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 1.29 by root, Wed Apr 16 16:45:30 2008 UTC

… …
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", 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 # AnyEvent integration (EV, Event, Glib, Tk, urxvt, pureperl...)
29 open my $fh, "<&=" . IO::AIO::poll_fileno or die "$!"; 30 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 my $w = AnyEvent->io (fh => $fh, poll => 'r', cb => sub { IO::AIO::poll_cb });
32
33 # EV integration
34 my $w = EV::io IO::AIO::poll_fileno, EV::READ, \&IO::AIO::poll_cb;
31 35
32 # Event integration 36 # Event integration
33 Event->io (fd => IO::AIO::poll_fileno, 37 Event->io (fd => IO::AIO::poll_fileno,
34 poll => 'r', 38 poll => 'r',
35 cb => \&IO::AIO::poll_cb); 39 cb => \&IO::AIO::poll_cb);
… …
58 but can also be used to easily do operations in parallel that are 62 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 63 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 64 faster on a RAID volume or over NFS when you do a number of stat
61 operations concurrently. 65 operations concurrently.
62 66
63 While this works on all types of file descriptors (for example sockets), 67 While most of this works on all types of file descriptors (for example
64 using these functions on file descriptors that support nonblocking 68 sockets), using these functions on file descriptors that support
65 operation (again, sockets, pipes etc.) is very inefficient. Use an event 69 nonblocking operation (again, sockets, pipes etc.) is very inefficient.
66 loop for that (such as the Event module): IO::AIO will naturally fit 70 Use an event loop for that (such as the Event module): IO::AIO will
67 into such an event loop itself. 71 naturally fit into such an event loop itself.
68 72
69 In this version, a number of threads are started that execute your 73 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 74 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 75 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 76 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 78 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 79 files currently, for example), and they would only support aio_read and
76 aio_write, so the remaining functionality would have to be implemented 80 aio_write, so the remaining functionality would have to be implemented
77 using threads anyway. 81 using threads anyway.
78 82
79 Although the module will work with in the presence of other (Perl-) 83 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 84 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 85 yourself, always call "poll_cb" from within the same thread, or never
82 never call "poll_cb" (or other "aio_" functions) recursively. 86 call "poll_cb" (or other "aio_" functions) recursively.
83 87
84 EXAMPLE 88 EXAMPLE
85 This is a simple example that uses the Event module and loads 89 This is a simple example that uses the Event module and loads
86 /etc/passwd asynchronously: 90 /etc/passwd asynchronously:
87 91
… …
94 poll => 'r', 98 poll => 'r',
95 cb => \&IO::AIO::poll_cb); 99 cb => \&IO::AIO::poll_cb);
96 100
97 # queue the request to open /etc/passwd 101 # queue the request to open /etc/passwd
98 aio_open "/etc/passwd", O_RDONLY, 0, sub { 102 aio_open "/etc/passwd", O_RDONLY, 0, sub {
99 my $fh = $_[0] 103 my $fh = shift
100 or die "error while opening: $!"; 104 or die "error while opening: $!";
101 105
102 # stat'ing filehandles is generally non-blocking 106 # stat'ing filehandles is generally non-blocking
103 my $size = -s $fh; 107 my $size = -s $fh;
104 108
… …
167 the actual aio request is severed and calling its methods will 171 the actual aio request is severed and calling its methods will
168 either do nothing or result in a runtime error). 172 either do nothing or result in a runtime error).
169 173
170FUNCTIONS 174FUNCTIONS
171 AIO REQUEST FUNCTIONS 175 AIO REQUEST FUNCTIONS
172 All the "aio_*" calls are more or less thin wrappers around the 176 All the "aio_*" calls are more or less thin wrappers around the syscall
173 syscall with the same name (sans "aio_"). The arguments are similar 177 with the same name (sans "aio_"). The arguments are similar or
174 or identical, and they all accept an additional (and optional) 178 identical, and they all accept an additional (and optional) $callback
175 $callback argument which must be a code reference. This code 179 argument which must be a code reference. This code reference will get
176 reference will get called with the syscall return code (e.g. most 180 called with the syscall return code (e.g. most syscalls return -1 on
177 syscalls return -1 on error, unlike perl, which usually delivers 181 error, unlike perl, which usually delivers "false") as it's sole
178 "false") as it's sole argument when the given syscall has been 182 argument when the given syscall has been executed asynchronously.
179 executed asynchronously.
180 183
181 All functions expecting a filehandle keep a copy of the filehandle 184 All functions expecting a filehandle keep a copy of the filehandle
182 internally until the request has finished. 185 internally until the request has finished.
183 186
184 All functions return request objects of type IO::AIO::REQ that allow 187 All functions return request objects of type IO::AIO::REQ that allow
185 further manipulation of those requests while they are in-flight. 188 further manipulation of those requests while they are in-flight.
186 189
187 The pathnames you pass to these routines *must* be absolute and 190 The pathnames you pass to these routines *must* be absolute and encoded
188 encoded as octets. The reason for the former is that at the time the 191 as octets. The reason for the former is that at the time the request is
189 request is being executed, the current working directory could have 192 being executed, the current working directory could have changed.
190 changed. Alternatively, you can make sure that you never change the 193 Alternatively, you can make sure that you never change the current
191 current working directory anywhere in the program and then use 194 working directory anywhere in the program and then use relative paths.
192 relative paths.
193 195
194 To encode pathnames as octets, either make sure you either: a) 196 To encode pathnames as octets, either make sure you either: a) always
195 always pass in filenames you got from outside (command line, readdir 197 pass in filenames you got from outside (command line, readdir etc.)
