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

NAME
    IO::AIO - Asynchronous Input/Output

SYNOPSIS
     use IO::AIO;

     aio_open "/etc/passwd", O_RDONLY, 0, sub {
        my ($fh) = @_;
        ...
     };

     aio_unlink "/tmp/file", sub { };

     aio_read $fh, 30000, 1024, $buffer, 0, sub {
        $_[0] > 0 or die "read error: $!";
     };

     # AnyEvent
     open my $fh, "<&=" . IO::AIO::poll_fileno or die "$!";
     my $w = AnyEvent->io (fh => $fh, poll => 'r', cb => sub { IO::AIO::poll_cb });

     # Event
     Event->io (fd => IO::AIO::poll_fileno,
                poll => 'r',
                cb => \&IO::AIO::poll_cb);

     # Glib/Gtk2
     add_watch Glib::IO IO::AIO::poll_fileno,
               in => sub { IO::AIO::poll_cb; 1 };

     # Tk
     Tk::Event::IO->fileevent (IO::AIO::poll_fileno, "",
                               readable => \&IO::AIO::poll_cb);

     # Danga::Socket
     Danga::Socket->AddOtherFds (IO::AIO::poll_fileno =>
                                 \&IO::AIO::poll_cb);

DESCRIPTION
    This module implements asynchronous I/O using whatever means your
    operating system supports.

    Currently, a number of threads are started that execute your read/writes
    and signal their completion. You don't need thread support in your libc
    or perl, and the threads created by this module will not be visible to
    the pthreads library. In the future, this module might make use of the
    native aio functions available on many operating systems. However, they
    are often not well-supported (Linux doesn't allow them on normal files
    currently, for example), and they would only support aio_read and
    aio_write, so the remaining functionality would have to be implemented
    using threads anyway.

    Although the module will work with in the presence of other threads, it
    is currently not reentrant, so use appropriate locking yourself, always
    call "poll_cb" from within the same thread, or never call "poll_cb" (or
    other "aio_" functions) recursively.

FUNCTIONS
  AIO FUNCTIONS
    All the "aio_*" calls are more or less thin wrappers around the syscall
    with the same name (sans "aio_"). The arguments are similar or
    identical, and they all accept an additional (and optional) $callback
    argument which must be a code reference. This code reference will get
    called with the syscall return code (e.g. most syscalls return -1 on
    error, unlike perl, which usually delivers "false") as it's sole
    argument when the given syscall has been executed asynchronously.

    All functions expecting a filehandle keep a copy of the filehandle
    internally until the request has finished.

    The pathnames you pass to these routines *must* be absolute and encoded
    in byte form. The reason for the former is that at the time the request
    is being executed, the current working directory could have changed.
    Alternatively, you can make sure that you never change the current
    working directory.

    To encode pathnames to byte form, either make sure you either: a) always
    pass in filenames you got from outside (command line, readdir etc.), b)
    are ASCII or ISO 8859-1, c) use the Encode module and encode your
    pathnames to the locale (or other) encoding in effect in the user
    environment, d) use Glib::filename_from_unicode on unicode filenames or
    e) use something else.

    aio_open $pathname, $flags, $mode, $callback->($fh)
        Asynchronously open or create a file and call the callback with a
        newly created filehandle for the file.

        The pathname passed to "aio_open" must be absolute. See API NOTES,
        above, for an explanation.

        The $flags argument is a bitmask. See the "Fcntl" module for a list.
        They are the same as used by "sysopen".

        Likewise, $mode specifies the mode of the newly created file, if it
        didn't exist and "O_CREAT" has been given, just like perl's
        "sysopen", except that it is mandatory (i.e. use 0 if you don't
        create new files, and 0666 or 0777 if you do).

        Example:

           aio_open "/etc/passwd", O_RDONLY, 0, sub {
              if ($_[0]) {
                 print "open successful, fh is $_[0]\n";
                 ...
              } else {
                 die "open failed: $!\n";
              }
           };

    aio_close $fh, $callback->($status)
        Asynchronously close a file and call the callback with the result
        code. *WARNING:* although accepted, you should not pass in a perl
        filehandle here, as perl will likely close the file descriptor
        another time when the filehandle is destroyed. Normally, you can
        safely call perls "close" or just let filehandles go out of scope.

