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 1.48 by root, Wed Jun 29 11:25:17 2011 UTC

… …
2 IO::AIO - Asynchronous Input/Output 2 IO::AIO - Asynchronous 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
49 Although the module will work with in the presence of other threads, it 60 Although the module will work in the presence of other (Perl-) threads,
50 is currently not reentrant, so use appropriate locking yourself, always 61 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 62 yourself, always call "poll_cb" from within the same thread, or never
52 other "aio_" functions) recursively. 63 call "poll_cb" (or other "aio_" functions) recursively.
64
65 EXAMPLE
66 This is a simple example that uses the EV module and loads /etc/passwd
67 asynchronously:
68
69 use Fcntl;
70 use EV;
71 use IO::AIO;
72
73 # register the IO::AIO callback with EV
74 my $aio_w = EV::io IO::AIO::poll_fileno, EV::READ, \&IO::AIO::poll_cb;
75
76 # queue the request to open /etc/passwd
77 aio_open "/etc/passwd", IO::AIO::O_RDONLY, 0, sub {
78 my $fh = shift
79 or die "error while opening: $!";
80
81 # stat'ing filehandles is generally non-blocking
82 my $size = -s $fh;
83
84 # queue a request to read the file
85 my $contents;
86 aio_read $fh, 0, $size, $contents, 0, sub {
87 $_[0] == $size
88 or die "short read: $!";
89
90 close $fh;
91
92 # file contents now in $contents
93 print $contents;
94
95 # exit event loop and program
96 EV::unloop;
97 };
98 };
99
100 # possibly queue up other requests, or open GUI windows,
101 # check for sockets etc. etc.
102
103 # process events as long as there are some:
104 EV::loop;
105
106REQUEST ANATOMY AND LIFETIME
107 Every "aio_*" function creates a request. which is a C data structure
108 not directly visible to Perl.
109
110 If called in non-void context, every request function returns a Perl
111 object representing the request. In void context, nothing is returned,
112 which saves a bit of memory.
113
114 The perl object is a fairly standard ref-to-hash object. The hash
115 contents are not used by IO::AIO so you are free to store anything you
116 like in it.
117
118 During their existance, aio requests travel through the following
119 states, in order:
120
121 ready
122 Immediately after a request is created it is put into the ready
123 state, waiting for a thread to execute it.
124
125 execute
126 A thread has accepted the request for processing and is currently
127 executing it (e.g. blocking in read).
128
129 pending
130 The request has been executed and is waiting for result processing.
131
132 While request submission and execution is fully asynchronous, result
133 processing is not and relies on the perl interpreter calling
134 "poll_cb" (or another function with the same effect).
135
136 result
137 The request results are processed synchronously by "poll_cb".
138
139 The "poll_cb" function will process all outstanding aio requests by
140 calling their callbacks, freeing memory associated with them and
141 managing any groups they are contained in.
142
143 done
144 Request has reached the end of its lifetime and holds no resources
145 anymore (except possibly for the Perl object, but its connection to
146 the actual aio request is severed and calling its methods will
147 either do nothing or result in a runtime error).
53 148
54FUNCTIONS 149FUNCTIONS
150 QUICK OVERVIEW
151 This section simply lists the prototypes of the most important functions
152 for quick reference. See the following sections for function-by-function
153 documentation.
154
155 aio_open $pathname, $flags, $mode, $callback->($fh)
156 aio_close $fh, $callback->($status)
157 aio_read $fh,$offset,$length, $data,$dataoffset, $callback->($retval)
158 aio_write $fh,$offset,$length, $data,$dataoffset, $callback->($retval)
159 aio_sendfile $out_fh, $in_fh, $in_offset, $length, $callback->($retval)
160 aio_readahead $fh,$offset,$length, $callback->($retval)
161 aio_stat $fh_or_path, $callback->($status)
162 aio_lstat $fh, $callback->($status)
163 aio_statvfs $fh_or_path, $callback->($statvfs)
164 aio_utime $fh_or_path, $atime, $mtime, $callback->($status)
165 aio_chown $fh_or_path, $uid, $gid, $callback->($status)
166 aio_truncate $fh_or_path, $offset, $callback->($status)
167 aio_chmod $fh_or_path, $mode, $callback->($status)
168 aio_unlink $pathname, $callback->($status)
169 aio_mknod $path, $mode, $dev, $callback->($status)
170 aio_link $srcpath, $dstpath, $callback->($status)
171 aio_symlink $srcpath, $dstpath, $callback->($status)
172 aio_readlink $path, $callback->($link)
173 aio_rename $srcpath, $dstpath, $callback->($status)
174 aio_mkdir $pathname, $mode, $callback->($status)
175 aio_rmdir $pathname, $callback->($status)
176 aio_readdir $pathname, $callback->($entries)
177 aio_readdirx $pathname, $flags, $callback->($entries, $flags)
178 IO::AIO::READDIR_DENTS IO::AIO::READDIR_DIRS_FIRST
179 IO::AIO::READDIR_STAT_ORDER IO::AIO::READDIR_FOUND_UNKNOWN
180 aio_load $path, $data, $callback->($status)
181 aio_copy $srcpath, $dstpath, $callback->($status)
182 aio_move $srcpath, $dstpath, $callback->($status)
183 aio_scandir $path, $maxreq, $callback->($dirs, $nondirs)
184 aio_rmtree $path, $callback->($status)
185 aio_sync $callback->($status)
186 aio_fsync $fh, $callback->($status)
187 aio_fdatasync $fh, $callback->($status)
188 aio_sync_file_range $fh, $offset, $nbytes, $flags, $callback->($status)
189 aio_pathsync $path, $callback->($status)
190 aio_msync $scalar, $offset = 0, $length = undef, flags = 0, $callback->($status)
191 aio_mtouch $scalar, $offset = 0, $length = undef, flags = 0, $callback->($status)
192 aio_mlock $scalar, $offset = 0, $length = undef, $callback->($status)
193 aio_mlockall $flags, $callback->($status)
194 aio_group $callback->(...)
195 aio_nop $callback->()
196
197 $prev_pri = aioreq_pri [$pri]
198 aioreq_nice $pri_adjust
199
200 IO::AIO::poll_wait
201 IO::AIO::poll_cb
202 IO::AIO::poll
203 IO::AIO::flush
204 IO::AIO::max_poll_reqs $nreqs
205 IO::AIO::max_poll_time $seconds
206 IO::AIO::min_parallel $nthreads
207 IO::AIO::max_parallel $nthreads
208 IO::AIO::max_idle $nthreads
209 IO::AIO::idle_timeout $seconds
210 IO::AIO::max_outstanding $maxreqs
211 IO::AIO::nreqs
212 IO::AIO::nready
213 IO::AIO::npending
214
215 IO::AIO::sendfile $ofh, $ifh, $offset, $count
216 IO::AIO::fadvise $fh, $offset, $len, $advice
217 IO::AIO::madvise $scalar, $offset, $length, $advice
218 IO::AIO::mprotect $scalar, $offset, $length, $protect
219 IO::AIO::munlock $scalar, $offset = 0, $length = undef
220 IO::AIO::munlockall
221
55 AIO FUNCTIONS 222 AIO REQUEST FUNCTIONS
56 All the "aio_*" calls are more or less thin wrappers around the syscall 223 All the "aio_*" calls are more or less thin wrappers around the syscall
57 with the same name (sans "aio_"). The arguments are similar or 224 with the same name (sans "aio_"). The arguments are similar or
58 identical, and they all accept an additional (and optional) $callback 225 identical, and they all accept an additional (and optional) $callback
59 argument which must be a code reference. This code reference will get 226 argument which must be a code reference. This code reference will get
60 called with the syscall return code (e.g. most syscalls return -1 on 227 called with the syscall return code (e.g. most syscalls return -1 on
61 error, unlike perl, which usually delivers "false") as it's sole 228 error, unlike perl, which usually delivers "false") as its sole argument
62 argument when the given syscall has been executed asynchronously. 229 after the given syscall has been executed asynchronously.
