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

Comparing IO-AIO/README (file contents):
Revision 1.19 by root, Sun Oct 29 01:03:13 2006 UTC vs.
Revision 1.58 by root, Sun May 1 17:19:39 2016 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
… …
23 $req->cancel; # cancel request if still in queue 24 $req->cancel; # cancel request if still in queue
24 25
25 my $grp = aio_group sub { print "all stats done\n" }; 26 my $grp = aio_group sub { print "all stats done\n" };
26 add $grp aio_stat "..." for ...; 27 add $grp aio_stat "..." for ...;
27 28
28 # AnyEvent integration
29 open my $fh, "<&=" . IO::AIO::poll_fileno or die "$!";
30 my $w = AnyEvent->io (fh => $fh, poll => 'r', cb => sub { IO::AIO::poll_cb });
31
32 # Event integration
33 Event->io (fd => IO::AIO::poll_fileno,
34 poll => 'r',
35 cb => \&IO::AIO::poll_cb);
36
37 # Glib/Gtk2 integration
38 add_watch Glib::IO IO::AIO::poll_fileno,
39 in => sub { IO::AIO::poll_cb; 1 };
40
41 # Tk integration
42 Tk::Event::IO->fileevent (IO::AIO::poll_fileno, "",
43 readable => \&IO::AIO::poll_cb);
44
45 # Danga::Socket integration
46 Danga::Socket->AddOtherFds (IO::AIO::poll_fileno =>
47 \&IO::AIO::poll_cb);
48
49DESCRIPTION 29DESCRIPTION
50 This module implements asynchronous I/O using whatever means your 30 This module implements asynchronous I/O using whatever means your
51 operating system supports. 31 operating system supports. It is implemented as an interface to "libeio"
32 (<http://software.schmorp.de/pkg/libeio.html>).
52 33
53 Asynchronous means that operations that can normally block your program 34 Asynchronous means that operations that can normally block your program
54 (e.g. reading from disk) will be done asynchronously: the operation will 35 (e.g. reading from disk) will be done asynchronously: the operation will
55 still block, but you can do something else in the meantime. This is 36 still block, but you can do something else in the meantime. This is
56 extremely useful for programs that need to stay interactive even when 37 extremely useful for programs that need to stay interactive even when
… …
58 but can also be used to easily do operations in parallel that are 39 but can also be used to easily do operations in parallel that are
59 normally done sequentially, e.g. stat'ing many files, which is much 40 normally done sequentially, e.g. stat'ing many files, which is much
60 faster on a RAID volume or over NFS when you do a number of stat 41 faster on a RAID volume or over NFS when you do a number of stat
61 operations concurrently. 42 operations concurrently.
62 43
63 While this works on all types of file descriptors (for example sockets), 44 While most of this works on all types of file descriptors (for example
64 using these functions on file descriptors that support nonblocking 45 sockets), using these functions on file descriptors that support
65 operation (again, sockets, pipes etc.) is very inefficient. Use an event 46 nonblocking operation (again, sockets, pipes etc.) is very inefficient.
66 loop for that (such as the Event module): IO::AIO will naturally fit 47 Use an event loop for that (such as the EV module): IO::AIO will
67 into such an event loop itself. 48 naturally fit into such an event loop itself.
68 49
69 In this version, a number of threads are started that execute your 50 In this version, a number of threads are started that execute your
70 requests and signal their completion. You don't need thread support in 51 requests and signal their completion. You don't need thread support in
71 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
72 perl. In the future, this module might make use of the native aio 53 perl. In the future, this module might make use of the native aio
… …
74 not well-supported or restricted (GNU/Linux doesn't allow them on normal 55 not well-supported or restricted (GNU/Linux doesn't allow them on normal
75 files currently, for example), and they would only support aio_read and 56 files currently, for example), and they would only support aio_read and
76 aio_write, so the remaining functionality would have to be implemented 57 aio_write, so the remaining functionality would have to be implemented
77 using threads anyway. 58 using threads anyway.
78 59
79 Although the module will work with in the presence of other (Perl-) 60 Although the module will work in the presence of other (Perl-) threads,
80 threads, it is currently not reentrant in any way, so use appropriate 61 it is currently not reentrant in any way, so use appropriate locking
81 locking yourself, always call "poll_cb" from within the same thread, or 62 yourself, always call "poll_cb" from within the same thread, or never
82 never call "poll_cb" (or other "aio_" functions) recursively. 63 call "poll_cb" (or other "aio_" functions) recursively.
83 64
84 EXAMPLE 65 EXAMPLE
85 This is a simple example that uses the Event module and loads 66 This is a simple example that uses the EV module and loads /etc/passwd
86 /etc/passwd asynchronously: 67 asynchronously:
87 68
88 use Fcntl;
89 use Event; 69 use EV;
90 use IO::AIO; 70 use IO::AIO;
91 71
92 # register the IO::AIO callback with Event 72 # register the IO::AIO callback with EV
93 Event->io (fd => IO::AIO::poll_fileno, 73 my $aio_w = EV::io IO::AIO::poll_fileno, EV::READ, \&IO::AIO::poll_cb;
94 poll => 'r',
95 cb => \&IO::AIO::poll_cb);
96 74
97 # queue the request to open /etc/passwd 75 # queue the request to open /etc/passwd
98 aio_open "/etc/passwd", O_RDONLY, 0, sub { 76 aio_open "/etc/passwd", IO::AIO::O_RDONLY, 0, sub {
99 my $fh = $_[0] 77 my $fh = shift
100 or die "error while opening: $!"; 78 or die "error while opening: $!";
101 79
102 # stat'ing filehandles is generally non-blocking 80 # stat'ing filehandles is generally non-blocking
103 my $size = -s $fh; 81 my $size = -s $fh;
104 82
… …
112 90
113 # file contents now in $contents 91 # file contents now in $contents
114 print $contents; 92 print $contents;
115 93
116 # exit event loop and program 94 # exit event loop and program
117 Event::unloop; 95 EV::break;
118 }; 96 };
119 }; 97 };
120 98
121 # possibly queue up other requests, or open GUI windows, 99 # possibly queue up other requests, or open GUI windows,
122 # check for sockets etc. etc. 100 # check for sockets etc. etc.
123 101
124 # process events as long as there are some: 102 # process events as long as there are some:
125 Event::loop; 103 EV::run;
126 104
127REQUEST ANATOMY AND LIFETIME 105REQUEST ANATOMY AND LIFETIME
128 Every "aio_*" function creates a request. which is a C data structure 106 Every "aio_*" function creates a request. which is a C data structure
129 not directly visible to Perl. 107 not directly visible to Perl.
130 108
… …
166 anymore (except possibly for the Perl object, but its connection to 144 anymore (except possibly for the Perl object, but its connection to
167 the actual aio request is severed and calling its methods will 145 the actual aio request is severed and calling its methods will
168 either do nothing or result in a runtime error). 146 either do nothing or result in a runtime error).
169 147
170FUNCTIONS 148FUNCTIONS
149 QUICK OVERVIEW
150 This section simply lists the prototypes most of the functions for quick
151 reference. See the following sections for function-by-function
152 documentation.
153
154 aio_wd $pathname, $callback->($wd)
155 aio_open $pathname, $flags, $mode, $callback->($fh)
156 aio_close $fh, $callback->($status)
157 aio_seek $fh,$offset,$whence, $callback->($offs)
158 aio_read $fh,$offset,$length, $data,$dataoffset, $callback->($retval)
159 aio_write $fh,$offset,$length, $data,$dataoffset, $callback->($retval)
160 aio_sendfile $out_fh, $in_fh, $in_offset, $length, $callback->($retval)
161 aio_readahead $fh,$offset,$length, $callback->($retval)
162 aio_stat $fh_or_path, $callback->($status)
163 aio_lstat $fh, $callback->($status)
164 aio_statvfs $fh_or_path, $callback->($statvfs)
165 aio_utime $fh_or_path, $atime, $mtime, $callback->($status)
166 aio_chown $fh_or_path, $uid, $gid, $callback->($status)
167 aio_chmod $fh_or_path, $mode, $callback->($status)
168 aio_truncate $fh_or_path, $offset, $callback->($status)
169 aio_allocate $fh, $mode, $offset, $len, $callback->($status)
170 aio_fiemap $fh, $start, $length, $flags, $count, $cb->(\@extents)
171 aio_unlink $pathname, $callback->($status)
172 aio_mknod $pathname, $mode, $dev, $callback->($status)
173 aio_link $srcpath, $dstpath, $callback->($status)
174 aio_symlink $srcpath, $dstpath, $callback->($status)
175 aio_readlink $pathname, $callback->($link)
176 aio_realpath $pathname, $callback->($path)
177 aio_rename $srcpath, $dstpath, $callback->($status)
178 aio_mkdir $pathname, $mode, $callback->($status)
179 aio_rmdir $pathname, $callback->($status)
180 aio_readdir $pathname, $callback->($entries)
181 aio_readdirx $pathname, $flags, $callback->($entries, $flags)
182 IO::AIO::READDIR_DENTS IO::AIO::READDIR_DIRS_FIRST
183 IO::AIO::READDIR_STAT_ORDER IO::AIO::READDIR_FOUND_UNKNOWN
184 aio_scandir $pathname, $maxreq, $callback->($dirs, $nondirs)
185 aio_load $pathname, $data, $callback->($status)
186 aio_copy $srcpath, $dstpath, $callback->($status)
187 aio_move $srcpath, $dstpath, $callback->($status)
188 aio_rmtree $pathname, $callback->($status)
189 aio_fcntl $fh, $cmd, $arg, $callback->($status)
190 aio_ioctl $fh, $request, $buf, $callback->($status)
191 aio_sync $callback->($status)
192 aio_syncfs $fh, $callback->($status)
193 aio_fsync $fh, $callback->($status)
194 aio_fdatasync $fh, $callback->($status)
195 aio_sync_file_range $fh, $offset, $nbytes, $flags, $callback->($status)
196 aio_pathsync $pathname, $callback->($status)
197 aio_msync $scalar, $offset = 0, $length = undef, flags = 0, $callback->($status)
198 aio_mtouch $scalar, $offset = 0, $length = undef, flags = 0, $callback->($status)
199 aio_mlock $scalar, $offset = 0, $length = undef, $callback->($status)
200 aio_mlockall $flags, $callback->($status)
201 aio_group $callback->(...)
202 aio_nop $callback->()
203
204 $prev_pri = aioreq_pri [$pri]
205 aioreq_nice $pri_adjust
206
207 IO::AIO::poll_wait
208 IO::AIO::poll_cb
209 IO::AIO::poll
210 IO::AIO::flush
211 IO::AIO::max_poll_reqs $nreqs
212 IO::AIO::max_poll_time $seconds
213 IO::AIO::min_parallel $nthreads
214 IO::AIO::max_parallel $nthreads
215 IO::AIO::max_idle $nthreads
216 IO::AIO::idle_timeout $seconds
217 IO::AIO::max_outstanding $maxreqs
218 IO::AIO::nreqs
219 IO::AIO::nready
220 IO::AIO::npending
221
222 IO::AIO::sendfile $ofh, $ifh, $offset, $count
223 IO::AIO::fadvise $fh, $offset, $len, $advice
224 IO::AIO::mmap $scalar, $length, $prot, $flags[, $fh[, $offset]]
225 IO::AIO::munmap $scalar
226 IO::AIO::madvise $scalar, $offset, $length, $advice
227 IO::AIO::mprotect $scalar, $offset, $length, $protect
228 IO::AIO::munlock $scalar, $offset = 0, $length = undef
229 IO::AIO::munlockall
230
231 API NOTES
232 All the "aio_*" calls are more or less thin wrappers around the syscall
233 with the same name (sans "aio_"). The arguments are similar or
234 identical, and they all accept an additional (and optional) $callback
235 argument which must be a code reference. This code reference will be
236 called after the syscall has been executed in an asynchronous fashion.
237 The results of the request will be passed as arguments to the callback
238 (and, if an error occured, in $!) - for most requests the syscall return
239 code (e.g. most syscalls return -1 on error, unlike perl, which usually
240 delivers "false").
241
242 Some requests (such as "aio_readdir") pass the actual results and
243 communicate failures by passing "undef".
244
245 All functions expecting a filehandle keep a copy of the filehandle
246 internally until the request has finished.
247
248 All functions return request objects of type IO::AIO::REQ that allow
249 further manipulation of those requests while they are in-flight.
250
251 The pathnames you pass to these routines *should* be absolute. The
252 reason for this is that at the time the request is being executed, the
253 current working directory could have changed. Alternatively, you can
254 make sure that you never change the current working directory anywhere
255 in the program and then use relative paths. You can also take advantage
256 of IO::AIOs working directory abstraction, that lets you specify paths
257 relative to some previously-opened "working directory object" - see the
258 description of the "IO::AIO::WD" class later in this document.
259
260 To encode pathnames as octets, either make sure you either: a) always
261 pass in filenames you got from outside (command line, readdir etc.)
262 without tinkering, b) are in your native filesystem encoding, c) use the
263 Encode module and encode your pathnames to the locale (or other)
264 encoding in effect in the user environment, d) use
265 Glib::filename_from_unicode on unicode filenames or e) use something
266 else to ensure your scalar has the correct contents.