        This is supposed to be a bug in the API, so that might change. It's
        therefore best to avoid this function.

    aio_read $fh,$offset,$length, $data,$dataoffset, $callback->($retval)
    aio_write $fh,$offset,$length, $data,$dataoffset, $callback->($retval)
        Reads or writes "length" bytes from the specified "fh" and "offset"
        into the scalar given by "data" and offset "dataoffset" and calls
        the callback without the actual number of bytes read (or -1 on
        error, just like the syscall).

        The $data scalar *MUST NOT* be modified in any way while the request
        is outstanding. Modifying it can result in segfaults or WW3 (if the
        necessary/optional hardware is installed).

        Example: Read 15 bytes at offset 7 into scalar $buffer, starting at
        offset 0 within the scalar:

           aio_read $fh, 7, 15, $buffer, 0, sub {
              $_[0] > 0 or die "read error: $!";
              print "read $_[0] bytes: <$buffer>\n";
           };

    aio_sendfile $out_fh, $in_fh, $in_offset, $length, $callback->($retval)
        Tries to copy $length bytes from $in_fh to $out_fh. It starts
        reading at byte offset $in_offset, and starts writing at the current
        file offset of $out_fh. Because of that, it is not safe to issue
        more than one "aio_sendfile" per $out_fh, as they will interfere
        with each other.

        This call tries to make use of a native "sendfile" syscall to
        provide zero-copy operation. For this to work, $out_fh should refer
        to a socket, and $in_fh should refer to mmap'able file.

        If the native sendfile call fails or is not implemented, it will be
        emulated, so you can call "aio_sendfile" on any type of filehandle
        regardless of the limitations of the operating system.

        Please note, however, that "aio_sendfile" can read more bytes from
        $in_fh than are written, and there is no way to find out how many
        bytes have been read from "aio_sendfile" alone, as "aio_sendfile"
        only provides the number of bytes written to $out_fh. Only if the
        result value equals $length one can assume that $length bytes have
        been read.

    aio_readahead $fh,$offset,$length, $callback->($retval)
        "aio_readahead" populates the page cache with data from a file so
        that subsequent reads from that file will not block on disk I/O. The
        $offset argument specifies the starting point from which data is to
        be read and $length specifies the number of bytes to be read. I/O is
        performed in whole pages, so that offset is effectively rounded down
        to a page boundary and bytes are read up to the next page boundary
        greater than or equal to (off-set+length). "aio_readahead" does not
        read beyond the end of the file. The current file offset of the file
        is left unchanged.

        If that syscall doesn't exist (likely if your OS isn't Linux) it
        will be emulated by simply reading the data, which would have a
        similar effect.

    aio_stat $fh_or_path, $callback->($status)
    aio_lstat $fh, $callback->($status)
        Works like perl's "stat" or "lstat" in void context. The callback
        will be called after the stat and the results will be available
        using "stat _" or "-s _" etc...

        The pathname passed to "aio_stat" must be absolute. See API NOTES,
        above, for an explanation.

        Currently, the stats are always 64-bit-stats, i.e. instead of
        returning an error when stat'ing a large file, the results will be
        silently truncated unless perl itself is compiled with large file
        support.

        Example: Print the length of /etc/passwd:

           aio_stat "/etc/passwd", sub {
              $_[0] and die "stat failed: $!";
              print "size is ", -s _, "\n";
           };

    aio_unlink $pathname, $callback->($status)
        Asynchronously unlink (delete) a file and call the callback with the
        result code.

    aio_rmdir $pathname, $callback->($status)
        Asynchronously rmdir (delete) a directory and call the callback with
        the result code.

    aio_readdir $pathname $callback->($entries)
        Unlike the POSIX call of the same name, "aio_readdir" reads an
        entire directory (i.e. opendir + readdir + closedir). The entries
        will not be sorted, and will NOT include the "." and ".." entries.

        The callback a single argument which is either "undef" or an
        array-ref with the filenames.

    aio_scandir $path, $maxreq, $callback->($dirs, $nondirs)
        Scans a directory (similar to "aio_readdir") and tries to separate
        the entries of directory $path into two sets of names, ones you can
        recurse into (directories), and ones you cannot recurse into
        (everything else).

        "aio_scandir" is a composite request that consists of many
        aio-primitives. $maxreq specifies the maximum number of outstanding
        aio requests that this function generates. If it is "<= 0", then a
        suitable default will be chosen (currently 8).

