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

NAME
    Linux::AIO - linux-specific aio implemented using clone

SYNOPSIS
     use Linux::AIO;

     # This module has been mostly superseded by IO::AIO.

DESCRIPTION
    *This module has been mostly superseded by IO::AIO, which is API
    compatible.*

    This module implements asynchronous I/O using the means available to
    Linux - clone. It does not hook into the POSIX aio_* functions because
    Linux does not yet support these in the kernel (even as of 2.6.12, only
    O_DIRECT files are supported) and even if, it would only allow aio_read
    and write, not open, stat and so on.

    Instead, in this module a number of (non-posix) 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.

    NOTICE: the threads created by this module will automatically be killed
    when the thread calling min_parallel exits. Make sure you only ever call
    min_parallel from the same thread that loaded this module.

    Although the module will work with in the presence of other threads, it
    is not reentrant, so use appropriate locking yourself.

  API NOTES
    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 $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 that expect a filehandle will also accept a file
    descriptor.

    The filenames you pass to these routines *must* be absolute. The reason
    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.

    Linux::AIO::min_parallel $nthreads
        Set the minimum number of AIO threads to $nthreads. The default is
        1, which means a single asynchronous operation can be done at one
        time (the number of outstanding operations, however, is unlimited).

        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).

        Under normal circumstances you don't need to call this function, as
        this module automatically starts a single async thread.

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

        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.

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

        See "poll_cb" for an example.

    Linux::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.

        You can use Event to multiplex, e.g.:

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

    Linux::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.

    Linux::AIO::nreqs
        Returns the number of requests currently outstanding.

        Example: wait till there are no outstanding requests anymore:

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

    aio_open $pathname, $flags, $mode, $callback
        Asynchronously open or create a file and call the callback with the
        filedescriptor (NOT a perl filehandle, sorry for that, but watch
        out, this might change in the future).

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

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

        Example:

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

    aio_close $fh, $callback
        Asynchronously close a file and call the callback with the result
        code.

    aio_read $fh,$offset,$length, $data,$dataoffset,$callback
    aio_write $fh,$offset,$length, $data,$dataoffset,$callback
        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).

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

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

    aio_readahead $fh,$offset,$length, $callback
        Asynchronously reads the specified byte range into the page cache,
        using the "readahead" syscall.

        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.

    aio_stat $fh_or_path, $callback
    aio_lstat $fh, $callback
        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
        Asynchronously unlink (delete) a file and call the callback with the
        result code.

    aio_fsync $fh, $callback
        Asynchronously call fsync on the given filehandle and call the
        callback with the fsync result code.

    aio_fdatasync $fh, $callback
        Asynchronously call fdatasync on the given filehandle and call the
        callback with the fdatasync result code.

BUGS
    This module has been extensively tested in a large and very busy
    webserver for many years now.

       - aio_open gives a fd, but all other functions expect a perl filehandle.

SEE ALSO
    Coro, IO::AIO.