267
268 This works, btw. independent of the internal UTF-8 bit, which IO::AIO
269 handles correctly whether it is set or not.
270
171 AIO REQUEST FUNCTIONS 271 AIO REQUEST FUNCTIONS
172 All the "aio_*" calls are more or less thin wrappers around the
173 syscall with the same name (sans "aio_"). The arguments are similar
174 or identical, and they all accept an additional (and optional)
175 $callback argument which must be a code reference. This code
176 reference will get called with the syscall return code (e.g. most
177 syscalls return -1 on error, unlike perl, which usually delivers
178 "false") as it's sole argument when the given syscall has been
179 executed asynchronously.
180
181 All functions expecting a filehandle keep a copy of the filehandle
182 internally until the request has finished.
183
184 All functions return request objects of type IO::AIO::REQ that allow
185 further manipulation of those requests while they are in-flight.
186
187 The pathnames you pass to these routines *must* be absolute and
188 encoded as octets. The reason for the former is that at the time the
189 request is being executed, the current working directory could have
190 changed. Alternatively, you can make sure that you never change the
191 current working directory anywhere in the program and then use
192 relative paths.
193
194 To encode pathnames as octets, either make sure you either: a)
195 always pass in filenames you got from outside (command line, readdir
196 etc.) without tinkering, b) are ASCII or ISO 8859-1, c) use the
197 Encode module and encode your pathnames to the locale (or other)
198 encoding in effect in the user environment, d) use
199 Glib::filename_from_unicode on unicode filenames or e) use something
200 else to ensure your scalar has the correct contents.
201
202 This works, btw. independent of the internal UTF-8 bit, which
203 IO::AIO handles correctly wether it is set or not.
204
205 $prev_pri = aioreq_pri [$pri] 272 $prev_pri = aioreq_pri [$pri]
206 Returns the priority value that would be used for the next 273 Returns the priority value that would be used for the next request
207 request and, if $pri is given, sets the priority for the next 274 and, if $pri is given, sets the priority for the next aio request.
208 aio request.
209 275
210 The default priority is 0, the minimum and maximum priorities 276 The default priority is 0, the minimum and maximum priorities are -4
211 are -4 and 4, respectively. Requests with higher priority will 277 and 4, respectively. Requests with higher priority will be serviced
212 be serviced first. 278 first.
213 279
214 The priority will be reset to 0 after each call to one of the 280 The priority will be reset to 0 after each call to one of the
215 "aio_*" functions. 281 "aio_*" functions.
216 282
217 Example: open a file with low priority, then read something from 283 Example: open a file with low priority, then read something from it
218 it with higher priority so the read request is serviced before 284 with higher priority so the read request is serviced before other
219 other low priority open requests (potentially spamming the 285 low priority open requests (potentially spamming the cache):
220 cache):
221 286
287 aioreq_pri -3;
288 aio_open ..., sub {
289 return unless $_[0];
290
222 aioreq_pri -3; 291 aioreq_pri -2;
223 aio_open ..., sub {
224 return unless $_[0];
225
226 aioreq_pri -2;
227 aio_read $_[0], ..., sub { 292 aio_read $_[0], ..., sub {
228 ...
229 };
230 };
231
232 aioreq_nice $pri_adjust
233 Similar to "aioreq_pri", but subtracts the given value from the
234 current priority, so the effect is cumulative.
235
236 aio_open $pathname, $flags, $mode, $callback->($fh)
237 Asynchronously open or create a file and call the callback with
238 a newly created filehandle for the file.
239
240 The pathname passed to "aio_open" must be absolute. See API
241 NOTES, above, for an explanation.
242
243 The $flags argument is a bitmask. See the "Fcntl" module for a
244 list. They are the same as used by "sysopen".
245
246 Likewise, $mode specifies the mode of the newly created file, if
247 it didn't exist and "O_CREAT" has been given, just like perl's
248 "sysopen", except that it is mandatory (i.e. use 0 if you don't
249 create new files, and 0666 or 0777 if you do).
250
251 Example:
252
253 aio_open "/etc/passwd", O_RDONLY, 0, sub {
254 if ($_[0]) {
255 print "open successful, fh is $_[0]\n";
256 ...
257 } else {
258 die "open failed: $!\n";
259 }
260 };
261
262 aio_close $fh, $callback->($status)
263 Asynchronously close a file and call the callback with the
264 result code. *WARNING:* although accepted, you should not pass
265 in a perl filehandle here, as perl will likely close the file
266 descriptor another time when the filehandle is destroyed.
267 Normally, you can safely call perls "close" or just let
268 filehandles go out of scope.
269
270 This is supposed to be a bug in the API, so that might change.
271 It's therefore best to avoid this function.
272
273 aio_read $fh,$offset,$length, $data,$dataoffset,
274 $callback->($retval)
275 aio_write $fh,$offset,$length, $data,$dataoffset,
276 $callback->($retval)
277 Reads or writes "length" bytes from the specified "fh" and
278 "offset" into the scalar given by "data" and offset "dataoffset"
279 and calls the callback without the actual number of bytes read
280 (or -1 on error, just like the syscall).
281
282 The $data scalar *MUST NOT* be modified in any way while the
283 request is outstanding. Modifying it can result in segfaults or
284 WW3 (if the necessary/optional hardware is installed).
285
286 Example: Read 15 bytes at offset 7 into scalar $buffer, starting
287 at offset 0 within the scalar:
288
289 aio_read $fh, 7, 15, $buffer, 0, sub {
290 $_[0] > 0 or die "read error: $!";
291 print "read $_[0] bytes: <$buffer>\n";
292 };
293
294 aio_sendfile $out_fh, $in_fh, $in_offset, $length,
295 $callback->($retval)
296 Tries to copy $length bytes from $in_fh to $out_fh. It starts
297 reading at byte offset $in_offset, and starts writing at the
298 current file offset of $out_fh. Because of that, it is not safe
299 to issue more than one "aio_sendfile" per $out_fh, as they will
300 interfere with each other.
301
302 This call tries to make use of a native "sendfile" syscall to
303 provide zero-copy operation. For this to work, $out_fh should
304 refer to a socket, and $in_fh should refer to mmap'able file.
305
306 If the native sendfile call fails or is not implemented, it will
307 be emulated, so you can call "aio_sendfile" on any type of
308 filehandle regardless of the limitations of the operating
309 system.
310
311 Please note, however, that "aio_sendfile" can read more bytes
312 from $in_fh than are written, and there is no way to find out
313 how many bytes have been read from "aio_sendfile" alone, as
314 "aio_sendfile" only provides the number of bytes written to
315 $out_fh. Only if the result value equals $length one can assume
316 that $length bytes have been read.
317
318 aio_readahead $fh,$offset,$length, $callback->($retval)
319 "aio_readahead" populates the page cache with data from a file
320 so that subsequent reads from that file will not block on disk
321 I/O. The $offset argument specifies the starting point from
322 which data is to be read and $length specifies the number of
323 bytes to be read. I/O is performed in whole pages, so that
324 offset is effectively rounded down to a page boundary and bytes
325 are read up to the next page boundary greater than or equal to
326 (off-set+length). "aio_readahead" does not read beyond the end
327 of the file. The current file offset of the file is left
328 unchanged.
329
330 If that syscall doesn't exist (likely if your OS isn't Linux) it
331 will be emulated by simply reading the data, which would have a
332 similar effect.
333
334 aio_stat $fh_or_path, $callback->($status)
335 aio_lstat $fh, $callback->($status)
336 Works like perl's "stat" or "lstat" in void context. The
337 callback will be called after the stat and the results will be
338 available using "stat _" or "-s _" etc...
339
340 The pathname passed to "aio_stat" must be absolute. See API
341 NOTES, above, for an explanation.
342
343 Currently, the stats are always 64-bit-stats, i.e. instead of
344 returning an error when stat'ing a large file, the results will
345 be silently truncated unless perl itself is compiled with large
346 file support.
347
348 Example: Print the length of /etc/passwd:
349
350 aio_stat "/etc/passwd", sub {
351 $_[0] and die "stat failed: $!";
352 print "size is ", -s _, "\n";
353 };
354
355 aio_unlink $pathname, $callback->($status)
356 Asynchronously unlink (delete) a file and call the callback with
357 the result code.
358
359 aio_mknod $path, $mode, $dev, $callback->($status)
360 [EXPERIMENTAL]
361
362 Asynchronously create a device node (or fifo). See mknod(2).
363
364 The only (POSIX-) portable way of calling this function is:
365
366 aio_mknod $path, IO::AIO::S_IFIFO | $mode, 0, sub { ...
367
368 aio_link $srcpath, $dstpath, $callback->($status)
369 Asynchronously create a new link to the existing object at
370 $srcpath at the path $dstpath and call the callback with the
371 result code.
372
373 aio_symlink $srcpath, $dstpath, $callback->($status)
374 Asynchronously create a new symbolic link to the existing object
375 at $srcpath at the path $dstpath and call the callback with the
376 result code.
377
378 aio_rename $srcpath, $dstpath, $callback->($status)
379 Asynchronously rename the object at $srcpath to $dstpath, just
380 as rename(2) and call the callback with the result code.
381
382 aio_rmdir $pathname, $callback->($status)
383 Asynchronously rmdir (delete) a directory and call the callback
384 with the result code.
385
386 aio_readdir $pathname, $callback->($entries)
387 Unlike the POSIX call of the same name, "aio_readdir" reads an
388 entire directory (i.e. opendir + readdir + closedir). The
389 entries will not be sorted, and will NOT include the "." and
390 ".." entries.
391
392 The callback a single argument which is either "undef" or an
393 array-ref with the filenames.
394
395 aio_copy $srcpath, $dstpath, $callback->($status)
396 Try to copy the *file* (directories not supported as either
397 source or destination) from $srcpath to $dstpath and call the
398 callback with the 0 (error) or -1 ok.
399
400 This is a composite request that it creates the destination file
401 with mode 0200 and copies the contents of the source file into
402 it using "aio_sendfile", followed by restoring atime, mtime,
403 access mode and uid/gid, in that order.
404
405 If an error occurs, the partial destination file will be
406 unlinked, if possible, except when setting atime, mtime, access
407 mode and uid/gid, where errors are being ignored.
408
409 aio_move $srcpath, $dstpath, $callback->($status)
410 Try to move the *file* (directories not supported as either
411 source or destination) from $srcpath to $dstpath and call the
412 callback with the 0 (error) or -1 ok.
413
414 This is a composite request that tries to rename(2) the file
415 first. If rename files with "EXDEV", it copies the file with
416 "aio_copy" and, if that is successful, unlinking the $srcpath.
417
418 aio_scandir $path, $maxreq, $callback->($dirs, $nondirs)
419 Scans a directory (similar to "aio_readdir") but additionally
420 tries to efficiently separate the entries of directory $path
421 into two sets of names, directories you can recurse into
422 (directories), and ones you cannot recurse into (everything
423 else, including symlinks to directories).
424
425 "aio_scandir" is a composite request that creates of many sub
426 requests_ $maxreq specifies the maximum number of outstanding
427 aio requests that this function generates. If it is "<= 0", then
428 a suitable default will be chosen (currently 4).
429
430 On error, the callback is called without arguments, otherwise it
431 receives two array-refs with path-relative entry names.
432
433 Example:
434
435 aio_scandir $dir, 0, sub {
436 my ($dirs, $nondirs) = @_;
437 print "real directories: @$dirs\n";
438 print "everything else: @$nondirs\n";
439 };
440
441 Implementation notes.
442
443 The "aio_readdir" cannot be avoided, but "stat()"'ing every
444 entry can.
445
446 After reading the directory, the modification time, size etc. of
447 the directory before and after the readdir is checked, and if
448 they match (and isn't the current time), the link count will be
449 used to decide how many entries are directories (if >= 2).
450 Otherwise, no knowledge of the number of subdirectories will be
451 assumed.
452
453 Then entries will be sorted into likely directories (everything
454 without a non-initial dot currently) and likely non-directories
455 (everything else). Then every entry plus an appended "/." will
456 be "stat"'ed, likely directories first. If that succeeds, it
457 assumes that the entry is a directory or a symlink to directory
458 (which will be checked seperately). This is often faster than
459 stat'ing the entry itself because filesystems might detect the
460 type of the entry without reading the inode data (e.g. ext2fs
461 filetype feature).
462
463 If the known number of directories (link count - 2) has been
464 reached, the rest of the entries is assumed to be
465 non-directories.
466
467 This only works with certainty on POSIX (= UNIX) filesystems,
468 which fortunately are the vast majority of filesystems around.
469
470 It will also likely work on non-POSIX filesystems with reduced
471 efficiency as those tend to return 0 or 1 as link counts, which
472 disables the directory counting heuristic.
473
474 aio_fsync $fh, $callback->($status)
475 Asynchronously call fsync on the given filehandle and call the
476 callback with the fsync result code.
477
478 aio_fdatasync $fh, $callback->($status)
479 Asynchronously call fdatasync on the given filehandle and call
480 the callback with the fdatasync result code.
481
482 If this call isn't available because your OS lacks it or it
483 couldn't be detected, it will be emulated by calling "fsync"
484 instead.
