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 1.174 by root, Sun Jan 10 20:28:43 2010 UTC vs.
Revision 1.228 by root, Sun Jun 17 17:07:25 2012 UTC

… …
4 4
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", IO::AIO::O_RDONLY, 0, sub {
10 my $fh = shift 10 my $fh = shift
11 or die "/etc/passwd: $!"; 11 or die "/etc/passwd: $!";
12 ... 12 ...
13 }; 13 };
14 14
… …
25 my $req = aio_unlink "/tmp/file", sub { }; 25 my $req = aio_unlink "/tmp/file", sub { };
26 $req->cancel; # cancel request if still in queue 26 $req->cancel; # cancel request if still in queue
27 27
28 my $grp = aio_group sub { print "all stats done\n" }; 28 my $grp = aio_group sub { print "all stats done\n" };
29 add $grp aio_stat "..." for ...; 29 add $grp aio_stat "..." for ...;
30
31 # AnyEvent integration (EV, Event, Glib, Tk, POE, urxvt, pureperl...)
32 use AnyEvent::AIO;
33
34 # EV integration
35 my $aio_w = EV::io IO::AIO::poll_fileno, EV::READ, \&IO::AIO::poll_cb;
36
37 # Event integration
38 Event->io (fd => IO::AIO::poll_fileno,
39 poll => 'r',
40 cb => \&IO::AIO::poll_cb);
41
42 # Glib/Gtk2 integration
43 add_watch Glib::IO IO::AIO::poll_fileno,
44 in => sub { IO::AIO::poll_cb; 1 };
45
46 # Tk integration
47 Tk::Event::IO->fileevent (IO::AIO::poll_fileno, "",
48 readable => \&IO::AIO::poll_cb);
49
50 # Danga::Socket integration
51 Danga::Socket->AddOtherFds (IO::AIO::poll_fileno =>
52 \&IO::AIO::poll_cb);
53 30
54=head1 DESCRIPTION 31=head1 DESCRIPTION
55 32
56This module implements asynchronous I/O using whatever means your 33This module implements asynchronous I/O using whatever means your
57operating system supports. It is implemented as an interface to C<libeio> 34operating system supports. It is implemented as an interface to C<libeio>
… …
99 76
100 # register the IO::AIO callback with EV 77 # register the IO::AIO callback with EV
101 my $aio_w = EV::io IO::AIO::poll_fileno, EV::READ, \&IO::AIO::poll_cb; 78 my $aio_w = EV::io IO::AIO::poll_fileno, EV::READ, \&IO::AIO::poll_cb;
102 79
103 # queue the request to open /etc/passwd 80 # queue the request to open /etc/passwd
104 aio_open "/etc/passwd", O_RDONLY, 0, sub { 81 aio_open "/etc/passwd", IO::AIO::O_RDONLY, 0, sub {
105 my $fh = shift 82 my $fh = shift
106 or die "error while opening: $!"; 83 or die "error while opening: $!";
107 84
108 # stat'ing filehandles is generally non-blocking 85 # stat'ing filehandles is generally non-blocking
109 my $size = -s $fh; 86 my $size = -s $fh;
… …
191use common::sense; 168use common::sense;
192 169
193use base 'Exporter'; 170use base 'Exporter';
194 171
195BEGIN { 172BEGIN {
196 our $VERSION = '3.5'; 173 our $VERSION = '4.15';
197 174
198 our @AIO_REQ = qw(aio_sendfile aio_read aio_write aio_open aio_close 175 our @AIO_REQ = qw(aio_sendfile aio_seek aio_read aio_write aio_open aio_close
199 aio_stat aio_lstat aio_unlink aio_rmdir aio_readdir aio_readdirx 176 aio_stat aio_lstat aio_unlink aio_rmdir aio_readdir aio_readdirx
200 aio_scandir aio_symlink aio_readlink aio_sync aio_fsync 177 aio_scandir aio_symlink aio_readlink aio_realpath aio_sync
201 aio_fdatasync aio_sync_file_range aio_pathsync aio_readahead 178 aio_fsync aio_syncfs aio_fdatasync aio_sync_file_range aio_fallocate
179 aio_pathsync aio_readahead aio_fiemap
202 aio_rename aio_link aio_move aio_copy aio_group 180 aio_rename aio_link aio_move aio_copy aio_group
203 aio_nop aio_mknod aio_load aio_rmtree aio_mkdir aio_chown 181 aio_nop aio_mknod aio_load aio_rmtree aio_mkdir aio_chown
204 aio_chmod aio_utime aio_truncate 182 aio_chmod aio_utime aio_truncate
205 aio_msync aio_mtouch aio_statvfs); 183 aio_msync aio_mtouch aio_mlock aio_mlockall
184 aio_statvfs
185 aio_wd);
206 186
207 our @EXPORT = (@AIO_REQ, qw(aioreq_pri aioreq_nice)); 187 our @EXPORT = (@AIO_REQ, qw(aioreq_pri aioreq_nice));
208 our @EXPORT_OK = qw(poll_fileno poll_cb poll_wait flush 188 our @EXPORT_OK = qw(poll_fileno poll_cb poll_wait flush
209 min_parallel max_parallel max_idle 189 min_parallel max_parallel max_idle idle_timeout
210 nreqs nready npending nthreads 190 nreqs nready npending nthreads
211 max_poll_time max_poll_reqs 191 max_poll_time max_poll_reqs
212 sendfile fadvise); 192 sendfile fadvise madvise
193 mmap munmap munlock munlockall);
213 194
214 push @AIO_REQ, qw(aio_busy); # not exported 195 push @AIO_REQ, qw(aio_busy); # not exported
215 196
216 @IO::AIO::GRP::ISA = 'IO::AIO::REQ'; 197 @IO::AIO::GRP::ISA = 'IO::AIO::REQ';
217 198
… …
219 XSLoader::load ("IO::AIO", $VERSION); 200 XSLoader::load ("IO::AIO", $VERSION);
220} 201}
221 202
222=head1 FUNCTIONS 203=head1 FUNCTIONS
223 204
224=head2 AIO REQUEST FUNCTIONS 205=head2 QUICK OVERVIEW
206
207This section simply lists the prototypes of the most important functions
208for quick reference. See the following sections for function-by-function
209documentation.
210
211 aio_wd $pathname, $callback->($wd)
212 aio_open $pathname, $flags, $mode, $callback->($fh)
213 aio_close $fh, $callback->($status)
214 aio_seek $fh,$offset,$whence, $callback->($offs)
215 aio_read $fh,$offset,$length, $data,$dataoffset, $callback->($retval)
216 aio_write $fh,$offset,$length, $data,$dataoffset, $callback->($retval)
217 aio_sendfile $out_fh, $in_fh, $in_offset, $length, $callback->($retval)
218 aio_readahead $fh,$offset,$length, $callback->($retval)
219 aio_stat $fh_or_path, $callback->($status)
220 aio_lstat $fh, $callback->($status)
221 aio_statvfs $fh_or_path, $callback->($statvfs)
222 aio_utime $fh_or_path, $atime, $mtime, $callback->($status)
223 aio_chown $fh_or_path, $uid, $gid, $callback->($status)
224 aio_chmod $fh_or_path, $mode, $callback->($status)
225 aio_truncate $fh_or_path, $offset, $callback->($status)
226 aio_unlink $pathname, $callback->($status)
227 aio_mknod $pathname, $mode, $dev, $callback->($status)
228 aio_link $srcpath, $dstpath, $callback->($status)
229 aio_symlink $srcpath, $dstpath, $callback->($status)
230 aio_readlink $pathname, $callback->($link)
231 aio_realpath $pathname, $callback->($link)
232 aio_rename $srcpath, $dstpath, $callback->($status)
233 aio_mkdir $pathname, $mode, $callback->($status)
234 aio_rmdir $pathname, $callback->($status)
235 aio_readdir $pathname, $callback->($entries)
236 aio_readdirx $pathname, $flags, $callback->($entries, $flags)
237 IO::AIO::READDIR_DENTS IO::AIO::READDIR_DIRS_FIRST
238 IO::AIO::READDIR_STAT_ORDER IO::AIO::READDIR_FOUND_UNKNOWN
239 aio_scandir $pathname, $maxreq, $callback->($dirs, $nondirs)
240 aio_load $pathname, $data, $callback->($status)
241 aio_copy $srcpath, $dstpath, $callback->($status)
242 aio_move $srcpath, $dstpath, $callback->($status)
243 aio_rmtree $pathname, $callback->($status)
244 aio_sync $callback->($status)
245 aio_syncfs $fh, $callback->($status)
246 aio_fsync $fh, $callback->($status)
247 aio_fdatasync $fh, $callback->($status)
248 aio_sync_file_range $fh, $offset, $nbytes, $flags, $callback->($status)
249 aio_pathsync $pathname, $callback->($status)
250 aio_msync $scalar, $offset = 0, $length = undef, flags = 0, $callback->($status)
251 aio_mtouch $scalar, $offset = 0, $length = undef, flags = 0, $callback->($status)
252 aio_mlock $scalar, $offset = 0, $length = undef, $callback->($status)
253 aio_mlockall $flags, $callback->($status)
254 aio_group $callback->(...)
