ViewVC Help
View File | Revision Log | Show Annotations | Download File
/cvs/IO-AIO/README
(Generate patch)

Comparing IO-AIO/README (file contents):
Revision 1.8 by root, Sun Jul 31 17:24:05 2005 UTC vs.
Revision 1.60 by root, Tue Jul 31 22:27:49 2018 UTC

1NAME 1NAME
2 IO::AIO - Asynchronous Input/Output 2 IO::AIO - Asynchronous/Advanced Input/Output
3 3
4SYNOPSIS 4SYNOPSIS
5 use IO::AIO; 5 use IO::AIO;
6 6
7 aio_open "/etc/passwd", O_RDONLY, 0, sub { 7 aio_open "/etc/passwd", IO::AIO::O_RDONLY, 0, sub {
8 my ($fh) = @_; 8 my $fh = shift
9 or die "/etc/passwd: $!";
9 ... 10 ...
10 }; 11 };
11 12
12 aio_unlink "/tmp/file", sub { }; 13 aio_unlink "/tmp/file", sub { };
13 14
14 aio_read $fh, 30000, 1024, $buffer, 0, sub { 15 aio_read $fh, 30000, 1024, $buffer, 0, sub {
15 $_[0] > 0 or die "read error: $!"; 16 $_[0] > 0 or die "read error: $!";
16 }; 17 };
17 18
18 # Event 19 # version 2+ has request and group objects
19 Event->io (fd => IO::AIO::poll_fileno, 20 use IO::AIO 2;
20 poll => 'r',
21 cb => \&IO::AIO::poll_cb);
22 21
23 # Glib/Gtk2 22 aioreq_pri 4; # give next request a very high priority
24 add_watch Glib::IO IO::AIO::poll_fileno, 23 my $req = aio_unlink "/tmp/file", sub { };
25 in => sub { IO::AIO::poll_cb; 1 }; 24 $req->cancel; # cancel request if still in queue
26 25
27 # Tk 26 my $grp = aio_group sub { print "all stats done\n" };
28 Tk::Event::IO->fileevent (IO::AIO::poll_fileno, "", 27 add $grp aio_stat "..." for ...;
29 readable => \&IO::AIO::poll_cb);
30
31 # Danga::Socket
32 Danga::Socket->AddOtherFds (IO::AIO::poll_fileno =>
33 \&IO::AIO::poll_cb);
34 28
35DESCRIPTION 29DESCRIPTION
36 This module implements asynchronous I/O using whatever means your 30 This module implements asynchronous I/O using whatever means your
37 operating system supports. 31 operating system supports. It is implemented as an interface to "libeio"
32 (<http://software.schmorp.de/pkg/libeio.html>).
38 33
34 Asynchronous means that operations that can normally block your program
35 (e.g. reading from disk) will be done asynchronously: the operation will
36 still block, but you can do something else in the meantime. This is
37 extremely useful for programs that need to stay interactive even when
38 doing heavy I/O (GUI programs, high performance network servers etc.),
39 but can also be used to easily do operations in parallel that are
40 normally done sequentially, e.g. stat'ing many files, which is much
41 faster on a RAID volume or over NFS when you do a number of stat
42 operations concurrently.
43
44 While most of this works on all types of file descriptors (for example
45 sockets), using these functions on file descriptors that support
46 nonblocking operation (again, sockets, pipes etc.) is very inefficient.
47 Use an event loop for that (such as the EV module): IO::AIO will
48 naturally fit into such an event loop itself.
49
39 Currently, a number of threads are started that execute your read/writes 50 In this version, a number of threads are started that execute your
40 and signal their completion. You don't need thread support in your libc 51 requests and signal their completion. You don't need thread support in
41 or perl, and the threads created by this module will not be visible to 52 perl, and the threads created by this module will not be visible to
42 the pthreads library. In the future, this module might make use of the 53 perl. In the future, this module might make use of the native aio
43 native aio functions available on many operating systems. However, they 54 functions available on many operating systems. However, they are often
44 are often not well-supported (Linux doesn't allow them on normal files 55 not well-supported or restricted (GNU/Linux doesn't allow them on normal
45 currently, for example), and they would only support aio_read and 56 files currently, for example), and they would only support aio_read and
46 aio_write, so the remaining functionality would have to be implemented 57 aio_write, so the remaining functionality would have to be implemented
47 using threads anyway. 58 using threads anyway.
48 59
60 In addition to asynchronous I/O, this module also exports some rather
61 arcane interfaces, such as "madvise" or linux's "splice" system call,
62 which is why the "A" in "AIO" can also mean *advanced*.
63
49 Although the module will work with in the presence of other threads, it 64 Although the module will work in the presence of other (Perl-) threads,
50 is currently not reentrant, so use appropriate locking yourself, always 65 it is currently not reentrant in any way, so use appropriate locking
51 call "poll_cb" from within the same thread, or never call "poll_cb" (or 66 yourself, always call "poll_cb" from within the same thread, or never
52 other "aio_" functions) recursively. 67 call "poll_cb" (or other "aio_" functions) recursively.
68
69 EXAMPLE
70 This is a simple example that uses the EV module and loads /etc/passwd
71 asynchronously:
72
73 use EV;
74 use IO::AIO;
75
76 # register the IO::AIO callback with EV
77 my $aio_w = EV::io IO::AIO::poll_fileno, EV::READ, \&IO::AIO::poll_cb;
78
79 # queue the request to open /etc/passwd
80 aio_open "/etc/passwd", IO::AIO::O_RDONLY, 0, sub {
81 my $fh = shift
82 or die "error while opening: $!";
83
84 # stat'ing filehandles is generally non-blocking
85 my $size = -s $fh;
86
87 # queue a request to read the file
88 my $contents;
89 aio_read $fh, 0, $size, $contents, 0, sub {
90 $_[0] == $size
91 or die "short read: $!";
92
93 close $fh;
94
95 # file contents now in $contents
96 print $contents;
97
98 # exit event loop and program
99 EV::break;
100 };
101 };
102
103 # possibly queue up other requests, or open GUI windows,
104 # check for sockets etc. etc.
105
106 # process events as long as there are some:
107 EV::run;
108
109REQUEST ANATOMY AND LIFETIME
110 Every "aio_*" function creates a request. which is a C data structure
111 not directly visible to Perl.
112
113 If called in non-void context, every request function returns a Perl
114 object representing the request. In void context, nothing is returned,
115 which saves a bit of memory.
116
117 The perl object is a fairly standard ref-to-hash object. The hash
118 contents are not used by IO::AIO so you are free to store anything you
119 like in it.
120
121 During their existance, aio requests travel through the following
122 states, in order:
123
124 ready
125 Immediately after a request is created it is put into the ready
126 state, waiting for a thread to execute it.
127
128 execute
129 A thread has accepted the request for processing and is currently
130 executing it (e.g. blocking in read).
131
132 pending
133 The request has been executed and is waiting for result processing.
134
135 While request submission and execution is fully asynchronous, result
136 processing is not and relies on the perl interpreter calling
137 "poll_cb" (or another function with the same effect).
138
139 result
140 The request results are processed synchronously by "poll_cb".
141
142 The "poll_cb" function will process all outstanding aio requests by
143 calling their callbacks, freeing memory associated with them and
144 managing any groups they are contained in.
145
146 done
147 Request has reached the end of its lifetime and holds no resources
148 anymore (except possibly for the Perl object, but its connection to
149 the actual aio request is severed and calling its methods will
150 either do nothing or result in a runtime error).
53 151
54FUNCTIONS 152FUNCTIONS
55 AIO FUNCTIONS 153 QUICK OVERVIEW
154 This section simply lists the prototypes most of the functions for quick
155 reference. See the following sections for function-by-function
156 documentation.
157
158 aio_wd $pathname, $callback->($wd)
159 aio_open $pathname, $flags, $mode, $callback->($fh)
160 aio_close $fh, $callback->($status)
161 aio_seek $fh,$offset,$whence, $callback->($offs)
162 aio_read $fh,$offset,$length, $data,$dataoffset, $callback->($retval)
163 aio_write $fh,$offset,$length, $data,$dataoffset, $callback->($retval)
164 aio_sendfile $out_fh, $in_fh, $in_offset, $length, $callback->($retval)
165 aio_readahead $fh,$offset,$length, $callback->($retval)
166 aio_stat $fh_or_path, $callback->($status)
167 aio_lstat $fh, $callback->($status)
168 aio_statvfs $fh_or_path, $callback->($statvfs)
169 aio_utime $fh_or_path, $atime, $mtime, $callback->($status)
170 aio_chown $fh_or_path, $uid, $gid, $callback->($status)
171 aio_chmod $fh_or_path, $mode, $callback->($status)
172 aio_truncate $fh_or_path, $offset, $callback->($status)
173 aio_allocate $fh, $mode, $offset, $len, $callback->($status)
174 aio_fiemap $fh, $start, $length, $flags, $count, $cb->(\@extents)
175 aio_unlink $pathname, $callback->($status)
176 aio_mknod $pathname, $mode, $dev, $callback->($status)
177 aio_link $srcpath, $dstpath, $callback->($status)
178 aio_symlink $srcpath, $dstpath, $callback->($status)
179 aio_readlink $pathname, $callback->($link)
180 aio_realpath $pathname, $callback->($path)
181 aio_rename $srcpath, $dstpath, $callback->($status)
182 aio_rename2 $srcpath, $dstpath, $flags, $callback->($status)
183 aio_mkdir $pathname, $mode, $callback->($status)
184 aio_rmdir $pathname, $callback->($status)
185 aio_readdir $pathname, $callback->($entries)
186 aio_readdirx $pathname, $flags, $callback->($entries, $flags)
187 IO::AIO::READDIR_DENTS IO::AIO::READDIR_DIRS_FIRST
188 IO::AIO::READDIR_STAT_ORDER IO::AIO::READDIR_FOUND_UNKNOWN
189 aio_scandir $pathname, $maxreq, $callback->($dirs, $nondirs)
190 aio_load $pathname, $data, $callback->($status)
191 aio_copy $srcpath, $dstpath, $callback->($status)
192 aio_move $srcpath, $dstpath, $callback->($status)
193 aio_rmtree $pathname, $callback->($status)
194 aio_fcntl $fh, $cmd, $arg, $callback->($status)
195 aio_ioctl $fh, $request, $buf, $callback->($status)
196 aio_sync $callback->($status)
197 aio_syncfs $fh, $callback->($status)
198 aio_fsync $fh, $callback->($status)
199 aio_fdatasync $fh, $callback->($status)
200 aio_sync_file_range $fh, $offset, $nbytes, $flags, $callback->($status)
201 aio_pathsync $pathname, $callback->($status)
202 aio_msync $scalar, $offset = 0, $length = undef, flags = MS_SYNC, $callback->($status)
203 aio_mtouch $scalar, $offset = 0, $length = undef, flags = 0, $callback->($status)
204 aio_mlock $scalar, $offset = 0, $length = undef, $callback->($status)
205 aio_mlockall $flags, $callback->($status)
206 aio_group $callback->(...)
207 aio_nop $callback->()
208
209 $prev_pri = aioreq_pri [$pri]
210 aioreq_nice $pri_adjust
211
212 IO::AIO::poll_wait
213 IO::AIO::poll_cb
214 IO::AIO::poll
215 IO::AIO::flush
216 IO::AIO::max_poll_reqs $nreqs
217 IO::AIO::max_poll_time $seconds
218 IO::AIO::min_parallel $nthreads
219 IO::AIO::max_parallel $nthreads
220 IO::AIO::max_idle $nthreads
221 IO::AIO::idle_timeout $seconds
222 IO::AIO::max_outstanding $maxreqs
223 IO::AIO::nreqs
224 IO::AIO::nready
225 IO::AIO::npending
226 $nfd = IO::AIO::get_fdlimit [EXPERIMENTAL]
227 IO::AIO::min_fdlimit $nfd [EXPERIMENTAL]
228
229 IO::AIO::sendfile $ofh, $ifh, $offset, $count
230 IO::AIO::fadvise $fh, $offset, $len, $advice
231 IO::AIO::mmap $scalar, $length, $prot, $flags[, $fh[, $offset]]
232 IO::AIO::munmap $scalar
233 IO::AIO::mremap $scalar, $new_length, $flags[, $new_address]
234 IO::AIO::madvise $scalar, $offset, $length, $advice
235 IO::AIO::mprotect $scalar, $offset, $length, $protect
236 IO::AIO::munlock $scalar, $offset = 0, $length = undef
237 IO::AIO::munlockall
238
239 API NOTES
56 All the "aio_*" calls are more or less thin wrappers around the syscall 240 All the "aio_*" calls are more or less thin wrappers around the syscall
57 with the same name (sans "aio_"). The arguments are similar or 241 with the same name (sans "aio_"). The arguments are similar or
58 identical, and they all accept an additional (and optional) $callback 242 identical, and they all accept an additional (and optional) $callback
59 argument which must be a code reference. This code reference will get 243 argument which must be a code reference. This code reference will be
60 called with the syscall return code (e.g. most syscalls return -1 on 244 called after the syscall has been executed in an asynchronous fashion.