485
486 aio_group $callback->(...)
487 This is a very special aio request: Instead of doing something,
488 it is a container for other aio requests, which is useful if you
489 want to bundle many requests into a single, composite, request
490 with a definite callback and the ability to cancel the whole
491 request with its subrequests.
492
493 Returns an object of class IO::AIO::GRP. See its documentation
494 below for more info.
495
496 Example:
497
498 my $grp = aio_group sub {
499 print "all stats done\n";
500 };
501
502 add $grp
503 (aio_stat ...),
504 (aio_stat ...),
505 ...; 293 ...
506
507 aio_nop $callback->()
508 This is a special request - it does nothing in itself and is
509 only used for side effects, such as when you want to add a dummy
510 request to a group so that finishing the requests in the group
511 depends on executing the given code.
512
513 While this request does nothing, it still goes through the
514 execution phase and still requires a worker thread. Thus, the
515 callback will not be executed immediately but only after other
516 requests in the queue have entered their execution phase. This
517 can be used to measure request latency.
518
519 IO::AIO::aio_busy $fractional_seconds, $callback->() *NOT EXPORTED*
520 Mainly used for debugging and benchmarking, this aio request
521 puts one of the request workers to sleep for the given time.
522
523 While it is theoretically handy to have simple I/O scheduling
524 requests like sleep and file handle readable/writable, the
525 overhead this creates is immense (it blocks a thread for a long
526 time) so do not use this function except to put your application
527 under artificial I/O pressure.
528
529 IO::AIO::REQ CLASS
530 All non-aggregate "aio_*" functions return an object of this class
531 when called in non-void context.
532
533 cancel $req
534 Cancels the request, if possible. Has the effect of skipping
535 execution when entering the execute state and skipping calling
536 the callback when entering the the result state, but will leave
537 the request otherwise untouched. That means that requests that
538 currently execute will not be stopped and resources held by the
539 request will not be freed prematurely.
540
541 cb $req $callback->(...)
542 Replace (or simply set) the callback registered to the request.
543
544 IO::AIO::GRP CLASS
545 This class is a subclass of IO::AIO::REQ, so all its methods apply
546 to objects of this class, too.
547
548 A IO::AIO::GRP object is a special request that can contain multiple
549 other aio requests.
550
551 You create one by calling the "aio_group" constructing function with
552 a callback that will be called when all contained requests have
553 entered the "done" state:
554
555 my $grp = aio_group sub {
556 print "all requests are done\n";
557 };
558
559 You add requests by calling the "add" method with one or more
560 "IO::AIO::REQ" objects:
561
562 $grp->add (aio_unlink "...");
563
564 add $grp aio_stat "...", sub {
565 $_[0] or return $grp->result ("error");
566
567 # add another request dynamically, if first succeeded
568 add $grp aio_open "...", sub {
569 $grp->result ("ok");
570 }; 294 };
571 }; 295 };
572 296
297 aioreq_nice $pri_adjust
298 Similar to "aioreq_pri", but subtracts the given value from the
299 current priority, so the effect is cumulative.
300
301 aio_open $pathname, $flags, $mode, $callback->($fh)
302 Asynchronously open or create a file and call the callback with a
303 newly created filehandle for the file (or "undef" in case of an
304 error).
305
306 The pathname passed to "aio_open" must be absolute. See API NOTES,
307 above, for an explanation.
308
309 The $flags argument is a bitmask. See the "Fcntl" module for a list.
310 They are the same as used by "sysopen".
311
312 Likewise, $mode specifies the mode of the newly created file, if it
313 didn't exist and "O_CREAT" has been given, just like perl's
314 "sysopen", except that it is mandatory (i.e. use 0 if you don't
315 create new files, and 0666 or 0777 if you do). Note that the $mode
316 will be modified by the umask in effect then the request is being
317 executed, so better never change the umask.
318
319 Example:
320
321 aio_open "/etc/passwd", IO::AIO::O_RDONLY, 0, sub {
322 if ($_[0]) {
323 print "open successful, fh is $_[0]\n";
324 ...
325 } else {
326 die "open failed: $!\n";
327 }
328 };
329
330 In addition to all the common open modes/flags ("O_RDONLY",
331 "O_WRONLY", "O_RDWR", "O_CREAT", "O_TRUNC", "O_EXCL" and
332 "O_APPEND"), the following POSIX and non-POSIX constants are
333 available (missing ones on your system are, as usual, 0):
334
335 "O_ASYNC", "O_DIRECT", "O_NOATIME", "O_CLOEXEC", "O_NOCTTY",
336 "O_NOFOLLOW", "O_NONBLOCK", "O_EXEC", "O_SEARCH", "O_DIRECTORY",
337 "O_DSYNC", "O_RSYNC", "O_SYNC", "O_PATH", "O_TMPFILE", and
338 "O_TTY_INIT".
339
340 aio_close $fh, $callback->($status)
341 Asynchronously close a file and call the callback with the result
342 code.
343
344 Unfortunately, you can't do this to perl. Perl *insists* very
345 strongly on closing the file descriptor associated with the
346 filehandle itself.
347
348 Therefore, "aio_close" will not close the filehandle - instead it
349 will use dup2 to overwrite the file descriptor with the write-end of
350 a pipe (the pipe fd will be created on demand and will be cached).
351
352 Or in other words: the file descriptor will be closed, but it will
353 not be free for reuse until the perl filehandle is closed.
354
355 aio_seek $fh, $offset, $whence, $callback->($offs)
356 Seeks the filehandle to the new $offset, similarly to perl's
357 "sysseek". The $whence can use the traditional values (0 for
358 "IO::AIO::SEEK_SET", 1 for "IO::AIO::SEEK_CUR" or 2 for
359 "IO::AIO::SEEK_END").
360
361 The resulting absolute offset will be passed to the callback, or -1
362 in case of an error.
363
364 In theory, the $whence constants could be different than the
365 corresponding values from Fcntl, but perl guarantees they are the
366 same, so don't panic.
367
368 As a GNU/Linux (and maybe Solaris) extension, also the constants
369 "IO::AIO::SEEK_DATA" and "IO::AIO::SEEK_HOLE" are available, if they
370 could be found. No guarantees about suitability for use in
371 "aio_seek" or Perl's "sysseek" can be made though, although I would
372 naively assume they "just work".
373
374 aio_read $fh,$offset,$length, $data,$dataoffset, $callback->($retval)
375 aio_write $fh,$offset,$length, $data,$dataoffset, $callback->($retval)
376 Reads or writes $length bytes from or to the specified $fh and
377 $offset into the scalar given by $data and offset $dataoffset and
378 calls the callback without the actual number of bytes read (or -1 on
379 error, just like the syscall).
380
381 "aio_read" will, like "sysread", shrink or grow the $data scalar to
382 offset plus the actual number of bytes read.
383
384 If $offset is undefined, then the current file descriptor offset
385 will be used (and updated), otherwise the file descriptor offset
386 will not be changed by these calls.
387
388 If $length is undefined in "aio_write", use the remaining length of
389 $data.
390
391 If $dataoffset is less than zero, it will be counted from the end of
392 $data.
393
394 The $data scalar *MUST NOT* be modified in any way while the request
395 is outstanding. Modifying it can result in segfaults or World War
396 III (if the necessary/optional hardware is installed).
397
398 Example: Read 15 bytes at offset 7 into scalar $buffer, starting at
399 offset 0 within the scalar:
400
401 aio_read $fh, 7, 15, $buffer, 0, sub {
402 $_[0] > 0 or die "read error: $!";
403 print "read $_[0] bytes: <$buffer>\n";
404 };
405
406 aio_sendfile $out_fh, $in_fh, $in_offset, $length, $callback->($retval)
407 Tries to copy $length bytes from $in_fh to $out_fh. It starts
408 reading at byte offset $in_offset, and starts writing at the current
409 file offset of $out_fh. Because of that, it is not safe to issue
410 more than one "aio_sendfile" per $out_fh, as they will interfere
411 with each other. The same $in_fh works fine though, as this function
412 does not move or use the file offset of $in_fh.
413
414 Please note that "aio_sendfile" can read more bytes from $in_fh than
415 are written, and there is no way to find out how many more bytes
416 have been read from "aio_sendfile" alone, as "aio_sendfile" only
417 provides the number of bytes written to $out_fh. Only if the result
418 value equals $length one can assume that $length bytes have been
419 read.
420
421 Unlike with other "aio_" functions, it makes a lot of sense to use
422 "aio_sendfile" on non-blocking sockets, as long as one end
423 (typically the $in_fh) is a file - the file I/O will then be
424 asynchronous, while the socket I/O will be non-blocking. Note,
425 however, that you can run into a trap where "aio_sendfile" reads
426 some data with readahead, then fails to write all data, and when the
427 socket is ready the next time, the data in the cache is already
428 lost, forcing "aio_sendfile" to again hit the disk. Explicit
429 "aio_read" + "aio_write" let's you better control resource usage.
430
431 This call tries to make use of a native "sendfile"-like syscall to
432 provide zero-copy operation. For this to work, $out_fh should refer
433 to a socket, and $in_fh should refer to an mmap'able file.
434
435 If a native sendfile cannot be found or it fails with "ENOSYS",
436 "EINVAL", "ENOTSUP", "EOPNOTSUPP", "EAFNOSUPPORT", "EPROTOTYPE" or
437 "ENOTSOCK", it will be emulated, so you can call "aio_sendfile" on
438 any type of filehandle regardless of the limitations of the
439 operating system.
440
441 As native sendfile syscalls (as practically any non-POSIX interface
442 hacked together in a hurry to improve benchmark numbers) tend to be
443 rather buggy on many systems, this implementation tries to work
444 around some known bugs in Linux and FreeBSD kernels (probably
445 others, too), but that might fail, so you really really should check
446 the return value of "aio_sendfile" - fewre bytes than expected might
447 have been transferred.
448
449 aio_readahead $fh,$offset,$length, $callback->($retval)
450 "aio_readahead" populates the page cache with data from a file so
451 that subsequent reads from that file will not block on disk I/O. The
452 $offset argument specifies the starting point from which data is to
453 be read and $length specifies the number of bytes to be read. I/O is
454 performed in whole pages, so that offset is effectively rounded down
455 to a page boundary and bytes are read up to the next page boundary
456 greater than or equal to (off-set+length). "aio_readahead" does not
457 read beyond the end of the file. The current file offset of the file
458 is left unchanged.
459
460 If that syscall doesn't exist (likely if your OS isn't Linux) it
461 will be emulated by simply reading the data, which would have a
462 similar effect.
463
464 aio_stat $fh_or_path, $callback->($status)
465 aio_lstat $fh, $callback->($status)
466 Works like perl's "stat" or "lstat" in void context. The callback
467 will be called after the stat and the results will be available
468 using "stat _" or "-s _" etc...
469
470 The pathname passed to "aio_stat" must be absolute. See API NOTES,
471 above, for an explanation.
472
473 Currently, the stats are always 64-bit-stats, i.e. instead of
474 returning an error when stat'ing a large file, the results will be
475 silently truncated unless perl itself is compiled with large file
476 support.
477
478 To help interpret the mode and dev/rdev stat values, IO::AIO offers
479 the following constants and functions (if not implemented, the
480 constants will be 0 and the functions will either "croak" or fall
481 back on traditional behaviour).
482
483 "S_IFMT", "S_IFIFO", "S_IFCHR", "S_IFBLK", "S_IFLNK", "S_IFREG",
484 "S_IFDIR", "S_IFWHT", "S_IFSOCK", "IO::AIO::major $dev_t",
485 "IO::AIO::minor $dev_t", "IO::AIO::makedev $major, $minor".
486
487 Example: Print the length of /etc/passwd:
488
489 aio_stat "/etc/passwd", sub {
490 $_[0] and die "stat failed: $!";
491 print "size is ", -s _, "\n";
492 };
493
494 aio_statvfs $fh_or_path, $callback->($statvfs)
495 Works like the POSIX "statvfs" or "fstatvfs" syscalls, depending on
496 whether a file handle or path was passed.
497
498 On success, the callback is passed a hash reference with the
499 following members: "bsize", "frsize", "blocks", "bfree", "bavail",
500 "files", "ffree", "favail", "fsid", "flag" and "namemax". On
501 failure, "undef" is passed.
502
503 The following POSIX IO::AIO::ST_* constants are defined: "ST_RDONLY"
504 and "ST_NOSUID".
505
506 The following non-POSIX IO::AIO::ST_* flag masks are defined to
507 their correct value when available, or to 0 on systems that do not
508 support them: "ST_NODEV", "ST_NOEXEC", "ST_SYNCHRONOUS",
509 "ST_MANDLOCK", "ST_WRITE", "ST_APPEND", "ST_IMMUTABLE",
510 "ST_NOATIME", "ST_NODIRATIME" and "ST_RELATIME".
