… | |
… | |
14 | aio_unlink "/tmp/file", sub { }; |
14 | aio_unlink "/tmp/file", sub { }; |
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 | |
|
|
20 | use IO::AIO 2; # version has aio objects |
|
|
21 | |
|
|
22 | my $req = aio_unlink "/tmp/file", sub { }; |
|
|
23 | $req->cancel; # cancel request if still in queue |
19 | |
24 | |
20 | # AnyEvent |
25 | # AnyEvent |
21 | open my $fh, "<&=" . IO::AIO::poll_fileno or die "$!"; |
26 | open my $fh, "<&=" . IO::AIO::poll_fileno or die "$!"; |
22 | my $w = AnyEvent->io (fh => $fh, poll => 'r', cb => sub { IO::AIO::poll_cb }); |
27 | my $w = AnyEvent->io (fh => $fh, poll => 'r', cb => sub { IO::AIO::poll_cb }); |
23 | |
28 | |
… | |
… | |
61 | =cut |
66 | =cut |
62 | |
67 | |
63 | package IO::AIO; |
68 | package IO::AIO; |
64 | |
69 | |
65 | no warnings; |
70 | no warnings; |
|
|
71 | use strict 'vars'; |
66 | |
72 | |
67 | use base 'Exporter'; |
73 | use base 'Exporter'; |
68 | |
74 | |
69 | use Fcntl (); |
|
|
70 | |
|
|
71 | BEGIN { |
75 | BEGIN { |
72 | $VERSION = '1.71'; |
76 | our $VERSION = '1.99'; |
73 | |
77 | |
74 | @EXPORT = qw(aio_sendfile aio_read aio_write aio_open aio_close aio_stat |
78 | our @EXPORT = qw(aio_sendfile aio_read aio_write aio_open aio_close aio_stat |
75 | aio_lstat aio_unlink aio_rmdir aio_readdir aio_scandir aio_symlink |
79 | aio_lstat aio_unlink aio_rmdir aio_readdir aio_scandir aio_symlink |
76 | aio_fsync aio_fdatasync aio_readahead); |
80 | aio_fsync aio_fdatasync aio_readahead aio_rename aio_link aio_move); |
77 | @EXPORT_OK = qw(poll_fileno poll_cb min_parallel max_parallel |
81 | our @EXPORT_OK = qw(poll_fileno poll_cb min_parallel max_parallel max_outstanding nreqs); |
78 | max_outstanding nreqs); |
|
|
79 | |
82 | |
80 | require XSLoader; |
83 | require XSLoader; |
81 | XSLoader::load IO::AIO, $VERSION; |
84 | XSLoader::load ("IO::AIO", $VERSION); |
82 | } |
85 | } |
83 | |
86 | |
84 | =head1 FUNCTIONS |
87 | =head1 FUNCTIONS |
85 | |
88 | |
86 | =head2 AIO FUNCTIONS |
89 | =head2 AIO FUNCTIONS |
… | |
… | |
93 | perl, which usually delivers "false") as it's sole argument when the given |
96 | perl, which usually delivers "false") as it's sole argument when the given |
94 | syscall has been executed asynchronously. |
97 | syscall has been executed asynchronously. |
95 | |
98 | |
96 | All functions expecting a filehandle keep a copy of the filehandle |
99 | All functions expecting a filehandle keep a copy of the filehandle |
97 | internally until the request has finished. |
100 | internally until the request has finished. |
|
|
101 | |
|
|
102 | All non-composite requests (requests that are not broken down into |
|
|
103 | multiple requests) return objects of type L<IO::AIO::REQ> that allow |
|
|
104 | further manipulation of running requests. |
98 | |
105 | |
99 | The pathnames you pass to these routines I<must> be absolute and |
106 | The pathnames you pass to these routines I<must> be absolute and |
100 | encoded in byte form. The reason for the former is that at the time the |
107 | encoded in byte form. The reason for the former is that at the time the |
101 | request is being executed, the current working directory could have |
108 | request is being executed, the current working directory could have |
102 | changed. Alternatively, you can make sure that you never change the |
109 | changed. Alternatively, you can make sure that you never change the |
… | |
… | |
168 | aio_read $fh, 7, 15, $buffer, 0, sub { |
175 | aio_read $fh, 7, 15, $buffer, 0, sub { |
169 | $_[0] > 0 or die "read error: $!"; |
176 | $_[0] > 0 or die "read error: $!"; |
170 | print "read $_[0] bytes: <$buffer>\n"; |
177 | print "read $_[0] bytes: <$buffer>\n"; |
