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 1.81 by root, Fri Oct 27 19:17:23 2006 UTC vs.
Revision 1.156 by root, Tue Jun 16 23:41:59 2009 UTC

… …
5=head1 SYNOPSIS 5=head1 SYNOPSIS
6 6
7 use IO::AIO; 7 use IO::AIO;
8 8
9 aio_open "/etc/passwd", O_RDONLY, 0, sub { 9 aio_open "/etc/passwd", O_RDONLY, 0, sub {
10 my ($fh) = @_; 10 my $fh = shift
11 or die "/etc/passwd: $!";
11 ... 12 ...
12 }; 13 };
13 14
14 aio_unlink "/tmp/file", sub { }; 15 aio_unlink "/tmp/file", sub { };
15 16
… …
25 $req->cancel; # cancel request if still in queue 26 $req->cancel; # cancel request if still in queue
26 27
27 my $grp = aio_group sub { print "all stats done\n" }; 28 my $grp = aio_group sub { print "all stats done\n" };
28 add $grp aio_stat "..." for ...; 29 add $grp aio_stat "..." for ...;
29 30
31 # AnyEvent integration (EV, Event, Glib, Tk, POE, urxvt, pureperl...)
32 use AnyEvent::AIO;
33
30 # AnyEvent integration 34 # EV integration
31 open my $fh, "<&=" . IO::AIO::poll_fileno or die "$!"; 35 my $aio_w = EV::io IO::AIO::poll_fileno, EV::READ, \&IO::AIO::poll_cb;
32 my $w = AnyEvent->io (fh => $fh, poll => 'r', cb => sub { IO::AIO::poll_cb });
33 36
34 # Event integration 37 # Event integration
35 Event->io (fd => IO::AIO::poll_fileno, 38 Event->io (fd => IO::AIO::poll_fileno,
36 poll => 'r', 39 poll => 'r',
37 cb => \&IO::AIO::poll_cb); 40 cb => \&IO::AIO::poll_cb);
… …
49 \&IO::AIO::poll_cb); 52 \&IO::AIO::poll_cb);
50 53
51=head1 DESCRIPTION 54=head1 DESCRIPTION
52 55
53This module implements asynchronous I/O using whatever means your 56This module implements asynchronous I/O using whatever means your
54operating system supports. 57operating system supports. It is implemented as an interface to C<libeio>
58(L<http://software.schmorp.de/pkg/libeio.html>).
59
60Asynchronous means that operations that can normally block your program
61(e.g. reading from disk) will be done asynchronously: the operation
62will still block, but you can do something else in the meantime. This
63is extremely useful for programs that need to stay interactive even
64when doing heavy I/O (GUI programs, high performance network servers
65etc.), but can also be used to easily do operations in parallel that are
66normally done sequentially, e.g. stat'ing many files, which is much faster
67on a RAID volume or over NFS when you do a number of stat operations
68concurrently.
69
70While most of this works on all types of file descriptors (for
71example sockets), using these functions on file descriptors that
72support nonblocking operation (again, sockets, pipes etc.) is
73very inefficient. Use an event loop for that (such as the L<EV>
74module): IO::AIO will naturally fit into such an event loop itself.
55 75
56In this version, a number of threads are started that execute your 76In this version, a number of threads are started that execute your
57requests and signal their completion. You don't need thread support 77requests and signal their completion. You don't need thread support
58in perl, and the threads created by this module will not be visible 78in perl, and the threads created by this module will not be visible
59to perl. In the future, this module might make use of the native aio 79to perl. In the future, this module might make use of the native aio
60functions available on many operating systems. However, they are often 80functions available on many operating systems. However, they are often
61not well-supported or restricted (Linux doesn't allow them on normal 81not well-supported or restricted (GNU/Linux doesn't allow them on normal
62files currently, for example), and they would only support aio_read and 82files currently, for example), and they would only support aio_read and
63aio_write, so the remaining functionality would have to be implemented 83aio_write, so the remaining functionality would have to be implemented
64using threads anyway. 84using threads anyway.