255 aio_nop $callback->()
256
257 $prev_pri = aioreq_pri [$pri]
258 aioreq_nice $pri_adjust
259
260 IO::AIO::poll_wait
261 IO::AIO::poll_cb
262 IO::AIO::poll
263 IO::AIO::flush
264 IO::AIO::max_poll_reqs $nreqs
265 IO::AIO::max_poll_time $seconds
266 IO::AIO::min_parallel $nthreads
267 IO::AIO::max_parallel $nthreads
268 IO::AIO::max_idle $nthreads
269 IO::AIO::idle_timeout $seconds
270 IO::AIO::max_outstanding $maxreqs
271 IO::AIO::nreqs
272 IO::AIO::nready
273 IO::AIO::npending
274
275 IO::AIO::sendfile $ofh, $ifh, $offset, $count
276 IO::AIO::fadvise $fh, $offset, $len, $advice
277 IO::AIO::mmap $scalar, $length, $prot, $flags[, $fh[, $offset]]
278 IO::AIO::munmap $scalar
279 IO::AIO::madvise $scalar, $offset, $length, $advice
280 IO::AIO::mprotect $scalar, $offset, $length, $protect
281 IO::AIO::munlock $scalar, $offset = 0, $length = undef
282 IO::AIO::munlockall
283
284=head2 API NOTES
225 285
226All the C<aio_*> calls are more or less thin wrappers around the syscall 286All the C<aio_*> calls are more or less thin wrappers around the syscall
227with the same name (sans C<aio_>). The arguments are similar or identical, 287with the same name (sans C<aio_>). The arguments are similar or identical,
228and they all accept an additional (and optional) C<$callback> argument 288and they all accept an additional (and optional) C<$callback> argument
229which must be a code reference. This code reference will get called with 289which must be a code reference. This code reference will be called after
230the syscall return code (e.g. most syscalls return C<-1> on error, unlike 290the syscall has been executed in an asynchronous fashion. The results
231perl, which usually delivers "false") as its sole argument after the given 291of the request will be passed as arguments to the callback (and, if an
232syscall has been executed asynchronously. 292error occured, in C<$!>) - for most requests the syscall return code (e.g.
293most syscalls return C<-1> on error, unlike perl, which usually delivers
294"false").
295
296Some requests (such as C<aio_readdir>) pass the actual results and
297communicate failures by passing C<undef>.
233 298
234All functions expecting a filehandle keep a copy of the filehandle 299All functions expecting a filehandle keep a copy of the filehandle
235internally until the request has finished. 300internally until the request has finished.
236 301
237All functions return request objects of type L<IO::AIO::REQ> that allow 302All functions return request objects of type L<IO::AIO::REQ> that allow
238further manipulation of those requests while they are in-flight. 303further manipulation of those requests while they are in-flight.
239 304
240The pathnames you pass to these routines I<must> be absolute and 305The pathnames you pass to these routines I<should> be absolute. The
241encoded as octets. The reason for the former is that at the time the 306reason for this is that at the time the request is being executed, the
242request is being executed, the current working directory could have 307current working directory could have changed. Alternatively, you can
243changed. Alternatively, you can make sure that you never change the 308make sure that you never change the current working directory anywhere
244current working directory anywhere in the program and then use relative 309in the program and then use relative paths. You can also take advantage
245paths. 310of IO::AIOs working directory abstraction, that lets you specify paths
311relative to some previously-opened "working directory object" - see the
312description of the C<IO::AIO::WD> class later in this document.
246 313
247To encode pathnames as octets, either make sure you either: a) always pass 314To encode pathnames as octets, either make sure you either: a) always pass
248in filenames you got from outside (command line, readdir etc.) without 315in filenames you got from outside (command line, readdir etc.) without
249tinkering, b) are ASCII or ISO 8859-1, c) use the Encode module and encode 316tinkering, b) are in your native filesystem encoding, c) use the Encode
250your pathnames to the locale (or other) encoding in effect in the user 317module and encode your pathnames to the locale (or other) encoding in
251environment, d) use Glib::filename_from_unicode on unicode filenames or e) 318effect in the user environment, d) use Glib::filename_from_unicode on
252use something else to ensure your scalar has the correct contents. 319unicode filenames or e) use something else to ensure your scalar has the
320correct contents.
253 321
254This works, btw. independent of the internal UTF-8 bit, which IO::AIO 322This works, btw. independent of the internal UTF-8 bit, which IO::AIO
255handles correctly whether it is set or not. 323handles correctly whether it is set or not.
324
325=head2 AIO REQUEST FUNCTIONS
256 326
257=over 4 327=over 4
258 328
259=item $prev_pri = aioreq_pri [$pri] 329=item $prev_pri = aioreq_pri [$pri]
260 330
… …
307by the umask in effect then the request is being executed, so better never 377by the umask in effect then the request is being executed, so better never
308change the umask. 378change the umask.
309 379
310Example: 380Example:
311 381
312 aio_open "/etc/passwd", O_RDONLY, 0, sub { 382 aio_open "/etc/passwd", IO::AIO::O_RDONLY, 0, sub {
313 if ($_[0]) { 383 if ($_[0]) {
314 print "open successful, fh is $_[0]\n"; 384 print "open successful, fh is $_[0]\n";
315 ... 385 ...
316 } else { 386 } else {
317 die "open failed: $!\n"; 387 die "open failed: $!\n";
318 } 388 }
319 }; 389 };
320 390
391In addition to all the common open modes/flags (C<O_RDONLY>, C<O_WRONLY>,
392C<O_RDWR>, C<O_CREAT>, C<O_TRUNC>, C<O_EXCL> and C<O_APPEND>), the
393following POSIX and non-POSIX constants are available (missing ones on
394your system are, as usual, C<0>):
395
396C<O_ASYNC>, C<O_DIRECT>, C<O_NOATIME>, C<O_CLOEXEC>, C<O_NOCTTY>, C<O_NOFOLLOW>,
397C<O_NONBLOCK>, C<O_EXEC>, C<O_SEARCH>, C<O_DIRECTORY>, C<O_DSYNC>,
398C<O_RSYNC>, C<O_SYNC> and C<O_TTY_INIT>.
399
321 400
322=item aio_close $fh, $callback->($status) 401=item aio_close $fh, $callback->($status)
323 402
324Asynchronously close a file and call the callback with the result 403Asynchronously close a file and call the callback with the result
325code. 404code.
… …
333 412
334Or in other words: the file descriptor will be closed, but it will not be 413Or in other words: the file descriptor will be closed, but it will not be
335free for reuse until the perl filehandle is closed. 414free for reuse until the perl filehandle is closed.
336 415
337=cut 416=cut
417
418=item aio_seek $fh, $offset, $whence, $callback->($offs)
419
420Seeks the filehandle to the new C<$offset>, similarly to perl's
421C<sysseek>. The C<$whence> can use the traditional values (C<0> for
422C<IO::AIO::SEEK_SET>, C<1> for C<IO::AIO::SEEK_CUR> or C<2> for
423C<IO::AIO::SEEK_END>).
424
425The resulting absolute offset will be passed to the callback, or C<-1> in
426case of an error.
427
428In theory, the C<$whence> constants could be different than the
429corresponding values from L<Fcntl>, but perl guarantees they are the same,
430so don't panic.
431
432As a GNU/Linux (and maybe Solaris) extension, also the constants
433C<IO::AIO::SEEK_DATA> and C<IO::AIO::SEEK_HOLE> are available, if they
434could be found. No guarantees about suitability for use in C<aio_seek> or
435Perl's C<sysseek> can be made though, although I would naively assume they
436"just work".
338 437
339=item aio_read $fh,$offset,$length, $data,$dataoffset, $callback->($retval) 438=item aio_read $fh,$offset,$length, $data,$dataoffset, $callback->($retval)
340 439
341=item aio_write $fh,$offset,$length, $data,$dataoffset, $callback->($retval) 440=item aio_write $fh,$offset,$length, $data,$dataoffset, $callback->($retval)
342 441
… …
375 474
376Tries to copy C<$length> bytes from C<$in_fh> to C<$out_fh>. It starts 475Tries to copy C<$length> bytes from C<$in_fh> to C<$out_fh>. It starts
377reading at byte offset C<$in_offset>, and starts writing at the current 476reading at byte offset C<$in_offset>, and starts writing at the current
378file offset of C<$out_fh>. Because of that, it is not safe to issue more 477file offset of C<$out_fh>. Because of that, it is not safe to issue more
379than one C<aio_sendfile> per C<$out_fh>, as they will interfere with each 478than one C<aio_sendfile> per C<$out_fh>, as they will interfere with each
380other. 479other. The same C<$in_fh> works fine though, as this function does not
480move or use the file offset of C<$in_fh>.
381 481
482Please note that C<aio_sendfile> can read more bytes from C<$in_fh> than
483are written, and there is no way to find out how many more bytes have been
484read from C<aio_sendfile> alone, as C<aio_sendfile> only provides the
485number of bytes written to C<$out_fh>. Only if the result value equals
486C<$length> one can assume that C<$length> bytes have been read.