63 230
64 All functions expecting a filehandle keep a copy of the filehandle 231 All functions expecting a filehandle keep a copy of the filehandle
65 internally until the request has finished. 232 internally until the request has finished.
66 233
234 All functions return request objects of type IO::AIO::REQ that allow
235 further manipulation of those requests while they are in-flight.
236
67 The filenames you pass to these routines *must* be absolute. The reason 237 The pathnames you pass to these routines *must* be absolute and encoded
68 for this is that at the time the request is being executed, the current 238 as octets. The reason for the former is that at the time the request is
69 working directory could have changed. Alternatively, you can make sure 239 being executed, the current working directory could have changed.
70 that you never change the current working directory. 240 Alternatively, you can make sure that you never change the current
241 working directory anywhere in the program and then use relative paths.
71 242
243 To encode pathnames as octets, either make sure you either: a) always
244 pass in filenames you got from outside (command line, readdir etc.)
245 without tinkering, b) are ASCII or ISO 8859-1, c) use the Encode module
246 and encode your pathnames to the locale (or other) encoding in effect in
247 the user environment, d) use Glib::filename_from_unicode on unicode
248 filenames or e) use something else to ensure your scalar has the correct
249 contents.
250
251 This works, btw. independent of the internal UTF-8 bit, which IO::AIO
252 handles correctly whether it is set or not.
253
254 $prev_pri = aioreq_pri [$pri]
255 Returns the priority value that would be used for the next request
256 and, if $pri is given, sets the priority for the next aio request.
257
258 The default priority is 0, the minimum and maximum priorities are -4
259 and 4, respectively. Requests with higher priority will be serviced
260 first.
261
262 The priority will be reset to 0 after each call to one of the
263 "aio_*" functions.
264
265 Example: open a file with low priority, then read something from it
266 with higher priority so the read request is serviced before other
267 low priority open requests (potentially spamming the cache):
268
269 aioreq_pri -3;
270 aio_open ..., sub {
271 return unless $_[0];
272
273 aioreq_pri -2;
274 aio_read $_[0], ..., sub {
275 ...
276 };
277 };
278
279 aioreq_nice $pri_adjust
280 Similar to "aioreq_pri", but subtracts the given value from the
281 current priority, so the effect is cumulative.
282
72 aio_open $pathname, $flags, $mode, $callback 283 aio_open $pathname, $flags, $mode, $callback->($fh)
73 Asynchronously open or create a file and call the callback with a 284 Asynchronously open or create a file and call the callback with a
74 newly created filehandle for the file. 285 newly created filehandle for the file.
75 286
76 The pathname passed to "aio_open" must be absolute. See API NOTES, 287 The pathname passed to "aio_open" must be absolute. See API NOTES,
77 above, for an explanation. 288 above, for an explanation.
… …
80 They are the same as used by "sysopen". 291 They are the same as used by "sysopen".
81 292
82 Likewise, $mode specifies the mode of the newly created file, if it 293 Likewise, $mode specifies the mode of the newly created file, if it
83 didn't exist and "O_CREAT" has been given, just like perl's 294 didn't exist and "O_CREAT" has been given, just like perl's
84 "sysopen", except that it is mandatory (i.e. use 0 if you don't 295 "sysopen", except that it is mandatory (i.e. use 0 if you don't
85 create new files, and 0666 or 0777 if you do). 296 create new files, and 0666 or 0777 if you do). Note that the $mode
297 will be modified by the umask in effect then the request is being
298 executed, so better never change the umask.
86 299
87 Example: 300 Example:
88 301
89 aio_open "/etc/passwd", O_RDONLY, 0, sub { 302 aio_open "/etc/passwd", IO::AIO::O_RDONLY, 0, sub {
90 if ($_[0]) { 303 if ($_[0]) {
91 print "open successful, fh is $_[0]\n"; 304 print "open successful, fh is $_[0]\n";
92 ... 305 ...
93 } else { 306 } else {
94 die "open failed: $!\n"; 307 die "open failed: $!\n";
95 } 308 }
96 }; 309 };
97 310
311 In addition to all the common open modes/flags ("O_RDONLY",
312 "O_WRONLY", "O_RDWR", "O_CREAT", "O_TRUNC", "O_EXCL" and
313 "O_APPEND"), the following POSIX and non-POSIX constants are
314 available (missing ones on your system are, as usual, 0):
315
316 "O_ASYNC", "O_DIRECT", "O_NOATIME", "O_CLOEXEC", "O_NOCTTY",
317 "O_NOFOLLOW", "O_NONBLOCK", "O_EXEC", "O_SEARCH", "O_DIRECTORY",
318 "O_DSYNC", "O_RSYNC", "O_SYNC" and "O_TTY_INIT".
319
98 aio_close $fh, $callback 320 aio_close $fh, $callback->($status)
99 Asynchronously close a file and call the callback with the result 321 Asynchronously close a file and call the callback with the result
100 code. *WARNING:* although accepted, you should not pass in a perl 322 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.
104 323
105 This is supposed to be a bug in the API, so that might change. It's 324 Unfortunately, you can't do this to perl. Perl *insists* very
106 therefore best to avoid this function. 325 strongly on closing the file descriptor associated with the
326 filehandle itself.
107 327
328 Therefore, "aio_close" will not close the filehandle - instead it
329 will use dup2 to overwrite the file descriptor with the write-end of
330 a pipe (the pipe fd will be created on demand and will be cached).
331
332 Or in other words: the file descriptor will be closed, but it will
333 not be free for reuse until the perl filehandle is closed.
334
108 aio_read $fh,$offset,$length, $data,$dataoffset,$callback 335 aio_read $fh,$offset,$length, $data,$dataoffset, $callback->($retval)
109 aio_write $fh,$offset,$length, $data,$dataoffset,$callback 336 aio_write $fh,$offset,$length, $data,$dataoffset, $callback->($retval)
110 Reads or writes "length" bytes from the specified "fh" and "offset" 337 Reads or writes $length bytes from or to the specified $fh and
111 into the scalar given by "data" and offset "dataoffset" and calls 338 $offset into the scalar given by $data and offset $dataoffset and
112 the callback without the actual number of bytes read (or -1 on 339 calls the callback without the actual number of bytes read (or -1 on
113 error, just like the syscall). 340 error, just like the syscall).
341
342 "aio_read" will, like "sysread", shrink or grow the $data scalar to
343 offset plus the actual number of bytes read.
344
345 If $offset is undefined, then the current file descriptor offset
346 will be used (and updated), otherwise the file descriptor offset
347 will not be changed by these calls.
348
349 If $length is undefined in "aio_write", use the remaining length of
350 $data.
351
352 If $dataoffset is less than zero, it will be counted from the end of
353 $data.
354
355 The $data scalar *MUST NOT* be modified in any way while the request
356 is outstanding. Modifying it can result in segfaults or World War
357 III (if the necessary/optional hardware is installed).
114 358
115 Example: Read 15 bytes at offset 7 into scalar $buffer, starting at 359 Example: Read 15 bytes at offset 7 into scalar $buffer, starting at
116 offset 0 within the scalar: 360 offset 0 within the scalar:
117 361
118 aio_read $fh, 7, 15, $buffer, 0, sub { 362 aio_read $fh, 7, 15, $buffer, 0, sub {
119 $_[0] > 0 or die "read error: $!"; 363 $_[0] > 0 or die "read error: $!";
120 print "read $_[0] bytes: <$buffer>\n"; 364 print "read $_[0] bytes: <$buffer>\n";
121 }; 365 };
122 366
367 aio_sendfile $out_fh, $in_fh, $in_offset, $length, $callback->($retval)
368 Tries to copy $length bytes from $in_fh to $out_fh. It starts
369 reading at byte offset $in_offset, and starts writing at the current
370 file offset of $out_fh. Because of that, it is not safe to issue
371 more than one "aio_sendfile" per $out_fh, as they will interfere
372 with each other. The same $in_fh works fine though, as this function
373 does not move or use the file offset of $in_fh.
