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 1.83 by root, Fri Oct 27 20:11:58 2006 UTC vs.
Revision 1.116 by root, Wed Oct 3 21:27:51 2007 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
… …
50 51
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.
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
68example sockets), using these functions on file descriptors that
69support nonblocking operation (again, sockets, pipes etc.) is very
70inefficient. Use an event loop for that (such as the L<Event|Event>
71module): IO::AIO will naturally fit into such an event loop itself.
55 72
56In this version, a number of threads are started that execute your 73In this version, a number of threads are started that execute your
57requests and signal their completion. You don't need thread support 74requests and signal their completion. You don't need thread support
58in perl, and the threads created by this module will not be visible 75in perl, and the threads created by this module will not be visible
59to perl. In the future, this module might make use of the native aio 76to perl. In the future, this module might make use of the native aio
60functions available on many operating systems. However, they are often 77functions available on many operating systems. However, they are often
61not well-supported or restricted (Linux doesn't allow them on normal 78not well-supported or restricted (GNU/Linux doesn't allow them on normal
62files currently, for example), and they would only support aio_read and 79files currently, for example), and they would only support aio_read and
63aio_write, so the remaining functionality would have to be implemented 80aio_write, so the remaining functionality would have to be implemented
64using threads anyway. 81using threads anyway.
65 82
66Although the module will work with in the presence of other (Perl-) 83Although the module will work in the presence of other (Perl-) threads,
67threads, it is currently not reentrant in any way, so use appropriate 84it is currently not reentrant in any way, so use appropriate locking
68locking yourself, always call C<poll_cb> from within the same thread, or 85yourself, always call C<poll_cb> from within the same thread, or never
69never call C<poll_cb> (or other C<aio_> functions) recursively. 86call C<poll_cb> (or other C<aio_> functions) recursively.
87
88=head2 EXAMPLE
89
90This is a simple example that uses the Event module and loads
91F</etc/passwd> asynchronously:
92
93 use Fcntl;
94 use Event;
95 use IO::AIO;
96
97 # register the IO::AIO callback with Event
98 Event->io (fd => IO::AIO::poll_fileno,
99 poll => 'r',
100 cb => \&IO::AIO::poll_cb);
101
102 # queue the request to open /etc/passwd
103 aio_open "/etc/passwd", O_RDONLY, 0, sub {
104 my $fh = shift
105 or die "error while opening: $!";
106
107 # stat'ing filehandles is generally non-blocking
108 my $size = -s $fh;
109
110 # queue a request to read the file
111 my $contents;
112 aio_read $fh, 0, $size, $contents, 0, sub {
113 $_[0] == $size
114 or die "short read: $!";
115
116 close $fh;
117
118 # file contents now in $contents
119 print $contents;
120
121 # exit event loop and program
122 Event::unloop;
123 };
124 };
125
126 # possibly queue up other requests, or open GUI windows,
127 # check for sockets etc. etc.
128
129 # process events as long as there are some:
130 Event::loop;
70 131
71=head1 REQUEST ANATOMY AND LIFETIME 132=head1 REQUEST ANATOMY AND LIFETIME
72 133
73Every C<aio_*> function creates a request. which is a C data structure not 134Every C<aio_*> function creates a request. which is a C data structure not
74directly visible to Perl. 135directly visible to Perl.
… …
116Request has reached the end of its lifetime and holds no resources anymore 177Request has reached the end of its lifetime and holds no resources anymore
117(except possibly for the Perl object, but its connection to the actual 178(except possibly for the Perl object, but its connection to the actual
118aio request is severed and calling its methods will either do nothing or 179aio request is severed and calling its methods will either do nothing or
119result in a runtime error). 180result in a runtime error).
120 181
182=back
183
121=cut 184=cut
122 185
123package IO::AIO; 186package IO::AIO;
124 187
125no warnings; 188no warnings;
126use strict 'vars'; 189use strict 'vars';
127 190
128use base 'Exporter'; 191use base 'Exporter';
129 192
130BEGIN { 193BEGIN {
131 our $VERSION = '2.0'; 194 our $VERSION = '2.5';
132 195
133 our @AIO_REQ = qw(aio_sendfile aio_read aio_write aio_open aio_close aio_stat 196 our @AIO_REQ = qw(aio_sendfile aio_read aio_write aio_open aio_close aio_stat
134 aio_lstat aio_unlink aio_rmdir aio_readdir aio_scandir aio_symlink 197 aio_lstat aio_unlink aio_rmdir aio_readdir aio_scandir aio_symlink
135 aio_fsync aio_fdatasync aio_readahead aio_rename aio_link aio_move 198 aio_readlink aio_fsync aio_fdatasync aio_readahead aio_rename aio_link
136 aio_copy aio_group aio_nop aio_mknod); 199 aio_move aio_copy aio_group aio_nop aio_mknod aio_load aio_rmtree aio_mkdir
200 aio_chown aio_chmod aio_utime aio_truncate);
137 our @EXPORT = (@AIO_REQ, qw(aioreq_pri aioreq_nice)); 201 our @EXPORT = (@AIO_REQ, qw(aioreq_pri aioreq_nice aio_block));
138 our @EXPORT_OK = qw(poll_fileno poll_cb poll_wait flush 202 our @EXPORT_OK = qw(poll_fileno poll_cb poll_wait flush
139 min_parallel max_parallel nreqs nready npending); 203 min_parallel max_parallel max_idle
204 nreqs nready npending nthreads
205 max_poll_time max_poll_reqs);
140 206
141 @IO::AIO::GRP::ISA = 'IO::AIO::REQ'; 207 @IO::AIO::GRP::ISA = 'IO::AIO::REQ';
142 208
143 require XSLoader; 209 require XSLoader;
144 XSLoader::load ("IO::AIO", $VERSION); 210 XSLoader::load ("IO::AIO", $VERSION);
145} 211}
146 212
147=head1 FUNCTIONS 213=head1 FUNCTIONS
148 214
149=head2 AIO FUNCTIONS 215=head2 AIO REQUEST FUNCTIONS
150 216
151All the C<aio_*> calls are more or less thin wrappers around the syscall 217All the C<aio_*> calls are more or less thin wrappers around the syscall
152with the same name (sans C<aio_>). The arguments are similar or identical, 218with the same name (sans C<aio_>). The arguments are similar or identical,
153and they all accept an additional (and optional) C<$callback> argument 219and they all accept an additional (and optional) C<$callback> argument
154which 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
… …
157syscall has been executed asynchronously. 223syscall has been executed asynchronously.
