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 1.119 by root, Sun Dec 2 20:54:33 2007 UTC vs.
Revision 1.179 by root, Sun Jan 10 23:44:01 2010 UTC

… …
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 30
31 # AnyEvent integration (EV, Event, Glib, Tk, urxvt, pureperl...)
32 open my $fh, "<&=" . IO::AIO::poll_fileno or die "$!";
33 my $w = AnyEvent->io (fh => $fh, poll => 'r', cb => sub { IO::AIO::poll_cb });
34
35 # EV integration
36 my $w = EV::io IO::AIO::poll_fileno, EV::READ, \&IO::AIO::poll_cb;
37
38 # Event integration
39 Event->io (fd => IO::AIO::poll_fileno,
40 poll => 'r',
41 cb => \&IO::AIO::poll_cb);
42
43 # Glib/Gtk2 integration
44 add_watch Glib::IO IO::AIO::poll_fileno,
45 in => sub { IO::AIO::poll_cb; 1 };
46
47 # Tk integration
48 Tk::Event::IO->fileevent (IO::AIO::poll_fileno, "",
49 readable => \&IO::AIO::poll_cb);
50
51 # Danga::Socket integration
52 Danga::Socket->AddOtherFds (IO::AIO::poll_fileno =>
53 \&IO::AIO::poll_cb);
54
55=head1 DESCRIPTION 31=head1 DESCRIPTION
56 32
57This module implements asynchronous I/O using whatever means your 33This module implements asynchronous I/O using whatever means your
58operating system supports. 34operating system supports. It is implemented as an interface to C<libeio>
35(L<http://software.schmorp.de/pkg/libeio.html>).
59 36
60Asynchronous means that operations that can normally block your program 37Asynchronous means that operations that can normally block your program
61(e.g. reading from disk) will be done asynchronously: the operation 38(e.g. reading from disk) will be done asynchronously: the operation
62will still block, but you can do something else in the meantime. This 39will still block, but you can do something else in the meantime. This
63is extremely useful for programs that need to stay interactive even 40is extremely useful for programs that need to stay interactive even
… …
67on a RAID volume or over NFS when you do a number of stat operations 44on a RAID volume or over NFS when you do a number of stat operations
68concurrently. 45concurrently.
69 46
70While most of this works on all types of file descriptors (for 47While most of this works on all types of file descriptors (for
71example sockets), using these functions on file descriptors that 48example sockets), using these functions on file descriptors that
72support nonblocking operation (again, sockets, pipes etc.) is very 49support nonblocking operation (again, sockets, pipes etc.) is
73inefficient. Use an event loop for that (such as the L<Event|Event> 50very inefficient. Use an event loop for that (such as the L<EV>
74module): IO::AIO will naturally fit into such an event loop itself. 51module): IO::AIO will naturally fit into such an event loop itself.
75 52
76In this version, a number of threads are started that execute your 53In this version, a number of threads are started that execute your
77requests and signal their completion. You don't need thread support 54requests and signal their completion. You don't need thread support
78in perl, and the threads created by this module will not be visible 55in perl, and the threads created by this module will not be visible
… …
88yourself, always call C<poll_cb> from within the same thread, or never 65yourself, always call C<poll_cb> from within the same thread, or never
89call C<poll_cb> (or other C<aio_> functions) recursively. 66call C<poll_cb> (or other C<aio_> functions) recursively.
90 67
91=head2 EXAMPLE 68=head2 EXAMPLE
92 69
93This is a simple example that uses the Event module and loads 70This is a simple example that uses the EV module and loads
94F</etc/passwd> asynchronously: 71F</etc/passwd> asynchronously:
95 72
96 use Fcntl; 73 use Fcntl;
97 use Event; 74 use EV;
98 use IO::AIO; 75 use IO::AIO;
99 76
100 # register the IO::AIO callback with Event 77 # register the IO::AIO callback with EV
101 Event->io (fd => IO::AIO::poll_fileno, 78 my $aio_w = EV::io IO::AIO::poll_fileno, EV::READ, \&IO::AIO::poll_cb;
102 poll => 'r',
103 cb => \&IO::AIO::poll_cb);
104 79
105 # queue the request to open /etc/passwd 80 # queue the request to open /etc/passwd
106 aio_open "/etc/passwd", O_RDONLY, 0, sub { 81 aio_open "/etc/passwd", O_RDONLY, 0, sub {
107 my $fh = shift 82 my $fh = shift
108 or die "error while opening: $!"; 83 or die "error while opening: $!";
… …
120 95
121 # file contents now in $contents 96 # file contents now in $contents
122 print $contents; 97 print $contents;
123 98
124 # exit event loop and program 99 # exit event loop and program
125 Event::unloop; 100 EV::unloop;
126 }; 101 };
127 }; 102 };
128 103
129 # possibly queue up other requests, or open GUI windows, 104 # possibly queue up other requests, or open GUI windows,
130 # check for sockets etc. etc. 105 # check for sockets etc. etc.
131 106
132 # process events as long as there are some: 107 # process events as long as there are some:
133 Event::loop; 108 EV::loop;
134 109
135=head1 REQUEST ANATOMY AND LIFETIME 110=head1 REQUEST ANATOMY AND LIFETIME
136 111
137Every C<aio_*> function creates a request. which is a C data structure not 112Every C<aio_*> function creates a request. which is a C data structure not
138directly visible to Perl. 113directly visible to Perl.
… …
188 163
189package IO::AIO; 164package IO::AIO;
190 165
191use Carp (); 166use Carp ();
192 167
193no warnings; 168use common::sense;
194use strict 'vars';
195 169
196use base 'Exporter'; 170use base 'Exporter';
197 171
198BEGIN { 172BEGIN {
199 our $VERSION = '2.6'; 173 our $VERSION = '3.6';
200 174
201 our @AIO_REQ = qw(aio_sendfile aio_read aio_write aio_open aio_close aio_stat 175 our @AIO_REQ = qw(aio_sendfile aio_read aio_write aio_open aio_close
202 aio_lstat aio_unlink aio_rmdir aio_readdir aio_scandir aio_symlink 176 aio_stat aio_lstat aio_unlink aio_rmdir aio_readdir aio_readdirx
203 aio_readlink aio_sync aio_fsync aio_fdatasync aio_readahead aio_rename aio_link 177 aio_scandir aio_symlink aio_readlink aio_sync aio_fsync
178 aio_fdatasync aio_sync_file_range aio_pathsync aio_readahead
179 aio_rename aio_link aio_move aio_copy aio_group
204 aio_move aio_copy aio_group aio_nop aio_mknod aio_load aio_rmtree aio_mkdir 180 aio_nop aio_mknod aio_load aio_rmtree aio_mkdir aio_chown
205 aio_chown aio_chmod aio_utime aio_truncate); 181 aio_chmod aio_utime aio_truncate
182 aio_msync aio_mtouch aio_statvfs);
183
206 our @EXPORT = (@AIO_REQ, qw(aioreq_pri aioreq_nice aio_block)); 184 our @EXPORT = (@AIO_REQ, qw(aioreq_pri aioreq_nice));
207 our @EXPORT_OK = qw(poll_fileno poll_cb poll_wait flush 185 our @EXPORT_OK = qw(poll_fileno poll_cb poll_wait flush
208 min_parallel max_parallel max_idle 186 min_parallel max_parallel max_idle
209 nreqs nready npending nthreads 187 nreqs nready npending nthreads
210 max_poll_time max_poll_reqs); 188 max_poll_time max_poll_reqs
189 sendfile fadvise);
190
191 push @AIO_REQ, qw(aio_busy); # not exported
211 192
212 @IO::AIO::GRP::ISA = 'IO::AIO::REQ'; 193 @IO::AIO::GRP::ISA = 'IO::AIO::REQ';
213 194
214 require XSLoader; 195 require XSLoader;
215 XSLoader::load ("IO::AIO", $VERSION); 196 XSLoader::load ("IO::AIO", $VERSION);
216} 197}
217 198
218=head1 FUNCTIONS 199=head1 FUNCTIONS
200
201=head2 QUICK OVERVIEW
202
203This section simply lists the prototypes of the most important functions
204for quick reference. See the following sections for function-by-function
205documentation.