374
375 Please note that "aio_sendfile" can read more bytes from $in_fh than
376 are written, and there is no way to find out how many more bytes
377 have been read from "aio_sendfile" alone, as "aio_sendfile" only
378 provides the number of bytes written to $out_fh. Only if the result
379 value equals $length one can assume that $length bytes have been
380 read.
381
382 Unlike with other "aio_" functions, it makes a lot of sense to use
383 "aio_sendfile" on non-blocking sockets, as long as one end
384 (typically the $in_fh) is a file - the file I/O will then be
385 asynchronous, while the socket I/O will be non-blocking. Note,
386 however, that you can run into a trap where "aio_sendfile" reads
387 some data with readahead, then fails to write all data, and when the
388 socket is ready the next time, the data in the cache is already
389 lost, forcing "aio_sendfile" to again hit the disk. Explicit
390 "aio_read" + "aio_write" let's you better control resource usage.
391
392 This call tries to make use of a native "sendfile"-like syscall to
393 provide zero-copy operation. For this to work, $out_fh should refer
394 to a socket, and $in_fh should refer to an mmap'able file.
395
396 If a native sendfile cannot be found or it fails with "ENOSYS",
397 "EINVAL", "ENOTSUP", "EOPNOTSUPP", "EAFNOSUPPORT", "EPROTOTYPE" or
398 "ENOTSOCK", it will be emulated, so you can call "aio_sendfile" on
399 any type of filehandle regardless of the limitations of the
400 operating system.
401
402 As native sendfile syscalls (as practically any non-POSIX interface
403 hacked together in a hurry to improve benchmark numbers) tend to be
404 rather buggy on many systems, this implementation tries to work
405 around some known bugs in Linux and FreeBSD kernels (probably
406 others, too), but that might fail, so you really really should check
407 the return value of "aio_sendfile" - fewre bytes than expected might
408 have been transferred.
409
123 aio_readahead $fh,$offset,$length, $callback 410 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 411 "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 412 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 413 $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 414 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 415 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 416 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 417 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 418 read beyond the end of the file. The current file offset of the file
137 is left unchanged. 419 is left unchanged.
138 420
421 If that syscall doesn't exist (likely if your OS isn't Linux) it
422 will be emulated by simply reading the data, which would have a
423 similar effect.
424
139 aio_stat $fh_or_path, $callback 425 aio_stat $fh_or_path, $callback->($status)
140 aio_lstat $fh, $callback 426 aio_lstat $fh, $callback->($status)
141 Works like perl's "stat" or "lstat" in void context. The callback 427 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 428 will be called after the stat and the results will be available
143 using "stat _" or "-s _" etc... 429 using "stat _" or "-s _" etc...
144 430
145 The pathname passed to "aio_stat" must be absolute. See API NOTES, 431 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 434 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 435 returning an error when stat'ing a large file, the results will be
150 silently truncated unless perl itself is compiled with large file 436 silently truncated unless perl itself is compiled with large file
151 support. 437 support.
152 438
439 To help interpret the mode and dev/rdev stat values, IO::AIO offers
440 the following constants and functions (if not implemented, the
441 constants will be 0 and the functions will either "croak" or fall
442 back on traditional behaviour).
443
444 "S_IFMT", "S_IFIFO", "S_IFCHR", "S_IFBLK", "S_IFLNK", "S_IFREG",
445 "S_IFDIR", "S_IFWHT", "S_IFSOCK", "IO::AIO::major $dev_t",
446 "IO::AIO::minor $dev_t", "IO::AIO::makedev $major, $minor".
447
153 Example: Print the length of /etc/passwd: 448 Example: Print the length of /etc/passwd:
154 449
155 aio_stat "/etc/passwd", sub { 450 aio_stat "/etc/passwd", sub {
156 $_[0] and die "stat failed: $!"; 451 $_[0] and die "stat failed: $!";
157 print "size is ", -s _, "\n"; 452 print "size is ", -s _, "\n";
158 }; 453 };
159 454
455 aio_statvfs $fh_or_path, $callback->($statvfs)
456 Works like the POSIX "statvfs" or "fstatvfs" syscalls, depending on
457 whether a file handle or path was passed.
458
459 On success, the callback is passed a hash reference with the
460 following members: "bsize", "frsize", "blocks", "bfree", "bavail",
461 "files", "ffree", "favail", "fsid", "flag" and "namemax". On
462 failure, "undef" is passed.
463
464 The following POSIX IO::AIO::ST_* constants are defined: "ST_RDONLY"
465 and "ST_NOSUID".
466
467 The following non-POSIX IO::AIO::ST_* flag masks are defined to
468 their correct value when available, or to 0 on systems that do not
469 support them: "ST_NODEV", "ST_NOEXEC", "ST_SYNCHRONOUS",
470 "ST_MANDLOCK", "ST_WRITE", "ST_APPEND", "ST_IMMUTABLE",
471 "ST_NOATIME", "ST_NODIRATIME" and "ST_RELATIME".
472
473 Example: stat "/wd" and dump out the data if successful.
474
475 aio_statvfs "/wd", sub {
476 my $f = $_[0]
477 or die "statvfs: $!";
478
479 use Data::Dumper;
480 say Dumper $f;
481 };
482
483 # result:
484 {
485 bsize => 1024,
486 bfree => 4333064312,
487 blocks => 10253828096,
488 files => 2050765568,
489 flag => 4096,
490 favail => 2042092649,
491 bavail => 4333064312,
492 ffree => 2042092649,
493 namemax => 255,
494 frsize => 1024,
495 fsid => 1810
496 }
497
498 aio_utime $fh_or_path, $atime, $mtime, $callback->($status)
499 Works like perl's "utime" function (including the special case of
500 $atime and $mtime being undef). Fractional times are supported if
501 the underlying syscalls support them.
502
503 When called with a pathname, uses utimes(2) if available, otherwise
504 utime(2). If called on a file descriptor, uses futimes(2) if
505 available, otherwise returns ENOSYS, so this is not portable.
506
507 Examples:
508
509 # set atime and mtime to current time (basically touch(1)):
510 aio_utime "path", undef, undef;
511 # set atime to current time and mtime to beginning of the epoch:
512 aio_utime "path", time, undef; # undef==0
513
514 aio_chown $fh_or_path, $uid, $gid, $callback->($status)
515 Works like perl's "chown" function, except that "undef" for either
516 $uid or $gid is being interpreted as "do not change" (but -1 can
517 also be used).
518
519 Examples:
520
521 # same as "chown root path" in the shell:
522 aio_chown "path", 0, -1;
523 # same as above:
524 aio_chown "path", 0, undef;
525
526 aio_truncate $fh_or_path, $offset, $callback->($status)
527 Works like truncate(2) or ftruncate(2).
528
529 aio_chmod $fh_or_path, $mode, $callback->($status)
530 Works like perl's "chmod" function.
531
160 aio_unlink $pathname, $callback 532 aio_unlink $pathname, $callback->($status)
161 Asynchronously unlink (delete) a file and call the callback with the 533 Asynchronously unlink (delete) a file and call the callback with the
162 result code. 534 result code.
163 535
536 aio_mknod $path, $mode, $dev, $callback->($status)
537 [EXPERIMENTAL]
538
539 Asynchronously create a device node (or fifo). See mknod(2).
540
541 The only (POSIX-) portable way of calling this function is:
542
543 aio_mknod $path, IO::AIO::S_IFIFO | $mode, 0, sub { ...
544
545 See "aio_stat" for info about some potentially helpful extra
546 constants and functions.