158 224
159All functions expecting a filehandle keep a copy of the filehandle 225All functions expecting a filehandle keep a copy of the filehandle
160internally until the request has finished. 226internally until the request has finished.
161 227
162All 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
163manipulation of those requests while they are in-flight. 229further manipulation of those requests while they are in-flight.
164 230
165The pathnames you pass to these routines I<must> be absolute and 231The pathnames you pass to these routines I<must> be absolute and
166encoded 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
167request is being executed, the current working directory could have 233request is being executed, the current working directory could have
168changed. Alternatively, you can make sure that you never change the 234changed. Alternatively, you can make sure that you never change the
169current working directory. 235current working directory anywhere in the program and then use relative
236paths.
170 237
171To encode pathnames to byte form, either make sure you either: a) 238To encode pathnames as octets, either make sure you either: a) always pass
172always pass in filenames you got from outside (command line, readdir 239in filenames you got from outside (command line, readdir etc.) without
173etc.), 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
174your 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
175environment, d) use Glib::filename_from_unicode on unicode filenames or e) 242environment, d) use Glib::filename_from_unicode on unicode filenames or e)
176use 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.
177 247
178=over 4 248=over 4
179 249
180=item $prev_pri = aioreq_pri [$pri] 250=item $prev_pri = aioreq_pri [$pri]
181 251
… …
201 aio_read $_[0], ..., sub { 271 aio_read $_[0], ..., sub {
202 ... 272 ...
203 }; 273 };
204 }; 274 };
205 275
276
206=item aioreq_nice $pri_adjust 277=item aioreq_nice $pri_adjust
207 278
208Similar to C<aioreq_pri>, but subtracts the given value from the current 279Similar to C<aioreq_pri>, but subtracts the given value from the current
209priority, so effects are cumulative. 280priority, so the effect is cumulative.
281
210 282
211=item aio_open $pathname, $flags, $mode, $callback->($fh) 283=item aio_open $pathname, $flags, $mode, $callback->($fh)
212 284
213Asynchronously open or create a file and call the callback with a newly 285Asynchronously open or create a file and call the callback with a newly
214created filehandle for the file. 286created filehandle for the file.
… …
220list. They are the same as used by C<sysopen>. 292list. They are the same as used by C<sysopen>.
221 293
222Likewise, C<$mode> specifies the mode of the newly created file, if it 294Likewise, C<$mode> specifies the mode of the newly created file, if it
223didn't exist and C<O_CREAT> has been given, just like perl's C<sysopen>, 295didn't exist and C<O_CREAT> has been given, just like perl's C<sysopen>,
224except that it is mandatory (i.e. use C<0> if you don't create new files, 296except that it is mandatory (i.e. use C<0> if you don't create new files,
225and C<0666> or C<0777> if you do). 297and C<0666> or C<0777> if you do). Note that the C<$mode> will be modified
298by the umask in effect then the request is being executed, so better never
299change the umask.
226 300
227Example: 301Example:
228 302
229 aio_open "/etc/passwd", O_RDONLY, 0, sub { 303 aio_open "/etc/passwd", O_RDONLY, 0, sub {
230 if ($_[0]) { 304 if ($_[0]) {
… …
233 } else { 307 } else {
234 die "open failed: $!\n"; 308 die "open failed: $!\n";
235 } 309 }
236 }; 310 };
237 311
312
238=item aio_close $fh, $callback->($status) 313=item aio_close $fh, $callback->($status)
239 314
240Asynchronously close a file and call the callback with the result 315Asynchronously close a file and call the callback with the result
241code. I<WARNING:> although accepted, you should not pass in a perl 316code.
242filehandle here, as perl will likely close the file descriptor another
243time when the filehandle is destroyed. Normally, you can safely call perls
244C<close> or just let filehandles go out of scope.
245 317
246This is supposed to be a bug in the API, so that might change. It's 318Unlike the other functions operating on files, this function uses the
247therefore best to avoid this function. 319PerlIO layer to close the filehandle. The reason is that the PerlIO API
320insists on closing the underlying fd itself, no matter what, and doesn't
321allow modifications to the fd. Unfortunately, it is not clear that you can
322call PerlIO from different threads (actually, its quite clear that this
323won't work in some cases), so while it likely works perfectly with simple
324file handles (such as the ones created by C<aio_open>) it might fail in
325interesting ways for others.
326
327Having said that, aio_close tries to clean up the filehandle as much as
328possible before handing it to an io thread, and generally does work.
329
248 330
249=item aio_read $fh,$offset,$length, $data,$dataoffset, $callback->($retval) 331=item aio_read $fh,$offset,$length, $data,$dataoffset, $callback->($retval)
250 332
251=item aio_write $fh,$offset,$length, $data,$dataoffset, $callback->($retval) 333=item aio_write $fh,$offset,$length, $data,$dataoffset, $callback->($retval)
252 334
253Reads or writes C<length> bytes from the specified C<fh> and C<offset> 335Reads or writes C<$length> bytes from the specified C<$fh> and C<$offset>
254into the scalar given by C<data> and offset C<dataoffset> and calls the 336into the scalar given by C<$data> and offset C<$dataoffset> and calls the
255callback without the actual number of bytes read (or -1 on error, just 337callback without the actual number of bytes read (or -1 on error, just
256like the syscall). 338like the syscall).