61 error, unlike perl, which usually delivers "false") as it's sole 245 The results of the request will be passed as arguments to the callback
62 argument when the given syscall has been executed asynchronously. 246 (and, if an error occured, in $!) - for most requests the syscall return
247 code (e.g. most syscalls return -1 on error, unlike perl, which usually
248 delivers "false").
249
250 Some requests (such as "aio_readdir") pass the actual results and
251 communicate failures by passing "undef".
63 252
64 All functions expecting a filehandle keep a copy of the filehandle 253 All functions expecting a filehandle keep a copy of the filehandle
65 internally until the request has finished. 254 internally until the request has finished.
66 255
256 All functions return request objects of type IO::AIO::REQ that allow
257 further manipulation of those requests while they are in-flight.
258
67 The filenames you pass to these routines *must* be absolute. The reason 259 The pathnames you pass to these routines *should* be absolute. The
68 for this is that at the time the request is being executed, the current 260 reason for this is that at the time the request is being executed, the
69 working directory could have changed. Alternatively, you can make sure 261 current working directory could have changed. Alternatively, you can
70 that you never change the current working directory. 262 make sure that you never change the current working directory anywhere
263 in the program and then use relative paths. You can also take advantage
264 of IO::AIOs working directory abstraction, that lets you specify paths
265 relative to some previously-opened "working directory object" - see the
266 description of the "IO::AIO::WD" class later in this document.
71 267
268 To encode pathnames as octets, either make sure you either: a) always
269 pass in filenames you got from outside (command line, readdir etc.)
270 without tinkering, b) are in your native filesystem encoding, c) use the
271 Encode module and encode your pathnames to the locale (or other)
272 encoding in effect in the user environment, d) use
273 Glib::filename_from_unicode on unicode filenames or e) use something
274 else to ensure your scalar has the correct contents.
275
276 This works, btw. independent of the internal UTF-8 bit, which IO::AIO
277 handles correctly whether it is set or not.
278
279 AIO REQUEST FUNCTIONS
280 $prev_pri = aioreq_pri [$pri]
281 Returns the priority value that would be used for the next request
282 and, if $pri is given, sets the priority for the next aio request.
283
284 The default priority is 0, the minimum and maximum priorities are -4
285 and 4, respectively. Requests with higher priority will be serviced
286 first.
287
288 The priority will be reset to 0 after each call to one of the
289 "aio_*" functions.
290
291 Example: open a file with low priority, then read something from it
292 with higher priority so the read request is serviced before other
293 low priority open requests (potentially spamming the cache):
294
295 aioreq_pri -3;
296 aio_open ..., sub {
297 return unless $_[0];
298
299 aioreq_pri -2;
300 aio_read $_[0], ..., sub {
301 ...
302 };
303 };
304
305 aioreq_nice $pri_adjust
306 Similar to "aioreq_pri", but subtracts the given value from the
307 current priority, so the effect is cumulative.
308
72 aio_open $pathname, $flags, $mode, $callback 309 aio_open $pathname, $flags, $mode, $callback->($fh)
73 Asynchronously open or create a file and call the callback with a 310 Asynchronously open or create a file and call the callback with a
74 newly created filehandle for the file. 311 newly created filehandle for the file (or "undef" in case of an
312 error).
75 313
76 The pathname passed to "aio_open" must be absolute. See API NOTES, 314 The pathname passed to "aio_open" must be absolute. See API NOTES,
77 above, for an explanation. 315 above, for an explanation.
78 316
79 The $flags argument is a bitmask. See the "Fcntl" module for a list. 317 The $flags argument is a bitmask. See the "Fcntl" module for a list.
80 They are the same as used by "sysopen". 318 They are the same as used by "sysopen".
81 319
82 Likewise, $mode specifies the mode of the newly created file, if it 320 Likewise, $mode specifies the mode of the newly created file, if it
83 didn't exist and "O_CREAT" has been given, just like perl's 321 didn't exist and "O_CREAT" has been given, just like perl's
84 "sysopen", except that it is mandatory (i.e. use 0 if you don't 322 "sysopen", except that it is mandatory (i.e. use 0 if you don't
85 create new files, and 0666 or 0777 if you do). 323 create new files, and 0666 or 0777 if you do). Note that the $mode
324 will be modified by the umask in effect then the request is being
325 executed, so better never change the umask.
86 326
87 Example: 327 Example:
88 328
89 aio_open "/etc/passwd", O_RDONLY, 0, sub { 329 aio_open "/etc/passwd", IO::AIO::O_RDONLY, 0, sub {
90 if ($_[0]) { 330 if ($_[0]) {
91 print "open successful, fh is $_[0]\n"; 331 print "open successful, fh is $_[0]\n";
92 ... 332 ...
93 } else { 333 } else {
94 die "open failed: $!\n"; 334 die "open failed: $!\n";
95 } 335 }
96 }; 336 };
97 337
338 In addition to all the common open modes/flags ("O_RDONLY",
339 "O_WRONLY", "O_RDWR", "O_CREAT", "O_TRUNC", "O_EXCL" and
340 "O_APPEND"), the following POSIX and non-POSIX constants are
341 available (missing ones on your system are, as usual, 0):
342
343 "O_ASYNC", "O_DIRECT", "O_NOATIME", "O_CLOEXEC", "O_NOCTTY",
344 "O_NOFOLLOW", "O_NONBLOCK", "O_EXEC", "O_SEARCH", "O_DIRECTORY",
345 "O_DSYNC", "O_RSYNC", "O_SYNC", "O_PATH", "O_TMPFILE", "O_TTY_INIT"
346 and "O_ACCMODE".
347
98 aio_close $fh, $callback 348 aio_close $fh, $callback->($status)
99 Asynchronously close a file and call the callback with the result 349 Asynchronously close a file and call the callback with the result
100 code. *WARNING:* although accepted, you should not pass in a perl 350 code.
101 filehandle here, as perl will likely close the file descriptor
102 another time when the filehandle is destroyed. Normally, you can
103 safely call perls "close" or just let filehandles go out of scope.
104 351
105 This is supposed to be a bug in the API, so that might change. It's 352 Unfortunately, you can't do this to perl. Perl *insists* very
106 therefore best to avoid this function. 353 strongly on closing the file descriptor associated with the
354 filehandle itself.
107 355
356 Therefore, "aio_close" will not close the filehandle - instead it
357 will use dup2 to overwrite the file descriptor with the write-end of
358 a pipe (the pipe fd will be created on demand and will be cached).
359
360 Or in other words: the file descriptor will be closed, but it will
361 not be free for reuse until the perl filehandle is closed.
362
363 aio_seek $fh, $offset, $whence, $callback->($offs)
364 Seeks the filehandle to the new $offset, similarly to perl's
365 "sysseek". The $whence can use the traditional values (0 for
366 "IO::AIO::SEEK_SET", 1 for "IO::AIO::SEEK_CUR" or 2 for
367 "IO::AIO::SEEK_END").
368
369 The resulting absolute offset will be passed to the callback, or -1
370 in case of an error.
371
372 In theory, the $whence constants could be different than the
373 corresponding values from Fcntl, but perl guarantees they are the
374 same, so don't panic.
375
376 As a GNU/Linux (and maybe Solaris) extension, also the constants
377 "IO::AIO::SEEK_DATA" and "IO::AIO::SEEK_HOLE" are available, if they
378 could be found. No guarantees about suitability for use in
379 "aio_seek" or Perl's "sysseek" can be made though, although I would
380 naively assume they "just work".
381
108 aio_read $fh,$offset,$length, $data,$dataoffset,$callback 382 aio_read $fh,$offset,$length, $data,$dataoffset, $callback->($retval)
109 aio_write $fh,$offset,$length, $data,$dataoffset,$callback 383 aio_write $fh,$offset,$length, $data,$dataoffset, $callback->($retval)
110 Reads or writes "length" bytes from the specified "fh" and "offset" 384 Reads or writes $length bytes from or to the specified $fh and
111 into the scalar given by "data" and offset "dataoffset" and calls 385 $offset into the scalar given by $data and offset $dataoffset and
112 the callback without the actual number of bytes read (or -1 on 386 calls the callback with the actual number of bytes transferred (or
113 error, just like the syscall). 387 -1 on error, just like the syscall).
388
389 "aio_read" will, like "sysread", shrink or grow the $data scalar to
390 offset plus the actual number of bytes read.
391
392 If $offset is undefined, then the current file descriptor offset
393 will be used (and updated), otherwise the file descriptor offset
394 will not be changed by these calls.
395
396 If $length is undefined in "aio_write", use the remaining length of
397 $data.
398
399 If $dataoffset is less than zero, it will be counted from the end of
400 $data.
401
402 The $data scalar *MUST NOT* be modified in any way while the request
403 is outstanding. Modifying it can result in segfaults or World War
404 III (if the necessary/optional hardware is installed).
114 405
115 Example: Read 15 bytes at offset 7 into scalar $buffer, starting at 406 Example: Read 15 bytes at offset 7 into scalar $buffer, starting at
116 offset 0 within the scalar: 407 offset 0 within the scalar:
117 408
118 aio_read $fh, 7, 15, $buffer, 0, sub { 409 aio_read $fh, 7, 15, $buffer, 0, sub {
119 $_[0] > 0 or die "read error: $!"; 410 $_[0] > 0 or die "read error: $!";
120 print "read $_[0] bytes: <$buffer>\n"; 411 print "read $_[0] bytes: <$buffer>\n";
121 }; 412 };
122 413
414 aio_sendfile $out_fh, $in_fh, $in_offset, $length, $callback->($retval)
415 Tries to copy $length bytes from $in_fh to $out_fh. It starts
416 reading at byte offset $in_offset, and starts writing at the current
417 file offset of $out_fh. Because of that, it is not safe to issue
418 more than one "aio_sendfile" per $out_fh, as they will interfere
419 with each other. The same $in_fh works fine though, as this function
420 does not move or use the file offset of $in_fh.
421
422 Please note that "aio_sendfile" can read more bytes from $in_fh than
423 are written, and there is no way to find out how many more bytes
424 have been read from "aio_sendfile" alone, as "aio_sendfile" only
425 provides the number of bytes written to $out_fh. Only if the result
426 value equals $length one can assume that $length bytes have been
427 read.
428
429 Unlike with other "aio_" functions, it makes a lot of sense to use
430 "aio_sendfile" on non-blocking sockets, as long as one end
431 (typically the $in_fh) is a file - the file I/O will then be
432 asynchronous, while the socket I/O will be non-blocking. Note,
433 however, that you can run into a trap where "aio_sendfile" reads
434 some data with readahead, then fails to write all data, and when the
435 socket is ready the next time, the data in the cache is already
436 lost, forcing "aio_sendfile" to again hit the disk. Explicit
437 "aio_read" + "aio_write" let's you better control resource usage.
438
439 This call tries to make use of a native "sendfile"-like syscall to
440 provide zero-copy operation. For this to work, $out_fh should refer
441 to a socket, and $in_fh should refer to an mmap'able file.
442
443 If a native sendfile cannot be found or it fails with "ENOSYS",
444 "EINVAL", "ENOTSUP", "EOPNOTSUPP", "EAFNOSUPPORT", "EPROTOTYPE" or
445 "ENOTSOCK", it will be emulated, so you can call "aio_sendfile" on
446 any type of filehandle regardless of the limitations of the
447 operating system.