547
548 aio_link $srcpath, $dstpath, $callback->($status)
549 Asynchronously create a new link to the existing object at $srcpath
550 at the path $dstpath and call the callback with the result code.
551
552 aio_symlink $srcpath, $dstpath, $callback->($status)
553 Asynchronously create a new symbolic link to the existing object at
554 $srcpath at the path $dstpath and call the callback with the result
555 code.
556
557 aio_readlink $path, $callback->($link)
558 Asynchronously read the symlink specified by $path and pass it to
559 the callback. If an error occurs, nothing or undef gets passed to
560 the callback.
561
562 aio_rename $srcpath, $dstpath, $callback->($status)
563 Asynchronously rename the object at $srcpath to $dstpath, just as
564 rename(2) and call the callback with the result code.
565
566 aio_mkdir $pathname, $mode, $callback->($status)
567 Asynchronously mkdir (create) a directory and call the callback with
568 the result code. $mode will be modified by the umask at the time the
569 request is executed, so do not change your umask.
570
571 aio_rmdir $pathname, $callback->($status)
572 Asynchronously rmdir (delete) a directory and call the callback with
573 the result code.
574
575 aio_readdir $pathname, $callback->($entries)
576 Unlike the POSIX call of the same name, "aio_readdir" reads an
577 entire directory (i.e. opendir + readdir + closedir). The entries
578 will not be sorted, and will NOT include the "." and ".." entries.
579
580 The callback is passed a single argument which is either "undef" or
581 an array-ref with the filenames.
582
583 aio_readdirx $pathname, $flags, $callback->($entries, $flags)
584 Quite similar to "aio_readdir", but the $flags argument allows to
585 tune behaviour and output format. In case of an error, $entries will
586 be "undef".
587
588 The flags are a combination of the following constants, ORed
589 together (the flags will also be passed to the callback, possibly
590 modified):
591
592 IO::AIO::READDIR_DENTS
593 When this flag is off, then the callback gets an arrayref
594 consisting of names only (as with "aio_readdir"), otherwise it
595 gets an arrayref with "[$name, $type, $inode]" arrayrefs, each
596 describing a single directory entry in more detail.
597
598 $name is the name of the entry.
599
600 $type is one of the "IO::AIO::DT_xxx" constants:
601
602 "IO::AIO::DT_UNKNOWN", "IO::AIO::DT_FIFO", "IO::AIO::DT_CHR",
603 "IO::AIO::DT_DIR", "IO::AIO::DT_BLK", "IO::AIO::DT_REG",
604 "IO::AIO::DT_LNK", "IO::AIO::DT_SOCK", "IO::AIO::DT_WHT".
605
606 "IO::AIO::DT_UNKNOWN" means just that: readdir does not know. If
607 you need to know, you have to run stat yourself. Also, for speed
608 reasons, the $type scalars are read-only: you can not modify
609 them.
610
611 $inode is the inode number (which might not be exact on systems
612 with 64 bit inode numbers and 32 bit perls). This field has
613 unspecified content on systems that do not deliver the inode
614 information.
615
616 IO::AIO::READDIR_DIRS_FIRST
617 When this flag is set, then the names will be returned in an
618 order where likely directories come first, in optimal stat
619 order. This is useful when you need to quickly find directories,
620 or you want to find all directories while avoiding to stat()
621 each entry.
622
623 If the system returns type information in readdir, then this is
624 used to find directories directly. Otherwise, likely directories
625 are names beginning with ".", or otherwise names with no dots,
626 of which names with short names are tried first.
627
628 IO::AIO::READDIR_STAT_ORDER
629 When this flag is set, then the names will be returned in an
630 order suitable for stat()'ing each one. That is, when you plan
631 to stat() all files in the given directory, then the returned
632 order will likely be fastest.
633
634 If both this flag and "IO::AIO::READDIR_DIRS_FIRST" are
635 specified, then the likely dirs come first, resulting in a less
636 optimal stat order.
637
638 IO::AIO::READDIR_FOUND_UNKNOWN
639 This flag should not be set when calling "aio_readdirx".
640 Instead, it is being set by "aio_readdirx", when any of the
641 $type's found were "IO::AIO::DT_UNKNOWN". The absense of this
642 flag therefore indicates that all $type's are known, which can
643 be used to speed up some algorithms.
644
645 aio_load $path, $data, $callback->($status)
646 This is a composite request that tries to fully load the given file
647 into memory. Status is the same as with aio_read.
648
649 aio_copy $srcpath, $dstpath, $callback->($status)
650 Try to copy the *file* (directories not supported as either source
651 or destination) from $srcpath to $dstpath and call the callback with
652 a status of 0 (ok) or -1 (error, see $!).
653
654 This is a composite request that creates the destination file with
655 mode 0200 and copies the contents of the source file into it using
656 "aio_sendfile", followed by restoring atime, mtime, access mode and
657 uid/gid, in that order.
658
659 If an error occurs, the partial destination file will be unlinked,
660 if possible, except when setting atime, mtime, access mode and
661 uid/gid, where errors are being ignored.
662
663 aio_move $srcpath, $dstpath, $callback->($status)
664 Try to move the *file* (directories not supported as either source
665 or destination) from $srcpath to $dstpath and call the callback with
666 a status of 0 (ok) or -1 (error, see $!).
667
668 This is a composite request that tries to rename(2) the file first;
669 if rename fails with "EXDEV", it copies the file with "aio_copy"
670 and, if that is successful, unlinks the $srcpath.
671
672 aio_scandir $path, $maxreq, $callback->($dirs, $nondirs)
673 Scans a directory (similar to "aio_readdir") but additionally tries
674 to efficiently separate the entries of directory $path into two sets
675 of names, directories you can recurse into (directories), and ones
676 you cannot recurse into (everything else, including symlinks to
677 directories).
678
679 "aio_scandir" is a composite request that creates of many sub
680 requests_ $maxreq specifies the maximum number of outstanding aio
681 requests that this function generates. If it is "<= 0", then a
682 suitable default will be chosen (currently 4).
683
684 On error, the callback is called without arguments, otherwise it
685 receives two array-refs with path-relative entry names.
686
687 Example:
688
689 aio_scandir $dir, 0, sub {
690 my ($dirs, $nondirs) = @_;
691 print "real directories: @$dirs\n";
692 print "everything else: @$nondirs\n";
693 };
694
695 Implementation notes.
696
697 The "aio_readdir" cannot be avoided, but "stat()"'ing every entry
698 can.
699
700 If readdir returns file type information, then this is used directly
701 to find directories.
702
703 Otherwise, after reading the directory, the modification time, size
704 etc. of the directory before and after the readdir is checked, and
705 if they match (and isn't the current time), the link count will be
706 used to decide how many entries are directories (if >= 2).
707 Otherwise, no knowledge of the number of subdirectories will be
708 assumed.
709
710 Then entries will be sorted into likely directories a non-initial
711 dot currently) and likely non-directories (see "aio_readdirx"). Then
712 every entry plus an appended "/." will be "stat"'ed, likely
713 directories first, in order of their inode numbers. If that
714 succeeds, it assumes that the entry is a directory or a symlink to
715 directory (which will be checked seperately). This is often faster
716 than stat'ing the entry itself because filesystems might detect the
717 type of the entry without reading the inode data (e.g. ext2fs
718 filetype feature), even on systems that cannot return the filetype
719 information on readdir.
720
721 If the known number of directories (link count - 2) has been
722 reached, the rest of the entries is assumed to be non-directories.
723
724 This only works with certainty on POSIX (= UNIX) filesystems, which
725 fortunately are the vast majority of filesystems around.
726
727 It will also likely work on non-POSIX filesystems with reduced
728 efficiency as those tend to return 0 or 1 as link counts, which
729 disables the directory counting heuristic.