196 etc.) without tinkering, b) are ASCII or ISO 8859-1, c) use the 198 without tinkering, b) are ASCII or ISO 8859-1, c) use the Encode module
197 Encode module and encode your pathnames to the locale (or other) 199 and encode your pathnames to the locale (or other) encoding in effect in
198 encoding in effect in the user environment, d) use 200 the user environment, d) use Glib::filename_from_unicode on unicode
199 Glib::filename_from_unicode on unicode filenames or e) use something 201 filenames or e) use something else to ensure your scalar has the correct
200 else to ensure your scalar has the correct contents. 202 contents.
201 203
202 This works, btw. independent of the internal UTF-8 bit, which 204 This works, btw. independent of the internal UTF-8 bit, which IO::AIO
203 IO::AIO handles correctly wether it is set or not. 205 handles correctly wether it is set or not.
204 206
205 $prev_pri = aioreq_pri [$pri] 207 $prev_pri = aioreq_pri [$pri]
206 Returns the priority value that would be used for the next 208 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 209 and, if $pri is given, sets the priority for the next aio request.
208 aio request.
209 210
210 The default priority is 0, the minimum and maximum priorities 211 The default priority is 0, the minimum and maximum priorities are -4
211 are -4 and 4, respectively. Requests with higher priority will 212 and 4, respectively. Requests with higher priority will be serviced
212 be serviced first. 213 first.
213 214
214 The priority will be reset to 0 after each call to one of the 215 The priority will be reset to 0 after each call to one of the
215 "aio_*" functions. 216 "aio_*" functions.
216 217
217 Example: open a file with low priority, then read something from 218 Example: open a file with low priority, then read something from it
218 it with higher priority so the read request is serviced before 219 with higher priority so the read request is serviced before other
219 other low priority open requests (potentially spamming the 220 low priority open requests (potentially spamming the cache):
220 cache):
221 221
222 aioreq_pri -3;
223 aio_open ..., sub {
224 return unless $_[0];
225
222 aioreq_pri -3; 226 aioreq_pri -2;
223 aio_open ..., sub {
224 return unless $_[0];
225
226 aioreq_pri -2;
227 aio_read $_[0], ..., sub { 227 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 ...; 228 ...
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 }; 229 };
571 }; 230 };
572 231
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 a
238 newly created filehandle for the file.
239
240 The pathname passed to "aio_open" must be absolute. See API NOTES,
241 above, for an explanation.
242
243 The $flags argument is a bitmask. See the "Fcntl" module for a list.
244 They are the same as used by "sysopen".
245
246 Likewise, $mode specifies the mode of the newly created file, if it
247 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). Note that the $mode
250 will be modified by the umask in effect then the request is being
251 executed, so better never change the umask.
252
253 Example:
254
255 aio_open "/etc/passwd", O_RDONLY, 0, sub {
256 if ($_[0]) {
257 print "open successful, fh is $_[0]\n";
258 ...
259 } else {
260 die "open failed: $!\n";
261 }
262 };
263
264 aio_close $fh, $callback->($status)
265 Asynchronously close a file and call the callback with the result
266 code.
267
268 Unfortunately, you can't do this to perl. Perl *insists* very
269 strongly on closing the file descriptor associated with the
270 filehandle itself.
271
272 Therefore, "aio_close" will not close the filehandle - instead it
273 will use dup2 to overwrite the file descriptor with the write-end of
274 a pipe (the pipe fd will be created on demand and will be cached).
275
276 Or in other words: the file descriptor will be closed, but it will
277 not be free for reuse until the perl filehandle is closed.
278
279 aio_read $fh,$offset,$length, $data,$dataoffset, $callback->($retval)
280 aio_write $fh,$offset,$length, $data,$dataoffset, $callback->($retval)
281 Reads or writes $length bytes from the specified $fh and $offset
282 into the scalar given by $data and offset $dataoffset and calls the
283 callback without the actual number of bytes read (or -1 on error,
284 just like the syscall).
285
286 If $offset is undefined, then the current file descriptor offset
287 will be used (and updated), otherwise the file descriptor offset
288 will not be changed by these calls.
289
290 If $length is undefined in "aio_write", use the remaining length of
291 $data.
292
293 If $dataoffset is less than zero, it will be counted from the end of
294 $data.
295
296 The $data scalar *MUST NOT* be modified in any way while the request
297 is outstanding. Modifying it can result in segfaults or World War
298 III (if the necessary/optional hardware is installed).
299
300 Example: Read 15 bytes at offset 7 into scalar $buffer, starting at
301 offset 0 within the scalar:
302
303 aio_read $fh, 7, 15, $buffer, 0, sub {
304 $_[0] > 0 or die "read error: $!";
305 print "read $_[0] bytes: <$buffer>\n";
306 };
307
308 aio_sendfile $out_fh, $in_fh, $in_offset, $length, $callback->($retval)
309 Tries to copy $length bytes from $in_fh to $out_fh. It starts
310 reading at byte offset $in_offset, and starts writing at the current
311 file offset of $out_fh. Because of that, it is not safe to issue
312 more than one "aio_sendfile" per $out_fh, as they will interfere
313 with each other.
314
315 This call tries to make use of a native "sendfile" syscall to
316 provide zero-copy operation. For this to work, $out_fh should refer
317 to a socket, and $in_fh should refer to mmap'able file.
318
319 If the native sendfile call fails or is not implemented, it will be
320 emulated, so you can call "aio_sendfile" on any type of filehandle
321 regardless of the limitations of the operating system.
