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

Comparing IO-AIO/AIO.pm (file contents):
Revision 1.55 by root, Sun Oct 22 00:49:29 2006 UTC vs.
Revision 1.86 by root, Sat Oct 28 23:32:29 2006 UTC

15 15
16 aio_read $fh, 30000, 1024, $buffer, 0, sub { 16 aio_read $fh, 30000, 1024, $buffer, 0, sub {
17 $_[0] > 0 or die "read error: $!"; 17 $_[0] > 0 or die "read error: $!";
18 }; 18 };
19 19
20 use IO::AIO 2; # version has aio objects 20 # version 2+ has request and group objects
21 use IO::AIO 2;
21 22
23 aioreq_pri 4; # give next request a very high priority
22 my $req = aio_unlink "/tmp/file", sub { }; 24 my $req = aio_unlink "/tmp/file", sub { };
23 $req->cancel; # cancel request if still in queue 25 $req->cancel; # cancel request if still in queue
24 26
25 # AnyEvent 27 my $grp = aio_group sub { print "all stats done\n" };
28 add $grp aio_stat "..." for ...;
29
30 # AnyEvent integration
26 open my $fh, "<&=" . IO::AIO::poll_fileno or die "$!"; 31 open my $fh, "<&=" . IO::AIO::poll_fileno or die "$!";
27 my $w = AnyEvent->io (fh => $fh, poll => 'r', cb => sub { IO::AIO::poll_cb }); 32 my $w = AnyEvent->io (fh => $fh, poll => 'r', cb => sub { IO::AIO::poll_cb });
28 33
29 # Event 34 # Event integration
30 Event->io (fd => IO::AIO::poll_fileno, 35 Event->io (fd => IO::AIO::poll_fileno,
31 poll => 'r', 36 poll => 'r',
32 cb => \&IO::AIO::poll_cb); 37 cb => \&IO::AIO::poll_cb);
33 38
34 # Glib/Gtk2 39 # Glib/Gtk2 integration
35 add_watch Glib::IO IO::AIO::poll_fileno, 40 add_watch Glib::IO IO::AIO::poll_fileno,
36 in => sub { IO::AIO::poll_cb; 1 }; 41 in => sub { IO::AIO::poll_cb; 1 };
37 42
38 # Tk 43 # Tk integration
39 Tk::Event::IO->fileevent (IO::AIO::poll_fileno, "", 44 Tk::Event::IO->fileevent (IO::AIO::poll_fileno, "",
40 readable => \&IO::AIO::poll_cb); 45 readable => \&IO::AIO::poll_cb);
41 46
42 # Danga::Socket 47 # Danga::Socket integration
43 Danga::Socket->AddOtherFds (IO::AIO::poll_fileno => 48 Danga::Socket->AddOtherFds (IO::AIO::poll_fileno =>
44 \&IO::AIO::poll_cb); 49 \&IO::AIO::poll_cb);
45 50
46
47=head1 DESCRIPTION 51=head1 DESCRIPTION
48 52
49This module implements asynchronous I/O using whatever means your 53This module implements asynchronous I/O using whatever means your
50operating system supports. 54operating system supports.
51 55
56Asynchronous means that operations that can normally block your program
57(e.g. reading from disk) will be done asynchronously: the operation
58will still block, but you can do something else in the meantime. This
59is extremely useful for programs that need to stay interactive even
60when doing heavy I/O (GUI programs, high performance network servers
61etc.), but can also be used to easily do operations in parallel that are
62normally done sequentially, e.g. stat'ing many files, which is much faster
63on a RAID volume or over NFS when you do a number of stat operations
64concurrently.
65
66While this works on all types of file descriptors (for example sockets),
67using these functions on file descriptors that support nonblocking
68operation (again, sockets, pipes etc.) is very inefficient. Use an event
69loop for that (such as the L<Event|Event> module): IO::AIO will naturally
70fit into such an event loop itself.
71
52Currently, a number of threads are started that execute your read/writes 72In this version, a number of threads are started that execute your
53and signal their completion. You don't need thread support in your libc or 73requests and signal their completion. You don't need thread support
54perl, and the threads created by this module will not be visible to the 74in perl, and the threads created by this module will not be visible
55pthreads library. In the future, this module might make use of the native 75to perl. In the future, this module might make use of the native aio
56aio functions available on many operating systems. However, they are often 76functions available on many operating systems. However, they are often
57not well-supported (Linux doesn't allow them on normal files currently, 77not well-supported or restricted (GNU/Linux doesn't allow them on normal
58for example), and they would only support aio_read and aio_write, so the 78files currently, for example), and they would only support aio_read and
59remaining functionality would have to be implemented using threads anyway. 79aio_write, so the remaining functionality would have to be implemented
80using threads anyway.
60 81
61Although the module will work with in the presence of other threads, it is 82Although the module will work with in the presence of other (Perl-)
62currently not reentrant, so use appropriate locking yourself, always call 83threads, it is currently not reentrant in any way, so use appropriate
63C<poll_cb> from within the same thread, or never call C<poll_cb> (or other 84locking yourself, always call C<poll_cb> from within the same thread, or
64C<aio_> functions) recursively. 85never call C<poll_cb> (or other C<aio_> functions) recursively.
