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 1.68 by root, Tue Oct 24 03:17:39 2006 UTC vs.
Revision 1.98 by root, Sun Dec 31 17:07:32 2006 UTC

5=head1 SYNOPSIS 5=head1 SYNOPSIS
6 6
7 use IO::AIO; 7 use IO::AIO;
8 8
9 aio_open "/etc/passwd", O_RDONLY, 0, sub { 9 aio_open "/etc/passwd", O_RDONLY, 0, sub {
10 my ($fh) = @_; 10 my $fh = shift
11 or die "/etc/passwd: $!";
11 ... 12 ...
12 }; 13 };
13 14
14 aio_unlink "/tmp/file", sub { }; 15 aio_unlink "/tmp/file", sub { };
15 16
51=head1 DESCRIPTION 52=head1 DESCRIPTION
52 53
53This module implements asynchronous I/O using whatever means your 54This module implements asynchronous I/O using whatever means your
54operating system supports. 55operating system supports.
55 56
57Asynchronous means that operations that can normally block your program
58(e.g. reading from disk) will be done asynchronously: the operation
59will still block, but you can do something else in the meantime. This
60is extremely useful for programs that need to stay interactive even
61when doing heavy I/O (GUI programs, high performance network servers
62etc.), but can also be used to easily do operations in parallel that are
63normally done sequentially, e.g. stat'ing many files, which is much faster
64on a RAID volume or over NFS when you do a number of stat operations
65concurrently.
66
67While most of this works on all types of file descriptors (for example
68sockets), using these functions on file descriptors that support
69nonblocking operation (again, sockets, pipes etc.) is very inefficient or
70might not work (aio_read fails on sockets/pipes/fifos). Use an event loop
71for that (such as the L<Event|Event> module): IO::AIO will naturally fit
72into such an event loop itself.
73
56Currently, a number of threads are started that execute your read/writes 74In this version, a number of threads are started that execute your
57and signal their completion. You don't need thread support in perl, and 75requests and signal their completion. You don't need thread support
58the threads created by this module will not be visible to perl. In the 76in perl, and the threads created by this module will not be visible
59future, this module might make use of the native aio functions available 77to perl. In the future, this module might make use of the native aio
60on many operating systems. However, they are often not well-supported 78functions available on many operating systems. However, they are often
61(Linux doesn't allow them on normal files currently, for example), 79not well-supported or restricted (GNU/Linux doesn't allow them on normal
62and they would only support aio_read and aio_write, so the remaining 80files currently, for example), and they would only support aio_read and
63functionality would have to be implemented using threads anyway. 81aio_write, so the remaining functionality would have to be implemented
82using threads anyway.
64 83
65Although the module will work with in the presence of other threads, 84Although the module will work with in the presence of other (Perl-)
66it is currently not reentrant in any way, so use appropriate locking 85threads, it is currently not reentrant in any way, so use appropriate
67yourself, always call C<poll_cb> from within the same thread, or never 86locking yourself, always call C<poll_cb> from within the same thread, or
68call C<poll_cb> (or other C<aio_> functions) recursively. 87never call C<poll_cb> (or other C<aio_> functions) recursively.
88
89=head2 EXAMPLE
90
91This is a simple example that uses the Event module and loads
92F</etc/passwd> asynchronously:
93
94 use Fcntl;
95 use Event;
96 use IO::AIO;
97
98 # register the IO::AIO callback with Event
99 Event->io (fd => IO::AIO::poll_fileno,
100 poll => 'r',
101 cb => \&IO::AIO::poll_cb);
102
103 # queue the request to open /etc/passwd
104 aio_open "/etc/passwd", O_RDONLY, 0, sub {
105 my $fh = shift
106 or die "error while opening: $!";
107
108 # stat'ing filehandles is generally non-blocking
109 my $size = -s $fh;
110
111 # queue a request to read the file
112 my $contents;
113 aio_read $fh, 0, $size, $contents, 0, sub {
114 $_[0] == $size
115 or die "short read: $!";
116
117 close $fh;
118
119 # file contents now in $contents
120 print $contents;
121
122 # exit event loop and program
123 Event::unloop;
124 };
125 };
126
127 # possibly queue up other requests, or open GUI windows,
128 # check for sockets etc. etc.
129
130 # process events as long as there are some:
131 Event::loop;
132
133=head1 REQUEST ANATOMY AND LIFETIME
134
135Every C<aio_*> function creates a request. which is a C data structure not
136directly visible to Perl.
137
138If called in non-void context, every request function returns a Perl
139object representing the request. In void context, nothing is returned,
140which saves a bit of memory.
141
142The perl object is a fairly standard ref-to-hash object. The hash contents
143are not used by IO::AIO so you are free to store anything you like in it.
144
145During their existance, aio requests travel through the following states,
146in order:
147
148=over 4
149
150=item ready
151
152Immediately after a request is created it is put into the ready state,
153waiting for a thread to execute it.
154
155=item execute
156
157A thread has accepted the request for processing and is currently
158executing it (e.g. blocking in read).
159
160=item pending
161
162The request has been executed and is waiting for result processing.
163
164While request submission and execution is fully asynchronous, result
165processing is not and relies on the perl interpreter calling C<poll_cb>
166(or another function with the same effect).
167
168=item result
169
170The request results are processed synchronously by C<poll_cb>.
171
172The C<poll_cb> function will process all outstanding aio requests by
173calling their callbacks, freeing memory associated with them and managing
174any groups they are contained in.
175
176=item done
177
178Request has reached the end of its lifetime and holds no resources anymore
179(except possibly for the Perl object, but its connection to the actual
180aio request is severed and calling its methods will either do nothing or
181result in a runtime error).
