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 1.113 by root, Thu Sep 20 14:06:21 2007 UTC vs.
Revision 1.191 by root, Thu May 19 23:57:12 2011 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
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
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 # Event integration
36 Event->io (fd => IO::AIO::poll_fileno,
37 poll => 'r',
38 cb => \&IO::AIO::poll_cb);
39
40 # Glib/Gtk2 integration
41 add_watch Glib::IO IO::AIO::poll_fileno,
42 in => sub { IO::AIO::poll_cb; 1 };
43
44 # Tk integration
45 Tk::Event::IO->fileevent (IO::AIO::poll_fileno, "",
46 readable => \&IO::AIO::poll_cb);
47
48 # Danga::Socket integration
49 Danga::Socket->AddOtherFds (IO::AIO::poll_fileno =>
50 \&IO::AIO::poll_cb);
51
52=head1 DESCRIPTION 31=head1 DESCRIPTION
53 32
54This module implements asynchronous I/O using whatever means your 33This module implements asynchronous I/O using whatever means your
55operating system supports. 34operating system supports. It is implemented as an interface to C<libeio>
35(L<http://software.schmorp.de/pkg/libeio.html>).
56 36
57Asynchronous means that operations that can normally block your program 37Asynchronous means that operations that can normally block your program
58(e.g. reading from disk) will be done asynchronously: the operation 38(e.g. reading from disk) will be done asynchronously: the operation
59will 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
60is extremely useful for programs that need to stay interactive even 40is extremely useful for programs that need to stay interactive even
64on 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
65concurrently. 45concurrently.
66 46
67While most of this works on all types of file descriptors (for 47While most of this works on all types of file descriptors (for
68example sockets), using these functions on file descriptors that 48example sockets), using these functions on file descriptors that
69support nonblocking operation (again, sockets, pipes etc.) is very 49support nonblocking operation (again, sockets, pipes etc.) is
70inefficient. 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>
71module): IO::AIO will naturally fit into such an event loop itself. 51module): IO::AIO will naturally fit into such an event loop itself.
72 52
73In this version, a number of threads are started that execute your 53In this version, a number of threads are started that execute your
74requests and signal their completion. You don't need thread support 54requests and signal their completion. You don't need thread support
75in 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
85yourself, 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
86call C<poll_cb> (or other C<aio_> functions) recursively. 66call C<poll_cb> (or other C<aio_> functions) recursively.
87 67
88=head2 EXAMPLE 68=head2 EXAMPLE
89 69
90This is a simple example that uses the Event module and loads 70This is a simple example that uses the EV module and loads
91F</etc/passwd> asynchronously: 71F</etc/passwd> asynchronously:
92 72
93 use Fcntl; 73 use Fcntl;
94 use Event; 74 use EV;
95 use IO::AIO; 75 use IO::AIO;
96 76
97 # register the IO::AIO callback with Event 77 # register the IO::AIO callback with EV
98 Event->io (fd => IO::AIO::poll_fileno, 78 my $aio_w = EV::io IO::AIO::poll_fileno, EV::READ, \&IO::AIO::poll_cb;
99 poll => 'r',
100 cb => \&IO::AIO::poll_cb);
101 79
102 # queue the request to open /etc/passwd 80 # queue the request to open /etc/passwd
103 aio_open "/etc/passwd", O_RDONLY, 0, sub { 81 aio_open "/etc/passwd", IO::AIO::O_RDONLY, 0, sub {
104 my $fh = shift 82 my $fh = shift
105 or die "error while opening: $!"; 83 or die "error while opening: $!";
106 84
107 # stat'ing filehandles is generally non-blocking 85 # stat'ing filehandles is generally non-blocking
108 my $size = -s $fh; 86 my $size = -s $fh;
117 95
118 # file contents now in $contents 96 # file contents now in $contents
119 print $contents; 97 print $contents;
120 98
121 # exit event loop and program 99 # exit event loop and program
122 Event::unloop; 100 EV::unloop;
123 }; 101 };
124 }; 102 };
125 103
126 # possibly queue up other requests, or open GUI windows, 104 # possibly queue up other requests, or open GUI windows,
127 # check for sockets etc. etc. 105 # check for sockets etc. etc.
128 106
129 # process events as long as there are some: 107 # process events as long as there are some:
130 Event::loop; 108 EV::loop;
131 109
132=head1 REQUEST ANATOMY AND LIFETIME 110=head1 REQUEST ANATOMY AND LIFETIME
133 111
134Every 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
135directly visible to Perl. 113directly visible to Perl.
183 161
184=cut 162=cut
185 163
186package IO::AIO; 164package IO::AIO;
187 165
188no warnings; 166use Carp ();
189use strict 'vars'; 167
168use common::sense;
190 169
191use base 'Exporter'; 170use base 'Exporter';
192 171
193BEGIN { 172BEGIN {
194 our $VERSION = '2.4'; 173 our $VERSION = '3.8';
195 174
196 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
197 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
198 aio_readlink 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
199 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
200 aio_chown aio_chmod aio_utime aio_truncate); 181 aio_chmod aio_utime aio_truncate
182 aio_msync aio_mtouch aio_mlock aio_mlockall
183 aio_statvfs);
184
201 our @EXPORT = (@AIO_REQ, qw(aioreq_pri aioreq_nice aio_block)); 185 our @EXPORT = (@AIO_REQ, qw(aioreq_pri aioreq_nice));
202 our @EXPORT_OK = qw(poll_fileno poll_cb poll_wait flush 186 our @EXPORT_OK = qw(poll_fileno poll_cb poll_wait flush
203 min_parallel max_parallel max_idle 187 min_parallel max_parallel max_idle idle_timeout
204 nreqs nready npending nthreads 188 nreqs nready npending nthreads
205 max_poll_time max_poll_reqs); 189 max_poll_time max_poll_reqs
190 sendfile fadvise madvise
191 mmap munmap munlock munlockall);
192
193 push @AIO_REQ, qw(aio_busy); # not exported
206 194
207 @IO::AIO::GRP::ISA = 'IO::AIO::REQ'; 195 @IO::AIO::GRP::ISA = 'IO::AIO::REQ';
208 196
209 require XSLoader; 197 require XSLoader;
210 XSLoader::load ("IO::AIO", $VERSION); 198 XSLoader::load ("IO::AIO", $VERSION);
211} 199}
212 200
213=head1 FUNCTIONS 201=head1 FUNCTIONS
202
203=head2 QUICK OVERVIEW
204
205This section simply lists the prototypes of the most important functions
206for quick reference. See the following sections for function-by-function
207documentation.
208
209 aio_open $pathname, $flags, $mode, $callback->($fh)
210 aio_close $fh, $callback->($status)
211 aio_read $fh,$offset,$length, $data,$dataoffset, $callback->($retval)
212 aio_write $fh,$offset,$length, $data,$dataoffset, $callback->($retval)
213 aio_sendfile $out_fh, $in_fh, $in_offset, $length, $callback->($retval)
214 aio_readahead $fh,$offset,$length, $callback->($retval)
215 aio_stat $fh_or_path, $callback->($status)
216 aio_lstat $fh, $callback->($status)
217 aio_statvfs $fh_or_path, $callback->($statvfs)
218 aio_utime $fh_or_path, $atime, $mtime, $callback->($status)
219 aio_chown $fh_or_path, $uid, $gid, $callback->($status)
220 aio_truncate $fh_or_path, $offset, $callback->($status)
221 aio_chmod $fh_or_path, $mode, $callback->($status)
222 aio_unlink $pathname, $callback->($status)
223 aio_mknod $path, $mode, $dev, $callback->($status)
224 aio_link $srcpath, $dstpath, $callback->($status)
225 aio_symlink $srcpath, $dstpath, $callback->($status)
226 aio_readlink $path, $callback->($link)
227 aio_rename $srcpath, $dstpath, $callback->($status)
228 aio_mkdir $pathname, $mode, $callback->($status)
229 aio_rmdir $pathname, $callback->($status)
230 aio_readdir $pathname, $callback->($entries)
231 aio_readdirx $pathname, $flags, $callback->($entries, $flags)
232 IO::AIO::READDIR_DENTS IO::AIO::READDIR_DIRS_FIRST
233 IO::AIO::READDIR_STAT_ORDER IO::AIO::READDIR_FOUND_UNKNOWN
234 aio_load $path, $data, $callback->($status)
235 aio_copy $srcpath, $dstpath, $callback->($status)
236 aio_move $srcpath, $dstpath, $callback->($status)
237 aio_scandir $path, $maxreq, $callback->($dirs, $nondirs)
238 aio_rmtree $path, $callback->($status)
239 aio_sync $callback->($status)
240 aio_fsync $fh, $callback->($status)
241 aio_fdatasync $fh, $callback->($status)
242 aio_sync_file_range $fh, $offset, $nbytes, $flags, $callback->($status)
243 aio_pathsync $path, $callback->($status)
244 aio_msync $scalar, $offset = 0, $length = undef, flags = 0, $callback->($status)
245 aio_mtouch $scalar, $offset = 0, $length = undef, flags = 0, $callback->($status)
246 aio_mlock $scalar, $offset = 0, $length = undef, $callback->($status)
247 aio_mlockall $flags, $callback->($status)
248 aio_group $callback->(...)