206
207 aio_open $pathname, $flags, $mode, $callback->($fh)
208 aio_close $fh, $callback->($status)
209 aio_read $fh,$offset,$length, $data,$dataoffset, $callback->($retval)
210 aio_write $fh,$offset,$length, $data,$dataoffset, $callback->($retval)
211 aio_sendfile $out_fh, $in_fh, $in_offset, $length, $callback->($retval)
212 aio_readahead $fh,$offset,$length, $callback->($retval)
213 aio_stat $fh_or_path, $callback->($status)
214 aio_lstat $fh, $callback->($status)
215 aio_statvfs $fh_or_path, $callback->($statvfs)
216 aio_utime $fh_or_path, $atime, $mtime, $callback->($status)
217 aio_chown $fh_or_path, $uid, $gid, $callback->($status)
218 aio_truncate $fh_or_path, $offset, $callback->($status)
219 aio_chmod $fh_or_path, $mode, $callback->($status)
220 aio_unlink $pathname, $callback->($status)
221 aio_mknod $path, $mode, $dev, $callback->($status)
222 aio_link $srcpath, $dstpath, $callback->($status)
223 aio_symlink $srcpath, $dstpath, $callback->($status)
224 aio_readlink $path, $callback->($link)
225 aio_rename $srcpath, $dstpath, $callback->($status)
226 aio_mkdir $pathname, $mode, $callback->($status)
227 aio_rmdir $pathname, $callback->($status)
228 aio_readdir $pathname, $callback->($entries)
229 aio_readdirx $pathname, $flags, $callback->($entries, $flags)
230 IO::AIO::READDIR_DENTS IO::AIO::READDIR_DIRS_FIRST
231 IO::AIO::READDIR_STAT_ORDER IO::AIO::READDIR_FOUND_UNKNOWN
232 aio_load $path, $data, $callback->($status)
233 aio_copy $srcpath, $dstpath, $callback->($status)
234 aio_move $srcpath, $dstpath, $callback->($status)
235 aio_scandir $path, $maxreq, $callback->($dirs, $nondirs)
236 aio_rmtree $path, $callback->($status)
237 aio_sync $callback->($status)
238 aio_fsync $fh, $callback->($status)
239 aio_fdatasync $fh, $callback->($status)
240 aio_sync_file_range $fh, $offset, $nbytes, $flags, $callback->($status)
241 aio_pathsync $path, $callback->($status)
242 aio_msync $scalar, $offset = 0, $length = undef, flags = 0, $callback->($status)
243 aio_mtouch $scalar, $offset = 0, $length = undef, flags = 0, $callback->($status)
244 aio_group $callback->(...)
245 aio_nop $callback->()
246
247 $prev_pri = aioreq_pri [$pri]
248 aioreq_nice $pri_adjust
249
250 IO::AIO::poll_wait
251 IO::AIO::poll_cb
252 IO::AIO::poll
253 IO::AIO::flush
254 IO::AIO::max_poll_reqs $nreqs
255 IO::AIO::max_poll_time $seconds
256 IO::AIO::min_parallel $nthreads
257 IO::AIO::max_parallel $nthreads
258 IO::AIO::max_idle $nthreads
259 IO::AIO::max_outstanding $maxreqs
260 IO::AIO::nreqs
261 IO::AIO::nready
262 IO::AIO::npending
263
264 IO::AIO::sendfile $ofh, $ifh, $offset, $count
265 IO::AIO::fadvise $fh, $offset, $len, $advice
266 IO::AIO::mlockall $flags
267 IO::AIO::munlockall
219 268
220=head2 AIO REQUEST FUNCTIONS 269=head2 AIO REQUEST FUNCTIONS
221 270
222All the C<aio_*> calls are more or less thin wrappers around the syscall 271All the C<aio_*> calls are more or less thin wrappers around the syscall
223with the same name (sans C<aio_>). The arguments are similar or identical, 272with the same name (sans C<aio_>). The arguments are similar or identical,
224and they all accept an additional (and optional) C<$callback> argument 273and they all accept an additional (and optional) C<$callback> argument
225which must be a code reference. This code reference will get called with 274which must be a code reference. This code reference will get called with
226the syscall return code (e.g. most syscalls return C<-1> on error, unlike 275the syscall return code (e.g. most syscalls return C<-1> on error, unlike
227perl, which usually delivers "false") as it's sole argument when the given 276perl, which usually delivers "false") as its sole argument after the given
228syscall has been executed asynchronously. 277syscall has been executed asynchronously.
229 278
230All functions expecting a filehandle keep a copy of the filehandle 279All functions expecting a filehandle keep a copy of the filehandle
231internally until the request has finished. 280internally until the request has finished.
232 281
… …
246your pathnames to the locale (or other) encoding in effect in the user 295your pathnames to the locale (or other) encoding in effect in the user
247environment, d) use Glib::filename_from_unicode on unicode filenames or e) 296environment, d) use Glib::filename_from_unicode on unicode filenames or e)
248use something else to ensure your scalar has the correct contents. 297use something else to ensure your scalar has the correct contents.
249 298
250This works, btw. independent of the internal UTF-8 bit, which IO::AIO 299This works, btw. independent of the internal UTF-8 bit, which IO::AIO
251handles correctly wether it is set or not. 300handles correctly whether it is set or not.
252 301
253=over 4 302=over 4
254 303
255=item $prev_pri = aioreq_pri [$pri] 304=item $prev_pri = aioreq_pri [$pri]
256 305
… …
319 368
320Asynchronously close a file and call the callback with the result 369Asynchronously close a file and call the callback with the result
321code. 370code.
322 371
323Unfortunately, you can't do this to perl. Perl I<insists> very strongly on 372Unfortunately, you can't do this to perl. Perl I<insists> very strongly on
324closing the file descriptor associated with the filehandle itself. Here is 373closing the file descriptor associated with the filehandle itself.
325what aio_close will try:
326 374
327 1. dup()licate the fd 375Therefore, C<aio_close> will not close the filehandle - instead it will
328 2. asynchronously close() the duplicated fd 376use dup2 to overwrite the file descriptor with the write-end of a pipe
329 3. dup()licate the fd once more 377(the pipe fd will be created on demand and will be cached).
330 4. let perl close() the filehandle
331 5. asynchronously close the duplicated fd
332 378
333The idea is that the first close() flushes stuff to disk that closing an 379Or in other words: the file descriptor will be closed, but it will not be
334fd will flush, so when perl closes the fd, nothing much will need to be 380free for reuse until the perl filehandle is closed.
335flushed. The second async. close() will then flush stuff to disk that
336closing the last fd to the file will flush.
337
338Just FYI, SuSv3 has this to say on close:
339
340 All outstanding record locks owned by the process on the file
341 associated with the file descriptor shall be removed.
342
343 If fildes refers to a socket, close() shall cause the socket to be
344 destroyed. ... close() shall block for up to the current linger
345 interval until all data is transmitted.
346 [this actually sounds like a specification bug, but who knows]
347
348And at least Linux additionally actually flushes stuff on every close,
349even when the file itself is still open.
350
351Sounds enourmously inefficient and complicated? Yes... please show me how
352to nuke perl's fd out of existence...
353 381
354=cut 382=cut
355 383
356sub aio_close($;$) {
357 aio_block {
358 my ($fh, $cb) = @_;
359
360 my $pri = aioreq_pri;
361 my $grp = aio_group $cb;
362
363 my $fd = fileno $fh;
364
365 defined $fd or Carp::croak "aio_close called with fd-less filehandle";
366
367 # if the dups fail we will simply get EBADF
368 my $fd2 = _dup $fd;
369 aioreq_pri $pri;
370 add $grp _aio_close $fd2, sub {
371 my $fd2 = _dup $fd;
372 close $fh;
373 aioreq_pri $pri;
374 add $grp _aio_close $fd2, sub {
375 $grp->result ($_[0]);
376 };
377 };
378
379 $grp
380 }
381}
382
383
384=item aio_read $fh,$offset,$length, $data,$dataoffset, $callback->($retval) 384=item aio_read $fh,$offset,$length, $data,$dataoffset, $callback->($retval)
385 385
386=item aio_write $fh,$offset,$length, $data,$dataoffset, $callback->($retval) 386=item aio_write $fh,$offset,$length, $data,$dataoffset, $callback->($retval)
387 387
388Reads or writes C<$length> bytes from the specified C<$fh> and C<$offset> 388Reads or writes C<$length> bytes from or to the specified C<$fh> and
389into the scalar given by C<$data> and offset C<$dataoffset> and calls the 389C<$offset> into the scalar given by C<$data> and offset C<$dataoffset>
390callback without the actual number of bytes read (or -1 on error, just 390and calls the callback without the actual number of bytes read (or -1 on
391like the syscall). 391error, just like the syscall).