448
449 As native sendfile syscalls (as practically any non-POSIX interface
450 hacked together in a hurry to improve benchmark numbers) tend to be
451 rather buggy on many systems, this implementation tries to work
452 around some known bugs in Linux and FreeBSD kernels (probably
453 others, too), but that might fail, so you really really should check
454 the return value of "aio_sendfile" - fewer bytes than expected might
455 have been transferred.
456
123 aio_readahead $fh,$offset,$length, $callback 457 aio_readahead $fh,$offset,$length, $callback->($retval)
124 Asynchronously reads the specified byte range into the page cache,
125 using the "readahead" syscall. If that syscall doesn't exist (likely
126 if your OS isn't Linux) the status will be -1 and $! is set to
127 "ENOSYS".
128
129 "aio_readahead" populates the page cache with data from a file so 458 "aio_readahead" populates the page cache with data from a file so
130 that subsequent reads from that file will not block on disk I/O. The 459 that subsequent reads from that file will not block on disk I/O. The
131 $offset argument specifies the starting point from which data is to 460 $offset argument specifies the starting point from which data is to
132 be read and $length specifies the number of bytes to be read. I/O is 461 be read and $length specifies the number of bytes to be read. I/O is
133 performed in whole pages, so that offset is effectively rounded down 462 performed in whole pages, so that offset is effectively rounded down
134 to a page boundary and bytes are read up to the next page boundary 463 to a page boundary and bytes are read up to the next page boundary
135 greater than or equal to (off-set+length). "aio_readahead" does not 464 greater than or equal to (off-set+length). "aio_readahead" does not
136 read beyond the end of the file. The current file offset of the file 465 read beyond the end of the file. The current file offset of the file
137 is left unchanged. 466 is left unchanged.
138 467
468 If that syscall doesn't exist (likely if your kernel isn't Linux) it
469 will be emulated by simply reading the data, which would have a
470 similar effect.
471
139 aio_stat $fh_or_path, $callback 472 aio_stat $fh_or_path, $callback->($status)
140 aio_lstat $fh, $callback 473 aio_lstat $fh, $callback->($status)
141 Works like perl's "stat" or "lstat" in void context. The callback 474 Works like perl's "stat" or "lstat" in void context. The callback
142 will be called after the stat and the results will be available 475 will be called after the stat and the results will be available
143 using "stat _" or "-s _" etc... 476 using "stat _" or "-s _" etc...
144 477
145 The pathname passed to "aio_stat" must be absolute. See API NOTES, 478 The pathname passed to "aio_stat" must be absolute. See API NOTES,
… …
148 Currently, the stats are always 64-bit-stats, i.e. instead of 481 Currently, the stats are always 64-bit-stats, i.e. instead of
149 returning an error when stat'ing a large file, the results will be 482 returning an error when stat'ing a large file, the results will be
150 silently truncated unless perl itself is compiled with large file 483 silently truncated unless perl itself is compiled with large file
151 support. 484 support.
152 485
486 To help interpret the mode and dev/rdev stat values, IO::AIO offers
487 the following constants and functions (if not implemented, the
488 constants will be 0 and the functions will either "croak" or fall
489 back on traditional behaviour).
490
491 "S_IFMT", "S_IFIFO", "S_IFCHR", "S_IFBLK", "S_IFLNK", "S_IFREG",
492 "S_IFDIR", "S_IFWHT", "S_IFSOCK", "IO::AIO::major $dev_t",
493 "IO::AIO::minor $dev_t", "IO::AIO::makedev $major, $minor".
494
153 Example: Print the length of /etc/passwd: 495 Example: Print the length of /etc/passwd:
154 496
155 aio_stat "/etc/passwd", sub { 497 aio_stat "/etc/passwd", sub {
156 $_[0] and die "stat failed: $!"; 498 $_[0] and die "stat failed: $!";
157 print "size is ", -s _, "\n"; 499 print "size is ", -s _, "\n";
158 }; 500 };
159 501
502 aio_statvfs $fh_or_path, $callback->($statvfs)
503 Works like the POSIX "statvfs" or "fstatvfs" syscalls, depending on
504 whether a file handle or path was passed.
505
506 On success, the callback is passed a hash reference with the
507 following members: "bsize", "frsize", "blocks", "bfree", "bavail",
508 "files", "ffree", "favail", "fsid", "flag" and "namemax". On
509 failure, "undef" is passed.
510
511 The following POSIX IO::AIO::ST_* constants are defined: "ST_RDONLY"
512 and "ST_NOSUID".
513
514 The following non-POSIX IO::AIO::ST_* flag masks are defined to
515 their correct value when available, or to 0 on systems that do not
516 support them: "ST_NODEV", "ST_NOEXEC", "ST_SYNCHRONOUS",
517 "ST_MANDLOCK", "ST_WRITE", "ST_APPEND", "ST_IMMUTABLE",
518 "ST_NOATIME", "ST_NODIRATIME" and "ST_RELATIME".
519
520 Example: stat "/wd" and dump out the data if successful.
521
522 aio_statvfs "/wd", sub {
523 my $f = $_[0]
524 or die "statvfs: $!";
525
526 use Data::Dumper;
527 say Dumper $f;
528 };
529
530 # result:
531 {
532 bsize => 1024,
533 bfree => 4333064312,
534 blocks => 10253828096,
535 files => 2050765568,
536 flag => 4096,
537 favail => 2042092649,
538 bavail => 4333064312,
539 ffree => 2042092649,
540 namemax => 255,
541 frsize => 1024,
542 fsid => 1810
543 }
544
545 aio_utime $fh_or_path, $atime, $mtime, $callback->($status)
546 Works like perl's "utime" function (including the special case of
547 $atime and $mtime being undef). Fractional times are supported if
548 the underlying syscalls support them.
549
550 When called with a pathname, uses utimes(2) if available, otherwise
551 utime(2). If called on a file descriptor, uses futimes(2) if
552 available, otherwise returns ENOSYS, so this is not portable.
553
554 Examples:
555
556 # set atime and mtime to current time (basically touch(1)):
557 aio_utime "path", undef, undef;
558 # set atime to current time and mtime to beginning of the epoch:
559 aio_utime "path", time, undef; # undef==0
560
561 aio_chown $fh_or_path, $uid, $gid, $callback->($status)
562 Works like perl's "chown" function, except that "undef" for either
563 $uid or $gid is being interpreted as "do not change" (but -1 can
564 also be used).
565
566 Examples:
567
568 # same as "chown root path" in the shell:
569 aio_chown "path", 0, -1;
570 # same as above:
571 aio_chown "path", 0, undef;
572
573 aio_truncate $fh_or_path, $offset, $callback->($status)
574 Works like truncate(2) or ftruncate(2).
575
576 aio_allocate $fh, $mode, $offset, $len, $callback->($status)
577 Allocates or frees disk space according to the $mode argument. See
578 the linux "fallocate" documentation for details.
579
580 $mode is usually 0 or "IO::AIO::FALLOC_FL_KEEP_SIZE" to allocate
581 space, or "IO::AIO::FALLOC_FL_PUNCH_HOLE |
582 IO::AIO::FALLOC_FL_KEEP_SIZE", to deallocate a file range.
583
584 IO::AIO also supports "FALLOC_FL_COLLAPSE_RANGE", to remove a range
585 (without leaving a hole), "FALLOC_FL_ZERO_RANGE", to zero a range,
586 "FALLOC_FL_INSERT_RANGE" to insert a range and
587 "FALLOC_FL_UNSHARE_RANGE" to unshare shared blocks (see your
588 fallocate(2) manpage).
589
590 The file system block size used by "fallocate" is presumably the
591 "f_bsize" returned by "statvfs", but different filesystems and
592 filetypes can dictate other limitations.
593
594 If "fallocate" isn't available or cannot be emulated (currently no
595 emulation will be attempted), passes -1 and sets $! to "ENOSYS".
596
597 aio_chmod $fh_or_path, $mode, $callback->($status)
598 Works like perl's "chmod" function.
599
160 aio_unlink $pathname, $callback 600 aio_unlink $pathname, $callback->($status)
161 Asynchronously unlink (delete) a file and call the callback with the 601 Asynchronously unlink (delete) a file and call the callback with the
162 result code. 602 result code.
163 603
604 aio_mknod $pathname, $mode, $dev, $callback->($status)
605 [EXPERIMENTAL]
606
607 Asynchronously create a device node (or fifo). See mknod(2).
608
609 The only (POSIX-) portable way of calling this function is:
610
611 aio_mknod $pathname, IO::AIO::S_IFIFO | $mode, 0, sub { ...
612
613 See "aio_stat" for info about some potentially helpful extra
614 constants and functions.
615
616 aio_link $srcpath, $dstpath, $callback->($status)
617 Asynchronously create a new link to the existing object at $srcpath
618 at the path $dstpath and call the callback with the result code.
619
620 aio_symlink $srcpath, $dstpath, $callback->($status)
621 Asynchronously create a new symbolic link to the existing object at
622 $srcpath at the path $dstpath and call the callback with the result
623 code.
624
625 aio_readlink $pathname, $callback->($link)
626 Asynchronously read the symlink specified by $path and pass it to
627 the callback. If an error occurs, nothing or undef gets passed to
628 the callback.
629
630 aio_realpath $pathname, $callback->($path)
631 Asynchronously make the path absolute and resolve any symlinks in
632 $path. The resulting path only consists of directories (same as
633 Cwd::realpath).
634
635 This request can be used to get the absolute path of the current
636 working directory by passing it a path of . (a single dot).
637
638 aio_rename $srcpath, $dstpath, $callback->($status)
639 Asynchronously rename the object at $srcpath to $dstpath, just as
640 rename(2) and call the callback with the result code.
641
642 On systems that support the AIO::WD working directory abstraction
643 natively, the case "[$wd, "."]" as $srcpath is specialcased -
644 instead of failing, "rename" is called on the absolute path of $wd.
645
646 aio_rename2 $srcpath, $dstpath, $flags, $callback->($status)
647 Basically a version of "aio_rename" with an additional $flags
648 argument. Calling this with "$flags=0" is the same as calling
649 "aio_rename".
650
651 Non-zero flags are currently only supported on GNU/Linux systems
652 that support renameat2. Other systems fail with "ENOSYS" in this
653 case.
654
655 The following constants are available (missing ones are, as usual
656 0), see renameat2(2) for details:
657
658 "IO::AIO::RENAME_NOREPLACE", "IO::AIO::RENAME_EXCHANGE" and
659 "IO::AIO::RENAME_WHITEOUT".
660
661 aio_mkdir $pathname, $mode, $callback->($status)
662 Asynchronously mkdir (create) a directory and call the callback with
663 the result code. $mode will be modified by the umask at the time the
664 request is executed, so do not change your umask.
665
666 aio_rmdir $pathname, $callback->($status)
667 Asynchronously rmdir (delete) a directory and call the callback with
668 the result code.
669
670 On systems that support the AIO::WD working directory abstraction
671 natively, the case "[$wd, "."]" is specialcased - instead of
672 failing, "rmdir" is called on the absolute path of $wd.
673
674 aio_readdir $pathname, $callback->($entries)
675 Unlike the POSIX call of the same name, "aio_readdir" reads an
676 entire directory (i.e. opendir + readdir + closedir). The entries
677 will not be sorted, and will NOT include the "." and ".." entries.
678
679 The callback is passed a single argument which is either "undef" or
680 an array-ref with the filenames.
681
682 aio_readdirx $pathname, $flags, $callback->($entries, $flags)
683 Quite similar to "aio_readdir", but the $flags argument allows one
684 to tune behaviour and output format. In case of an error, $entries
685 will be "undef".
686
687 The flags are a combination of the following constants, ORed
688 together (the flags will also be passed to the callback, possibly
689 modified):
690
691 IO::AIO::READDIR_DENTS
692 Normally the callback gets an arrayref consisting of names only
693 (as with "aio_readdir"). If this flag is set, then the callback
694 gets an arrayref with "[$name, $type, $inode]" arrayrefs, each
695 describing a single directory entry in more detail:
696
697 $name is the name of the entry.