65 85
66Although the module will work with in the presence of other (Perl-) 86Although the module will work in the presence of other (Perl-) threads,
67threads, it is currently not reentrant in any way, so use appropriate 87it is currently not reentrant in any way, so use appropriate locking
68locking yourself, always call C<poll_cb> from within the same thread, or 88yourself, always call C<poll_cb> from within the same thread, or never
69never call C<poll_cb> (or other C<aio_> functions) recursively. 89call C<poll_cb> (or other C<aio_> functions) recursively.
90
91=head2 EXAMPLE
92
93This is a simple example that uses the EV module and loads
94F</etc/passwd> asynchronously:
95
96 use Fcntl;
97 use EV;
98 use IO::AIO;
99
100 # register the IO::AIO callback with EV
101 my $aio_w = EV::io IO::AIO::poll_fileno, EV::READ, \&IO::AIO::poll_cb;
102
103 # queue the request to open /etc/passwd
104 aio_open "/etc/passwd", O_RDONLY, 0, sub {
105 my $fh = shift
106 or die "error while opening: $!";
107
108 # stat'ing filehandles is generally non-blocking
109 my $size = -s $fh;
110
111 # queue a request to read the file
112 my $contents;
113 aio_read $fh, 0, $size, $contents, 0, sub {
114 $_[0] == $size
115 or die "short read: $!";
116
117 close $fh;
118
119 # file contents now in $contents
120 print $contents;
121
122 # exit event loop and program
123 EV::unloop;
124 };
125 };
126
127 # possibly queue up other requests, or open GUI windows,
128 # check for sockets etc. etc.
129
130 # process events as long as there are some:
131 EV::loop;
70 132
71=head1 REQUEST ANATOMY AND LIFETIME 133=head1 REQUEST ANATOMY AND LIFETIME
72 134
73Every C<aio_*> function creates a request. which is a C data structure not 135Every C<aio_*> function creates a request. which is a C data structure not
74directly visible to Perl. 136directly visible to Perl.
… …
116Request has reached the end of its lifetime and holds no resources anymore 178Request has reached the end of its lifetime and holds no resources anymore
117(except possibly for the Perl object, but its connection to the actual 179(except possibly for the Perl object, but its connection to the actual
118aio request is severed and calling its methods will either do nothing or 180aio request is severed and calling its methods will either do nothing or
119result in a runtime error). 181result in a runtime error).
120 182
183=back
184
121=cut 185=cut
122 186
123package IO::AIO; 187package IO::AIO;
188
189use Carp ();
124 190
125no warnings; 191no warnings;
126use strict 'vars'; 192use strict 'vars';
127 193
128use base 'Exporter'; 194use base 'Exporter';
129 195
130BEGIN { 196BEGIN {
131 our $VERSION = '2.0'; 197 our $VERSION = '3.23';
132 198
133 our @AIO_REQ = qw(aio_sendfile aio_read aio_write aio_open aio_close aio_stat 199 our @AIO_REQ = qw(aio_sendfile aio_read aio_write aio_open aio_close
134 aio_lstat aio_unlink aio_rmdir aio_readdir aio_scandir aio_symlink 200 aio_stat aio_lstat aio_unlink aio_rmdir aio_readdir aio_readdirx
135 aio_fsync aio_fdatasync aio_readahead aio_rename aio_link aio_move 201 aio_scandir aio_symlink aio_readlink aio_sync aio_fsync
136 aio_group aio_nop); 202 aio_fdatasync aio_sync_file_range aio_pathsync aio_readahead
203 aio_rename aio_link aio_move aio_copy aio_group
204 aio_nop aio_mknod aio_load aio_rmtree aio_mkdir aio_chown
205 aio_chmod aio_utime aio_truncate);
206
137 our @EXPORT = (@AIO_REQ, qw(aioreq_pri aioreq_nice)); 207 our @EXPORT = (@AIO_REQ, qw(aioreq_pri aioreq_nice));
138 our @EXPORT_OK = qw(poll_fileno poll_cb poll_wait flush 208 our @EXPORT_OK = qw(poll_fileno poll_cb poll_wait flush
139 min_parallel max_parallel nreqs nready npending); 209 min_parallel max_parallel max_idle
210 nreqs nready npending nthreads
211 max_poll_time max_poll_reqs);
212