249 aio_nop $callback->()
250
251 $prev_pri = aioreq_pri [$pri]
252 aioreq_nice $pri_adjust
253
254 IO::AIO::poll_wait
255 IO::AIO::poll_cb
256 IO::AIO::poll
257 IO::AIO::flush
258 IO::AIO::max_poll_reqs $nreqs
259 IO::AIO::max_poll_time $seconds
260 IO::AIO::min_parallel $nthreads
261 IO::AIO::max_parallel $nthreads
262 IO::AIO::max_idle $nthreads
263 IO::AIO::idle_timeout $seconds
264 IO::AIO::max_outstanding $maxreqs
265 IO::AIO::nreqs
266 IO::AIO::nready
267 IO::AIO::npending
268
269 IO::AIO::sendfile $ofh, $ifh, $offset, $count
270 IO::AIO::fadvise $fh, $offset, $len, $advice
271 IO::AIO::madvise $scalar, $offset, $length, $advice
272 IO::AIO::mprotect $scalar, $offset, $length, $protect
273 IO::AIO::munlock $scalar, $offset = 0, $length = undef
274 IO::AIO::munlockall
214 275
215=head2 AIO REQUEST FUNCTIONS 276=head2 AIO REQUEST FUNCTIONS
216 277
217All the C<aio_*> calls are more or less thin wrappers around the syscall 278All the C<aio_*> calls are more or less thin wrappers around the syscall
218with the same name (sans C<aio_>). The arguments are similar or identical, 279with the same name (sans C<aio_>). The arguments are similar or identical,
219and they all accept an additional (and optional) C<$callback> argument 280and they all accept an additional (and optional) C<$callback> argument
220which must be a code reference. This code reference will get called with 281which must be a code reference. This code reference will get called with
221the syscall return code (e.g. most syscalls return C<-1> on error, unlike 282the syscall return code (e.g. most syscalls return C<-1> on error, unlike
222perl, which usually delivers "false") as it's sole argument when the given 283perl, which usually delivers "false") as its sole argument after the given
223syscall has been executed asynchronously. 284syscall has been executed asynchronously.
224 285
225All functions expecting a filehandle keep a copy of the filehandle 286All functions expecting a filehandle keep a copy of the filehandle
226internally until the request has finished. 287internally until the request has finished.
227 288
241your pathnames to the locale (or other) encoding in effect in the user 302your pathnames to the locale (or other) encoding in effect in the user
242environment, d) use Glib::filename_from_unicode on unicode filenames or e) 303environment, d) use Glib::filename_from_unicode on unicode filenames or e)
243use something else to ensure your scalar has the correct contents. 304use something else to ensure your scalar has the correct contents.
244 305
245This works, btw. independent of the internal UTF-8 bit, which IO::AIO 306This works, btw. independent of the internal UTF-8 bit, which IO::AIO
246handles correctly wether it is set or not. 307handles correctly whether it is set or not.
247 308
248=over 4 309=over 4
249 310
250=item $prev_pri = aioreq_pri [$pri] 311=item $prev_pri = aioreq_pri [$pri]
251 312
298by the umask in effect then the request is being executed, so better never 359by the umask in effect then the request is being executed, so better never
299change the umask. 360change the umask.
300 361
301Example: 362Example:
302 363
303 aio_open "/etc/passwd", O_RDONLY, 0, sub { 364 aio_open "/etc/passwd", IO::AIO::O_RDONLY, 0, sub {
304 if ($_[0]) { 365 if ($_[0]) {
305 print "open successful, fh is $_[0]\n"; 366 print "open successful, fh is $_[0]\n";
306 ... 367 ...
307 } else { 368 } else {
308 die "open failed: $!\n"; 369 die "open failed: $!\n";
311 372
312 373
313=item aio_close $fh, $callback->($status) 374=item aio_close $fh, $callback->($status)
314 375
315Asynchronously close a file and call the callback with the result 376Asynchronously close a file and call the callback with the result
316code. I<WARNING:> although accepted, you should not pass in a perl 377code.
317filehandle here, as perl will likely close the file descriptor another
318time when the filehandle is destroyed. Normally, you can safely call perls
319C<close> or just let filehandles go out of scope.
320 378
321This is supposed to be a bug in the API, so that might change. It's 379Unfortunately, you can't do this to perl. Perl I<insists> very strongly on
322therefore best to avoid this function. 380closing the file descriptor associated with the filehandle itself.
323 381
382Therefore, C<aio_close> will not close the filehandle - instead it will
383use dup2 to overwrite the file descriptor with the write-end of a pipe
384(the pipe fd will be created on demand and will be cached).
385
386Or in other words: the file descriptor will be closed, but it will not be
387free for reuse until the perl filehandle is closed.
388
389=cut
324 390
325=item aio_read $fh,$offset,$length, $data,$dataoffset, $callback->($retval) 391=item aio_read $fh,$offset,$length, $data,$dataoffset, $callback->($retval)
326 392
327=item aio_write $fh,$offset,$length, $data,$dataoffset, $callback->($retval) 393=item aio_write $fh,$offset,$length, $data,$dataoffset, $callback->($retval)
328 394
329Reads or writes C<$length> bytes from the specified C<$fh> and C<$offset> 395Reads or writes C<$length> bytes from or to the specified C<$fh> and
330into the scalar given by C<$data> and offset C<$dataoffset> and calls the 396C<$offset> into the scalar given by C<$data> and offset C<$dataoffset>
331callback without the actual number of bytes read (or -1 on error, just 397and calls the callback without the actual number of bytes read (or -1 on
332like the syscall). 398error, just like the syscall).
399
400C<aio_read> will, like C<sysread>, shrink or grow the C<$data> scalar to
401offset plus the actual number of bytes read.
333 402
334If C<$offset> is undefined, then the current file descriptor offset will 403If C<$offset> is undefined, then the current file descriptor offset will
335be used (and updated), otherwise the file descriptor offset will not be 404be used (and updated), otherwise the file descriptor offset will not be
336changed by these calls. 405changed by these calls.
337 406
338If C<$length> is undefined in C<aio_write>, use the remaining length of C<$data>. 407If C<$length> is undefined in C<aio_write>, use the remaining length of
408C<$data>.
339 409
340If C<$dataoffset> is less than zero, it will be counted from the end of 410If C<$dataoffset> is less than zero, it will be counted from the end of
341C<$data>. 411C<$data>.
342 412
343The C<$data> scalar I<MUST NOT> be modified in any way while the request 413The C<$data> scalar I<MUST NOT> be modified in any way while the request
359reading at byte offset C<$in_offset>, and starts writing at the current 429reading at byte offset C<$in_offset>, and starts writing at the current
360file offset of C<$out_fh>. Because of that, it is not safe to issue more 430file offset of C<$out_fh>. Because of that, it is not safe to issue more
361than one C<aio_sendfile> per C<$out_fh>, as they will interfere with each 431than one C<aio_sendfile> per C<$out_fh>, as they will interfere with each
362other. 432other.
363 433
434Please note that C<aio_sendfile> can read more bytes from C<$in_fh> than
435are written, and there is no way to find out how many bytes have been read
436from C<aio_sendfile> alone, as C<aio_sendfile> only provides the number of
437bytes written to C<$out_fh>. Only if the result value equals C<$length>
438one can assume that C<$length> bytes have been read.
439
440Unlike with other C<aio_> functions, it makes a lot of sense to use
441C<aio_sendfile> on non-blocking sockets, as long as one end (typically
442the C<$in_fh>) is a file - the file I/O will then be asynchronous, while
443the socket I/O will be non-blocking. Note, however, that you can run into
444a trap where C<aio_sendfile> reads some data with readahead, then fails
445to write all data, and when the socket is ready the next time, the data
446in the cache is already lost, forcing C<aio_sendfile> to again hit the
447disk. Explicit C<aio_read> + C<aio_write> let's you control resource usage
448much better.
449
364This call tries to make use of a native C<sendfile> syscall to provide 450This call tries to make use of a native C<sendfile> syscall to provide
365zero-copy operation. For this to work, C<$out_fh> should refer to a 451zero-copy operation. For this to work, C<$out_fh> should refer to a
366socket, and C<$in_fh> should refer to mmap'able file. 452socket, and C<$in_fh> should refer to an mmap'able file.
367 453
368If the native sendfile call fails or is not implemented, it will be 454If a native sendfile cannot be found or it fails with C<ENOSYS>,
455C<ENOTSUP>, C<EOPNOTSUPP>, C<EAFNOSUPPORT>, C<EPROTOTYPE> or C<ENOTSOCK>,
369emulated, so you can call C<aio_sendfile> on any type of filehandle 456it will be emulated, so you can call C<aio_sendfile> on any type of
370regardless of the limitations of the operating system. 457filehandle regardless of the limitations of the operating system.