322
323 Please note, however, that "aio_sendfile" can read more bytes from
324 $in_fh than are written, and there is no way to find out how many
325 bytes have been read from "aio_sendfile" alone, as "aio_sendfile"
326 only provides the number of bytes written to $out_fh. Only if the
327 result value equals $length one can assume that $length bytes have
328 been read.
329
330 aio_readahead $fh,$offset,$length, $callback->($retval)
331 "aio_readahead" populates the page cache with data from a file so
332 that subsequent reads from that file will not block on disk I/O. The
333 $offset argument specifies the starting point from which data is to
334 be read and $length specifies the number of bytes to be read. I/O is
335 performed in whole pages, so that offset is effectively rounded down
336 to a page boundary and bytes are read up to the next page boundary
337 greater than or equal to (off-set+length). "aio_readahead" does not
338 read beyond the end of the file. The current file offset of the file
339 is left unchanged.
340
341 If that syscall doesn't exist (likely if your OS isn't Linux) it
342 will be emulated by simply reading the data, which would have a
343 similar effect.
344
345 aio_stat $fh_or_path, $callback->($status)
346 aio_lstat $fh, $callback->($status)
347 Works like perl's "stat" or "lstat" in void context. The callback
348 will be called after the stat and the results will be available
349 using "stat _" or "-s _" etc...
350
351 The pathname passed to "aio_stat" must be absolute. See API NOTES,
352 above, for an explanation.
353
354 Currently, the stats are always 64-bit-stats, i.e. instead of
355 returning an error when stat'ing a large file, the results will be
356 silently truncated unless perl itself is compiled with large file
357 support.
358
359 Example: Print the length of /etc/passwd:
360
361 aio_stat "/etc/passwd", sub {
362 $_[0] and die "stat failed: $!";
363 print "size is ", -s _, "\n";
364 };
365
366 aio_utime $fh_or_path, $atime, $mtime, $callback->($status)
367 Works like perl's "utime" function (including the special case of
368 $atime and $mtime being undef). Fractional times are supported if
369 the underlying syscalls support them.
370
371 When called with a pathname, uses utimes(2) if available, otherwise
372 utime(2). If called on a file descriptor, uses futimes(2) if
373 available, otherwise returns ENOSYS, so this is not portable.
374
375 Examples:
376
377 # set atime and mtime to current time (basically touch(1)):
378 aio_utime "path", undef, undef;
379 # set atime to current time and mtime to beginning of the epoch:
380 aio_utime "path", time, undef; # undef==0
381
382 aio_chown $fh_or_path, $uid, $gid, $callback->($status)
383 Works like perl's "chown" function, except that "undef" for either
384 $uid or $gid is being interpreted as "do not change" (but -1 can
385 also be used).
386
387 Examples:
388
389 # same as "chown root path" in the shell:
390 aio_chown "path", 0, -1;
391 # same as above:
392 aio_chown "path", 0, undef;
393
394 aio_truncate $fh_or_path, $offset, $callback->($status)
395 Works like truncate(2) or ftruncate(2).
396
397 aio_chmod $fh_or_path, $mode, $callback->($status)
398 Works like perl's "chmod" function.
399
400 aio_unlink $pathname, $callback->($status)
401 Asynchronously unlink (delete) a file and call the callback with the
402 result code.
403
404 aio_mknod $path, $mode, $dev, $callback->($status)
405 [EXPERIMENTAL]
406
407 Asynchronously create a device node (or fifo). See mknod(2).
408
409 The only (POSIX-) portable way of calling this function is:
410
411 aio_mknod $path, IO::AIO::S_IFIFO | $mode, 0, sub { ...
412
413 aio_link $srcpath, $dstpath, $callback->($status)
414 Asynchronously create a new link to the existing object at $srcpath
415 at the path $dstpath and call the callback with the result code.
416
417 aio_symlink $srcpath, $dstpath, $callback->($status)
418 Asynchronously create a new symbolic link to the existing object at
419 $srcpath at the path $dstpath and call the callback with the result
420 code.
421
422 aio_readlink $path, $callback->($link)
423 Asynchronously read the symlink specified by $path and pass it to
424 the callback. If an error occurs, nothing or undef gets passed to
425 the callback.
426
427 aio_rename $srcpath, $dstpath, $callback->($status)
428 Asynchronously rename the object at $srcpath to $dstpath, just as
429 rename(2) and call the callback with the result code.
430
431 aio_mkdir $pathname, $mode, $callback->($status)
432 Asynchronously mkdir (create) a directory and call the callback with
433 the result code. $mode will be modified by the umask at the time the
434 request is executed, so do not change your umask.
435
436 aio_rmdir $pathname, $callback->($status)
437 Asynchronously rmdir (delete) a directory and call the callback with
438 the result code.
439
440 aio_readdir $pathname, $callback->($entries)
441 Unlike the POSIX call of the same name, "aio_readdir" reads an
442 entire directory (i.e. opendir + readdir + closedir). The entries
443 will not be sorted, and will NOT include the "." and ".." entries.
444
445 The callback a single argument which is either "undef" or an
446 array-ref with the filenames.
447
448 aio_load $path, $data, $callback->($status)
449 This is a composite request that tries to fully load the given file
450 into memory. Status is the same as with aio_read.
451
452 aio_copy $srcpath, $dstpath, $callback->($status)
453 Try to copy the *file* (directories not supported as either source
454 or destination) from $srcpath to $dstpath and call the callback with
455 the 0 (error) or -1 ok.
456
457 This is a composite request that it creates the destination file
458 with mode 0200 and copies the contents of the source file into it
459 using "aio_sendfile", followed by restoring atime, mtime, access
460 mode and uid/gid, in that order.