213 push @AIO_REQ, qw(aio_busy); # not exported
140 214
141 @IO::AIO::GRP::ISA = 'IO::AIO::REQ'; 215 @IO::AIO::GRP::ISA = 'IO::AIO::REQ';
142 216
143 require XSLoader; 217 require XSLoader;
144 XSLoader::load ("IO::AIO", $VERSION); 218 XSLoader::load ("IO::AIO", $VERSION);
145} 219}
146 220
147=head1 FUNCTIONS 221=head1 FUNCTIONS
148 222
149=head2 AIO FUNCTIONS 223=head2 AIO REQUEST FUNCTIONS
150 224
151All the C<aio_*> calls are more or less thin wrappers around the syscall 225All the C<aio_*> calls are more or less thin wrappers around the syscall
152with the same name (sans C<aio_>). The arguments are similar or identical, 226with the same name (sans C<aio_>). The arguments are similar or identical,
153and they all accept an additional (and optional) C<$callback> argument 227and they all accept an additional (and optional) C<$callback> argument
154which must be a code reference. This code reference will get called with 228which must be a code reference. This code reference will get called with
155the syscall return code (e.g. most syscalls return C<-1> on error, unlike 229the syscall return code (e.g. most syscalls return C<-1> on error, unlike
156perl, which usually delivers "false") as it's sole argument when the given 230perl, which usually delivers "false") as its sole argument after the given
157syscall has been executed asynchronously. 231syscall has been executed asynchronously.
158 232
159All functions expecting a filehandle keep a copy of the filehandle 233All functions expecting a filehandle keep a copy of the filehandle
160internally until the request has finished. 234internally until the request has finished.
161 235
162All requests return objects of type L<IO::AIO::REQ> that allow further 236All functions return request objects of type L<IO::AIO::REQ> that allow
163manipulation of those requests while they are in-flight. 237further manipulation of those requests while they are in-flight.
164 238
165The pathnames you pass to these routines I<must> be absolute and 239The pathnames you pass to these routines I<must> be absolute and
166encoded in byte form. The reason for the former is that at the time the 240encoded as octets. The reason for the former is that at the time the
167request is being executed, the current working directory could have 241request is being executed, the current working directory could have
168changed. Alternatively, you can make sure that you never change the 242changed. Alternatively, you can make sure that you never change the
169current working directory. 243current working directory anywhere in the program and then use relative
244paths.
170 245
171To encode pathnames to byte form, either make sure you either: a) 246To encode pathnames as octets, either make sure you either: a) always pass
172always pass in filenames you got from outside (command line, readdir 247in filenames you got from outside (command line, readdir etc.) without
173etc.), b) are ASCII or ISO 8859-1, c) use the Encode module and encode 248tinkering, b) are ASCII or ISO 8859-1, c) use the Encode module and encode
174your pathnames to the locale (or other) encoding in effect in the user 249your pathnames to the locale (or other) encoding in effect in the user
175environment, d) use Glib::filename_from_unicode on unicode filenames or e) 250environment, d) use Glib::filename_from_unicode on unicode filenames or e)
176use something else. 251use something else to ensure your scalar has the correct contents.
252
253This works, btw. independent of the internal UTF-8 bit, which IO::AIO
254handles correctly whether it is set or not.
177 255
178=over 4 256=over 4
179 257
180=item $prev_pri = aioreq_pri [$pri] 258=item $prev_pri = aioreq_pri [$pri]
181 259
… …
201 aio_read $_[0], ..., sub { 279 aio_read $_[0], ..., sub {
202 ... 280 ...