371
372Please note, however, that C<aio_sendfile> can read more bytes from
373C<$in_fh> than are written, and there is no way to find out how many
374bytes have been read from C<aio_sendfile> alone, as C<aio_sendfile> only
375provides the number of bytes written to C<$out_fh>. Only if the result
376value equals C<$length> one can assume that C<$length> bytes have been
377read.
378 458
379 459
380=item aio_readahead $fh,$offset,$length, $callback->($retval) 460=item aio_readahead $fh,$offset,$length, $callback->($retval)
381 461
382C<aio_readahead> populates the page cache with data from a file so that 462C<aio_readahead> populates the page cache with data from a file so that
405 485
406Currently, the stats are always 64-bit-stats, i.e. instead of returning an 486Currently, the stats are always 64-bit-stats, i.e. instead of returning an
407error when stat'ing a large file, the results will be silently truncated 487error when stat'ing a large file, the results will be silently truncated
408unless perl itself is compiled with large file support. 488unless perl itself is compiled with large file support.
409 489
490To help interpret the mode and dev/rdev stat values, IO::AIO offers the
491following constants and functions (if not implemented, the constants will
492be C<0> and the functions will either C<croak> or fall back on traditional
493behaviour).
494
495C<S_IFMT>, C<S_IFIFO>, C<S_IFCHR>, C<S_IFBLK>, C<S_IFLNK>, C<S_IFREG>,
496C<S_IFDIR>, C<S_IFWHT>, C<S_IFSOCK>, C<IO::AIO::major $dev_t>,
497C<IO::AIO::minor $dev_t>, C<IO::AIO::makedev $major, $minor>.
498
410Example: Print the length of F</etc/passwd>: 499Example: Print the length of F</etc/passwd>:
411 500
412 aio_stat "/etc/passwd", sub { 501 aio_stat "/etc/passwd", sub {
413 $_[0] and die "stat failed: $!"; 502 $_[0] and die "stat failed: $!";
414 print "size is ", -s _, "\n"; 503 print "size is ", -s _, "\n";
415 }; 504 };
416 505
417 506
507=item aio_statvfs $fh_or_path, $callback->($statvfs)
508
509Works like the POSIX C<statvfs> or C<fstatvfs> syscalls, depending on
510whether a file handle or path was passed.
511
512On success, the callback is passed a hash reference with the following
513members: C<bsize>, C<frsize>, C<blocks>, C<bfree>, C<bavail>, C<files>,
514C<ffree>, C<favail>, C<fsid>, C<flag> and C<namemax>. On failure, C<undef>
515is passed.
516
517The following POSIX IO::AIO::ST_* constants are defined: C<ST_RDONLY> and
518C<ST_NOSUID>.
519
520The following non-POSIX IO::AIO::ST_* flag masks are defined to
521their correct value when available, or to C<0> on systems that do
522not support them: C<ST_NODEV>, C<ST_NOEXEC>, C<ST_SYNCHRONOUS>,
523C<ST_MANDLOCK>, C<ST_WRITE>, C<ST_APPEND>, C<ST_IMMUTABLE>, C<ST_NOATIME>,
524C<ST_NODIRATIME> and C<ST_RELATIME>.
525
526Example: stat C</wd> and dump out the data if successful.
527
528 aio_statvfs "/wd", sub {
529 my $f = $_[0]
530 or die "statvfs: $!";
531
532 use Data::Dumper;
533 say Dumper $f;
534 };
535
536 # result:
537 {
538 bsize => 1024,
539 bfree => 4333064312,
540 blocks => 10253828096,
541 files => 2050765568,
542 flag => 4096,
543 favail => 2042092649,
544 bavail => 4333064312,
545 ffree => 2042092649,
546 namemax => 255,
547 frsize => 1024,
548 fsid => 1810
549 }
550
551
418=item aio_utime $fh_or_path, $atime, $mtime, $callback->($status) 552=item aio_utime $fh_or_path, $atime, $mtime, $callback->($status)
419 553
420Works like perl's C<utime> function (including the special case of $atime 554Works like perl's C<utime> function (including the special case of $atime
421and $mtime being undef). Fractional times are supported if the underlying 555and $mtime being undef). Fractional times are supported if the underlying
422syscalls support them. 556syscalls support them.
470 604
471The only (POSIX-) portable way of calling this function is: 605The only (POSIX-) portable way of calling this function is:
472 606
473 aio_mknod $path, IO::AIO::S_IFIFO | $mode, 0, sub { ... 607 aio_mknod $path, IO::AIO::S_IFIFO | $mode, 0, sub { ...
474 608
609See C<aio_stat> for info about some potentially helpful extra constants
610and functions.
475 611
476=item aio_link $srcpath, $dstpath, $callback->($status) 612=item aio_link $srcpath, $dstpath, $callback->($status)
477 613
478Asynchronously create a new link to the existing object at C<$srcpath> at 614Asynchronously create a new link to the existing object at C<$srcpath> at
479the path C<$dstpath> and call the callback with the result code. 615the path C<$dstpath> and call the callback with the result code.
515 651
516Unlike the POSIX call of the same name, C<aio_readdir> reads an entire 652Unlike the POSIX call of the same name, C<aio_readdir> reads an entire
517directory (i.e. opendir + readdir + closedir). The entries will not be 653directory (i.e. opendir + readdir + closedir). The entries will not be
518sorted, and will B<NOT> include the C<.> and C<..> entries. 654sorted, and will B<NOT> include the C<.> and C<..> entries.
519 655
520The callback a single argument which is either C<undef> or an array-ref 656The callback is passed a single argument which is either C<undef> or an
521with the filenames. 657array-ref with the filenames.
658
659
660=item aio_readdirx $pathname, $flags, $callback->($entries, $flags)
661
662Quite similar to C<aio_readdir>, but the C<$flags> argument allows to tune
663behaviour and output format. In case of an error, C<$entries> will be
664C<undef>.
665
666The flags are a combination of the following constants, ORed together (the
667flags will also be passed to the callback, possibly modified):
668
669=over 4
670
671=item IO::AIO::READDIR_DENTS
672
673When this flag is off, then the callback gets an arrayref consisting of
674names only (as with C<aio_readdir>), otherwise it gets an arrayref with
675C<[$name, $type, $inode]> arrayrefs, each describing a single directory
676entry in more detail.
677
678C<$name> is the name of the entry.
679
680C<$type> is one of the C<IO::AIO::DT_xxx> constants:
681
682C<IO::AIO::DT_UNKNOWN>, C<IO::AIO::DT_FIFO>, C<IO::AIO::DT_CHR>, C<IO::AIO::DT_DIR>,
683C<IO::AIO::DT_BLK>, C<IO::AIO::DT_REG>, C<IO::AIO::DT_LNK>, C<IO::AIO::DT_SOCK>,
684C<IO::AIO::DT_WHT>.
685
686C<IO::AIO::DT_UNKNOWN> means just that: readdir does not know. If you need to
687know, you have to run stat yourself. Also, for speed reasons, the C<$type>
688scalars are read-only: you can not modify them.
689
690C<$inode> is the inode number (which might not be exact on systems with 64
691bit inode numbers and 32 bit perls). This field has unspecified content on
692systems that do not deliver the inode information.
693
694=item IO::AIO::READDIR_DIRS_FIRST
695
696When this flag is set, then the names will be returned in an order where
697likely directories come first. This is useful when you need to quickly
698find directories, or you want to find all directories while avoiding to
699stat() each entry.
700
701If the system returns type information in readdir, then this is used
702to find directories directly. Otherwise, likely directories are files
703beginning with ".", or otherwise files with no dots, of which files with
704short names are tried first.
705
706=item IO::AIO::READDIR_STAT_ORDER
707
708When this flag is set, then the names will be returned in an order
709suitable for stat()'ing each one. That is, when you plan to stat()
710all files in the given directory, then the returned order will likely
711be fastest.
712
713If both this flag and C<IO::AIO::READDIR_DIRS_FIRST> are specified, then
714the likely dirs come first, resulting in a less optimal stat order.
715
716=item IO::AIO::READDIR_FOUND_UNKNOWN
717
718This flag should not be set when calling C<aio_readdirx>. Instead, it
719is being set by C<aio_readdirx>, when any of the C<$type>'s found were
720C<IO::AIO::DT_UNKNOWN>. The absense of this flag therefore indicates that all
721C<$type>'s are known, which can be used to speed up some algorithms.
722
723=back
522 724
523 725
524=item aio_load $path, $data, $callback->($status) 726=item aio_load $path, $data, $callback->($status)
525 727
526This is a composite request that tries to fully load the given file into 728This is a composite request that tries to fully load the given file into
527memory. Status is the same as with aio_read. 729memory. Status is the same as with aio_read.