392
393C<aio_read> will, like C<sysread>, shrink or grow the C<$data> scalar to
394offset plus the actual number of bytes read.
392 395
393If C<$offset> is undefined, then the current file descriptor offset will 396If C<$offset> is undefined, then the current file descriptor offset will
394be used (and updated), otherwise the file descriptor offset will not be 397be used (and updated), otherwise the file descriptor offset will not be
395changed by these calls. 398changed by these calls.
396 399
397If C<$length> is undefined in C<aio_write>, use the remaining length of C<$data>. 400If C<$length> is undefined in C<aio_write>, use the remaining length of
401C<$data>.
398 402
399If C<$dataoffset> is less than zero, it will be counted from the end of 403If C<$dataoffset> is less than zero, it will be counted from the end of
400C<$data>. 404C<$data>.
401 405
402The C<$data> scalar I<MUST NOT> be modified in any way while the request 406The C<$data> scalar I<MUST NOT> be modified in any way while the request
… …
420than one C<aio_sendfile> per C<$out_fh>, as they will interfere with each 424than one C<aio_sendfile> per C<$out_fh>, as they will interfere with each
421other. 425other.
422 426
423This call tries to make use of a native C<sendfile> syscall to provide 427This call tries to make use of a native C<sendfile> syscall to provide
424zero-copy operation. For this to work, C<$out_fh> should refer to a 428zero-copy operation. For this to work, C<$out_fh> should refer to a
425socket, and C<$in_fh> should refer to mmap'able file. 429socket, and C<$in_fh> should refer to an mmap'able file.
426 430
427If the native sendfile call fails or is not implemented, it will be 431If a native sendfile cannot be found or it fails with C<ENOSYS>,
432C<ENOTSUP>, C<EOPNOTSUPP>, C<EAFNOSUPPORT>, C<EPROTOTYPE> or C<ENOTSOCK>,
428emulated, so you can call C<aio_sendfile> on any type of filehandle 433it will be emulated, so you can call C<aio_sendfile> on any type of
429regardless of the limitations of the operating system. 434filehandle regardless of the limitations of the operating system.
430 435
431Please note, however, that C<aio_sendfile> can read more bytes from 436Please note, however, that C<aio_sendfile> can read more bytes from
432C<$in_fh> than are written, and there is no way to find out how many 437C<$in_fh> than are written, and there is no way to find out how many
433bytes have been read from C<aio_sendfile> alone, as C<aio_sendfile> only 438bytes have been read from C<aio_sendfile> alone, as C<aio_sendfile> only
434provides the number of bytes written to C<$out_fh>. Only if the result 439provides the number of bytes written to C<$out_fh>. Only if the result
… …
472 $_[0] and die "stat failed: $!"; 477 $_[0] and die "stat failed: $!";
473 print "size is ", -s _, "\n"; 478 print "size is ", -s _, "\n";
474 }; 479 };
475 480
476 481
482=item aio_statvfs $fh_or_path, $callback->($statvfs)
483
484Works like the POSIX C<statvfs> or C<fstatvfs> syscalls, depending on
485whether a file handle or path was passed.
486
487On success, the callback is passed a hash reference with the following
488members: C<bsize>, C<frsize>, C<blocks>, C<bfree>, C<bavail>, C<files>,
489C<ffree>, C<favail>, C<fsid>, C<flag> and C<namemax>. On failure, C<undef>
490is passed.
491
492The following POSIX IO::AIO::ST_* constants are defined: C<ST_RDONLY> and
493C<ST_NOSUID>.
494
495The following non-POSIX IO::AIO::ST_* flag masks are defined to
496their correct value when available, or to C<0> on systems that do
497not support them: C<ST_NODEV>, C<ST_NOEXEC>, C<ST_SYNCHRONOUS>,
498C<ST_MANDLOCK>, C<ST_WRITE>, C<ST_APPEND>, C<ST_IMMUTABLE>, C<ST_NOATIME>,
499C<ST_NODIRATIME> and C<ST_RELATIME>.
500
501Example: stat C</wd> and dump out the data if successful.
502
503 aio_statvfs "/wd", sub {
504 my $f = $_[0]
505 or die "statvfs: $!";
506
507 use Data::Dumper;
508 say Dumper $f;
509 };
510
511 # result:
512 {
513 bsize => 1024,
514 bfree => 4333064312,
515 blocks => 10253828096,
516 files => 2050765568,
517 flag => 4096,
518 favail => 2042092649,
519 bavail => 4333064312,
520 ffree => 2042092649,
521 namemax => 255,
522 frsize => 1024,
523 fsid => 1810
524 }
525
526
477=item aio_utime $fh_or_path, $atime, $mtime, $callback->($status) 527=item aio_utime $fh_or_path, $atime, $mtime, $callback->($status)
478 528
479Works like perl's C<utime> function (including the special case of $atime 529Works like perl's C<utime> function (including the special case of $atime
480and $mtime being undef). Fractional times are supported if the underlying 530and $mtime being undef). Fractional times are supported if the underlying
481syscalls support them. 531syscalls support them.
… …
574 624
575Unlike the POSIX call of the same name, C<aio_readdir> reads an entire 625Unlike the POSIX call of the same name, C<aio_readdir> reads an entire
576directory (i.e. opendir + readdir + closedir). The entries will not be 626directory (i.e. opendir + readdir + closedir). The entries will not be
577sorted, and will B<NOT> include the C<.> and C<..> entries. 627sorted, and will B<NOT> include the C<.> and C<..> entries.
578 628
579The callback a single argument which is either C<undef> or an array-ref 629The callback is passed a single argument which is either C<undef> or an
580with the filenames. 630array-ref with the filenames.
631
632
633=item aio_readdirx $pathname, $flags, $callback->($entries, $flags)
634
635Quite similar to C<aio_readdir>, but the C<$flags> argument allows to tune
636behaviour and output format. In case of an error, C<$entries> will be
637C<undef>.
638
639The flags are a combination of the following constants, ORed together (the
640flags will also be passed to the callback, possibly modified):
641
642=over 4
643
644=item IO::AIO::READDIR_DENTS
645
646When this flag is off, then the callback gets an arrayref with of names
647only (as with C<aio_readdir>), otherwise it gets an arrayref with
648C<[$name, $type, $inode]> arrayrefs, each describing a single directory
649entry in more detail.
650
651C<$name> is the name of the entry.
652
653C<$type> is one of the C<IO::AIO::DT_xxx> constants:
654
655C<IO::AIO::DT_UNKNOWN>, C<IO::AIO::DT_FIFO>, C<IO::AIO::DT_CHR>, C<IO::AIO::DT_DIR>,
656C<IO::AIO::DT_BLK>, C<IO::AIO::DT_REG>, C<IO::AIO::DT_LNK>, C<IO::AIO::DT_SOCK>,
657C<IO::AIO::DT_WHT>.
658
659C<IO::AIO::DT_UNKNOWN> means just that: readdir does not know. If you need to
660know, you have to run stat yourself. Also, for speed reasons, the C<$type>
661scalars are read-only: you can not modify them.
662
663C<$inode> is the inode number (which might not be exact on systems with 64
664bit inode numbers and 32 bit perls). This field has unspecified content on
665systems that do not deliver the inode information.
666
667=item IO::AIO::READDIR_DIRS_FIRST
668
669When this flag is set, then the names will be returned in an order where
670likely directories come first. This is useful when you need to quickly
671find directories, or you want to find all directories while avoiding to
672stat() each entry.
673
674If the system returns type information in readdir, then this is used
675to find directories directly. Otherwise, likely directories are files
676beginning with ".", or otherwise files with no dots, of which files with
677short names are tried first.
678
679=item IO::AIO::READDIR_STAT_ORDER
680
681When this flag is set, then the names will be returned in an order
682suitable for stat()'ing each one. That is, when you plan to stat()
683all files in the given directory, then the returned order will likely
684be fastest.
685
686If both this flag and C<IO::AIO::READDIR_DIRS_FIRST> are specified, then
687the likely dirs come first, resulting in a less optimal stat order.