730
731 aio_rmtree $path, $callback->($status)
732 Delete a directory tree starting (and including) $path, return the
733 status of the final "rmdir" only. This is a composite request that
734 uses "aio_scandir" to recurse into and rmdir directories, and unlink
735 everything else.
736
737 aio_sync $callback->($status)
738 Asynchronously call sync and call the callback when finished.
739
164 aio_fsync $fh, $callback 740 aio_fsync $fh, $callback->($status)
165 Asynchronously call fsync on the given filehandle and call the 741 Asynchronously call fsync on the given filehandle and call the
166 callback with the fsync result code. 742 callback with the fsync result code.
167 743
168 aio_fdatasync $fh, $callback 744 aio_fdatasync $fh, $callback->($status)
169 Asynchronously call fdatasync on the given filehandle and call the 745 Asynchronously call fdatasync on the given filehandle and call the
170 callback with the fdatasync result code. Might set $! to "ENOSYS" if 746 callback with the fdatasync result code.
171 "fdatasync" is not available. 747
748 If this call isn't available because your OS lacks it or it couldn't
749 be detected, it will be emulated by calling "fsync" instead.
750
751 aio_sync_file_range $fh, $offset, $nbytes, $flags, $callback->($status)
752 Sync the data portion of the file specified by $offset and $length
753 to disk (but NOT the metadata), by calling the Linux-specific
754 sync_file_range call. If sync_file_range is not available or it
755 returns ENOSYS, then fdatasync or fsync is being substituted.
756
757 $flags can be a combination of
758 "IO::AIO::SYNC_FILE_RANGE_WAIT_BEFORE",
759 "IO::AIO::SYNC_FILE_RANGE_WRITE" and
760 "IO::AIO::SYNC_FILE_RANGE_WAIT_AFTER": refer to the sync_file_range
761 manpage for details.
762
763 aio_pathsync $path, $callback->($status)
764 This request tries to open, fsync and close the given path. This is
765 a composite request intended to sync directories after directory
766 operations (E.g. rename). This might not work on all operating
767 systems or have any specific effect, but usually it makes sure that
768 directory changes get written to disc. It works for anything that
769 can be opened for read-only, not just directories.
770
771 Future versions of this function might fall back to other methods
772 when "fsync" on the directory fails (such as calling "sync").
773
774 Passes 0 when everything went ok, and -1 on error.
775
776 aio_msync $scalar, $offset = 0, $length = undef, flags = 0,
777 $callback->($status)
778 This is a rather advanced IO::AIO call, which only works on
779 mmap(2)ed scalars (see the "IO::AIO::mmap" function, although it
780 also works on data scalars managed by the Sys::Mmap or Mmap modules,
781 note that the scalar must only be modified in-place while an aio
782 operation is pending on it).
783
784 It calls the "msync" function of your OS, if available, with the
785 memory area starting at $offset in the string and ending $length
786 bytes later. If $length is negative, counts from the end, and if
787 $length is "undef", then it goes till the end of the string. The
788 flags can be a combination of "IO::AIO::MS_ASYNC",
789 "IO::AIO::MS_INVALIDATE" and "IO::AIO::MS_SYNC".
790
791 aio_mtouch $scalar, $offset = 0, $length = undef, flags = 0,
792 $callback->($status)
793 This is a rather advanced IO::AIO call, which works best on
794 mmap(2)ed scalars.
795
796 It touches (reads or writes) all memory pages in the specified range
797 inside the scalar. All caveats and parameters are the same as for
798 "aio_msync", above, except for flags, which must be either 0 (which
799 reads all pages and ensures they are instantiated) or
800 "IO::AIO::MT_MODIFY", which modifies the memory page s(by reading
801 and writing an octet from it, which dirties the page).
802
803 aio_mlock $scalar, $offset = 0, $length = undef, $callback->($status)
804 This is a rather advanced IO::AIO call, which works best on
805 mmap(2)ed scalars.
806
807 It reads in all the pages of the underlying storage into memory (if
808 any) and locks them, so they are not getting swapped/paged out or
809 removed.
810
811 If $length is undefined, then the scalar will be locked till the
812 end.
813
814 On systems that do not implement "mlock", this function returns -1
815 and sets errno to "ENOSYS".
816
817 Note that the corresponding "munlock" is synchronous and is
818 documented under "MISCELLANEOUS FUNCTIONS".
819
820 Example: open a file, mmap and mlock it - both will be undone when
821 $data gets destroyed.
822
823 open my $fh, "<", $path or die "$path: $!";
824 my $data;
825 IO::AIO::mmap $data, -s $fh, IO::AIO::PROT_READ, IO::AIO::MAP_SHARED, $fh;
826 aio_mlock $data; # mlock in background
827
828 aio_mlockall $flags, $callback->($status)
829 Calls the "mlockall" function with the given $flags (a combination
830 of "IO::AIO::MCL_CURRENT" and "IO::AIO::MCL_FUTURE").
831
832 On systems that do not implement "mlockall", this function returns
833 -1 and sets errno to "ENOSYS".
834
835 Note that the corresponding "munlockall" is synchronous and is
836 documented under "MISCELLANEOUS FUNCTIONS".
837
838 Example: asynchronously lock all current and future pages into
839 memory.
840
841 aio_mlockall IO::AIO::MCL_FUTURE;
842
843 aio_group $callback->(...)
844 This is a very special aio request: Instead of doing something, it
845 is a container for other aio requests, which is useful if you want
846 to bundle many requests into a single, composite, request with a
847 definite callback and the ability to cancel the whole request with
848 its subrequests.
849
850 Returns an object of class IO::AIO::GRP. See its documentation below
851 for more info.
852
853 Example:
854
855 my $grp = aio_group sub {
856 print "all stats done\n";
857 };
858
859 add $grp
860 (aio_stat ...),
861 (aio_stat ...),
862 ...;
863
864 aio_nop $callback->()
865 This is a special request - it does nothing in itself and is only
866 used for side effects, such as when you want to add a dummy request
867 to a group so that finishing the requests in the group depends on
868 executing the given code.
869
870 While this request does nothing, it still goes through the execution
871 phase and still requires a worker thread. Thus, the callback will
872 not be executed immediately but only after other requests in the
873 queue have entered their execution phase. This can be used to
874 measure request latency.
875
876 IO::AIO::aio_busy $fractional_seconds, $callback->() *NOT EXPORTED*
877 Mainly used for debugging and benchmarking, this aio request puts
878 one of the request workers to sleep for the given time.
879
880 While it is theoretically handy to have simple I/O scheduling
881 requests like sleep and file handle readable/writable, the overhead
882 this creates is immense (it blocks a thread for a long time) so do
883 not use this function except to put your application under
884 artificial I/O pressure.
885
886 IO::AIO::REQ CLASS
887 All non-aggregate "aio_*" functions return an object of this class when
888 called in non-void context.
889
890 cancel $req
891 Cancels the request, if possible. Has the effect of skipping
892 execution when entering the execute state and skipping calling the
893 callback when entering the the result state, but will leave the
894 request otherwise untouched (with the exception of readdir). That
895 means that requests that currently execute will not be stopped and
896 resources held by the request will not be freed prematurely.
897
898 cb $req $callback->(...)
899 Replace (or simply set) the callback registered to the request.
900
901 IO::AIO::GRP CLASS
902 This class is a subclass of IO::AIO::REQ, so all its methods apply to
903 objects of this class, too.
904
905 A IO::AIO::GRP object is a special request that can contain multiple
906 other aio requests.
907
908 You create one by calling the "aio_group" constructing function with a
909 callback that will be called when all contained requests have entered
910 the "done" state:
911
912 my $grp = aio_group sub {
913 print "all requests are done\n";
914 };
915
916 You add requests by calling the "add" method with one or more
917 "IO::AIO::REQ" objects:
918
919 $grp->add (aio_unlink "...");
920
921 add $grp aio_stat "...", sub {
922 $_[0] or return $grp->result ("error");
923
924 # add another request dynamically, if first succeeded
925 add $grp aio_open "...", sub {
926 $grp->result ("ok");
927 };
928 };
929
930 This makes it very easy to create composite requests (see the source of
931 "aio_move" for an application) that work and feel like simple requests.