528 730
529=cut 731=cut
530 732
531sub aio_load($$;$) { 733sub aio_load($$;$) {
532 aio_block {
533 my ($path, undef, $cb) = @_; 734 my ($path, undef, $cb) = @_;
534 my $data = \$_[1]; 735 my $data = \$_[1];
535 736
536 my $pri = aioreq_pri; 737 my $pri = aioreq_pri;
537 my $grp = aio_group $cb; 738 my $grp = aio_group $cb;
739
740 aioreq_pri $pri;
741 add $grp aio_open $path, O_RDONLY, 0, sub {
742 my $fh = shift
743 or return $grp->result (-1);
538 744
539 aioreq_pri $pri; 745 aioreq_pri $pri;
540 add $grp aio_open $path, O_RDONLY, 0, sub {
541 my $fh = shift
542 or return $grp->result (-1);
543
544 aioreq_pri $pri;
545 add $grp aio_read $fh, 0, (-s $fh), $$data, 0, sub { 746 add $grp aio_read $fh, 0, (-s $fh), $$data, 0, sub {
546 $grp->result ($_[0]); 747 $grp->result ($_[0]);
547 };
548 }; 748 };
549
550 $grp
551 } 749 };
750
751 $grp
552} 752}
553 753
554=item aio_copy $srcpath, $dstpath, $callback->($status) 754=item aio_copy $srcpath, $dstpath, $callback->($status)
555 755
556Try to copy the I<file> (directories not supported as either source or 756Try to copy the I<file> (directories not supported as either source or
557destination) from C<$srcpath> to C<$dstpath> and call the callback with 757destination) from C<$srcpath> to C<$dstpath> and call the callback with
558the C<0> (error) or C<-1> ok. 758a status of C<0> (ok) or C<-1> (error, see C<$!>).
559 759
560This is a composite request that it creates the destination file with 760This is a composite request that creates the destination file with
561mode 0200 and copies the contents of the source file into it using 761mode 0200 and copies the contents of the source file into it using
562C<aio_sendfile>, followed by restoring atime, mtime, access mode and 762C<aio_sendfile>, followed by restoring atime, mtime, access mode and
563uid/gid, in that order. 763uid/gid, in that order.
564 764
565If an error occurs, the partial destination file will be unlinked, if 765If an error occurs, the partial destination file will be unlinked, if
567errors are being ignored. 767errors are being ignored.
568 768
569=cut 769=cut
570 770
571sub aio_copy($$;$) { 771sub aio_copy($$;$) {
572 aio_block {
573 my ($src, $dst, $cb) = @_; 772 my ($src, $dst, $cb) = @_;
574 773
575 my $pri = aioreq_pri; 774 my $pri = aioreq_pri;
576 my $grp = aio_group $cb; 775 my $grp = aio_group $cb;
577 776
578 aioreq_pri $pri; 777 aioreq_pri $pri;
579 add $grp aio_open $src, O_RDONLY, 0, sub { 778 add $grp aio_open $src, O_RDONLY, 0, sub {
580 if (my $src_fh = $_[0]) { 779 if (my $src_fh = $_[0]) {
581 my @stat = stat $src_fh; 780 my @stat = stat $src_fh; # hmm, might block over nfs?
582 781
583 aioreq_pri $pri; 782 aioreq_pri $pri;
584 add $grp aio_open $dst, O_CREAT | O_WRONLY | O_TRUNC, 0200, sub { 783 add $grp aio_open $dst, O_CREAT | O_WRONLY | O_TRUNC, 0200, sub {
585 if (my $dst_fh = $_[0]) { 784 if (my $dst_fh = $_[0]) {
586 aioreq_pri $pri; 785 aioreq_pri $pri;
587 add $grp aio_sendfile $dst_fh, $src_fh, 0, $stat[7], sub { 786 add $grp aio_sendfile $dst_fh, $src_fh, 0, $stat[7], sub {
588 if ($_[0] == $stat[7]) { 787 if ($_[0] == $stat[7]) {
589 $grp->result (0); 788 $grp->result (0);
590 close $src_fh; 789 close $src_fh;
591 790
592 # those should not normally block. should. should.
593 utime $stat[8], $stat[9], $dst;
594 chmod $stat[2] & 07777, $dst_fh;
595 chown $stat[4], $stat[5], $dst_fh;
596 close $dst_fh;
597 } else { 791 my $ch = sub {
598 $grp->result (-1);
599 close $src_fh;
600 close $dst_fh;
601
602 aioreq $pri; 792 aioreq_pri $pri;
793 add $grp aio_chmod $dst_fh, $stat[2] & 07777, sub {
794 aioreq_pri $pri;
795 add $grp aio_chown $dst_fh, $stat[4], $stat[5], sub {
796 aioreq_pri $pri;
603 add $grp aio_unlink $dst; 797 add $grp aio_close $dst_fh;
798 }
799 };
604 } 800 };
801
802 aioreq_pri $pri;
803 add $grp aio_utime $dst_fh, $stat[8], $stat[9], sub {
804 if ($_[0] < 0 && $! == ENOSYS) {
805 aioreq_pri $pri;
806 add $grp aio_utime $dst, $stat[8], $stat[9], $ch;
807 } else {
808 $ch->();
809 }
810 };
811 } else {
812 $grp->result (-1);
813 close $src_fh;
814 close $dst_fh;
815
816 aioreq $pri;
817 add $grp aio_unlink $dst;
605 }; 818 }
606 } else {
607 $grp->result (-1);
608 } 819 };
820 } else {
821 $grp->result (-1);
609 }, 822 }
610
611 } else {
612 $grp->result (-1);
613 } 823 },
824
825 } else {
826 $grp->result (-1);
614 }; 827 }
615
616 $grp
617 } 828 };
829
830 $grp
618} 831}
619 832
620=item aio_move $srcpath, $dstpath, $callback->($status) 833=item aio_move $srcpath, $dstpath, $callback->($status)
621 834
622Try to move the I<file> (directories not supported as either source or 835Try to move the I<file> (directories not supported as either source or
623destination) from C<$srcpath> to C<$dstpath> and call the callback with 836destination) from C<$srcpath> to C<$dstpath> and call the callback with
624the C<0> (error) or C<-1> ok. 837a status of C<0> (ok) or C<-1> (error, see C<$!>).
625 838
626This is a composite request that tries to rename(2) the file first. If 839This is a composite request that tries to rename(2) the file first; if
627rename files with C<EXDEV>, it copies the file with C<aio_copy> and, if 840rename fails with C<EXDEV>, it copies the file with C<aio_copy> and, if
628that is successful, unlinking the C<$srcpath>. 841that is successful, unlinks the C<$srcpath>.
629 842
630=cut 843=cut
631 844
632sub aio_move($$;$) { 845sub aio_move($$;$) {
633 aio_block {
634 my ($src, $dst, $cb) = @_; 846 my ($src, $dst, $cb) = @_;
635 847
636 my $pri = aioreq_pri; 848 my $pri = aioreq_pri;
637 my $grp = aio_group $cb; 849 my $grp = aio_group $cb;
638 850
639 aioreq_pri $pri; 851 aioreq_pri $pri;
640 add $grp aio_rename $src, $dst, sub { 852 add $grp aio_rename $src, $dst, sub {
641 if ($_[0] && $! == EXDEV) { 853 if ($_[0] && $! == EXDEV) {
642 aioreq_pri $pri; 854 aioreq_pri $pri;
643 add $grp aio_copy $src, $dst, sub { 855 add $grp aio_copy $src, $dst, sub {
644 $grp->result ($_[0]);
645
646 if (!$_[0]) {
647 aioreq_pri $pri;
648 add $grp aio_unlink $src;
649 }
650 };
651 } else {
652 $grp->result ($_[0]); 856 $grp->result ($_[0]);
857
858 if (!$_[0]) {
859 aioreq_pri $pri;
860 add $grp aio_unlink $src;
861 }
653 } 862 };
863 } else {
864 $grp->result ($_[0]);
654 }; 865 }
655
656 $grp
657 } 866 };
867
868 $grp
658} 869}
659 870
660=item aio_scandir $path, $maxreq, $callback->($dirs, $nondirs) 871=item aio_scandir $path, $maxreq, $callback->($dirs, $nondirs)
661 872
662Scans a directory (similar to C<aio_readdir>) but additionally tries to 873Scans a directory (similar to C<aio_readdir>) but additionally tries to
682 893
683Implementation notes. 894Implementation notes.
684 895
685The C<aio_readdir> cannot be avoided, but C<stat()>'ing every entry can. 896The C<aio_readdir> cannot be avoided, but C<stat()>'ing every entry can.
686 897
898If readdir returns file type information, then this is used directly to
899find directories.
900
687After reading the directory, the modification time, size etc. of the 901Otherwise, after reading the directory, the modification time, size etc.
688directory before and after the readdir is checked, and if they match (and 902of the directory before and after the readdir is checked, and if they
689isn't the current time), the link count will be used to decide how many 903match (and isn't the current time), the link count will be used to decide
690entries are directories (if >= 2). Otherwise, no knowledge of the number 904how many entries are directories (if >= 2). Otherwise, no knowledge of the
691of subdirectories will be assumed. 905number of subdirectories will be assumed.