182
183=back
69 184
70=cut 185=cut
71 186
72package IO::AIO; 187package IO::AIO;
73 188
75use strict 'vars'; 190use strict 'vars';
76 191
77use base 'Exporter'; 192use base 'Exporter';
78 193
79BEGIN { 194BEGIN {
80 our $VERSION = '2.0'; 195 our $VERSION = '2.31';
81 196
82 our @AIO_REQ = qw(aio_sendfile aio_read aio_write aio_open aio_close aio_stat 197 our @AIO_REQ = qw(aio_sendfile aio_read aio_write aio_open aio_close aio_stat
83 aio_lstat aio_unlink aio_rmdir aio_readdir aio_scandir aio_symlink 198 aio_lstat aio_unlink aio_rmdir aio_readdir aio_scandir aio_symlink
84 aio_fsync aio_fdatasync aio_readahead aio_rename aio_link aio_move 199 aio_readlink aio_fsync aio_fdatasync aio_readahead aio_rename aio_link
85 aio_group aio_nop); 200 aio_move aio_copy aio_group aio_nop aio_mknod aio_load);
86 our @EXPORT = (@AIO_REQ, qw(aioreq_pri)); 201 our @EXPORT = (@AIO_REQ, qw(aioreq_pri aioreq_nice aio_block));
87 our @EXPORT_OK = qw(poll_fileno poll_cb poll_wait flush 202 our @EXPORT_OK = qw(poll_fileno poll_cb poll_wait flush
88 min_parallel max_parallel max_outstanding nreqs); 203 min_parallel max_parallel max_idle
204 nreqs nready npending nthreads
205 max_poll_time max_poll_reqs);
89 206
90 @IO::AIO::GRP::ISA = 'IO::AIO::REQ'; 207 @IO::AIO::GRP::ISA = 'IO::AIO::REQ';
91 208
92 require XSLoader; 209 require XSLoader;
93 XSLoader::load ("IO::AIO", $VERSION); 210 XSLoader::load ("IO::AIO", $VERSION);
94} 211}
95 212
96=head1 FUNCTIONS 213=head1 FUNCTIONS
97 214
98=head2 AIO FUNCTIONS 215=head2 AIO REQUEST FUNCTIONS
99 216
100All the C<aio_*> calls are more or less thin wrappers around the syscall 217All the C<aio_*> calls are more or less thin wrappers around the syscall
101with the same name (sans C<aio_>). The arguments are similar or identical, 218with the same name (sans C<aio_>). The arguments are similar or identical,
102and they all accept an additional (and optional) C<$callback> argument 219and they all accept an additional (and optional) C<$callback> argument
103which must be a code reference. This code reference will get called with 220which must be a code reference. This code reference will get called with
106syscall has been executed asynchronously. 223syscall has been executed asynchronously.
107 224
108All functions expecting a filehandle keep a copy of the filehandle 225All functions expecting a filehandle keep a copy of the filehandle
109internally until the request has finished. 226internally until the request has finished.
110 227
111All requests return objects of type L<IO::AIO::REQ> that allow further 228All functions return request objects of type L<IO::AIO::REQ> that allow
112manipulation of those requests while they are in-flight. 229further manipulation of those requests while they are in-flight.
113 230
114The pathnames you pass to these routines I<must> be absolute and 231The pathnames you pass to these routines I<must> be absolute and
115encoded in byte form. The reason for the former is that at the time the 232encoded as octets. The reason for the former is that at the time the
116request is being executed, the current working directory could have 233request is being executed, the current working directory could have
117changed. Alternatively, you can make sure that you never change the 234changed. Alternatively, you can make sure that you never change the
118current working directory. 235current working directory anywhere in the program and then use relative
236paths.
119 237
120To encode pathnames to byte form, either make sure you either: a) 238To encode pathnames as octets, either make sure you either: a) always pass
121always pass in filenames you got from outside (command line, readdir 239in filenames you got from outside (command line, readdir etc.) without
122etc.), b) are ASCII or ISO 8859-1, c) use the Encode module and encode 240tinkering, b) are ASCII or ISO 8859-1, c) use the Encode module and encode
123your pathnames to the locale (or other) encoding in effect in the user 241your pathnames to the locale (or other) encoding in effect in the user
124environment, d) use Glib::filename_from_unicode on unicode filenames or e) 242environment, d) use Glib::filename_from_unicode on unicode filenames or e)
125use something else. 243use something else to ensure your scalar has the correct contents.
244
245This works, btw. independent of the internal UTF-8 bit, which IO::AIO
246handles correctly wether it is set or not.
126 247
127=over 4 248=over 4
128 249
129=item aioreq_pri $pri 250=item $prev_pri = aioreq_pri [$pri]
130 251
131Sets the priority for the next aio request. The default priority 252Returns the priority value that would be used for the next request and, if
253C<$pri> is given, sets the priority for the next aio request.
254
132is C<0>, the minimum and maximum priorities are C<-4> and C<4>, 255The default priority is C<0>, the minimum and maximum priorities are C<-4>
133respectively. Requests with higher priority will be serviced first. 256and C<4>, respectively. Requests with higher priority will be serviced
257first.
134 258
135The priority will be reset to C<0> after each call to one of the C<aio_> 259The priority will be reset to C<0> after each call to one of the C<aio_*>
136functions. 260functions.
261
262Example: open a file with low priority, then read something from it with
263higher priority so the read request is serviced before other low priority
264open requests (potentially spamming the cache):
265
266 aioreq_pri -3;
267 aio_open ..., sub {
268 return unless $_[0];
269
270 aioreq_pri -2;
271 aio_read $_[0], ..., sub {
272 ...
273 };
274 };
275
276=item aioreq_nice $pri_adjust
277
278Similar to C<aioreq_pri>, but subtracts the given value from the current
279priority, so the effect is cumulative.
137 280
138=item aio_open $pathname, $flags, $mode, $callback->($fh) 281=item aio_open $pathname, $flags, $mode, $callback->($fh)
139 282
140Asynchronously open or create a file and call the callback with a newly 283Asynchronously open or create a file and call the callback with a newly
141created filehandle for the file. 284created filehandle for the file.