688
689=item IO::AIO::READDIR_FOUND_UNKNOWN
690
691This flag should not be set when calling C<aio_readdirx>. Instead, it
692is being set by C<aio_readdirx>, when any of the C<$type>'s found were
693C<IO::AIO::DT_UNKNOWN>. The absense of this flag therefore indicates that all
694C<$type>'s are known, which can be used to speed up some algorithms.
695
696=back
581 697
582 698
583=item aio_load $path, $data, $callback->($status) 699=item aio_load $path, $data, $callback->($status)
584 700
585This is a composite request that tries to fully load the given file into 701This is a composite request that tries to fully load the given file into
586memory. Status is the same as with aio_read. 702memory. Status is the same as with aio_read.
587 703
588=cut 704=cut
589 705
590sub aio_load($$;$) { 706sub aio_load($$;$) {
591 aio_block {
592 my ($path, undef, $cb) = @_; 707 my ($path, undef, $cb) = @_;
593 my $data = \$_[1]; 708 my $data = \$_[1];
594 709
595 my $pri = aioreq_pri; 710 my $pri = aioreq_pri;
596 my $grp = aio_group $cb; 711 my $grp = aio_group $cb;
712
713 aioreq_pri $pri;
714 add $grp aio_open $path, O_RDONLY, 0, sub {
715 my $fh = shift
716 or return $grp->result (-1);
597 717
598 aioreq_pri $pri; 718 aioreq_pri $pri;
599 add $grp aio_open $path, O_RDONLY, 0, sub {
600 my $fh = shift
601 or return $grp->result (-1);
602
603 aioreq_pri $pri;
604 add $grp aio_read $fh, 0, (-s $fh), $$data, 0, sub { 719 add $grp aio_read $fh, 0, (-s $fh), $$data, 0, sub {
605 $grp->result ($_[0]); 720 $grp->result ($_[0]);
606 };
607 }; 721 };
608
609 $grp
610 } 722 };
723
724 $grp
611} 725}
612 726
613=item aio_copy $srcpath, $dstpath, $callback->($status) 727=item aio_copy $srcpath, $dstpath, $callback->($status)
614 728
615Try to copy the I<file> (directories not supported as either source or 729Try to copy the I<file> (directories not supported as either source or
616destination) from C<$srcpath> to C<$dstpath> and call the callback with 730destination) from C<$srcpath> to C<$dstpath> and call the callback with
617the C<0> (error) or C<-1> ok. 731a status of C<0> (ok) or C<-1> (error, see C<$!>).
618 732
619This is a composite request that it creates the destination file with 733This is a composite request that creates the destination file with
620mode 0200 and copies the contents of the source file into it using 734mode 0200 and copies the contents of the source file into it using
621C<aio_sendfile>, followed by restoring atime, mtime, access mode and 735C<aio_sendfile>, followed by restoring atime, mtime, access mode and
622uid/gid, in that order. 736uid/gid, in that order.
623 737
624If an error occurs, the partial destination file will be unlinked, if 738If an error occurs, the partial destination file will be unlinked, if
… …
626errors are being ignored. 740errors are being ignored.
627 741
628=cut 742=cut
629 743
630sub aio_copy($$;$) { 744sub aio_copy($$;$) {
631 aio_block {
632 my ($src, $dst, $cb) = @_; 745 my ($src, $dst, $cb) = @_;
633 746
634 my $pri = aioreq_pri; 747 my $pri = aioreq_pri;
635 my $grp = aio_group $cb; 748 my $grp = aio_group $cb;
636 749
637 aioreq_pri $pri; 750 aioreq_pri $pri;
638 add $grp aio_open $src, O_RDONLY, 0, sub { 751 add $grp aio_open $src, O_RDONLY, 0, sub {
639 if (my $src_fh = $_[0]) { 752 if (my $src_fh = $_[0]) {
640 my @stat = stat $src_fh; 753 my @stat = stat $src_fh; # hmm, might block over nfs?
641 754
642 aioreq_pri $pri; 755 aioreq_pri $pri;
643 add $grp aio_open $dst, O_CREAT | O_WRONLY | O_TRUNC, 0200, sub { 756 add $grp aio_open $dst, O_CREAT | O_WRONLY | O_TRUNC, 0200, sub {
644 if (my $dst_fh = $_[0]) { 757 if (my $dst_fh = $_[0]) {
645 aioreq_pri $pri; 758 aioreq_pri $pri;
646 add $grp aio_sendfile $dst_fh, $src_fh, 0, $stat[7], sub { 759 add $grp aio_sendfile $dst_fh, $src_fh, 0, $stat[7], sub {
647 if ($_[0] == $stat[7]) { 760 if ($_[0] == $stat[7]) {
648 $grp->result (0); 761 $grp->result (0);
649 close $src_fh; 762 close $src_fh;
650 763
651 # those should not normally block. should. should.
652 utime $stat[8], $stat[9], $dst;
653 chmod $stat[2] & 07777, $dst_fh;
654 chown $stat[4], $stat[5], $dst_fh;
655 close $dst_fh;
656 } else { 764 my $ch = sub {
657 $grp->result (-1);
658 close $src_fh;
659 close $dst_fh;
660
661 aioreq $pri; 765 aioreq_pri $pri;
766 add $grp aio_chmod $dst_fh, $stat[2] & 07777, sub {
767 aioreq_pri $pri;
768 add $grp aio_chown $dst_fh, $stat[4], $stat[5], sub {
769 aioreq_pri $pri;
662 add $grp aio_unlink $dst; 770 add $grp aio_close $dst_fh;
771 }
772 };
663 } 773 };
774
775 aioreq_pri $pri;
776 add $grp aio_utime $dst_fh, $stat[8], $stat[9], sub {
777 if ($_[0] < 0 && $! == ENOSYS) {
778 aioreq_pri $pri;
779 add $grp aio_utime $dst, $stat[8], $stat[9], $ch;
780 } else {
781 $ch->();
782 }
783 };
784 } else {
785 $grp->result (-1);
786 close $src_fh;
787 close $dst_fh;
788
789 aioreq $pri;
790 add $grp aio_unlink $dst;
664 }; 791 }
665 } else {
666 $grp->result (-1);
667 } 792 };
793 } else {
794 $grp->result (-1);
668 }, 795 }
669
670 } else {
671 $grp->result (-1);
672 } 796 },
797
798 } else {
799 $grp->result (-1);
673 }; 800 }
674
675 $grp
676 } 801 };
802
803 $grp
677} 804}
678 805
679=item aio_move $srcpath, $dstpath, $callback->($status) 806=item aio_move $srcpath, $dstpath, $callback->($status)
680 807
681Try to move the I<file> (directories not supported as either source or 808Try to move the I<file> (directories not supported as either source or
682destination) from C<$srcpath> to C<$dstpath> and call the callback with 809destination) from C<$srcpath> to C<$dstpath> and call the callback with
683the C<0> (error) or C<-1> ok. 810a status of C<0> (ok) or C<-1> (error, see C<$!>).
684 811
685This is a composite request that tries to rename(2) the file first. If 812This is a composite request that tries to rename(2) the file first; if
686rename files with C<EXDEV>, it copies the file with C<aio_copy> and, if 813rename fails with C<EXDEV>, it copies the file with C<aio_copy> and, if
687that is successful, unlinking the C<$srcpath>. 814that is successful, unlinks the C<$srcpath>.
688 815
689=cut 816=cut
690 817
691sub aio_move($$;$) { 818sub aio_move($$;$) {
692 aio_block {
693 my ($src, $dst, $cb) = @_; 819 my ($src, $dst, $cb) = @_;
694 820
695 my $pri = aioreq_pri; 821 my $pri = aioreq_pri;
696 my $grp = aio_group $cb; 822 my $grp = aio_group $cb;
697 823
698 aioreq_pri $pri; 824 aioreq_pri $pri;
699 add $grp aio_rename $src, $dst, sub { 825 add $grp aio_rename $src, $dst, sub {
700 if ($_[0] && $! == EXDEV) { 826 if ($_[0] && $! == EXDEV) {
701 aioreq_pri $pri; 827 aioreq_pri $pri;
702 add $grp aio_copy $src, $dst, sub { 828 add $grp aio_copy $src, $dst, sub {
703 $grp->result ($_[0]);
704
705 if (!$_[0]) {
706 aioreq_pri $pri;
707 add $grp aio_unlink $src;
708 }
709 };
710 } else {
711 $grp->result ($_[0]); 829 $grp->result ($_[0]);
830
831 if (!$_[0]) {
832 aioreq_pri $pri;
833 add $grp aio_unlink $src;
834 }
712 } 835 };
836 } else {
837 $grp->result ($_[0]);
713 }; 838 }
714
715 $grp
716 } 839 };
840
841 $grp
717} 842}
718 843
719=item aio_scandir $path, $maxreq, $callback->($dirs, $nondirs) 844=item aio_scandir $path, $maxreq, $callback->($dirs, $nondirs)
720 845
721Scans a directory (similar to C<aio_readdir>) but additionally tries to 846Scans a directory (similar to C<aio_readdir>) but additionally tries to
… …
741 866
742Implementation notes. 867Implementation notes.