692 906
693Then entries will be sorted into likely directories (everything without 907Then entries will be sorted into likely directories a non-initial dot
694a non-initial dot currently) and likely non-directories (everything 908currently) and likely non-directories (see C<aio_readdirx>). Then every
695else). Then every entry plus an appended C</.> will be C<stat>'ed, 909entry plus an appended C</.> will be C<stat>'ed, likely directories first,
696likely directories first. If that succeeds, it assumes that the entry 910in order of their inode numbers. If that succeeds, it assumes that the
697is a directory or a symlink to directory (which will be checked 911entry is a directory or a symlink to directory (which will be checked
698seperately). This is often faster than stat'ing the entry itself because 912seperately). This is often faster than stat'ing the entry itself because
699filesystems might detect the type of the entry without reading the inode 913filesystems might detect the type of the entry without reading the inode
700data (e.g. ext2fs filetype feature). 914data (e.g. ext2fs filetype feature), even on systems that cannot return
915the filetype information on readdir.
701 916
702If the known number of directories (link count - 2) has been reached, the 917If the known number of directories (link count - 2) has been reached, the
703rest of the entries is assumed to be non-directories. 918rest of the entries is assumed to be non-directories.
704 919
705This only works with certainty on POSIX (= UNIX) filesystems, which 920This only works with certainty on POSIX (= UNIX) filesystems, which
710directory counting heuristic. 925directory counting heuristic.
711 926
712=cut 927=cut
713 928
714sub aio_scandir($$;$) { 929sub aio_scandir($$;$) {
715 aio_block {
716 my ($path, $maxreq, $cb) = @_; 930 my ($path, $maxreq, $cb) = @_;
717 931
718 my $pri = aioreq_pri; 932 my $pri = aioreq_pri;
719 933
720 my $grp = aio_group $cb; 934 my $grp = aio_group $cb;
721 935
722 $maxreq = 4 if $maxreq <= 0; 936 $maxreq = 4 if $maxreq <= 0;
723 937
724 # stat once 938 # stat once
939 aioreq_pri $pri;
940 add $grp aio_stat $path, sub {
941 return $grp->result () if $_[0];
942 my $now = time;
943 my $hash1 = join ":", (stat _)[0,1,3,7,9];
944
945 # read the directory entries
725 aioreq_pri $pri; 946 aioreq_pri $pri;
726 add $grp aio_stat $path, sub { 947 add $grp aio_readdirx $path, READDIR_DIRS_FIRST, sub {
948 my $entries = shift
727 return $grp->result () if $_[0]; 949 or return $grp->result ();
728 my $now = time;
729 my $hash1 = join ":", (stat _)[0,1,3,7,9];
730 950
731 # read the directory entries 951 # stat the dir another time
732 aioreq_pri $pri; 952 aioreq_pri $pri;
733 add $grp aio_readdir $path, sub {
734 my $entries = shift
735 or return $grp->result ();
736
737 # stat the dir another time
738 aioreq_pri $pri;
739 add $grp aio_stat $path, sub { 953 add $grp aio_stat $path, sub {
740 my $hash2 = join ":", (stat _)[0,1,3,7,9]; 954 my $hash2 = join ":", (stat _)[0,1,3,7,9];
741 955
742 my $ndirs; 956 my $ndirs;
743 957
744 # take the slow route if anything looks fishy 958 # take the slow route if anything looks fishy
745 if ($hash1 ne $hash2 or (stat _)[9] == $now) { 959 if ($hash1 ne $hash2 or (stat _)[9] == $now) {
746 $ndirs = -1; 960 $ndirs = -1;
747 } else { 961 } else {
748 # if nlink == 2, we are finished 962 # if nlink == 2, we are finished
749 # on non-posix-fs's, we rely on nlink < 2 963 # for non-posix-fs's, we rely on nlink < 2
750 $ndirs = (stat _)[3] - 2 964 $ndirs = (stat _)[3] - 2
751 or return $grp->result ([], $entries); 965 or return $grp->result ([], $entries);
752 } 966 }
753 967
754 # sort into likely dirs and likely nondirs
755 # dirs == files without ".", short entries first
756 $entries = [map $_->[0],
757 sort { $b->[1] cmp $a->[1] }
758 map [$_, sprintf "%s%04d", (/.\./ ? "1" : "0"), length],
759 @$entries];
760
761 my (@dirs, @nondirs); 968 my (@dirs, @nondirs);
762 969
763 my $statgrp = add $grp aio_group sub { 970 my $statgrp = add $grp aio_group sub {
764 $grp->result (\@dirs, \@nondirs); 971 $grp->result (\@dirs, \@nondirs);
765 }; 972 };
766 973
767 limit $statgrp $maxreq; 974 limit $statgrp $maxreq;
768 feed $statgrp sub { 975 feed $statgrp sub {
769 return unless @$entries; 976 return unless @$entries;
770 my $entry = pop @$entries; 977 my $entry = shift @$entries;
771 978
772 aioreq_pri $pri; 979 aioreq_pri $pri;
773 add $statgrp aio_stat "$path/$entry/.", sub { 980 add $statgrp aio_stat "$path/$entry/.", sub {
774 if ($_[0] < 0) { 981 if ($_[0] < 0) {
775 push @nondirs, $entry; 982 push @nondirs, $entry;
776 } else { 983 } else {
777 # need to check for real directory 984 # need to check for real directory
778 aioreq_pri $pri; 985 aioreq_pri $pri;
779 add $statgrp aio_lstat "$path/$entry", sub { 986 add $statgrp aio_lstat "$path/$entry", sub {
780 if (-d _) { 987 if (-d _) {
781 push @dirs, $entry; 988 push @dirs, $entry;
782 989
783 unless (--$ndirs) { 990 unless (--$ndirs) {
784 push @nondirs, @$entries; 991 push @nondirs, @$entries;
785 feed $statgrp; 992 feed $statgrp;
786 }
787 } else {
788 push @nondirs, $entry;
789 } 993 }
994 } else {
995 push @nondirs, $entry;
790 } 996 }
791 } 997 }
792 }; 998 }
793 }; 999 };
794 }; 1000 };
795 }; 1001 };
796 }; 1002 };
797
798 $grp
799 } 1003 };
1004
1005 $grp
800} 1006}
801 1007
802=item aio_rmtree $path, $callback->($status) 1008=item aio_rmtree $path, $callback->($status)
803 1009
804Delete a directory tree starting (and including) C<$path>, return the 1010Delete a directory tree starting (and including) C<$path>, return the
808 1014
809=cut 1015=cut
810 1016
811sub aio_rmtree; 1017sub aio_rmtree;
812sub aio_rmtree($;$) { 1018sub aio_rmtree($;$) {
813 aio_block {
814 my ($path, $cb) = @_; 1019 my ($path, $cb) = @_;
815 1020
816 my $pri = aioreq_pri; 1021 my $pri = aioreq_pri;
817 my $grp = aio_group $cb; 1022 my $grp = aio_group $cb;
818 1023
819 aioreq_pri $pri; 1024 aioreq_pri $pri;
820 add $grp aio_scandir $path, 0, sub { 1025 add $grp aio_scandir $path, 0, sub {
821 my ($dirs, $nondirs) = @_; 1026 my ($dirs, $nondirs) = @_;
822 1027
823 my $dirgrp = aio_group sub { 1028 my $dirgrp = aio_group sub {
824 add $grp aio_rmdir $path, sub { 1029 add $grp aio_rmdir $path, sub {
825 $grp->result ($_[0]); 1030 $grp->result ($_[0]);
826 };
827 }; 1031 };
828
829 (aioreq_pri $pri), add $dirgrp aio_rmtree "$path/$_" for @$dirs;
830 (aioreq_pri $pri), add $dirgrp aio_unlink "$path/$_" for @$nondirs;
831
832 add $grp $dirgrp;
833 }; 1032 };
834 1033
835 $grp 1034 (aioreq_pri $pri), add $dirgrp aio_rmtree "$path/$_" for @$dirs;
1035 (aioreq_pri $pri), add $dirgrp aio_unlink "$path/$_" for @$nondirs;
1036
1037 add $grp $dirgrp;
836 } 1038 };
1039
1040 $grp
837} 1041}
1042
1043=item aio_sync $callback->($status)
1044
1045Asynchronously call sync and call the callback when finished.
838 1046
839=item aio_fsync $fh, $callback->($status) 1047=item aio_fsync $fh, $callback->($status)
840 1048
841Asynchronously call fsync on the given filehandle and call the callback 1049Asynchronously call fsync on the given filehandle and call the callback
842with the fsync result code. 1050with the fsync result code.
846Asynchronously call fdatasync on the given filehandle and call the 1054Asynchronously call fdatasync on the given filehandle and call the
847callback with the fdatasync result code. 1055callback with the fdatasync result code.
848 1056
849If this call isn't available because your OS lacks it or it couldn't be 1057If this call isn't available because your OS lacks it or it couldn't be
850detected, it will be emulated by calling C<fsync> instead. 1058detected, it will be emulated by calling C<fsync> instead.
1059
1060=item aio_sync_file_range $fh, $offset, $nbytes, $flags, $callback->($status)
1061
1062Sync the data portion of the file specified by C<$offset> and C<$length>
1063to disk (but NOT the metadata), by calling the Linux-specific
1064sync_file_range call. If sync_file_range is not available or it returns
1065ENOSYS, then fdatasync or fsync is being substituted.