698
699 $type is one of the "IO::AIO::DT_xxx" constants:
700
701 "IO::AIO::DT_UNKNOWN", "IO::AIO::DT_FIFO", "IO::AIO::DT_CHR",
702 "IO::AIO::DT_DIR", "IO::AIO::DT_BLK", "IO::AIO::DT_REG",
703 "IO::AIO::DT_LNK", "IO::AIO::DT_SOCK", "IO::AIO::DT_WHT".
704
705 "IO::AIO::DT_UNKNOWN" means just that: readdir does not know. If
706 you need to know, you have to run stat yourself. Also, for
707 speed/memory reasons, the $type scalars are read-only: you must
708 not modify them.
709
710 $inode is the inode number (which might not be exact on systems
711 with 64 bit inode numbers and 32 bit perls). This field has
712 unspecified content on systems that do not deliver the inode
713 information.
714
715 IO::AIO::READDIR_DIRS_FIRST
716 When this flag is set, then the names will be returned in an
717 order where likely directories come first, in optimal stat
718 order. This is useful when you need to quickly find directories,
719 or you want to find all directories while avoiding to stat()
720 each entry.
721
722 If the system returns type information in readdir, then this is
723 used to find directories directly. Otherwise, likely directories
724 are names beginning with ".", or otherwise names with no dots,
725 of which names with short names are tried first.
726
727 IO::AIO::READDIR_STAT_ORDER
728 When this flag is set, then the names will be returned in an
729 order suitable for stat()'ing each one. That is, when you plan
730 to stat() most or all files in the given directory, then the
731 returned order will likely be faster.
732
733 If both this flag and "IO::AIO::READDIR_DIRS_FIRST" are
734 specified, then the likely dirs come first, resulting in a less
735 optimal stat order for stat'ing all entries, but likely a more
736 optimal order for finding subdirectories.
737
738 IO::AIO::READDIR_FOUND_UNKNOWN
739 This flag should not be set when calling "aio_readdirx".
740 Instead, it is being set by "aio_readdirx", when any of the
741 $type's found were "IO::AIO::DT_UNKNOWN". The absence of this
742 flag therefore indicates that all $type's are known, which can
743 be used to speed up some algorithms.
744
745 aio_slurp $pathname, $offset, $length, $data, $callback->($status)
746 Opens, reads and closes the given file. The data is put into $data,
747 which is resized as required.
748
749 If $offset is negative, then it is counted from the end of the file.
750
751 If $length is zero, then the remaining length of the file is used.
752 Also, in this case, the same limitations to modifying $data apply as
753 when IO::AIO::mmap is used, i.e. it must only be modified in-place
754 with "substr". If the size of the file is known, specifying a
755 non-zero $length results in a performance advantage.
756
757 This request is similar to the older "aio_load" request, but since
758 it is a single request, it might be more efficient to use.
759
760 Example: load /etc/passwd into $passwd.
761
762 my $passwd;
763 aio_slurp "/etc/passwd", 0, 0, $passwd, sub {
764 $_[0] >= 0
765 or die "/etc/passwd: $!\n";
766
767 printf "/etc/passwd is %d bytes long, and contains:\n", length $passwd;
768 print $passwd;
769 };
770 IO::AIO::flush;
771
772 aio_load $pathname, $data, $callback->($status)
773 This is a composite request that tries to fully load the given file
774 into memory. Status is the same as with aio_read.
775
776 Using "aio_slurp" might be more efficient, as it is a single
777 request.
778
779 aio_copy $srcpath, $dstpath, $callback->($status)
780 Try to copy the *file* (directories not supported as either source
781 or destination) from $srcpath to $dstpath and call the callback with
782 a status of 0 (ok) or -1 (error, see $!).
783
784 Existing destination files will be truncated.
785
786 This is a composite request that creates the destination file with
787 mode 0200 and copies the contents of the source file into it using
788 "aio_sendfile", followed by restoring atime, mtime, access mode and
789 uid/gid, in that order.
790
791 If an error occurs, the partial destination file will be unlinked,
792 if possible, except when setting atime, mtime, access mode and
793 uid/gid, where errors are being ignored.
794
795 aio_move $srcpath, $dstpath, $callback->($status)
796 Try to move the *file* (directories not supported as either source
797 or destination) from $srcpath to $dstpath and call the callback with
798 a status of 0 (ok) or -1 (error, see $!).
799
800 This is a composite request that tries to rename(2) the file first;
801 if rename fails with "EXDEV", it copies the file with "aio_copy"
802 and, if that is successful, unlinks the $srcpath.
803
804 aio_scandir $pathname, $maxreq, $callback->($dirs, $nondirs)
805 Scans a directory (similar to "aio_readdir") but additionally tries
806 to efficiently separate the entries of directory $path into two sets
807 of names, directories you can recurse into (directories), and ones
808 you cannot recurse into (everything else, including symlinks to
809 directories).
810
811 "aio_scandir" is a composite request that generates many sub
812 requests. $maxreq specifies the maximum number of outstanding aio
813 requests that this function generates. If it is "<= 0", then a
814 suitable default will be chosen (currently 4).
815
816 On error, the callback is called without arguments, otherwise it
817 receives two array-refs with path-relative entry names.
818
819 Example:
820
821 aio_scandir $dir, 0, sub {
822 my ($dirs, $nondirs) = @_;
823 print "real directories: @$dirs\n";
824 print "everything else: @$nondirs\n";
825 };
826
827 Implementation notes.
828
829 The "aio_readdir" cannot be avoided, but "stat()"'ing every entry
830 can.
831
832 If readdir returns file type information, then this is used directly
833 to find directories.
834
835 Otherwise, after reading the directory, the modification time, size
836 etc. of the directory before and after the readdir is checked, and
837 if they match (and isn't the current time), the link count will be
838 used to decide how many entries are directories (if >= 2).
839 Otherwise, no knowledge of the number of subdirectories will be
840 assumed.
841
842 Then entries will be sorted into likely directories a non-initial
843 dot currently) and likely non-directories (see "aio_readdirx"). Then
844 every entry plus an appended "/." will be "stat"'ed, likely
845 directories first, in order of their inode numbers. If that
846 succeeds, it assumes that the entry is a directory or a symlink to
847 directory (which will be checked separately). This is often faster
848 than stat'ing the entry itself because filesystems might detect the
849 type of the entry without reading the inode data (e.g. ext2fs
850 filetype feature), even on systems that cannot return the filetype
851 information on readdir.
852
853 If the known number of directories (link count - 2) has been
854 reached, the rest of the entries is assumed to be non-directories.
855
856 This only works with certainty on POSIX (= UNIX) filesystems, which
857 fortunately are the vast majority of filesystems around.
858
859 It will also likely work on non-POSIX filesystems with reduced
860 efficiency as those tend to return 0 or 1 as link counts, which
861 disables the directory counting heuristic.
862
863 aio_rmtree $pathname, $callback->($status)
864 Delete a directory tree starting (and including) $path, return the
865 status of the final "rmdir" only. This is a composite request that
866 uses "aio_scandir" to recurse into and rmdir directories, and unlink
867 everything else.
868
869 aio_fcntl $fh, $cmd, $arg, $callback->($status)
870 aio_ioctl $fh, $request, $buf, $callback->($status)
871 These work just like the "fcntl" and "ioctl" built-in functions,
872 except they execute asynchronously and pass the return value to the
873 callback.
874
875 Both calls can be used for a lot of things, some of which make more
876 sense to run asynchronously in their own thread, while some others
877 make less sense. For example, calls that block waiting for external
878 events, such as locking, will also lock down an I/O thread while it
879 is waiting, which can deadlock the whole I/O system. At the same
880 time, there might be no alternative to using a thread to wait.
881
882 So in general, you should only use these calls for things that do
883 (filesystem) I/O, not for things that wait for other events
884 (network, other processes), although if you are careful and know
885 what you are doing, you still can.
886
887 The following constants are available (missing ones are, as usual
888 0):
889
890 "F_DUPFD_CLOEXEC",
891
892 "F_OFD_GETLK", "F_OFD_SETLK", "F_OFD_GETLKW",
893
894 "FIFREEZE", "FITHAW", "FITRIM", "FICLONE", "FICLONERANGE",
895 "FIDEDUPERANGE".
896
897 "FS_IOC_GETFLAGS", "FS_IOC_SETFLAGS", "FS_IOC_GETVERSION",
898 "FS_IOC_SETVERSION", "FS_IOC_FIEMAP".
899
900 "FS_IOC_FSGETXATTR", "FS_IOC_FSSETXATTR",
901 "FS_IOC_SET_ENCRYPTION_POLICY", "FS_IOC_GET_ENCRYPTION_PWSALT",
902 "FS_IOC_GET_ENCRYPTION_POLICY", "FS_KEY_DESCRIPTOR_SIZE".
903
904 "FS_SECRM_FL", "FS_UNRM_FL", "FS_COMPR_FL", "FS_SYNC_FL",
905 "FS_IMMUTABLE_FL", "FS_APPEND_FL", "FS_NODUMP_FL", "FS_NOATIME_FL",
906 "FS_DIRTY_FL", "FS_COMPRBLK_FL", "FS_NOCOMP_FL", "FS_ENCRYPT_FL",
907 "FS_BTREE_FL", "FS_INDEX_FL", "FS_JOURNAL_DATA_FL", "FS_NOTAIL_FL",
908 "FS_DIRSYNC_FL", "FS_TOPDIR_FL", "FS_FL_USER_MODIFIABLE".
909
910 "FS_XFLAG_REALTIME", "FS_XFLAG_PREALLOC", "FS_XFLAG_IMMUTABLE",
911 "FS_XFLAG_APPEND", "FS_XFLAG_SYNC", "FS_XFLAG_NOATIME",
912 "FS_XFLAG_NODUMP", "FS_XFLAG_RTINHERIT", "FS_XFLAG_PROJINHERIT",
913 "FS_XFLAG_NOSYMLINKS", "FS_XFLAG_EXTSIZE", "FS_XFLAG_EXTSZINHERIT",
914 "FS_XFLAG_NODEFRAG", "FS_XFLAG_FILESTREAM", "FS_XFLAG_DAX",
915 "FS_XFLAG_HASATTR",
916
917 aio_sync $callback->($status)
918 Asynchronously call sync and call the callback when finished.
919
164 aio_fsync $fh, $callback 920 aio_fsync $fh, $callback->($status)
165 Asynchronously call fsync on the given filehandle and call the 921 Asynchronously call fsync on the given filehandle and call the
166 callback with the fsync result code. 922 callback with the fsync result code.
167 923
168 aio_fdatasync $fh, $callback 924 aio_fdatasync $fh, $callback->($status)
169 Asynchronously call fdatasync on the given filehandle and call the 925 Asynchronously call fdatasync on the given filehandle and call the
170 callback with the fdatasync result code. Might set $! to "ENOSYS" if 926 callback with the fdatasync result code.
171 "fdatasync" is not available. 927
928 If this call isn't available because your OS lacks it or it couldn't
929 be detected, it will be emulated by calling "fsync" instead.
930
931 aio_syncfs $fh, $callback->($status)
932 Asynchronously call the syncfs syscall to sync the filesystem
933 associated to the given filehandle and call the callback with the
934 syncfs result code. If syncfs is not available, calls sync(), but
935 returns -1 and sets errno to "ENOSYS" nevertheless.
936
937 aio_sync_file_range $fh, $offset, $nbytes, $flags, $callback->($status)
938 Sync the data portion of the file specified by $offset and $length
939 to disk (but NOT the metadata), by calling the Linux-specific
940 sync_file_range call. If sync_file_range is not available or it
941 returns ENOSYS, then fdatasync or fsync is being substituted.
942
943 $flags can be a combination of
944 "IO::AIO::SYNC_FILE_RANGE_WAIT_BEFORE",
945 "IO::AIO::SYNC_FILE_RANGE_WRITE" and
946 "IO::AIO::SYNC_FILE_RANGE_WAIT_AFTER": refer to the sync_file_range
947 manpage for details.
948
949 aio_pathsync $pathname, $callback->($status)
950 This request tries to open, fsync and close the given path. This is
951 a composite request intended to sync directories after directory
952 operations (E.g. rename). This might not work on all operating
953 systems or have any specific effect, but usually it makes sure that
954 directory changes get written to disc. It works for anything that
955 can be opened for read-only, not just directories.