171 | }; |
178 | }; |
172 | |
179 | |
|
|
180 | =item aio_move $srcpath, $dstpath, $callback->($status) |
|
|
181 | |
|
|
182 | Try to move the I<file> (directories not supported as either source or |
|
|
183 | destination) from C<$srcpath> to C<$dstpath> and call the callback with |
|
|
184 | the C<0> (error) or C<-1> ok. |
|
|
185 | |
|
|
186 | This is a composite request that tries to rename(2) the file first. If |
|
|
187 | rename files with C<EXDEV>, it creates the destination file with mode 0200 |
|
|
188 | and copies the contents of the source file into it using C<aio_sendfile>, |
|
|
189 | followed by restoring atime, mtime, access mode and uid/gid, in that |
|
|
190 | order, and unlinking the C<$srcpath>. |
|
|
191 | |
|
|
192 | If an error occurs, the partial destination file will be unlinked, if |
|
|
193 | possible, except when setting atime, mtime, access mode and uid/gid, where |
|
|
194 | errors are being ignored. |
|
|
195 | |
|
|
196 | =cut |
|
|
197 | |
|
|
198 | sub aio_move($$$) { |
|
|
199 | my ($src, $dst, $cb) = @_; |
|
|
200 | |
|
|
201 | aio_rename $src, $dst, sub { |
|
|
202 | if ($_[0] && $! == EXDEV) { |
|
|
203 | aio_open $src, O_RDONLY, 0, sub { |
|
|
204 | if (my $src_fh = $_[0]) { |
|
|
205 | my @stat = stat $src_fh; |
|
|
206 | |
|
|
207 | aio_open $dst, O_WRONLY, 0200, sub { |
|
|
208 | if (my $dst_fh = $_[0]) { |
|
|
209 | aio_sendfile $dst_fh, $src_fh, 0, $stat[7], sub { |
|
|
210 | close $src_fh; |
|
|
211 | |
|
|
212 | if ($_[0] == $stat[7]) { |
|
|
213 | utime $stat[8], $stat[9], $dst; |
|
|
214 | chmod $stat[2] & 07777, $dst_fh; |
|
|
215 | chown $stat[4], $stat[5], $dst_fh; |
|
|
216 | close $dst_fh; |
|
|
217 | |
|
|
218 | aio_unlink $src, sub { |
|
|
219 | $cb->($_[0]); |
|
|
220 | }; |
|
|
221 | } else { |
|
|
222 | my $errno = $!; |
|
|
223 | aio_unlink $dst, sub { |
|
|
224 | $! = $errno; |
|
|
225 | $cb->(-1); |
|
|
226 | }; |
|
|
227 | } |
|
|
228 | }; |
|
|
229 | } else { |
|
|
230 | $cb->(-1); |
|
|
231 | } |
|
|
232 | }, |
|
|
233 | |
|
|
234 | } else { |
|
|
235 | $cb->(-1); |
|
|
236 | } |
|
|
237 | }; |
|
|
238 | } else { |
|
|
239 | $cb->($_[0]); |
|
|
240 | } |
|
|
241 | }; |
|
|
242 | } |
|
|
243 | |
173 | =item aio_sendfile $out_fh, $in_fh, $in_offset, $length, $callback->($retval) |
244 | =item aio_sendfile $out_fh, $in_fh, $in_offset, $length, $callback->($retval) |
174 | |
245 | |
175 | Tries to copy C<$length> bytes from C<$in_fh> to C<$out_fh>. It starts |
246 | Tries to copy C<$length> bytes from C<$in_fh> to C<$out_fh>. It starts |
176 | reading at byte offset C<$in_offset>, and starts writing at the current |
247 | reading at byte offset C<$in_offset>, and starts writing at the current |
177 | file offset of C<$out_fh>. Because of that, it is not safe to issue more |
248 | file offset of C<$out_fh>. Because of that, it is not safe to issue more |
… | |
… | |
232 | =item aio_unlink $pathname, $callback->($status) |
303 | =item aio_unlink $pathname, $callback->($status) |
233 | |
304 | |
234 | Asynchronously unlink (delete) a file and call the callback with the |
305 | Asynchronously unlink (delete) a file and call the callback with the |
235 | result code. |
306 | result code. |
236 | |
307 | |
|
|
308 | =item aio_link $srcpath, $dstpath, $callback->($status) |
|
|
309 | |
|
|
310 | Asynchronously create a new link to the existing object at C<$srcpath> at |
|
|
311 | the path C<$dstpath> and call the callback with the result code. |
|
|
312 | |
|
|
313 | =item aio_symlink $srcpath, $dstpath, $callback->($status) |
|
|
314 | |
|
|
315 | Asynchronously create a new symbolic link to the existing object at C<$srcpath> at |
|
|