1066
1067C<$flags> can be a combination of C<IO::AIO::SYNC_FILE_RANGE_WAIT_BEFORE>,
1068C<IO::AIO::SYNC_FILE_RANGE_WRITE> and
1069C<IO::AIO::SYNC_FILE_RANGE_WAIT_AFTER>: refer to the sync_file_range
1070manpage for details.
1071
1072=item aio_pathsync $path, $callback->($status)
1073
1074This request tries to open, fsync and close the given path. This is a
1075composite request intended to sync directories after directory operations
1076(E.g. rename). This might not work on all operating systems or have any
1077specific effect, but usually it makes sure that directory changes get
1078written to disc. It works for anything that can be opened for read-only,
1079not just directories.
1080
1081Future versions of this function might fall back to other methods when
1082C<fsync> on the directory fails (such as calling C<sync>).
1083
1084Passes C<0> when everything went ok, and C<-1> on error.
1085
1086=cut
1087
1088sub aio_pathsync($;$) {
1089 my ($path, $cb) = @_;
1090
1091 my $pri = aioreq_pri;
1092 my $grp = aio_group $cb;
1093
1094 aioreq_pri $pri;
1095 add $grp aio_open $path, O_RDONLY, 0, sub {
1096 my ($fh) = @_;
1097 if ($fh) {
1098 aioreq_pri $pri;
1099 add $grp aio_fsync $fh, sub {
1100 $grp->result ($_[0]);
1101
1102 aioreq_pri $pri;
1103 add $grp aio_close $fh;
1104 };
1105 } else {
1106 $grp->result (-1);
1107 }
1108 };
1109
1110 $grp
1111}
1112
1113=item aio_msync $scalar, $offset = 0, $length = undef, flags = 0, $callback->($status)
1114
1115This is a rather advanced IO::AIO call, which only works on mmap(2)ed
1116scalars (see the C<IO::AIO::mmap> function, although it also works on data
1117scalars managed by the L<Sys::Mmap> or L<Mmap> modules, note that the
1118scalar must only be modified in-place while an aio operation is pending on
1119it).
1120
1121It calls the C<msync> function of your OS, if available, with the memory
1122area starting at C<$offset> in the string and ending C<$length> bytes
1123later. If C<$length> is negative, counts from the end, and if C<$length>
1124is C<undef>, then it goes till the end of the string. The flags can be
1125a combination of C<IO::AIO::MS_ASYNC>, C<IO::AIO::MS_INVALIDATE> and
1126C<IO::AIO::MS_SYNC>.
1127
1128=item aio_mtouch $scalar, $offset = 0, $length = undef, flags = 0, $callback->($status)
1129
1130This is a rather advanced IO::AIO call, which works best on mmap(2)ed
1131scalars.
1132
1133It touches (reads or writes) all memory pages in the specified
1134range inside the scalar. All caveats and parameters are the same
1135as for C<aio_msync>, above, except for flags, which must be either
1136C<0> (which reads all pages and ensures they are instantiated) or
1137C<IO::AIO::MT_MODIFY>, which modifies the memory page s(by reading and
1138writing an octet from it, which dirties the page).
1139
1140=item aio_mlock $scalar, $offset = 0, $length = undef, $callback->($status)
1141
1142This is a rather advanced IO::AIO call, which works best on mmap(2)ed
1143scalars.
1144
1145It reads in all the pages of the underlying storage into memory (if any)
1146and locks them, so they are not getting swapped/paged out or removed.
1147
1148If C<$length> is undefined, then the scalar will be locked till the end.
1149
1150On systems that do not implement C<mlock>, this function returns C<-1>
1151and sets errno to C<ENOSYS>.
1152
1153Note that the corresponding C<munlock> is synchronous and is
1154documented under L<MISCELLANEOUS FUNCTIONS>.
1155
1156Example: open a file, mmap and mlock it - both will be undone when
1157C<$data> gets destroyed.
1158
1159 open my $fh, "<", $path or die "$path: $!";
1160 my $data;
1161 IO::AIO::mmap $data, -s $fh, IO::AIO::PROT_READ, IO::AIO::MAP_SHARED, $fh;
1162 aio_mlock $data; # mlock in background
1163
1164=item aio_mlockall $flags, $callback->($status)
1165
1166Calls the C<mlockall> function with the given C<$flags> (a combination of
1167C<IO::AIO::MCL_CURRENT> and C<IO::AIO::MCL_FUTURE>).
1168
1169On systems that do not implement C<mlockall>, this function returns C<-1>
1170and sets errno to C<ENOSYS>.
1171
1172Note that the corresponding C<munlockall> is synchronous and is
1173documented under L<MISCELLANEOUS FUNCTIONS>.
1174
1175Example: asynchronously lock all current and future pages into memory.
1176
1177 aio_mlockall IO::AIO::MCL_FUTURE;
851 1178
852=item aio_group $callback->(...) 1179=item aio_group $callback->(...)
853 1180
854This is a very special aio request: Instead of doing something, it is a 1181This is a very special aio request: Instead of doing something, it is a
855container for other aio requests, which is useful if you want to bundle 1182container for other aio requests, which is useful if you want to bundle
905=item cancel $req 1232=item cancel $req
906 1233
907Cancels the request, if possible. Has the effect of skipping execution 1234Cancels the request, if possible. Has the effect of skipping execution
908when entering the B<execute> state and skipping calling the callback when 1235when entering the B<execute> state and skipping calling the callback when
909entering the the B<result> state, but will leave the request otherwise 1236entering the the B<result> state, but will leave the request otherwise
910untouched. That means that requests that currently execute will not be 1237untouched (with the exception of readdir). That means that requests that
911stopped and resources held by the request will not be freed prematurely. 1238currently execute will not be stopped and resources held by the request
1239will not be freed prematurely.
912 1240
913=item cb $req $callback->(...) 1241=item cb $req $callback->(...)
914 1242
915Replace (or simply set) the callback registered to the request. 1243Replace (or simply set) the callback registered to the request.
916 1244
967Their lifetime, simplified, looks like this: when they are empty, they 1295Their lifetime, simplified, looks like this: when they are empty, they
968will finish very quickly. If they contain only requests that are in the 1296will finish very quickly. If they contain only requests that are in the
969C<done> state, they will also finish. Otherwise they will continue to 1297C<done> state, they will also finish. Otherwise they will continue to
970exist. 1298exist.
971 1299
972That means after creating a group you have some time to add requests. And 1300That means after creating a group you have some time to add requests
973in the callbacks of those requests, you can add further requests to the 1301(precisely before the callback has been invoked, which is only done within
974group. And only when all those requests have finished will the the group 1302the C<poll_cb>). And in the callbacks of those requests, you can add
975itself finish. 1303further requests to the group. And only when all those requests have
1304finished will the the group itself finish.
976 1305
977=over 4 1306=over 4
978 1307
979=item add $grp ... 1308=item add $grp ...
980 1309
989=item $grp->cancel_subs 1318=item $grp->cancel_subs
990 1319
991Cancel all subrequests and clears any feeder, but not the group request 1320Cancel all subrequests and clears any feeder, but not the group request
992itself. Useful when you queued a lot of events but got a result early. 1321itself. Useful when you queued a lot of events but got a result early.
993 1322
1323The group request will finish normally (you cannot add requests to the
1324group).
1325
994=item $grp->result (...) 1326=item $grp->result (...)
995 1327
996Set the result value(s) that will be passed to the group callback when all 1328Set the result value(s) that will be passed to the group callback when all
997subrequests have finished and set thre groups errno to the current value 1329subrequests have finished and set the groups errno to the current value
998of errno (just like calling C<errno> without an error number). By default, 1330of errno (just like calling C<errno> without an error number). By default,
999no argument will be passed and errno is zero. 1331no argument will be passed and errno is zero.
1000 1332
1001=item $grp->errno ([$errno]) 1333=item $grp->errno ([$errno])
1002 1334
1013=item feed $grp $callback->($grp) 1345=item feed $grp $callback->($grp)
1014 1346
1015Sets a feeder/generator on this group: every group can have an attached 1347Sets a feeder/generator on this group: every group can have an attached
1016generator that generates requests if idle. The idea behind this is that, 1348generator that generates requests if idle. The idea behind this is that,
1017although you could just queue as many requests as you want in a group, 1349although you could just queue as many requests as you want in a group,
1018this might starve other requests for a potentially long time. For 1350this might starve other requests for a potentially long time. For example,
1019example, C<aio_scandir> might generate hundreds of thousands C<aio_stat> 1351C<aio_scandir> might generate hundreds of thousands C<aio_stat> requests,
1020requests, delaying any later requests for a long time. 1352delaying any later requests for a long time.
1021 1353
1022To avoid this, and allow incremental generation of requests, you can 1354To avoid this, and allow incremental generation of requests, you can
1023instead a group and set a feeder on it that generates those requests. The 1355instead a group and set a feeder on it that generates those requests. The
1024feed callback will be called whenever there are few enough (see C<limit>, 1356feed callback will be called whenever there are few enough (see C<limit>,
1025below) requests active in the group itself and is expected to queue more 1357below) requests active in the group itself and is expected to queue more
1029not impose any limits). 1361not impose any limits).