        On error, the callback is called without arguments, otherwise it
        receives two array-refs with path-relative entry names.

        Example:

           aio_scandir $dir, 0, sub {
              my ($dirs, $nondirs) = @_;
              print "real directories: @$dirs\n";
              print "everything else: @$nondirs\n";
           };

        Implementation notes.

        The "aio_readdir" cannot be avoided, but "stat()"'ing every entry
        can.

        After reading the directory, the modification time, size etc. of the
        directory before and after the readdir is checked, and if they
        match, the link count will be used to decide how many entries are
        directories (if >= 2). Otherwise, no knowledge of the number of
        subdirectories will be assumed.

        Then entires will be sorted into likely directories (everything
        without a non-initial dot) and likely non-directories (everything
        else). Then every entry + "/." will be "stat"'ed, likely directories
        first. This is often faster because filesystems might detect the
        type of the entry without reading the inode data (e.g. ext2s
        filetype feature). If that succeeds, it assumes that the entry is a
        directory or a symlink to directory (which will be checked
        seperately).

        If the known number of directories has been reached, the rest of the
        entries is assumed to be non-directories.

    aio_fsync $fh, $callback->($status)
        Asynchronously call fsync on the given filehandle and call the
        callback with the fsync result code.

    aio_fdatasync $fh, $callback->($status)
        Asynchronously call fdatasync on the given filehandle and call the
        callback with the fdatasync result code.

        If this call isn't available because your OS lacks it or it couldn't
        be detected, it will be emulated by calling "fsync" instead.

  SUPPORT FUNCTIONS
    $fileno = IO::AIO::poll_fileno
        Return the *request result pipe file descriptor*. This filehandle
        must be polled for reading by some mechanism outside this module
        (e.g. Event or select, see below or the SYNOPSIS). If the pipe
        becomes readable you have to call "poll_cb" to check the results.

        See "poll_cb" for an example.

    IO::AIO::poll_cb
        Process all outstanding events on the result pipe. You have to call
        this regularly. Returns the number of events processed. Returns
        immediately when no events are outstanding.

        Example: Install an Event watcher that automatically calls
        IO::AIO::poll_cb with high priority:

           Event->io (fd => IO::AIO::poll_fileno,
                      poll => 'r', async => 1,
                      cb => \&IO::AIO::poll_cb);

    IO::AIO::poll_wait
        Wait till the result filehandle becomes ready for reading (simply
        does a "select" on the filehandle. This is useful if you want to
        synchronously wait for some requests to finish).

        See "nreqs" for an example.

    IO::AIO::nreqs
        Returns the number of requests currently outstanding (i.e. for which
        their callback has not been invoked yet).

        Example: wait till there are no outstanding requests anymore:

           IO::AIO::poll_wait, IO::AIO::poll_cb
              while IO::AIO::nreqs;

    IO::AIO::flush
        Wait till all outstanding AIO requests have been handled.

        Strictly equivalent to:

           IO::AIO::poll_wait, IO::AIO::poll_cb
              while IO::AIO::nreqs;

    IO::AIO::poll
        Waits until some requests have been handled.

        Strictly equivalent to:

           IO::AIO::poll_wait, IO::AIO::poll_cb
              if IO::AIO::nreqs;

    IO::AIO::min_parallel $nthreads
        Set the minimum number of AIO threads to $nthreads. The current
        default is 4, which means four asynchronous operations can be done
        at one time (the number of outstanding operations, however, is
        unlimited).

        IO::AIO starts threads only on demand, when an AIO request is queued
        and no free thread exists.

        It is recommended to keep the number of threads low, as some Linux
        kernel versions will scale negatively with the number of threads
        (higher parallelity => MUCH higher latency). With current Linux 2.6
        versions, 4-32 threads should be fine.

        Under most circumstances you don't need to call this function, as
        the module selects a default that is suitable for low to moderate
        load.

    IO::AIO::max_parallel $nthreads
        Sets the maximum number of AIO threads to $nthreads. If more than
        the specified number of threads are currently running, this function
        kills them. This function blocks until the limit is reached.

        While $nthreads are zero, aio requests get queued but not executed
        until the number of threads has been increased again.

        This module automatically runs "max_parallel 0" at program end, to
        ensure that all threads are killed and that there are no outstanding
        requests.