257 339
340If C<$offset> is undefined, then the current file descriptor offset will
341be used (and updated), otherwise the file descriptor offset will not be
342changed by these calls.
343
344If C<$length> is undefined in C<aio_write>, use the remaining length of C<$data>.
345
346If C<$dataoffset> is less than zero, it will be counted from the end of
347C<$data>.
348
258The C<$data> scalar I<MUST NOT> be modified in any way while the request 349The C<$data> scalar I<MUST NOT> be modified in any way while the request
259is outstanding. Modifying it can result in segfaults or WW3 (if the 350is outstanding. Modifying it can result in segfaults or World War III (if
260necessary/optional hardware is installed). 351the necessary/optional hardware is installed).
261 352
262Example: Read 15 bytes at offset 7 into scalar C<$buffer>, starting at 353Example: Read 15 bytes at offset 7 into scalar C<$buffer>, starting at
263offset C<0> within the scalar: 354offset C<0> within the scalar:
264 355
265 aio_read $fh, 7, 15, $buffer, 0, sub { 356 aio_read $fh, 7, 15, $buffer, 0, sub {
266 $_[0] > 0 or die "read error: $!"; 357 $_[0] > 0 or die "read error: $!";
267 print "read $_[0] bytes: <$buffer>\n"; 358 print "read $_[0] bytes: <$buffer>\n";
268 }; 359 };
360
269 361
270=item aio_sendfile $out_fh, $in_fh, $in_offset, $length, $callback->($retval) 362=item aio_sendfile $out_fh, $in_fh, $in_offset, $length, $callback->($retval)
271 363
272Tries to copy C<$length> bytes from C<$in_fh> to C<$out_fh>. It starts 364Tries to copy C<$length> bytes from C<$in_fh> to C<$out_fh>. It starts
273reading at byte offset C<$in_offset>, and starts writing at the current 365reading at byte offset C<$in_offset>, and starts writing at the current
… …
287C<$in_fh> than are written, and there is no way to find out how many 379C<$in_fh> than are written, and there is no way to find out how many
288bytes have been read from C<aio_sendfile> alone, as C<aio_sendfile> only 380bytes have been read from C<aio_sendfile> alone, as C<aio_sendfile> only
289provides the number of bytes written to C<$out_fh>. Only if the result 381provides the number of bytes written to C<$out_fh>. Only if the result
290value equals C<$length> one can assume that C<$length> bytes have been 382value equals C<$length> one can assume that C<$length> bytes have been
291read. 383read.
384
292 385
293=item aio_readahead $fh,$offset,$length, $callback->($retval) 386=item aio_readahead $fh,$offset,$length, $callback->($retval)
294 387
295C<aio_readahead> populates the page cache with data from a file so that 388C<aio_readahead> populates the page cache with data from a file so that
296subsequent reads from that file will not block on disk I/O. The C<$offset> 389subsequent reads from that file will not block on disk I/O. The C<$offset>
… …
302file. The current file offset of the file is left unchanged. 395file. The current file offset of the file is left unchanged.
303 396
304If that syscall doesn't exist (likely if your OS isn't Linux) it will be 397If that syscall doesn't exist (likely if your OS isn't Linux) it will be
305emulated by simply reading the data, which would have a similar effect. 398emulated by simply reading the data, which would have a similar effect.
306 399
400
307=item aio_stat $fh_or_path, $callback->($status) 401=item aio_stat $fh_or_path, $callback->($status)
308 402
309=item aio_lstat $fh, $callback->($status) 403=item aio_lstat $fh, $callback->($status)
310 404
311Works like perl's C<stat> or C<lstat> in void context. The callback will 405Works like perl's C<stat> or C<lstat> in void context. The callback will
… …
324 aio_stat "/etc/passwd", sub { 418 aio_stat "/etc/passwd", sub {
325 $_[0] and die "stat failed: $!"; 419 $_[0] and die "stat failed: $!";
326 print "size is ", -s _, "\n"; 420 print "size is ", -s _, "\n";
327 }; 421 };
328 422
423
424=item aio_utime $fh_or_path, $atime, $mtime, $callback->($status)
425
426Works like perl's C<utime> function (including the special case of $atime
427and $mtime being undef). Fractional times are supported if the underlying
428syscalls support them.
429
430When called with a pathname, uses utimes(2) if available, otherwise
431utime(2). If called on a file descriptor, uses futimes(2) if available,
432otherwise returns ENOSYS, so this is not portable.
433
434Examples:
435
436 # set atime and mtime to current time (basically touch(1)):
437 aio_utime "path", undef, undef;
438 # set atime to current time and mtime to beginning of the epoch:
439 aio_utime "path", time, undef; # undef==0
440
441
442=item aio_chown $fh_or_path, $uid, $gid, $callback->($status)
443
444Works like perl's C<chown> function, except that C<undef> for either $uid
445or $gid is being interpreted as "do not change" (but -1 can also be used).
446
447Examples:
448
449 # same as "chown root path" in the shell:
450 aio_chown "path", 0, -1;
451 # same as above:
452 aio_chown "path", 0, undef;
453
454
455=item aio_truncate $fh_or_path, $offset, $callback->($status)
456
457Works like truncate(2) or ftruncate(2).
458
459
460=item aio_chmod $fh_or_path, $mode, $callback->($status)
461
462Works like perl's C<chmod> function.
463
464
329=item aio_unlink $pathname, $callback->($status) 465=item aio_unlink $pathname, $callback->($status)
330 466
331Asynchronously unlink (delete) a file and call the callback with the 467Asynchronously unlink (delete) a file and call the callback with the
332result code. 468result code.