1030 1362
1031If the feed does not queue more requests when called, it will be 1363If the feed does not queue more requests when called, it will be
1032automatically removed from the group. 1364automatically removed from the group.
1033 1365
1034If the feed limit is C<0>, it will be set to C<2> automatically. 1366If the feed limit is C<0> when this method is called, it will be set to
1367C<2> automatically.
1035 1368
1036Example: 1369Example:
1037 1370
1038 # stat all files in @files, but only ever use four aio requests concurrently: 1371 # stat all files in @files, but only ever use four aio requests concurrently:
1039 1372
1051Sets the feeder limit for the group: The feeder will be called whenever 1384Sets the feeder limit for the group: The feeder will be called whenever
1052the group contains less than this many requests. 1385the group contains less than this many requests.
1053 1386
1054Setting the limit to C<0> will pause the feeding process. 1387Setting the limit to C<0> will pause the feeding process.
1055 1388
1389The default value for the limit is C<0>, but note that setting a feeder
1390automatically bumps it up to C<2>.
1391
1056=back 1392=back
1057 1393
1058=head2 SUPPORT FUNCTIONS 1394=head2 SUPPORT FUNCTIONS
1059 1395
1060=head3 EVENT PROCESSING AND EVENT LOOP INTEGRATION 1396=head3 EVENT PROCESSING AND EVENT LOOP INTEGRATION
1062=over 4 1398=over 4
1063 1399
1064=item $fileno = IO::AIO::poll_fileno 1400=item $fileno = IO::AIO::poll_fileno
1065 1401
1066Return the I<request result pipe file descriptor>. This filehandle must be 1402Return the I<request result pipe file descriptor>. This filehandle must be
1067polled for reading by some mechanism outside this module (e.g. Event or 1403polled for reading by some mechanism outside this module (e.g. EV, Glib,
1068select, see below or the SYNOPSIS). If the pipe becomes readable you have 1404select and so on, see below or the SYNOPSIS). If the pipe becomes readable
1069to call C<poll_cb> to check the results. 1405you have to call C<poll_cb> to check the results.
1070 1406
1071See C<poll_cb> for an example. 1407See C<poll_cb> for an example.
1072 1408
1073=item IO::AIO::poll_cb 1409=item IO::AIO::poll_cb
1074 1410
1075Process some outstanding events on the result pipe. You have to call this 1411Process some outstanding events on the result pipe. You have to call
1076regularly. Returns the number of events processed. Returns immediately 1412this regularly. Returns C<0> if all events could be processed (or there
1077when no events are outstanding. The amount of events processed depends on 1413were no events to process), or C<-1> if it returned earlier for whatever
1078the settings of C<IO::AIO::max_poll_req> and C<IO::AIO::max_poll_time>. 1414reason. Returns immediately when no events are outstanding. The amount of
1415events processed depends on the settings of C<IO::AIO::max_poll_req> and
1416C<IO::AIO::max_poll_time>.
1079 1417
1080If not all requests were processed for whatever reason, the filehandle 1418If not all requests were processed for whatever reason, the filehandle
1081will still be ready when C<poll_cb> returns. 1419will still be ready when C<poll_cb> returns, so normally you don't have to
1420do anything special to have it called later.
1082 1421
1083Example: Install an Event watcher that automatically calls 1422Example: Install an Event watcher that automatically calls
1084IO::AIO::poll_cb with high priority: 1423IO::AIO::poll_cb with high priority (more examples can be found in the
1424SYNOPSIS section, at the top of this document):
1085 1425
1086 Event->io (fd => IO::AIO::poll_fileno, 1426 Event->io (fd => IO::AIO::poll_fileno,
1087 poll => 'r', async => 1, 1427 poll => 'r', async => 1,
1088 cb => \&IO::AIO::poll_cb); 1428 cb => \&IO::AIO::poll_cb);
1429
1430=item IO::AIO::poll_wait
1431
1432If there are any outstanding requests and none of them in the result
1433phase, wait till the result filehandle becomes ready for reading (simply
1434does a C<select> on the filehandle. This is useful if you want to
1435synchronously wait for some requests to finish).
1436
1437See C<nreqs> for an example.
1438
1439=item IO::AIO::poll
1440
1441Waits until some requests have been handled.
1442
1443Returns the number of requests processed, but is otherwise strictly
1444equivalent to:
1445
1446 IO::AIO::poll_wait, IO::AIO::poll_cb
1447
1448=item IO::AIO::flush
1449
1450Wait till all outstanding AIO requests have been handled.
1451
1452Strictly equivalent to:
1453
1454 IO::AIO::poll_wait, IO::AIO::poll_cb
1455 while IO::AIO::nreqs;
1089 1456
1090=item IO::AIO::max_poll_reqs $nreqs 1457=item IO::AIO::max_poll_reqs $nreqs
1091 1458
1092=item IO::AIO::max_poll_time $seconds 1459=item IO::AIO::max_poll_time $seconds
1093 1460
1118 # use a low priority so other tasks have priority 1485 # use a low priority so other tasks have priority
1119 Event->io (fd => IO::AIO::poll_fileno, 1486 Event->io (fd => IO::AIO::poll_fileno,
1120 poll => 'r', nice => 1, 1487 poll => 'r', nice => 1,
1121 cb => &IO::AIO::poll_cb); 1488 cb => &IO::AIO::poll_cb);
1122 1489
1123=item IO::AIO::poll_wait
1124
1125If there are any outstanding requests and none of them in the result
1126phase, wait till the result filehandle becomes ready for reading (simply
1127does a C<select> on the filehandle. This is useful if you want to
1128synchronously wait for some requests to finish).
1129
1130See C<nreqs> for an example.
1131
1132=item IO::AIO::poll
1133
1134Waits until some requests have been handled.
1135
1136Returns the number of requests processed, but is otherwise strictly
1137equivalent to:
1138
1139 IO::AIO::poll_wait, IO::AIO::poll_cb
1140
1141=item IO::AIO::flush
1142
1143Wait till all outstanding AIO requests have been handled.
1144
1145Strictly equivalent to:
1146
1147 IO::AIO::poll_wait, IO::AIO::poll_cb
1148 while IO::AIO::nreqs;
1149
1150=back 1490=back
1151 1491
1152=head3 CONTROLLING THE NUMBER OF THREADS 1492=head3 CONTROLLING THE NUMBER OF THREADS
1153 1493
1154=over 1494=over
1187 1527
1188Under normal circumstances you don't need to call this function. 1528Under normal circumstances you don't need to call this function.
1189 1529
1190=item IO::AIO::max_idle $nthreads 1530=item IO::AIO::max_idle $nthreads
1191 1531
1192Limit the number of threads (default: 4) that are allowed to idle (i.e., 1532Limit the number of threads (default: 4) that are allowed to idle
1193threads that did not get a request to process within 10 seconds). That 1533(i.e., threads that did not get a request to process within the idle
1194means if a thread becomes idle while C<$nthreads> other threads are also 1534timeout (default: 10 seconds). That means if a thread becomes idle while
1195idle, it will free its resources and exit. 1535C<$nthreads> other threads are also idle, it will free its resources and
1536exit.
1196 1537
1197This is useful when you allow a large number of threads (e.g. 100 or 1000) 1538This is useful when you allow a large number of threads (e.g. 100 or 1000)
1198to allow for extremely high load situations, but want to free resources 1539to allow for extremely high load situations, but want to free resources
1199under normal circumstances (1000 threads can easily consume 30MB of RAM). 1540under normal circumstances (1000 threads can easily consume 30MB of RAM).
1200 1541
1201The default is probably ok in most situations, especially if thread 1542The default is probably ok in most situations, especially if thread
1202creation is fast. If thread creation is very slow on your system you might 1543creation is fast. If thread creation is very slow on your system you might
1203want to use larger values. 1544want to use larger values.
1204 1545
1546=item IO::AIO::idle_timeout $seconds
1547
1548Sets the minimum idle timeout (default 10) after which worker threads are
1549allowed to exit. SEe C<IO::AIO::max_idle>.
1550
1205=item $oldmaxreqs = IO::AIO::max_outstanding $maxreqs 1551=item IO::AIO::max_outstanding $maxreqs
1206 1552
1207This is a very bad function to use in interactive programs because it 1553This is a very bad function to use in interactive programs because it
1208blocks, and a bad way to reduce concurrency because it is inexact: Better 1554blocks, and a bad way to reduce concurrency because it is inexact: Better
1209use an C<aio_group> together with a feed callback. 1555use an C<aio_group> together with a feed callback.
1210 1556
1215 1561
1216The default value is very large, so there is no practical limit on the 1562The default value is very large, so there is no practical limit on the
1217number of outstanding requests. 1563number of outstanding requests.