192 aio_read $fh, 7, 15, $buffer, 0, sub { 335 aio_read $fh, 7, 15, $buffer, 0, sub {
193 $_[0] > 0 or die "read error: $!"; 336 $_[0] > 0 or die "read error: $!";
194 print "read $_[0] bytes: <$buffer>\n"; 337 print "read $_[0] bytes: <$buffer>\n";
195 }; 338 };
196 339
197=item aio_move $srcpath, $dstpath, $callback->($status)
198
199[EXPERIMENTAL due to internal aio_group use]
200
201Try to move the I<file> (directories not supported as either source or
202destination) from C<$srcpath> to C<$dstpath> and call the callback with
203the C<0> (error) or C<-1> ok.
204
205This is a composite request that tries to rename(2) the file first. If
206rename files with C<EXDEV>, it creates the destination file with mode 0200
207and copies the contents of the source file into it using C<aio_sendfile>,
208followed by restoring atime, mtime, access mode and uid/gid, in that
209order, and unlinking the C<$srcpath>.
210
211If an error occurs, the partial destination file will be unlinked, if
212possible, except when setting atime, mtime, access mode and uid/gid, where
213errors are being ignored.
214
215=cut
216
217sub aio_move($$$) {
218 my ($src, $dst, $cb) = @_;
219
220 my $grp = aio_group $cb;
221
222 add $grp aio_rename $src, $dst, sub {
223 if ($_[0] && $! == EXDEV) {
224 add $grp aio_open $src, O_RDONLY, 0, sub {
225 if (my $src_fh = $_[0]) {
226 my @stat = stat $src_fh;
227
228 add $grp aio_open $dst, O_WRONLY, 0200, sub {
229 if (my $dst_fh = $_[0]) {
230 add $grp aio_sendfile $dst_fh, $src_fh, 0, $stat[7], sub {
231 close $src_fh;
232
233 if ($_[0] == $stat[7]) {
234 utime $stat[8], $stat[9], $dst;
235 chmod $stat[2] & 07777, $dst_fh;
236 chown $stat[4], $stat[5], $dst_fh;
237 close $dst_fh;
238
239 add $grp aio_unlink $src, sub {
240 $grp->result ($_[0]);
241 };
242 } else {
243 my $errno = $!;
244 add $grp aio_unlink $dst, sub {
245 $! = $errno;
246 $grp->result (-1);
247 };
248 }
249 };
250 } else {
251 $grp->result (-1);
252 }
253 },
254
255 } else {
256 $grp->result (-1);
257 }
258 };
259 } else {
260 $grp->result ($_[0]);
261 }
262 };
263
264 $grp
265}
266
267=item aio_sendfile $out_fh, $in_fh, $in_offset, $length, $callback->($retval) 340=item aio_sendfile $out_fh, $in_fh, $in_offset, $length, $callback->($retval)
268 341
269Tries to copy C<$length> bytes from C<$in_fh> to C<$out_fh>. It starts 342Tries to copy C<$length> bytes from C<$in_fh> to C<$out_fh>. It starts
270reading at byte offset C<$in_offset>, and starts writing at the current 343reading at byte offset C<$in_offset>, and starts writing at the current
271file offset of C<$out_fh>. Because of that, it is not safe to issue more 344file offset of C<$out_fh>. Because of that, it is not safe to issue more
326=item aio_unlink $pathname, $callback->($status) 399=item aio_unlink $pathname, $callback->($status)
327 400
328Asynchronously unlink (delete) a file and call the callback with the 401Asynchronously unlink (delete) a file and call the callback with the
329result code. 402result code.
330 403
404=item aio_mknod $path, $mode, $dev, $callback->($status)
405
406[EXPERIMENTAL]
407
408Asynchronously create a device node (or fifo). See mknod(2).
409
410The only (POSIX-) portable way of calling this function is:
411
412 aio_mknod $path, IO::AIO::S_IFIFO | $mode, 0, sub { ...
413
331=item aio_link $srcpath, $dstpath, $callback->($status) 414=item aio_link $srcpath, $dstpath, $callback->($status)
332 415
333Asynchronously create a new link to the existing object at C<$srcpath> at 416Asynchronously create a new link to the existing object at C<$srcpath> at
334the path C<$dstpath> and call the callback with the result code. 417the path C<$dstpath> and call the callback with the result code.
335 418
336=item aio_symlink $srcpath, $dstpath, $callback->($status) 419=item aio_symlink $srcpath, $dstpath, $callback->($status)
337 420
338Asynchronously create a new symbolic link to the existing object at C<$srcpath> at 421Asynchronously create a new symbolic link to the existing object at C<$srcpath> at
339the path C<$dstpath> and call the callback with the result code. 422the path C<$dstpath> and call the callback with the result code.
423
424=item aio_readlink $path, $callback->($link)
425
426Asynchronously read the symlink specified by C<$path> and pass it to
427the callback. If an error occurs, nothing or undef gets passed to the
428callback.
340 429
341=item aio_rename $srcpath, $dstpath, $callback->($status) 430=item aio_rename $srcpath, $dstpath, $callback->($status)
342 431
343Asynchronously rename the object at C<$srcpath> to C<$dstpath>, just as 432Asynchronously rename the object at C<$srcpath> to C<$dstpath>, just as
344rename(2) and call the callback with the result code. 433rename(2) and call the callback with the result code.
355sorted, and will B<NOT> include the C<.> and C<..> entries. 444sorted, and will B<NOT> include the C<.> and C<..> entries.
356 445
357The callback a single argument which is either C<undef> or an array-ref 446The callback a single argument which is either C<undef> or an array-ref
358with the filenames. 447with the filenames.