        Under normal circumstances you don't need to call this function.

    $oldnreqs = IO::AIO::max_outstanding $nreqs
        Sets the maximum number of outstanding requests to $nreqs. If you
        try to queue up more than this number of requests, the caller will
        block until some requests have been handled.

        The default is very large, so normally there is no practical limit.
        If you queue up many requests in a loop it often improves speed if
        you set this to a relatively low number, such as 100.

        Under normal circumstances you don't need to call this function.

  FORK BEHAVIOUR
    Before the fork, IO::AIO enters a quiescent state where no requests can
    be added in other threads and no results will be processed. After the
    fork the parent simply leaves the quiescent state and continues
    request/result processing, while the child clears the request/result
    queue (so the requests started before the fork will only be handled in
    the parent). Threats will be started on demand until the limit ste in
    the parent process has been reached again.

SEE ALSO
    Coro, Linux::AIO.

AUTHOR
     Marc Lehmann <schmorp@schmorp.de>
     http://home.schmorp.de/

Revision:	1.14
Committed:	Mon Dec 26 18:31:12 2005 UTC (20 years, 8 months ago) by root
Branch:	MAIN
CVS Tags:	rel-1_71, rel-1_7
Changes since 1.13:	+4 -0 lines
Log Message:	* empty log message *
#	User	Rev	Content
1	root	1.1	NAME
2			IO::AIO - Asynchronous Input/Output
3
4			SYNOPSIS
5			use IO::AIO;
6
7	root	1.5	aio_open "/etc/passwd", O_RDONLY, 0, sub {
8			my ($fh) = @_;
9			...
10			};
11
12			aio_unlink "/tmp/file", sub { };
13
14			aio_read $fh, 30000, 1024, $buffer, 0, sub {
15			$_[0] > 0 or die "read error: $!";
16			};
17
18	root	1.14	# AnyEvent
19			open my $fh, "<&=" . IO::AIO::poll_fileno or die "$!";
20			my $w = AnyEvent->io (fh => $fh, poll => 'r', cb => sub { IO::AIO::poll_cb });
21
22	root	1.5	# Event
23			Event->io (fd => IO::AIO::poll_fileno,
24			poll => 'r',
25			cb => \&IO::AIO::poll_cb);
26
27			# Glib/Gtk2
28			add_watch Glib::IO IO::AIO::poll_fileno,
29	root	1.7	in => sub { IO::AIO::poll_cb; 1 };
30	root	1.5
31			# Tk
32			Tk::Event::IO->fileevent (IO::AIO::poll_fileno, "",
33			readable => \&IO::AIO::poll_cb);
34
35	root	1.6	# Danga::Socket
36			Danga::Socket->AddOtherFds (IO::AIO::poll_fileno =>
37			\&IO::AIO::poll_cb);
38
39	root	1.1	DESCRIPTION
40			This module implements asynchronous I/O using whatever means your
41	root	1.2	operating system supports.
42	root	1.1
43	root	1.2	Currently, a number of threads are started that execute your read/writes
44			and signal their completion. You don't need thread support in your libc
45			or 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
47			native aio functions available on many operating systems. However, they
48			are often not well-supported (Linux doesn't allow them on normal files
49			currently, for example), and they would only support aio_read and
50			aio_write, so the remaining functionality would have to be implemented
51			using threads anyway.
52	root	1.1
53			Although the module will work with in the presence of other threads, it
54	root	1.7	is currently not reentrant, so use appropriate locking yourself, always
55			call "poll_cb" from within the same thread, or never call "poll_cb" (or
56			other "aio_" functions) recursively.
57	root	1.1
58	root	1.4	FUNCTIONS
59			AIO FUNCTIONS
60	root	1.1	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
62	root	1.6	identical, and they all accept an additional (and optional) $callback
63			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
65			error, unlike perl, which usually delivers "false") as it's sole
66			argument when the given syscall has been executed asynchronously.
67	root	1.1
68	root	1.8	All functions expecting a filehandle keep a copy of the filehandle
69			internally until the request has finished.
70	root	1.1
71	root	1.9	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
73			is being executed, the current working directory could have changed.
74			Alternatively, you can make sure that you never change the current
75			working directory.
76
77			To encode pathnames to byte form, either make sure you either: a) always
78			pass in filenames you got from outside (command line, readdir etc.), b)
79			are ASCII or ISO 8859-1, c) use the Encode module and encode your