743 868
744The C<aio_readdir> cannot be avoided, but C<stat()>'ing every entry can. 869The C<aio_readdir> cannot be avoided, but C<stat()>'ing every entry can.
745 870
871If readdir returns file type information, then this is used directly to
872find directories.
873
746After reading the directory, the modification time, size etc. of the 874Otherwise, after reading the directory, the modification time, size etc.
747directory before and after the readdir is checked, and if they match (and 875of the directory before and after the readdir is checked, and if they
748isn't the current time), the link count will be used to decide how many 876match (and isn't the current time), the link count will be used to decide
749entries are directories (if >= 2). Otherwise, no knowledge of the number 877how many entries are directories (if >= 2). Otherwise, no knowledge of the
750of subdirectories will be assumed. 878number of subdirectories will be assumed.
751 879
752Then entries will be sorted into likely directories (everything without 880Then entries will be sorted into likely directories a non-initial dot
753a non-initial dot currently) and likely non-directories (everything 881currently) and likely non-directories (see C<aio_readdirx>). Then every
754else). Then every entry plus an appended C</.> will be C<stat>'ed, 882entry plus an appended C</.> will be C<stat>'ed, likely directories first,
755likely directories first. If that succeeds, it assumes that the entry 883in order of their inode numbers. If that succeeds, it assumes that the
756is a directory or a symlink to directory (which will be checked 884entry is a directory or a symlink to directory (which will be checked
757seperately). This is often faster than stat'ing the entry itself because 885seperately). This is often faster than stat'ing the entry itself because
758filesystems might detect the type of the entry without reading the inode 886filesystems might detect the type of the entry without reading the inode
759data (e.g. ext2fs filetype feature). 887data (e.g. ext2fs filetype feature), even on systems that cannot return
888the filetype information on readdir.
760 889
761If the known number of directories (link count - 2) has been reached, the 890If the known number of directories (link count - 2) has been reached, the
762rest of the entries is assumed to be non-directories. 891rest of the entries is assumed to be non-directories.
763 892
764This only works with certainty on POSIX (= UNIX) filesystems, which 893This only works with certainty on POSIX (= UNIX) filesystems, which
… …
769directory counting heuristic. 898directory counting heuristic.
770 899
771=cut 900=cut
772 901
773sub aio_scandir($$;$) { 902sub aio_scandir($$;$) {
774 aio_block {
775 my ($path, $maxreq, $cb) = @_; 903 my ($path, $maxreq, $cb) = @_;
776 904
777 my $pri = aioreq_pri; 905 my $pri = aioreq_pri;
778 906
779 my $grp = aio_group $cb; 907 my $grp = aio_group $cb;
780 908
781 $maxreq = 4 if $maxreq <= 0; 909 $maxreq = 4 if $maxreq <= 0;
782 910
783 # stat once 911 # stat once
912 aioreq_pri $pri;
913 add $grp aio_stat $path, sub {
914 return $grp->result () if $_[0];
915 my $now = time;
916 my $hash1 = join ":", (stat _)[0,1,3,7,9];
917
918 # read the directory entries
784 aioreq_pri $pri; 919 aioreq_pri $pri;
785 add $grp aio_stat $path, sub { 920 add $grp aio_readdirx $path, READDIR_DIRS_FIRST, sub {
921 my $entries = shift
786 return $grp->result () if $_[0]; 922 or return $grp->result ();
787 my $now = time;
788 my $hash1 = join ":", (stat _)[0,1,3,7,9];
789 923
790 # read the directory entries 924 # stat the dir another time
791 aioreq_pri $pri; 925 aioreq_pri $pri;
792 add $grp aio_readdir $path, sub {
793 my $entries = shift
794 or return $grp->result ();
795
796 # stat the dir another time
797 aioreq_pri $pri;
798 add $grp aio_stat $path, sub { 926 add $grp aio_stat $path, sub {
799 my $hash2 = join ":", (stat _)[0,1,3,7,9]; 927 my $hash2 = join ":", (stat _)[0,1,3,7,9];
800 928
801 my $ndirs; 929 my $ndirs;
802 930
803 # take the slow route if anything looks fishy 931 # take the slow route if anything looks fishy
804 if ($hash1 ne $hash2 or (stat _)[9] == $now) { 932 if ($hash1 ne $hash2 or (stat _)[9] == $now) {
805 $ndirs = -1; 933 $ndirs = -1;
806 } else { 934 } else {
807 # if nlink == 2, we are finished 935 # if nlink == 2, we are finished
808 # on non-posix-fs's, we rely on nlink < 2 936 # for non-posix-fs's, we rely on nlink < 2
809 $ndirs = (stat _)[3] - 2 937 $ndirs = (stat _)[3] - 2
810 or return $grp->result ([], $entries); 938 or return $grp->result ([], $entries);
811 } 939 }
812 940
813 # sort into likely dirs and likely nondirs
814 # dirs == files without ".", short entries first
815 $entries = [map $_->[0],
816 sort { $b->[1] cmp $a->[1] }
817 map [$_, sprintf "%s%04d", (/.\./ ? "1" : "0"), length],
818 @$entries];
819
820 my (@dirs, @nondirs); 941 my (@dirs, @nondirs);
821 942
822 my $statgrp = add $grp aio_group sub { 943 my $statgrp = add $grp aio_group sub {
823 $grp->result (\@dirs, \@nondirs); 944 $grp->result (\@dirs, \@nondirs);
824 }; 945 };
825 946
826 limit $statgrp $maxreq; 947 limit $statgrp $maxreq;
827 feed $statgrp sub { 948 feed $statgrp sub {
828 return unless @$entries; 949 return unless @$entries;
829 my $entry = pop @$entries; 950 my $entry = shift @$entries;
830 951
831 aioreq_pri $pri; 952 aioreq_pri $pri;
832 add $statgrp aio_stat "$path/$entry/.", sub { 953 add $statgrp aio_stat "$path/$entry/.", sub {
833 if ($_[0] < 0) { 954 if ($_[0] < 0) {
834 push @nondirs, $entry; 955 push @nondirs, $entry;
835 } else { 956 } else {
836 # need to check for real directory 957 # need to check for real directory
837 aioreq_pri $pri; 958 aioreq_pri $pri;
838 add $statgrp aio_lstat "$path/$entry", sub { 959 add $statgrp aio_lstat "$path/$entry", sub {
839 if (-d _) { 960 if (-d _) {
840 push @dirs, $entry; 961 push @dirs, $entry;
841 962
842 unless (--$ndirs) { 963 unless (--$ndirs) {
843 push @nondirs, @$entries; 964 push @nondirs, @$entries;
844 feed $statgrp; 965 feed $statgrp;
845 }
846 } else {
847 push @nondirs, $entry;
848 } 966 }
967 } else {
968 push @nondirs, $entry;
849 } 969 }
850 } 970 }
851 }; 971 }
852 }; 972 };
853 }; 973 };
854 }; 974 };
855 }; 975 };
856
857 $grp
858 } 976 };
977
978 $grp
859} 979}
860 980
861=item aio_rmtree $path, $callback->($status) 981=item aio_rmtree $path, $callback->($status)
862 982
863Delete a directory tree starting (and including) C<$path>, return the 983Delete a directory tree starting (and including) C<$path>, return the
… …
867 987
868=cut 988=cut
869 989
870sub aio_rmtree; 990sub aio_rmtree;
871sub aio_rmtree($;$) { 991sub aio_rmtree($;$) {
872 aio_block {
873 my ($path, $cb) = @_; 992 my ($path, $cb) = @_;
874 993
875 my $pri = aioreq_pri; 994 my $pri = aioreq_pri;
876 my $grp = aio_group $cb; 995 my $grp = aio_group $cb;
877 996
878 aioreq_pri $pri; 997 aioreq_pri $pri;
879 add $grp aio_scandir $path, 0, sub { 998 add $grp aio_scandir $path, 0, sub {
880 my ($dirs, $nondirs) = @_; 999 my ($dirs, $nondirs) = @_;
881 1000
882 my $dirgrp = aio_group sub { 1001 my $dirgrp = aio_group sub {
883 add $grp aio_rmdir $path, sub { 1002 add $grp aio_rmdir $path, sub {
884 $grp->result ($_[0]); 1003 $grp->result ($_[0]);
885 };
886 }; 1004 };
887
888 (aioreq_pri $pri), add $dirgrp aio_rmtree "$path/$_" for @$dirs;
889 (aioreq_pri $pri), add $dirgrp aio_unlink "$path/$_" for @$nondirs;
890
891 add $grp $dirgrp;
892 }; 1005 };
893 1006
894 $grp 1007 (aioreq_pri $pri), add $dirgrp aio_rmtree "$path/$_" for @$dirs;
1008 (aioreq_pri $pri), add $dirgrp aio_unlink "$path/$_" for @$nondirs;
1009
1010 add $grp $dirgrp;
895 } 1011 };
1012
1013 $grp
896} 1014}
897 1015
898=item aio_sync $callback->($status) 1016=item aio_sync $callback->($status)
899 1017
900Asynchronously call sync and call the callback when finished. 1018Asynchronously call sync and call the callback when finished.