86
87=head2 EXAMPLE
88
89This is a simple example that uses the Event module and loads
90F</etc/passwd> asynchronously:
91
92 use Fcntl;
93 use Event;
94 use IO::AIO;
95
96 # register the IO::AIO callback with Event
97 Event->io (fd => IO::AIO::poll_fileno,
98 poll => 'r',
99 cb => \&IO::AIO::poll_cb);
100
101 # queue the request to open /etc/passwd
102 aio_open "/etc/passwd", O_RDONLY, 0, sub {
103 my $fh = $_[0]
104 or die "error while opening: $!";
105
106 # stat'ing filehandles is generally non-blocking
107 my $size = -s $fh;
108
109 # queue a request to read the file
110 my $contents;
111 aio_read $fh, 0, $size, $contents, 0, sub {
112 $_[0] == $size
113 or die "short read: $!";
114
115 close $fh;
116
117 # file contents now in $contents
118 print $contents;
119
120 # exit event loop and program
121 Event::unloop;
122 };
123 };
124
125 # possibly queue up other requests, or open GUI windows,
126 # check for sockets etc. etc.
127
128 # process events as long as there are some:
129 Event::loop;
130
131=head1 REQUEST ANATOMY AND LIFETIME
132
133Every C<aio_*> function creates a request. which is a C data structure not
134directly visible to Perl.
135
136If called in non-void context, every request function returns a Perl
137object representing the request. In void context, nothing is returned,
138which saves a bit of memory.
139
140The perl object is a fairly standard ref-to-hash object. The hash contents
141are not used by IO::AIO so you are free to store anything you like in it.
142
143During their existance, aio requests travel through the following states,
144in order:
145
146=over 4
147
148=item ready
149
150Immediately after a request is created it is put into the ready state,
151waiting for a thread to execute it.
152
153=item execute
154
155A thread has accepted the request for processing and is currently
156executing it (e.g. blocking in read).
157
158=item pending
159
160The request has been executed and is waiting for result processing.
161
162While request submission and execution is fully asynchronous, result
163processing is not and relies on the perl interpreter calling C<poll_cb>
164(or another function with the same effect).
165
166=item result
167
168The request results are processed synchronously by C<poll_cb>.
169
170The C<poll_cb> function will process all outstanding aio requests by
171calling their callbacks, freeing memory associated with them and managing
172any groups they are contained in.
173
174=item done
175
176Request has reached the end of its lifetime and holds no resources anymore
177(except possibly for the Perl object, but its connection to the actual
178aio request is severed and calling its methods will either do nothing or
179result in a runtime error).
65 180
66=cut 181=cut
67 182
68package IO::AIO; 183package IO::AIO;
69 184
71use strict 'vars'; 186use strict 'vars';
72 187
73use base 'Exporter'; 188use base 'Exporter';
74 189
75BEGIN { 190BEGIN {
76 our $VERSION = '2.0'; 191 our $VERSION = '2.1';
77 192
78 our @EXPORT = qw(aio_sendfile aio_read aio_write aio_open aio_close aio_stat 193 our @AIO_REQ = qw(aio_sendfile aio_read aio_write aio_open aio_close aio_stat
79 aio_lstat aio_unlink aio_rmdir aio_readdir aio_scandir aio_symlink 194 aio_lstat aio_unlink aio_rmdir aio_readdir aio_scandir aio_symlink
80 aio_fsync aio_fdatasync aio_readahead aio_rename aio_link aio_move 195 aio_fsync aio_fdatasync aio_readahead aio_rename aio_link aio_move
81 aio_group); 196 aio_copy aio_group aio_nop aio_mknod);
82 our @EXPORT_OK = qw(poll_fileno poll_cb min_parallel max_parallel max_outstanding nreqs); 197 our @EXPORT = (@AIO_REQ, qw(aioreq_pri aioreq_nice));
198 our @EXPORT_OK = qw(poll_fileno poll_cb poll_wait flush
199 min_parallel max_parallel max_idle
200 nreqs nready npending nthreads
201 max_poll_time max_poll_reqs);
83 202
84 @IO::AIO::GRP::ISA = 'IO::AIO::REQ'; 203 @IO::AIO::GRP::ISA = 'IO::AIO::REQ';
85 204
86 require XSLoader; 205 require XSLoader;
87 XSLoader::load ("IO::AIO", $VERSION); 206 XSLoader::load ("IO::AIO", $VERSION);
118environment, d) use Glib::filename_from_unicode on unicode filenames or e) 237environment, d) use Glib::filename_from_unicode on unicode filenames or e)
119use something else. 238use something else.
120 239
121=over 4 240=over 4
122 241
242=item $prev_pri = aioreq_pri [$pri]
243
244Returns the priority value that would be used for the next request and, if
245C<$pri> is given, sets the priority for the next aio request.
246
247The default priority is C<0>, the minimum and maximum priorities are C<-4>
248and C<4>, respectively. Requests with higher priority will be serviced
249first.
250
251The priority will be reset to C<0> after each call to one of the C<aio_*>
252functions.
253
254Example: open a file with low priority, then read something from it with
255higher priority so the read request is serviced before other low priority
256open requests (potentially spamming the cache):
257
258 aioreq_pri -3;
259 aio_open ..., sub {
260 return unless $_[0];
261
262 aioreq_pri -2;
263 aio_read $_[0], ..., sub {
264 ...
265 };
266 };
267
268=item aioreq_nice $pri_adjust
269
270Similar to C<aioreq_pri>, but subtracts the given value from the current
271priority, so effects are cumulative.
272
123=item aio_open $pathname, $flags, $mode, $callback->($fh) 273=item aio_open $pathname, $flags, $mode, $callback->($fh)
124 274
125Asynchronously open or create a file and call the callback with a newly 275Asynchronously open or create a file and call the callback with a newly
126created filehandle for the file. 276created filehandle for the file.
127 277
176 326
177 aio_read $fh, 7, 15, $buffer, 0, sub { 327 aio_read $fh, 7, 15, $buffer, 0, sub {
178 $_[0] > 0 or die "read error: $!"; 328 $_[0] > 0 or die "read error: $!";
179 print "read $_[0] bytes: <$buffer>\n"; 329 print "read $_[0] bytes: <$buffer>\n";
180 }; 330 };
181
182=item aio_move $srcpath, $dstpath, $callback->($status)
183
184Try to move the I<file> (directories not supported as either source or
185destination) from C<$srcpath> to C<$dstpath> and call the callback with
186the C<0> (error) or C<-1> ok.