487
488Unlike with other C<aio_> functions, it makes a lot of sense to use
489C<aio_sendfile> on non-blocking sockets, as long as one end (typically
490the C<$in_fh>) is a file - the file I/O will then be asynchronous, while
491the socket I/O will be non-blocking. Note, however, that you can run
492into a trap where C<aio_sendfile> reads some data with readahead, then
493fails to write all data, and when the socket is ready the next time, the
494data in the cache is already lost, forcing C<aio_sendfile> to again hit
495the disk. Explicit C<aio_read> + C<aio_write> let's you better control
496resource usage.
497
382This call tries to make use of a native C<sendfile> syscall to provide 498This call tries to make use of a native C<sendfile>-like syscall to
383zero-copy operation. For this to work, C<$out_fh> should refer to a 499provide zero-copy operation. For this to work, C<$out_fh> should refer to
384socket, and C<$in_fh> should refer to mmap'able file. 500a socket, and C<$in_fh> should refer to an mmap'able file.
385 501
386If a native sendfile cannot be found or it fails with C<ENOSYS>, 502If a native sendfile cannot be found or it fails with C<ENOSYS>,
387C<ENOTSUP>, C<EOPNOTSUPP>, C<EAFNOSUPPORT>, C<EPROTOTYPE> or C<ENOTSOCK>, 503C<EINVAL>, C<ENOTSUP>, C<EOPNOTSUPP>, C<EAFNOSUPPORT>, C<EPROTOTYPE> or
388it will be emulated, so you can call C<aio_sendfile> on any type of 504C<ENOTSOCK>, it will be emulated, so you can call C<aio_sendfile> on any
389filehandle regardless of the limitations of the operating system. 505type of filehandle regardless of the limitations of the operating system.
390 506
391Please note, however, that C<aio_sendfile> can read more bytes from 507As native sendfile syscalls (as practically any non-POSIX interface hacked
392C<$in_fh> than are written, and there is no way to find out how many 508together in a hurry to improve benchmark numbers) tend to be rather buggy
393bytes have been read from C<aio_sendfile> alone, as C<aio_sendfile> only 509on many systems, this implementation tries to work around some known bugs
394provides the number of bytes written to C<$out_fh>. Only if the result 510in Linux and FreeBSD kernels (probably others, too), but that might fail,
395value equals C<$length> one can assume that C<$length> bytes have been 511so you really really should check the return value of C<aio_sendfile> -
396read. 512fewre bytes than expected might have been transferred.
397 513
398 514
399=item aio_readahead $fh,$offset,$length, $callback->($retval) 515=item aio_readahead $fh,$offset,$length, $callback->($retval)
400 516
401C<aio_readahead> populates the page cache with data from a file so that 517C<aio_readahead> populates the page cache with data from a file so that
… …
424 540
425Currently, the stats are always 64-bit-stats, i.e. instead of returning an 541Currently, the stats are always 64-bit-stats, i.e. instead of returning an
426error when stat'ing a large file, the results will be silently truncated 542error when stat'ing a large file, the results will be silently truncated
427unless perl itself is compiled with large file support. 543unless perl itself is compiled with large file support.
428 544
545To help interpret the mode and dev/rdev stat values, IO::AIO offers the
546following constants and functions (if not implemented, the constants will
547be C<0> and the functions will either C<croak> or fall back on traditional
548behaviour).
549
550C<S_IFMT>, C<S_IFIFO>, C<S_IFCHR>, C<S_IFBLK>, C<S_IFLNK>, C<S_IFREG>,
551C<S_IFDIR>, C<S_IFWHT>, C<S_IFSOCK>, C<IO::AIO::major $dev_t>,
552C<IO::AIO::minor $dev_t>, C<IO::AIO::makedev $major, $minor>.
553
429Example: Print the length of F</etc/passwd>: 554Example: Print the length of F</etc/passwd>:
430 555
431 aio_stat "/etc/passwd", sub { 556 aio_stat "/etc/passwd", sub {
432 $_[0] and die "stat failed: $!"; 557 $_[0] and die "stat failed: $!";
433 print "size is ", -s _, "\n"; 558 print "size is ", -s _, "\n";
434 }; 559 };
435 560
436 561
437=item aio_statvfs $fh_or_path, $callback->($statvfs) 562=item aio_statvfs $fh_or_path, $callback->($statvfs)
438 563
439Works like the POSIX C<statvfs> or C<fstatvfs> syscalls, depending on 564Works like the POSIX C<statvfs> or C<fstatvfs> syscalls, depending on
440whether a file handle or path was passed. 565whether a file handle or path was passed.
441 566
442On success, the callback is passed a hash reference with the following 567On success, the callback is passed a hash reference with the following
… …
524 649
525Asynchronously unlink (delete) a file and call the callback with the 650Asynchronously unlink (delete) a file and call the callback with the
526result code. 651result code.
527 652
528 653
529=item aio_mknod $path, $mode, $dev, $callback->($status) 654=item aio_mknod $pathname, $mode, $dev, $callback->($status)
530 655
531[EXPERIMENTAL] 656[EXPERIMENTAL]
532 657
533Asynchronously create a device node (or fifo). See mknod(2). 658Asynchronously create a device node (or fifo). See mknod(2).
534 659
535The only (POSIX-) portable way of calling this function is: 660The only (POSIX-) portable way of calling this function is:
536 661
537 aio_mknod $path, IO::AIO::S_IFIFO | $mode, 0, sub { ... 662 aio_mknod $pathname, IO::AIO::S_IFIFO | $mode, 0, sub { ...
538 663
664See C<aio_stat> for info about some potentially helpful extra constants
665and functions.
539 666
540=item aio_link $srcpath, $dstpath, $callback->($status) 667=item aio_link $srcpath, $dstpath, $callback->($status)
541 668
542Asynchronously create a new link to the existing object at C<$srcpath> at 669Asynchronously create a new link to the existing object at C<$srcpath> at
543the path C<$dstpath> and call the callback with the result code. 670the path C<$dstpath> and call the callback with the result code.
… …
547 674
548Asynchronously create a new symbolic link to the existing object at C<$srcpath> at 675Asynchronously create a new symbolic link to the existing object at C<$srcpath> at
549the path C<$dstpath> and call the callback with the result code. 676the path C<$dstpath> and call the callback with the result code.
550 677
551 678
552=item aio_readlink $path, $callback->($link) 679=item aio_readlink $pathname, $callback->($link)
553 680
554Asynchronously read the symlink specified by C<$path> and pass it to 681Asynchronously read the symlink specified by C<$path> and pass it to
555the callback. If an error occurs, nothing or undef gets passed to the 682the callback. If an error occurs, nothing or undef gets passed to the
556callback. 683callback.
557 684
558 685
686=item aio_realpath $pathname, $callback->($path)
687
688Asynchronously make the path absolute and resolve any symlinks in
689C<$path>. The resulting path only consists of directories (Same as
690L<Cwd::realpath>).
691
692This request can be used to get the absolute path of the current working
693directory by passing it a path of F<.> (a single dot).
694
695
559=item aio_rename $srcpath, $dstpath, $callback->($status) 696=item aio_rename $srcpath, $dstpath, $callback->($status)
560 697
561Asynchronously rename the object at C<$srcpath> to C<$dstpath>, just as 698Asynchronously rename the object at C<$srcpath> to C<$dstpath>, just as
562rename(2) and call the callback with the result code. 699rename(2) and call the callback with the result code.
563 700
… …
585array-ref with the filenames. 722array-ref with the filenames.
586 723
587 724
588=item aio_readdirx $pathname, $flags, $callback->($entries, $flags) 725=item aio_readdirx $pathname, $flags, $callback->($entries, $flags)
589 726
590Quite similar to C<aio_readdir>, but the C<$flags> argument allows to tune 727Quite similar to C<aio_readdir>, but the C<$flags> argument allows one to
591behaviour and output format. In case of an error, C<$entries> will be 728tune behaviour and output format. In case of an error, C<$entries> will be
592C<undef>. 729C<undef>.
593 730
594The flags are a combination of the following constants, ORed together (the 731The flags are a combination of the following constants, ORed together (the
595flags will also be passed to the callback, possibly modified): 732flags will also be passed to the callback, possibly modified):
596 733
597=over 4 734=over 4
598 735
599=item IO::AIO::READDIR_DENTS 736=item IO::AIO::READDIR_DENTS
600 737
601When this flag is off, then the callback gets an arrayref with of names 738When this flag is off, then the callback gets an arrayref consisting of
602only (as with C<aio_readdir>), otherwise it gets an arrayref with 739names only (as with C<aio_readdir>), otherwise it gets an arrayref with
603C<[$name, $type, $inode]> arrayrefs, each describing a single directory 740C<[$name, $type, $inode]> arrayrefs, each describing a single directory
604entry in more detail. 741entry in more detail.
605 742
606C<$name> is the name of the entry. 743C<$name> is the name of the entry.
607 744
… …
620systems that do not deliver the inode information. 757systems that do not deliver the inode information.