359 448
449=item aio_load $path, $data, $callback->($status)
450
451This is a composite request that tries to fully load the given file into
452memory. Status is the same as with aio_read.
453
454=cut
455
456sub aio_load($$;$) {
457 aio_block {
458 my ($path, undef, $cb) = @_;
459 my $data = \$_[1];
460
461 my $pri = aioreq_pri;
462 my $grp = aio_group $cb;
463
464 aioreq_pri $pri;
465 add $grp aio_open $path, O_RDONLY, 0, sub {
466 my ($fh) = @_
467 or return $grp->result (-1);
468
469 aioreq_pri $pri;
470 add $grp aio_read $fh, 0, (-s $fh), $$data, 0, sub {
471 $grp->result ($_[0]);
472 };
473 };
474
475 $grp
476 }
477}
478
479=item aio_copy $srcpath, $dstpath, $callback->($status)
480
481Try to copy the I<file> (directories not supported as either source or
482destination) from C<$srcpath> to C<$dstpath> and call the callback with
483the C<0> (error) or C<-1> ok.
484
485This is a composite request that it creates the destination file with
486mode 0200 and copies the contents of the source file into it using
487C<aio_sendfile>, followed by restoring atime, mtime, access mode and
488uid/gid, in that order.
489
490If an error occurs, the partial destination file will be unlinked, if
491possible, except when setting atime, mtime, access mode and uid/gid, where
492errors are being ignored.
493
494=cut
495
496sub aio_copy($$;$) {
497 aio_block {
498 my ($src, $dst, $cb) = @_;
499
500 my $pri = aioreq_pri;
501 my $grp = aio_group $cb;
502
503 aioreq_pri $pri;
504 add $grp aio_open $src, O_RDONLY, 0, sub {
505 if (my $src_fh = $_[0]) {
506 my @stat = stat $src_fh;
507
508 aioreq_pri $pri;
509 add $grp aio_open $dst, O_CREAT | O_WRONLY | O_TRUNC, 0200, sub {
510 if (my $dst_fh = $_[0]) {
511 aioreq_pri $pri;
512 add $grp aio_sendfile $dst_fh, $src_fh, 0, $stat[7], sub {
513 if ($_[0] == $stat[7]) {
514 $grp->result (0);
515 close $src_fh;
516
517 # those should not normally block. should. should.
518 utime $stat[8], $stat[9], $dst;
519 chmod $stat[2] & 07777, $dst_fh;
520 chown $stat[4], $stat[5], $dst_fh;
521 close $dst_fh;
522 } else {
523 $grp->result (-1);
524 close $src_fh;
525 close $dst_fh;
526
527 aioreq $pri;
528 add $grp aio_unlink $dst;
529 }
530 };
531 } else {
532 $grp->result (-1);
533 }
534 },
535
536 } else {
537 $grp->result (-1);
538 }
539 };
540
541 $grp
542 }
543}
544
545=item aio_move $srcpath, $dstpath, $callback->($status)
546
547Try to move the I<file> (directories not supported as either source or
548destination) from C<$srcpath> to C<$dstpath> and call the callback with
549the C<0> (error) or C<-1> ok.
550
551This is a composite request that tries to rename(2) the file first. If
552rename files with C<EXDEV>, it copies the file with C<aio_copy> and, if
553that is successful, unlinking the C<$srcpath>.
554
555=cut
556
557sub aio_move($$;$) {
558 aio_block {
559 my ($src, $dst, $cb) = @_;
560
561 my $pri = aioreq_pri;
562 my $grp = aio_group $cb;
563
564 aioreq_pri $pri;
565 add $grp aio_rename $src, $dst, sub {
566 if ($_[0] && $! == EXDEV) {
567 aioreq_pri $pri;
568 add $grp aio_copy $src, $dst, sub {
569 $grp->result ($_[0]);
570
571 if (!$_[0]) {
572 aioreq_pri $pri;
573 add $grp aio_unlink $src;
574 }
575 };
576 } else {
577 $grp->result ($_[0]);
578 }
579 };
580
581 $grp
582 }
583}
584
360=item aio_scandir $path, $maxreq, $callback->($dirs, $nondirs) 585=item aio_scandir $path, $maxreq, $callback->($dirs, $nondirs)
361 586
362[EXPERIMENTAL due to internal aio_group use]
363
364Scans a directory (similar to C<aio_readdir>) but additionally tries to 587Scans a directory (similar to C<aio_readdir>) but additionally tries to
365separate the entries of directory C<$path> into two sets of names, ones 588efficiently separate the entries of directory C<$path> into two sets of
366you can recurse into (directories or links to them), and ones you cannot 589names, directories you can recurse into (directories), and ones you cannot
367recurse into (everything else). 590recurse into (everything else, including symlinks to directories).
368 591
369C<aio_scandir> is a composite request that creates of many sub requests_ 592C<aio_scandir> is a composite request that creates of many sub requests_
370C<$maxreq> specifies the maximum number of outstanding aio requests that 593C<$maxreq> specifies the maximum number of outstanding aio requests that
371this function generates. If it is C<< <= 0 >>, then a suitable default 594this function generates. If it is C<< <= 0 >>, then a suitable default
372will be chosen (currently 6). 595will be chosen (currently 4).
373 596
374On error, the callback is called without arguments, otherwise it receives 597On error, the callback is called without arguments, otherwise it receives
375two array-refs with path-relative entry names. 598two array-refs with path-relative entry names.
376 599
377Example: 600Example:
412directory counting heuristic. 635directory counting heuristic.