187
188This is a composite request that tries to rename(2) the file first. If
189rename files with C<EXDEV>, it creates the destination file with mode 0200
190and copies the contents of the source file into it using C<aio_sendfile>,
191followed by restoring atime, mtime, access mode and uid/gid, in that
192order, and unlinking the C<$srcpath>.
193
194If an error occurs, the partial destination file will be unlinked, if
195possible, except when setting atime, mtime, access mode and uid/gid, where
196errors are being ignored.
197
198=cut
199
200sub aio_move($$$) {
201 my ($src, $dst, $cb) = @_;
202
203 my $grp = aio_group;
204
205 add $grp aio_rename $src, $dst, sub {
206 if ($_[0] && $! == EXDEV) {
207 add $grp aio_open $src, O_RDONLY, 0, sub {
208 if (my $src_fh = $_[0]) {
209 my @stat = stat $src_fh;
210
211 add $grp aio_open $dst, O_WRONLY, 0200, sub {
212 if (my $dst_fh = $_[0]) {
213 add $grp aio_sendfile $dst_fh, $src_fh, 0, $stat[7], sub {
214 close $src_fh;
215
216 if ($_[0] == $stat[7]) {
217 utime $stat[8], $stat[9], $dst;
218 chmod $stat[2] & 07777, $dst_fh;
219 chown $stat[4], $stat[5], $dst_fh;
220 close $dst_fh;
221
222 add $grp aio_unlink $src, sub {
223 $cb->($_[0]);
224 };
225 } else {
226 my $errno = $!;
227 add $grp aio_unlink $dst, sub {
228 $! = $errno;
229 $cb->(-1);
230 };
231 }
232 };
233 } else {
234 $cb->(-1);
235 }
236 },
237
238 } else {
239 $cb->(-1);
240 }
241 };
242 } else {
243 $cb->($_[0]);
244 }
245 };
246
247 $grp
248}
249 331
250=item aio_sendfile $out_fh, $in_fh, $in_offset, $length, $callback->($retval) 332=item aio_sendfile $out_fh, $in_fh, $in_offset, $length, $callback->($retval)
251 333
252Tries to copy C<$length> bytes from C<$in_fh> to C<$out_fh>. It starts 334Tries to copy C<$length> bytes from C<$in_fh> to C<$out_fh>. It starts
253reading at byte offset C<$in_offset>, and starts writing at the current 335reading at byte offset C<$in_offset>, and starts writing at the current
309=item aio_unlink $pathname, $callback->($status) 391=item aio_unlink $pathname, $callback->($status)
310 392
311Asynchronously unlink (delete) a file and call the callback with the 393Asynchronously unlink (delete) a file and call the callback with the
312result code. 394result code.
313 395
396=item aio_mknod $path, $mode, $dev, $callback->($status)
397
398[EXPERIMENTAL]
399
400Asynchronously create a device node (or fifo). See mknod(2).
401
402The only (POSIX-) portable way of calling this function is:
403
404 aio_mknod $path, IO::AIO::S_IFIFO | $mode, 0, sub { ...
405
314=item aio_link $srcpath, $dstpath, $callback->($status) 406=item aio_link $srcpath, $dstpath, $callback->($status)
315 407
316Asynchronously create a new link to the existing object at C<$srcpath> at 408Asynchronously create a new link to the existing object at C<$srcpath> at
317the path C<$dstpath> and call the callback with the result code. 409the path C<$dstpath> and call the callback with the result code.
318 410
338sorted, and will B<NOT> include the C<.> and C<..> entries. 430sorted, and will B<NOT> include the C<.> and C<..> entries.
339 431
340The callback a single argument which is either C<undef> or an array-ref 432The callback a single argument which is either C<undef> or an array-ref
341with the filenames. 433with the filenames.
342 434
435=item aio_copy $srcpath, $dstpath, $callback->($status)
436
437Try to copy the I<file> (directories not supported as either source or
438destination) from C<$srcpath> to C<$dstpath> and call the callback with
439the C<0> (error) or C<-1> ok.
440
441This is a composite request that it creates the destination file with
442mode 0200 and copies the contents of the source file into it using
443C<aio_sendfile>, followed by restoring atime, mtime, access mode and
444uid/gid, in that order.
445
446If an error occurs, the partial destination file will be unlinked, if
447possible, except when setting atime, mtime, access mode and uid/gid, where
448errors are being ignored.
449
450=cut
451
452sub aio_copy($$;$) {
453 my ($src, $dst, $cb) = @_;
454
455 my $pri = aioreq_pri;
456 my $grp = aio_group $cb;
457
458 aioreq_pri $pri;
459 add $grp aio_open $src, O_RDONLY, 0, sub {
460 if (my $src_fh = $_[0]) {
461 my @stat = stat $src_fh;
462
463 aioreq_pri $pri;
464 add $grp aio_open $dst, O_CREAT | O_WRONLY | O_TRUNC, 0200, sub {
465 if (my $dst_fh = $_[0]) {
466 aioreq_pri $pri;
467 add $grp aio_sendfile $dst_fh, $src_fh, 0, $stat[7], sub {
468 if ($_[0] == $stat[7]) {
469 $grp->result (0);
470 close $src_fh;
471
472 # those should not normally block. should. should.