621 758
622=item IO::AIO::READDIR_DIRS_FIRST 759=item IO::AIO::READDIR_DIRS_FIRST
623 760
624When this flag is set, then the names will be returned in an order where 761When this flag is set, then the names will be returned in an order where
625likely directories come first. This is useful when you need to quickly 762likely directories come first, in optimal stat order. This is useful when
626find directories, or you want to find all directories while avoiding to 763you need to quickly find directories, or you want to find all directories
627stat() each entry. 764while avoiding to stat() each entry.
628 765
629If the system returns type information in readdir, then this is used 766If the system returns type information in readdir, then this is used
630to find directories directly. Otherwise, likely directories are files 767to find directories directly. Otherwise, likely directories are names
631beginning with ".", or otherwise files with no dots, of which files with 768beginning with ".", or otherwise names with no dots, of which names with
632short names are tried first. 769short names are tried first.
633 770
634=item IO::AIO::READDIR_STAT_ORDER 771=item IO::AIO::READDIR_STAT_ORDER
635 772
636When this flag is set, then the names will be returned in an order 773When this flag is set, then the names will be returned in an order
… …
643 780
644=item IO::AIO::READDIR_FOUND_UNKNOWN 781=item IO::AIO::READDIR_FOUND_UNKNOWN
645 782
646This flag should not be set when calling C<aio_readdirx>. Instead, it 783This flag should not be set when calling C<aio_readdirx>. Instead, it
647is being set by C<aio_readdirx>, when any of the C<$type>'s found were 784is being set by C<aio_readdirx>, when any of the C<$type>'s found were
648C<IO::AIO::DT_UNKNOWN>. The absense of this flag therefore indicates that all 785C<IO::AIO::DT_UNKNOWN>. The absence of this flag therefore indicates that all
649C<$type>'s are known, which can be used to speed up some algorithms. 786C<$type>'s are known, which can be used to speed up some algorithms.
650 787
651=back 788=back
652 789
653 790
654=item aio_load $path, $data, $callback->($status) 791=item aio_load $pathname, $data, $callback->($status)
655 792
656This is a composite request that tries to fully load the given file into 793This is a composite request that tries to fully load the given file into
657memory. Status is the same as with aio_read. 794memory. Status is the same as with aio_read.
658 795
659=cut 796=cut
… …
781 if ($_[0] && $! == EXDEV) { 918 if ($_[0] && $! == EXDEV) {
782 aioreq_pri $pri; 919 aioreq_pri $pri;
783 add $grp aio_copy $src, $dst, sub { 920 add $grp aio_copy $src, $dst, sub {
784 $grp->result ($_[0]); 921 $grp->result ($_[0]);
785 922
786 if (!$_[0]) { 923 unless ($_[0]) {
787 aioreq_pri $pri; 924 aioreq_pri $pri;
788 add $grp aio_unlink $src; 925 add $grp aio_unlink $src;
789 } 926 }
790 }; 927 };
791 } else { 928 } else {
… …
794 }; 931 };
795 932
796 $grp 933 $grp
797} 934}
798 935
799=item aio_scandir $path, $maxreq, $callback->($dirs, $nondirs) 936=item aio_scandir $pathname, $maxreq, $callback->($dirs, $nondirs)
800 937
801Scans a directory (similar to C<aio_readdir>) but additionally tries to 938Scans a directory (similar to C<aio_readdir>) but additionally tries to
802efficiently separate the entries of directory C<$path> into two sets of 939efficiently separate the entries of directory C<$path> into two sets of
803names, directories you can recurse into (directories), and ones you cannot 940names, directories you can recurse into (directories), and ones you cannot
804recurse into (everything else, including symlinks to directories). 941recurse into (everything else, including symlinks to directories).
… …
835Then entries will be sorted into likely directories a non-initial dot 972Then entries will be sorted into likely directories a non-initial dot
836currently) and likely non-directories (see C<aio_readdirx>). Then every 973currently) and likely non-directories (see C<aio_readdirx>). Then every
837entry plus an appended C</.> will be C<stat>'ed, likely directories first, 974entry plus an appended C</.> will be C<stat>'ed, likely directories first,
838in order of their inode numbers. If that succeeds, it assumes that the 975in order of their inode numbers. If that succeeds, it assumes that the
839entry is a directory or a symlink to directory (which will be checked 976entry is a directory or a symlink to directory (which will be checked
840seperately). This is often faster than stat'ing the entry itself because 977separately). This is often faster than stat'ing the entry itself because
841filesystems might detect the type of the entry without reading the inode 978filesystems might detect the type of the entry without reading the inode
842data (e.g. ext2fs filetype feature), even on systems that cannot return 979data (e.g. ext2fs filetype feature), even on systems that cannot return
843the filetype information on readdir. 980the filetype information on readdir.
844 981
845If the known number of directories (link count - 2) has been reached, the 982If the known number of directories (link count - 2) has been reached, the
… …
861 998
862 my $grp = aio_group $cb; 999 my $grp = aio_group $cb;
863 1000
864 $maxreq = 4 if $maxreq <= 0; 1001 $maxreq = 4 if $maxreq <= 0;
865 1002
866 # stat once 1003 # get a wd object
867 aioreq_pri $pri; 1004 aioreq_pri $pri;
868 add $grp aio_stat $path, sub { 1005 add $grp aio_wd $path, sub {
1006 $_[0]
869 return $grp->result () if $_[0]; 1007 or return $grp->result ();
870 my $now = time;
871 my $hash1 = join ":", (stat _)[0,1,3,7,9];
872 1008
873 # read the directory entries 1009 my $wd = [shift, "."];
1010
1011 # stat once
874 aioreq_pri $pri; 1012 aioreq_pri $pri;
875 add $grp aio_readdirx $path, READDIR_DIRS_FIRST, sub { 1013 add $grp aio_stat $wd, sub {
876 my $entries = shift
877 or return $grp->result (); 1014 return $grp->result () if $_[0];
1015 my $now = time;
1016 my $hash1 = join ":", (stat _)[0,1,3,7,9];
878 1017
879 # stat the dir another time 1018 # read the directory entries
880 aioreq_pri $pri; 1019 aioreq_pri $pri;
1020 add $grp aio_readdirx $wd, READDIR_DIRS_FIRST, sub {
1021 my $entries = shift
1022 or return $grp->result ();
1023
1024 # stat the dir another time
1025 aioreq_pri $pri;
881 add $grp aio_stat $path, sub { 1026 add $grp aio_stat $wd, sub {
882 my $hash2 = join ":", (stat _)[0,1,3,7,9]; 1027 my $hash2 = join ":", (stat _)[0,1,3,7,9];
883 1028
884 my $ndirs; 1029 my $ndirs;
885 1030
886 # take the slow route if anything looks fishy 1031 # take the slow route if anything looks fishy
887 if ($hash1 ne $hash2 or (stat _)[9] == $now) { 1032 if ($hash1 ne $hash2 or (stat _)[9] == $now) {
888 $ndirs = -1; 1033 $ndirs = -1;
889 } else { 1034 } else {
890 # if nlink == 2, we are finished 1035 # if nlink == 2, we are finished
891 # for non-posix-fs's, we rely on nlink < 2 1036 # for non-posix-fs's, we rely on nlink < 2
892 $ndirs = (stat _)[3] - 2 1037 $ndirs = (stat _)[3] - 2
893 or return $grp->result ([], $entries); 1038 or return $grp->result ([], $entries);
894 } 1039 }
895 1040
896 my (@dirs, @nondirs); 1041 my (@dirs, @nondirs);
897 1042
898 my $statgrp = add $grp aio_group sub { 1043 my $statgrp = add $grp aio_group sub {
899 $grp->result (\@dirs, \@nondirs); 1044 $grp->result (\@dirs, \@nondirs);
900 }; 1045 };
901 1046
902 limit $statgrp $maxreq; 1047 limit $statgrp $maxreq;
903 feed $statgrp sub { 1048 feed $statgrp sub {
904 return unless @$entries; 1049 return unless @$entries;
905 my $entry = shift @$entries; 1050 my $entry = shift @$entries;
906 1051
907 aioreq_pri $pri; 1052 aioreq_pri $pri;
1053 $wd->[1] = "$entry/.";
908 add $statgrp aio_stat "$path/$entry/.", sub { 1054 add $statgrp aio_stat $wd, sub {
909 if ($_[0] < 0) { 1055 if ($_[0] < 0) {
910 push @nondirs, $entry; 1056 push @nondirs, $entry;
911 } else { 1057 } else {
912 # need to check for real directory 1058 # need to check for real directory
913 aioreq_pri $pri; 1059 aioreq_pri $pri;
1060 $wd->[1] = $entry;
914 add $statgrp aio_lstat "$path/$entry", sub { 1061 add $statgrp aio_lstat $wd, sub {
915 if (-d _) { 1062 if (-d _) {
916 push @dirs, $entry; 1063 push @dirs, $entry;
917 1064
918 unless (--$ndirs) { 1065 unless (--$ndirs) {
919 push @nondirs, @$entries; 1066 push @nondirs, @$entries;
920 feed $statgrp; 1067 feed $statgrp;
1068 }
1069 } else {
1070 push @nondirs, $entry;
921 } 1071 }
922 } else {
923 push @nondirs, $entry;
924 } 1072 }
925 } 1073 }
926 } 1074 };
927 }; 1075 };
928 }; 1076 };
929 }; 1077 };
930 }; 1078 };
931 }; 1079 };
932 1080
933 $grp 1081 $grp
934} 1082}
935 1083
936=item aio_rmtree $path, $callback->($status) 1084=item aio_rmtree $pathname, $callback->($status)
937 1085
938Delete a directory tree starting (and including) C<$path>, return the 1086Delete a directory tree starting (and including) C<$path>, return the
939status of the final C<rmdir> only. This is a composite request that 1087status of the final C<rmdir> only. This is a composite request that
940uses C<aio_scandir> to recurse into and rmdir directories, and unlink 1088uses C<aio_scandir> to recurse into and rmdir directories, and unlink
941everything else. 1089everything else.