333 469
470
334=item aio_mknod $path, $mode, $dev, $callback->($status) 471=item aio_mknod $path, $mode, $dev, $callback->($status)
335 472
473[EXPERIMENTAL]
474
336Asynchronously create a device node (or fifo). See mknod(2). 475Asynchronously create a device node (or fifo). See mknod(2).
337 476
338The only portable (POSIX) way of calling this function is: 477The only (POSIX-) portable way of calling this function is:
339 478
340 aio_mknod $path, IO::AIO::S_IFIFO | $mode, 0, sub { ... 479 aio_mknod $path, IO::AIO::S_IFIFO | $mode, 0, sub { ...
480
341 481
342=item aio_link $srcpath, $dstpath, $callback->($status) 482=item aio_link $srcpath, $dstpath, $callback->($status)
343 483
344Asynchronously create a new link to the existing object at C<$srcpath> at 484Asynchronously create a new link to the existing object at C<$srcpath> at
345the path C<$dstpath> and call the callback with the result code. 485the path C<$dstpath> and call the callback with the result code.
346 486
487
347=item aio_symlink $srcpath, $dstpath, $callback->($status) 488=item aio_symlink $srcpath, $dstpath, $callback->($status)
348 489
349Asynchronously create a new symbolic link to the existing object at C<$srcpath> at 490Asynchronously create a new symbolic link to the existing object at C<$srcpath> at
350the path C<$dstpath> and call the callback with the result code. 491the path C<$dstpath> and call the callback with the result code.
351 492
493
494=item aio_readlink $path, $callback->($link)
495
496Asynchronously read the symlink specified by C<$path> and pass it to
497the callback. If an error occurs, nothing or undef gets passed to the
498callback.
499
500
352=item aio_rename $srcpath, $dstpath, $callback->($status) 501=item aio_rename $srcpath, $dstpath, $callback->($status)
353 502
354Asynchronously rename the object at C<$srcpath> to C<$dstpath>, just as 503Asynchronously rename the object at C<$srcpath> to C<$dstpath>, just as
355rename(2) and call the callback with the result code. 504rename(2) and call the callback with the result code.
356 505
506
507=item aio_mkdir $pathname, $mode, $callback->($status)
508
509Asynchronously mkdir (create) a directory and call the callback with
510the result code. C<$mode> will be modified by the umask at the time the
511request is executed, so do not change your umask.
512
513
357=item aio_rmdir $pathname, $callback->($status) 514=item aio_rmdir $pathname, $callback->($status)
358 515
359Asynchronously rmdir (delete) a directory and call the callback with the 516Asynchronously rmdir (delete) a directory and call the callback with the
360result code. 517result code.
518
361 519
362=item aio_readdir $pathname, $callback->($entries) 520=item aio_readdir $pathname, $callback->($entries)
363 521
364Unlike the POSIX call of the same name, C<aio_readdir> reads an entire 522Unlike the POSIX call of the same name, C<aio_readdir> reads an entire
365directory (i.e. opendir + readdir + closedir). The entries will not be 523directory (i.e. opendir + readdir + closedir). The entries will not be
366sorted, and will B<NOT> include the C<.> and C<..> entries. 524sorted, and will B<NOT> include the C<.> and C<..> entries.
367 525
368The callback a single argument which is either C<undef> or an array-ref 526The callback a single argument which is either C<undef> or an array-ref
369with the filenames. 527with the filenames.
528
529
530=item aio_load $path, $data, $callback->($status)
531
532This is a composite request that tries to fully load the given file into
533memory. Status is the same as with aio_read.
534
535=cut
536
537sub aio_load($$;$) {
538 aio_block {
539 my ($path, undef, $cb) = @_;
540 my $data = \$_[1];
541
542 my $pri = aioreq_pri;
543 my $grp = aio_group $cb;
544
545 aioreq_pri $pri;
546 add $grp aio_open $path, O_RDONLY, 0, sub {
547 my $fh = shift
548 or return $grp->result (-1);
549
550 aioreq_pri $pri;
551 add $grp aio_read $fh, 0, (-s $fh), $$data, 0, sub {
552 $grp->result ($_[0]);
553 };
554 };
555
556 $grp
557 }
558}
370 559
371=item aio_copy $srcpath, $dstpath, $callback->($status) 560=item aio_copy $srcpath, $dstpath, $callback->($status)
372 561
373Try to copy the I<file> (directories not supported as either source or 562Try to copy the I<file> (directories not supported as either source or
374destination) from C<$srcpath> to C<$dstpath> and call the callback with 563destination) from C<$srcpath> to C<$dstpath> and call the callback with
… …
384errors are being ignored. 573errors are being ignored.
385 574
386=cut 575=cut
387 576
388sub aio_copy($$;$) { 577sub aio_copy($$;$) {
578 aio_block {
389 my ($src, $dst, $cb) = @_; 579 my ($src, $dst, $cb) = @_;
390 580
391 my $pri = aioreq_pri; 581 my $pri = aioreq_pri;
392 my $grp = aio_group $cb; 582 my $grp = aio_group $cb;
393 583
394 aioreq_pri $pri; 584 aioreq_pri $pri;
395 add $grp aio_open $src, O_RDONLY, 0, sub { 585 add $grp aio_open $src, O_RDONLY, 0, sub {
396 if (my $src_fh = $_[0]) { 586 if (my $src_fh = $_[0]) {
397 my @stat = stat $src_fh; 587 my @stat = stat $src_fh;
398 588
399 aioreq_pri $pri; 589 aioreq_pri $pri;
400 add $grp aio_open $dst, O_CREAT | O_WRONLY | O_TRUNC, 0200, sub { 590 add $grp aio_open $dst, O_CREAT | O_WRONLY | O_TRUNC, 0200, sub {
401 if (my $dst_fh = $_[0]) { 591 if (my $dst_fh = $_[0]) {
402 aioreq_pri $pri; 592 aioreq_pri $pri;
403 add $grp aio_sendfile $dst_fh, $src_fh, 0, $stat[7], sub { 593 add $grp aio_sendfile $dst_fh, $src_fh, 0, $stat[7], sub {
404 if ($_[0] == $stat[7]) { 594 if ($_[0] == $stat[7]) {
405 $grp->result (0); 595 $grp->result (0);
406 close $src_fh; 596 close $src_fh;
407 597
408 # those should not normally block. should. should. 598 # those should not normally block. should. should.