461
462 If an error occurs, the partial destination file will be unlinked,
463 if possible, except when setting atime, mtime, access mode and
464 uid/gid, where errors are being ignored.
465
466 aio_move $srcpath, $dstpath, $callback->($status)
467 Try to move the *file* (directories not supported as either source
468 or destination) from $srcpath to $dstpath and call the callback with
469 the 0 (error) or -1 ok.
470
471 This is a composite request that tries to rename(2) the file first.
472 If rename files with "EXDEV", it copies the file with "aio_copy"
473 and, if that is successful, unlinking the $srcpath.
474
475 aio_scandir $path, $maxreq, $callback->($dirs, $nondirs)
476 Scans a directory (similar to "aio_readdir") but additionally tries
477 to efficiently separate the entries of directory $path into two sets
478 of names, directories you can recurse into (directories), and ones
479 you cannot recurse into (everything else, including symlinks to
480 directories).
481
482 "aio_scandir" is a composite request that creates of many sub
483 requests_ $maxreq specifies the maximum number of outstanding aio
484 requests that this function generates. If it is "<= 0", then a
485 suitable default will be chosen (currently 4).
486
487 On error, the callback is called without arguments, otherwise it
488 receives two array-refs with path-relative entry names.
489
490 Example:
491
492 aio_scandir $dir, 0, sub {
493 my ($dirs, $nondirs) = @_;
494 print "real directories: @$dirs\n";
495 print "everything else: @$nondirs\n";
496 };
497
498 Implementation notes.
499
500 The "aio_readdir" cannot be avoided, but "stat()"'ing every entry
501 can.
502
503 After reading the directory, the modification time, size etc. of the
504 directory before and after the readdir is checked, and if they match
505 (and isn't the current time), the link count will be used to decide
506 how many entries are directories (if >= 2). Otherwise, no knowledge
507 of the number of subdirectories will be assumed.
508
509 Then entries will be sorted into likely directories (everything
510 without a non-initial dot currently) and likely non-directories
511 (everything else). Then every entry plus an appended "/." will be
512 "stat"'ed, likely directories first. If that succeeds, it assumes
513 that the entry is a directory or a symlink to directory (which will
514 be checked seperately). This is often faster than stat'ing the entry
515 itself because filesystems might detect the type of the entry
516 without reading the inode data (e.g. ext2fs filetype feature).
517
518 If the known number of directories (link count - 2) has been
519 reached, the rest of the entries is assumed to be non-directories.
520
521 This only works with certainty on POSIX (= UNIX) filesystems, which
522 fortunately are the vast majority of filesystems around.
523
524 It will also likely work on non-POSIX filesystems with reduced
525 efficiency as those tend to return 0 or 1 as link counts, which
526 disables the directory counting heuristic.
527
528 aio_rmtree $path, $callback->($status)
529 Delete a directory tree starting (and including) $path, return the
530 status of the final "rmdir" only. This is a composite request that
531 uses "aio_scandir" to recurse into and rmdir directories, and unlink
532 everything else.
533
534 aio_sync $callback->($status)
535 Asynchronously call sync and call the callback when finished.
536
537 aio_fsync $fh, $callback->($status)
538 Asynchronously call fsync on the given filehandle and call the
539 callback with the fsync result code.
540
541 aio_fdatasync $fh, $callback->($status)
542 Asynchronously call fdatasync on the given filehandle and call the
543 callback with the fdatasync result code.
544
545 If this call isn't available because your OS lacks it or it couldn't
546 be detected, it will be emulated by calling "fsync" instead.
547
548 aio_pathsync $path, $callback->($status)
549 This request tries to open, fsync and close the given path. This is
550 a composite request intended tosync directories after directory
551 operations (E.g. rename). This might not work on all operating
552 systems or have any specific effect, but usually it makes sure that
553 directory changes get written to disc. It works for anything that
554 can be opened for read-only, not just directories.
555
556 Passes 0 when everything went ok, and -1 on error.
557
558 aio_group $callback->(...)
559 This is a very special aio request: Instead of doing something, it
560 is a container for other aio requests, which is useful if you want
561 to bundle many requests into a single, composite, request with a
562 definite callback and the ability to cancel the whole request with
563 its subrequests.
564
565 Returns an object of class IO::AIO::GRP. See its documentation below
566 for more info.
567
568 Example:
569
570 my $grp = aio_group sub {
571 print "all stats done\n";
572 };
573
574 add $grp
575 (aio_stat ...),
576 (aio_stat ...),
577 ...;
578
579 aio_nop $callback->()
580 This is a special request - it does nothing in itself and is only
581 used for side effects, such as when you want to add a dummy request
582 to a group so that finishing the requests in the group depends on
583 executing the given code.
584
585 While this request does nothing, it still goes through the execution
586 phase and still requires a worker thread. Thus, the callback will
587 not be executed immediately but only after other requests in the
588 queue have entered their execution phase. This can be used to
589 measure request latency.
590
591 IO::AIO::aio_busy $fractional_seconds, $callback->() *NOT EXPORTED*
592 Mainly used for debugging and benchmarking, this aio request puts
593 one of the request workers to sleep for the given time.
594
595 While it is theoretically handy to have simple I/O scheduling
596 requests like sleep and file handle readable/writable, the overhead
597 this creates is immense (it blocks a thread for a long time) so do
598 not use this function except to put your application under
599 artificial I/O pressure.
600
601 IO::AIO::REQ CLASS
602 All non-aggregate "aio_*" functions return an object of this class when
603 called in non-void context.
604
605 cancel $req
606 Cancels the request, if possible. Has the effect of skipping
607 execution when entering the execute state and skipping calling the
608 callback when entering the the result state, but will leave the
609 request otherwise untouched. That means that requests that currently
610 execute will not be stopped and resources held by the request will
611 not be freed prematurely.