932
933 * The IO::AIO::GRP objects will be cleaned up during calls to
934 "IO::AIO::poll_cb", just like any other request.
935
936 * They can be canceled like any other request. Canceling will cancel
937 not only the request itself, but also all requests it contains.
938
939 * They can also can also be added to other IO::AIO::GRP objects.
940
941 * You must not add requests to a group from within the group callback
942 (or any later time).
943
944 Their lifetime, simplified, looks like this: when they are empty, they
945 will finish very quickly. If they contain only requests that are in the
946 "done" state, they will also finish. Otherwise they will continue to
947 exist.
948
949 That means after creating a group you have some time to add requests
950 (precisely before the callback has been invoked, which is only done
951 within the "poll_cb"). And in the callbacks of those requests, you can
952 add further requests to the group. And only when all those requests have
953 finished will the the group itself finish.
954
955 add $grp ...
956 $grp->add (...)
957 Add one or more requests to the group. Any type of IO::AIO::REQ can
958 be added, including other groups, as long as you do not create
959 circular dependencies.
960
961 Returns all its arguments.
962
963 $grp->cancel_subs
964 Cancel all subrequests and clears any feeder, but not the group
965 request itself. Useful when you queued a lot of events but got a
966 result early.
967
968 The group request will finish normally (you cannot add requests to
969 the group).
970
971 $grp->result (...)
972 Set the result value(s) that will be passed to the group callback
973 when all subrequests have finished and set the groups errno to the
974 current value of errno (just like calling "errno" without an error
975 number). By default, no argument will be passed and errno is zero.
976
977 $grp->errno ([$errno])
978 Sets the group errno value to $errno, or the current value of errno
979 when the argument is missing.
980
981 Every aio request has an associated errno value that is restored
982 when the callback is invoked. This method lets you change this value
983 from its default (0).
984
985 Calling "result" will also set errno, so make sure you either set $!
986 before the call to "result", or call c<errno> after it.
987
988 feed $grp $callback->($grp)
989 Sets a feeder/generator on this group: every group can have an
990 attached generator that generates requests if idle. The idea behind
991 this is that, although you could just queue as many requests as you
992 want in a group, this might starve other requests for a potentially
993 long time. For example, "aio_scandir" might generate hundreds of
994 thousands "aio_stat" requests, delaying any later requests for a
995 long time.
996
997 To avoid this, and allow incremental generation of requests, you can
998 instead a group and set a feeder on it that generates those
999 requests. The feed callback will be called whenever there are few
1000 enough (see "limit", below) requests active in the group itself and
1001 is expected to queue more requests.
1002
1003 The feed callback can queue as many requests as it likes (i.e. "add"
1004 does not impose any limits).
1005
1006 If the feed does not queue more requests when called, it will be
1007 automatically removed from the group.
1008
1009 If the feed limit is 0 when this method is called, it will be set to
1010 2 automatically.
1011
1012 Example:
1013
1014 # stat all files in @files, but only ever use four aio requests concurrently:
1015
1016 my $grp = aio_group sub { print "finished\n" };
1017 limit $grp 4;
1018 feed $grp sub {
1019 my $file = pop @files
1020 or return;
1021
1022 add $grp aio_stat $file, sub { ... };
1023 };
1024
1025 limit $grp $num
1026 Sets the feeder limit for the group: The feeder will be called
1027 whenever the group contains less than this many requests.
1028
1029 Setting the limit to 0 will pause the feeding process.
1030
1031 The default value for the limit is 0, but note that setting a feeder
1032 automatically bumps it up to 2.
172 1033
173 SUPPORT FUNCTIONS 1034 SUPPORT FUNCTIONS
1035 EVENT PROCESSING AND EVENT LOOP INTEGRATION
174 $fileno = IO::AIO::poll_fileno 1036 $fileno = IO::AIO::poll_fileno
175 Return the *request result pipe file descriptor*. This filehandle 1037 Return the *request result pipe file descriptor*. This filehandle
176 must be polled for reading by some mechanism outside this module 1038 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 1039 (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. 1040 pipe becomes readable you have to call "poll_cb" to check the
1041 results.
179 1042
180 See "poll_cb" for an example. 1043 See "poll_cb" for an example.
181 1044
182 IO::AIO::poll_cb 1045 IO::AIO::poll_cb
183 Process all outstanding events on the result pipe. You have to call 1046 Process some outstanding events on the result pipe. You have to call
184 this regularly. Returns the number of events processed. Returns 1047 this regularly. Returns 0 if all events could be processed (or there
1048 were no events to process), or -1 if it returned earlier for
185 immediately when no events are outstanding. 1049 whatever reason. Returns immediately when no events are outstanding.
1050 The amount of events processed depends on the settings of
1051 "IO::AIO::max_poll_req" and "IO::AIO::max_poll_time".
1052
1053 If not all requests were processed for whatever reason, the
1054 filehandle will still be ready when "poll_cb" returns, so normally
1055 you don't have to do anything special to have it called later.
1056
1057 Apart from calling "IO::AIO::poll_cb" when the event filehandle
1058 becomes ready, it can be beneficial to call this function from loops
1059 which submit a lot of requests, to make sure the results get
1060 processed when they become available and not just when the loop is
1061 finished and the event loop takes over again. This function returns
1062 very fast when there are no outstanding requests.
186 1063
187 Example: Install an Event watcher that automatically calls 1064 Example: Install an Event watcher that automatically calls
188 IO::AIO::poll_cb with high priority: 1065 IO::AIO::poll_cb with high priority (more examples can be found in
1066 the SYNOPSIS section, at the top of this document):
189 1067
190 Event->io (fd => IO::AIO::poll_fileno, 1068 Event->io (fd => IO::AIO::poll_fileno,
191 poll => 'r', async => 1, 1069 poll => 'r', async => 1,
192 cb => \&IO::AIO::poll_cb); 1070 cb => \&IO::AIO::poll_cb);
193 1071
194 IO::AIO::poll_wait 1072 IO::AIO::poll_wait
1073 If there are any outstanding requests and none of them in the result
195 Wait till the result filehandle becomes ready for reading (simply 1074 phase, wait till the result filehandle becomes ready for reading
196 does a "select" on the filehandle. This is useful if you want to 1075 (simply does a "select" on the filehandle. This is useful if you
197 synchronously wait for some requests to finish). 1076 want to synchronously wait for some requests to finish).
198 1077
199 See "nreqs" for an example. 1078 See "nreqs" for an example.
200 1079
1080 IO::AIO::poll
1081 Waits until some requests have been handled.
1082
1083 Returns the number of requests processed, but is otherwise strictly
1084 equivalent to:
1085
1086 IO::AIO::poll_wait, IO::AIO::poll_cb
1087
201 IO::AIO::nreqs 1088 IO::AIO::flush
202 Returns the number of requests currently outstanding (i.e. for which 1089 Wait till all outstanding AIO requests have been handled.
203 their callback has not been invoked yet).
204 1090
205 Example: wait till there are no outstanding requests anymore: 1091 Strictly equivalent to:
206 1092
207 IO::AIO::poll_wait, IO::AIO::poll_cb 1093 IO::AIO::poll_wait, IO::AIO::poll_cb
208 while IO::AIO::nreqs; 1094 while IO::AIO::nreqs;
209 1095
210 IO::AIO::flush 1096 IO::AIO::max_poll_reqs $nreqs
211 Wait till all outstanding AIO requests have been handled. 1097 IO::AIO::max_poll_time $seconds
1098 These set the maximum number of requests (default 0, meaning
1099 infinity) that are being processed by "IO::AIO::poll_cb" in one
1100 call, respectively the maximum amount of time (default 0, meaning
1101 infinity) spent in "IO::AIO::poll_cb" to process requests (more
1102 correctly the mininum amount of time "poll_cb" is allowed to use).