316 | the path C<$dstpath> and call the callback with the result code. |
|
|
317 | |
|
|
318 | =item aio_rename $srcpath, $dstpath, $callback->($status) |
|
|
319 | |
|
|
320 | Asynchronously rename the object at C<$srcpath> to C<$dstpath>, just as |
|
|
321 | rename(2) and call the callback with the result code. |
|
|
322 | |
237 | =item aio_rmdir $pathname, $callback->($status) |
323 | =item aio_rmdir $pathname, $callback->($status) |
238 | |
324 | |
239 | Asynchronously rmdir (delete) a directory and call the callback with the |
325 | Asynchronously rmdir (delete) a directory and call the callback with the |
240 | result code. |
326 | result code. |
241 | |
327 | |
242 | =item aio_readdir $pathname $callback->($entries) |
328 | =item aio_readdir $pathname, $callback->($entries) |
243 | |
329 | |
244 | Unlike the POSIX call of the same name, C<aio_readdir> reads an entire |
330 | Unlike the POSIX call of the same name, C<aio_readdir> reads an entire |
245 | directory (i.e. opendir + readdir + closedir). The entries will not be |
331 | directory (i.e. opendir + readdir + closedir). The entries will not be |
246 | sorted, and will B<NOT> include the C<.> and C<..> entries. |
332 | sorted, and will B<NOT> include the C<.> and C<..> entries. |
247 | |
333 | |
248 | The callback a single argument which is either C<undef> or an array-ref |
334 | The callback a single argument which is either C<undef> or an array-ref |
249 | with the filenames. |
335 | with the filenames. |
250 | |
336 | |
251 | =item aio_scandir $path, $maxreq, $callback->($dirs, $nondirs) |
337 | =item aio_scandir $path, $maxreq, $callback->($dirs, $nondirs) |
252 | |
338 | |
253 | Scans a directory (similar to C<aio_readdir>) and tries to separate the |
339 | Scans a directory (similar to C<aio_readdir>) but additionally tries to |
254 | entries of directory C<$path> into two sets of names, ones you can recurse |
340 | separate the entries of directory C<$path> into two sets of names, ones |
255 | into (directories), and ones you cannot recurse into (everything else). |
341 | you can recurse into (directories or links to them), and ones you cannot |
|
|
342 | recurse into (everything else). |
256 | |
343 | |
257 | C<aio_scandir> is a composite request that consists of many |
344 | C<aio_scandir> is a composite request that consists of many sub |
258 | aio-primitives. C<$maxreq> specifies the maximum number of outstanding |
345 | requests. C<$maxreq> specifies the maximum number of outstanding aio |
259 | aio requests that this function generates. If it is C<< <= 0 >>, then a |
346 | requests that this function generates. If it is C<< <= 0 >>, then a |
260 | suitable default will be chosen (currently 8). |
347 | suitable default will be chosen (currently 8). |
261 | |
348 | |
262 | On error, the callback is called without arguments, otherwise it receives |
349 | On error, the callback is called without arguments, otherwise it receives |
263 | two array-refs with path-relative entry names. |
350 | two array-refs with path-relative entry names. |
264 | |
351 | |
… | |
… | |
273 | Implementation notes. |
360 | Implementation notes. |
274 | |
361 | |
275 | The C<aio_readdir> cannot be avoided, but C<stat()>'ing every entry can. |
362 | The C<aio_readdir> cannot be avoided, but C<stat()>'ing every entry can. |
276 | |
363 | |
277 | After reading the directory, the modification time, size etc. of the |
364 | After reading the directory, the modification time, size etc. of the |
278 | directory before and after the readdir is checked, and if they match, the |
365 | directory before and after the readdir is checked, and if they match (and |
279 | link count will be used to decide how many entries are directories (if |