612
613 cb $req $callback->(...)
614 Replace (or simply set) the callback registered to the request.
615
616 IO::AIO::GRP CLASS
617 This class is a subclass of IO::AIO::REQ, so all its methods apply to
618 objects of this class, too.
619
620 A IO::AIO::GRP object is a special request that can contain multiple
621 other aio requests.
622
623 You create one by calling the "aio_group" constructing function with a
624 callback that will be called when all contained requests have entered
625 the "done" state:
626
627 my $grp = aio_group sub {
628 print "all requests are done\n";
629 };
630
631 You add requests by calling the "add" method with one or more
632 "IO::AIO::REQ" objects:
633
634 $grp->add (aio_unlink "...");
635
636 add $grp aio_stat "...", sub {
637 $_[0] or return $grp->result ("error");
638
639 # add another request dynamically, if first succeeded
640 add $grp aio_open "...", sub {
641 $grp->result ("ok");
642 };
643 };
644
573 This makes it very easy to create composite requests (see the source 645 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 646 "aio_move" for an application) that work and feel like simple requests.
575 requests.
576 647
577 * The IO::AIO::GRP objects will be cleaned up during calls to 648 * The IO::AIO::GRP objects will be cleaned up during calls to
578 "IO::AIO::poll_cb", just like any other request. 649 "IO::AIO::poll_cb", just like any other request.
650
579 * They can be canceled like any other request. Canceling will cancel 651 * They can be canceled like any other request. Canceling will cancel
580 not only the request itself, but also all requests it contains. 652 not only the request itself, but also all requests it contains.
653
581 * They can also can also be added to other IO::AIO::GRP objects. 654 * They can also can also be added to other IO::AIO::GRP objects.
655
582 * You must not add requests to a group from within the group 656 * You must not add requests to a group from within the group callback
583 callback (or any later time). 657 (or any later time).
584 658
585 Their lifetime, simplified, looks like this: when they are empty, 659 Their lifetime, simplified, looks like this: when they are empty, they
586 they will finish very quickly. If they contain only requests that 660 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 661 "done" state, they will also finish. Otherwise they will continue to
588 continue to exist. 662 exist.
589 663
590 That means after creating a group you have some time to add 664 That means after creating a group you have some time to add requests.
591 requests. And in the callbacks of those requests, you can add 665 And in the callbacks of those requests, you can add further requests to
592 further requests to the group. And only when all those requests have 666 the group. And only when all those requests have finished will the the
593 finished will the the group itself finish. 667 group itself finish.
594 668
595 add $grp ... 669 add $grp ...
596 $grp->add (...) 670 $grp->add (...)
597 Add one or more requests to the group. Any type of IO::AIO::REQ 671 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 672 be added, including other groups, as long as you do not create
599 create circular dependencies. 673 circular dependencies.
600 674
601 Returns all its arguments. 675 Returns all its arguments.
602 676
603 $grp->cancel_subs 677 $grp->cancel_subs
604 Cancel all subrequests and clears any feeder, but not the group 678 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 679 request itself. Useful when you queued a lot of events but got a
606 result early. 680 result early.
607 681
608 $grp->result (...) 682 $grp->result (...)
609 Set the result value(s) that will be passed to the group 683 Set the result value(s) that will be passed to the group callback
610 callback when all subrequests have finished and set thre groups 684 when all subrequests have finished and set the groups errno to the
611 errno to the current value of errno (just like calling "errno" 685 current value of errno (just like calling "errno" without an error
612 without an error number). By default, no argument will be passed 686 number). By default, no argument will be passed and errno is zero.
613 and errno is zero.
614 687
615 $grp->errno ([$errno]) 688 $grp->errno ([$errno])
616 Sets the group errno value to $errno, or the current value of 689 Sets the group errno value to $errno, or the current value of errno
617 errno when the argument is missing. 690 when the argument is missing.
618 691
619 Every aio request has an associated errno value that is restored 692 Every aio request has an associated errno value that is restored
620 when the callback is invoked. This method lets you change this 693 when the callback is invoked. This method lets you change this value
621 value from its default (0). 694 from its default (0).
622 695
623 Calling "result" will also set errno, so make sure you either 696 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. 697 before the call to "result", or call c<errno> after it.
625 698
626 feed $grp $callback->($grp) 699 feed $grp $callback->($grp)
627 Sets a feeder/generator on this group: every group can have an 700 Sets a feeder/generator on this group: every group can have an
628 attached generator that generates requests if idle. The idea 701 attached generator that generates requests if idle. The idea behind
629 behind this is that, although you could just queue as many 702 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 703 want in a group, this might starve other requests for a potentially
631 requests for a potentially long time. For example, "aio_scandir" 704 long time. For example, "aio_scandir" might generate hundreds of
632 might generate hundreds of thousands "aio_stat" requests, 705 thousands "aio_stat" requests, delaying any later requests for a
633 delaying any later requests for a long time. 706 long time.
634 707
635 To avoid this, and allow incremental generation of requests, you 708 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 709 instead a group and set a feeder on it that generates those
637 requests. The feed callback will be called whenever there are 710 requests. The feed callback will be called whenever there are few
638 few enough (see "limit", below) requests active in the group 711 enough (see "limit", below) requests active in the group itself and
639 itself and is expected to queue more requests. 712 is expected to queue more requests.
640 713
641 The feed callback can queue as many requests as it likes (i.e. 714 The feed callback can queue as many requests as it likes (i.e. "add"
642 "add" does not impose any limits). 715 does not impose any limits).