… …
983callback with the fdatasync result code. 1131callback with the fdatasync result code.
984 1132
985If this call isn't available because your OS lacks it or it couldn't be 1133If this call isn't available because your OS lacks it or it couldn't be
986detected, it will be emulated by calling C<fsync> instead. 1134detected, it will be emulated by calling C<fsync> instead.
987 1135
1136=item aio_syncfs $fh, $callback->($status)
1137
1138Asynchronously call the syncfs syscall to sync the filesystem associated
1139to the given filehandle and call the callback with the syncfs result
1140code. If syncfs is not available, calls sync(), but returns C<-1> and sets
1141errno to C<ENOSYS> nevertheless.
1142
988=item aio_sync_file_range $fh, $offset, $nbytes, $flags, $callback->($status) 1143=item aio_sync_file_range $fh, $offset, $nbytes, $flags, $callback->($status)
989 1144
990Sync the data portion of the file specified by C<$offset> and C<$length> 1145Sync the data portion of the file specified by C<$offset> and C<$length>
991to disk (but NOT the metadata), by calling the Linux-specific 1146to disk (but NOT the metadata), by calling the Linux-specific
992sync_file_range call. If sync_file_range is not available or it returns 1147sync_file_range call. If sync_file_range is not available or it returns
… …
995C<$flags> can be a combination of C<IO::AIO::SYNC_FILE_RANGE_WAIT_BEFORE>, 1150C<$flags> can be a combination of C<IO::AIO::SYNC_FILE_RANGE_WAIT_BEFORE>,
996C<IO::AIO::SYNC_FILE_RANGE_WRITE> and 1151C<IO::AIO::SYNC_FILE_RANGE_WRITE> and
997C<IO::AIO::SYNC_FILE_RANGE_WAIT_AFTER>: refer to the sync_file_range 1152C<IO::AIO::SYNC_FILE_RANGE_WAIT_AFTER>: refer to the sync_file_range
998manpage for details. 1153manpage for details.
999 1154
1000=item aio_pathsync $path, $callback->($status) 1155=item aio_pathsync $pathname, $callback->($status)
1001 1156
1002This request tries to open, fsync and close the given path. This is a 1157This request tries to open, fsync and close the given path. This is a
1003composite request intended to sync directories after directory operations 1158composite request intended to sync directories after directory operations
1004(E.g. rename). This might not work on all operating systems or have any 1159(E.g. rename). This might not work on all operating systems or have any
1005specific effect, but usually it makes sure that directory changes get 1160specific effect, but usually it makes sure that directory changes get
… …
1039} 1194}
1040 1195
1041=item aio_msync $scalar, $offset = 0, $length = undef, flags = 0, $callback->($status) 1196=item aio_msync $scalar, $offset = 0, $length = undef, flags = 0, $callback->($status)
1042 1197
1043This is a rather advanced IO::AIO call, which only works on mmap(2)ed 1198This is a rather advanced IO::AIO call, which only works on mmap(2)ed
1044scalars (see the L<Sys::Mmap> or L<Mmap> modules for details on this, note 1199scalars (see the C<IO::AIO::mmap> function, although it also works on data
1200scalars managed by the L<Sys::Mmap> or L<Mmap> modules, note that the
1045that the scalar must only be modified in-place while an aio operation is 1201scalar must only be modified in-place while an aio operation is pending on
1046pending on it). 1202it).
1047 1203
1048It calls the C<msync> function of your OS, if available, with the memory 1204It calls the C<msync> function of your OS, if available, with the memory
1049area starting at C<$offset> in the string and ending C<$length> bytes 1205area starting at C<$offset> in the string and ending C<$length> bytes
1050later. If C<$length> is negative, counts from the end, and if C<$length> 1206later. If C<$length> is negative, counts from the end, and if C<$length>
1051is C<undef>, then it goes till the end of the string. The flags can be 1207is C<undef>, then it goes till the end of the string. The flags can be
… …
1062as for C<aio_msync>, above, except for flags, which must be either 1218as for C<aio_msync>, above, except for flags, which must be either
1063C<0> (which reads all pages and ensures they are instantiated) or 1219C<0> (which reads all pages and ensures they are instantiated) or
1064C<IO::AIO::MT_MODIFY>, which modifies the memory page s(by reading and 1220C<IO::AIO::MT_MODIFY>, which modifies the memory page s(by reading and
1065writing an octet from it, which dirties the page). 1221writing an octet from it, which dirties the page).
1066 1222
1223=item aio_mlock $scalar, $offset = 0, $length = undef, $callback->($status)
1224
1225This is a rather advanced IO::AIO call, which works best on mmap(2)ed
1226scalars.
1227
1228It reads in all the pages of the underlying storage into memory (if any)
1229and locks them, so they are not getting swapped/paged out or removed.
1230
1231If C<$length> is undefined, then the scalar will be locked till the end.
1232
1233On systems that do not implement C<mlock>, this function returns C<-1>
1234and sets errno to C<ENOSYS>.
1235
1236Note that the corresponding C<munlock> is synchronous and is
1237documented under L<MISCELLANEOUS FUNCTIONS>.
1238
1239Example: open a file, mmap and mlock it - both will be undone when
1240C<$data> gets destroyed.
1241
1242 open my $fh, "<", $path or die "$path: $!";
1243 my $data;
1244 IO::AIO::mmap $data, -s $fh, IO::AIO::PROT_READ, IO::AIO::MAP_SHARED, $fh;
1245 aio_mlock $data; # mlock in background
1246
1247=item aio_mlockall $flags, $callback->($status)
1248
1249Calls the C<mlockall> function with the given C<$flags> (a combination of
1250C<IO::AIO::MCL_CURRENT> and C<IO::AIO::MCL_FUTURE>).
1251
1252On systems that do not implement C<mlockall>, this function returns C<-1>
1253and sets errno to C<ENOSYS>.
1254
1255Note that the corresponding C<munlockall> is synchronous and is
1256documented under L<MISCELLANEOUS FUNCTIONS>.
1257
1258Example: asynchronously lock all current and future pages into memory.
1259
1260 aio_mlockall IO::AIO::MCL_FUTURE;
1261
1262=item aio_fiemap $fh, $start, $length, $flags, $count, $cb->(\@extents)
1263
1264Queries the extents of the given file (by calling the Linux FIEMAP ioctl,
1265see L<http://cvs.schmorp.de/IO-AIO/doc/fiemap.txt> for details). If the
1266C<ioctl> is not available on your OS, then this rquiest will fail with
1267C<ENOSYS>.
1268
1269C<$start> is the starting offset to query extents for, C<$length> is the
1270size of the range to query - if it is C<undef>, then the whole file will
1271be queried.
1272
1273C<$flags> is a combination of flags (C<IO::AIO::FIEMAP_FLAG_SYNC> or
1274C<IO::AIO::FIEMAP_FLAG_XATTR> - C<IO::AIO::FIEMAP_FLAGS_COMPAT> is also
1275exported), and is normally C<0> or C<IO::AIO::FIEMAP_FLAG_SYNC> to query
1276the data portion.
1277
1278C<$count> is the maximum number of extent records to return. If it is
1279C<undef>, then IO::AIO queries all extents of the file. As a very special
1280case, if it is C<0>, then the callback receives the number of extents
1281instead of the extents themselves.
1282
1283If an error occurs, the callback receives no arguments. The special
1284C<errno> value C<IO::AIO::EBADR> is available to test for flag errors.
1285
1286Otherwise, the callback receives an array reference with extent
1287structures. Each extent structure is an array reference itself, with the
1288following members:
1289
1290 [$logical, $physical, $length, $flags]
1291
1292Flags is any combination of the following flag values (typically either C<0>
1293or C<IO::AIO::FIEMAP_EXTENT_LAST>):
1294
1295C<IO::AIO::FIEMAP_EXTENT_LAST>, C<IO::AIO::FIEMAP_EXTENT_UNKNOWN>,
1296C<IO::AIO::FIEMAP_EXTENT_DELALLOC>, C<IO::AIO::FIEMAP_EXTENT_ENCODED>,
1297C<IO::AIO::FIEMAP_EXTENT_DATA_ENCRYPTED>, C<IO::AIO::FIEMAP_EXTENT_NOT_ALIGNED>,
1298C<IO::AIO::FIEMAP_EXTENT_DATA_INLINE>, C<IO::AIO::FIEMAP_EXTENT_DATA_TAIL>,
1299C<IO::AIO::FIEMAP_EXTENT_UNWRITTEN>, C<IO::AIO::FIEMAP_EXTENT_MERGED> or
1300C<IO::AIO::FIEMAP_EXTENT_SHARED>.