956
957 Future versions of this function might fall back to other methods
958 when "fsync" on the directory fails (such as calling "sync").
959
960 Passes 0 when everything went ok, and -1 on error.
961
962 aio_msync $scalar, $offset = 0, $length = undef, flags = MS_SYNC,
963 $callback->($status)
964 This is a rather advanced IO::AIO call, which only works on
965 mmap(2)ed scalars (see the "IO::AIO::mmap" function, although it
966 also works on data scalars managed by the Sys::Mmap or Mmap modules,
967 note that the scalar must only be modified in-place while an aio
968 operation is pending on it).
969
970 It calls the "msync" function of your OS, if available, with the
971 memory area starting at $offset in the string and ending $length
972 bytes later. If $length is negative, counts from the end, and if
973 $length is "undef", then it goes till the end of the string. The
974 flags can be either "IO::AIO::MS_ASYNC" or "IO::AIO::MS_SYNC", plus
975 an optional "IO::AIO::MS_INVALIDATE".
976
977 aio_mtouch $scalar, $offset = 0, $length = undef, flags = 0,
978 $callback->($status)
979 This is a rather advanced IO::AIO call, which works best on
980 mmap(2)ed scalars.
981
982 It touches (reads or writes) all memory pages in the specified range
983 inside the scalar. All caveats and parameters are the same as for
984 "aio_msync", above, except for flags, which must be either 0 (which
985 reads all pages and ensures they are instantiated) or
986 "IO::AIO::MT_MODIFY", which modifies the memory pages (by reading
987 and writing an octet from it, which dirties the page).
988
989 aio_mlock $scalar, $offset = 0, $length = undef, $callback->($status)
990 This is a rather advanced IO::AIO call, which works best on
991 mmap(2)ed scalars.
992
993 It reads in all the pages of the underlying storage into memory (if
994 any) and locks them, so they are not getting swapped/paged out or
995 removed.
996
997 If $length is undefined, then the scalar will be locked till the
998 end.
999
1000 On systems that do not implement "mlock", this function returns -1
1001 and sets errno to "ENOSYS".
1002
1003 Note that the corresponding "munlock" is synchronous and is
1004 documented under "MISCELLANEOUS FUNCTIONS".
1005
1006 Example: open a file, mmap and mlock it - both will be undone when
1007 $data gets destroyed.
1008
1009 open my $fh, "<", $path or die "$path: $!";
1010 my $data;
1011 IO::AIO::mmap $data, -s $fh, IO::AIO::PROT_READ, IO::AIO::MAP_SHARED, $fh;
1012 aio_mlock $data; # mlock in background
1013
1014 aio_mlockall $flags, $callback->($status)
1015 Calls the "mlockall" function with the given $flags (a combination
1016 of "IO::AIO::MCL_CURRENT" and "IO::AIO::MCL_FUTURE").
1017
1018 On systems that do not implement "mlockall", this function returns
1019 -1 and sets errno to "ENOSYS".
1020
1021 Note that the corresponding "munlockall" is synchronous and is
1022 documented under "MISCELLANEOUS FUNCTIONS".
1023
1024 Example: asynchronously lock all current and future pages into
1025 memory.
1026
1027 aio_mlockall IO::AIO::MCL_FUTURE;
1028
1029 aio_fiemap $fh, $start, $length, $flags, $count, $cb->(\@extents)
1030 Queries the extents of the given file (by calling the Linux "FIEMAP"
1031 ioctl, see <http://cvs.schmorp.de/IO-AIO/doc/fiemap.txt> for
1032 details). If the ioctl is not available on your OS, then this
1033 request will fail with "ENOSYS".
1034
1035 $start is the starting offset to query extents for, $length is the
1036 size of the range to query - if it is "undef", then the whole file
1037 will be queried.
1038
1039 $flags is a combination of flags ("IO::AIO::FIEMAP_FLAG_SYNC" or
1040 "IO::AIO::FIEMAP_FLAG_XATTR" - "IO::AIO::FIEMAP_FLAGS_COMPAT" is
1041 also exported), and is normally 0 or "IO::AIO::FIEMAP_FLAG_SYNC" to
1042 query the data portion.
1043
1044 $count is the maximum number of extent records to return. If it is
1045 "undef", then IO::AIO queries all extents of the range. As a very
1046 special case, if it is 0, then the callback receives the number of
1047 extents instead of the extents themselves (which is unreliable, see
1048 below).
1049
1050 If an error occurs, the callback receives no arguments. The special
1051 "errno" value "IO::AIO::EBADR" is available to test for flag errors.
1052
1053 Otherwise, the callback receives an array reference with extent
1054 structures. Each extent structure is an array reference itself, with
1055 the following members:
1056
1057 [$logical, $physical, $length, $flags]
1058
1059 Flags is any combination of the following flag values (typically
1060 either 0 or "IO::AIO::FIEMAP_EXTENT_LAST" (1)):
1061
1062 "IO::AIO::FIEMAP_EXTENT_LAST", "IO::AIO::FIEMAP_EXTENT_UNKNOWN",
1063 "IO::AIO::FIEMAP_EXTENT_DELALLOC", "IO::AIO::FIEMAP_EXTENT_ENCODED",
1064 "IO::AIO::FIEMAP_EXTENT_DATA_ENCRYPTED",
1065 "IO::AIO::FIEMAP_EXTENT_NOT_ALIGNED",
1066 "IO::AIO::FIEMAP_EXTENT_DATA_INLINE",
1067 "IO::AIO::FIEMAP_EXTENT_DATA_TAIL",
1068 "IO::AIO::FIEMAP_EXTENT_UNWRITTEN", "IO::AIO::FIEMAP_EXTENT_MERGED"
1069 or "IO::AIO::FIEMAP_EXTENT_SHARED".
1070
1071 At the time of this writing (Linux 3.2), this request is unreliable
1072 unless $count is "undef", as the kernel has all sorts of bugs
1073 preventing it to return all extents of a range for files with a
1074 large number of extents. The code (only) works around all these
1075 issues if $count is "undef".
1076
1077 aio_group $callback->(...)
1078 This is a very special aio request: Instead of doing something, it
1079 is a container for other aio requests, which is useful if you want
1080 to bundle many requests into a single, composite, request with a
1081 definite callback and the ability to cancel the whole request with
1082 its subrequests.
1083
1084 Returns an object of class IO::AIO::GRP. See its documentation below
1085 for more info.
1086
1087 Example:
1088
1089 my $grp = aio_group sub {
1090 print "all stats done\n";
1091 };
1092
1093 add $grp
1094 (aio_stat ...),
1095 (aio_stat ...),
1096 ...;
1097
1098 aio_nop $callback->()
1099 This is a special request - it does nothing in itself and is only
1100 used for side effects, such as when you want to add a dummy request
1101 to a group so that finishing the requests in the group depends on
1102 executing the given code.
1103
1104 While this request does nothing, it still goes through the execution
1105 phase and still requires a worker thread. Thus, the callback will
1106 not be executed immediately but only after other requests in the
1107 queue have entered their execution phase. This can be used to
1108 measure request latency.
1109
1110 IO::AIO::aio_busy $fractional_seconds, $callback->() *NOT EXPORTED*
1111 Mainly used for debugging and benchmarking, this aio request puts
1112 one of the request workers to sleep for the given time.
1113
1114 While it is theoretically handy to have simple I/O scheduling
1115 requests like sleep and file handle readable/writable, the overhead
1116 this creates is immense (it blocks a thread for a long time) so do
1117 not use this function except to put your application under
1118 artificial I/O pressure.
1119
1120 IO::AIO::WD - multiple working directories
1121 Your process only has one current working directory, which is used by
1122 all threads. This makes it hard to use relative paths (some other
1123 component could call "chdir" at any time, and it is hard to control when
1124 the path will be used by IO::AIO).
1125
1126 One solution for this is to always use absolute paths. This usually
1127 works, but can be quite slow (the kernel has to walk the whole path on
1128 every access), and can also be a hassle to implement.
1129
1130 Newer POSIX systems have a number of functions (openat, fdopendir,
1131 futimensat and so on) that make it possible to specify working
1132 directories per operation.
1133
1134 For portability, and because the clowns who "designed", or shall I
1135 write, perpetrated this new interface were obviously half-drunk, this
1136 abstraction cannot be perfect, though.
1137
1138 IO::AIO allows you to convert directory paths into a so-called
1139 IO::AIO::WD object. This object stores the canonicalised, absolute
1140 version of the path, and on systems that allow it, also a directory file
1141 descriptor.
1142
1143 Everywhere where a pathname is accepted by IO::AIO (e.g. in "aio_stat"
1144 or "aio_unlink"), one can specify an array reference with an IO::AIO::WD
1145 object and a pathname instead (or the IO::AIO::WD object alone, which
1146 gets interpreted as "[$wd, "."]"). If the pathname is absolute, the
1147 IO::AIO::WD object is ignored, otherwise the pathname is resolved
1148 relative to that IO::AIO::WD object.
1149
1150 For example, to get a wd object for /etc and then stat passwd inside,
1151 you would write:
1152
1153 aio_wd "/etc", sub {
1154 my $etcdir = shift;
1155
1156 # although $etcdir can be undef on error, there is generally no reason
1157 # to check for errors here, as aio_stat will fail with ENOENT
1158 # when $etcdir is undef.
1159
1160 aio_stat [$etcdir, "passwd"], sub {
1161 # yay
1162 };
1163 };
1164
1165 The fact that "aio_wd" is a request and not a normal function shows that
1166 creating an IO::AIO::WD object is itself a potentially blocking
1167 operation, which is why it is done asynchronously.
1168
1169 To stat the directory obtained with "aio_wd" above, one could write
1170 either of the following three request calls:
1171
1172 aio_lstat "/etc" , sub { ... # pathname as normal string
1173 aio_lstat [$wd, "."], sub { ... # "." relative to $wd (i.e. $wd itself)
1174 aio_lstat $wd , sub { ... # shorthand for the previous
1175
1176 As with normal pathnames, IO::AIO keeps a copy of the working directory
1177 object and the pathname string, so you could write the following without
1178 causing any issues due to $path getting reused:
1179
1180 my $path = [$wd, undef];
1181
1182 for my $name (qw(abc def ghi)) {
1183 $path->[1] = $name;
1184 aio_stat $path, sub {
1185 # ...
1186 };
1187 }
1188
1189 There are some caveats: when directories get renamed (or deleted), the
1190 pathname string doesn't change, so will point to the new directory (or
1191 nowhere at all), while the directory fd, if available on the system,
1192 will still point to the original directory. Most functions accepting a
1193 pathname will use the directory fd on newer systems, and the string on
1194 older systems. Some functions (such as "aio_realpath") will always rely
1195 on the string form of the pathname.
1196
1197 So this functionality is mainly useful to get some protection against
1198 "chdir", to easily get an absolute path out of a relative path for
1199 future reference, and to speed up doing many operations in the same
1200 directory (e.g. when stat'ing all files in a directory).
1201
1202 The following functions implement this working directory abstraction:
1203
1204 aio_wd $pathname, $callback->($wd)
1205 Asynchonously canonicalise the given pathname and convert it to an
1206 IO::AIO::WD object representing it. If possible and supported on the
1207 system, also open a directory fd to speed up pathname resolution
1208 relative to this working directory.
1209
1210 If something goes wrong, then "undef" is passwd to the callback
1211 instead of a working directory object and $! is set appropriately.
1212 Since passing "undef" as working directory component of a pathname
1213 fails the request with "ENOENT", there is often no need for error
1214 checking in the "aio_wd" callback, as future requests using the
1215 value will fail in the expected way.
1216
1217 IO::AIO::CWD
1218 This is a compiletime constant (object) that represents the process
1219 current working directory.
1220
1221 Specifying this object as working directory object for a pathname is
1222 as if the pathname would be specified directly, without a directory
1223 object. For example, these calls are functionally identical:
1224
1225 aio_stat "somefile", sub { ... };
1226 aio_stat [IO::AIO::CWD, "somefile"], sub { ... };
1227
1228 To recover the path associated with an IO::AIO::WD object, you can use
1229 "aio_realpath":
1230
1231 aio_realpath $wd, sub {
1232 warn "path is $_[0]\n";
1233 };
1234
1235 Currently, "aio_statvfs" always, and "aio_rename" and "aio_rmdir"
1236 sometimes, fall back to using an absolue path.