413 636
414=cut 637=cut
415 638
416sub aio_scandir($$$) { 639sub aio_scandir($$$) {
640 aio_block {
417 my ($path, $maxreq, $cb) = @_; 641 my ($path, $maxreq, $cb) = @_;
418 642
643 my $pri = aioreq_pri;
644
419 my $grp = aio_group $cb; 645 my $grp = aio_group $cb;
420 646
421 $maxreq = 6 if $maxreq <= 0; 647 $maxreq = 4 if $maxreq <= 0;
422 648
423 # stat once 649 # stat once
650 aioreq_pri $pri;
424 add $grp aio_stat $path, sub { 651 add $grp aio_stat $path, sub {
425 return $grp->result () if $_[0]; 652 return $grp->result () if $_[0];
426 my $now = time; 653 my $now = time;
427 my $hash1 = join ":", (stat _)[0,1,3,7,9]; 654 my $hash1 = join ":", (stat _)[0,1,3,7,9];
428 655
429 # read the directory entries 656 # read the directory entries
657 aioreq_pri $pri;
430 add $grp aio_readdir $path, sub { 658 add $grp aio_readdir $path, sub {
431 my $entries = shift 659 my $entries = shift
432 or return $grp->result (); 660 or return $grp->result ();
433 661
434 # stat the dir another time 662 # stat the dir another time
663 aioreq_pri $pri;
435 add $grp aio_stat $path, sub { 664 add $grp aio_stat $path, sub {
436 my $hash2 = join ":", (stat _)[0,1,3,7,9]; 665 my $hash2 = join ":", (stat _)[0,1,3,7,9];
437 666
438 my $ndirs; 667 my $ndirs;
439 668
440 # take the slow route if anything looks fishy 669 # take the slow route if anything looks fishy
441 if ($hash1 ne $hash2 or (stat _)[9] == $now) { 670 if ($hash1 ne $hash2 or (stat _)[9] == $now) {
442 $ndirs = -1; 671 $ndirs = -1;
443 } else { 672 } else {
444 # if nlink == 2, we are finished 673 # if nlink == 2, we are finished
445 # on non-posix-fs's, we rely on nlink < 2 674 # on non-posix-fs's, we rely on nlink < 2
446 $ndirs = (stat _)[3] - 2 675 $ndirs = (stat _)[3] - 2
447 or return $grp->result ([], $entries); 676 or return $grp->result ([], $entries);
448 } 677 }
449 678
450 # sort into likely dirs and likely nondirs 679 # sort into likely dirs and likely nondirs
451 # dirs == files without ".", short entries first 680 # dirs == files without ".", short entries first
452 $entries = [map $_->[0], 681 $entries = [map $_->[0],
453 sort { $b->[1] cmp $a->[1] } 682 sort { $b->[1] cmp $a->[1] }
454 map [$_, sprintf "%s%04d", (/.\./ ? "1" : "0"), length], 683 map [$_, sprintf "%s%04d", (/.\./ ? "1" : "0"), length],
455 @$entries]; 684 @$entries];
456 685
457 my (@dirs, @nondirs); 686 my (@dirs, @nondirs);
458 687
459 my ($statcb, $schedcb);
460 my $nreq = 0;
461
462 my $statgrp = add $grp aio_group; 688 my $statgrp = add $grp aio_group sub {
463
464 $schedcb = sub {
465 if (@$entries) {
466 if ($nreq < $maxreq) {
467 my $ent = pop @$entries;
468 $nreq++;
469 add $statgrp aio_stat "$path/$ent/.", sub { $statcb->($_[0], $ent) };
470 }
471 } elsif (!$nreq) {
472 # finished
473 $statgrp->cancel;
474 undef $statcb;
475 undef $schedcb;
476 $grp->result (\@dirs, \@nondirs); 689 $grp->result (\@dirs, \@nondirs);
477 } 690 };
691
692 limit $statgrp $maxreq;
693 feed $statgrp sub {
694 return unless @$entries;
695 my $entry = pop @$entries;
696
697 aioreq_pri $pri;
698 add $statgrp aio_stat "$path/$entry/.", sub {
699 if ($_[0] < 0) {
700 push @nondirs, $entry;
701 } else {
702 # need to check for real directory
703 aioreq_pri $pri;
704 add $statgrp aio_lstat "$path/$entry", sub {
705 if (-d _) {
706 push @dirs, $entry;
707
708 unless (--$ndirs) {
709 push @nondirs, @$entries;
710 feed $statgrp;
711 }
712 } else {
713 push @nondirs, $entry;
714 }
715 }
716 }
717 };
718 };
478 }; 719 };
479 $statcb = sub {
480 my ($status, $entry) = @_;
481
482 if ($status < 0) {
483 $nreq--;
484 push @nondirs, $entry;
485 &$schedcb;
486 } else {
487 # need to check for real directory
488 add $grp aio_lstat "$path/$entry", sub {
489 $nreq--;
490
491 if (-d _) {
492 push @dirs, $entry;
493
494 if (!--$ndirs) {
495 push @nondirs, @$entries;
496 $entries = [];
497 }
498 } else {
499 push @nondirs, $entry;
500 }
501
502 &$schedcb;
503 }
504 }
505 };
506
507 &$schedcb while @$entries && $nreq < $maxreq;
508 }; 720 };
509 }; 721 };
722
723 $grp
510 }; 724 }
511
512 $grp
513} 725}
514 726
515=item aio_fsync $fh, $callback->($status) 727=item aio_fsync $fh, $callback->($status)
516 728
517Asynchronously call fsync on the given filehandle and call the callback 729Asynchronously call fsync on the given filehandle and call the callback
525If this call isn't available because your OS lacks it or it couldn't be 737If this call isn't available because your OS lacks it or it couldn't be
526detected, it will be emulated by calling C<fsync> instead. 738detected, it will be emulated by calling C<fsync> instead.
527 739
528=item aio_group $callback->(...) 740=item aio_group $callback->(...)
529 741
530[EXPERIMENTAL]
531
532This is a very special aio request: Instead of doing something, it is a 742This is a very special aio request: Instead of doing something, it is a
533container for other aio requests, which is useful if you want to bundle 743container for other aio requests, which is useful if you want to bundle
534many requests into a single, composite, request. 744many requests into a single, composite, request with a definite callback
745and the ability to cancel the whole request with its subrequests.