1301
1067=item aio_group $callback->(...) 1302=item aio_group $callback->(...)
1068 1303
1069This is a very special aio request: Instead of doing something, it is a 1304This is a very special aio request: Instead of doing something, it is a
1070container for other aio requests, which is useful if you want to bundle 1305container for other aio requests, which is useful if you want to bundle
1071many requests into a single, composite, request with a definite callback 1306many requests into a single, composite, request with a definite callback
… …
1107like sleep and file handle readable/writable, the overhead this creates is 1342like sleep and file handle readable/writable, the overhead this creates is
1108immense (it blocks a thread for a long time) so do not use this function 1343immense (it blocks a thread for a long time) so do not use this function
1109except to put your application under artificial I/O pressure. 1344except to put your application under artificial I/O pressure.
1110 1345
1111=back 1346=back
1347
1348
1349=head2 IO::AIO::WD - multiple working directories
1350
1351Your process only has one current working directory, which is used by all
1352threads. This makes it hard to use relative paths (some other component
1353could call C<chdir> at any time, and it is hard to control when the path
1354will be used by IO::AIO).
1355
1356One solution for this is to always use absolute paths. This usually works,
1357but can be quite slow (the kernel has to walk the whole path on every
1358access), and can also be a hassle to implement.
1359
1360Newer POSIX systems have a number of functions (openat, fdopendir,
1361futimensat and so on) that make it possible to specify working directories
1362per operation.
1363
1364For portability, and because the clowns who "designed", or shall I write,
1365perpetrated this new interface were obviously half-drunk, this abstraction
1366cannot be perfect, though.
1367
1368IO::AIO allows you to convert directory paths into a so-called IO::AIO::WD
1369object. This object stores the canonicalised, absolute version of the
1370path, and on systems that allow it, also a directory file descriptor.
1371
1372Everywhere where a pathname is accepted by IO::AIO (e.g. in C<aio_stat>
1373or C<aio_unlink>), one can specify an array reference with an IO::AIO::WD
1374object and a pathname instead (or the IO::AIO::WD object alone, which
1375gets interpreted as C<[$wd, "."]>). If the pathname is absolute, the
1376IO::AIO::WD object is ignored, otherwise the pathname is resolved relative
1377to that IO::AIO::WD object.
1378
1379For example, to get a wd object for F</etc> and then stat F<passwd>
1380inside, you would write:
1381
1382 aio_wd "/etc", sub {
1383 my $etcdir = shift;
1384
1385 # although $etcdir can be undef on error, there is generally no reason
1386 # to check for errors here, as aio_stat will fail with ENOENT
1387 # when $etcdir is undef.
1388
1389 aio_stat [$etcdir, "passwd"], sub {
1390 # yay
1391 };
1392 };
1393
1394That C<aio_wd> is a request and not a normal function shows that creating
1395an IO::AIO::WD object is itself a potentially blocking operation, which is
1396why it is done asynchronously.
1397
1398To stat the directory obtained with C<aio_wd> above, one could write
1399either of the following three request calls:
1400
1401 aio_lstat "/etc" , sub { ... # pathname as normal string
1402 aio_lstat [$wd, "."], sub { ... # "." relative to $wd (i.e. $wd itself)
1403 aio_lstat $wd , sub { ... # shorthand for the previous
1404
1405As with normal pathnames, IO::AIO keeps a copy of the working directory
1406object and the pathname string, so you could write the following without
1407causing any issues due to C<$path> getting reused:
1408
1409 my $path = [$wd, undef];
1410
1411 for my $name (qw(abc def ghi)) {
1412 $path->[1] = $name;
1413 aio_stat $path, sub {
1414 # ...
1415 };
1416 }
1417
1418There are some caveats: when directories get renamed (or deleted), the
1419pathname string doesn't change, so will point to the new directory (or
1420nowhere at all), while the directory fd, if available on the system,
1421will still point to the original directory. Most functions accepting a
1422pathname will use the directory fd on newer systems, and the string on
1423older systems. Some functions (such as realpath) will always rely on the
1424string form of the pathname.
1425
1426So this fucntionality is mainly useful to get some protection against
1427C<chdir>, to easily get an absolute path out of a relative path for future
1428reference, and to speed up doing many operations in the same directory
1429(e.g. when stat'ing all files in a directory).
1430
1431The following functions implement this working directory abstraction:
1432
1433=over 4
1434
1435=item aio_wd $pathname, $callback->($wd)
1436
1437Asynchonously canonicalise the given pathname and convert it to an
1438IO::AIO::WD object representing it. If possible and supported on the
1439system, also open a directory fd to speed up pathname resolution relative
1440to this working directory.
1441
1442If something goes wrong, then C<undef> is passwd to the callback instead
1443of a working directory object and C<$!> is set appropriately. Since
1444passing C<undef> as working directory component of a pathname fails the
1445request with C<ENOENT>, there is often no need for error checking in the
1446C<aio_wd> callback, as future requests using the value will fail in the
1447expected way.
1448
1449If this call isn't available because your OS lacks it or it couldn't be
1450detected, it will be emulated by calling C<fsync> instead.
1451
1452=item IO::AIO::CWD
1453
1454This is a compiletime constant (object) that represents the process
1455current working directory.
1456
1457Specifying this object as working directory object for a pathname is as
1458if the pathname would be specified directly, without a directory object,
1459e.g., these calls are functionally identical:
1460
1461 aio_stat "somefile", sub { ... };
1462 aio_stat [IO::AIO::CWD, "somefile"], sub { ... };
1463
1464=back
1465
1112 1466
1113=head2 IO::AIO::REQ CLASS 1467=head2 IO::AIO::REQ CLASS
1114 1468
1115All non-aggregate C<aio_*> functions return an object of this class when 1469All non-aggregate C<aio_*> functions return an object of this class when
1116called in non-void context. 1470called in non-void context.
… …
1234 1588
1235Sets a feeder/generator on this group: every group can have an attached 1589Sets a feeder/generator on this group: every group can have an attached
1236generator that generates requests if idle. The idea behind this is that, 1590generator that generates requests if idle. The idea behind this is that,
1237although you could just queue as many requests as you want in a group, 1591although you could just queue as many requests as you want in a group,
1238this might starve other requests for a potentially long time. For example, 1592this might starve other requests for a potentially long time. For example,
1239C<aio_scandir> might generate hundreds of thousands C<aio_stat> requests, 1593C<aio_scandir> might generate hundreds of thousands of C<aio_stat>
1240delaying any later requests for a long time. 1594requests, delaying any later requests for a long time.
1241 1595
1242To avoid this, and allow incremental generation of requests, you can 1596To avoid this, and allow incremental generation of requests, you can
1243instead a group and set a feeder on it that generates those requests. The 1597instead a group and set a feeder on it that generates those requests. The
1244feed callback will be called whenever there are few enough (see C<limit>, 1598feed callback will be called whenever there are few enough (see C<limit>,
1245below) requests active in the group itself and is expected to queue more 1599below) requests active in the group itself and is expected to queue more
… …
1294 1648
1295See C<poll_cb> for an example. 1649See C<poll_cb> for an example.
1296 1650
1297=item IO::AIO::poll_cb 1651=item IO::AIO::poll_cb
1298 1652
1299Process some outstanding events on the result pipe. You have to call this 1653Process some outstanding events on the result pipe. You have to call
1300regularly. Returns C<0> if all events could be processed, or C<-1> if it 1654this regularly. Returns C<0> if all events could be processed (or there
1301returned earlier for whatever reason. Returns immediately when no events 1655were no events to process), or C<-1> if it returned earlier for whatever
1302are outstanding. The amount of events processed depends on the settings of 1656reason. Returns immediately when no events are outstanding. The amount of
1303C<IO::AIO::max_poll_req> and C<IO::AIO::max_poll_time>. 1657events processed depends on the settings of C<IO::AIO::max_poll_req> and
1658C<IO::AIO::max_poll_time>.
1304 1659
1305If not all requests were processed for whatever reason, the filehandle 1660If not all requests were processed for whatever reason, the filehandle
1306will still be ready when C<poll_cb> returns, so normally you don't have to 1661will still be ready when C<poll_cb> returns, so normally you don't have to
1307do anything special to have it called later. 1662do anything special to have it called later.
1308 1663
1664Apart from calling C<IO::AIO::poll_cb> when the event filehandle becomes
1665ready, it can be beneficial to call this function from loops which submit
1666a lot of requests, to make sure the results get processed when they become
1667available and not just when the loop is finished and the event loop takes
1668over again. This function returns very fast when there are no outstanding
1669requests.