1237
1238 IO::AIO::REQ CLASS
1239 All non-aggregate "aio_*" functions return an object of this class when
1240 called in non-void context.
1241
1242 cancel $req
1243 Cancels the request, if possible. Has the effect of skipping
1244 execution when entering the execute state and skipping calling the
1245 callback when entering the the result state, but will leave the
1246 request otherwise untouched (with the exception of readdir). That
1247 means that requests that currently execute will not be stopped and
1248 resources held by the request will not be freed prematurely.
1249
1250 cb $req $callback->(...)
1251 Replace (or simply set) the callback registered to the request.
1252
1253 IO::AIO::GRP CLASS
1254 This class is a subclass of IO::AIO::REQ, so all its methods apply to
1255 objects of this class, too.
1256
1257 A IO::AIO::GRP object is a special request that can contain multiple
1258 other aio requests.
1259
1260 You create one by calling the "aio_group" constructing function with a
1261 callback that will be called when all contained requests have entered
1262 the "done" state:
1263
1264 my $grp = aio_group sub {
1265 print "all requests are done\n";
1266 };
1267
1268 You add requests by calling the "add" method with one or more
1269 "IO::AIO::REQ" objects:
1270
1271 $grp->add (aio_unlink "...");
1272
1273 add $grp aio_stat "...", sub {
1274 $_[0] or return $grp->result ("error");
1275
1276 # add another request dynamically, if first succeeded
1277 add $grp aio_open "...", sub {
1278 $grp->result ("ok");
1279 };
1280 };
1281
1282 This makes it very easy to create composite requests (see the source of
1283 "aio_move" for an application) that work and feel like simple requests.
1284
1285 * The IO::AIO::GRP objects will be cleaned up during calls to
1286 "IO::AIO::poll_cb", just like any other request.
1287
1288 * They can be canceled like any other request. Canceling will cancel
1289 not only the request itself, but also all requests it contains.
1290
1291 * They can also can also be added to other IO::AIO::GRP objects.
1292
1293 * You must not add requests to a group from within the group callback
1294 (or any later time).
1295
1296 Their lifetime, simplified, looks like this: when they are empty, they
1297 will finish very quickly. If they contain only requests that are in the
1298 "done" state, they will also finish. Otherwise they will continue to
1299 exist.
1300
1301 That means after creating a group you have some time to add requests
1302 (precisely before the callback has been invoked, which is only done
1303 within the "poll_cb"). And in the callbacks of those requests, you can
1304 add further requests to the group. And only when all those requests have
1305 finished will the the group itself finish.
1306
1307 add $grp ...
1308 $grp->add (...)
1309 Add one or more requests to the group. Any type of IO::AIO::REQ can
1310 be added, including other groups, as long as you do not create
1311 circular dependencies.
1312
1313 Returns all its arguments.
1314
1315 $grp->cancel_subs
1316 Cancel all subrequests and clears any feeder, but not the group
1317 request itself. Useful when you queued a lot of events but got a
1318 result early.
1319
1320 The group request will finish normally (you cannot add requests to
1321 the group).
1322
1323 $grp->result (...)
1324 Set the result value(s) that will be passed to the group callback
1325 when all subrequests have finished and set the groups errno to the
1326 current value of errno (just like calling "errno" without an error
1327 number). By default, no argument will be passed and errno is zero.
1328
1329 $grp->errno ([$errno])
1330 Sets the group errno value to $errno, or the current value of errno
1331 when the argument is missing.
1332
1333 Every aio request has an associated errno value that is restored
1334 when the callback is invoked. This method lets you change this value
1335 from its default (0).
1336
1337 Calling "result" will also set errno, so make sure you either set $!
1338 before the call to "result", or call c<errno> after it.
1339
1340 feed $grp $callback->($grp)
1341 Sets a feeder/generator on this group: every group can have an
1342 attached generator that generates requests if idle. The idea behind
1343 this is that, although you could just queue as many requests as you
1344 want in a group, this might starve other requests for a potentially
1345 long time. For example, "aio_scandir" might generate hundreds of
1346 thousands of "aio_stat" requests, delaying any later requests for a
1347 long time.
1348
1349 To avoid this, and allow incremental generation of requests, you can
1350 instead a group and set a feeder on it that generates those
1351 requests. The feed callback will be called whenever there are few
1352 enough (see "limit", below) requests active in the group itself and
1353 is expected to queue more requests.
1354
1355 The feed callback can queue as many requests as it likes (i.e. "add"
1356 does not impose any limits).
1357
1358 If the feed does not queue more requests when called, it will be
1359 automatically removed from the group.
1360
1361 If the feed limit is 0 when this method is called, it will be set to
1362 2 automatically.
1363
1364 Example:
1365
1366 # stat all files in @files, but only ever use four aio requests concurrently:
1367
1368 my $grp = aio_group sub { print "finished\n" };
1369 limit $grp 4;
1370 feed $grp sub {
1371 my $file = pop @files
1372 or return;
1373
1374 add $grp aio_stat $file, sub { ... };
1375 };
1376
1377 limit $grp $num
1378 Sets the feeder limit for the group: The feeder will be called
1379 whenever the group contains less than this many requests.
1380
1381 Setting the limit to 0 will pause the feeding process.
1382
1383 The default value for the limit is 0, but note that setting a feeder
1384 automatically bumps it up to 2.
172 1385
173 SUPPORT FUNCTIONS 1386 SUPPORT FUNCTIONS
1387 EVENT PROCESSING AND EVENT LOOP INTEGRATION
174 $fileno = IO::AIO::poll_fileno 1388 $fileno = IO::AIO::poll_fileno
175 Return the *request result pipe file descriptor*. This filehandle 1389 Return the *request result pipe file descriptor*. This filehandle
176 must be polled for reading by some mechanism outside this module 1390 must be polled for reading by some mechanism outside this module
177 (e.g. Event or select, see below or the SYNOPSIS). If the pipe 1391 (e.g. EV, Glib, select and so on, see below or the SYNOPSIS). If the
178 becomes readable you have to call "poll_cb" to check the results. 1392 pipe becomes readable you have to call "poll_cb" to check the
1393 results.
179 1394
180 See "poll_cb" for an example. 1395 See "poll_cb" for an example.
181 1396
182 IO::AIO::poll_cb 1397 IO::AIO::poll_cb
183 Process all outstanding events on the result pipe. You have to call 1398 Process some requests that have reached the result phase (i.e. they
184 this regularly. Returns the number of events processed. Returns 1399 have been executed but the results are not yet reported). You have
1400 to call this "regularly" to finish outstanding requests.
1401
1402 Returns 0 if all events could be processed (or there were no events
1403 to process), or -1 if it returned earlier for whatever reason.
185 immediately when no events are outstanding. 1404 Returns immediately when no events are outstanding. The amount of
1405 events processed depends on the settings of "IO::AIO::max_poll_req",
1406 "IO::AIO::max_poll_time" and "IO::AIO::max_outstanding".
1407
1408 If not all requests were processed for whatever reason, the poll
1409 file descriptor will still be ready when "poll_cb" returns, so
1410 normally you don't have to do anything special to have it called
1411 later.
1412
1413 Apart from calling "IO::AIO::poll_cb" when the event filehandle
1414 becomes ready, it can be beneficial to call this function from loops
1415 which submit a lot of requests, to make sure the results get
1416 processed when they become available and not just when the loop is
1417 finished and the event loop takes over again. This function returns
1418 very fast when there are no outstanding requests.
186 1419
187 Example: Install an Event watcher that automatically calls 1420 Example: Install an Event watcher that automatically calls
188 IO::AIO::poll_cb with high priority: 1421 IO::AIO::poll_cb with high priority (more examples can be found in
1422 the SYNOPSIS section, at the top of this document):
189 1423
190 Event->io (fd => IO::AIO::poll_fileno, 1424 Event->io (fd => IO::AIO::poll_fileno,
191 poll => 'r', async => 1, 1425 poll => 'r', async => 1,
192 cb => \&IO::AIO::poll_cb); 1426 cb => \&IO::AIO::poll_cb);
193 1427
194 IO::AIO::poll_wait 1428 IO::AIO::poll_wait
195 Wait till the result filehandle becomes ready for reading (simply 1429 Wait until either at least one request is in the result phase or no
196 does a "select" on the filehandle. This is useful if you want to 1430 requests are outstanding anymore.
197 synchronously wait for some requests to finish). 1431
1432 This is useful if you want to synchronously wait for some requests
1433 to become ready, without actually handling them.
198 1434
199 See "nreqs" for an example. 1435 See "nreqs" for an example.
200 1436
1437 IO::AIO::poll
1438 Waits until some requests have been handled.
1439
1440 Returns the number of requests processed, but is otherwise strictly
1441 equivalent to:
1442
1443 IO::AIO::poll_wait, IO::AIO::poll_cb
1444
201 IO::AIO::nreqs 1445 IO::AIO::flush
202 Returns the number of requests currently outstanding (i.e. for which 1446 Wait till all outstanding AIO requests have been handled.
203 their callback has not been invoked yet).
204 1447
205 Example: wait till there are no outstanding requests anymore: 1448 Strictly equivalent to:
206 1449
207 IO::AIO::poll_wait, IO::AIO::poll_cb 1450 IO::AIO::poll_wait, IO::AIO::poll_cb
208 while IO::AIO::nreqs; 1451 while IO::AIO::nreqs;
209 1452
210 IO::AIO::flush 1453 IO::AIO::max_poll_reqs $nreqs
211 Wait till all outstanding AIO requests have been handled. 1454 IO::AIO::max_poll_time $seconds
1455 These set the maximum number of requests (default 0, meaning
1456 infinity) that are being processed by "IO::AIO::poll_cb" in one
1457 call, respectively the maximum amount of time (default 0, meaning
1458 infinity) spent in "IO::AIO::poll_cb" to process requests (more
1459 correctly the mininum amount of time "poll_cb" is allowed to use).
212 1460
213 Strictly equivalent to: 1461 Setting "max_poll_time" to a non-zero value creates an overhead of
1462 one syscall per request processed, which is not normally a problem
1463 unless your callbacks are really really fast or your OS is really
1464 really slow (I am not mentioning Solaris here). Using
1465 "max_poll_reqs" incurs no overhead.
214 1466
215 IO::AIO::poll_wait, IO::AIO::poll_cb 1467 Setting these is useful if you want to ensure some level of
216 while IO::AIO::nreqs; 1468 interactiveness when perl is not fast enough to process all requests
1469 in time.
217 1470
218 IO::AIO::poll 1471 For interactive programs, values such as 0.01 to 0.1 should be fine.
219 Waits until some requests have been handled.
220 1472
221 Strictly equivalent to: 1473 Example: Install an Event watcher that automatically calls
1474 IO::AIO::poll_cb with low priority, to ensure that other parts of
1475 the program get the CPU sometimes even under high AIO load.
222 1476
223 IO::AIO::poll_wait, IO::AIO::poll_cb 1477 # try not to spend much more than 0.1s in poll_cb
224 if IO::AIO::nreqs; 1478 IO::AIO::max_poll_time 0.1;
225 1479
1480 # use a low priority so other tasks have priority
1481 Event->io (fd => IO::AIO::poll_fileno,
1482 poll => 'r', nice => 1,
1483 cb => &IO::AIO::poll_cb);
1484
1485 CONTROLLING THE NUMBER OF THREADS
226 IO::AIO::min_parallel $nthreads 1486 IO::AIO::min_parallel $nthreads
227 Set the minimum number of AIO threads to $nthreads. The default is 1487 Set the minimum number of AIO threads to $nthreads. The current
228 1, which means a single asynchronous operation can be done at one 1488 default is 8, which means eight asynchronous operations can execute
229 time (the number of outstanding operations, however, is unlimited). 1489 concurrently at any one time (the number of outstanding requests,
1490 however, is unlimited).