643 716
644 If the feed does not queue more requests when called, it will be 717 If the feed does not queue more requests when called, it will be
645 automatically removed from the group. 718 automatically removed from the group.
646 719
647 If the feed limit is 0, it will be set to 2 automatically. 720 If the feed limit is 0, it will be set to 2 automatically.
648 721
649 Example: 722 Example:
650 723
651 # stat all files in @files, but only ever use four aio requests concurrently: 724 # stat all files in @files, but only ever use four aio requests concurrently:
652 725
653 my $grp = aio_group sub { print "finished\n" }; 726 my $grp = aio_group sub { print "finished\n" };
654 limit $grp 4; 727 limit $grp 4;
655 feed $grp sub { 728 feed $grp sub {
656 my $file = pop @files 729 my $file = pop @files
657 or return; 730 or return;
658 731
659 add $grp aio_stat $file, sub { ... }; 732 add $grp aio_stat $file, sub { ... };
660 }; 733 };
661 734
662 limit $grp $num 735 limit $grp $num
663 Sets the feeder limit for the group: The feeder will be called 736 Sets the feeder limit for the group: The feeder will be called
664 whenever the group contains less than this many requests. 737 whenever the group contains less than this many requests.
665 738
666 Setting the limit to 0 will pause the feeding process. 739 Setting the limit to 0 will pause the feeding process.
667 740
668 SUPPORT FUNCTIONS 741 SUPPORT FUNCTIONS
669 EVENT PROCESSING AND EVENT LOOP INTEGRATION 742 EVENT PROCESSING AND EVENT LOOP INTEGRATION
670 $fileno = IO::AIO::poll_fileno 743 $fileno = IO::AIO::poll_fileno
671 Return the *request result pipe file descriptor*. This 744 Return the *request result pipe file descriptor*. This filehandle
672 filehandle must be polled for reading by some mechanism outside 745 must be polled for reading by some mechanism outside this module
673 this module (e.g. Event or select, see below or the SYNOPSIS). 746 (e.g. Event or select, see below or the SYNOPSIS). If the pipe
674 If the pipe becomes readable you have to call "poll_cb" to check 747 becomes readable you have to call "poll_cb" to check the results.
675 the results.
676 748
677 See "poll_cb" for an example. 749 See "poll_cb" for an example.
678 750
679 IO::AIO::poll_cb 751 IO::AIO::poll_cb
680 Process some outstanding events on the result pipe. You have to 752 Process some outstanding events on the result pipe. You have to call
681 call this regularly. Returns the number of events processed. 753 this regularly. Returns the number of events processed. Returns
682 Returns immediately when no events are outstanding. The amount 754 immediately when no events are outstanding. The amount of events
683 of events processed depends on the settings of 755 processed depends on the settings of "IO::AIO::max_poll_req" and
684 "IO::AIO::max_poll_req" and "IO::AIO::max_poll_time". 756 "IO::AIO::max_poll_time".
685 757
686 If not all requests were processed for whatever reason, the 758 If not all requests were processed for whatever reason, the
687 filehandle will still be ready when "poll_cb" returns. 759 filehandle will still be ready when "poll_cb" returns.
688 760
689 Example: Install an Event watcher that automatically calls 761 Example: Install an Event watcher that automatically calls
690 IO::AIO::poll_cb with high priority: 762 IO::AIO::poll_cb with high priority:
691 763
692 Event->io (fd => IO::AIO::poll_fileno, 764 Event->io (fd => IO::AIO::poll_fileno,
693 poll => 'r', async => 1, 765 poll => 'r', async => 1,
694 cb => \&IO::AIO::poll_cb); 766 cb => \&IO::AIO::poll_cb);
695 767
696 IO::AIO::max_poll_reqs $nreqs 768 IO::AIO::max_poll_reqs $nreqs
697 IO::AIO::max_poll_time $seconds 769 IO::AIO::max_poll_time $seconds
698 These set the maximum number of requests (default 0, meaning 770 These set the maximum number of requests (default 0, meaning
699 infinity) that are being processed by "IO::AIO::poll_cb" in one 771 infinity) that are being processed by "IO::AIO::poll_cb" in one
700 call, respectively the maximum amount of time (default 0, 772 call, respectively the maximum amount of time (default 0, meaning
701 meaning infinity) spent in "IO::AIO::poll_cb" to process 773 infinity) spent in "IO::AIO::poll_cb" to process requests (more
702 requests (more correctly the mininum amount of time "poll_cb" is 774 correctly the mininum amount of time "poll_cb" is allowed to use).
703 allowed to use).
704 775
776 Setting "max_poll_time" to a non-zero value creates an overhead of
777 one syscall per request processed, which is not normally a problem
778 unless your callbacks are really really fast or your OS is really
779 really slow (I am not mentioning Solaris here). Using
780 "max_poll_reqs" incurs no overhead.
781
705 Setting these is useful if you want to ensure some level of 782 Setting these is useful if you want to ensure some level of
706 interactiveness when perl is not fast enough to process all 783 interactiveness when perl is not fast enough to process all requests
707 requests in time. 784 in time.
708 785
709 For interactive programs, values such as 0.01 to 0.1 should be 786 For interactive programs, values such as 0.01 to 0.1 should be fine.
710 fine.
711 787
712 Example: Install an Event watcher that automatically calls 788 Example: Install an Event watcher that automatically calls
713 IO::AIO::poll_some with low priority, to ensure that other parts 789 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. 790 the program get the CPU sometimes even under high AIO load.