… …
909Asynchronously call fdatasync on the given filehandle and call the 1027Asynchronously call fdatasync on the given filehandle and call the
910callback with the fdatasync result code. 1028callback with the fdatasync result code.
911 1029
912If this call isn't available because your OS lacks it or it couldn't be 1030If this call isn't available because your OS lacks it or it couldn't be
913detected, it will be emulated by calling C<fsync> instead. 1031detected, it will be emulated by calling C<fsync> instead.
1032
1033=item aio_sync_file_range $fh, $offset, $nbytes, $flags, $callback->($status)
1034
1035Sync the data portion of the file specified by C<$offset> and C<$length>
1036to disk (but NOT the metadata), by calling the Linux-specific
1037sync_file_range call. If sync_file_range is not available or it returns
1038ENOSYS, then fdatasync or fsync is being substituted.
1039
1040C<$flags> can be a combination of C<IO::AIO::SYNC_FILE_RANGE_WAIT_BEFORE>,
1041C<IO::AIO::SYNC_FILE_RANGE_WRITE> and
1042C<IO::AIO::SYNC_FILE_RANGE_WAIT_AFTER>: refer to the sync_file_range
1043manpage for details.
1044
1045=item aio_pathsync $path, $callback->($status)
1046
1047This request tries to open, fsync and close the given path. This is a
1048composite request intended to sync directories after directory operations
1049(E.g. rename). This might not work on all operating systems or have any
1050specific effect, but usually it makes sure that directory changes get
1051written to disc. It works for anything that can be opened for read-only,
1052not just directories.
1053
1054Future versions of this function might fall back to other methods when
1055C<fsync> on the directory fails (such as calling C<sync>).
1056
1057Passes C<0> when everything went ok, and C<-1> on error.
1058
1059=cut
1060
1061sub aio_pathsync($;$) {
1062 my ($path, $cb) = @_;
1063
1064 my $pri = aioreq_pri;
1065 my $grp = aio_group $cb;
1066
1067 aioreq_pri $pri;
1068 add $grp aio_open $path, O_RDONLY, 0, sub {
1069 my ($fh) = @_;
1070 if ($fh) {
1071 aioreq_pri $pri;
1072 add $grp aio_fsync $fh, sub {
1073 $grp->result ($_[0]);
1074
1075 aioreq_pri $pri;
1076 add $grp aio_close $fh;
1077 };
1078 } else {
1079 $grp->result (-1);
1080 }
1081 };
1082
1083 $grp
1084}
1085
1086=item aio_msync $scalar, $offset = 0, $length = undef, flags = 0, $callback->($status)
1087
1088This is a rather advanced IO::AIO call, which only works on mmap(2)ed
1089scalars (see the C<IO::AIO::mmap> function, although it also works on data
1090scalars managed by the L<Sys::Mmap> or L<Mmap> modules, note that the
1091scalar must only be modified in-place while an aio operation is pending on
1092it).
1093
1094It calls the C<msync> function of your OS, if available, with the memory
1095area starting at C<$offset> in the string and ending C<$length> bytes
1096later. If C<$length> is negative, counts from the end, and if C<$length>
1097is C<undef>, then it goes till the end of the string. The flags can be
1098a combination of C<IO::AIO::MS_ASYNC>, C<IO::AIO::MS_INVALIDATE> and
1099C<IO::AIO::MS_SYNC>.
1100
1101=item aio_mtouch $scalar, $offset = 0, $length = undef, flags = 0, $callback->($status)
1102
1103This is a rather advanced IO::AIO call, which works best on mmap(2)ed
1104scalars.
1105
1106It touches (reads or writes) all memory pages in the specified
1107range inside the scalar. All caveats and parameters are the same
1108as for C<aio_msync>, above, except for flags, which must be either
1109C<0> (which reads all pages and ensures they are instantiated) or
1110C<IO::AIO::MT_MODIFY>, which modifies the memory page s(by reading and
1111writing an octet from it, which dirties the page).
914 1112
915=item aio_group $callback->(...) 1113=item aio_group $callback->(...)
916 1114
917This is a very special aio request: Instead of doing something, it is a 1115This is a very special aio request: Instead of doing something, it is a
918container for other aio requests, which is useful if you want to bundle 1116container for other aio requests, which is useful if you want to bundle
… …
968=item cancel $req 1166=item cancel $req
969 1167
970Cancels the request, if possible. Has the effect of skipping execution 1168Cancels the request, if possible. Has the effect of skipping execution
971when entering the B<execute> state and skipping calling the callback when 1169when entering the B<execute> state and skipping calling the callback when
972entering the the B<result> state, but will leave the request otherwise 1170entering the the B<result> state, but will leave the request otherwise
973untouched. That means that requests that currently execute will not be 1171untouched (with the exception of readdir). That means that requests that
974stopped and resources held by the request will not be freed prematurely. 1172currently execute will not be stopped and resources held by the request
1173will not be freed prematurely.
975 1174
976=item cb $req $callback->(...) 1175=item cb $req $callback->(...)
977 1176
978Replace (or simply set) the callback registered to the request. 1177Replace (or simply set) the callback registered to the request.
979 1178
… …
1030Their lifetime, simplified, looks like this: when they are empty, they 1229Their lifetime, simplified, looks like this: when they are empty, they
1031will finish very quickly. If they contain only requests that are in the 1230will finish very quickly. If they contain only requests that are in the
1032C<done> state, they will also finish. Otherwise they will continue to 1231C<done> state, they will also finish. Otherwise they will continue to
1033exist. 1232exist.
1034 1233
1035That means after creating a group you have some time to add requests. And 1234That means after creating a group you have some time to add requests
1036in the callbacks of those requests, you can add further requests to the 1235(precisely before the callback has been invoked, which is only done within
1037group. And only when all those requests have finished will the the group 1236the C<poll_cb>). And in the callbacks of those requests, you can add
1038itself finish. 1237further requests to the group. And only when all those requests have
1238finished will the the group itself finish.
1039 1239
1040=over 4 1240=over 4
1041 1241
1042=item add $grp ... 1242=item add $grp ...
1043 1243
… …
1052=item $grp->cancel_subs 1252=item $grp->cancel_subs
1053 1253
1054Cancel all subrequests and clears any feeder, but not the group request 1254Cancel all subrequests and clears any feeder, but not the group request
1055itself. Useful when you queued a lot of events but got a result early. 1255itself. Useful when you queued a lot of events but got a result early.
1056 1256
1257The group request will finish normally (you cannot add requests to the
1258group).
1259
1057=item $grp->result (...) 1260=item $grp->result (...)