1670
1309Example: Install an Event watcher that automatically calls 1671Example: Install an Event watcher that automatically calls
1310IO::AIO::poll_cb with high priority (more examples can be found in the 1672IO::AIO::poll_cb with high priority (more examples can be found in the
1311SYNOPSIS section, at the top of this document): 1673SYNOPSIS section, at the top of this document):
1312 1674
1313 Event->io (fd => IO::AIO::poll_fileno, 1675 Event->io (fd => IO::AIO::poll_fileno,
1314 poll => 'r', async => 1, 1676 poll => 'r', async => 1,
1315 cb => \&IO::AIO::poll_cb); 1677 cb => \&IO::AIO::poll_cb);
1678
1679=item IO::AIO::poll_wait
1680
1681If there are any outstanding requests and none of them in the result
1682phase, wait till the result filehandle becomes ready for reading (simply
1683does a C<select> on the filehandle. This is useful if you want to
1684synchronously wait for some requests to finish).
1685
1686See C<nreqs> for an example.
1687
1688=item IO::AIO::poll
1689
1690Waits until some requests have been handled.
1691
1692Returns the number of requests processed, but is otherwise strictly
1693equivalent to:
1694
1695 IO::AIO::poll_wait, IO::AIO::poll_cb
1696
1697=item IO::AIO::flush
1698
1699Wait till all outstanding AIO requests have been handled.
1700
1701Strictly equivalent to:
1702
1703 IO::AIO::poll_wait, IO::AIO::poll_cb
1704 while IO::AIO::nreqs;
1316 1705
1317=item IO::AIO::max_poll_reqs $nreqs 1706=item IO::AIO::max_poll_reqs $nreqs
1318 1707
1319=item IO::AIO::max_poll_time $seconds 1708=item IO::AIO::max_poll_time $seconds
1320 1709
… …
1345 # use a low priority so other tasks have priority 1734 # use a low priority so other tasks have priority
1346 Event->io (fd => IO::AIO::poll_fileno, 1735 Event->io (fd => IO::AIO::poll_fileno,
1347 poll => 'r', nice => 1, 1736 poll => 'r', nice => 1,
1348 cb => &IO::AIO::poll_cb); 1737 cb => &IO::AIO::poll_cb);
1349 1738
1350=item IO::AIO::poll_wait
1351
1352If there are any outstanding requests and none of them in the result
1353phase, wait till the result filehandle becomes ready for reading (simply
1354does a C<select> on the filehandle. This is useful if you want to
1355synchronously wait for some requests to finish).
1356
1357See C<nreqs> for an example.
1358
1359=item IO::AIO::poll
1360
1361Waits until some requests have been handled.
1362
1363Returns the number of requests processed, but is otherwise strictly
1364equivalent to:
1365
1366 IO::AIO::poll_wait, IO::AIO::poll_cb
1367
1368=item IO::AIO::flush
1369
1370Wait till all outstanding AIO requests have been handled.
1371
1372Strictly equivalent to:
1373
1374 IO::AIO::poll_wait, IO::AIO::poll_cb
1375 while IO::AIO::nreqs;
1376
1377=back 1739=back
1378 1740
1379=head3 CONTROLLING THE NUMBER OF THREADS 1741=head3 CONTROLLING THE NUMBER OF THREADS
1380 1742
1381=over 1743=over
… …
1414 1776
1415Under normal circumstances you don't need to call this function. 1777Under normal circumstances you don't need to call this function.
1416 1778
1417=item IO::AIO::max_idle $nthreads 1779=item IO::AIO::max_idle $nthreads
1418 1780
1419Limit the number of threads (default: 4) that are allowed to idle (i.e., 1781Limit the number of threads (default: 4) that are allowed to idle
1420threads that did not get a request to process within 10 seconds). That 1782(i.e., threads that did not get a request to process within the idle
1421means if a thread becomes idle while C<$nthreads> other threads are also 1783timeout (default: 10 seconds). That means if a thread becomes idle while
1422idle, it will free its resources and exit. 1784C<$nthreads> other threads are also idle, it will free its resources and
1785exit.
1423 1786
1424This is useful when you allow a large number of threads (e.g. 100 or 1000) 1787This is useful when you allow a large number of threads (e.g. 100 or 1000)
1425to allow for extremely high load situations, but want to free resources 1788to allow for extremely high load situations, but want to free resources
1426under normal circumstances (1000 threads can easily consume 30MB of RAM). 1789under normal circumstances (1000 threads can easily consume 30MB of RAM).
1427 1790
1428The default is probably ok in most situations, especially if thread 1791The default is probably ok in most situations, especially if thread
1429creation is fast. If thread creation is very slow on your system you might 1792creation is fast. If thread creation is very slow on your system you might
1430want to use larger values. 1793want to use larger values.
1431 1794
1795=item IO::AIO::idle_timeout $seconds
1796
1797Sets the minimum idle timeout (default 10) after which worker threads are
1798allowed to exit. SEe C<IO::AIO::max_idle>.
1799
1432=item IO::AIO::max_outstanding $maxreqs 1800=item IO::AIO::max_outstanding $maxreqs
1801
1802Sets the maximum number of outstanding requests to C<$nreqs>. If
1803you do queue up more than this number of requests, the next call to
1804C<IO::AIO::poll_cb> (and other functions calling C<poll_cb>, such as
1805C<IO::AIO::flush> or C<IO::AIO::poll>) will block until the limit is no
1806longer exceeded.
1807
1808In other words, this setting does not enforce a queue limit, but can be
1809used to make poll functions block if the limit is exceeded.
1433 1810
1434This is a very bad function to use in interactive programs because it 1811This is a very bad function to use in interactive programs because it
1435blocks, and a bad way to reduce concurrency because it is inexact: Better 1812blocks, and a bad way to reduce concurrency because it is inexact: Better
1436use an C<aio_group> together with a feed callback. 1813use an C<aio_group> together with a feed callback.
1437 1814
1438Sets the maximum number of outstanding requests to C<$nreqs>. If you 1815It's main use is in scripts without an event loop - when you want to stat
1439do queue up more than this number of requests, the next call to the 1816a lot of files, you can write somehting like this:
1440C<poll_cb> (and C<poll_some> and other functions calling C<poll_cb>)
1441function will block until the limit is no longer exceeded.
1442 1817
1443The default value is very large, so there is no practical limit on the 1818 IO::AIO::max_outstanding 32;
1444number of outstanding requests.
1445 1819
1446You can still queue as many requests as you want. Therefore, 1820 for my $path (...) {
1447C<max_outstanding> is mainly useful in simple scripts (with low values) or 1821 aio_stat $path , ...;
1448as a stop gap to shield against fatal memory overflow (with large values). 1822 IO::AIO::poll_cb;
1823 }
1824
1825 IO::AIO::flush;
1826
1827The call to C<poll_cb> inside the loop will normally return instantly, but
1828as soon as more thna C<32> reqeusts are in-flight, it will block until
1829some requests have been handled. This keeps the loop from pushing a large
1830number of C<aio_stat> requests onto the queue.
1831
1832The default value for C<max_outstanding> is very large, so there is no
1833practical limit on the number of outstanding requests.
1449 1834
1450=back 1835=back
1451 1836
1452=head3 STATISTICAL INFORMATION 1837=head3 STATISTICAL INFORMATION
1453 1838
… …
1491 1876
1492Returns the number of bytes copied, or C<-1> on error. 1877Returns the number of bytes copied, or C<-1> on error.
1493 1878
1494=item IO::AIO::fadvise $fh, $offset, $len, $advice 1879=item IO::AIO::fadvise $fh, $offset, $len, $advice
1495 1880
1496Simply calls the C<posix_fadvise> function (see it's 1881Simply calls the C<posix_fadvise> function (see its
1497manpage for details). The following advice constants are 1882manpage for details). The following advice constants are
1498avaiable: C<IO::AIO::FADV_NORMAL>, C<IO::AIO::FADV_SEQUENTIAL>, 1883available: C<IO::AIO::FADV_NORMAL>, C<IO::AIO::FADV_SEQUENTIAL>,
1499C<IO::AIO::FADV_RANDOM>, C<IO::AIO::FADV_NOREUSE>, 1884C<IO::AIO::FADV_RANDOM>, C<IO::AIO::FADV_NOREUSE>,
1500C<IO::AIO::FADV_WILLNEED>, C<IO::AIO::FADV_DONTNEED>. 1885C<IO::AIO::FADV_WILLNEED>, C<IO::AIO::FADV_DONTNEED>.
1501 1886
1502On systems that do not implement C<posix_fadvise>, this function returns 1887On systems that do not implement C<posix_fadvise>, this function returns
1503ENOSYS, otherwise the return value of C<posix_fadvise>. 1888ENOSYS, otherwise the return value of C<posix_fadvise>.
1504 1889
1505=item IO::AIO::mlockall $flags 1890=item IO::AIO::madvise $scalar, $offset, $len, $advice
1506 1891
1507Calls the C<mlockall> function with the given C<$flags> (a combination of 1892Simply calls the C<posix_madvise> function (see its
1508C<IO::AIO::MCL_CURRENT> and C<IO::AIO::MCL__FUTURE>). 1893manpage for details). The following advice constants are
1894available: C<IO::AIO::MADV_NORMAL>, C<IO::AIO::MADV_SEQUENTIAL>,
1895C<IO::AIO::MADV_RANDOM>, C<IO::AIO::MADV_WILLNEED>, C<IO::AIO::MADV_DONTNEED>.