1218 1564
1219You can still queue as many requests as you want. Therefore, 1565You can still queue as many requests as you want. Therefore,
1220C<max_oustsanding> is mainly useful in simple scripts (with low values) or 1566C<max_outstanding> is mainly useful in simple scripts (with low values) or
1221as a stop gap to shield against fatal memory overflow (with large values). 1567as a stop gap to shield against fatal memory overflow (with large values).
1222 1568
1223=back 1569=back
1224 1570
1225=head3 STATISTICAL INFORMATION 1571=head3 STATISTICAL INFORMATION
1246Returns the number of requests currently in the pending state (executed, 1592Returns the number of requests currently in the pending state (executed,
1247but not yet processed by poll_cb). 1593but not yet processed by poll_cb).
1248 1594
1249=back 1595=back
1250 1596
1597=head3 MISCELLANEOUS FUNCTIONS
1598
1599IO::AIO implements some functions that might be useful, but are not
1600asynchronous.
1601
1602=over 4
1603
1604=item IO::AIO::sendfile $ofh, $ifh, $offset, $count
1605
1606Calls the C<eio_sendfile_sync> function, which is like C<aio_sendfile>,
1607but is blocking (this makes most sense if you know the input data is
1608likely cached already and the output filehandle is set to non-blocking
1609operations).
1610
1611Returns the number of bytes copied, or C<-1> on error.
1612
1613=item IO::AIO::fadvise $fh, $offset, $len, $advice
1614
1615Simply calls the C<posix_fadvise> function (see its
1616manpage for details). The following advice constants are
1617avaiable: C<IO::AIO::FADV_NORMAL>, C<IO::AIO::FADV_SEQUENTIAL>,
1618C<IO::AIO::FADV_RANDOM>, C<IO::AIO::FADV_NOREUSE>,
1619C<IO::AIO::FADV_WILLNEED>, C<IO::AIO::FADV_DONTNEED>.
1620
1621On systems that do not implement C<posix_fadvise>, this function returns
1622ENOSYS, otherwise the return value of C<posix_fadvise>.
1623
1624=item IO::AIO::madvise $scalar, $offset, $len, $advice
1625
1626Simply calls the C<posix_madvise> function (see its
1627manpage for details). The following advice constants are
1628avaiable: C<IO::AIO::MADV_NORMAL>, C<IO::AIO::MADV_SEQUENTIAL>,
1629C<IO::AIO::MADV_RANDOM>, C<IO::AIO::MADV_WILLNEED>, C<IO::AIO::MADV_DONTNEED>.
1630
1631On systems that do not implement C<posix_madvise>, this function returns
1632ENOSYS, otherwise the return value of C<posix_madvise>.
1633
1634=item IO::AIO::mprotect $scalar, $offset, $len, $protect
1635
1636Simply calls the C<mprotect> function on the preferably AIO::mmap'ed
1637$scalar (see its manpage for details). The following protect
1638constants are avaiable: C<IO::AIO::PROT_NONE>, C<IO::AIO::PROT_READ>,
1639C<IO::AIO::PROT_WRITE>, C<IO::AIO::PROT_EXEC>.
1640
1641On systems that do not implement C<mprotect>, this function returns
1642ENOSYS, otherwise the return value of C<mprotect>.
1643
1644=item IO::AIO::mmap $scalar, $length, $prot, $flags, $fh[, $offset]
1645
1646Memory-maps a file (or anonymous memory range) and attaches it to the
1647given C<$scalar>, which will act like a string scalar.
1648
1649The only operations allowed on the scalar are C<substr>/C<vec> that don't
1650change the string length, and most read-only operations such as copying it
1651or searching it with regexes and so on.
1652
1653Anything else is unsafe and will, at best, result in memory leaks.
1654
1655The memory map associated with the C<$scalar> is automatically removed
1656when the C<$scalar> is destroyed, or when the C<IO::AIO::mmap> or
1657C<IO::AIO::munmap> functions are called.
1658
1659This calls the C<mmap>(2) function internally. See your system's manual
1660page for details on the C<$length>, C<$prot> and C<$flags> parameters.
1661
1662The C<$length> must be larger than zero and smaller than the actual
1663filesize.
1664
1665C<$prot> is a combination of C<IO::AIO::PROT_NONE>, C<IO::AIO::PROT_EXEC>,
1666C<IO::AIO::PROT_READ> and/or C<IO::AIO::PROT_WRITE>,
1667
1668C<$flags> can be a combination of C<IO::AIO::MAP_SHARED> or
1669C<IO::AIO::MAP_PRIVATE>, or a number of system-specific flags (when
1670not available, the are defined as 0): C<IO::AIO::MAP_ANONYMOUS>
1671(which is set to C<MAP_ANON> if your system only provides this
1672constant), C<IO::AIO::MAP_HUGETLB>, C<IO::AIO::MAP_LOCKED>,
1673C<IO::AIO::MAP_NORESERVE>, C<IO::AIO::MAP_POPULATE> or
1674C<IO::AIO::MAP_NONBLOCK>
1675
1676If C<$fh> is C<undef>, then a file descriptor of C<-1> is passed.
1677
1678C<$offset> is the offset from the start of the file - it generally must be
1679a multiple of C<IO::AIO::PAGESIZE> and defaults to C<0>.
1680
1681Example:
1682
1683 use Digest::MD5;
1684 use IO::AIO;
1685
1686 open my $fh, "<verybigfile"
1687 or die "$!";
1688
1689 IO::AIO::mmap my $data, -s $fh, IO::AIO::PROT_READ, IO::AIO::MAP_SHARED, $fh
1690 or die "verybigfile: $!";
1691
1692 my $fast_md5 = md5 $data;
1693
1694=item IO::AIO::munmap $scalar
1695
1696Removes a previous mmap and undefines the C<$scalar>.
1697
1698=item IO::AIO::munlock $scalar, $offset = 0, $length = undef
1699
1700Calls the C<munlock> function, undoing the effects of a previous
1701C<aio_mlock> call (see its description for details).
1702
1703=item IO::AIO::munlockall
1704
1705Calls the C<munlockall> function.
1706
1707On systems that do not implement C<munlockall>, this function returns
1708ENOSYS, otherwise the return value of C<munlockall>.
1709
1710=back
1711
1251=cut 1712=cut
1252 1713
1253# support function to convert a fd into a perl filehandle
1254sub _fd2fh {
1255 return undef if $_[0] < 0;
1256
1257 # try to generate nice filehandles
1258 my $sym = "IO::AIO::fd#$_[0]";
1259 local *$sym;
1260
1261 open *$sym, "+<&=$_[0]" # usually works under any unix
1262 or open *$sym, "<&=$_[0]" # cygwin needs this
1263 or open *$sym, ">&=$_[0]" # or this
1264 or return undef;
1265
1266 *$sym
1267}
1268
1269min_parallel 8; 1714min_parallel 8;
1270 1715
1271END { flush } 1716END { flush }
1272 1717
12731; 17181;
1719
1720=head1 EVENT LOOP INTEGRATION
1721
1722It is recommended to use L<AnyEvent::AIO> to integrate IO::AIO
1723automatically into many event loops:
1724
1725 # AnyEvent integration (EV, Event, Glib, Tk, POE, urxvt, pureperl...)
1726 use AnyEvent::AIO;
1727
1728You can also integrate IO::AIO manually into many event loops, here are
1729some examples of how to do this:
1730
1731 # EV integration
1732 my $aio_w = EV::io IO::AIO::poll_fileno, EV::READ, \&IO::AIO::poll_cb;
1733
1734 # Event integration
1735 Event->io (fd => IO::AIO::poll_fileno,
1736 poll => 'r',
1737 cb => \&IO::AIO::poll_cb);
1738
1739 # Glib/Gtk2 integration
1740 add_watch Glib::IO IO::AIO::poll_fileno,
1741 in => sub { IO::AIO::poll_cb; 1 };
1742
1743 # Tk integration
1744 Tk::Event::IO->fileevent (IO::AIO::poll_fileno, "",
1745 readable => \&IO::AIO::poll_cb);
1746
1747 # Danga::Socket integration
1748 Danga::Socket->AddOtherFds (IO::AIO::poll_fileno =>
1749 \&IO::AIO::poll_cb);
1274 1750
1275=head2 FORK BEHAVIOUR 1751=head2 FORK BEHAVIOUR
1276 1752
1277This module should do "the right thing" when the process using it forks: 1753This module should do "the right thing" when the process using it forks:
1278 1754
1311 1787
1312Known bugs will be fixed in the next release. 1788Known bugs will be fixed in the next release.
1313 1789
1314=head1 SEE ALSO 1790=head1 SEE ALSO
1315 1791
1316L<Coro::AIO>. 1792L<AnyEvent::AIO> for easy integration into event loops, L<Coro::AIO> for a
1793more natural syntax.
1317 1794
1318=head1 AUTHOR 1795=head1 AUTHOR
1319 1796
1320 Marc Lehmann <schmorp@schmorp.de> 1797 Marc Lehmann <schmorp@schmorp.de>
1321 http://home.schmorp.de/ 1798 http://home.schmorp.de/

Diff Legend

Removed lines
+ Added lines
< Changed lines
> Changed lines