80			pathnames to the locale (or other) encoding in effect in the user
81			environment, d) use Glib::filename_from_unicode on unicode filenames or
82			e) use something else.
83	root	1.1
84	root	1.13	aio_open $pathname, $flags, $mode, $callback->($fh)
85	root	1.2	Asynchronously open or create a file and call the callback with a
86			newly created filehandle for the file.
87
88			The pathname passed to "aio_open" must be absolute. See API NOTES,
89			above, for an explanation.
90	root	1.1
91	root	1.7	The $flags argument is a bitmask. See the "Fcntl" module for a list.
92			They are the same as used by "sysopen".
93
94			Likewise, $mode specifies the mode of the newly created file, if it
95			didn't exist and "O_CREAT" has been given, just like perl's
96			"sysopen", except that it is mandatory (i.e. use 0 if you don't
97			create new files, and 0666 or 0777 if you do).
98	root	1.1
99			Example:
100
101			aio_open "/etc/passwd", O_RDONLY, 0, sub {
102	root	1.2	if ($_[0]) {
103			print "open successful, fh is $_[0]\n";
104	root	1.1	...
105			} else {
106			die "open failed: $!\n";
107			}
108			};
109
110	root	1.13	aio_close $fh, $callback->($status)
111	root	1.1	Asynchronously close a file and call the callback with the result
112	root	1.2	code. WARNING: although accepted, you should not pass in a perl
113			filehandle here, as perl will likely close the file descriptor
114	root	1.7	another time when the filehandle is destroyed. Normally, you can
115			safely call perls "close" or just let filehandles go out of scope.
116
117			This is supposed to be a bug in the API, so that might change. It's
118			therefore best to avoid this function.
119	root	1.1
120	root	1.13	aio_read $fh,$offset,$length, $data,$dataoffset, $callback->($retval)
121			aio_write $fh,$offset,$length, $data,$dataoffset, $callback->($retval)
122	root	1.1	Reads or writes "length" bytes from the specified "fh" and "offset"
123			into the scalar given by "data" and offset "dataoffset" and calls
124			the callback without the actual number of bytes read (or -1 on
125			error, just like the syscall).
126
127	root	1.10	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
129			necessary/optional hardware is installed).
130
131	root	1.6	Example: Read 15 bytes at offset 7 into scalar $buffer, starting at
132	root	1.1	offset 0 within the scalar:
133
134			aio_read $fh, 7, 15, $buffer, 0, sub {
135	root	1.5	$_[0] > 0 or die "read error: $!";
136			print "read $_[0] bytes: <$buffer>\n";
137	root	1.1	};
138
139	root	1.13	aio_sendfile $out_fh, $in_fh, $in_offset, $length, $callback->($retval)
140	root	1.12	Tries to copy $length bytes from $in_fh to $out_fh. It starts
141			reading at byte offset $in_offset, and starts writing at the current
142			file offset of $out_fh. Because of that, it is not safe to issue
143			more than one "aio_sendfile" per $out_fh, as they will interfere
144			with each other.
145
146			This call tries to make use of a native "sendfile" syscall to
147			provide zero-copy operation. For this to work, $out_fh should refer
148			to a socket, and $in_fh should refer to mmap'able file.
149
150			If the native sendfile call fails or is not implemented, it will be
151			emulated, so you can call "aio_sendfile" on any type of filehandle
152			regardless of the limitations of the operating system.
153
154			Please note, however, that "aio_sendfile" can read more bytes from
155			$in_fh than are written, and there is no way to find out how many
156			bytes have been read from "aio_sendfile" alone, as "aio_sendfile"
157			only provides the number of bytes written to $out_fh. Only if the
158			result value equals $length one can assume that $length bytes have
159			been read.
160
161	root	1.13	aio_readahead $fh,$offset,$length, $callback->($retval)
162	root	1.7	"aio_readahead" populates the page cache with data from a file so
163			that subsequent reads from that file will not block on disk I/O. The
164	root	1.1	$offset argument specifies the starting point from which data is to
165			be read and $length specifies the number of bytes to be read. I/O is
166			performed in whole pages, so that offset is effectively rounded down
167			to a page boundary and bytes are read up to the next page boundary
168	root	1.7	greater than or equal to (off-set+length). "aio_readahead" does not
169	root	1.1	read beyond the end of the file. The current file offset of the file
170			is left unchanged.