409 utime $stat[8], $stat[9], $dst; 599 utime $stat[8], $stat[9], $dst;
410 chmod $stat[2] & 07777, $dst_fh; 600 chmod $stat[2] & 07777, $dst_fh;
411 chown $stat[4], $stat[5], $dst_fh; 601 chown $stat[4], $stat[5], $dst_fh;
412 close $dst_fh; 602 close $dst_fh;
413 } else { 603 } else {
414 $grp->result (-1); 604 $grp->result (-1);
415 close $src_fh; 605 close $src_fh;
416 close $dst_fh; 606 close $dst_fh;
417 607
418 aioreq $pri; 608 aioreq $pri;
419 add $grp aio_unlink $dst; 609 add $grp aio_unlink $dst;
610 }
420 } 611 };
612 } else {
613 $grp->result (-1);
421 }; 614 }
422 } else {
423 $grp->result (-1);
424 } 615 },
616
617 } else {
618 $grp->result (-1);
425 }, 619 }
426
427 } else {
428 $grp->result (-1);
429 } 620 };
621
622 $grp
430 }; 623 }
431
432 $grp
433} 624}
434 625
435=item aio_move $srcpath, $dstpath, $callback->($status) 626=item aio_move $srcpath, $dstpath, $callback->($status)
436 627
437Try to move the I<file> (directories not supported as either source or 628Try to move the I<file> (directories not supported as either source or
… …
443that is successful, unlinking the C<$srcpath>. 634that is successful, unlinking the C<$srcpath>.
444 635
445=cut 636=cut
446 637
447sub aio_move($$;$) { 638sub aio_move($$;$) {
639 aio_block {
448 my ($src, $dst, $cb) = @_; 640 my ($src, $dst, $cb) = @_;
449 641
450 my $pri = aioreq_pri; 642 my $pri = aioreq_pri;
451 my $grp = aio_group $cb; 643 my $grp = aio_group $cb;
452 644
453 aioreq_pri $pri; 645 aioreq_pri $pri;
454 add $grp aio_rename $src, $dst, sub { 646 add $grp aio_rename $src, $dst, sub {
455 if ($_[0] && $! == EXDEV) { 647 if ($_[0] && $! == EXDEV) {
456 aioreq_pri $pri; 648 aioreq_pri $pri;
457 add $grp aio_copy $src, $dst, sub { 649 add $grp aio_copy $src, $dst, sub {
650 $grp->result ($_[0]);
651
652 if (!$_[0]) {
653 aioreq_pri $pri;
654 add $grp aio_unlink $src;
655 }
656 };
657 } else {
458 $grp->result ($_[0]); 658 $grp->result ($_[0]);
459
460 if (!$_[0]) {
461 aioreq_pri $pri;
462 add $grp aio_unlink $src;
463 }
464 }; 659 }
465 } else {
466 $grp->result ($_[0]);
467 } 660 };
661
662 $grp
468 }; 663 }
469
470 $grp
471} 664}
472 665
473=item aio_scandir $path, $maxreq, $callback->($dirs, $nondirs) 666=item aio_scandir $path, $maxreq, $callback->($dirs, $nondirs)
474 667
475Scans a directory (similar to C<aio_readdir>) but additionally tries to 668Scans a directory (similar to C<aio_readdir>) but additionally tries to
… …
522as those tend to return 0 or 1 as link counts, which disables the 715as those tend to return 0 or 1 as link counts, which disables the
523directory counting heuristic. 716directory counting heuristic.