203 }; 281 };
204 }; 282 };
205 283
284
206=item aioreq_nice $pri_adjust 285=item aioreq_nice $pri_adjust
207 286
208Similar to C<aioreq_pri>, but subtracts the given value from the current 287Similar to C<aioreq_pri>, but subtracts the given value from the current
209priority, so effects are cumulative. 288priority, so the effect is cumulative.
289
210 290
211=item aio_open $pathname, $flags, $mode, $callback->($fh) 291=item aio_open $pathname, $flags, $mode, $callback->($fh)
212 292
213Asynchronously open or create a file and call the callback with a newly 293Asynchronously open or create a file and call the callback with a newly
214created filehandle for the file. 294created filehandle for the file.
… …
220list. They are the same as used by C<sysopen>. 300list. They are the same as used by C<sysopen>.
221 301
222Likewise, C<$mode> specifies the mode of the newly created file, if it 302Likewise, C<$mode> specifies the mode of the newly created file, if it
223didn't exist and C<O_CREAT> has been given, just like perl's C<sysopen>, 303didn't exist and C<O_CREAT> has been given, just like perl's C<sysopen>,
224except that it is mandatory (i.e. use C<0> if you don't create new files, 304except that it is mandatory (i.e. use C<0> if you don't create new files,
225and C<0666> or C<0777> if you do). 305and C<0666> or C<0777> if you do). Note that the C<$mode> will be modified
306by the umask in effect then the request is being executed, so better never
307change the umask.
226 308
227Example: 309Example:
228 310
229 aio_open "/etc/passwd", O_RDONLY, 0, sub { 311 aio_open "/etc/passwd", O_RDONLY, 0, sub {
230 if ($_[0]) { 312 if ($_[0]) {
… …
233 } else { 315 } else {
234 die "open failed: $!\n"; 316 die "open failed: $!\n";
235 } 317 }
236 }; 318 };
237 319
320
238=item aio_close $fh, $callback->($status) 321=item aio_close $fh, $callback->($status)
239 322
240Asynchronously close a file and call the callback with the result 323Asynchronously close a file and call the callback with the result
241code. I<WARNING:> although accepted, you should not pass in a perl 324code.
242filehandle here, as perl will likely close the file descriptor another
243time when the filehandle is destroyed. Normally, you can safely call perls
244C<close> or just let filehandles go out of scope.
245 325
246This is supposed to be a bug in the API, so that might change. It's 326Unfortunately, you can't do this to perl. Perl I<insists> very strongly on
247therefore best to avoid this function. 327closing the file descriptor associated with the filehandle itself.
328
329Therefore, C<aio_close> will not close the filehandle - instead it will
330use dup2 to overwrite the file descriptor with the write-end of a pipe
331(the pipe fd will be created on demand and will be cached).
332
333Or in other words: the file descriptor will be closed, but it will not be
334free for reuse until the perl filehandle is closed.
335
336=cut
248 337
249=item aio_read $fh,$offset,$length, $data,$dataoffset, $callback->($retval) 338=item aio_read $fh,$offset,$length, $data,$dataoffset, $callback->($retval)
250 339
251=item aio_write $fh,$offset,$length, $data,$dataoffset, $callback->($retval) 340=item aio_write $fh,$offset,$length, $data,$dataoffset, $callback->($retval)
252 341
253Reads or writes C<length> bytes from the specified C<fh> and C<offset> 342Reads or writes C<$length> bytes from or to the specified C<$fh> and
254into the scalar given by C<data> and offset C<dataoffset> and calls the 343C<$offset> into the scalar given by C<$data> and offset C<$dataoffset>
255callback without the actual number of bytes read (or -1 on error, just 344and calls the callback without the actual number of bytes read (or -1 on
256like the syscall). 345error, just like the syscall).