171
172	root	1.9	If that syscall doesn't exist (likely if your OS isn't Linux) it
173			will be emulated by simply reading the data, which would have a
174			similar effect.
175
176	root	1.13	aio_stat $fh_or_path, $callback->($status)
177			aio_lstat $fh, $callback->($status)
178	root	1.1	Works like perl's "stat" or "lstat" in void context. The callback
179			will be called after the stat and the results will be available
180			using "stat _" or "-s _" etc...
181
182			The pathname passed to "aio_stat" must be absolute. See API NOTES,
183			above, for an explanation.
184
185			Currently, the stats are always 64-bit-stats, i.e. instead of
186			returning an error when stat'ing a large file, the results will be
187			silently truncated unless perl itself is compiled with large file
188			support.
189
190			Example: Print the length of /etc/passwd:
191
192			aio_stat "/etc/passwd", sub {
193			$_[0] and die "stat failed: $!";
194			print "size is ", -s _, "\n";
195			};
196
197	root	1.13	aio_unlink $pathname, $callback->($status)
198	root	1.1	Asynchronously unlink (delete) a file and call the callback with the
199			result code.
200
201	root	1.13	aio_rmdir $pathname, $callback->($status)
202	root	1.9	Asynchronously rmdir (delete) a directory and call the callback with
203			the result code.
204
205	root	1.13	aio_readdir $pathname $callback->($entries)
206			Unlike the POSIX call of the same name, "aio_readdir" reads an
207			entire directory (i.e. opendir + readdir + closedir). The entries
208			will not be sorted, and will NOT include the "." and ".." entries.
209
210			The callback a single argument which is either "undef" or an
211			array-ref with the filenames.
212
213			aio_scandir $path, $maxreq, $callback->($dirs, $nondirs)
214			Scans a directory (similar to "aio_readdir") and tries to separate
215			the entries of directory $path into two sets of names, ones you can
216			recurse into (directories), and ones you cannot recurse into
217			(everything else).
218
219			"aio_scandir" is a composite request that consists of many
220			aio-primitives. $maxreq specifies the maximum number of outstanding
221			aio requests that this function generates. If it is "<= 0", then a
222			suitable default will be chosen (currently 8).
223
224			On error, the callback is called without arguments, otherwise it
225			receives two array-refs with path-relative entry names.
226
227			Example:
228
229			aio_scandir $dir, 0, sub {
230			my ($dirs, $nondirs) = @_;
231			print "real directories: @$dirs\n";
232			print "everything else: @$nondirs\n";
233			};
234
235			Implementation notes.
236
237			The "aio_readdir" cannot be avoided, but "stat()"'ing every entry
238			can.
239
240			After reading the directory, the modification time, size etc. of the
241			directory before and after the readdir is checked, and if they
242			match, the link count will be used to decide how many entries are
243			directories (if >= 2). Otherwise, no knowledge of the number of
244			subdirectories will be assumed.
245
246			Then entires will be sorted into likely directories (everything
247			without a non-initial dot) and likely non-directories (everything
248			else). Then every entry + "/." will be "stat"'ed, likely directories
249			first. This is often faster because filesystems might detect the
250			type of the entry without reading the inode data (e.g. ext2s
251			filetype feature). If that succeeds, it assumes that the entry is a
252			directory or a symlink to directory (which will be checked
253			seperately).
254
255			If the known number of directories has been reached, the rest of the
256			entries is assumed to be non-directories.
257
258			aio_fsync $fh, $callback->($status)
259	root	1.1	Asynchronously call fsync on the given filehandle and call the
260			callback with the fsync result code.
261
262	root	1.13	aio_fdatasync $fh, $callback->($status)
263	root	1.1	Asynchronously call fdatasync on the given filehandle and call the
264	root	1.9	callback with the fdatasync result code.
265
266			If this call isn't available because your OS lacks it or it couldn't
267			be detected, it will be emulated by calling "fsync" instead.
268	root	1.1
269	root	1.4	SUPPORT FUNCTIONS
270			$fileno = IO::AIO::poll_fileno
271	root	1.7	Return the request result pipe file descriptor. This filehandle
272			must be polled for reading by some mechanism outside this module