511
512 Example: stat "/wd" and dump out the data if successful.
513
514 aio_statvfs "/wd", sub {
515 my $f = $_[0]
516 or die "statvfs: $!";
517
518 use Data::Dumper;
519 say Dumper $f;
520 };
521
522 # result:
523 {
524 bsize => 1024,
525 bfree => 4333064312,
526 blocks => 10253828096,
527 files => 2050765568,
528 flag => 4096,
529 favail => 2042092649,
530 bavail => 4333064312,
531 ffree => 2042092649,
532 namemax => 255,
533 frsize => 1024,
534 fsid => 1810
535 }
536
537 Here is a (likely partial - send me updates!) list of fsid values
538 used by Linux - it is safe to hardcode these when $^O is "linux":
539
540 0x0000adf5 adfs
541 0x0000adff affs
542 0x5346414f afs
543 0x09041934 anon-inode filesystem
544 0x00000187 autofs
545 0x42465331 befs
546 0x1badface bfs
547 0x42494e4d binfmt_misc
548 0x9123683e btrfs
549 0x0027e0eb cgroupfs
550 0xff534d42 cifs
551 0x73757245 coda
552 0x012ff7b7 coh
553 0x28cd3d45 cramfs
554 0x453dcd28 cramfs-wend (wrong endianness)
555 0x64626720 debugfs
556 0x00001373 devfs
557 0x00001cd1 devpts
558 0x0000f15f ecryptfs
559 0x00414a53 efs
560 0x0000137d ext
561 0x0000ef53 ext2/ext3/ext4
562 0x0000ef51 ext2
563 0xf2f52010 f2fs
564 0x00004006 fat
565 0x65735546 fuseblk
566 0x65735543 fusectl
567 0x0bad1dea futexfs
568 0x01161970 gfs2
569 0x47504653 gpfs
570 0x00004244 hfs
571 0xf995e849 hpfs
572 0x00c0ffee hostfs
573 0x958458f6 hugetlbfs
574 0x2bad1dea inotifyfs
575 0x00009660 isofs
576 0x000072b6 jffs2
577 0x3153464a jfs
578 0x6b414653 k-afs
579 0x0bd00bd0 lustre
580 0x0000137f minix
581 0x0000138f minix 30 char names
582 0x00002468 minix v2
583 0x00002478 minix v2 30 char names
584 0x00004d5a minix v3
585 0x19800202 mqueue
586 0x00004d44 msdos
587 0x0000564c novell
588 0x00006969 nfs
589 0x6e667364 nfsd
590 0x00003434 nilfs
591 0x5346544e ntfs
592 0x00009fa1 openprom
593 0x7461636F ocfs2
594 0x00009fa0 proc
595 0x6165676c pstorefs
596 0x0000002f qnx4
597 0x68191122 qnx6
598 0x858458f6 ramfs
599 0x52654973 reiserfs
600 0x00007275 romfs
601 0x67596969 rpc_pipefs
602 0x73636673 securityfs
603 0xf97cff8c selinux
604 0x0000517b smb
605 0x534f434b sockfs
606 0x73717368 squashfs
607 0x62656572 sysfs
608 0x012ff7b6 sysv2
609 0x012ff7b5 sysv4
610 0x01021994 tmpfs
611 0x15013346 udf
612 0x00011954 ufs
613 0x54190100 ufs byteswapped
614 0x00009fa2 usbdevfs
615 0x01021997 v9fs
616 0xa501fcf5 vxfs
617 0xabba1974 xenfs
618 0x012ff7b4 xenix
619 0x58465342 xfs
620 0x012fd16d xia
621
622 aio_utime $fh_or_path, $atime, $mtime, $callback->($status)
623 Works like perl's "utime" function (including the special case of
624 $atime and $mtime being undef). Fractional times are supported if
625 the underlying syscalls support them.
626
627 When called with a pathname, uses utimes(2) if available, otherwise
628 utime(2). If called on a file descriptor, uses futimes(2) if
629 available, otherwise returns ENOSYS, so this is not portable.
630
631 Examples:
632
633 # set atime and mtime to current time (basically touch(1)):
634 aio_utime "path", undef, undef;
635 # set atime to current time and mtime to beginning of the epoch:
636 aio_utime "path", time, undef; # undef==0
637
638 aio_chown $fh_or_path, $uid, $gid, $callback->($status)
639 Works like perl's "chown" function, except that "undef" for either
640 $uid or $gid is being interpreted as "do not change" (but -1 can
641 also be used).
642
643 Examples:
644
645 # same as "chown root path" in the shell:
646 aio_chown "path", 0, -1;
647 # same as above:
648 aio_chown "path", 0, undef;
649
650 aio_truncate $fh_or_path, $offset, $callback->($status)
651 Works like truncate(2) or ftruncate(2).
652
653 aio_allocate $fh, $mode, $offset, $len, $callback->($status)
654 Allocates or frees disk space according to the $mode argument. See
655 the linux "fallocate" documentation for details.
656
657 $mode is usually 0 or "IO::AIO::FALLOC_FL_KEEP_SIZE" to allocate
658 space, or "IO::AIO::FALLOC_FL_PUNCH_HOLE |
659 IO::AIO::FALLOC_FL_KEEP_SIZE", to deallocate a file range.
660
661 IO::AIO also supports "FALLOC_FL_COLLAPSE_RANGE", to remove a range
662 (without leaving a hole) and "FALLOC_FL_ZERO_RANGE", to zero a range
663 (see your fallocate(2) manpage).
664
665 The file system block size used by "fallocate" is presumably the
666 "f_bsize" returned by "statvfs".
667
668 If "fallocate" isn't available or cannot be emulated (currently no
669 emulation will be attempted), passes -1 and sets $! to "ENOSYS".
670
671 aio_chmod $fh_or_path, $mode, $callback->($status)
672 Works like perl's "chmod" function.
673
674 aio_unlink $pathname, $callback->($status)
675 Asynchronously unlink (delete) a file and call the callback with the
676 result code.
677
678 aio_mknod $pathname, $mode, $dev, $callback->($status)
679 [EXPERIMENTAL]
680
681 Asynchronously create a device node (or fifo). See mknod(2).
682
683 The only (POSIX-) portable way of calling this function is:
684
685 aio_mknod $pathname, IO::AIO::S_IFIFO | $mode, 0, sub { ...
686
687 See "aio_stat" for info about some potentially helpful extra
688 constants and functions.
689
690 aio_link $srcpath, $dstpath, $callback->($status)
691 Asynchronously create a new link to the existing object at $srcpath
692 at the path $dstpath and call the callback with the result code.
693
694 aio_symlink $srcpath, $dstpath, $callback->($status)
695 Asynchronously create a new symbolic link to the existing object at
696 $srcpath at the path $dstpath and call the callback with the result
697 code.
698
699 aio_readlink $pathname, $callback->($link)
700 Asynchronously read the symlink specified by $path and pass it to
701 the callback. If an error occurs, nothing or undef gets passed to
702 the callback.
703
704 aio_realpath $pathname, $callback->($path)
705 Asynchronously make the path absolute and resolve any symlinks in
706 $path. The resulting path only consists of directories (same as
707 Cwd::realpath).
708
709 This request can be used to get the absolute path of the current
710 working directory by passing it a path of . (a single dot).
711
712 aio_rename $srcpath, $dstpath, $callback->($status)
713 Asynchronously rename the object at $srcpath to $dstpath, just as
714 rename(2) and call the callback with the result code.
715
716 On systems that support the AIO::WD working directory abstraction
717 natively, the case "[$wd, "."]" as $srcpath is specialcased -
718 instead of failing, "rename" is called on the absolute path of $wd.
719
720 aio_mkdir $pathname, $mode, $callback->($status)
721 Asynchronously mkdir (create) a directory and call the callback with
722 the result code. $mode will be modified by the umask at the time the
723 request is executed, so do not change your umask.
724
725 aio_rmdir $pathname, $callback->($status)
726 Asynchronously rmdir (delete) a directory and call the callback with
727 the result code.
728
729 On systems that support the AIO::WD working directory abstraction
730 natively, the case "[$wd, "."]" is specialcased - instead of
731 failing, "rmdir" is called on the absolute path of $wd.
732
733 aio_readdir $pathname, $callback->($entries)
734 Unlike the POSIX call of the same name, "aio_readdir" reads an
735 entire directory (i.e. opendir + readdir + closedir). The entries
736 will not be sorted, and will NOT include the "." and ".." entries.
737
738 The callback is passed a single argument which is either "undef" or
739 an array-ref with the filenames.
740
741 aio_readdirx $pathname, $flags, $callback->($entries, $flags)
742 Quite similar to "aio_readdir", but the $flags argument allows one
743 to tune behaviour and output format. In case of an error, $entries
744 will be "undef".
745
746 The flags are a combination of the following constants, ORed
747 together (the flags will also be passed to the callback, possibly
748 modified):
749
750 IO::AIO::READDIR_DENTS
751 When this flag is off, then the callback gets an arrayref
752 consisting of names only (as with "aio_readdir"), otherwise it
753 gets an arrayref with "[$name, $type, $inode]" arrayrefs, each
754 describing a single directory entry in more detail.
755
756 $name is the name of the entry.
757
758 $type is one of the "IO::AIO::DT_xxx" constants:
759
760 "IO::AIO::DT_UNKNOWN", "IO::AIO::DT_FIFO", "IO::AIO::DT_CHR",
761 "IO::AIO::DT_DIR", "IO::AIO::DT_BLK", "IO::AIO::DT_REG",
762 "IO::AIO::DT_LNK", "IO::AIO::DT_SOCK", "IO::AIO::DT_WHT".
763
764 "IO::AIO::DT_UNKNOWN" means just that: readdir does not know. If
765 you need to know, you have to run stat yourself. Also, for speed
766 reasons, the $type scalars are read-only: you can not modify
767 them.
768
769 $inode is the inode number (which might not be exact on systems
770 with 64 bit inode numbers and 32 bit perls). This field has
771 unspecified content on systems that do not deliver the inode
772 information.
773
774 IO::AIO::READDIR_DIRS_FIRST
775 When this flag is set, then the names will be returned in an
776 order where likely directories come first, in optimal stat
777 order. This is useful when you need to quickly find directories,
778 or you want to find all directories while avoiding to stat()
779 each entry.
780
781 If the system returns type information in readdir, then this is
782 used to find directories directly. Otherwise, likely directories
783 are names beginning with ".", or otherwise names with no dots,
784 of which names with short names are tried first.
785
786 IO::AIO::READDIR_STAT_ORDER
787 When this flag is set, then the names will be returned in an
788 order suitable for stat()'ing each one. That is, when you plan
789 to stat() all files in the given directory, then the returned
790 order will likely be fastest.
791
792 If both this flag and "IO::AIO::READDIR_DIRS_FIRST" are
793 specified, then the likely dirs come first, resulting in a less
794 optimal stat order.
795
796 IO::AIO::READDIR_FOUND_UNKNOWN
797 This flag should not be set when calling "aio_readdirx".
798 Instead, it is being set by "aio_readdirx", when any of the
799 $type's found were "IO::AIO::DT_UNKNOWN". The absence of this
800 flag therefore indicates that all $type's are known, which can
801 be used to speed up some algorithms.
802
803 aio_load $pathname, $data, $callback->($status)
804 This is a composite request that tries to fully load the given file
805 into memory. Status is the same as with aio_read.
806
807 aio_copy $srcpath, $dstpath, $callback->($status)
808 Try to copy the *file* (directories not supported as either source
809 or destination) from $srcpath to $dstpath and call the callback with
810 a status of 0 (ok) or -1 (error, see $!).
811
812 This is a composite request that creates the destination file with
813 mode 0200 and copies the contents of the source file into it using
814 "aio_sendfile", followed by restoring atime, mtime, access mode and
815 uid/gid, in that order.
816
817 If an error occurs, the partial destination file will be unlinked,
818 if possible, except when setting atime, mtime, access mode and
819 uid/gid, where errors are being ignored.
820
821 aio_move $srcpath, $dstpath, $callback->($status)
822 Try to move the *file* (directories not supported as either source
823 or destination) from $srcpath to $dstpath and call the callback with
824 a status of 0 (ok) or -1 (error, see $!).
825
826 This is a composite request that tries to rename(2) the file first;
827 if rename fails with "EXDEV", it copies the file with "aio_copy"
828 and, if that is successful, unlinks the $srcpath.
829
830 aio_scandir $pathname, $maxreq, $callback->($dirs, $nondirs)
831 Scans a directory (similar to "aio_readdir") but additionally tries
832 to efficiently separate the entries of directory $path into two sets
833 of names, directories you can recurse into (directories), and ones
834 you cannot recurse into (everything else, including symlinks to
835 directories).
836
837 "aio_scandir" is a composite request that creates of many sub
838 requests_ $maxreq specifies the maximum number of outstanding aio
839 requests that this function generates. If it is "<= 0", then a
840 suitable default will be chosen (currently 4).
841
842 On error, the callback is called without arguments, otherwise it
843 receives two array-refs with path-relative entry names.
844
845 Example:
846
847 aio_scandir $dir, 0, sub {
848 my ($dirs, $nondirs) = @_;
849 print "real directories: @$dirs\n";
850 print "everything else: @$nondirs\n";
851 };
852
853 Implementation notes.
854
855 The "aio_readdir" cannot be avoided, but "stat()"'ing every entry
856 can.
857
858 If readdir returns file type information, then this is used directly
859 to find directories.
860
861 Otherwise, after reading the directory, the modification time, size
862 etc. of the directory before and after the readdir is checked, and
863 if they match (and isn't the current time), the link count will be
864 used to decide how many entries are directories (if >= 2).