1509 1896
1897On systems that do not implement C<posix_madvise>, this function returns
1898ENOSYS, otherwise the return value of C<posix_madvise>.
1899
1900=item IO::AIO::mprotect $scalar, $offset, $len, $protect
1901
1902Simply calls the C<mprotect> function on the preferably AIO::mmap'ed
1903$scalar (see its manpage for details). The following protect
1904constants are available: C<IO::AIO::PROT_NONE>, C<IO::AIO::PROT_READ>,
1905C<IO::AIO::PROT_WRITE>, C<IO::AIO::PROT_EXEC>.
1906
1510On systems that do not implement C<mlockall>, this function returns 1907On systems that do not implement C<mprotect>, this function returns
1511ENOSYS, otherwise the return value of C<mlockall>. 1908ENOSYS, otherwise the return value of C<mprotect>.
1909
1910=item IO::AIO::mmap $scalar, $length, $prot, $flags, $fh[, $offset]
1911
1912Memory-maps a file (or anonymous memory range) and attaches it to the
1913given C<$scalar>, which will act like a string scalar. Returns true on
1914success, and false otherwise.
1915
1916The only operations allowed on the scalar are C<substr>/C<vec> that don't
1917change the string length, and most read-only operations such as copying it
1918or searching it with regexes and so on.
1919
1920Anything else is unsafe and will, at best, result in memory leaks.
1921
1922The memory map associated with the C<$scalar> is automatically removed
1923when the C<$scalar> is destroyed, or when the C<IO::AIO::mmap> or
1924C<IO::AIO::munmap> functions are called.
1925
1926This calls the C<mmap>(2) function internally. See your system's manual
1927page for details on the C<$length>, C<$prot> and C<$flags> parameters.
1928
1929The C<$length> must be larger than zero and smaller than the actual
1930filesize.
1931
1932C<$prot> is a combination of C<IO::AIO::PROT_NONE>, C<IO::AIO::PROT_EXEC>,
1933C<IO::AIO::PROT_READ> and/or C<IO::AIO::PROT_WRITE>,
1934
1935C<$flags> can be a combination of C<IO::AIO::MAP_SHARED> or
1936C<IO::AIO::MAP_PRIVATE>, or a number of system-specific flags (when
1937not available, the are defined as 0): C<IO::AIO::MAP_ANONYMOUS>
1938(which is set to C<MAP_ANON> if your system only provides this
1939constant), C<IO::AIO::MAP_HUGETLB>, C<IO::AIO::MAP_LOCKED>,
1940C<IO::AIO::MAP_NORESERVE>, C<IO::AIO::MAP_POPULATE> or
1941C<IO::AIO::MAP_NONBLOCK>
1942
1943If C<$fh> is C<undef>, then a file descriptor of C<-1> is passed.
1944
1945C<$offset> is the offset from the start of the file - it generally must be
1946a multiple of C<IO::AIO::PAGESIZE> and defaults to C<0>.
1947
1948Example:
1949
1950 use Digest::MD5;
1951 use IO::AIO;
1952
1953 open my $fh, "<verybigfile"
1954 or die "$!";
1955
1956 IO::AIO::mmap my $data, -s $fh, IO::AIO::PROT_READ, IO::AIO::MAP_SHARED, $fh
1957 or die "verybigfile: $!";
1958
1959 my $fast_md5 = md5 $data;
1960
1961=item IO::AIO::munmap $scalar
1962
1963Removes a previous mmap and undefines the C<$scalar>.
1964
1965=item IO::AIO::munlock $scalar, $offset = 0, $length = undef
1966
1967Calls the C<munlock> function, undoing the effects of a previous
1968C<aio_mlock> call (see its description for details).
1512 1969
1513=item IO::AIO::munlockall 1970=item IO::AIO::munlockall
1514 1971
1515Calls the C<munlockall> function. 1972Calls the C<munlockall> function.
1516 1973
1517On systems that do not implement C<munlockall>, this function returns 1974On systems that do not implement C<munlockall>, this function returns
1518ENOSYS, otherwise the return value of C<munlockall>. 1975ENOSYS, otherwise the return value of C<munlockall>.
1519 1976
1977=item IO::AIO::splice $r_fh, $r_off, $w_fh, $w_off, $length, $flags
1978
1979Calls the GNU/Linux C<splice(2)> syscall, if available. If C<$r_off> or
1980C<$w_off> are C<undef>, then C<NULL> is passed for these, otherwise they
1981should be the file offset.
1982
1983C<$r_fh> and C<$w_fh> should not refer to the same file, as splice might
1984silently corrupt the data in this case.
1985
1986The following symbol flag values are available: C<IO::AIO::SPLICE_F_MOVE>,
1987C<IO::AIO::SPLICE_F_NONBLOCK>, C<IO::AIO::SPLICE_F_MORE> and
1988C<IO::AIO::SPLICE_F_GIFT>.
1989
1990See the C<splice(2)> manpage for details.
1991
1992=item IO::AIO::tee $r_fh, $w_fh, $length, $flags
1993
1994Calls the GNU/Linux C<tee(2)> syscall, see it's manpage and the
1995description for C<IO::AIO::splice> above for details.
1996
1520=back 1997=back
1521 1998
1522=cut 1999=cut
1523 2000
1524min_parallel 8; 2001min_parallel 8;
1525 2002
1526END { flush } 2003END { flush }
1527 2004
15281; 20051;
1529 2006
2007=head1 EVENT LOOP INTEGRATION
2008
2009It is recommended to use L<AnyEvent::AIO> to integrate IO::AIO
2010automatically into many event loops:
2011
2012 # AnyEvent integration (EV, Event, Glib, Tk, POE, urxvt, pureperl...)
2013 use AnyEvent::AIO;
2014
2015You can also integrate IO::AIO manually into many event loops, here are
2016some examples of how to do this:
2017
2018 # EV integration
2019 my $aio_w = EV::io IO::AIO::poll_fileno, EV::READ, \&IO::AIO::poll_cb;
2020
2021 # Event integration
2022 Event->io (fd => IO::AIO::poll_fileno,
2023 poll => 'r',
2024 cb => \&IO::AIO::poll_cb);
2025
2026 # Glib/Gtk2 integration
2027 add_watch Glib::IO IO::AIO::poll_fileno,
2028 in => sub { IO::AIO::poll_cb; 1 };
2029
2030 # Tk integration
2031 Tk::Event::IO->fileevent (IO::AIO::poll_fileno, "",
2032 readable => \&IO::AIO::poll_cb);
2033
2034 # Danga::Socket integration
2035 Danga::Socket->AddOtherFds (IO::AIO::poll_fileno =>
2036 \&IO::AIO::poll_cb);
2037
1530=head2 FORK BEHAVIOUR 2038=head2 FORK BEHAVIOUR
1531 2039
1532This module should do "the right thing" when the process using it forks: 2040Usage of pthreads in a program changes the semantics of fork
2041considerably. Specifically, only async-safe functions can be called after
2042fork. Perl doesn't know about this, so in general, you cannot call fork
2043with defined behaviour in perl if pthreads are involved. IO::AIO uses
2044pthreads, so this applies, but many other extensions and (for inexplicable
2045reasons) perl itself often is linked against pthreads, so this limitation
2046applies to quite a lot of perls.
1533 2047
1534Before the fork, IO::AIO enters a quiescent state where no requests 2048This module no longer tries to fight your OS, or POSIX. That means IO::AIO
1535can be added in other threads and no results will be processed. After 2049only works in the process that loaded it. Forking is fully supported, but
1536the fork the parent simply leaves the quiescent state and continues 2050using IO::AIO in the child is not.
1537request/result processing, while the child frees the request/result queue
1538(so that the requests started before the fork will only be handled in the
1539parent). Threads will be started on demand until the limit set in the
1540parent process has been reached again.
1541 2051
1542In short: the parent will, after a short pause, continue as if fork had 2052You might get around by not I<using> IO::AIO before (or after)
1543not been called, while the child will act as if IO::AIO has not been used 2053forking. You could also try to call the L<IO::AIO::reinit> function in the
1544yet. 2054child:
2055
2056=over 4
2057
2058=item IO::AIO::reinit
2059
2060Abandons all current requests and I/O threads and simply reinitialises all
2061data structures. This is not an operation supported by any standards, but
2062happens to work on GNU/Linux and some newer BSD systems.
2063
2064The only reasonable use for this function is to call it after forking, if
2065C<IO::AIO> was used in the parent. Calling it while IO::AIO is active in
2066the process will result in undefined behaviour. Calling it at any time
2067will also result in any undefined (by POSIX) behaviour.
2068
2069=back
1545 2070
1546=head2 MEMORY USAGE 2071=head2 MEMORY USAGE
1547 2072
1548Per-request usage: 2073Per-request usage:
1549 2074

Diff Legend

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