273			(e.g. Event or select, see below or the SYNOPSIS). If the pipe
274			becomes readable you have to call "poll_cb" to check the results.
275	root	1.4
276			See "poll_cb" for an example.
277
278			IO::AIO::poll_cb
279			Process all outstanding events on the result pipe. You have to call
280			this regularly. Returns the number of events processed. Returns
281			immediately when no events are outstanding.
282
283	root	1.7	Example: Install an Event watcher that automatically calls
284			IO::AIO::poll_cb with high priority:
285	root	1.4
286			Event->io (fd => IO::AIO::poll_fileno,
287			poll => 'r', async => 1,
288			cb => \&IO::AIO::poll_cb);
289
290			IO::AIO::poll_wait
291			Wait till the result filehandle becomes ready for reading (simply
292	root	1.7	does a "select" on the filehandle. This is useful if you want to
293	root	1.4	synchronously wait for some requests to finish).
294
295			See "nreqs" for an example.
296
297			IO::AIO::nreqs
298	root	1.7	Returns the number of requests currently outstanding (i.e. for which
299			their callback has not been invoked yet).
300	root	1.4
301			Example: wait till there are no outstanding requests anymore:
302
303			IO::AIO::poll_wait, IO::AIO::poll_cb
304			while IO::AIO::nreqs;
305
306	root	1.6	IO::AIO::flush
307			Wait till all outstanding AIO requests have been handled.
308
309			Strictly equivalent to:
310
311			IO::AIO::poll_wait, IO::AIO::poll_cb
312			while IO::AIO::nreqs;
313
314			IO::AIO::poll
315			Waits until some requests have been handled.
316
317			Strictly equivalent to:
318
319			IO::AIO::poll_wait, IO::AIO::poll_cb
320			if IO::AIO::nreqs;
321
322	root	1.4	IO::AIO::min_parallel $nthreads
323	root	1.11	Set the minimum number of AIO threads to $nthreads. The current
324			default is 4, which means four asynchronous operations can be done
325			at one time (the number of outstanding operations, however, is
326			unlimited).
327
328			IO::AIO starts threads only on demand, when an AIO request is queued
329			and no free thread exists.
330	root	1.4
331			It is recommended to keep the number of threads low, as some Linux
332			kernel versions will scale negatively with the number of threads
333			(higher parallelity => MUCH higher latency). With current Linux 2.6
334			versions, 4-32 threads should be fine.
335
336	root	1.11	Under most circumstances you don't need to call this function, as
337			the module selects a default that is suitable for low to moderate
338			load.
339	root	1.4
340			IO::AIO::max_parallel $nthreads
341			Sets the maximum number of AIO threads to $nthreads. If more than
342	root	1.11	the specified number of threads are currently running, this function
343			kills them. This function blocks until the limit is reached.
344
345			While $nthreads are zero, aio requests get queued but not executed
346			until the number of threads has been increased again.
347	root	1.4
348			This module automatically runs "max_parallel 0" at program end, to
349			ensure that all threads are killed and that there are no outstanding
350			requests.
351
352			Under normal circumstances you don't need to call this function.
353
354			$oldnreqs = IO::AIO::max_outstanding $nreqs
355			Sets the maximum number of outstanding requests to $nreqs. If you
356			try to queue up more than this number of requests, the caller will
357			block until some requests have been handled.
358
359			The default is very large, so normally there is no practical limit.
360	root	1.11	If you queue up many requests in a loop it often improves speed if
361			you set this to a relatively low number, such as 100.
362	root	1.4
363			Under normal circumstances you don't need to call this function.
364	root	1.1
365	root	1.9	FORK BEHAVIOUR
366	root	1.11	Before the fork, IO::AIO enters a quiescent state where no requests can
367	root	1.10	be added in other threads and no results will be processed. After the
368			fork the parent simply leaves the quiescent state and continues
369			request/result processing, while the child clears the request/result
370	root	1.11	queue (so the requests started before the fork will only be handled in
371			the parent). Threats will be started on demand until the limit ste in
372			the parent process has been reached again.
373	root	1.9
374	root	1.1	SEE ALSO
375	root	1.2	Coro, Linux::AIO.
376	root	1.1
377			AUTHOR
378			Marc Lehmann <schmorp@schmorp.de>
379			http://home.schmorp.de/
380