524 717
525=cut 718=cut
526 719
527sub aio_scandir($$$) { 720sub aio_scandir($$;$) {
721 aio_block {
528 my ($path, $maxreq, $cb) = @_; 722 my ($path, $maxreq, $cb) = @_;
529 723
530 my $pri = aioreq_pri; 724 my $pri = aioreq_pri;
531 725
532 my $grp = aio_group $cb; 726 my $grp = aio_group $cb;
533 727
534 $maxreq = 4 if $maxreq <= 0; 728 $maxreq = 4 if $maxreq <= 0;
535 729
536 # stat once 730 # stat once
537 aioreq_pri $pri;
538 add $grp aio_stat $path, sub {
539 return $grp->result () if $_[0];
540 my $now = time;
541 my $hash1 = join ":", (stat _)[0,1,3,7,9];
542
543 # read the directory entries
544 aioreq_pri $pri; 731 aioreq_pri $pri;
545 add $grp aio_readdir $path, sub { 732 add $grp aio_stat $path, sub {
546 my $entries = shift
547 or return $grp->result (); 733 return $grp->result () if $_[0];
734 my $now = time;
735 my $hash1 = join ":", (stat _)[0,1,3,7,9];
548 736
549 # stat the dir another time 737 # read the directory entries
550 aioreq_pri $pri; 738 aioreq_pri $pri;
739 add $grp aio_readdir $path, sub {
740 my $entries = shift
741 or return $grp->result ();
742
743 # stat the dir another time
744 aioreq_pri $pri;
551 add $grp aio_stat $path, sub { 745 add $grp aio_stat $path, sub {
552 my $hash2 = join ":", (stat _)[0,1,3,7,9]; 746 my $hash2 = join ":", (stat _)[0,1,3,7,9];
553 747
554 my $ndirs; 748 my $ndirs;
555 749
556 # take the slow route if anything looks fishy 750 # take the slow route if anything looks fishy
557 if ($hash1 ne $hash2 or (stat _)[9] == $now) { 751 if ($hash1 ne $hash2 or (stat _)[9] == $now) {
558 $ndirs = -1; 752 $ndirs = -1;
559 } else { 753 } else {
560 # if nlink == 2, we are finished 754 # if nlink == 2, we are finished
561 # on non-posix-fs's, we rely on nlink < 2 755 # on non-posix-fs's, we rely on nlink < 2
562 $ndirs = (stat _)[3] - 2 756 $ndirs = (stat _)[3] - 2
563 or return $grp->result ([], $entries); 757 or return $grp->result ([], $entries);
564 } 758 }
565 759
566 # sort into likely dirs and likely nondirs 760 # sort into likely dirs and likely nondirs
567 # dirs == files without ".", short entries first 761 # dirs == files without ".", short entries first
568 $entries = [map $_->[0], 762 $entries = [map $_->[0],
569 sort { $b->[1] cmp $a->[1] } 763 sort { $b->[1] cmp $a->[1] }
570 map [$_, sprintf "%s%04d", (/.\./ ? "1" : "0"), length], 764 map [$_, sprintf "%s%04d", (/.\./ ? "1" : "0"), length],
571 @$entries]; 765 @$entries];
572 766
573 my (@dirs, @nondirs); 767 my (@dirs, @nondirs);
574 768
575 my $statgrp = add $grp aio_group sub { 769 my $statgrp = add $grp aio_group sub {
576 $grp->result (\@dirs, \@nondirs); 770 $grp->result (\@dirs, \@nondirs);
577 }; 771 };
578 772
579 limit $statgrp $maxreq; 773 limit $statgrp $maxreq;
580 feed $statgrp sub { 774 feed $statgrp sub {
581 return unless @$entries; 775 return unless @$entries;
582 my $entry = pop @$entries; 776 my $entry = pop @$entries;
583 777
584 aioreq_pri $pri; 778 aioreq_pri $pri;
585 add $statgrp aio_stat "$path/$entry/.", sub { 779 add $statgrp aio_stat "$path/$entry/.", sub {
586 if ($_[0] < 0) { 780 if ($_[0] < 0) {
587 push @nondirs, $entry; 781 push @nondirs, $entry;
588 } else { 782 } else {
589 # need to check for real directory 783 # need to check for real directory
590 aioreq_pri $pri; 784 aioreq_pri $pri;
591 add $statgrp aio_lstat "$path/$entry", sub { 785 add $statgrp aio_lstat "$path/$entry", sub {
592 if (-d _) { 786 if (-d _) {
593 push @dirs, $entry; 787 push @dirs, $entry;
594 788
595 unless (--$ndirs) { 789 unless (--$ndirs) {
596 push @nondirs, @$entries; 790 push @nondirs, @$entries;
597 feed $statgrp; 791 feed $statgrp;
792 }
793 } else {
794 push @nondirs, $entry;
598 } 795 }
599 } else {
600 push @nondirs, $entry;
601 } 796 }
602 } 797 }
603 } 798 };
604 }; 799 };
605 }; 800 };
606 }; 801 };
607 }; 802 };
803
804 $grp
608 }; 805 }
806}
609 807
808=item aio_rmtree $path, $callback->($status)
809
810Delete a directory tree starting (and including) C<$path>, return the
811status of the final C<rmdir> only. This is a composite request that
812uses C<aio_scandir> to recurse into and rmdir directories, and unlink
813everything else.
814
815=cut
816
817sub aio_rmtree;
818sub aio_rmtree($;$) {
819 aio_block {
820 my ($path, $cb) = @_;
821
822 my $pri = aioreq_pri;
823 my $grp = aio_group $cb;
824
825 aioreq_pri $pri;
826 add $grp aio_scandir $path, 0, sub {
827 my ($dirs, $nondirs) = @_;
828
829 my $dirgrp = aio_group sub {
830 add $grp aio_rmdir $path, sub {
831 $grp->result ($_[0]);
832 };
833 };
834
835 (aioreq_pri $pri), add $dirgrp aio_rmtree "$path/$_" for @$dirs;
836 (aioreq_pri $pri), add $dirgrp aio_unlink "$path/$_" for @$nondirs;
837
838 add $grp $dirgrp;
839 };
840
610 $grp 841 $grp
842 }
611} 843}
612 844
613=item aio_fsync $fh, $callback->($status) 845=item aio_fsync $fh, $callback->($status)
614 846
615Asynchronously call fsync on the given filehandle and call the callback 847Asynchronously call fsync on the given filehandle and call the callback
… …
829 1061
830=back 1062=back
831 1063
832=head2 SUPPORT FUNCTIONS 1064=head2 SUPPORT FUNCTIONS
833 1065
1066=head3 EVENT PROCESSING AND EVENT LOOP INTEGRATION
1067
834=over 4 1068=over 4
835 1069
836=item $fileno = IO::AIO::poll_fileno 1070=item $fileno = IO::AIO::poll_fileno
837 1071
838Return the I<request result pipe file descriptor>. This filehandle must be 1072Return the I<request result pipe file descriptor>. This filehandle must be
… …
842 1076
843See C<poll_cb> for an example. 1077See C<poll_cb> for an example.
844 1078
845=item IO::AIO::poll_cb 1079=item IO::AIO::poll_cb
846 1080
847Process all outstanding events on the result pipe. You have to call this 1081Process some outstanding events on the result pipe. You have to call this
848regularly. Returns the number of events processed. Returns immediately 1082regularly. Returns the number of events processed. Returns immediately
849when no events are outstanding. 1083when no events are outstanding. The amount of events processed depends on
1084the settings of C<IO::AIO::max_poll_req> and C<IO::AIO::max_poll_time>.