535 746
536Returns an object of class L<IO::AIO::GRP>. See its documentation below 747Returns an object of class L<IO::AIO::GRP>. See its documentation below
537for more info. 748for more info.
538 749
539Example: 750Example:
558phase and still requires a worker thread. Thus, the callback will not 769phase and still requires a worker thread. Thus, the callback will not
559be executed immediately but only after other requests in the queue have 770be executed immediately but only after other requests in the queue have
560entered their execution phase. This can be used to measure request 771entered their execution phase. This can be used to measure request
561latency. 772latency.
562 773
563=item IO::AIO::aio_sleep $fractional_seconds, $callback->() *NOT EXPORTED* 774=item IO::AIO::aio_busy $fractional_seconds, $callback->() *NOT EXPORTED*
564 775
565Mainly used for debugging and benchmarking, this aio request puts one of 776Mainly used for debugging and benchmarking, this aio request puts one of
566the request workers to sleep for the given time. 777the request workers to sleep for the given time.
567 778
568While it is theoretically handy to have simple I/O scheduling requests 779While it is theoretically handy to have simple I/O scheduling requests
569like sleep and file handle readable/writable, the overhead this creates 780like sleep and file handle readable/writable, the overhead this creates is
570is immense, so do not use this function except to put your application 781immense (it blocks a thread for a long time) so do not use this function
571under artificial I/O pressure. 782except to put your application under artificial I/O pressure.
572 783
573=back 784=back
574 785
575=head2 IO::AIO::REQ CLASS 786=head2 IO::AIO::REQ CLASS
576 787
577All non-aggregate C<aio_*> functions return an object of this class when 788All non-aggregate C<aio_*> functions return an object of this class when
578called in non-void context. 789called in non-void context.
579
580A request always moves through the following five states in its lifetime,
581in order: B<ready> (request has been created, but has not been executed
582yet), B<execute> (request is currently being executed), B<pending>
583(request has been executed but callback has not been called yet),
584B<result> (results are being processed synchronously, includes calling the
585callback) and B<done> (request has reached the end of its lifetime and
586holds no resources anymore).
587 790
588=over 4 791=over 4
589 792
590=item cancel $req 793=item cancel $req
591 794
645=item * They can also can also be added to other IO::AIO::GRP objects. 848=item * They can also can also be added to other IO::AIO::GRP objects.
646 849
647=item * You must not add requests to a group from within the group callback (or 850=item * You must not add requests to a group from within the group callback (or
648any later time). 851any later time).
649 852
650=item * This does not harmonise well with C<max_outstanding>, so best do
651not combine C<aio_group> with it. Groups and feeders are recommended for
652this kind of concurrency-limiting.
653
654=back 853=back
655 854
656Their lifetime, simplified, looks like this: when they are empty, they 855Their lifetime, simplified, looks like this: when they are empty, they
657will finish very quickly. If they contain only requests that are in the 856will finish very quickly. If they contain only requests that are in the
658C<done> state, they will also finish. Otherwise they will continue to 857C<done> state, they will also finish. Otherwise they will continue to
673be added, including other groups, as long as you do not create circular 872be added, including other groups, as long as you do not create circular
674dependencies. 873dependencies.
675 874
676Returns all its arguments. 875Returns all its arguments.
677 876
877=item $grp->cancel_subs
878
879Cancel all subrequests and clears any feeder, but not the group request
880itself. Useful when you queued a lot of events but got a result early.
881
678=item $grp->result (...) 882=item $grp->result (...)
679 883
680Set the result value(s) that will be passed to the group callback when all 884Set the result value(s) that will be passed to the group callback when all
681subrequests have finished. By default, no argument will be passed. 885subrequests have finished and set thre groups errno to the current value
886of errno (just like calling C<errno> without an error number). By default,
887no argument will be passed and errno is zero.
888
889=item $grp->errno ([$errno])
890
891Sets the group errno value to C<$errno>, or the current value of errno
892when the argument is missing.
893
894Every aio request has an associated errno value that is restored when
895the callback is invoked. This method lets you change this value from its
896default (0).
897
898Calling C<result> will also set errno, so make sure you either set C<$!>
899before the call to C<result>, or call c<errno> after it.
682 900
683=item feed $grp $callback->($grp) 901=item feed $grp $callback->($grp)
684
685[VERY EXPERIMENTAL]
686 902
687Sets a feeder/generator on this group: every group can have an attached 903Sets a feeder/generator on this group: every group can have an attached
688generator that generates requests if idle. The idea behind this is that, 904generator that generates requests if idle. The idea behind this is that,
689although you could just queue as many requests as you want in a group, 905although you could just queue as many requests as you want in a group,
690this might starve other requests for a potentially long time. For 906this might starve other requests for a potentially long time. For
727 943
728=back 944=back
729 945
730=head2 SUPPORT FUNCTIONS 946=head2 SUPPORT FUNCTIONS
731 947
948=head3 EVENT PROCESSING AND EVENT LOOP INTEGRATION
949
732=over 4 950=over 4
733 951
734=item $fileno = IO::AIO::poll_fileno 952=item $fileno = IO::AIO::poll_fileno
735 953
736Return the I<request result pipe file descriptor>. This filehandle must be 954Return the I<request result pipe file descriptor>. This filehandle must be
740 958
741See C<poll_cb> for an example. 959See C<poll_cb> for an example.
742 960
743=item IO::AIO::poll_cb 961=item IO::AIO::poll_cb
744 962
745Process all outstanding events on the result pipe. You have to call this 963Process some outstanding events on the result pipe. You have to call this
746regularly. Returns the number of events processed. Returns immediately 964regularly. Returns the number of events processed. Returns immediately
747when no events are outstanding. 965when no events are outstanding. The amount of events processed depends on
966the settings of C<IO::AIO::max_poll_req> and C<IO::AIO::max_poll_time>.
