE" or
		1318	"IO::AIO::MAP_NONBLOCK"
		1319
		1320	If $fh is "undef", then a file descriptor of -1 is passed.
		1321
		1322	$offset is the offset from the start of the file - it generally must
		1323	be a multiple of "IO::AIO::PAGESIZE" and defaults to 0.
		1324
		1325	Example:
		1326
		1327	use Digest::MD5;
		1328	use IO::AIO;
		1329
		1330	open my $fh, "<verybigfile"
		1331	or die "$!";
		1332
		1333	IO::AIO::mmap my $data, -s $fh, IO::AIO::PROT_READ, IO::AIO::MAP_SHARED, $fh
		1334	or die "verybigfile: $!";
		1335
		1336	my $fast_md5 = md5 $data;
		1337
		1338	IO::AIO::munmap $scalar
		1339	Removes a previous mmap and undefines the $scalar.
		1340
		1341	IO::AIO::munlock $scalar, $offset = 0, $length = undef
		1342	Calls the "munlock" function, undoing the effects of a previous
		1343	"aio_mlock" call (see its description for details).
		1344
		1345	IO::AIO::munlockall
		1346	Calls the "munlockall" function.
		1347
		1348	On systems that do not implement "munlockall", this function returns
		1349	ENOSYS, otherwise the return value of "munlockall".
		1350
		1351	EVENT LOOP INTEGRATION
		1352	It is recommended to use AnyEvent::AIO to integrate IO::AIO
		1353	automatically into many event loops:
		1354
		1355	# AnyEvent integration (EV, Event, Glib, Tk, POE, urxvt, pureperl...)
		1356	use AnyEvent::AIO;
		1357
		1358	You can also integrate IO::AIO manually into many event loops, here are
		1359	some examples of how to do this:
		1360
		1361	# EV integration
		1362	my $aio_w = EV::io IO::AIO::poll_fileno, EV::READ, \&IO::AIO::poll_cb;
		1363
		1364	# Event integration
		1365	Event->io (fd => IO::AIO::poll_fileno,
		1366	poll => 'r',
		1367	cb => \&IO::AIO::poll_cb);
		1368
		1369	# Glib/Gtk2 integration
		1370	add_watch Glib::IO IO::AIO::poll_fileno,
		1371	in => sub { IO::AIO::poll_cb; 1 };
		1372
		1373	# Tk integration
		1374	Tk::Event::IO->fileevent (IO::AIO::poll_fileno, "",
		1375	readable => \&IO::AIO::poll_cb);
		1376
		1377	# Danga::Socket integration
		1378	Danga::Socket->AddOtherFds (IO::AIO::poll_fileno =>
		1379	\&IO::AIO::poll_cb);
		1380
930	FORK BEHAVIOUR	1381	FORK BEHAVIOUR
931	This module should do "the right thing" when the process using it forks:	1382	Usage of pthreads in a program changes the semantics of fork
		1383	considerably. Specifically, only async-safe functions can be called
		1384	after fork. Perl doesn't know about this, so in general, you cannot call
		1385	fork with defined behaviour in perl if pthreads are involved. IO::AIO
		1386	uses pthreads, so this applies, but many other extensions and (for
		1387	inexplicable reasons) perl itself often is linked against pthreads, so
		1388	this limitation applies to quite a lot of perls.
932		1389
933	Before the fork, IO::AIO enters a quiescent state where no requests can	1390	This module no longer tries to fight your OS, or POSIX. That means
934	be added in other threads and no results will be processed. After the	1391	IO::AIO only works in the process that loaded it. Forking is fully
935	fork the parent simply leaves the quiescent state and continues	1392	supported, but using IO::AIO in the child is not.
936	request/result processing, while the child frees the request/result
937	queue (so that the requests started before the fork will only be handled
938	in the parent). Threads will be started on demand until the limit set in
939	the parent process has been reached again.
940		1393
941	In short: the parent will, after a short pause, continue as if fork had	1394	You might get around by not using IO::AIO before (or after) forking.
942	not been called, while the child will act as if IO::AIO has not been	1395	You could also try to call the IO::AIO::reinit function in the child:
943	used yet.	1396
		1397	IO::AIO::reinit
		1398	Abondons all current requests and I/O threads and simply
		1399	reinitialises all data structures. This is not an operation
		1400	suppported by any standards, but happens to work on GNU/Linux and
		1401	some newer BSD systems.
		1402
		1403	The only reasonable use for this function is to call it after
		1404	forking, if "IO::AIO" was used in the parent. Calling it while
		1405	IO::AIO is active in the process will result in undefined behaviour.
		1406	Calling it at any time will also result in any undefined (by POSIX)
		1407	behaviour.
944		1408
945	MEMORY USAGE	1409	MEMORY USAGE
946	Per-request usage:	1410	Per-request usage:
947		1411
948	Each aio request uses - depending on your architecture - around 100-200	1412	Each aio request uses - depending on your architecture - around 100-200

Diff Legend

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

Comparing IO-AIO/README (file contents): Revision 1.34 by root, Thu Nov 20 09:01:40 2008 UTC vs. Revision 1.49 by root, Mon Jul 18 03:09:06 2011 UTC

Diff Legend

Comparing IO-AIO/README (file contents):
Revision 1.34 by root, Thu Nov 20 09:01:40 2008 UTC vs.
Revision 1.49 by root, Mon Jul 18 03:09:06 2011 UTC