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

Revision:	1.7
Committed:	Wed Aug 17 16:57:53 2005 UTC (18 years, 9 months ago) by root
Branch:	MAIN
CVS Tags:	rel-1_9, rel-1_8, HEAD
Changes since 1.6:	+5 -0 lines
Log Message:	* empty log message *
#	User	Rev	Content
1	root	1.1	NAME
2			Linux::AIO - linux-specific aio implemented using clone
3
4			SYNOPSIS
5			use Linux::AIO;
6
7	root	1.7	# This module has been mostly superseded by IO::AIO.
8
9	root	1.1	DESCRIPTION
10	root	1.7	*This module has been mostly superseded by IO::AIO, which is API
11			compatible.*
12
13	root	1.3	This module implements asynchronous I/O using the means available to
14			Linux - clone. It does not hook into the POSIX aio_* functions because
15			Linux does not yet support these in the kernel (even as of 2.6.12, only
16			O_DIRECT files are supported) and even if, it would only allow aio_read
17			and write, not open, stat and so on.
18	root	1.1
19			Instead, in this module a number of (non-posix) threads are started that
20			execute your read/writes and signal their completion. You don't need
21			thread support in your libc or perl, and the threads created by this
22			module will not be visible to the pthreads library.
23
24			NOTICE: the threads created by this module will automatically be killed
25			when the thread calling min_parallel exits. Make sure you only ever call
26			min_parallel from the same thread that loaded this module.
27
28	root	1.3	Although the module will work with in the presence of other threads, it
29			is not reentrant, so use appropriate locking yourself.
30	root	1.1
31	root	1.4	API NOTES
32			All the "aio_*" calls are more or less thin wrappers around the syscall
33			with the same name (sans "aio_"). The arguments are similar or
34			identical, and they all accept an additional $callback argument which
35			must be a code reference. This code reference will get called with the
36			syscall return code (e.g. most syscalls return -1 on error, unlike perl,
37			which usually delivers "false") as it's sole argument when the given
38			syscall has been executed asynchronously.
39
40			All functions that expect a filehandle will also accept a file
41			descriptor.
42
43	root	1.5	The filenames you pass to these routines must be absolute. The reason
44			is that at the time the request is being executed, the current working
45			directory could have changed. Alternatively, you can make sure that you
46			never change the current working directory.
47
48	root	1.1	Linux::AIO::min_parallel $nthreads
49			Set the minimum number of AIO threads to $nthreads. The default is
50			1, which means a single asynchronous operation can be done at one
51			time (the number of outstanding operations, however, is unlimited).
52
53	root	1.2	It is recommended to keep the number of threads low, as some linux
54	root	1.1	kernel versions will scale negatively with the number of threads
55			(higher parallelity => MUCH higher latency).
56
57	root	1.4	Under normal circumstances you don't need to call this function, as
58			this module automatically starts a single async thread.
59
60	root	1.2	Linux::AIO::max_parallel $nthreads
61			Sets the maximum number of AIO threads to $nthreads. If more than
62			the specified number of threads are currently running, kill them.
63			This function blocks until the limit is reached.
64
65			This module automatically runs "max_parallel 0" at program end, to
66			ensure that all threads are killed and that there are no outstanding
67			requests.
68
69	root	1.4	Under normal circumstances you don't need to call this function.
70
71	root	1.1	$fileno = Linux::AIO::poll_fileno
72			Return the request result pipe filehandle. This filehandle must be
73			polled for reading by some mechanism outside this module (e.g. Event
74			or select, see below). If the pipe becomes readable you have to call
75			"poll_cb" to check the results.
76
77	root	1.4	See "poll_cb" for an example.
78
79	root	1.1	Linux::AIO::poll_cb
80			Process all outstanding events on the result pipe. You have to call
81			this regularly. Returns the number of events processed. Returns
82			immediately when no events are outstanding.
83
84			You can use Event to multiplex, e.g.:
85
86			Event->io (fd => Linux::AIO::poll_fileno,
87			poll => 'r', async => 1,
88	root	1.4	cb => \&Linux::AIO::poll_cb);
89	root	1.1
90	root	1.2	Linux::AIO::poll_wait
91			Wait till the result filehandle becomes ready for reading (simply
92			does a select on the filehandle. This is useful if you want to
93			synchronously wait for some requests to finish).
94
95	root	1.4	See "nreqs" for an example.
96
97	root	1.1	Linux::AIO::nreqs
98			Returns the number of requests currently outstanding.
99
100	root	1.4	Example: wait till there are no outstanding requests anymore:
101
102	root	1.5	Linux::AIO::poll_wait, Linux::AIO::poll_cb
103			while Linux::AIO::nreqs;
104	root	1.4
105	root	1.1	aio_open $pathname, $flags, $mode, $callback
106			Asynchronously open or create a file and call the callback with the
107			filedescriptor (NOT a perl filehandle, sorry for that, but watch
108			out, this might change in the future).
109
110	root	1.5	The pathname passed to "aio_open" must be absolute. See API NOTES,
111			above, for an explanation.
112
113	root	1.4	The $mode argument is a bitmask. See the "Fcntl" module for a list.
114			They are the same as used in "sysopen".
115
116			Example:
117
118			aio_open "/etc/passwd", O_RDONLY, 0, sub {
119			if ($_[0] >= 0) {
120	root	1.6	open my $fh, "<&=$_[0]";
121	root	1.4	print "open successful, fh is $fh\n";
122			...
123			} else {
124			die "open failed: $!\n";
125			}
126			};
127
128	root	1.1	aio_close $fh, $callback
129			Asynchronously close a file and call the callback with the result
130			code.
131
132			aio_read $fh,$offset,$length, $data,$dataoffset,$callback
133			aio_write $fh,$offset,$length, $data,$dataoffset,$callback
134			Reads or writes "length" bytes from the specified "fh" and "offset"
135			into the scalar given by "data" and offset "dataoffset" and calls
136	root	1.4	the callback without the actual number of bytes read (or -1 on
137			error, just like the syscall).
138
139			Example: Read 15 bytes at offset 7 into scalar $buffer, strating at
140			offset 0 within the scalar:
141
142			aio_read $fh, 7, 15, $buffer, 0, sub {
143			$_[0] >= 0 or die "read error: $!";
144			print "read <$buffer>\n";
145			};
146
147			aio_readahead $fh,$offset,$length, $callback
148			Asynchronously reads the specified byte range into the page cache,
149			using the "readahead" syscall.
150
151			readahead() populates the page cache with data from a file so that
152			subsequent reads from that file will not block on disk I/O. The
153			$offset argument specifies the starting point from which data is to
154			be read and $length specifies the number of bytes to be read. I/O is
155			performed in whole pages, so that offset is effectively rounded down
156			to a page boundary and bytes are read up to the next page boundary
157			greater than or equal to (off-set+length). aio_readahead() does not
158			read beyond the end of the file. The current file offset of the file
159			is left unchanged.
160	root	1.1
161			aio_stat $fh_or_path, $callback
162			aio_lstat $fh, $callback
163			Works like perl's "stat" or "lstat" in void context. The callback
164			will be called after the stat and the results will be available
165			using "stat _" or "-s _" etc...
166
167	root	1.5	The pathname passed to "aio_stat" must be absolute. See API NOTES,
168			above, for an explanation.
169
170	root	1.1	Currently, the stats are always 64-bit-stats, i.e. instead of
171			returning an error when stat'ing a large file, the results will be
172			silently truncated unless perl itself is compiled with large file
173			support.
174
175	root	1.4	Example: Print the length of /etc/passwd:
176
177			aio_stat "/etc/passwd", sub {
178			$_[0] and die "stat failed: $!";
179			print "size is ", -s _, "\n";
180			};
181
182	root	1.1	aio_unlink $pathname, $callback
183	root	1.4	Asynchronously unlink (delete) a file and call the callback with the
184			result code.
185
186			aio_fsync $fh, $callback
187			Asynchronously call fsync on the given filehandle and call the
188			callback with the fsync result code.
189
190			aio_fdatasync $fh, $callback
191			Asynchronously call fdatasync on the given filehandle and call the
192			callback with the fdatasync result code.
193	root	1.1
194			BUGS
195			This module has been extensively tested in a large and very busy
196			webserver for many years now.
197
198			- aio_open gives a fd, but all other functions expect a perl filehandle.
199
200			SEE ALSO
201	root	1.5	Coro, IO::AIO.
202	root	1.1
203			AUTHOR
204			Marc Lehmann <schmorp@schmorp.de>
205			http://home.schmorp.de/
206