346
347C<aio_read> will, like C<sysread>, shrink or grow the C<$data> scalar to
348offset plus the actual number of bytes read.
349
350If C<$offset> is undefined, then the current file descriptor offset will
351be used (and updated), otherwise the file descriptor offset will not be
352changed by these calls.
353
354If C<$length> is undefined in C<aio_write>, use the remaining length of
355C<$data>.
356
357If C<$dataoffset> is less than zero, it will be counted from the end of
358C<$data>.
257 359
258The C<$data> scalar I<MUST NOT> be modified in any way while the request 360The C<$data> scalar I<MUST NOT> be modified in any way while the request
259is outstanding. Modifying it can result in segfaults or WW3 (if the 361is outstanding. Modifying it can result in segfaults or World War III (if
260necessary/optional hardware is installed). 362the necessary/optional hardware is installed).
261 363
262Example: Read 15 bytes at offset 7 into scalar C<$buffer>, starting at 364Example: Read 15 bytes at offset 7 into scalar C<$buffer>, starting at
263offset C<0> within the scalar: 365offset C<0> within the scalar:
264 366
265 aio_read $fh, 7, 15, $buffer, 0, sub { 367 aio_read $fh, 7, 15, $buffer, 0, sub {
266 $_[0] > 0 or die "read error: $!"; 368 $_[0] > 0 or die "read error: $!";
267 print "read $_[0] bytes: <$buffer>\n"; 369 print "read $_[0] bytes: <$buffer>\n";
268 }; 370 };
269 371
270=item aio_move $srcpath, $dstpath, $callback->($status)
271
272Try to move the I<file> (directories not supported as either source or
273destination) from C<$srcpath> to C<$dstpath> and call the callback with
274the C<0> (error) or C<-1> ok.
275
276This is a composite request that tries to rename(2) the file first. If
277rename files with C<EXDEV>, it creates the destination file with mode 0200
278and copies the contents of the source file into it using C<aio_sendfile>,
279followed by restoring atime, mtime, access mode and uid/gid, in that
280order, and unlinking the C<$srcpath>.
281
282If an error occurs, the partial destination file will be unlinked, if
283possible, except when setting atime, mtime, access mode and uid/gid, where
284errors are being ignored.
285
286=cut
287
288sub aio_move($$$) {
289 my ($src, $dst, $cb) = @_;
290
291 my $pri = aioreq_pri;
292 my $grp = aio_group $cb;
293
294 aioreq_pri $pri;
295 add $grp aio_rename $src, $dst, sub {
296 if ($_[0] && $! == EXDEV) {
297 aioreq_pri $pri;
298 add $grp aio_open $src, O_RDONLY, 0, sub {
299 if (my $src_fh = $_[0]) {
300 my @stat = stat $src_fh;
301
302 aioreq_pri $pri;
303 add $grp aio_open $dst, O_WRONLY, 0200, sub {
304 if (my $dst_fh = $_[0]) {
305 aioreq_pri $pri;
306 add $grp aio_sendfile $dst_fh, $src_fh, 0, $stat[7], sub {
307 close $src_fh;
308
309 if ($_[0] == $stat[7]) {
310 utime $stat[8], $stat[9], $dst;
311 chmod $stat[2] & 07777, $dst_fh;
312 chown $stat[4], $stat[5], $dst_fh;
313 close $dst_fh;
314
315 aioreq_pri $pri;
316 add $grp aio_unlink $src, sub {
317 $grp->result ($_[0]);
318 };
319 } else {
320 my $errno = $!;
321 aioreq_pri $pri;
322 add $grp aio_unlink $dst, sub {
323 $! = $errno;
324 $grp->result (-1);
325 };
326 }
327 };
328 } else {
329 $grp->result (-1);
330 }
331 },
332
333 } else {
334 $grp->result (-1);
335 }
336 };
337 } else {
338 $grp->result ($_[0]);
339 }
340 };
341
342 $grp
343}
344 372
345=item aio_sendfile $out_fh, $in_fh, $in_offset, $length, $callback->($retval) 373=item aio_sendfile $out_fh, $in_fh, $in_offset, $length, $callback->($retval)
346 374
347Tries to copy C<$length> bytes from C<$in_fh> to C<$out_fh>. It starts 375Tries to copy C<$length> bytes from C<$in_fh> to C<$out_fh>. It starts
348reading at byte offset C<$in_offset>, and starts writing at the current 376reading at byte offset C<$in_offset>, and starts writing at the current
… …
362C<$in_fh> than are written, and there is no way to find out how many 390C<$in_fh> than are written, and there is no way to find out how many
363bytes have been read from C<aio_sendfile> alone, as C<aio_sendfile> only 391bytes have been read from C<aio_sendfile> alone, as C<aio_sendfile> only
364provides the number of bytes written to C<$out_fh>. Only if the result 392provides the number of bytes written to C<$out_fh>. Only if the result
365value equals C<$length> one can assume that C<$length> bytes have been 393value equals C<$length> one can assume that C<$length> bytes have been
366read. 394read.