212 1103
213 Strictly equivalent to: 1104 Setting "max_poll_time" to a non-zero value creates an overhead of
1105 one syscall per request processed, which is not normally a problem
1106 unless your callbacks are really really fast or your OS is really
1107 really slow (I am not mentioning Solaris here). Using
1108 "max_poll_reqs" incurs no overhead.
214 1109
215 IO::AIO::poll_wait, IO::AIO::poll_cb 1110 Setting these is useful if you want to ensure some level of
216 while IO::AIO::nreqs; 1111 interactiveness when perl is not fast enough to process all requests
1112 in time.
217 1113
218 IO::AIO::poll 1114 For interactive programs, values such as 0.01 to 0.1 should be fine.
219 Waits until some requests have been handled.
220 1115
221 Strictly equivalent to: 1116 Example: Install an Event watcher that automatically calls
1117 IO::AIO::poll_cb with low priority, to ensure that other parts of
1118 the program get the CPU sometimes even under high AIO load.
222 1119
223 IO::AIO::poll_wait, IO::AIO::poll_cb 1120 # try not to spend much more than 0.1s in poll_cb
224 if IO::AIO::nreqs; 1121 IO::AIO::max_poll_time 0.1;
225 1122
1123 # use a low priority so other tasks have priority
1124 Event->io (fd => IO::AIO::poll_fileno,
1125 poll => 'r', nice => 1,
1126 cb => &IO::AIO::poll_cb);
1127
1128 CONTROLLING THE NUMBER OF THREADS
226 IO::AIO::min_parallel $nthreads 1129 IO::AIO::min_parallel $nthreads
227 Set the minimum number of AIO threads to $nthreads. The default is 1130 Set the minimum number of AIO threads to $nthreads. The current
228 1, which means a single asynchronous operation can be done at one 1131 default is 8, which means eight asynchronous operations can execute
229 time (the number of outstanding operations, however, is unlimited). 1132 concurrently at any one time (the number of outstanding requests,
1133 however, is unlimited).
230 1134
1135 IO::AIO starts threads only on demand, when an AIO request is queued
1136 and no free thread exists. Please note that queueing up a hundred
1137 requests can create demand for a hundred threads, even if it turns
1138 out that everything is in the cache and could have been processed
1139 faster by a single thread.
1140
231 It is recommended to keep the number of threads low, as some Linux 1141 It is recommended to keep the number of threads relatively low, as
232 kernel versions will scale negatively with the number of threads 1142 some Linux kernel versions will scale negatively with the number of
233 (higher parallelity => MUCH higher latency). With current Linux 2.6 1143 threads (higher parallelity => MUCH higher latency). With current
234 versions, 4-32 threads should be fine. 1144 Linux 2.6 versions, 4-32 threads should be fine.
235 1145
236 Under normal circumstances you don't need to call this function, as 1146 Under most circumstances you don't need to call this function, as
237 this module automatically starts some threads (the exact number 1147 the module selects a default that is suitable for low to moderate
238 might change, and is currently 4). 1148 load.
239 1149
240 IO::AIO::max_parallel $nthreads 1150 IO::AIO::max_parallel $nthreads
241 Sets the maximum number of AIO threads to $nthreads. If more than 1151 Sets the maximum number of AIO threads to $nthreads. If more than
242 the specified number of threads are currently running, kill them. 1152 the specified number of threads are currently running, this function
243 This function blocks until the limit is reached. 1153 kills them. This function blocks until the limit is reached.
1154
1155 While $nthreads are zero, aio requests get queued but not executed
1156 until the number of threads has been increased again.
244 1157
245 This module automatically runs "max_parallel 0" at program end, to 1158 This module automatically runs "max_parallel 0" at program end, to
246 ensure that all threads are killed and that there are no outstanding 1159 ensure that all threads are killed and that there are no outstanding
247 requests. 1160 requests.
248 1161
249 Under normal circumstances you don't need to call this function. 1162 Under normal circumstances you don't need to call this function.
250 1163
1164 IO::AIO::max_idle $nthreads
1165 Limit the number of threads (default: 4) that are allowed to idle
1166 (i.e., threads that did not get a request to process within the idle
1167 timeout (default: 10 seconds). That means if a thread becomes idle
1168 while $nthreads other threads are also idle, it will free its
1169 resources and exit.
1170
1171 This is useful when you allow a large number of threads (e.g. 100 or
1172 1000) to allow for extremely high load situations, but want to free
1173 resources under normal circumstances (1000 threads can easily
1174 consume 30MB of RAM).
1175
1176 The default is probably ok in most situations, especially if thread
1177 creation is fast. If thread creation is very slow on your system you
1178 might want to use larger values.
1179
1180 IO::AIO::idle_timeout $seconds
1181 Sets the minimum idle timeout (default 10) after which worker
1182 threads are allowed to exit. SEe "IO::AIO::max_idle".
1183
251 $oldnreqs = IO::AIO::max_outstanding $nreqs 1184 IO::AIO::max_outstanding $maxreqs
252 Sets the maximum number of outstanding requests to $nreqs. If you 1185 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 1186 queue up more than this number of requests, the next call to
1187 "IO::AIO::poll_cb" (and other functions calling "poll_cb", such as
1188 "IO::AIO::flush" or "IO::AIO::poll") will block until the limit is
1189 no longer exceeded.
1190
1191 In other words, this setting does not enforce a queue limit, but can
1192 be used to make poll functions block if the limit is exceeded.
1193
1194 This is a very bad function to use in interactive programs because
1195 it blocks, and a bad way to reduce concurrency because it is
1196 inexact: Better use an "aio_group" together with a feed callback.
1197
1198 It's main use is in scripts without an event loop - when you want to
1199 stat a lot of files, you can write somehting like this:
1200
1201 IO::AIO::max_outstanding 32;
1202
1203 for my $path (...) {
1204 aio_stat $path , ...;
1205 IO::AIO::poll_cb;
1206 }
1207
1208 IO::AIO::flush;
1209
1210 The call to "poll_cb" inside the loop will normally return
1211 instantly, but as soon as more thna 32 reqeusts are in-flight, it
254 block until some requests have been handled. 1212 will block until some requests have been handled. This keeps the
1213 loop from pushing a large number of "aio_stat" requests onto the
1214 queue.
255 1215
256 The default is very large, so normally there is no practical limit. 1216 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 1217 no practical limit on the number of outstanding requests.
258 if you set this to a relatively low number, such as 100.
259 1218
260 Under normal circumstances you don't need to call this function. 1219 STATISTICAL INFORMATION
1220 IO::AIO::nreqs
1221 Returns the number of requests currently in the ready, execute or
1222 pending states (i.e. for which their callback has not been invoked
1223 yet).
1224
1225 Example: wait till there are no outstanding requests anymore:
1226
1227 IO::AIO::poll_wait, IO::AIO::poll_cb
1228 while IO::AIO::nreqs;
1229
1230 IO::AIO::nready
1231 Returns the number of requests currently in the ready state (not yet
1232 executed).
1233
1234 IO::AIO::npending
1235 Returns the number of requests currently in the pending state
1236 (executed, but not yet processed by poll_cb).
1237
1238 MISCELLANEOUS FUNCTIONS
1239 IO::AIO implements some functions that might be useful, but are not
1240 asynchronous.
1241
1242 IO::AIO::sendfile $ofh, $ifh, $offset, $count
1243 Calls the "eio_sendfile_sync" function, which is like
1244 "aio_sendfile", but is blocking (this makes most sense if you know
1245 the input data is likely cached already and the output filehandle is
1246 set to non-blocking operations).