1058 1261
1059Set the result value(s) that will be passed to the group callback when all 1262Set the result value(s) that will be passed to the group callback when all
1060subrequests have finished and set thre groups errno to the current value 1263subrequests have finished and set the groups errno to the current value
1061of errno (just like calling C<errno> without an error number). By default, 1264of errno (just like calling C<errno> without an error number). By default,
1062no argument will be passed and errno is zero. 1265no argument will be passed and errno is zero.
1063 1266
1064=item $grp->errno ([$errno]) 1267=item $grp->errno ([$errno])
1065 1268
… …
1076=item feed $grp $callback->($grp) 1279=item feed $grp $callback->($grp)
1077 1280
1078Sets a feeder/generator on this group: every group can have an attached 1281Sets a feeder/generator on this group: every group can have an attached
1079generator that generates requests if idle. The idea behind this is that, 1282generator that generates requests if idle. The idea behind this is that,
1080although you could just queue as many requests as you want in a group, 1283although you could just queue as many requests as you want in a group,
1081this might starve other requests for a potentially long time. For 1284this might starve other requests for a potentially long time. For example,
1082example, C<aio_scandir> might generate hundreds of thousands C<aio_stat> 1285C<aio_scandir> might generate hundreds of thousands C<aio_stat> requests,
1083requests, delaying any later requests for a long time. 1286delaying any later requests for a long time.
1084 1287
1085To avoid this, and allow incremental generation of requests, you can 1288To avoid this, and allow incremental generation of requests, you can
1086instead a group and set a feeder on it that generates those requests. The 1289instead a group and set a feeder on it that generates those requests. The
1087feed callback will be called whenever there are few enough (see C<limit>, 1290feed callback will be called whenever there are few enough (see C<limit>,
1088below) requests active in the group itself and is expected to queue more 1291below) requests active in the group itself and is expected to queue more
… …
1092not impose any limits). 1295not impose any limits).
1093 1296
1094If the feed does not queue more requests when called, it will be 1297If the feed does not queue more requests when called, it will be
1095automatically removed from the group. 1298automatically removed from the group.
1096 1299
1097If the feed limit is C<0>, it will be set to C<2> automatically. 1300If the feed limit is C<0> when this method is called, it will be set to
1301C<2> automatically.
1098 1302
1099Example: 1303Example:
1100 1304
1101 # stat all files in @files, but only ever use four aio requests concurrently: 1305 # stat all files in @files, but only ever use four aio requests concurrently:
1102 1306
… …
1114Sets the feeder limit for the group: The feeder will be called whenever 1318Sets the feeder limit for the group: The feeder will be called whenever
1115the group contains less than this many requests. 1319the group contains less than this many requests.
1116 1320
1117Setting the limit to C<0> will pause the feeding process. 1321Setting the limit to C<0> will pause the feeding process.
1118 1322
1323The default value for the limit is C<0>, but note that setting a feeder
1324automatically bumps it up to C<2>.
1325
1119=back 1326=back
1120 1327
1121=head2 SUPPORT FUNCTIONS 1328=head2 SUPPORT FUNCTIONS
1122 1329
1123=head3 EVENT PROCESSING AND EVENT LOOP INTEGRATION 1330=head3 EVENT PROCESSING AND EVENT LOOP INTEGRATION
… …
1125=over 4 1332=over 4
1126 1333
1127=item $fileno = IO::AIO::poll_fileno 1334=item $fileno = IO::AIO::poll_fileno
1128 1335
1129Return the I<request result pipe file descriptor>. This filehandle must be 1336Return the I<request result pipe file descriptor>. This filehandle must be
1130polled for reading by some mechanism outside this module (e.g. Event or 1337polled for reading by some mechanism outside this module (e.g. EV, Glib,
1131select, see below or the SYNOPSIS). If the pipe becomes readable you have 1338select and so on, see below or the SYNOPSIS). If the pipe becomes readable
1132to call C<poll_cb> to check the results. 1339you have to call C<poll_cb> to check the results.
1133 1340
1134See C<poll_cb> for an example. 1341See C<poll_cb> for an example.
1135 1342
1136=item IO::AIO::poll_cb 1343=item IO::AIO::poll_cb
1137 1344
1138Process some outstanding events on the result pipe. You have to call this 1345Process some outstanding events on the result pipe. You have to call this
1139regularly. Returns the number of events processed. Returns immediately 1346regularly. Returns C<0> if all events could be processed, or C<-1> if it
1347returned earlier for whatever reason. Returns immediately when no events
1140when no events are outstanding. The amount of events processed depends on 1348are outstanding. The amount of events processed depends on the settings of
1141the settings of C<IO::AIO::max_poll_req> and C<IO::AIO::max_poll_time>. 1349C<IO::AIO::max_poll_req> and C<IO::AIO::max_poll_time>.
1142 1350
1143If not all requests were processed for whatever reason, the filehandle 1351If not all requests were processed for whatever reason, the filehandle
1144will still be ready when C<poll_cb> returns. 1352will still be ready when C<poll_cb> returns, so normally you don't have to
1353do anything special to have it called later.
1145 1354
1146Example: Install an Event watcher that automatically calls 1355Example: Install an Event watcher that automatically calls
1147IO::AIO::poll_cb with high priority: 1356IO::AIO::poll_cb with high priority (more examples can be found in the
1357SYNOPSIS section, at the top of this document):
1148 1358
1149 Event->io (fd => IO::AIO::poll_fileno, 1359 Event->io (fd => IO::AIO::poll_fileno,
1150 poll => 'r', async => 1, 1360 poll => 'r', async => 1,
1151 cb => \&IO::AIO::poll_cb); 1361 cb => \&IO::AIO::poll_cb);
1362
1363=item IO::AIO::poll_wait
1364
1365If there are any outstanding requests and none of them in the result
1366phase, wait till the result filehandle becomes ready for reading (simply
1367does a C<select> on the filehandle. This is useful if you want to
1368synchronously wait for some requests to finish).
1369
1370See C<nreqs> for an example.
1371
1372=item IO::AIO::poll
1373
1374Waits until some requests have been handled.
1375
1376Returns the number of requests processed, but is otherwise strictly
1377equivalent to:
1378
1379 IO::AIO::poll_wait, IO::AIO::poll_cb
1380
1381=item IO::AIO::flush
1382
1383Wait till all outstanding AIO requests have been handled.
1384
1385Strictly equivalent to:
1386
1387 IO::AIO::poll_wait, IO::AIO::poll_cb
1388 while IO::AIO::nreqs;
1152 1389
1153=item IO::AIO::max_poll_reqs $nreqs 1390=item IO::AIO::max_poll_reqs $nreqs
1154 1391
1155=item IO::AIO::max_poll_time $seconds 1392=item IO::AIO::max_poll_time $seconds
1156 1393
… …
1181 # use a low priority so other tasks have priority 1418 # use a low priority so other tasks have priority
1182 Event->io (fd => IO::AIO::poll_fileno, 1419 Event->io (fd => IO::AIO::poll_fileno,
1183 poll => 'r', nice => 1, 1420 poll => 'r', nice => 1,
1184 cb => &IO::AIO::poll_cb); 1421 cb => &IO::AIO::poll_cb);
1185 1422
1186=item IO::AIO::poll_wait
1187
1188If there are any outstanding requests and none of them in the result
1189phase, wait till the result filehandle becomes ready for reading (simply
1190does a C<select> on the filehandle. This is useful if you want to
1191synchronously wait for some requests to finish).
1192
1193See C<nreqs> for an example.
1194
1195=item IO::AIO::poll
1196
1197Waits until some requests have been handled.
1198
1199Returns the number of requests processed, but is otherwise strictly
1200equivalent to:
1201
1202 IO::AIO::poll_wait, IO::AIO::poll_cb
1203
1204=item IO::AIO::flush
1205
1206Wait till all outstanding AIO requests have been handled.
1207
1208Strictly equivalent to:
1209
1210 IO::AIO::poll_wait, IO::AIO::poll_cb
1211 while IO::AIO::nreqs;
1212
1213=back 1423=back
1214 1424
1215=head3 CONTROLLING THE NUMBER OF THREADS 1425=head3 CONTROLLING THE NUMBER OF THREADS
1216 1426
1217=over 1427=over
… …
1263 1473
1264The default is probably ok in most situations, especially if thread 1474The default is probably ok in most situations, especially if thread
1265creation is fast. If thread creation is very slow on your system you might 1475creation is fast. If thread creation is very slow on your system you might
1266want to use larger values. 1476want to use larger values.