865 Otherwise, no knowledge of the number of subdirectories will be
866 assumed.
867
868 Then entries will be sorted into likely directories a non-initial
869 dot currently) and likely non-directories (see "aio_readdirx"). Then
870 every entry plus an appended "/." will be "stat"'ed, likely
871 directories first, in order of their inode numbers. If that
872 succeeds, it assumes that the entry is a directory or a symlink to
873 directory (which will be checked separately). This is often faster
874 than stat'ing the entry itself because filesystems might detect the
875 type of the entry without reading the inode data (e.g. ext2fs
876 filetype feature), even on systems that cannot return the filetype
877 information on readdir.
878
879 If the known number of directories (link count - 2) has been
880 reached, the rest of the entries is assumed to be non-directories.
881
882 This only works with certainty on POSIX (= UNIX) filesystems, which
883 fortunately are the vast majority of filesystems around.
884
885 It will also likely work on non-POSIX filesystems with reduced
886 efficiency as those tend to return 0 or 1 as link counts, which
887 disables the directory counting heuristic.
888
889 aio_rmtree $pathname, $callback->($status)
890 Delete a directory tree starting (and including) $path, return the
891 status of the final "rmdir" only. This is a composite request that
892 uses "aio_scandir" to recurse into and rmdir directories, and unlink
893 everything else.
894
895 aio_fcntl $fh, $cmd, $arg, $callback->($status)
896 aio_ioctl $fh, $request, $buf, $callback->($status)
897 These work just like the "fcntl" and "ioctl" built-in functions,
898 except they execute asynchronously and pass the return value to the
899 callback.
900
901 Both calls can be used for a lot of things, some of which make more
902 sense to run asynchronously in their own thread, while some others
903 make less sense. For example, calls that block waiting for external
904 events, such as locking, will also lock down an I/O thread while it
905 is waiting, which can deadlock the whole I/O system. At the same
906 time, there might be no alternative to using a thread to wait.
907
908 So in general, you should only use these calls for things that do
909 (filesystem) I/O, not for things that wait for other events
910 (network, other processes), although if you are careful and know
911 what you are doing, you still can.
912
913 aio_sync $callback->($status)
914 Asynchronously call sync and call the callback when finished.
915
916 aio_fsync $fh, $callback->($status)
917 Asynchronously call fsync on the given filehandle and call the
918 callback with the fsync result code.
919
920 aio_fdatasync $fh, $callback->($status)
921 Asynchronously call fdatasync on the given filehandle and call the
922 callback with the fdatasync result code.
923
924 If this call isn't available because your OS lacks it or it couldn't
925 be detected, it will be emulated by calling "fsync" instead.
926
927 aio_syncfs $fh, $callback->($status)
928 Asynchronously call the syncfs syscall to sync the filesystem
929 associated to the given filehandle and call the callback with the
930 syncfs result code. If syncfs is not available, calls sync(), but
931 returns -1 and sets errno to "ENOSYS" nevertheless.
932
933 aio_sync_file_range $fh, $offset, $nbytes, $flags, $callback->($status)
934 Sync the data portion of the file specified by $offset and $length
935 to disk (but NOT the metadata), by calling the Linux-specific
936 sync_file_range call. If sync_file_range is not available or it
937 returns ENOSYS, then fdatasync or fsync is being substituted.
938
939 $flags can be a combination of
940 "IO::AIO::SYNC_FILE_RANGE_WAIT_BEFORE",
941 "IO::AIO::SYNC_FILE_RANGE_WRITE" and
942 "IO::AIO::SYNC_FILE_RANGE_WAIT_AFTER": refer to the sync_file_range
943 manpage for details.
944
945 aio_pathsync $pathname, $callback->($status)
946 This request tries to open, fsync and close the given path. This is
947 a composite request intended to sync directories after directory
948 operations (E.g. rename). This might not work on all operating
949 systems or have any specific effect, but usually it makes sure that
950 directory changes get written to disc. It works for anything that
951 can be opened for read-only, not just directories.
952
953 Future versions of this function might fall back to other methods
954 when "fsync" on the directory fails (such as calling "sync").
955
956 Passes 0 when everything went ok, and -1 on error.
957
958 aio_msync $scalar, $offset = 0, $length = undef, flags = 0,
959 $callback->($status)
960 This is a rather advanced IO::AIO call, which only works on
961 mmap(2)ed scalars (see the "IO::AIO::mmap" function, although it
962 also works on data scalars managed by the Sys::Mmap or Mmap modules,
963 note that the scalar must only be modified in-place while an aio
964 operation is pending on it).
965
966 It calls the "msync" function of your OS, if available, with the
967 memory area starting at $offset in the string and ending $length
968 bytes later. If $length is negative, counts from the end, and if
969 $length is "undef", then it goes till the end of the string. The
970 flags can be a combination of "IO::AIO::MS_ASYNC",
971 "IO::AIO::MS_INVALIDATE" and "IO::AIO::MS_SYNC".
972
973 aio_mtouch $scalar, $offset = 0, $length = undef, flags = 0,
974 $callback->($status)
975 This is a rather advanced IO::AIO call, which works best on
976 mmap(2)ed scalars.
977
978 It touches (reads or writes) all memory pages in the specified range
979 inside the scalar. All caveats and parameters are the same as for
980 "aio_msync", above, except for flags, which must be either 0 (which
981 reads all pages and ensures they are instantiated) or
982 "IO::AIO::MT_MODIFY", which modifies the memory pages (by reading
983 and writing an octet from it, which dirties the page).
984
985 aio_mlock $scalar, $offset = 0, $length = undef, $callback->($status)
986 This is a rather advanced IO::AIO call, which works best on
987 mmap(2)ed scalars.
988
989 It reads in all the pages of the underlying storage into memory (if
990 any) and locks them, so they are not getting swapped/paged out or
991 removed.
992
993 If $length is undefined, then the scalar will be locked till the
994 end.
995
996 On systems that do not implement "mlock", this function returns -1
997 and sets errno to "ENOSYS".
998
999 Note that the corresponding "munlock" is synchronous and is
1000 documented under "MISCELLANEOUS FUNCTIONS".
1001
1002 Example: open a file, mmap and mlock it - both will be undone when
1003 $data gets destroyed.
1004
1005 open my $fh, "<", $path or die "$path: $!";
1006 my $data;
1007 IO::AIO::mmap $data, -s $fh, IO::AIO::PROT_READ, IO::AIO::MAP_SHARED, $fh;
1008 aio_mlock $data; # mlock in background
1009
1010 aio_mlockall $flags, $callback->($status)
1011 Calls the "mlockall" function with the given $flags (a combination
1012 of "IO::AIO::MCL_CURRENT" and "IO::AIO::MCL_FUTURE").
1013
1014 On systems that do not implement "mlockall", this function returns
1015 -1 and sets errno to "ENOSYS".
1016
1017 Note that the corresponding "munlockall" is synchronous and is
1018 documented under "MISCELLANEOUS FUNCTIONS".
1019
1020 Example: asynchronously lock all current and future pages into
1021 memory.
1022
1023 aio_mlockall IO::AIO::MCL_FUTURE;
1024
1025 aio_fiemap $fh, $start, $length, $flags, $count, $cb->(\@extents)
1026 Queries the extents of the given file (by calling the Linux "FIEMAP"
1027 ioctl, see <http://cvs.schmorp.de/IO-AIO/doc/fiemap.txt> for
1028 details). If the ioctl is not available on your OS, then this
1029 request will fail with "ENOSYS".
1030
1031 $start is the starting offset to query extents for, $length is the
1032 size of the range to query - if it is "undef", then the whole file
1033 will be queried.
1034
1035 $flags is a combination of flags ("IO::AIO::FIEMAP_FLAG_SYNC" or
1036 "IO::AIO::FIEMAP_FLAG_XATTR" - "IO::AIO::FIEMAP_FLAGS_COMPAT" is
1037 also exported), and is normally 0 or "IO::AIO::FIEMAP_FLAG_SYNC" to
1038 query the data portion.
1039
1040 $count is the maximum number of extent records to return. If it is
1041 "undef", then IO::AIO queries all extents of the range. As a very
1042 special case, if it is 0, then the callback receives the number of
1043 extents instead of the extents themselves (which is unreliable, see
1044 below).
1045
1046 If an error occurs, the callback receives no arguments. The special
1047 "errno" value "IO::AIO::EBADR" is available to test for flag errors.
1048
1049 Otherwise, the callback receives an array reference with extent
1050 structures. Each extent structure is an array reference itself, with
1051 the following members:
1052
1053 [$logical, $physical, $length, $flags]
1054
1055 Flags is any combination of the following flag values (typically
1056 either 0 or "IO::AIO::FIEMAP_EXTENT_LAST" (1)):
1057
1058 "IO::AIO::FIEMAP_EXTENT_LAST", "IO::AIO::FIEMAP_EXTENT_UNKNOWN",
1059 "IO::AIO::FIEMAP_EXTENT_DELALLOC", "IO::AIO::FIEMAP_EXTENT_ENCODED",
1060 "IO::AIO::FIEMAP_EXTENT_DATA_ENCRYPTED",
1061 "IO::AIO::FIEMAP_EXTENT_NOT_ALIGNED",
1062 "IO::AIO::FIEMAP_EXTENT_DATA_INLINE",
1063 "IO::AIO::FIEMAP_EXTENT_DATA_TAIL",
1064 "IO::AIO::FIEMAP_EXTENT_UNWRITTEN", "IO::AIO::FIEMAP_EXTENT_MERGED"
1065 or "IO::AIO::FIEMAP_EXTENT_SHARED".
1066
1067 At the time of this writing (Linux 3.2), this requets is unreliable
1068 unless $count is "undef", as the kernel has all sorts of bugs
1069 preventing it to return all extents of a range for files with large
1070 number of extents. The code works around all these issues if $count
1071 is undef.
1072
1073 aio_group $callback->(...)
1074 This is a very special aio request: Instead of doing something, it
1075 is a container for other aio requests, which is useful if you want
1076 to bundle many requests into a single, composite, request with a
1077 definite callback and the ability to cancel the whole request with
1078 its subrequests.
1079
1080 Returns an object of class IO::AIO::GRP. See its documentation below
1081 for more info.
1082
1083 Example:
1084
1085 my $grp = aio_group sub {
1086 print "all stats done\n";
1087 };
1088
1089 add $grp
1090 (aio_stat ...),
1091 (aio_stat ...),
1092 ...;
1093
1094 aio_nop $callback->()
1095 This is a special request - it does nothing in itself and is only
1096 used for side effects, such as when you want to add a dummy request
1097 to a group so that finishing the requests in the group depends on
1098 executing the given code.
1099
1100 While this request does nothing, it still goes through the execution
1101 phase and still requires a worker thread. Thus, the callback will
1102 not be executed immediately but only after other requests in the
1103 queue have entered their execution phase. This can be used to
1104 measure request latency.
1105
1106 IO::AIO::aio_busy $fractional_seconds, $callback->() *NOT EXPORTED*
1107 Mainly used for debugging and benchmarking, this aio request puts
1108 one of the request workers to sleep for the given time.
1109
1110 While it is theoretically handy to have simple I/O scheduling
1111 requests like sleep and file handle readable/writable, the overhead
1112 this creates is immense (it blocks a thread for a long time) so do
1113 not use this function except to put your application under
1114 artificial I/O pressure.
1115
1116 IO::AIO::WD - multiple working directories
1117 Your process only has one current working directory, which is used by
1118 all threads. This makes it hard to use relative paths (some other
1119 component could call "chdir" at any time, and it is hard to control when
1120 the path will be used by IO::AIO).
1121
1122 One solution for this is to always use absolute paths. This usually
1123 works, but can be quite slow (the kernel has to walk the whole path on
1124 every access), and can also be a hassle to implement.
1125
1126 Newer POSIX systems have a number of functions (openat, fdopendir,
1127 futimensat and so on) that make it possible to specify working
1128 directories per operation.
1129
1130 For portability, and because the clowns who "designed", or shall I
1131 write, perpetrated this new interface were obviously half-drunk, this
1132 abstraction cannot be perfect, though.
1133
1134 IO::AIO allows you to convert directory paths into a so-called
1135 IO::AIO::WD object. This object stores the canonicalised, absolute
1136 version of the path, and on systems that allow it, also a directory file
1137 descriptor.
1138
1139 Everywhere where a pathname is accepted by IO::AIO (e.g. in "aio_stat"
1140 or "aio_unlink"), one can specify an array reference with an IO::AIO::WD
1141 object and a pathname instead (or the IO::AIO::WD object alone, which
1142 gets interpreted as "[$wd, "."]"). If the pathname is absolute, the
1143 IO::AIO::WD object is ignored, otherwise the pathname is resolved
1144 relative to that IO::AIO::WD object.
1145
1146 For example, to get a wd object for /etc and then stat passwd inside,
1147 you would write:
1148
1149 aio_wd "/etc", sub {
1150 my $etcdir = shift;
1151
1152 # although $etcdir can be undef on error, there is generally no reason
1153 # to check for errors here, as aio_stat will fail with ENOENT
1154 # when $etcdir is undef.