473 utime $stat[8], $stat[9], $dst;
474 chmod $stat[2] & 07777, $dst_fh;
475 chown $stat[4], $stat[5], $dst_fh;
476 close $dst_fh;
477 } else {
478 $grp->result (-1);
479 close $src_fh;
480 close $dst_fh;
481
482 aioreq $pri;
483 add $grp aio_unlink $dst;
484 }
485 };
486 } else {
487 $grp->result (-1);
488 }
489 },
490
491 } else {
492 $grp->result (-1);
493 }
494 };
495
496 $grp
497}
498
499=item aio_move $srcpath, $dstpath, $callback->($status)
500
501Try to move the I<file> (directories not supported as either source or
502destination) from C<$srcpath> to C<$dstpath> and call the callback with
503the C<0> (error) or C<-1> ok.
504
505This is a composite request that tries to rename(2) the file first. If
506rename files with C<EXDEV>, it copies the file with C<aio_copy> and, if
507that is successful, unlinking the C<$srcpath>.
508
509=cut
510
511sub aio_move($$;$) {
512 my ($src, $dst, $cb) = @_;
513
514 my $pri = aioreq_pri;
515 my $grp = aio_group $cb;
516
517 aioreq_pri $pri;
518 add $grp aio_rename $src, $dst, sub {
519 if ($_[0] && $! == EXDEV) {
520 aioreq_pri $pri;
521 add $grp aio_copy $src, $dst, sub {
522 $grp->result ($_[0]);
523
524 if (!$_[0]) {
525 aioreq_pri $pri;
526 add $grp aio_unlink $src;
527 }
528 };
529 } else {
530 $grp->result ($_[0]);
531 }
532 };
533
534 $grp
535}
536
343=item aio_scandir $path, $maxreq, $callback->($dirs, $nondirs) 537=item aio_scandir $path, $maxreq, $callback->($dirs, $nondirs)
344 538
345Scans a directory (similar to C<aio_readdir>) but additionally tries to 539Scans a directory (similar to C<aio_readdir>) but additionally tries to
346separate the entries of directory C<$path> into two sets of names, ones 540efficiently separate the entries of directory C<$path> into two sets of
347you can recurse into (directories or links to them), and ones you cannot 541names, directories you can recurse into (directories), and ones you cannot
348recurse into (everything else). 542recurse into (everything else, including symlinks to directories).
349 543
350C<aio_scandir> is a composite request that consists of many sub 544C<aio_scandir> is a composite request that creates of many sub requests_
351requests. C<$maxreq> specifies the maximum number of outstanding aio 545C<$maxreq> specifies the maximum number of outstanding aio requests that
352requests that this function generates. If it is C<< <= 0 >>, then a 546this function generates. If it is C<< <= 0 >>, then a suitable default
353suitable default will be chosen (currently 8). 547will be chosen (currently 4).
354 548
355On error, the callback is called without arguments, otherwise it receives 549On error, the callback is called without arguments, otherwise it receives
356two array-refs with path-relative entry names. 550two array-refs with path-relative entry names.
357 551
358Example: 552Example:
395=cut 589=cut
396 590
397sub aio_scandir($$$) { 591sub aio_scandir($$$) {
398 my ($path, $maxreq, $cb) = @_; 592 my ($path, $maxreq, $cb) = @_;
399 593
594 my $pri = aioreq_pri;
595
400 my $grp = aio_group; 596 my $grp = aio_group $cb;
401 597
402 $maxreq = 8 if $maxreq <= 0; 598 $maxreq = 4 if $maxreq <= 0;
403 599
404 # stat once 600 # stat once
601 aioreq_pri $pri;
405 add $grp aio_stat $path, sub { 602 add $grp aio_stat $path, sub {
406 return $cb->() if $_[0]; 603 return $grp->result () if $_[0];
407 my $now = time; 604 my $now = time;
408 my $hash1 = join ":", (stat _)[0,1,3,7,9]; 605 my $hash1 = join ":", (stat _)[0,1,3,7,9];
409 606
410 # read the directory entries 607 # read the directory entries
608 aioreq_pri $pri;
411 add $grp aio_readdir $path, sub { 609 add $grp aio_readdir $path, sub {
412 my $entries = shift 610 my $entries = shift
413 or return $cb->(); 611 or return $grp->result ();
414 612
415 # stat the dir another time 613 # stat the dir another time
614 aioreq_pri $pri;
416 add $grp aio_stat $path, sub { 615 add $grp aio_stat $path, sub {
417 my $hash2 = join ":", (stat _)[0,1,3,7,9]; 616 my $hash2 = join ":", (stat _)[0,1,3,7,9];
418 617
419 my $ndirs; 618 my $ndirs;
420 619
423 $ndirs = -1; 622 $ndirs = -1;
424 } else { 623 } else {
425 # if nlink == 2, we are finished 624 # if nlink == 2, we are finished
426 # on non-posix-fs's, we rely on nlink < 2 625 # on non-posix-fs's, we rely on nlink < 2
427 $ndirs = (stat _)[3] - 2 626 $ndirs = (stat _)[3] - 2
428 or return $cb->([], $entries); 627 or return $grp->result ([], $entries);
429 } 628 }
430 629
431 # sort into likely dirs and likely nondirs 630 # sort into likely dirs and likely nondirs
432 # dirs == files without ".", short entries first 631 # dirs == files without ".", short entries first
433 $entries = [map $_->[0], 632 $entries = [map $_->[0],
435 map [$_, sprintf "%s%04d", (/.\./ ? "1" : "0"), length], 634 map [$_, sprintf "%s%04d", (/.\./ ? "1" : "0"), length],
436 @$entries]; 635 @$entries];
437 636
438 my (@dirs, @nondirs); 637 my (@dirs, @nondirs);
439 638
440 my ($statcb, $schedcb); 639 my $statgrp = add $grp aio_group sub {
441 my $nreq = 0; 640 $grp->result (\@dirs, \@nondirs);
641 };
442 642
443 $schedcb = sub { 643 limit $statgrp $maxreq;
444 if (@$entries) { 644 feed $statgrp sub {
445 if ($nreq < $maxreq) { 645 return unless @$entries;
446 my $ent = pop @$entries; 646 my $entry = pop @$entries;
647
648 aioreq_pri $pri;
649 add $statgrp aio_stat "$path/$entry/.", sub {
650 if ($_[0] < 0) {
651 push @nondirs, $entry;
652 } else {
653 # need to check for real directory