715 791
716 # try not to spend much more than 0.1s in poll_cb 792 # try not to spend much more than 0.1s in poll_cb
717 IO::AIO::max_poll_time 0.1; 793 IO::AIO::max_poll_time 0.1;
718 794
719 # use a low priority so other tasks have priority 795 # use a low priority so other tasks have priority
720 Event->io (fd => IO::AIO::poll_fileno, 796 Event->io (fd => IO::AIO::poll_fileno,
721 poll => 'r', nice => 1, 797 poll => 'r', nice => 1,
722 cb => &IO::AIO::poll_cb); 798 cb => &IO::AIO::poll_cb);
723 799
724 IO::AIO::poll_wait 800 IO::AIO::poll_wait
801 If there are any outstanding requests and none of them in the result
725 Wait till the result filehandle becomes ready for reading 802 phase, wait till the result filehandle becomes ready for reading
726 (simply does a "select" on the filehandle. This is useful if you 803 (simply does a "select" on the filehandle. This is useful if you
727 want to synchronously wait for some requests to finish). 804 want to synchronously wait for some requests to finish).
728 805
729 See "nreqs" for an example. 806 See "nreqs" for an example.
730 807
731 IO::AIO::poll 808 IO::AIO::poll
732 Waits until some requests have been handled. 809 Waits until some requests have been handled.
733 810
811 Returns the number of requests processed, but is otherwise strictly
734 Strictly equivalent to: 812 equivalent to:
735 813
736 IO::AIO::poll_wait, IO::AIO::poll_cb 814 IO::AIO::poll_wait, IO::AIO::poll_cb
737 if IO::AIO::nreqs;
738 815
739 IO::AIO::flush 816 IO::AIO::flush
740 Wait till all outstanding AIO requests have been handled. 817 Wait till all outstanding AIO requests have been handled.
741 818
742 Strictly equivalent to: 819 Strictly equivalent to:
743 820
744 IO::AIO::poll_wait, IO::AIO::poll_cb 821 IO::AIO::poll_wait, IO::AIO::poll_cb
745 while IO::AIO::nreqs; 822 while IO::AIO::nreqs;
746 823
747 CONTROLLING THE NUMBER OF THREADS 824 CONTROLLING THE NUMBER OF THREADS
748 IO::AIO::min_parallel $nthreads 825 IO::AIO::min_parallel $nthreads
749 Set the minimum number of AIO threads to $nthreads. The current 826 Set the minimum number of AIO threads to $nthreads. The current
750 default is 8, which means eight asynchronous operations can 827 default is 8, which means eight asynchronous operations can execute
751 execute concurrently at any one time (the number of outstanding 828 concurrently at any one time (the number of outstanding requests,
752 requests, however, is unlimited). 829 however, is unlimited).
753 830
754 IO::AIO starts threads only on demand, when an AIO request is 831 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 832 and no free thread exists. Please note that queueing up a hundred
756 hundred requests can create demand for a hundred threads, even 833 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 834 out that everything is in the cache and could have been processed
758 been processed faster by a single thread. 835 faster by a single thread.
759 836
760 It is recommended to keep the number of threads relatively low, 837 It is recommended to keep the number of threads relatively low, as
761 as some Linux kernel versions will scale negatively with the 838 some Linux kernel versions will scale negatively with the number of
762 number of threads (higher parallelity => MUCH higher latency). 839 threads (higher parallelity => MUCH higher latency). With current
763 With current Linux 2.6 versions, 4-32 threads should be fine. 840 Linux 2.6 versions, 4-32 threads should be fine.
764 841
765 Under most circumstances you don't need to call this function, 842 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 843 the module selects a default that is suitable for low to moderate
767 moderate load. 844 load.
768 845
769 IO::AIO::max_parallel $nthreads 846 IO::AIO::max_parallel $nthreads
770 Sets the maximum number of AIO threads to $nthreads. If more 847 Sets the maximum number of AIO threads to $nthreads. If more than
771 than the specified number of threads are currently running, this 848 the specified number of threads are currently running, this function
772 function kills them. This function blocks until the limit is 849 kills them. This function blocks until the limit is reached.
773 reached.
774 850
775 While $nthreads are zero, aio requests get queued but not 851 While $nthreads are zero, aio requests get queued but not executed
776 executed until the number of threads has been increased again. 852 until the number of threads has been increased again.
777 853
778 This module automatically runs "max_parallel 0" at program end, 854 This module automatically runs "max_parallel 0" at program end, to
779 to ensure that all threads are killed and that there are no 855 ensure that all threads are killed and that there are no outstanding
780 outstanding requests. 856 requests.
781 857
782 Under normal circumstances you don't need to call this function. 858 Under normal circumstances you don't need to call this function.
783 859
784 IO::AIO::max_idle $nthreads 860 IO::AIO::max_idle $nthreads
785 Limit the number of threads (default: 4) that are allowed to 861 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 862 (i.e., threads that did not get a request to process within 10
787 10 seconds). That means if a thread becomes idle while $nthreads 863 seconds). That means if a thread becomes idle while $nthreads other
788 other threads are also idle, it will free its resources and 864 threads are also idle, it will free its resources and exit.
789 exit.
790 865
791 This is useful when you allow a large number of threads (e.g. 866 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 867 1000) to allow for extremely high load situations, but want to free
793 want to free resources under normal circumstances (1000 threads 868 resources under normal circumstances (1000 threads can easily
794 can easily consume 30MB of RAM). 869 consume 30MB of RAM).
795 870
796 The default is probably ok in most situations, especially if 871 The default is probably ok in most situations, especially if thread
797 thread creation is fast. If thread creation is very slow on your 872 creation is fast. If thread creation is very slow on your system you
798 system you might want to use larger values. 873 might want to use larger values.