1155
1156 aio_stat [$etcdir, "passwd"], sub {
1157 # yay
1158 };
1159 };
1160
1161 The fact that "aio_wd" is a request and not a normal function shows that
1162 creating an IO::AIO::WD object is itself a potentially blocking
1163 operation, which is why it is done asynchronously.
1164
1165 To stat the directory obtained with "aio_wd" above, one could write
1166 either of the following three request calls:
1167
1168 aio_lstat "/etc" , sub { ... # pathname as normal string
1169 aio_lstat [$wd, "."], sub { ... # "." relative to $wd (i.e. $wd itself)
1170 aio_lstat $wd , sub { ... # shorthand for the previous
1171
1172 As with normal pathnames, IO::AIO keeps a copy of the working directory
1173 object and the pathname string, so you could write the following without
1174 causing any issues due to $path getting reused:
1175
1176 my $path = [$wd, undef];
1177
1178 for my $name (qw(abc def ghi)) {
1179 $path->[1] = $name;
1180 aio_stat $path, sub {
1181 # ...
1182 };
1183 }
1184
1185 There are some caveats: when directories get renamed (or deleted), the
1186 pathname string doesn't change, so will point to the new directory (or
1187 nowhere at all), while the directory fd, if available on the system,
1188 will still point to the original directory. Most functions accepting a
1189 pathname will use the directory fd on newer systems, and the string on
1190 older systems. Some functions (such as realpath) will always rely on the
1191 string form of the pathname.
1192
1193 So this functionality is mainly useful to get some protection against
1194 "chdir", to easily get an absolute path out of a relative path for
1195 future reference, and to speed up doing many operations in the same
1196 directory (e.g. when stat'ing all files in a directory).
1197
1198 The following functions implement this working directory abstraction:
1199
1200 aio_wd $pathname, $callback->($wd)
1201 Asynchonously canonicalise the given pathname and convert it to an
1202 IO::AIO::WD object representing it. If possible and supported on the
1203 system, also open a directory fd to speed up pathname resolution
1204 relative to this working directory.
1205
1206 If something goes wrong, then "undef" is passwd to the callback
1207 instead of a working directory object and $! is set appropriately.
1208 Since passing "undef" as working directory component of a pathname
1209 fails the request with "ENOENT", there is often no need for error
1210 checking in the "aio_wd" callback, as future requests using the
1211 value will fail in the expected way.
1212
1213 IO::AIO::CWD
1214 This is a compiletime constant (object) that represents the process
1215 current working directory.
1216
1217 Specifying this object as working directory object for a pathname is
1218 as if the pathname would be specified directly, without a directory
1219 object. For example, these calls are functionally identical:
1220
1221 aio_stat "somefile", sub { ... };
1222 aio_stat [IO::AIO::CWD, "somefile"], sub { ... };
1223
1224 To recover the path associated with an IO::AIO::WD object, you can use
1225 "aio_realpath":
1226
1227 aio_realpath $wd, sub {
1228 warn "path is $_[0]\n";
1229 };
1230
1231 Currently, "aio_statvfs" always, and "aio_rename" and "aio_rmdir"
1232 sometimes, fall back to using an absolue path.
1233
1234 IO::AIO::REQ CLASS
1235 All non-aggregate "aio_*" functions return an object of this class when
1236 called in non-void context.
1237
1238 cancel $req
1239 Cancels the request, if possible. Has the effect of skipping
1240 execution when entering the execute state and skipping calling the
1241 callback when entering the the result state, but will leave the
1242 request otherwise untouched (with the exception of readdir). That
1243 means that requests that currently execute will not be stopped and
1244 resources held by the request will not be freed prematurely.
1245
1246 cb $req $callback->(...)
1247 Replace (or simply set) the callback registered to the request.
1248
1249 IO::AIO::GRP CLASS
1250 This class is a subclass of IO::AIO::REQ, so all its methods apply to
1251 objects of this class, too.
1252
1253 A IO::AIO::GRP object is a special request that can contain multiple
1254 other aio requests.
1255
1256 You create one by calling the "aio_group" constructing function with a
1257 callback that will be called when all contained requests have entered
1258 the "done" state:
1259
1260 my $grp = aio_group sub {
1261 print "all requests are done\n";
1262 };
1263
1264 You add requests by calling the "add" method with one or more
1265 "IO::AIO::REQ" objects:
1266
1267 $grp->add (aio_unlink "...");
1268
1269 add $grp aio_stat "...", sub {
1270 $_[0] or return $grp->result ("error");
1271
1272 # add another request dynamically, if first succeeded
1273 add $grp aio_open "...", sub {
1274 $grp->result ("ok");
1275 };
1276 };
1277
573 This makes it very easy to create composite requests (see the source 1278 This makes it very easy to create composite requests (see the source of
574 of "aio_move" for an application) that work and feel like simple 1279 "aio_move" for an application) that work and feel like simple requests.
575 requests.
576 1280
577 * The IO::AIO::GRP objects will be cleaned up during calls to 1281 * The IO::AIO::GRP objects will be cleaned up during calls to
578 "IO::AIO::poll_cb", just like any other request. 1282 "IO::AIO::poll_cb", just like any other request.
1283
579 * They can be canceled like any other request. Canceling will cancel 1284 * They can be canceled like any other request. Canceling will cancel
580 not only the request itself, but also all requests it contains. 1285 not only the request itself, but also all requests it contains.
1286
581 * They can also can also be added to other IO::AIO::GRP objects. 1287 * They can also can also be added to other IO::AIO::GRP objects.
1288
582 * You must not add requests to a group from within the group 1289 * You must not add requests to a group from within the group callback
583 callback (or any later time). 1290 (or any later time).
584 1291
585 Their lifetime, simplified, looks like this: when they are empty, 1292 Their lifetime, simplified, looks like this: when they are empty, they
586 they will finish very quickly. If they contain only requests that 1293 will finish very quickly. If they contain only requests that are in the
587 are in the "done" state, they will also finish. Otherwise they will 1294 "done" state, they will also finish. Otherwise they will continue to
588 continue to exist. 1295 exist.
589 1296
590 That means after creating a group you have some time to add 1297 That means after creating a group you have some time to add requests
1298 (precisely before the callback has been invoked, which is only done
591 requests. And in the callbacks of those requests, you can add 1299 within the "poll_cb"). And in the callbacks of those requests, you can
592 further requests to the group. And only when all those requests have 1300 add further requests to the group. And only when all those requests have
593 finished will the the group itself finish. 1301 finished will the the group itself finish.
594 1302
595 add $grp ... 1303 add $grp ...
596 $grp->add (...) 1304 $grp->add (...)
597 Add one or more requests to the group. Any type of IO::AIO::REQ 1305 Add one or more requests to the group. Any type of IO::AIO::REQ can
598 can be added, including other groups, as long as you do not 1306 be added, including other groups, as long as you do not create
599 create circular dependencies. 1307 circular dependencies.
600 1308
601 Returns all its arguments. 1309 Returns all its arguments.
602 1310
603 $grp->cancel_subs 1311 $grp->cancel_subs
604 Cancel all subrequests and clears any feeder, but not the group 1312 Cancel all subrequests and clears any feeder, but not the group
605 request itself. Useful when you queued a lot of events but got a 1313 request itself. Useful when you queued a lot of events but got a
606 result early. 1314 result early.
607 1315
1316 The group request will finish normally (you cannot add requests to
1317 the group).
1318
608 $grp->result (...) 1319 $grp->result (...)
609 Set the result value(s) that will be passed to the group 1320 Set the result value(s) that will be passed to the group callback
610 callback when all subrequests have finished and set thre groups 1321 when all subrequests have finished and set the groups errno to the
611 errno to the current value of errno (just like calling "errno" 1322 current value of errno (just like calling "errno" without an error
612 without an error number). By default, no argument will be passed 1323 number). By default, no argument will be passed and errno is zero.
613 and errno is zero.
614 1324
615 $grp->errno ([$errno]) 1325 $grp->errno ([$errno])
616 Sets the group errno value to $errno, or the current value of 1326 Sets the group errno value to $errno, or the current value of errno
617 errno when the argument is missing. 1327 when the argument is missing.
618 1328
619 Every aio request has an associated errno value that is restored 1329 Every aio request has an associated errno value that is restored
620 when the callback is invoked. This method lets you change this 1330 when the callback is invoked. This method lets you change this value
621 value from its default (0). 1331 from its default (0).
622 1332
623 Calling "result" will also set errno, so make sure you either 1333 Calling "result" will also set errno, so make sure you either set $!
624 set $! before the call to "result", or call c<errno> after it. 1334 before the call to "result", or call c<errno> after it.
625 1335
626 feed $grp $callback->($grp) 1336 feed $grp $callback->($grp)
627 Sets a feeder/generator on this group: every group can have an 1337 Sets a feeder/generator on this group: every group can have an
628 attached generator that generates requests if idle. The idea 1338 attached generator that generates requests if idle. The idea behind
629 behind this is that, although you could just queue as many 1339 this is that, although you could just queue as many requests as you
630 requests as you want in a group, this might starve other 1340 want in a group, this might starve other requests for a potentially
631 requests for a potentially long time. For example, "aio_scandir" 1341 long time. For example, "aio_scandir" might generate hundreds of
632 might generate hundreds of thousands "aio_stat" requests, 1342 thousands of "aio_stat" requests, delaying any later requests for a
633 delaying any later requests for a long time. 1343 long time.
634 1344
635 To avoid this, and allow incremental generation of requests, you 1345 To avoid this, and allow incremental generation of requests, you can
636 can instead a group and set a feeder on it that generates those 1346 instead a group and set a feeder on it that generates those
637 requests. The feed callback will be called whenever there are 1347 requests. The feed callback will be called whenever there are few
638 few enough (see "limit", below) requests active in the group 1348 enough (see "limit", below) requests active in the group itself and
639 itself and is expected to queue more requests. 1349 is expected to queue more requests.
640 1350
641 The feed callback can queue as many requests as it likes (i.e. 1351 The feed callback can queue as many requests as it likes (i.e. "add"
642 "add" does not impose any limits). 1352 does not impose any limits).
643 1353
644 If the feed does not queue more requests when called, it will be 1354 If the feed does not queue more requests when called, it will be
645 automatically removed from the group. 1355 automatically removed from the group.
646 1356
647 If the feed limit is 0, it will be set to 2 automatically. 1357 If the feed limit is 0 when this method is called, it will be set to
1358 2 automatically.
648 1359
649 Example: 1360 Example:
650 1361
651 # stat all files in @files, but only ever use four aio requests concurrently: 1362 # stat all files in @files, but only ever use four aio requests concurrently:
652 1363
653 my $grp = aio_group sub { print "finished\n" }; 1364 my $grp = aio_group sub { print "finished\n" };
654 limit $grp 4; 1365 limit $grp 4;
655 feed $grp sub { 1366 feed $grp sub {
656 my $file = pop @files 1367 my $file = pop @files
657 or return; 1368 or return;
658 1369
659 add $grp aio_stat $file, sub { ... }; 1370 add $grp aio_stat $file, sub { ... };
660 }; 1371 };
661 1372
662 limit $grp $num 1373 limit $grp $num
663 Sets the feeder limit for the group: The feeder will be called 1374 Sets the feeder limit for the group: The feeder will be called
664 whenever the group contains less than this many requests. 1375 whenever the group contains less than this many requests.
665 1376
666 Setting the limit to 0 will pause the feeding process. 1377 Setting the limit to 0 will pause the feeding process.
1378
1379 The default value for the limit is 0, but note that setting a feeder
1380 automatically bumps it up to 2.
667 1381
668 SUPPORT FUNCTIONS 1382 SUPPORT FUNCTIONS
669 EVENT PROCESSING AND EVENT LOOP INTEGRATION 1383 EVENT PROCESSING AND EVENT LOOP INTEGRATION
670 $fileno = IO::AIO::poll_fileno 1384 $fileno = IO::AIO::poll_fileno
671 Return the *request result pipe file descriptor*. This 1385 Return the *request result pipe file descriptor*. This filehandle
672 filehandle must be polled for reading by some mechanism outside 1386 must be polled for reading by some mechanism outside this module
673 this module (e.g. Event or select, see below or the SYNOPSIS). 1387 (e.g. EV, Glib, select and so on, see below or the SYNOPSIS). If the
674 If the pipe becomes readable you have to call "poll_cb" to check 1388 pipe becomes readable you have to call "poll_cb" to check the
675 the results. 1389 results.
676 1390
677 See "poll_cb" for an example. 1391 See "poll_cb" for an example.
678 1392
679 IO::AIO::poll_cb 1393 IO::AIO::poll_cb
680 Process some outstanding events on the result pipe. You have to 1394 Process some requests that have reached the result phase (i.e. they
681 call this regularly. Returns the number of events processed. 1395 have been executed but the results are not yet reported). You have
1396 to call this "regularly" to finish outstanding requests.
1397
1398 Returns 0 if all events could be processed (or there were no events
1399 to process), or -1 if it returned earlier for whatever reason.