654 aioreq_pri $pri;
655 add $statgrp aio_lstat "$path/$entry", sub {
656 if (-d _) {
657 push @dirs, $entry;
658
659 unless (--$ndirs) {
660 push @nondirs, @$entries;
661 feed $statgrp;
662 }
663 } else {
664 push @nondirs, $entry;
665 }
447 $nreq++; 666 }
448 add $grp aio_stat "$path/$ent/.", sub { $statcb->($_[0], $ent) };
449 } 667 }
450 } elsif (!$nreq) {
451 # finished
452 undef $statcb;
453 undef $schedcb;
454 $cb->(\@dirs, \@nondirs) if $cb;
455 undef $cb;
456 } 668 };
457 }; 669 };
458 $statcb = sub {
459 my ($status, $entry) = @_;
460
461 if ($status < 0) {
462 $nreq--;
463 push @nondirs, $entry;
464 &$schedcb;
465 } else {
466 # need to check for real directory
467 add $grp aio_lstat "$path/$entry", sub {
468 $nreq--;
469
470 if (-d _) {
471 push @dirs, $entry;
472
473 if (!--$ndirs) {
474 push @nondirs, @$entries;
475 $entries = [];
476 }
477 } else {
478 push @nondirs, $entry;
479 }
480
481 &$schedcb;
482 }
483 }
484 };
485
486 &$schedcb while @$entries && $nreq < $maxreq;
487 }; 670 };
488 }; 671 };
489 }; 672 };
490 673
491 $grp 674 $grp
502callback with the fdatasync result code. 685callback with the fdatasync result code.
503 686
504If this call isn't available because your OS lacks it or it couldn't be 687If this call isn't available because your OS lacks it or it couldn't be
505detected, it will be emulated by calling C<fsync> instead. 688detected, it will be emulated by calling C<fsync> instead.
506 689
507=item aio_group $callback->() 690=item aio_group $callback->(...)
508
509[EXPERIMENTAL]
510 691
511This is a very special aio request: Instead of doing something, it is a 692This is a very special aio request: Instead of doing something, it is a
512container for other aio requests, which is useful if you want to bundle 693container for other aio requests, which is useful if you want to bundle
513many requests into a single, composite, request. 694many requests into a single, composite, request with a definite callback
695and the ability to cancel the whole request with its subrequests.
514 696
515Returns an object of class L<IO::AIO::GRP>. See its documentation below 697Returns an object of class L<IO::AIO::GRP>. See its documentation below
516for more info. 698for more info.
517 699
518Example: 700Example:
524 add $grp 706 add $grp
525 (aio_stat ...), 707 (aio_stat ...),
526 (aio_stat ...), 708 (aio_stat ...),
527 ...; 709 ...;
528 710
711=item aio_nop $callback->()
712
713This is a special request - it does nothing in itself and is only used for
714side effects, such as when you want to add a dummy request to a group so
715that finishing the requests in the group depends on executing the given
716code.
717
718While this request does nothing, it still goes through the execution
719phase and still requires a worker thread. Thus, the callback will not
720be executed immediately but only after other requests in the queue have
721entered their execution phase. This can be used to measure request
722latency.
723
529=item aio_sleep $fractional_seconds, $callback->() *NOT EXPORTED* 724=item IO::AIO::aio_busy $fractional_seconds, $callback->() *NOT EXPORTED*
530 725
531Mainly used for debugging and benchmarking, this aio request puts one of 726Mainly used for debugging and benchmarking, this aio request puts one of
532the request workers to sleep for the given time. 727the request workers to sleep for the given time.
533 728
729While it is theoretically handy to have simple I/O scheduling requests
730like sleep and file handle readable/writable, the overhead this creates is
731immense (it blocks a thread for a long time) so do not use this function
732except to put your application under artificial I/O pressure.
733
534=back 734=back
535 735
536=head2 IO::AIO::REQ CLASS 736=head2 IO::AIO::REQ CLASS
537 737
538All non-aggregate C<aio_*> functions return an object of this class when 738All non-aggregate C<aio_*> functions return an object of this class when
539called in non-void context. 739called in non-void context.
540 740
541A request always moves through the following five states in its lifetime,
542in order: B<ready> (request has been created, but has not been executed
543yet), B<execute> (request is currently being executed), B<pending>
544(request has been executed but callback has not been called yet),
545B<result> (results are being processed synchronously, includes calling the
546callback) and B<done> (request has reached the end of its lifetime and
547holds no resources anymore).
548
549=over 4 741=over 4
550 742
551=item $req->cancel 743=item cancel $req
552 744
553Cancels the request, if possible. Has the effect of skipping execution 745Cancels the request, if possible. Has the effect of skipping execution
554when entering the B<execute> state and skipping calling the callback when 746when entering the B<execute> state and skipping calling the callback when
555entering the the B<result> state, but will leave the request otherwise 747entering the the B<result> state, but will leave the request otherwise
556untouched. That means that requests that currently execute will not be 748untouched. That means that requests that currently execute will not be
557stopped and resources held by the request will not be freed prematurely. 749stopped and resources held by the request will not be freed prematurely.
558 750
751=item cb $req $callback->(...)
752
753Replace (or simply set) the callback registered to the request.