366 | isn't the current time), the link count will be used to decide how many |
280 | >= 2). Otherwise, no knowledge of the number of subdirectories will be |
367 | entries are directories (if >= 2). Otherwise, no knowledge of the number |
281 | assumed. |
368 | of subdirectories will be assumed. |
282 | |
369 | |
283 | Then entires will be sorted into likely directories (everything without a |
370 | Then entries will be sorted into likely directories (everything without |
284 | non-initial dot) and likely non-directories (everything else). Then every |
371 | a non-initial dot currently) and likely non-directories (everything |
285 | entry + C</.> will be C<stat>'ed, likely directories first. This is often |
372 | else). Then every entry plus an appended C</.> will be C<stat>'ed, |
|
|
373 | likely directories first. If that succeeds, it assumes that the entry |
|
|
374 | is a directory or a symlink to directory (which will be checked |
|
|
375 | seperately). This is often faster than stat'ing the entry itself because |
286 | faster because filesystems might detect the type of the entry without |
376 | filesystems might detect the type of the entry without reading the inode |
287 | reading the inode data (e.g. ext2s filetype feature). If that succeeds, |
377 | data (e.g. ext2fs filetype feature). |
288 | it assumes that the entry is a directory or a symlink to directory (which |
|
|
289 | will be checked seperately). |
|
|
290 | |
378 | |
291 | If the known number of directories has been reached, the rest of the |
379 | If the known number of directories (link count - 2) has been reached, the |
292 | entries is assumed to be non-directories. |
380 | rest of the entries is assumed to be non-directories. |
|
|
381 | |
|
|
382 | This only works with certainty on POSIX (= UNIX) filesystems, which |
|
|
383 | fortunately are the vast majority of filesystems around. |
|
|
384 | |
|
|
385 | It will also likely work on non-POSIX filesystems with reduced efficiency |
|
|
386 | as those tend to return 0 or 1 as link counts, which disables the |
|
|
387 | directory counting heuristic. |
293 | |
388 | |
294 | =cut |
389 | =cut |
295 | |
390 | |
296 | sub aio_scandir($$$) { |
391 | sub aio_scandir($$$) { |
297 | my ($path, $maxreq, $cb) = @_; |
392 | my ($path, $maxreq, $cb) = @_; |
298 | |
393 | |
299 | $maxreq = 8 if $maxreq <= 0; |
394 | $maxreq = 8 if $maxreq <= 0; |
300 | |
395 | |
301 | # stat once |
396 | # stat once |
302 | aio_stat $path, sub { |
397 | aio_stat $path, sub { |
303 | $cb->() if $_[0]; |
398 | return $cb->() if $_[0]; |
|
|
399 | my $now = time; |
304 | my $hash1 = join ":", (stat _)[0,1,3,7,9]; |
400 | my $hash1 = join ":", (stat _)[0,1,3,7,9]; |
305 | |
401 | |
306 | # read the directory entries |
402 | # read the directory entries |
307 | aio_readdir $path, sub { |
403 | aio_readdir $path, sub { |
308 | my $entries = shift |
404 | my $entries = shift |
… | |
… | |
313 | my $hash2 = join ":", (stat _)[0,1,3,7,9]; |
409 | my $hash2 = join ":", (stat _)[0,1,3,7,9]; |
314 | |
410 | |
315 | my $ndirs; |
411 | my $ndirs; |
316 | |
412 | |
317 | # take the slow route if anything looks fishy |
413 | # take the slow route if anything looks fishy |
318 | if ($hash1 ne $hash2) { |
414 | if ($hash1 ne $hash2 or (stat _)[9] == $now) { |
319 | $ndirs = -1; |
415 | $ndirs = -1; |
320 | } else { |
416 | } else { |
321 | # if nlink == 2, we are finished |
417 | # if nlink == 2, we are finished |
322 | # on non-posix-fs's, we rely on nlink < 2 |
418 | # on non-posix-fs's, we rely on nlink < 2 |
323 | $ndirs = (stat _)[3] - 2 |
419 | $ndirs = (stat _)[3] - 2 |
… | |
… | |