230 1491
1492 IO::AIO starts threads only on demand, when an AIO request is queued
1493 and no free thread exists. Please note that queueing up a hundred
1494 requests can create demand for a hundred threads, even if it turns
1495 out that everything is in the cache and could have been processed
1496 faster by a single thread.
1497
231 It is recommended to keep the number of threads low, as some Linux 1498 It is recommended to keep the number of threads relatively low, as
232 kernel versions will scale negatively with the number of threads 1499 some Linux kernel versions will scale negatively with the number of
233 (higher parallelity => MUCH higher latency). With current Linux 2.6 1500 threads (higher parallelity => MUCH higher latency). With current
234 versions, 4-32 threads should be fine. 1501 Linux 2.6 versions, 4-32 threads should be fine.
235 1502
236 Under normal circumstances you don't need to call this function, as 1503 Under most circumstances you don't need to call this function, as
237 this module automatically starts some threads (the exact number 1504 the module selects a default that is suitable for low to moderate
238 might change, and is currently 4). 1505 load.
239 1506
240 IO::AIO::max_parallel $nthreads 1507 IO::AIO::max_parallel $nthreads
241 Sets the maximum number of AIO threads to $nthreads. If more than 1508 Sets the maximum number of AIO threads to $nthreads. If more than
242 the specified number of threads are currently running, kill them. 1509 the specified number of threads are currently running, this function
243 This function blocks until the limit is reached. 1510 kills them. This function blocks until the limit is reached.
1511
1512 While $nthreads are zero, aio requests get queued but not executed
1513 until the number of threads has been increased again.
244 1514
245 This module automatically runs "max_parallel 0" at program end, to 1515 This module automatically runs "max_parallel 0" at program end, to
246 ensure that all threads are killed and that there are no outstanding 1516 ensure that all threads are killed and that there are no outstanding
247 requests. 1517 requests.
248 1518
249 Under normal circumstances you don't need to call this function. 1519 Under normal circumstances you don't need to call this function.
250 1520
1521 IO::AIO::max_idle $nthreads
1522 Limit the number of threads (default: 4) that are allowed to idle
1523 (i.e., threads that did not get a request to process within the idle
1524 timeout (default: 10 seconds). That means if a thread becomes idle
1525 while $nthreads other threads are also idle, it will free its
1526 resources and exit.
1527
1528 This is useful when you allow a large number of threads (e.g. 100 or
1529 1000) to allow for extremely high load situations, but want to free
1530 resources under normal circumstances (1000 threads can easily
1531 consume 30MB of RAM).
1532
1533 The default is probably ok in most situations, especially if thread
1534 creation is fast. If thread creation is very slow on your system you
1535 might want to use larger values.
1536
1537 IO::AIO::idle_timeout $seconds
1538 Sets the minimum idle timeout (default 10) after which worker
1539 threads are allowed to exit. SEe "IO::AIO::max_idle".
1540
251 $oldnreqs = IO::AIO::max_outstanding $nreqs 1541 IO::AIO::max_outstanding $maxreqs
252 Sets the maximum number of outstanding requests to $nreqs. If you 1542 Sets the maximum number of outstanding requests to $nreqs. If you do
253 try to queue up more than this number of requests, the caller will 1543 queue up more than this number of requests, the next call to
1544 "IO::AIO::poll_cb" (and other functions calling "poll_cb", such as
1545 "IO::AIO::flush" or "IO::AIO::poll") will block until the limit is
1546 no longer exceeded.
1547
1548 In other words, this setting does not enforce a queue limit, but can
1549 be used to make poll functions block if the limit is exceeded.
1550
1551 This is a very bad function to use in interactive programs because
1552 it blocks, and a bad way to reduce concurrency because it is
1553 inexact: Better use an "aio_group" together with a feed callback.
1554
1555 Its main use is in scripts without an event loop - when you want to
1556 stat a lot of files, you can write something like this:
1557
1558 IO::AIO::max_outstanding 32;
1559
1560 for my $path (...) {
1561 aio_stat $path , ...;
1562 IO::AIO::poll_cb;
1563 }
1564
1565 IO::AIO::flush;
1566
1567 The call to "poll_cb" inside the loop will normally return
1568 instantly, but as soon as more thna 32 reqeusts are in-flight, it
254 block until some requests have been handled. 1569 will block until some requests have been handled. This keeps the
1570 loop from pushing a large number of "aio_stat" requests onto the
1571 queue.
255 1572
256 The default is very large, so normally there is no practical limit. 1573 The default value for "max_outstanding" is very large, so there is
257 If you queue up many requests in a loop it it often improves speed 1574 no practical limit on the number of outstanding requests.
258 if you set this to a relatively low number, such as 100.
259 1575
260 Under normal circumstances you don't need to call this function. 1576 STATISTICAL INFORMATION
1577 IO::AIO::nreqs
1578 Returns the number of requests currently in the ready, execute or
1579 pending states (i.e. for which their callback has not been invoked
1580 yet).
1581
1582 Example: wait till there are no outstanding requests anymore:
1583
1584 IO::AIO::poll_wait, IO::AIO::poll_cb
1585 while IO::AIO::nreqs;
1586
1587 IO::AIO::nready
1588 Returns the number of requests currently in the ready state (not yet
1589 executed).
1590
1591 IO::AIO::npending
1592 Returns the number of requests currently in the pending state
1593 (executed, but not yet processed by poll_cb).
1594
1595 MISCELLANEOUS FUNCTIONS
1596 IO::AIO implements some functions that are useful when you want to use
1597 some "Advanced I/O" function not available to in Perl, without going the
1598 "Asynchronous I/O" route. Many of these have an asynchronous "aio_*"
1599 counterpart.
1600
1601 $numfd = IO::AIO::get_fdlimit
1602 This function is *EXPERIMENTAL* and subject to change.
1603
1604 Tries to find the current file descriptor limit and returns it, or
1605 "undef" and sets $! in case of an error. The limit is one larger
1606 than the highest valid file descriptor number.
1607
1608 IO::AIO::min_fdlimit [$numfd]
1609 This function is *EXPERIMENTAL* and subject to change.
1610
1611 Try to increase the current file descriptor limit(s) to at least
1612 $numfd by changing the soft or hard file descriptor resource limit.
1613 If $numfd is missing, it will try to set a very high limit, although
1614 this is not recommended when you know the actual minimum that you
1615 require.
1616
1617 If the limit cannot be raised enough, the function makes a
1618 best-effort attempt to increase the limit as much as possible, using
1619 various tricks, while still failing. You can query the resulting
1620 limit using "IO::AIO::get_fdlimit".
1621
1622 If an error occurs, returns "undef" and sets $!, otherwise returns
1623 true.
1624
1625 IO::AIO::sendfile $ofh, $ifh, $offset, $count
1626 Calls the "eio_sendfile_sync" function, which is like
1627 "aio_sendfile", but is blocking (this makes most sense if you know
1628 the input data is likely cached already and the output filehandle is
1629 set to non-blocking operations).
1630
1631 Returns the number of bytes copied, or -1 on error.
1632
1633 IO::AIO::fadvise $fh, $offset, $len, $advice
1634 Simply calls the "posix_fadvise" function (see its manpage for
1635 details). The following advice constants are available:
1636 "IO::AIO::FADV_NORMAL", "IO::AIO::FADV_SEQUENTIAL",
1637 "IO::AIO::FADV_RANDOM", "IO::AIO::FADV_NOREUSE",
1638 "IO::AIO::FADV_WILLNEED", "IO::AIO::FADV_DONTNEED".
1639
1640 On systems that do not implement "posix_fadvise", this function
1641 returns ENOSYS, otherwise the return value of "posix_fadvise".
1642
1643 IO::AIO::madvise $scalar, $offset, $len, $advice
1644 Simply calls the "posix_madvise" function (see its manpage for
1645 details). The following advice constants are available:
1646 "IO::AIO::MADV_NORMAL", "IO::AIO::MADV_SEQUENTIAL",
1647 "IO::AIO::MADV_RANDOM", "IO::AIO::MADV_WILLNEED",
1648 "IO::AIO::MADV_DONTNEED".
1649
1650 If $offset is negative, counts from the end. If $length is negative,
1651 the remaining length of the $scalar is used. If possible, $length
1652 will be reduced to fit into the $scalar.
1653
1654 On systems that do not implement "posix_madvise", this function
1655 returns ENOSYS, otherwise the return value of "posix_madvise".
1656
1657 IO::AIO::mprotect $scalar, $offset, $len, $protect
1658 Simply calls the "mprotect" function on the preferably AIO::mmap'ed
1659 $scalar (see its manpage for details). The following protect
1660 constants are available: "IO::AIO::PROT_NONE", "IO::AIO::PROT_READ",
1661 "IO::AIO::PROT_WRITE", "IO::AIO::PROT_EXEC".
1662
1663 If $offset is negative, counts from the end. If $length is negative,
1664 the remaining length of the $scalar is used. If possible, $length
1665 will be reduced to fit into the $scalar.
1666
1667 On systems that do not implement "mprotect", this function returns
1668 ENOSYS, otherwise the return value of "mprotect".
1669
1670 IO::AIO::mmap $scalar, $length, $prot, $flags, $fh[, $offset]
1671 Memory-maps a file (or anonymous memory range) and attaches it to
1672 the given $scalar, which will act like a string scalar. Returns true
1673 on success, and false otherwise.
1674
1675 The scalar must exist, but its contents do not matter - this means
1676 you cannot use a nonexistant array or hash element. When in doubt,
1677 "undef" the scalar first.
1678
1679 The only operations allowed on the mmapped scalar are
1680 "substr"/"vec", which don't change the string length, and most
1681 read-only operations such as copying it or searching it with regexes
1682 and so on.
1683
1684 Anything else is unsafe and will, at best, result in memory leaks.
1685
1686 The memory map associated with the $scalar is automatically removed
1687 when the $scalar is undef'd or destroyed, or when the
1688 "IO::AIO::mmap" or "IO::AIO::munmap" functions are called on it.
1689
1690 This calls the "mmap"(2) function internally. See your system's
1691 manual page for details on the $length, $prot and $flags parameters.
1692
1693 The $length must be larger than zero and smaller than the actual
1694 filesize.
1695
1696 $prot is a combination of "IO::AIO::PROT_NONE",
1697 "IO::AIO::PROT_EXEC", "IO::AIO::PROT_READ" and/or
1698 "IO::AIO::PROT_WRITE",
1699
1700 $flags can be a combination of "IO::AIO::MAP_SHARED" or
1701 "IO::AIO::MAP_PRIVATE", or a number of system-specific flags (when
1702 not available, the are 0): "IO::AIO::MAP_ANONYMOUS" (which is set to
1703 "MAP_ANON" if your system only provides this constant),
1704 "IO::AIO::MAP_LOCKED", "IO::AIO::MAP_NORESERVE",
1705 "IO::AIO::MAP_POPULATE", "IO::AIO::MAP_NONBLOCK",
1706 "IO::AIO::MAP_FIXED", "IO::AIO::MAP_GROWSDOWN",
1707 "IO::AIO::MAP_32BIT", "IO::AIO::MAP_HUGETLB" or
1708 "IO::AIO::MAP_STACK".
1709
1710 If $fh is "undef", then a file descriptor of -1 is passed.
1711
1712 $offset is the offset from the start of the file - it generally must
1713 be a multiple of "IO::AIO::PAGESIZE" and defaults to 0.
1714
1715 Example:
1716
1717 use Digest::MD5;
1718 use IO::AIO;
1719
1720 open my $fh, "<verybigfile"
1721 or die "$!";
1722
1723 IO::AIO::mmap my $data, -s $fh, IO::AIO::PROT_READ, IO::AIO::MAP_SHARED, $fh
1724 or die "verybigfile: $!";
1725
1726 my $fast_md5 = md5 $data;
1727
1728 IO::AIO::munmap $scalar
1729 Removes a previous mmap and undefines the $scalar.
1730
1731 IO::AIO::mremap $scalar, $new_length, $flags = MREMAP_MAYMOVE[,
1732 $new_address = 0]
1733 Calls the Linux-specific mremap(2) system call. The $scalar must
1734 have been mapped by "IO::AIO::mmap", and $flags must currently
1735 either be 0 or "IO::AIO::MREMAP_MAYMOVE".