754
559=back 755=back
560 756
561=head2 IO::AIO::GRP CLASS 757=head2 IO::AIO::GRP CLASS
562 758
563This class is a subclass of L<IO::AIO::REQ>, so all its methods apply to 759This class is a subclass of L<IO::AIO::REQ>, so all its methods apply to
577You add requests by calling the C<add> method with one or more 773You add requests by calling the C<add> method with one or more
578C<IO::AIO::REQ> objects: 774C<IO::AIO::REQ> objects:
579 775
580 $grp->add (aio_unlink "..."); 776 $grp->add (aio_unlink "...");
581 777
582 add $grp aio_stat "...", sub { ... }; 778 add $grp aio_stat "...", sub {
779 $_[0] or return $grp->result ("error");
780
781 # add another request dynamically, if first succeeded
782 add $grp aio_open "...", sub {
783 $grp->result ("ok");
784 };
785 };
583 786
584This makes it very easy to create composite requests (see the source of 787This makes it very easy to create composite requests (see the source of
585C<aio_move> for an application) that work and feel like simple requests. 788C<aio_move> for an application) that work and feel like simple requests.
586 789
790=over 4
791
587The IO::AIO::GRP objects will be cleaned up during calls to 792=item * The IO::AIO::GRP objects will be cleaned up during calls to
588C<IO::AIO::poll_cb>, just like any other request. 793C<IO::AIO::poll_cb>, just like any other request.
589 794
590They can be canceled like any other request. Canceling will cancel not 795=item * They can be canceled like any other request. Canceling will cancel not
591just the request itself, but also all requests it contains. 796only the request itself, but also all requests it contains.
592 797
593They can also can also be added to other IO::AIO::GRP objects. 798=item * They can also can also be added to other IO::AIO::GRP objects.
799
800=item * You must not add requests to a group from within the group callback (or
801any later time).
802
803=back
594 804
595Their lifetime, simplified, looks like this: when they are empty, they 805Their lifetime, simplified, looks like this: when they are empty, they
596will finish very quickly. If they contain only requests that are in the 806will finish very quickly. If they contain only requests that are in the
597C<done> state, they will also finish. Otherwise they will continue to 807C<done> state, they will also finish. Otherwise they will continue to
598exist. 808exist.
599 809
810That means after creating a group you have some time to add requests. And
811in the callbacks of those requests, you can add further requests to the
812group. And only when all those requests have finished will the the group
813itself finish.
814
600=over 4 815=over 4
601 816
817=item add $grp ...
818
602=item $grp->add (...) 819=item $grp->add (...)
603 820
604=item add $grp ... 821Add one or more requests to the group. Any type of L<IO::AIO::REQ> can
822be added, including other groups, as long as you do not create circular
823dependencies.
605 824
606Add one or more 825Returns all its arguments.
607Cancels the request, if possible. Has the effect of skipping execution 826
608when entering the B<execute> state and skipping calling the callback when 827=item $grp->cancel_subs
609entering the the B<result> state, but will leave the request otherwise 828
610untouched. That means that requests that currently execute will not be 829Cancel all subrequests and clears any feeder, but not the group request
611stopped and resources held by the request will not be freed prematurely. 830itself. Useful when you queued a lot of events but got a result early.
831
832=item $grp->result (...)
833
834Set the result value(s) that will be passed to the group callback when all
835subrequests have finished and set thre groups errno to the current value
836of errno (just like calling C<errno> without an error number). By default,
837no argument will be passed and errno is zero.
838
839=item $grp->errno ([$errno])
840
841Sets the group errno value to C<$errno>, or the current value of errno
842when the argument is missing.
843
844Every aio request has an associated errno value that is restored when
845the callback is invoked. This method lets you change this value from its
846default (0).
847
848Calling C<result> will also set errno, so make sure you either set C<$!>
849before the call to C<result>, or call c<errno> after it.
850
851=item feed $grp $callback->($grp)
852
853Sets a feeder/generator on this group: every group can have an attached
854generator that generates requests if idle. The idea behind this is that,
855although you could just queue as many requests as you want in a group,
856this might starve other requests for a potentially long time. For
857example, C<aio_scandir> might generate hundreds of thousands C<aio_stat>
858requests, delaying any later requests for a long time.
859
860To avoid this, and allow incremental generation of requests, you can
861instead a group and set a feeder on it that generates those requests. The
862feed callback will be called whenever there are few enough (see C<limit>,
863below) requests active in the group itself and is expected to queue more
864requests.
865
866The feed callback can queue as many requests as it likes (i.e. C<add> does
867not impose any limits).
868
869If the feed does not queue more requests when called, it will be
870automatically removed from the group.
871
872If the feed limit is C<0>, it will be set to C<2> automatically.
873
874Example:
875
876 # stat all files in @files, but only ever use four aio requests concurrently:
877
878 my $grp = aio_group sub { print "finished\n" };
879 limit $grp 4;
880 feed $grp sub {
881 my $file = pop @files
882 or return;
883
884 add $grp aio_stat $file, sub { ... };
885 };
886
887=item limit $grp $num
888
889Sets the feeder limit for the group: The feeder will be called whenever
890the group contains less than this many requests.
891
892Setting the limit to C<0> will pause the feeding process.
612 893
613=back 894=back
614 895
615=head2 SUPPORT FUNCTIONS 896=head2 SUPPORT FUNCTIONS
897
898=head3 EVENT PROCESSING AND EVENT LOOP INTEGRATION
616 899
617=over 4 900=over 4
618 901
619=item $fileno = IO::AIO::poll_fileno 902=item $fileno = IO::AIO::poll_fileno
620 903
625 908
626See C<poll_cb> for an example. 909See C<poll_cb> for an example.