799 874
800 $oldmaxreqs = IO::AIO::max_outstanding $maxreqs 875 $oldmaxreqs = IO::AIO::max_outstanding $maxreqs
801 This is a very bad function to use in interactive programs 876 This is a very bad function to use in interactive programs because
802 because it blocks, and a bad way to reduce concurrency because 877 it blocks, and a bad way to reduce concurrency because it is
803 it is inexact: Better use an "aio_group" together with a feed 878 inexact: Better use an "aio_group" together with a feed callback.
804 callback.
805 879
806 Sets the maximum number of outstanding requests to $nreqs. If 880 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 881 queue up more than this number of requests, the next call to the
808 to the "poll_cb" (and "poll_some" and other functions calling 882 "poll_cb" (and "poll_some" and other functions calling "poll_cb")
809 "poll_cb") function will block until the limit is no longer 883 function will block until the limit is no longer exceeded.
810 exceeded.
811 884
812 The default value is very large, so there is no practical limit 885 The default value is very large, so there is no practical limit on
813 on the number of outstanding requests. 886 the number of outstanding requests.
814 887
815 You can still queue as many requests as you want. Therefore, 888 You can still queue as many requests as you want. Therefore,
816 "max_oustsanding" is mainly useful in simple scripts (with low 889 "max_oustsanding" is mainly useful in simple scripts (with low
817 values) or as a stop gap to shield against fatal memory overflow 890 values) or as a stop gap to shield against fatal memory overflow
818 (with large values). 891 (with large values).
819 892
820 STATISTICAL INFORMATION 893 STATISTICAL INFORMATION
821 IO::AIO::nreqs 894 IO::AIO::nreqs
822 Returns the number of requests currently in the ready, execute 895 Returns the number of requests currently in the ready, execute or
823 or pending states (i.e. for which their callback has not been 896 pending states (i.e. for which their callback has not been invoked
824 invoked yet). 897 yet).
825 898
826 Example: wait till there are no outstanding requests anymore: 899 Example: wait till there are no outstanding requests anymore:
827 900
828 IO::AIO::poll_wait, IO::AIO::poll_cb 901 IO::AIO::poll_wait, IO::AIO::poll_cb
829 while IO::AIO::nreqs; 902 while IO::AIO::nreqs;
830 903
831 IO::AIO::nready 904 IO::AIO::nready
832 Returns the number of requests currently in the ready state (not 905 Returns the number of requests currently in the ready state (not yet
833 yet executed). 906 executed).
834 907
835 IO::AIO::npending 908 IO::AIO::npending
836 Returns the number of requests currently in the pending state 909 Returns the number of requests currently in the pending state
837 (executed, but not yet processed by poll_cb). 910 (executed, but not yet processed by poll_cb).
838 911
839 FORK BEHAVIOUR 912 FORK BEHAVIOUR
840 This module should do "the right thing" when the process using it 913 This module should do "the right thing" when the process using it forks:
841 forks:
842 914
843 Before the fork, IO::AIO enters a quiescent state where no requests 915 Before the fork, IO::AIO enters a quiescent state where no requests can
844 can be added in other threads and no results will be processed. 916 be added in other threads and no results will be processed. After the
845 After the fork the parent simply leaves the quiescent state and 917 fork the parent simply leaves the quiescent state and continues
846 continues request/result processing, while the child frees the 918 request/result processing, while the child frees the request/result
847 request/result queue (so that the requests started before the fork 919 queue (so that the requests started before the fork will only be handled
848 will only be handled in the parent). Threads will be started on 920 in the parent). Threads will be started on demand until the limit set in
849 demand until the limit set in the parent process has been reached 921 the parent process has been reached again.
850 again.
851 922
852 In short: the parent will, after a short pause, continue as if fork 923 In short: the parent will, after a short pause, continue as if fork had
853 had not been called, while the child will act as if IO::AIO has not 924 not been called, while the child will act as if IO::AIO has not been
854 been used yet. 925 used yet.
855 926
856 MEMORY USAGE 927 MEMORY USAGE
857 Per-request usage: 928 Per-request usage:
858 929
859 Each aio request uses - depending on your architecture - around 930 Each aio request uses - depending on your architecture - around 100-200
860 100-200 bytes of memory. In addition, stat requests need a stat 931 bytes of memory. In addition, stat requests need a stat buffer (possibly
861 buffer (possibly a few hundred bytes), readdir requires a result 932 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 933 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 934 will consume memory till the request has entered the done state.
864 request has entered the done state.
865 935
866 This is now awfully much, so queuing lots of requests is not usually 936 This is not awfully much, so queuing lots of requests is not usually a
867 a problem. 937 problem.
868 938
869 Per-thread usage: 939 Per-thread usage:
870 940
871 In the execution phase, some aio requests require more memory for 941 In the execution phase, some aio requests require more memory for
872 temporary buffers, and each thread requires a stack and other data 942 temporary buffers, and each thread requires a stack and other data
873 structures (usually around 16k-128k, depending on the OS). 943 structures (usually around 16k-128k, depending on the OS).
874 944
875KNOWN BUGS 945KNOWN BUGS
876 Known bugs will be fixed in the next release. 946 Known bugs will be fixed in the next release.
877 947
878SEE ALSO 948SEE ALSO
879 Coro::AIO. 949 Coro::AIO.
880 950
881AUTHOR 951AUTHOR
882 Marc Lehmann <schmorp@schmorp.de> 952 Marc Lehmann <schmorp@schmorp.de>
883 http://home.schmorp.de/ 953 http://home.schmorp.de/
884 954

Diff Legend

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