1736
1737 Returns true if successful, and false otherwise. If the underlying
1738 mmapped region has changed address, then the true value has the
1739 numerical value 1, otherwise it has the numerical value 0:
1740
1741 my $success = IO::AIO::mremap $mmapped, 8192, IO::AIO::MREMAP_MAYMOVE
1742 or die "mremap: $!";
1743
1744 if ($success*1) {
1745 warn "scalar has chanegd address in memory\n";
1746 }
1747
1748 "IO::AIO::MREMAP_FIXED" and the $new_address argument are currently
1749 implemented, but not supported and might go away in a future
1750 version.
1751
1752 On systems where this call is not supported or is not emulated, this
1753 call returns falls and sets $! to "ENOSYS".
1754
1755 IO::AIO::munlock $scalar, $offset = 0, $length = undef
1756 Calls the "munlock" function, undoing the effects of a previous
1757 "aio_mlock" call (see its description for details).
1758
1759 IO::AIO::munlockall
1760 Calls the "munlockall" function.
1761
1762 On systems that do not implement "munlockall", this function returns
1763 ENOSYS, otherwise the return value of "munlockall".
1764
1765 IO::AIO::splice $r_fh, $r_off, $w_fh, $w_off, $length, $flags
1766 Calls the GNU/Linux splice(2) syscall, if available. If $r_off or
1767 $w_off are "undef", then "NULL" is passed for these, otherwise they
1768 should be the file offset.
1769
1770 $r_fh and $w_fh should not refer to the same file, as splice might
1771 silently corrupt the data in this case.
1772
1773 The following symbol flag values are available:
1774 "IO::AIO::SPLICE_F_MOVE", "IO::AIO::SPLICE_F_NONBLOCK",
1775 "IO::AIO::SPLICE_F_MORE" and "IO::AIO::SPLICE_F_GIFT".
1776
1777 See the splice(2) manpage for details.
1778
1779 IO::AIO::tee $r_fh, $w_fh, $length, $flags
1780 Calls the GNU/Linux tee(2) syscall, see its manpage and the
1781 description for "IO::AIO::splice" above for details.
1782
1783 $actual_size = IO::AIO::pipesize $r_fh[, $new_size]
1784 Attempts to query or change the pipe buffer size. Obviously works
1785 only on pipes, and currently works only on GNU/Linux systems, and
1786 fails with -1/"ENOSYS" everywhere else. If anybody knows how to
1787 influence pipe buffer size on other systems, drop me a note.
1788
1789 ($rfh, $wfh) = IO::AIO::pipe2 [$flags]
1790 This is a direct interface to the Linux pipe2(2) system call. If
1791 $flags is missing or 0, then this should be the same as a call to
1792 perl's built-in "pipe" function and create a new pipe, and works on
1793 systems that lack the pipe2 syscall. On win32, this case invokes
1794 "_pipe (..., 4096, O_BINARY)".
1795
1796 If $flags is non-zero, it tries to invoke the pipe2 system call with
1797 the given flags (Linux 2.6.27, glibc 2.9).
1798
1799 On success, the read and write file handles are returned.
1800
1801 On error, nothing will be returned. If the pipe2 syscall is missing
1802 and $flags is non-zero, fails with "ENOSYS".
1803
1804 Please refer to pipe2(2) for more info on the $flags, but at the
1805 time of this writing, "IO::AIO::O_CLOEXEC", "IO::AIO::O_NONBLOCK"
1806 and "IO::AIO::O_DIRECT" (Linux 3.4, for packet-based pipes) were
1807 supported.
1808
1809 Example: create a pipe race-free w.r.t. threads and fork:
1810
1811 my ($rfh, $wfh) = IO::AIO::pipe2 IO::AIO::O_CLOEXEC
1812 or die "pipe2: $!\n";
1813
1814 $fh = IO::AIO::eventfd [$initval, [$flags]]
1815 This is a direct interface to the Linux eventfd(2) system call. The
1816 (unhelpful) defaults for $initval and $flags are 0 for both.
1817
1818 On success, the new eventfd filehandle is returned, otherwise
1819 returns "undef". If the eventfd syscall is missing, fails with
1820 "ENOSYS".
1821
1822 Please refer to eventfd(2) for more info on this call.
1823
1824 The following symbol flag values are available:
1825 "IO::AIO::EFD_CLOEXEC", "IO::AIO::EFD_NONBLOCK" and
1826 "IO::AIO::EFD_SEMAPHORE" (Linux 2.6.30).
1827
1828 Example: create a new eventfd filehandle:
1829
1830 $fh = IO::AIO::eventfd 0, IO::AIO::O_CLOEXEC
1831 or die "eventfd: $!\n";
1832
1833 $fh = IO::AIO::timerfd_create $clockid[, $flags]
1834 This is a direct interface to the Linux timerfd_create(2) system
1835 call. The (unhelpful) default for $flags is 0.
1836
1837 On success, the new timerfd filehandle is returned, otherwise
1838 returns "undef". If the eventfd syscall is missing, fails with
1839 "ENOSYS".
1840
1841 Please refer to timerfd_create(2) for more info on this call.
1842
1843 The following $clockid values are available:
1844 "IO::AIO::CLOCK_REALTIME", "IO::AIO::CLOCK_MONOTONIC"
1845 "IO::AIO::CLOCK_CLOCK_BOOTTIME" (Linux 3.15)
1846 "IO::AIO::CLOCK_CLOCK_REALTIME_ALARM" (Linux 3.11) and
1847 "IO::AIO::CLOCK_CLOCK_BOOTTIME_ALARM" (Linux 3.11).
1848
1849 The following $flags values are available (Linux 2.6.27):
1850 "IO::AIO::TFD_NONBLOCK" and "IO::AIO::TFD_CLOEXEC".
1851
1852 Example: create a new timerfd and set it to one-second repeated
1853 alarms, then wait for two alarms:
1854
1855 my $fh = IO::AIO::timerfd_create IO::AIO::CLOCK_BOOTTIME, IO::AIO::TFD_CLOEXEC
1856 or die "timerfd_create: $!\n";
1857
1858 defined IO::AIO::timerfd_settime $fh, 0, 1, 1
1859 or die "timerfd_settime: $!\n";
1860
1861 for (1..2) {
1862 8 == sysread $fh, my $buf, 8
1863 or die "timerfd read failure\n";
1864
1865 printf "number of expirations (likely 1): %d\n",
1866 unpack "Q", $buf;
1867 }
1868
1869 ($cur_interval, $cur_value) = IO::AIO::timerfd_settime $fh, $flags,
1870 $new_interval, $nbw_value
1871 This is a direct interface to the Linux timerfd_settime(2) system
1872 call. Please refer to its manpage for more info on this call.
1873
1874 The new itimerspec is specified using two (possibly fractional)
1875 second values, $new_interval and $new_value).
1876
1877 On success, the current interval and value are returned (as per
1878 "timerfd_gettime"). On failure, the empty list is returned.
1879
1880 The following $flags values are available:
1881 "IO::AIO::TFD_TIMER_ABSTIME" and "IO::AIO::TFD_TIMER_CANCEL_ON_SET".
1882
1883 See "IO::AIO::timerfd_create" for a full example.
1884
1885 ($cur_interval, $cur_value) = IO::AIO::timerfd_gettime $fh
1886 This is a direct interface to the Linux timerfd_gettime(2) system
1887 call. Please refer to its manpage for more info on this call.
1888
1889 On success, returns the current values of interval and value for the
1890 given timerfd (as potentially fractional second values). On failure,
1891 the empty list is returned.
1892
1893EVENT LOOP INTEGRATION
1894 It is recommended to use AnyEvent::AIO to integrate IO::AIO
1895 automatically into many event loops:
1896
1897 # AnyEvent integration (EV, Event, Glib, Tk, POE, urxvt, pureperl...)
1898 use AnyEvent::AIO;
1899
1900 You can also integrate IO::AIO manually into many event loops, here are
1901 some examples of how to do this:
1902
1903 # EV integration
1904 my $aio_w = EV::io IO::AIO::poll_fileno, EV::READ, \&IO::AIO::poll_cb;
1905
1906 # Event integration
1907 Event->io (fd => IO::AIO::poll_fileno,
1908 poll => 'r',
1909 cb => \&IO::AIO::poll_cb);
1910
1911 # Glib/Gtk2 integration
1912 add_watch Glib::IO IO::AIO::poll_fileno,
1913 in => sub { IO::AIO::poll_cb; 1 };
1914
1915 # Tk integration
1916 Tk::Event::IO->fileevent (IO::AIO::poll_fileno, "",
1917 readable => \&IO::AIO::poll_cb);
1918
1919 # Danga::Socket integration
1920 Danga::Socket->AddOtherFds (IO::AIO::poll_fileno =>
1921 \&IO::AIO::poll_cb);
1922
1923 FORK BEHAVIOUR
1924 Usage of pthreads in a program changes the semantics of fork
1925 considerably. Specifically, only async-safe functions can be called
1926 after fork. Perl doesn't know about this, so in general, you cannot call
1927 fork with defined behaviour in perl if pthreads are involved. IO::AIO
1928 uses pthreads, so this applies, but many other extensions and (for
1929 inexplicable reasons) perl itself often is linked against pthreads, so
1930 this limitation applies to quite a lot of perls.
1931
1932 This module no longer tries to fight your OS, or POSIX. That means
1933 IO::AIO only works in the process that loaded it. Forking is fully
1934 supported, but using IO::AIO in the child is not.
1935
1936 You might get around by not *using* IO::AIO before (or after) forking.
1937 You could also try to call the IO::AIO::reinit function in the child:
1938
1939 IO::AIO::reinit
1940 Abandons all current requests and I/O threads and simply
1941 reinitialises all data structures. This is not an operation
1942 supported by any standards, but happens to work on GNU/Linux and
1943 some newer BSD systems.
1944
1945 The only reasonable use for this function is to call it after
1946 forking, if "IO::AIO" was used in the parent. Calling it while
1947 IO::AIO is active in the process will result in undefined behaviour.
1948 Calling it at any time will also result in any undefined (by POSIX)
1949 behaviour.
1950
1951 LINUX-SPECIFIC CALLS
1952 When a call is documented as "linux-specific" then this means it
1953 originated on GNU/Linux. "IO::AIO" will usually try to autodetect the
1954 availability and compatibility of such calls regardless of the platform
1955 it is compiled on, so platforms such as FreeBSD which often implement
1956 these calls will work. When in doubt, call them and see if they fail wth
1957 "ENOSYS".
1958
1959 MEMORY USAGE
1960 Per-request usage:
1961
1962 Each aio request uses - depending on your architecture - around 100-200
1963 bytes of memory. In addition, stat requests need a stat buffer (possibly
1964 a few hundred bytes), readdir requires a result buffer and so on. Perl
1965 scalars and other data passed into aio requests will also be locked and
1966 will consume memory till the request has entered the done state.
1967
1968 This is not awfully much, so queuing lots of requests is not usually a
1969 problem.
1970
1971 Per-thread usage:
1972
1973 In the execution phase, some aio requests require more memory for
1974 temporary buffers, and each thread requires a stack and other data
1975 structures (usually around 16k-128k, depending on the OS).
1976
1977KNOWN BUGS
1978 Known bugs will be fixed in the next release :)
1979
1980KNOWN ISSUES
1981 Calls that try to "import" foreign memory areas (such as "IO::AIO::mmap"
1982 or "IO::AIO::aio_slurp") do not work with generic lvalues, such as
1983 non-created hash slots or other scalars I didn't think of. It's best to
1984 avoid such and either use scalar variables or making sure that the
1985 scalar exists (e.g. by storing "undef") and isn't "funny" (e.g. tied).
1986
1987 I am not sure anything can be done about this, so this is considered a
1988 known issue, rather than a bug.
261 1989
262SEE ALSO 1990SEE ALSO
263 Coro, Linux::AIO. 1991 AnyEvent::AIO for easy integration into event loops, Coro::AIO for a
1992 more natural syntax.
264 1993
265AUTHOR 1994AUTHOR
266 Marc Lehmann <schmorp@schmorp.de> 1995 Marc Lehmann <schmorp@schmorp.de>
267 http://home.schmorp.de/ 1996 http://home.schmorp.de/
268 1997

Diff Legend

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