398 | If this call isn't available because your OS lacks it or it couldn't be |
494 | If this call isn't available because your OS lacks it or it couldn't be |
399 | detected, it will be emulated by calling C<fsync> instead. |
495 | detected, it will be emulated by calling C<fsync> instead. |
400 | |
496 | |
401 | =back |
497 | =back |
402 | |
498 | |
|
|
499 | =head2 IO::AIO::REQ CLASS |
|
|
500 | |
|
|
501 | All non-aggregate C<aio_*> functions return an object of this class when |
|
|
502 | called in non-void context. |
|
|
503 | |
|
|
504 | A request always moves through the following five states in its lifetime, |
|
|
505 | in order: B<ready> (request has been created, but has not been executed |
|
|
506 | yet), B<execute> (request is currently being executed), B<pending> |
|
|
507 | (request has been executed but callback has not been called yet), |
|
|
508 | B<result> (results are being processed synchronously, includes calling the |
|
|
509 | callback) and B<done> (request has reached the end of its lifetime and |
|
|
510 | holds no resources anymore). |
|
|
511 | |
|
|
512 | =over 4 |
|
|
513 | |
|
|
514 | =item $req->cancel |
|
|
515 | |
|
|
516 | Cancels the request, if possible. Has the effect of skipping execution |
|
|
517 | when entering the B<execute> state and skipping calling the callback when |
|
|
518 | entering the the B<result> state, but will leave the request otherwise |
|
|
519 | untouched. That means that requests that currently execute will not be |
|
|
520 | stopped and resources held by the request will not be freed prematurely. |
|
|
521 | |
|
|
522 | =back |
|
|
523 | |
403 | =head2 SUPPORT FUNCTIONS |
524 | =head2 SUPPORT FUNCTIONS |
404 | |
525 | |
405 | =over 4 |
526 | =over 4 |
406 | |
527 | |
407 | =item $fileno = IO::AIO::poll_fileno |
528 | =item $fileno = IO::AIO::poll_fileno |
… | |
… | |
532 | } |
653 | } |
533 | |
654 | |
534 | 1; |
655 | 1; |
535 | |
656 | |
536 | =head2 FORK BEHAVIOUR |
657 | =head2 FORK BEHAVIOUR |
|
|
658 | |
|
|
659 | This module should do "the right thing" when the process using it forks: |
537 | |
660 | |
538 | Before the fork, IO::AIO enters a quiescent state where no requests |
661 | Before the fork, IO::AIO enters a quiescent state where no requests |
539 | can be added in other threads and no results will be processed. After |
662 | can be added in other threads and no results will be processed. After |
540 | the fork the parent simply leaves the quiescent state and continues |
663 | the fork the parent simply leaves the quiescent state and continues |
541 | request/result processing, while the child clears the request/result |
664 | request/result processing, while the child clears the request/result |
542 | queue (so the requests started before the fork will only be handled in |
665 | queue (so the requests started before the fork will only be handled in |
543 | the parent). Threats will be started on demand until the limit ste in the |
666 | the parent). Threads will be started on demand until the limit ste in the |
544 | parent process has been reached again. |
667 | parent process has been reached again. |
545 | |
668 | |
|
|
669 | In short: the parent will, after a short pause, continue as if fork had |
|
|
670 | not been called, while the child will act as if IO::AIO has not been used |
|
|
671 | yet. |
|
|
672 | |
546 | =head1 SEE ALSO |
673 | =head1 SEE ALSO |
547 | |
674 | |
548 | L<Coro>, L<Linux::AIO>. |
675 | L<Coro>, L<Linux::AIO> (obsolete). |
549 | |
676 | |
550 | =head1 AUTHOR |
677 | =head1 AUTHOR |
551 | |
678 | |
552 | Marc Lehmann <schmorp@schmorp.de> |
679 | Marc Lehmann <schmorp@schmorp.de> |
553 | http://home.schmorp.de/ |
680 | http://home.schmorp.de/ |