682 Returns immediately when no events are outstanding. The amount 1400 Returns immediately when no events are outstanding. The amount of
683 of events processed depends on the settings of 1401 events processed depends on the settings of "IO::AIO::max_poll_req",
684 "IO::AIO::max_poll_req" and "IO::AIO::max_poll_time". 1402 "IO::AIO::max_poll_time" and "IO::AIO::max_outstanding".
685 1403
686 If not all requests were processed for whatever reason, the 1404 If not all requests were processed for whatever reason, the poll
687 filehandle will still be ready when "poll_cb" returns. 1405 file descriptor will still be ready when "poll_cb" returns, so
1406 normally you don't have to do anything special to have it called
1407 later.
688 1408
1409 Apart from calling "IO::AIO::poll_cb" when the event filehandle
1410 becomes ready, it can be beneficial to call this function from loops
1411 which submit a lot of requests, to make sure the results get
1412 processed when they become available and not just when the loop is
1413 finished and the event loop takes over again. This function returns
1414 very fast when there are no outstanding requests.
1415
689 Example: Install an Event watcher that automatically calls 1416 Example: Install an Event watcher that automatically calls
690 IO::AIO::poll_cb with high priority: 1417 IO::AIO::poll_cb with high priority (more examples can be found in
1418 the SYNOPSIS section, at the top of this document):
691 1419
692 Event->io (fd => IO::AIO::poll_fileno, 1420 Event->io (fd => IO::AIO::poll_fileno,
693 poll => 'r', async => 1, 1421 poll => 'r', async => 1,
694 cb => \&IO::AIO::poll_cb); 1422 cb => \&IO::AIO::poll_cb);
695 1423
1424 IO::AIO::poll_wait
1425 Wait until either at least one request is in the result phase or no
1426 requests are outstanding anymore.
1427
1428 This is useful if you want to synchronously wait for some requests
1429 to become ready, without actually handling them.
1430
1431 See "nreqs" for an example.
1432
1433 IO::AIO::poll
1434 Waits until some requests have been handled.
1435
1436 Returns the number of requests processed, but is otherwise strictly
1437 equivalent to:
1438
1439 IO::AIO::poll_wait, IO::AIO::poll_cb
1440
1441 IO::AIO::flush
1442 Wait till all outstanding AIO requests have been handled.
1443
1444 Strictly equivalent to:
1445
1446 IO::AIO::poll_wait, IO::AIO::poll_cb
1447 while IO::AIO::nreqs;
1448
696 IO::AIO::max_poll_reqs $nreqs 1449 IO::AIO::max_poll_reqs $nreqs
697 IO::AIO::max_poll_time $seconds 1450 IO::AIO::max_poll_time $seconds
698 These set the maximum number of requests (default 0, meaning 1451 These set the maximum number of requests (default 0, meaning
699 infinity) that are being processed by "IO::AIO::poll_cb" in one 1452 infinity) that are being processed by "IO::AIO::poll_cb" in one
700 call, respectively the maximum amount of time (default 0, 1453 call, respectively the maximum amount of time (default 0, meaning
701 meaning infinity) spent in "IO::AIO::poll_cb" to process 1454 infinity) spent in "IO::AIO::poll_cb" to process requests (more
702 requests (more correctly the mininum amount of time "poll_cb" is 1455 correctly the mininum amount of time "poll_cb" is allowed to use).
703 allowed to use).
704 1456
1457 Setting "max_poll_time" to a non-zero value creates an overhead of
1458 one syscall per request processed, which is not normally a problem
1459 unless your callbacks are really really fast or your OS is really
1460 really slow (I am not mentioning Solaris here). Using
1461 "max_poll_reqs" incurs no overhead.
1462
705 Setting these is useful if you want to ensure some level of 1463 Setting these is useful if you want to ensure some level of
706 interactiveness when perl is not fast enough to process all 1464 interactiveness when perl is not fast enough to process all requests
707 requests in time. 1465 in time.
708 1466
709 For interactive programs, values such as 0.01 to 0.1 should be 1467 For interactive programs, values such as 0.01 to 0.1 should be fine.
710 fine.
711 1468
712 Example: Install an Event watcher that automatically calls 1469 Example: Install an Event watcher that automatically calls
713 IO::AIO::poll_some with low priority, to ensure that other parts 1470 IO::AIO::poll_cb with low priority, to ensure that other parts of
714 of the program get the CPU sometimes even under high AIO load. 1471 the program get the CPU sometimes even under high AIO load.
715 1472
716 # try not to spend much more than 0.1s in poll_cb 1473 # try not to spend much more than 0.1s in poll_cb
717 IO::AIO::max_poll_time 0.1; 1474 IO::AIO::max_poll_time 0.1;
718 1475
719 # use a low priority so other tasks have priority 1476 # use a low priority so other tasks have priority
720 Event->io (fd => IO::AIO::poll_fileno, 1477 Event->io (fd => IO::AIO::poll_fileno,
721 poll => 'r', nice => 1, 1478 poll => 'r', nice => 1,
722 cb => &IO::AIO::poll_cb); 1479 cb => &IO::AIO::poll_cb);
723
724 IO::AIO::poll_wait
725 Wait till the result filehandle becomes ready for reading
726 (simply does a "select" on the filehandle. This is useful if you
727 want to synchronously wait for some requests to finish).
728
729 See "nreqs" for an example.
730
731 IO::AIO::poll
732 Waits until some requests have been handled.
733
734 Strictly equivalent to:
735
736 IO::AIO::poll_wait, IO::AIO::poll_cb
737 if IO::AIO::nreqs;
738
739 IO::AIO::flush
740 Wait till all outstanding AIO requests have been handled.
741
742 Strictly equivalent to:
743
744 IO::AIO::poll_wait, IO::AIO::poll_cb
745 while IO::AIO::nreqs;
746 1480
747 CONTROLLING THE NUMBER OF THREADS 1481 CONTROLLING THE NUMBER OF THREADS
748 IO::AIO::min_parallel $nthreads 1482 IO::AIO::min_parallel $nthreads
749 Set the minimum number of AIO threads to $nthreads. The current 1483 Set the minimum number of AIO threads to $nthreads. The current
750 default is 8, which means eight asynchronous operations can 1484 default is 8, which means eight asynchronous operations can execute
751 execute concurrently at any one time (the number of outstanding 1485 concurrently at any one time (the number of outstanding requests,
752 requests, however, is unlimited). 1486 however, is unlimited).
753 1487
754 IO::AIO starts threads only on demand, when an AIO request is 1488 IO::AIO starts threads only on demand, when an AIO request is queued
755 queued and no free thread exists. Please note that queueing up a 1489 and no free thread exists. Please note that queueing up a hundred
756 hundred requests can create demand for a hundred threads, even 1490 requests can create demand for a hundred threads, even if it turns
757 if it turns out that everything is in the cache and could have 1491 out that everything is in the cache and could have been processed
758 been processed faster by a single thread. 1492 faster by a single thread.
759 1493
760 It is recommended to keep the number of threads relatively low, 1494 It is recommended to keep the number of threads relatively low, as
761 as some Linux kernel versions will scale negatively with the 1495 some Linux kernel versions will scale negatively with the number of
762 number of threads (higher parallelity => MUCH higher latency). 1496 threads (higher parallelity => MUCH higher latency). With current
763 With current Linux 2.6 versions, 4-32 threads should be fine. 1497 Linux 2.6 versions, 4-32 threads should be fine.
764 1498
765 Under most circumstances you don't need to call this function, 1499 Under most circumstances you don't need to call this function, as
766 as the module selects a default that is suitable for low to 1500 the module selects a default that is suitable for low to moderate
767 moderate load. 1501 load.
768 1502
769 IO::AIO::max_parallel $nthreads 1503 IO::AIO::max_parallel $nthreads
770 Sets the maximum number of AIO threads to $nthreads. If more 1504 Sets the maximum number of AIO threads to $nthreads. If more than
771 than the specified number of threads are currently running, this 1505 the specified number of threads are currently running, this function
772 function kills them. This function blocks until the limit is 1506 kills them. This function blocks until the limit is reached.
773 reached.
774 1507
775 While $nthreads are zero, aio requests get queued but not 1508 While $nthreads are zero, aio requests get queued but not executed
776 executed until the number of threads has been increased again. 1509 until the number of threads has been increased again.
777 1510
778 This module automatically runs "max_parallel 0" at program end, 1511 This module automatically runs "max_parallel 0" at program end, to
779 to ensure that all threads are killed and that there are no 1512 ensure that all threads are killed and that there are no outstanding
780 outstanding requests. 1513 requests.
781 1514
782 Under normal circumstances you don't need to call this function. 1515 Under normal circumstances you don't need to call this function.
783 1516
784 IO::AIO::max_idle $nthreads 1517 IO::AIO::max_idle $nthreads
785 Limit the number of threads (default: 4) that are allowed to 1518 Limit the number of threads (default: 4) that are allowed to idle
786 idle (i.e., threads that did not get a request to process within 1519 (i.e., threads that did not get a request to process within the idle
787 10 seconds). That means if a thread becomes idle while $nthreads 1520 timeout (default: 10 seconds). That means if a thread becomes idle
788 other threads are also idle, it will free its resources and 1521 while $nthreads other threads are also idle, it will free its
789 exit. 1522 resources and exit.
790 1523
791 This is useful when you allow a large number of threads (e.g. 1524 This is useful when you allow a large number of threads (e.g. 100 or
792 100 or 1000) to allow for extremely high load situations, but 1525 1000) to allow for extremely high load situations, but want to free
793 want to free resources under normal circumstances (1000 threads 1526 resources under normal circumstances (1000 threads can easily
794 can easily consume 30MB of RAM). 1527 consume 30MB of RAM).
795 1528
796 The default is probably ok in most situations, especially if 1529 The default is probably ok in most situations, especially if thread
797 thread creation is fast. If thread creation is very slow on your 1530 creation is fast. If thread creation is very slow on your system you
798 system you might want to use larger values. 1531 might want to use larger values.
799 1532
1533 IO::AIO::idle_timeout $seconds
1534 Sets the minimum idle timeout (default 10) after which worker
1535 threads are allowed to exit. SEe "IO::AIO::max_idle".
1536
800 $oldmaxreqs = IO::AIO::max_outstanding $maxreqs 1537 IO::AIO::max_outstanding $maxreqs
801 This is a very bad function to use in interactive programs
802 because it blocks, and a bad way to reduce concurrency because
803 it is inexact: Better use an "aio_group" together with a feed
804 callback.
805
806 Sets the maximum number of outstanding requests to $nreqs. If 1538 Sets the maximum number of outstanding requests to $nreqs. If you do
807 you to queue up more than this number of requests, the next call 1539 queue up more than this number of requests, the next call to
808 to the "poll_cb" (and "poll_some" and other functions calling 1540 "IO::AIO::poll_cb" (and other functions calling "poll_cb", such as
809 "poll_cb") function will block until the limit is no longer 1541 "IO::AIO::flush" or "IO::AIO::poll") will block until the limit is
810 exceeded. 1542 no longer exceeded.
811 1543
812 The default value is very large, so there is no practical limit 1544 In other words, this setting does not enforce a queue limit, but can
1545 be used to make poll functions block if the limit is exceeded.
1546
1547 This is a very bad function to use in interactive programs because
1548 it blocks, and a bad way to reduce concurrency because it is
1549 inexact: Better use an "aio_group" together with a feed callback.
1550
1551 Its main use is in scripts without an event loop - when you want to
1552 stat a lot of files, you can write somehting like this:
1553
1554 IO::AIO::max_outstanding 32;
1555
1556 for my $path (...) {
1557 aio_stat $path , ...;
1558 IO::AIO::poll_cb;
1559 }
1560
1561 IO::AIO::flush;
1562
1563 The call to "poll_cb" inside the loop will normally return
1564 instantly, but as soon as more thna 32 reqeusts are in-flight, it
1565 will block until some requests have been handled. This keeps the
1566 loop from pushing a large number of "aio_stat" requests onto the
1567 queue.
1568
1569 The default value for "max_outstanding" is very large, so there is
813 on the number of outstanding requests. 1570 no practical limit on the number of outstanding requests.
814
815 You can still queue as many requests as you want. Therefore,
816 "max_oustsanding" is mainly useful in simple scripts (with low
817 values) or as a stop gap to shield against fatal memory overflow
818 (with large values).
819 1571
820 STATISTICAL INFORMATION 1572 STATISTICAL INFORMATION
821 IO::AIO::nreqs 1573 IO::AIO::nreqs
822 Returns the number of requests currently in the ready, execute 1574 Returns the number of requests currently in the ready, execute or
823 or pending states (i.e. for which their callback has not been 1575 pending states (i.e. for which their callback has not been invoked
824 invoked yet). 1576 yet).
825 1577
826 Example: wait till there are no outstanding requests anymore: 1578 Example: wait till there are no outstanding requests anymore:
827 1579
828 IO::AIO::poll_wait, IO::AIO::poll_cb 1580 IO::AIO::poll_wait, IO::AIO::poll_cb
829 while IO::AIO::nreqs; 1581 while IO::AIO::nreqs;
830 1582
831 IO::AIO::nready 1583 IO::AIO::nready
832 Returns the number of requests currently in the ready state (not 1584 Returns the number of requests currently in the ready state (not yet
833 yet executed). 1585 executed).