1247
1248 Returns the number of bytes copied, or -1 on error.
1249
1250 IO::AIO::fadvise $fh, $offset, $len, $advice
1251 Simply calls the "posix_fadvise" function (see its manpage for
1252 details). The following advice constants are avaiable:
1253 "IO::AIO::FADV_NORMAL", "IO::AIO::FADV_SEQUENTIAL",
1254 "IO::AIO::FADV_RANDOM", "IO::AIO::FADV_NOREUSE",
1255 "IO::AIO::FADV_WILLNEED", "IO::AIO::FADV_DONTNEED".
1256
1257 On systems that do not implement "posix_fadvise", this function
1258 returns ENOSYS, otherwise the return value of "posix_fadvise".
1259
1260 IO::AIO::madvise $scalar, $offset, $len, $advice
1261 Simply calls the "posix_madvise" function (see its manpage for
1262 details). The following advice constants are avaiable:
1263 "IO::AIO::MADV_NORMAL", "IO::AIO::MADV_SEQUENTIAL",
1264 "IO::AIO::MADV_RANDOM", "IO::AIO::MADV_WILLNEED",
1265 "IO::AIO::MADV_DONTNEED".
1266
1267 On systems that do not implement "posix_madvise", this function
1268 returns ENOSYS, otherwise the return value of "posix_madvise".
1269
1270 IO::AIO::mprotect $scalar, $offset, $len, $protect
1271 Simply calls the "mprotect" function on the preferably AIO::mmap'ed
1272 $scalar (see its manpage for details). The following protect
1273 constants are avaiable: "IO::AIO::PROT_NONE", "IO::AIO::PROT_READ",
1274 "IO::AIO::PROT_WRITE", "IO::AIO::PROT_EXEC".
1275
1276 On systems that do not implement "mprotect", this function returns
1277 ENOSYS, otherwise the return value of "mprotect".
1278
1279 IO::AIO::mmap $scalar, $length, $prot, $flags, $fh[, $offset]
1280 Memory-maps a file (or anonymous memory range) and attaches it to
1281 the given $scalar, which will act like a string scalar.
1282
1283 The only operations allowed on the scalar are "substr"/"vec" that
1284 don't change the string length, and most read-only operations such
1285 as copying it or searching it with regexes and so on.
1286
1287 Anything else is unsafe and will, at best, result in memory leaks.
1288
1289 The memory map associated with the $scalar is automatically removed
1290 when the $scalar is destroyed, or when the "IO::AIO::mmap" or
1291 "IO::AIO::munmap" functions are called.
1292
1293 This calls the "mmap"(2) function internally. See your system's
1294 manual page for details on the $length, $prot and $flags parameters.
1295
1296 The $length must be larger than zero and smaller than the actual
1297 filesize.
1298
1299 $prot is a combination of "IO::AIO::PROT_NONE",
1300 "IO::AIO::PROT_EXEC", "IO::AIO::PROT_READ" and/or
1301 "IO::AIO::PROT_WRITE",
1302
1303 $flags can be a combination of "IO::AIO::MAP_SHARED" or
1304 "IO::AIO::MAP_PRIVATE", or a number of system-specific flags (when
1305 not available, the are defined as 0): "IO::AIO::MAP_ANONYMOUS"
1306 (which is set to "MAP_ANON" if your system only provides this
1307 constant), "IO::AIO::MAP_HUGETLB", "IO::AIO::MAP_LOCKED",
1308 "IO::AIO::MAP_NORESERVE", "IO::AIO::MAP_POPULATE" or
1309 "IO::AIO::MAP_NONBLOCK"
1310
1311 If $fh is "undef", then a file descriptor of -1 is passed.
1312
1313 $offset is the offset from the start of the file - it generally must
1314 be a multiple of "IO::AIO::PAGESIZE" and defaults to 0.
1315
1316 Example:
1317
1318 use Digest::MD5;
1319 use IO::AIO;
1320
1321 open my $fh, "<verybigfile"
1322 or die "$!";
1323
1324 IO::AIO::mmap my $data, -s $fh, IO::AIO::PROT_READ, IO::AIO::MAP_SHARED, $fh
1325 or die "verybigfile: $!";
1326
1327 my $fast_md5 = md5 $data;
1328
1329 IO::AIO::munmap $scalar
1330 Removes a previous mmap and undefines the $scalar.
1331
1332 IO::AIO::munlock $scalar, $offset = 0, $length = undef
1333 Calls the "munlock" function, undoing the effects of a previous
1334 "aio_mlock" call (see its description for details).
1335
1336 IO::AIO::munlockall
1337 Calls the "munlockall" function.
1338
1339 On systems that do not implement "munlockall", this function returns
1340 ENOSYS, otherwise the return value of "munlockall".
1341
1342EVENT LOOP INTEGRATION
1343 It is recommended to use AnyEvent::AIO to integrate IO::AIO
1344 automatically into many event loops:
1345
1346 # AnyEvent integration (EV, Event, Glib, Tk, POE, urxvt, pureperl...)
1347 use AnyEvent::AIO;
1348
1349 You can also integrate IO::AIO manually into many event loops, here are
1350 some examples of how to do this:
1351
1352 # EV integration
1353 my $aio_w = EV::io IO::AIO::poll_fileno, EV::READ, \&IO::AIO::poll_cb;
1354
1355 # Event integration
1356 Event->io (fd => IO::AIO::poll_fileno,
1357 poll => 'r',
1358 cb => \&IO::AIO::poll_cb);
1359
1360 # Glib/Gtk2 integration
1361 add_watch Glib::IO IO::AIO::poll_fileno,
1362 in => sub { IO::AIO::poll_cb; 1 };
1363
1364 # Tk integration
1365 Tk::Event::IO->fileevent (IO::AIO::poll_fileno, "",
1366 readable => \&IO::AIO::poll_cb);
1367
1368 # Danga::Socket integration
1369 Danga::Socket->AddOtherFds (IO::AIO::poll_fileno =>
1370 \&IO::AIO::poll_cb);
1371
1372 FORK BEHAVIOUR
1373 Usage of pthreads in a program changes the semantics of fork
1374 considerably. Specifically, only async-safe functions can be called
1375 after fork. Perl doesn't know about this, so in general, you cannot call
1376 fork with defined behaviour in perl. IO::AIO uses pthreads, so this
1377 applies, but many other extensions and (for inexplicable reasons) perl
1378 itself often is linked against pthreads, so this limitation applies.
1379
1380 Some operating systems have extensions that allow safe use of fork, and
1381 this module should do "the right thing" on those, and tries on others.
1382 At the time of this writing (2011) only GNU/Linux supports these
1383 extensions to POSIX.
1384
1385 MEMORY USAGE
1386 Per-request usage:
1387
1388 Each aio request uses - depending on your architecture - around 100-200
1389 bytes of memory. In addition, stat requests need a stat buffer (possibly
1390 a few hundred bytes), readdir requires a result buffer and so on. Perl
1391 scalars and other data passed into aio requests will also be locked and
1392 will consume memory till the request has entered the done state.
1393
1394 This is not awfully much, so queuing lots of requests is not usually a
1395 problem.
1396
1397 Per-thread usage:
1398
1399 In the execution phase, some aio requests require more memory for
1400 temporary buffers, and each thread requires a stack and other data
1401 structures (usually around 16k-128k, depending on the OS).
1402
1403KNOWN BUGS
1404 Known bugs will be fixed in the next release.
261 1405
262SEE ALSO 1406SEE ALSO
263 Coro, Linux::AIO. 1407 AnyEvent::AIO for easy integration into event loops, Coro::AIO for a
1408 more natural syntax.
264 1409
265AUTHOR 1410AUTHOR
266 Marc Lehmann <schmorp@schmorp.de> 1411 Marc Lehmann <schmorp@schmorp.de>
267 http://home.schmorp.de/ 1412 http://home.schmorp.de/
268 1413

Diff Legend

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