1267 1477
1268=item $oldmaxreqs = IO::AIO::max_outstanding $maxreqs 1478=item IO::AIO::max_outstanding $maxreqs
1269 1479
1270This is a very bad function to use in interactive programs because it 1480This is a very bad function to use in interactive programs because it
1271blocks, and a bad way to reduce concurrency because it is inexact: Better 1481blocks, and a bad way to reduce concurrency because it is inexact: Better
1272use an C<aio_group> together with a feed callback. 1482use an C<aio_group> together with a feed callback.
1273 1483
… …
1278 1488
1279The default value is very large, so there is no practical limit on the 1489The default value is very large, so there is no practical limit on the
1280number of outstanding requests. 1490number of outstanding requests.
1281 1491
1282You can still queue as many requests as you want. Therefore, 1492You can still queue as many requests as you want. Therefore,
1283C<max_oustsanding> is mainly useful in simple scripts (with low values) or 1493C<max_outstanding> is mainly useful in simple scripts (with low values) or
1284as a stop gap to shield against fatal memory overflow (with large values). 1494as a stop gap to shield against fatal memory overflow (with large values).
1285 1495
1286=back 1496=back
1287 1497
1288=head3 STATISTICAL INFORMATION 1498=head3 STATISTICAL INFORMATION
… …
1309Returns the number of requests currently in the pending state (executed, 1519Returns the number of requests currently in the pending state (executed,
1310but not yet processed by poll_cb). 1520but not yet processed by poll_cb).
1311 1521
1312=back 1522=back
1313 1523
1524=head3 MISCELLANEOUS FUNCTIONS
1525
1526IO::AIO implements some functions that might be useful, but are not
1527asynchronous.
1528
1529=over 4
1530
1531=item IO::AIO::sendfile $ofh, $ifh, $offset, $count
1532
1533Calls the C<eio_sendfile_sync> function, which is like C<aio_sendfile>,
1534but is blocking (this makes most sense if you know the input data is
1535likely cached already and the output filehandle is set to non-blocking
1536operations).
1537
1538Returns the number of bytes copied, or C<-1> on error.
1539
1540=item IO::AIO::fadvise $fh, $offset, $len, $advice
1541
1542Simply calls the C<posix_fadvise> function (see it's
1543manpage for details). The following advice constants are
1544avaiable: C<IO::AIO::FADV_NORMAL>, C<IO::AIO::FADV_SEQUENTIAL>,
1545C<IO::AIO::FADV_RANDOM>, C<IO::AIO::FADV_NOREUSE>,
1546C<IO::AIO::FADV_WILLNEED>, C<IO::AIO::FADV_DONTNEED>.
1547
1548On systems that do not implement C<posix_fadvise>, this function returns
1549ENOSYS, otherwise the return value of C<posix_fadvise>.
1550
1551=item IO::AIO::mmap $scalar, $length, $prot, $flags, $fh[, $offset]
1552
1553Memory-maps a file (or anonymous memory range) and attaches it to the
1554given C<$scalar>, which will act like a string scalar.
1555
1556The only operations allowed on the scalar are C<substr>/C<vec> that don't
1557change the string length, and most read-only operations such as copying it
1558or searching it with regexes and so on.
1559
1560Anything else is unsafe and will, at best, result in memory leaks.
1561
1562The memory map associated with the C<$scalar> is automatically removed
1563when the C<$scalar> is destroyed, or when the C<IO::AIO::mmap> or
1564C<IO::AIO::munmap> functions are called.
1565
1566This calls the C<mmap>(2) function internally. See your system's manual
1567page for details on the C<$length>, C<$prot> and C<$flags> parameters.
1568
1569The C<$length> must be larger than zero and smaller than the actual
1570filesize.
1571
1572C<$prot> is a combination of C<IO::AIO::PROT_NONE>, C<IO::AIO::PROT_EXEC>,
1573C<IO::AIO::PROT_READ> and/or C<IO::AIO::PROT_WRITE>,
1574
1575C<$flags> can be a combination of C<IO::AIO::MAP_SHARED> or
1576C<IO::AIO::MAP_PRIVATE>, or a number of system-specific flags (when
1577not available, the are defined as 0): C<IO::AIO::MAP_ANONYMOUS>
1578(which is set to C<MAP_ANON> if your system only provides this
1579constant), C<IO::AIO::MAP_HUGETLB>, C<IO::AIO::MAP_LOCKED>,
1580C<IO::AIO::MAP_NORESERVE>, C<IO::AIO::MAP_POPULATE> or
1581C<IO::AIO::MAP_NONBLOCK>
1582
1583If C<$fh> is C<undef>, then a file descriptor of C<-1> is passed.
1584
1585C<$offset> is the offset from the start of the file - it generally must be
1586a multiple of C<IO::AIO::PAGESIZE> and defaults to C<0>.
1587
1588Example:
1589
1590 use Digest::MD5;
1591 use IO::AIO;
1592
1593 open my $fh, "<verybigfile"
1594 or die "$!";
1595
1596 IO::AIO::mmap my $data, -s $fh, IO::AIO::PROT_READ, IO::AIO::MAP_SHARED, $fh
1597 or die "verybigfile: $!";
1598
1599 my $fast_md5 = md5 $data;
1600
1601=item IO::AIO::munmap $scalar
1602
1603Removes a previous mmap and undefines the C<$scalar>.
1604
1605=item IO::AIO::mlockall $flags
1606
1607Calls the C<mlockall> function with the given C<$flags> (a combination of
1608C<IO::AIO::MCL_CURRENT> and C<IO::AIO::MCL__FUTURE>).
1609
1610On systems that do not implement C<mlockall>, this function returns
1611ENOSYS, otherwise the return value of C<mlockall>.
1612
1613=item IO::AIO::munlockall
1614
1615Calls the C<munlockall> function.
1616
1617On systems that do not implement C<munlockall>, this function returns
1618ENOSYS, otherwise the return value of C<munlockall>.
1619
1620=back
1621
1314=cut 1622=cut
1315 1623
1316min_parallel 8; 1624min_parallel 8;
1317 1625
1318END { flush } 1626END { flush }
1319 1627
13201; 16281;
1629
1630=head1 EVENT LOOP INTEGRATION
1631
1632It is recommended to use L<AnyEvent::AIO> to integrate IO::AIO
1633automatically into many event loops:
1634
1635 # AnyEvent integration (EV, Event, Glib, Tk, POE, urxvt, pureperl...)
1636 use AnyEvent::AIO;
1637
1638You can also integrate IO::AIO manually into many event loops, here are
1639some examples of how to do this:
1640
1641 # EV integration
1642 my $aio_w = EV::io IO::AIO::poll_fileno, EV::READ, \&IO::AIO::poll_cb;
1643
1644 # Event integration
1645 Event->io (fd => IO::AIO::poll_fileno,
1646 poll => 'r',
1647 cb => \&IO::AIO::poll_cb);
1648
1649 # Glib/Gtk2 integration
1650 add_watch Glib::IO IO::AIO::poll_fileno,
1651 in => sub { IO::AIO::poll_cb; 1 };
1652
1653 # Tk integration
1654 Tk::Event::IO->fileevent (IO::AIO::poll_fileno, "",
1655 readable => \&IO::AIO::poll_cb);
1656
1657 # Danga::Socket integration
1658 Danga::Socket->AddOtherFds (IO::AIO::poll_fileno =>
1659 \&IO::AIO::poll_cb);
1321 1660
1322=head2 FORK BEHAVIOUR 1661=head2 FORK BEHAVIOUR
1323 1662
1324This module should do "the right thing" when the process using it forks: 1663This module should do "the right thing" when the process using it forks:
1325 1664
… …
1358 1697
1359Known bugs will be fixed in the next release. 1698Known bugs will be fixed in the next release.
1360 1699
1361=head1 SEE ALSO 1700=head1 SEE ALSO
1362 1701
1363L<Coro::AIO>. 1702L<AnyEvent::AIO> for easy integration into event loops, L<Coro::AIO> for a
1703more natural syntax.
1364 1704
1365=head1 AUTHOR 1705=head1 AUTHOR
1366 1706
1367 Marc Lehmann <schmorp@schmorp.de> 1707 Marc Lehmann <schmorp@schmorp.de>
1368 http://home.schmorp.de/ 1708 http://home.schmorp.de/

Diff Legend

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