967
968If not all requests were processed for whatever reason, the filehandle
969will still be ready when C<poll_cb> returns.
748 970
749Example: Install an Event watcher that automatically calls 971Example: Install an Event watcher that automatically calls
750IO::AIO::poll_cb with high priority: 972IO::AIO::poll_cb with high priority:
751 973
752 Event->io (fd => IO::AIO::poll_fileno, 974 Event->io (fd => IO::AIO::poll_fileno,
753 poll => 'r', async => 1, 975 poll => 'r', async => 1,
754 cb => \&IO::AIO::poll_cb); 976 cb => \&IO::AIO::poll_cb);
755 977
978=item IO::AIO::max_poll_reqs $nreqs
979
980=item IO::AIO::max_poll_time $seconds
981
982These set the maximum number of requests (default C<0>, meaning infinity)
983that are being processed by C<IO::AIO::poll_cb> in one call, respectively
984the maximum amount of time (default C<0>, meaning infinity) spent in
985C<IO::AIO::poll_cb> to process requests (more correctly the mininum amount
986of time C<poll_cb> is allowed to use).
987
988Setting C<max_poll_time> to a non-zero value creates an overhead of one
989syscall per request processed, which is not normally a problem unless your
990callbacks are really really fast or your OS is really really slow (I am
991not mentioning Solaris here). Using C<max_poll_reqs> incurs no overhead.
992
993Setting these is useful if you want to ensure some level of
994interactiveness when perl is not fast enough to process all requests in
995time.
996
997For interactive programs, values such as C<0.01> to C<0.1> should be fine.
998
999Example: Install an Event watcher that automatically calls
1000IO::AIO::poll_cb with low priority, to ensure that other parts of the
1001program get the CPU sometimes even under high AIO load.
1002
1003 # try not to spend much more than 0.1s in poll_cb
1004 IO::AIO::max_poll_time 0.1;
1005
1006 # use a low priority so other tasks have priority
1007 Event->io (fd => IO::AIO::poll_fileno,
1008 poll => 'r', nice => 1,
1009 cb => &IO::AIO::poll_cb);
1010
756=item IO::AIO::poll_wait 1011=item IO::AIO::poll_wait
757 1012
1013If there are any outstanding requests and none of them in the result
758Wait till the result filehandle becomes ready for reading (simply does a 1014phase, wait till the result filehandle becomes ready for reading (simply
759C<select> on the filehandle. This is useful if you want to synchronously wait 1015does a C<select> on the filehandle. This is useful if you want to
760for some requests to finish). 1016synchronously wait for some requests to finish).
761 1017
762See C<nreqs> for an example. 1018See C<nreqs> for an example.
763 1019
1020=item IO::AIO::poll
1021
1022Waits until some requests have been handled.
1023
1024Returns the number of requests processed, but is otherwise strictly
1025equivalent to:
1026
1027 IO::AIO::poll_wait, IO::AIO::poll_cb
1028
764=item IO::AIO::nreqs 1029=item IO::AIO::flush
765 1030
766Returns the number of requests currently outstanding (i.e. for which their 1031Wait till all outstanding AIO requests have been handled.
767callback has not been invoked yet).
768 1032
769Example: wait till there are no outstanding requests anymore: 1033Strictly equivalent to:
770 1034
771 IO::AIO::poll_wait, IO::AIO::poll_cb 1035 IO::AIO::poll_wait, IO::AIO::poll_cb
772 while IO::AIO::nreqs; 1036 while IO::AIO::nreqs;
773 1037
774=item IO::AIO::flush 1038=head3 CONTROLLING THE NUMBER OF THREADS
775
776Wait till all outstanding AIO requests have been handled.
777
778Strictly equivalent to:
779
780 IO::AIO::poll_wait, IO::AIO::poll_cb
781 while IO::AIO::nreqs;
782
783=item IO::AIO::poll
784
785Waits until some requests have been handled.
786
787Strictly equivalent to:
788
789 IO::AIO::poll_wait, IO::AIO::poll_cb
790 if IO::AIO::nreqs;
791 1039
792=item IO::AIO::min_parallel $nthreads 1040=item IO::AIO::min_parallel $nthreads
793 1041
794Set the minimum number of AIO threads to C<$nthreads>. The current 1042Set the minimum number of AIO threads to C<$nthreads>. The current
795default is C<8>, which means eight asynchronous operations can execute 1043default is C<8>, which means eight asynchronous operations can execute
796concurrently at any one time (the number of outstanding requests, 1044concurrently at any one time (the number of outstanding requests,
797however, is unlimited). 1045however, is unlimited).
798 1046
799IO::AIO starts threads only on demand, when an AIO request is queued and 1047IO::AIO starts threads only on demand, when an AIO request is queued and
800no free thread exists. 1048no free thread exists. Please note that queueing up a hundred requests can
1049create demand for a hundred threads, even if it turns out that everything
1050is in the cache and could have been processed faster by a single thread.
801 1051
802It is recommended to keep the number of threads relatively low, as some 1052It is recommended to keep the number of threads relatively low, as some
803Linux kernel versions will scale negatively with the number of threads 1053Linux kernel versions will scale negatively with the number of threads
804(higher parallelity => MUCH higher latency). With current Linux 2.6 1054(higher parallelity => MUCH higher latency). With current Linux 2.6
805versions, 4-32 threads should be fine. 1055versions, 4-32 threads should be fine.
819This module automatically runs C<max_parallel 0> at program end, to ensure 1069This module automatically runs C<max_parallel 0> at program end, to ensure
820that all threads are killed and that there are no outstanding requests. 1070that all threads are killed and that there are no outstanding requests.