627 910
628=item IO::AIO::poll_cb 911=item IO::AIO::poll_cb
629 912
630Process all outstanding events on the result pipe. You have to call this 913Process some outstanding events on the result pipe. You have to call this
631regularly. Returns the number of events processed. Returns immediately 914regularly. Returns the number of events processed. Returns immediately
632when no events are outstanding. 915when no events are outstanding. The amount of events processed depends on
916the settings of C<IO::AIO::max_poll_req> and C<IO::AIO::max_poll_time>.
917
918If not all requests were processed for whatever reason, the filehandle
919will still be ready when C<poll_cb> returns.
633 920
634Example: Install an Event watcher that automatically calls 921Example: Install an Event watcher that automatically calls
635IO::AIO::poll_cb with high priority: 922IO::AIO::poll_cb with high priority:
636 923
637 Event->io (fd => IO::AIO::poll_fileno, 924 Event->io (fd => IO::AIO::poll_fileno,
638 poll => 'r', async => 1, 925 poll => 'r', async => 1,
639 cb => \&IO::AIO::poll_cb); 926 cb => \&IO::AIO::poll_cb);
640 927
928=item IO::AIO::max_poll_reqs $nreqs
929
930=item IO::AIO::max_poll_time $seconds
931
932These set the maximum number of requests (default C<0>, meaning infinity)
933that are being processed by C<IO::AIO::poll_cb> in one call, respectively
934the maximum amount of time (default C<0>, meaning infinity) spent in
935C<IO::AIO::poll_cb> to process requests (more correctly the mininum amount
936of time C<poll_cb> is allowed to use).
937
938Setting these is useful if you want to ensure some level of
939interactiveness when perl is not fast enough to process all requests in
940time.
941
942For interactive programs, values such as C<0.01> to C<0.1> should be fine.
943
944Example: Install an Event watcher that automatically calls
945IO::AIO::poll_some with low priority, to ensure that other parts of the
946program get the CPU sometimes even under high AIO load.
947
948 # try not to spend much more than 0.1s in poll_cb
949 IO::AIO::max_poll_time 0.1;
950
951 # use a low priority so other tasks have priority
952 Event->io (fd => IO::AIO::poll_fileno,
953 poll => 'r', nice => 1,
954 cb => &IO::AIO::poll_cb);
955
641=item IO::AIO::poll_wait 956=item IO::AIO::poll_wait
642 957
643Wait till the result filehandle becomes ready for reading (simply does a 958Wait till the result filehandle becomes ready for reading (simply does a
644C<select> on the filehandle. This is useful if you want to synchronously wait 959C<select> on the filehandle. This is useful if you want to synchronously
645for some requests to finish). 960wait for some requests to finish).
646 961
647See C<nreqs> for an example. 962See C<nreqs> for an example.
648 963
964=item IO::AIO::poll
965
966Waits until some requests have been handled.
967
968Strictly equivalent to:
969
970 IO::AIO::poll_wait, IO::AIO::poll_cb
971 if IO::AIO::nreqs;
972
649=item IO::AIO::nreqs 973=item IO::AIO::flush
650 974
651Returns the number of requests currently outstanding (i.e. for which their 975Wait till all outstanding AIO requests have been handled.
652callback has not been invoked yet).
653 976
654Example: wait till there are no outstanding requests anymore: 977Strictly equivalent to:
655 978
656 IO::AIO::poll_wait, IO::AIO::poll_cb 979 IO::AIO::poll_wait, IO::AIO::poll_cb
657 while IO::AIO::nreqs; 980 while IO::AIO::nreqs;
658 981
659=item IO::AIO::flush 982=head3 CONTROLLING THE NUMBER OF THREADS
660
661Wait till all outstanding AIO requests have been handled.
662
663Strictly equivalent to:
664
665 IO::AIO::poll_wait, IO::AIO::poll_cb
666 while IO::AIO::nreqs;
667
668=item IO::AIO::poll
669
670Waits until some requests have been handled.
671
672Strictly equivalent to:
673
674 IO::AIO::poll_wait, IO::AIO::poll_cb
675 if IO::AIO::nreqs;
676 983
677=item IO::AIO::min_parallel $nthreads 984=item IO::AIO::min_parallel $nthreads
678 985
679Set the minimum number of AIO threads to C<$nthreads>. The current default 986Set the minimum number of AIO threads to C<$nthreads>. The current
680is C<4>, which means four asynchronous operations can be done at one time 987default is C<8>, which means eight asynchronous operations can execute
681(the number of outstanding operations, however, is unlimited). 988concurrently at any one time (the number of outstanding requests,
989however, is unlimited).
682 990
683IO::AIO starts threads only on demand, when an AIO request is queued and 991IO::AIO starts threads only on demand, when an AIO request is queued and
684no free thread exists. 992no free thread exists. Please note that queueing up a hundred requests can
993create demand for a hundred threads, even if it turns out that everything
994is in the cache and could have been processed faster by a single thread.
685 995
686It is recommended to keep the number of threads low, as some Linux 996It is recommended to keep the number of threads relatively low, as some
687kernel versions will scale negatively with the number of threads (higher 997Linux kernel versions will scale negatively with the number of threads
688parallelity => MUCH higher latency). With current Linux 2.6 versions, 4-32 998(higher parallelity => MUCH higher latency). With current Linux 2.6
689threads should be fine. 999versions, 4-32 threads should be fine.
690 1000
691Under most circumstances you don't need to call this function, as the 1001Under most circumstances you don't need to call this function, as the
692module selects a default that is suitable for low to moderate load. 1002module selects a default that is suitable for low to moderate load.
693 1003
694=item IO::AIO::max_parallel $nthreads 1004=item IO::AIO::max_parallel $nthreads
703This module automatically runs C<max_parallel 0> at program end, to ensure 1013This module automatically runs C<max_parallel 0> at program end, to ensure
704that all threads are killed and that there are no outstanding requests. 1014that all threads are killed and that there are no outstanding requests.