850 1085
851If not all requests were processed for whatever reason, the filehandle 1086If not all requests were processed for whatever reason, the filehandle
852will still be ready when C<poll_cb> returns. 1087will still be ready when C<poll_cb> returns.
853 1088
854Example: Install an Event watcher that automatically calls 1089Example: Install an Event watcher that automatically calls
… …
856 1091
857 Event->io (fd => IO::AIO::poll_fileno, 1092 Event->io (fd => IO::AIO::poll_fileno,
858 poll => 'r', async => 1, 1093 poll => 'r', async => 1,
859 cb => \&IO::AIO::poll_cb); 1094 cb => \&IO::AIO::poll_cb);
860 1095
861=item IO::AIO::poll_some $max_requests 1096=item IO::AIO::max_poll_reqs $nreqs
862 1097
863Similar to C<poll_cb>, but only processes up to C<$max_requests> requests 1098=item IO::AIO::max_poll_time $seconds
864at a time.
865 1099
866Useful if you want to ensure some level of interactiveness when perl is 1100These set the maximum number of requests (default C<0>, meaning infinity)
867not fast enough to process all requests in time. 1101that are being processed by C<IO::AIO::poll_cb> in one call, respectively
1102the maximum amount of time (default C<0>, meaning infinity) spent in
1103C<IO::AIO::poll_cb> to process requests (more correctly the mininum amount
1104of time C<poll_cb> is allowed to use).
1105
1106Setting C<max_poll_time> to a non-zero value creates an overhead of one
1107syscall per request processed, which is not normally a problem unless your
1108callbacks are really really fast or your OS is really really slow (I am
1109not mentioning Solaris here). Using C<max_poll_reqs> incurs no overhead.
1110
1111Setting these is useful if you want to ensure some level of
1112interactiveness when perl is not fast enough to process all requests in
1113time.
1114
1115For interactive programs, values such as C<0.01> to C<0.1> should be fine.
868 1116
869Example: Install an Event watcher that automatically calls 1117Example: Install an Event watcher that automatically calls
870IO::AIO::poll_some with low priority, to ensure that other parts of the 1118IO::AIO::poll_cb with low priority, to ensure that other parts of the
871program get the CPU sometimes even under high AIO load. 1119program get the CPU sometimes even under high AIO load.
872 1120
1121 # try not to spend much more than 0.1s in poll_cb
1122 IO::AIO::max_poll_time 0.1;
1123
1124 # use a low priority so other tasks have priority
873 Event->io (fd => IO::AIO::poll_fileno, 1125 Event->io (fd => IO::AIO::poll_fileno,
874 poll => 'r', nice => 1, 1126 poll => 'r', nice => 1,
875 cb => sub { IO::AIO::poll_some 256 }); 1127 cb => &IO::AIO::poll_cb);
876 1128
877=item IO::AIO::poll_wait 1129=item IO::AIO::poll_wait
878 1130
1131If there are any outstanding requests and none of them in the result
879Wait till the result filehandle becomes ready for reading (simply does a 1132phase, wait till the result filehandle becomes ready for reading (simply
880C<select> on the filehandle. This is useful if you want to synchronously wait 1133does a C<select> on the filehandle. This is useful if you want to
881for some requests to finish). 1134synchronously wait for some requests to finish).
882 1135
883See C<nreqs> for an example. 1136See C<nreqs> for an example.
884 1137
1138=item IO::AIO::poll
1139
1140Waits until some requests have been handled.
1141
1142Returns the number of requests processed, but is otherwise strictly
1143equivalent to:
1144
1145 IO::AIO::poll_wait, IO::AIO::poll_cb
1146
885=item IO::AIO::nreqs 1147=item IO::AIO::flush
886 1148
887Returns the number of requests currently in the ready, execute or pending 1149Wait till all outstanding AIO requests have been handled.
888states (i.e. for which their callback has not been invoked yet).
889 1150
890Example: wait till there are no outstanding requests anymore: 1151Strictly equivalent to:
891 1152
892 IO::AIO::poll_wait, IO::AIO::poll_cb 1153 IO::AIO::poll_wait, IO::AIO::poll_cb
893 while IO::AIO::nreqs; 1154 while IO::AIO::nreqs;
894 1155
895=item IO::AIO::nready 1156=back
896 1157
897Returns the number of requests currently in the ready state (not yet 1158=head3 CONTROLLING THE NUMBER OF THREADS
898executed).
899 1159
900=item IO::AIO::npending 1160=over
901
902Returns the number of requests currently in the pending state (executed,
903but not yet processed by poll_cb).
904
905=item IO::AIO::flush
906
907Wait till all outstanding AIO requests have been handled.
908
909Strictly equivalent to:
910
911 IO::AIO::poll_wait, IO::AIO::poll_cb
912 while IO::AIO::nreqs;
913
914=item IO::AIO::poll
915
916Waits until some requests have been handled.
917
918Strictly equivalent to:
919
920 IO::AIO::poll_wait, IO::AIO::poll_cb
921 if IO::AIO::nreqs;
922 1161
923=item IO::AIO::min_parallel $nthreads 1162=item IO::AIO::min_parallel $nthreads
924 1163
925Set the minimum number of AIO threads to C<$nthreads>. The current 1164Set the minimum number of AIO threads to C<$nthreads>. The current
926default is C<8>, which means eight asynchronous operations can execute 1165default is C<8>, which means eight asynchronous operations can execute
927concurrently at any one time (the number of outstanding requests, 1166concurrently at any one time (the number of outstanding requests,
928however, is unlimited). 1167however, is unlimited).