834 1586
835 IO::AIO::npending 1587 IO::AIO::npending
836 Returns the number of requests currently in the pending state 1588 Returns the number of requests currently in the pending state
837 (executed, but not yet processed by poll_cb). 1589 (executed, but not yet processed by poll_cb).
1590
1591 MISCELLANEOUS FUNCTIONS
1592 IO::AIO implements some functions that are useful when you want to use
1593 some "Advanced I/O" function not available to in Perl, without going the
1594 "Asynchronous I/O" route. Many of these have an asynchronous "aio_*"
1595 counterpart.
1596
1597 IO::AIO::sendfile $ofh, $ifh, $offset, $count
1598 Calls the "eio_sendfile_sync" function, which is like
1599 "aio_sendfile", but is blocking (this makes most sense if you know
1600 the input data is likely cached already and the output filehandle is
1601 set to non-blocking operations).
1602
1603 Returns the number of bytes copied, or -1 on error.
1604
1605 IO::AIO::fadvise $fh, $offset, $len, $advice
1606 Simply calls the "posix_fadvise" function (see its manpage for
1607 details). The following advice constants are available:
1608 "IO::AIO::FADV_NORMAL", "IO::AIO::FADV_SEQUENTIAL",
1609 "IO::AIO::FADV_RANDOM", "IO::AIO::FADV_NOREUSE",
1610 "IO::AIO::FADV_WILLNEED", "IO::AIO::FADV_DONTNEED".
1611
1612 On systems that do not implement "posix_fadvise", this function
1613 returns ENOSYS, otherwise the return value of "posix_fadvise".
1614
1615 IO::AIO::madvise $scalar, $offset, $len, $advice
1616 Simply calls the "posix_madvise" function (see its manpage for
1617 details). The following advice constants are available:
1618 "IO::AIO::MADV_NORMAL", "IO::AIO::MADV_SEQUENTIAL",
1619 "IO::AIO::MADV_RANDOM", "IO::AIO::MADV_WILLNEED",
1620 "IO::AIO::MADV_DONTNEED".
1621
1622 On systems that do not implement "posix_madvise", this function
1623 returns ENOSYS, otherwise the return value of "posix_madvise".
1624
1625 IO::AIO::mprotect $scalar, $offset, $len, $protect
1626 Simply calls the "mprotect" function on the preferably AIO::mmap'ed
1627 $scalar (see its manpage for details). The following protect
1628 constants are available: "IO::AIO::PROT_NONE", "IO::AIO::PROT_READ",
1629 "IO::AIO::PROT_WRITE", "IO::AIO::PROT_EXEC".
1630
1631 On systems that do not implement "mprotect", this function returns
1632 ENOSYS, otherwise the return value of "mprotect".
1633
1634 IO::AIO::mmap $scalar, $length, $prot, $flags, $fh[, $offset]
1635 Memory-maps a file (or anonymous memory range) and attaches it to
1636 the given $scalar, which will act like a string scalar. Returns true
1637 on success, and false otherwise.
1638
1639 The only operations allowed on the scalar are "substr"/"vec" that
1640 don't change the string length, and most read-only operations such
1641 as copying it or searching it with regexes and so on.
1642
1643 Anything else is unsafe and will, at best, result in memory leaks.
1644
1645 The memory map associated with the $scalar is automatically removed
1646 when the $scalar is destroyed, or when the "IO::AIO::mmap" or
1647 "IO::AIO::munmap" functions are called.
1648
1649 This calls the "mmap"(2) function internally. See your system's
1650 manual page for details on the $length, $prot and $flags parameters.
1651
1652 The $length must be larger than zero and smaller than the actual
1653 filesize.
1654
1655 $prot is a combination of "IO::AIO::PROT_NONE",
1656 "IO::AIO::PROT_EXEC", "IO::AIO::PROT_READ" and/or
1657 "IO::AIO::PROT_WRITE",
1658
1659 $flags can be a combination of "IO::AIO::MAP_SHARED" or
1660 "IO::AIO::MAP_PRIVATE", or a number of system-specific flags (when
1661 not available, the are 0): "IO::AIO::MAP_ANONYMOUS" (which is set to
1662 "MAP_ANON" if your system only provides this constant),
1663 "IO::AIO::MAP_LOCKED", "IO::AIO::MAP_NORESERVE",
1664 "IO::AIO::MAP_POPULATE", "IO::AIO::MAP_NONBLOCK",
1665 "IO::AIO::MAP_FIXED", "IO::AIO::MAP_GROWSDOWN",
1666 "IO::AIO::MAP_32BIT", "IO::AIO::MAP_HUGETLB" or
1667 "IO::AIO::MAP_STACK".
1668
1669 If $fh is "undef", then a file descriptor of -1 is passed.
1670
1671 $offset is the offset from the start of the file - it generally must
1672 be a multiple of "IO::AIO::PAGESIZE" and defaults to 0.
1673
1674 Example:
1675
1676 use Digest::MD5;
1677 use IO::AIO;
1678
1679 open my $fh, "<verybigfile"
1680 or die "$!";
1681
1682 IO::AIO::mmap my $data, -s $fh, IO::AIO::PROT_READ, IO::AIO::MAP_SHARED, $fh
1683 or die "verybigfile: $!";
1684
1685 my $fast_md5 = md5 $data;
1686
1687 IO::AIO::munmap $scalar
1688 Removes a previous mmap and undefines the $scalar.
1689
1690 IO::AIO::munlock $scalar, $offset = 0, $length = undef
1691 Calls the "munlock" function, undoing the effects of a previous
1692 "aio_mlock" call (see its description for details).
1693
1694 IO::AIO::munlockall
1695 Calls the "munlockall" function.
1696
1697 On systems that do not implement "munlockall", this function returns
1698 ENOSYS, otherwise the return value of "munlockall".
1699
1700 IO::AIO::splice $r_fh, $r_off, $w_fh, $w_off, $length, $flags
1701 Calls the GNU/Linux splice(2) syscall, if available. If $r_off or
1702 $w_off are "undef", then "NULL" is passed for these, otherwise they
1703 should be the file offset.
1704
1705 $r_fh and $w_fh should not refer to the same file, as splice might
1706 silently corrupt the data in this case.
1707
1708 The following symbol flag values are available:
1709 "IO::AIO::SPLICE_F_MOVE", "IO::AIO::SPLICE_F_NONBLOCK",
1710 "IO::AIO::SPLICE_F_MORE" and "IO::AIO::SPLICE_F_GIFT".
1711
1712 See the splice(2) manpage for details.
1713
1714 IO::AIO::tee $r_fh, $w_fh, $length, $flags
1715 Calls the GNU/Linux tee(2) syscall, see its manpage and the
1716 description for "IO::AIO::splice" above for details.
1717
1718 $actual_size = IO::AIO::pipesize $r_fh[, $new_size]
1719 Attempts to query or change the pipe buffer size. Obviously works
1720 only on pipes, and currently works only on GNU/Linux systems, and
1721 fails with -1/"ENOSYS" everywhere else. If anybody knows how to
1722 influence pipe buffer size on other systems, drop me a note.
1723
1724 ($rfh, $wfh) = IO::AIO::pipe2 [$flags]
1725 This is a direct interface to the Linux pipe2(2) system call. If
1726 $flags is missing or 0, then this should be the same as a call to
1727 perl's built-in "pipe" function and create a new pipe, and works on
1728 systems that lack the pipe2 syscall. On win32, this case invokes
1729 "_pipe (..., 4096, O_BINARY)".
1730
1731 If $flags is non-zero, it tries to invoke the pipe2 system call with
1732 the given flags (Linux 2.6.27, glibc 2.9).
1733
1734 On success, the read and write file handles are returned.
1735
1736 On error, nothing will be returned. If the pipe2 syscall is missing
1737 and $flags is non-zero, fails with "ENOSYS".
1738
1739 Please refer to pipe2(2) for more info on the $flags, but at the
1740 time of this writing, "IO::AIO::O_CLOEXEC", "IO::AIO::O_NONBLOCK"
1741 and "IO::AIO::O_DIRECT" (Linux 3.4, for packet-based pipes) were
1742 supported.
1743
1744EVENT LOOP INTEGRATION
1745 It is recommended to use AnyEvent::AIO to integrate IO::AIO
1746 automatically into many event loops:
1747
1748 # AnyEvent integration (EV, Event, Glib, Tk, POE, urxvt, pureperl...)
1749 use AnyEvent::AIO;
1750
1751 You can also integrate IO::AIO manually into many event loops, here are
1752 some examples of how to do this:
1753
1754 # EV integration
1755 my $aio_w = EV::io IO::AIO::poll_fileno, EV::READ, \&IO::AIO::poll_cb;
1756
1757 # Event integration
1758 Event->io (fd => IO::AIO::poll_fileno,
1759 poll => 'r',
1760 cb => \&IO::AIO::poll_cb);
1761
1762 # Glib/Gtk2 integration
1763 add_watch Glib::IO IO::AIO::poll_fileno,
1764 in => sub { IO::AIO::poll_cb; 1 };
1765
1766 # Tk integration
1767 Tk::Event::IO->fileevent (IO::AIO::poll_fileno, "",
1768 readable => \&IO::AIO::poll_cb);
1769
1770 # Danga::Socket integration
1771 Danga::Socket->AddOtherFds (IO::AIO::poll_fileno =>
1772 \&IO::AIO::poll_cb);
838 1773
839 FORK BEHAVIOUR 1774 FORK BEHAVIOUR
840 This module should do "the right thing" when the process using it 1775 Usage of pthreads in a program changes the semantics of fork
841 forks: 1776 considerably. Specifically, only async-safe functions can be called
1777 after fork. Perl doesn't know about this, so in general, you cannot call
1778 fork with defined behaviour in perl if pthreads are involved. IO::AIO
1779 uses pthreads, so this applies, but many other extensions and (for
1780 inexplicable reasons) perl itself often is linked against pthreads, so
1781 this limitation applies to quite a lot of perls.
842 1782
843 Before the fork, IO::AIO enters a quiescent state where no requests 1783 This module no longer tries to fight your OS, or POSIX. That means
844 can be added in other threads and no results will be processed. 1784 IO::AIO only works in the process that loaded it. Forking is fully
845 After the fork the parent simply leaves the quiescent state and 1785 supported, but using IO::AIO in the child is not.
846 continues request/result processing, while the child frees the
847 request/result queue (so that the requests started before the fork
848 will only be handled in the parent). Threads will be started on
849 demand until the limit set in the parent process has been reached
850 again.
851 1786
852 In short: the parent will, after a short pause, continue as if fork 1787 You might get around by not *using* IO::AIO before (or after) forking.
853 had not been called, while the child will act as if IO::AIO has not 1788 You could also try to call the IO::AIO::reinit function in the child:
854 been used yet. 1789
1790 IO::AIO::reinit
1791 Abandons all current requests and I/O threads and simply
1792 reinitialises all data structures. This is not an operation
1793 supported by any standards, but happens to work on GNU/Linux and
1794 some newer BSD systems.
1795
1796 The only reasonable use for this function is to call it after
1797 forking, if "IO::AIO" was used in the parent. Calling it while
1798 IO::AIO is active in the process will result in undefined behaviour.
1799 Calling it at any time will also result in any undefined (by POSIX)
1800 behaviour.
855 1801
856 MEMORY USAGE 1802 MEMORY USAGE
857 Per-request usage: 1803 Per-request usage:
858 1804
859 Each aio request uses - depending on your architecture - around 1805 Each aio request uses - depending on your architecture - around 100-200
860 100-200 bytes of memory. In addition, stat requests need a stat 1806 bytes of memory. In addition, stat requests need a stat buffer (possibly
861 buffer (possibly a few hundred bytes), readdir requires a result 1807 a few hundred bytes), readdir requires a result buffer and so on. Perl
862 buffer and so on. Perl scalars and other data passed into aio 1808 scalars and other data passed into aio requests will also be locked and
863 requests will also be locked and will consume memory till the 1809 will consume memory till the request has entered the done state.
864 request has entered the done state.
865 1810
866 This is now awfully much, so queuing lots of requests is not usually 1811 This is not awfully much, so queuing lots of requests is not usually a
867 a problem. 1812 problem.
868 1813
869 Per-thread usage: 1814 Per-thread usage:
870 1815
871 In the execution phase, some aio requests require more memory for 1816 In the execution phase, some aio requests require more memory for
872 temporary buffers, and each thread requires a stack and other data 1817 temporary buffers, and each thread requires a stack and other data
873 structures (usually around 16k-128k, depending on the OS). 1818 structures (usually around 16k-128k, depending on the OS).
874 1819
875KNOWN BUGS 1820KNOWN BUGS
876 Known bugs will be fixed in the next release. 1821 Known bugs will be fixed in the next release.
877 1822
878SEE ALSO 1823SEE ALSO
879 Coro::AIO. 1824 AnyEvent::AIO for easy integration into event loops, Coro::AIO for a
1825 more natural syntax.
880 1826
881AUTHOR 1827AUTHOR
882 Marc Lehmann <schmorp@schmorp.de> 1828 Marc Lehmann <schmorp@schmorp.de>
883 http://home.schmorp.de/ 1829 http://home.schmorp.de/
884 1830

Diff Legend

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