705 1015
706Under normal circumstances you don't need to call this function. 1016Under normal circumstances you don't need to call this function.
707 1017
1018=item IO::AIO::max_idle $nthreads
1019
1020Limit the number of threads (default: 4) that are allowed to idle (i.e.,
1021threads that did not get a request to process within 10 seconds). That
1022means if a thread becomes idle while C<$nthreads> other threads are also
1023idle, it will free its resources and exit.
1024
1025This is useful when you allow a large number of threads (e.g. 100 or 1000)
1026to allow for extremely high load situations, but want to free resources
1027under normal circumstances (1000 threads can easily consume 30MB of RAM).
1028
1029The default is probably ok in most situations, especially if thread
1030creation is fast. If thread creation is very slow on your system you might
1031want to use larger values.
1032
708=item $oldnreqs = IO::AIO::max_outstanding $nreqs 1033=item $oldmaxreqs = IO::AIO::max_outstanding $maxreqs
1034
1035This is a very bad function to use in interactive programs because it
1036blocks, and a bad way to reduce concurrency because it is inexact: Better
1037use an C<aio_group> together with a feed callback.
709 1038
710Sets the maximum number of outstanding requests to C<$nreqs>. If you 1039Sets the maximum number of outstanding requests to C<$nreqs>. If you
711try to queue up more than this number of requests, the caller will block until 1040to queue up more than this number of requests, the next call to the
712some requests have been handled. 1041C<poll_cb> (and C<poll_some> and other functions calling C<poll_cb>)
1042function will block until the limit is no longer exceeded.
713 1043
714The default is very large, so normally there is no practical limit. If you 1044The default value is very large, so there is no practical limit on the
715queue up many requests in a loop it often improves speed if you set 1045number of outstanding requests.
716this to a relatively low number, such as C<100>.
717 1046
718Under normal circumstances you don't need to call this function. 1047You can still queue as many requests as you want. Therefore,
1048C<max_oustsanding> is mainly useful in simple scripts (with low values) or
1049as a stop gap to shield against fatal memory overflow (with large values).
1050
1051=head3 STATISTICAL INFORMATION
1052
1053=item IO::AIO::nreqs
1054
1055Returns the number of requests currently in the ready, execute or pending
1056states (i.e. for which their callback has not been invoked yet).
1057
1058Example: wait till there are no outstanding requests anymore:
1059
1060 IO::AIO::poll_wait, IO::AIO::poll_cb
1061 while IO::AIO::nreqs;
1062
1063=item IO::AIO::nready
1064
1065Returns the number of requests currently in the ready state (not yet
1066executed).
1067
1068=item IO::AIO::npending
1069
1070Returns the number of requests currently in the pending state (executed,
1071but not yet processed by poll_cb).
719 1072
720=back 1073=back
721 1074
722=cut 1075=cut
723 1076
735 or return undef; 1088 or return undef;
736 1089
737 *$sym 1090 *$sym
738} 1091}
739 1092
740min_parallel 4; 1093min_parallel 8;
741 1094
742END { 1095END {
743 max_parallel 0; 1096 min_parallel 1;
744} 1097 flush;
1098};
745 1099
7461; 11001;
747 1101
748=head2 FORK BEHAVIOUR 1102=head2 FORK BEHAVIOUR
749 1103
750This module should do "the right thing" when the process using it forks: 1104This module should do "the right thing" when the process using it forks:
751 1105
752Before the fork, IO::AIO enters a quiescent state where no requests 1106Before the fork, IO::AIO enters a quiescent state where no requests
753can be added in other threads and no results will be processed. After 1107can be added in other threads and no results will be processed. After
754the fork the parent simply leaves the quiescent state and continues 1108the fork the parent simply leaves the quiescent state and continues
755request/result processing, while the child clears the request/result 1109request/result processing, while the child frees the request/result queue
756queue (so the requests started before the fork will only be handled in 1110(so that the requests started before the fork will only be handled in the
757the parent). Threads will be started on demand until the limit ste in the 1111parent). Threads will be started on demand until the limit set in the
758parent process has been reached again. 1112parent process has been reached again.
759 1113
760In short: the parent will, after a short pause, continue as if fork had 1114In short: the parent will, after a short pause, continue as if fork had
761not been called, while the child will act as if IO::AIO has not been used 1115not been called, while the child will act as if IO::AIO has not been used
762yet. 1116yet.
763 1117
1118=head2 MEMORY USAGE
1119
1120Per-request usage:
1121
1122Each aio request uses - depending on your architecture - around 100-200
1123bytes of memory. In addition, stat requests need a stat buffer (possibly
1124a few hundred bytes), readdir requires a result buffer and so on. Perl
1125scalars and other data passed into aio requests will also be locked and
1126will consume memory till the request has entered the done state.
1127
1128This is now awfully much, so queuing lots of requests is not usually a
1129problem.
1130
1131Per-thread usage:
1132
1133In the execution phase, some aio requests require more memory for
1134temporary buffers, and each thread requires a stack and other data
1135structures (usually around 16k-128k, depending on the OS).
1136
1137=head1 KNOWN BUGS
1138
1139Known bugs will be fixed in the next release.
1140
764=head1 SEE ALSO 1141=head1 SEE ALSO
765 1142
766L<Coro>, L<Linux::AIO> (obsolete). 1143L<Coro::AIO>.
767 1144
768=head1 AUTHOR 1145=head1 AUTHOR
769 1146
770 Marc Lehmann <schmorp@schmorp.de> 1147 Marc Lehmann <schmorp@schmorp.de>
771 http://home.schmorp.de/ 1148 http://home.schmorp.de/

Diff Legend

Removed lines
+ Added lines
< Changed lines
> Changed lines