929 1168
930IO::AIO starts threads only on demand, when an AIO request is queued and 1169IO::AIO starts threads only on demand, when an AIO request is queued and
931no free thread exists. 1170no free thread exists. Please note that queueing up a hundred requests can
1171create demand for a hundred threads, even if it turns out that everything
1172is in the cache and could have been processed faster by a single thread.
932 1173
933It is recommended to keep the number of threads relatively low, as some 1174It is recommended to keep the number of threads relatively low, as some
934Linux kernel versions will scale negatively with the number of threads 1175Linux kernel versions will scale negatively with the number of threads
935(higher parallelity => MUCH higher latency). With current Linux 2.6 1176(higher parallelity => MUCH higher latency). With current Linux 2.6
936versions, 4-32 threads should be fine. 1177versions, 4-32 threads should be fine.
… …
950This module automatically runs C<max_parallel 0> at program end, to ensure 1191This module automatically runs C<max_parallel 0> at program end, to ensure
951that all threads are killed and that there are no outstanding requests. 1192that all threads are killed and that there are no outstanding requests.
952 1193
953Under normal circumstances you don't need to call this function. 1194Under normal circumstances you don't need to call this function.
954 1195
1196=item IO::AIO::max_idle $nthreads
1197
1198Limit the number of threads (default: 4) that are allowed to idle (i.e.,
1199threads that did not get a request to process within 10 seconds). That
1200means if a thread becomes idle while C<$nthreads> other threads are also
1201idle, it will free its resources and exit.
1202
1203This is useful when you allow a large number of threads (e.g. 100 or 1000)
1204to allow for extremely high load situations, but want to free resources
1205under normal circumstances (1000 threads can easily consume 30MB of RAM).
1206
1207The default is probably ok in most situations, especially if thread
1208creation is fast. If thread creation is very slow on your system you might
1209want to use larger values.
1210
955=item $oldmaxreqs = IO::AIO::max_outstanding $maxreqs 1211=item $oldmaxreqs = IO::AIO::max_outstanding $maxreqs
956 1212
957This is a very bad function to use in interactive programs because it 1213This is a very bad function to use in interactive programs because it
958blocks, and a bad way to reduce concurrency because it is inexact: Better 1214blocks, and a bad way to reduce concurrency because it is inexact: Better
959use an C<aio_group> together with a feed callback. 1215use an C<aio_group> together with a feed callback.
960 1216
961Sets the maximum number of outstanding requests to C<$nreqs>. If you 1217Sets the maximum number of outstanding requests to C<$nreqs>. If you
962to queue up more than this number of requests, the next call to the 1218do queue up more than this number of requests, the next call to the
963C<poll_cb> (and C<poll_some> and other functions calling C<poll_cb>) 1219C<poll_cb> (and C<poll_some> and other functions calling C<poll_cb>)
964function will block until the limit is no longer exceeded. 1220function will block until the limit is no longer exceeded.
965 1221
966The default value is very large, so there is no practical limit on the 1222The default value is very large, so there is no practical limit on the
967number of outstanding requests. 1223number of outstanding requests.
… …
970C<max_oustsanding> is mainly useful in simple scripts (with low values) or 1226C<max_oustsanding> is mainly useful in simple scripts (with low values) or
971as a stop gap to shield against fatal memory overflow (with large values). 1227as a stop gap to shield against fatal memory overflow (with large values).
972 1228
973=back 1229=back
974 1230
1231=head3 STATISTICAL INFORMATION
1232
1233=over
1234
1235=item IO::AIO::nreqs
1236
1237Returns the number of requests currently in the ready, execute or pending
1238states (i.e. for which their callback has not been invoked yet).
1239
1240Example: wait till there are no outstanding requests anymore:
1241
1242 IO::AIO::poll_wait, IO::AIO::poll_cb
1243 while IO::AIO::nreqs;
1244
1245=item IO::AIO::nready
1246
1247Returns the number of requests currently in the ready state (not yet
1248executed).
1249
1250=item IO::AIO::npending
1251
1252Returns the number of requests currently in the pending state (executed,
1253but not yet processed by poll_cb).
1254
1255=back
1256
975=cut 1257=cut
976 1258
977# support function to convert a fd into a perl filehandle
978sub _fd2fh {
979 return undef if $_[0] < 0;
980
981 # try to generate nice filehandles
982 my $sym = "IO::AIO::fd#$_[0]";
983 local *$sym;
984
985 open *$sym, "+<&=$_[0]" # usually works under any unix
986 or open *$sym, "<&=$_[0]" # cygwin needs this
987 or open *$sym, ">&=$_[0]" # or this
988 or return undef;
989
990 *$sym
991}
992
993min_parallel 8; 1259min_parallel 8;
994 1260
995END { 1261END { flush }
996 flush;
997};
998 1262
9991; 12631;
1000 1264
1001=head2 FORK BEHAVIOUR 1265=head2 FORK BEHAVIOUR
1002 1266
… …
1022bytes of memory. In addition, stat requests need a stat buffer (possibly 1286bytes of memory. In addition, stat requests need a stat buffer (possibly
1023a few hundred bytes), readdir requires a result buffer and so on. Perl 1287a few hundred bytes), readdir requires a result buffer and so on. Perl
1024scalars and other data passed into aio requests will also be locked and 1288scalars and other data passed into aio requests will also be locked and
1025will consume memory till the request has entered the done state. 1289will consume memory till the request has entered the done state.
1026 1290
1027This is now awfully much, so queuing lots of requests is not usually a 1291This is not awfully much, so queuing lots of requests is not usually a
1028problem. 1292problem.
1029 1293
1030Per-thread usage: 1294Per-thread usage:
1031 1295
1032In the execution phase, some aio requests require more memory for 1296In the execution phase, some aio requests require more memory for

Diff Legend

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