821 1071
822Under normal circumstances you don't need to call this function. 1072Under normal circumstances you don't need to call this function.
823 1073
1074=item IO::AIO::max_idle $nthreads
1075
1076Limit the number of threads (default: 4) that are allowed to idle (i.e.,
1077threads that did not get a request to process within 10 seconds). That
1078means if a thread becomes idle while C<$nthreads> other threads are also
1079idle, it will free its resources and exit.
1080
1081This is useful when you allow a large number of threads (e.g. 100 or 1000)
1082to allow for extremely high load situations, but want to free resources
1083under normal circumstances (1000 threads can easily consume 30MB of RAM).
1084
1085The default is probably ok in most situations, especially if thread
1086creation is fast. If thread creation is very slow on your system you might
1087want to use larger values.
1088
824=item $oldnreqs = IO::AIO::max_outstanding $nreqs 1089=item $oldmaxreqs = IO::AIO::max_outstanding $maxreqs
825 1090
826[DEPRECATED] 1091This is a very bad function to use in interactive programs because it
1092blocks, and a bad way to reduce concurrency because it is inexact: Better
1093use an C<aio_group> together with a feed callback.
827 1094
828Sets the maximum number of outstanding requests to C<$nreqs>. If you 1095Sets the maximum number of outstanding requests to C<$nreqs>. If you
829try to queue up more than this number of requests, the caller will block until 1096to queue up more than this number of requests, the next call to the
830some requests have been handled. 1097C<poll_cb> (and C<poll_some> and other functions calling C<poll_cb>)
1098function will block until the limit is no longer exceeded.
831 1099
832The default is very large, so normally there is no practical limit. If you 1100The default value is very large, so there is no practical limit on the
833queue up many requests in a loop it often improves speed if you set 1101number of outstanding requests.
834this to a relatively low number, such as C<100>.
835 1102
836This function does not work well together with C<aio_group>'s, and their 1103You can still queue as many requests as you want. Therefore,
837feeder interface is better suited to limiting concurrency, so do not use 1104C<max_oustsanding> is mainly useful in simple scripts (with low values) or
838this function. 1105as a stop gap to shield against fatal memory overflow (with large values).
839 1106
840Under normal circumstances you don't need to call this function. 1107=head3 STATISTICAL INFORMATION
1108
1109=item IO::AIO::nreqs
1110
1111Returns the number of requests currently in the ready, execute or pending
1112states (i.e. for which their callback has not been invoked yet).
1113
1114Example: wait till there are no outstanding requests anymore:
1115
1116 IO::AIO::poll_wait, IO::AIO::poll_cb
1117 while IO::AIO::nreqs;
1118
1119=item IO::AIO::nready
1120
1121Returns the number of requests currently in the ready state (not yet
1122executed).
1123
1124=item IO::AIO::npending
1125
1126Returns the number of requests currently in the pending state (executed,
1127but not yet processed by poll_cb).
841 1128
842=back 1129=back
843 1130
844=cut 1131=cut
845 1132
859 *$sym 1146 *$sym
860} 1147}
861 1148
862min_parallel 8; 1149min_parallel 8;
863 1150
864END { 1151END { flush }
865 max_parallel 0;
866}
867 1152
8681; 11531;
869 1154
870=head2 FORK BEHAVIOUR 1155=head2 FORK BEHAVIOUR
871 1156
872This module should do "the right thing" when the process using it forks: 1157This module should do "the right thing" when the process using it forks:
873 1158
874Before the fork, IO::AIO enters a quiescent state where no requests 1159Before the fork, IO::AIO enters a quiescent state where no requests
875can be added in other threads and no results will be processed. After 1160can be added in other threads and no results will be processed. After
876the fork the parent simply leaves the quiescent state and continues 1161the fork the parent simply leaves the quiescent state and continues
877request/result processing, while the child clears the request/result 1162request/result processing, while the child frees the request/result queue
878queue (so the requests started before the fork will only be handled in 1163(so that the requests started before the fork will only be handled in the
879the parent). Threads will be started on demand until the limit ste in the 1164parent). Threads will be started on demand until the limit set in the
880parent process has been reached again. 1165parent process has been reached again.
881 1166
882In short: the parent will, after a short pause, continue as if fork had 1167In short: the parent will, after a short pause, continue as if fork had
883not been called, while the child will act as if IO::AIO has not been used 1168not been called, while the child will act as if IO::AIO has not been used
884yet. 1169yet.
885 1170
886=head2 MEMORY USAGE 1171=head2 MEMORY USAGE
887 1172
1173Per-request usage:
1174
888Each aio request uses - depending on your architecture - around 128 bytes 1175Each aio request uses - depending on your architecture - around 100-200
889of memory. In addition, stat requests need a stat buffer (possibly a few 1176bytes of memory. In addition, stat requests need a stat buffer (possibly
890hundred bytes). Perl scalars and other data passed into aio requests will 1177a few hundred bytes), readdir requires a result buffer and so on. Perl
891also be locked. 1178scalars and other data passed into aio requests will also be locked and
1179will consume memory till the request has entered the done state.
892 1180
893This is now awfully much, so queuing lots of requests is not usually a 1181This is now awfully much, so queuing lots of requests is not usually a
894problem. 1182problem.
895 1183
896Each thread needs a stack area which is usually around 16k, sometimes much 1184Per-thread usage:
897larger, depending on the OS. 1185
1186In the execution phase, some aio requests require more memory for
1187temporary buffers, and each thread requires a stack and other data
1188structures (usually around 16k-128k, depending on the OS).
1189
1190=head1 KNOWN BUGS
1191
1192Known bugs will be fixed in the next release.
898 1193
899=head1 SEE ALSO 1194=head1 SEE ALSO
900 1195
901L<Coro::AIO>. 1196L<Coro::AIO>.
902 1197

Diff Legend

Removed lines
+ Added lines
< Changed lines
> Changed lines