395
367 396
368=item aio_readahead $fh,$offset,$length, $callback->($retval) 397=item aio_readahead $fh,$offset,$length, $callback->($retval)
369 398
370C<aio_readahead> populates the page cache with data from a file so that 399C<aio_readahead> populates the page cache with data from a file so that
371subsequent reads from that file will not block on disk I/O. The C<$offset> 400subsequent reads from that file will not block on disk I/O. The C<$offset>
… …
377file. The current file offset of the file is left unchanged. 406file. The current file offset of the file is left unchanged.
378 407
379If that syscall doesn't exist (likely if your OS isn't Linux) it will be 408If that syscall doesn't exist (likely if your OS isn't Linux) it will be
380emulated by simply reading the data, which would have a similar effect. 409emulated by simply reading the data, which would have a similar effect.
381 410
411
382=item aio_stat $fh_or_path, $callback->($status) 412=item aio_stat $fh_or_path, $callback->($status)
383 413
384=item aio_lstat $fh, $callback->($status) 414=item aio_lstat $fh, $callback->($status)
385 415
386Works like perl's C<stat> or C<lstat> in void context. The callback will 416Works like perl's C<stat> or C<lstat> in void context. The callback will
… …
399 aio_stat "/etc/passwd", sub { 429 aio_stat "/etc/passwd", sub {
400 $_[0] and die "stat failed: $!"; 430 $_[0] and die "stat failed: $!";
401 print "size is ", -s _, "\n"; 431 print "size is ", -s _, "\n";
402 }; 432 };
403 433
434
435=item aio_utime $fh_or_path, $atime, $mtime, $callback->($status)
436
437Works like perl's C<utime> function (including the special case of $atime
438and $mtime being undef). Fractional times are supported if the underlying
439syscalls support them.
440
441When called with a pathname, uses utimes(2) if available, otherwise
442utime(2). If called on a file descriptor, uses futimes(2) if available,
443otherwise returns ENOSYS, so this is not portable.
444
445Examples:
446
447 # set atime and mtime to current time (basically touch(1)):
448 aio_utime "path", undef, undef;
449 # set atime to current time and mtime to beginning of the epoch:
450 aio_utime "path", time, undef; # undef==0
451
452
453=item aio_chown $fh_or_path, $uid, $gid, $callback->($status)
454
455Works like perl's C<chown> function, except that C<undef> for either $uid
456or $gid is being interpreted as "do not change" (but -1 can also be used).
457
458Examples:
459
460 # same as "chown root path" in the shell:
461 aio_chown "path", 0, -1;
462 # same as above:
463 aio_chown "path", 0, undef;
464
465
466=item aio_truncate $fh_or_path, $offset, $callback->($status)
467
468Works like truncate(2) or ftruncate(2).
469
470
471=item aio_chmod $fh_or_path, $mode, $callback->($status)
472
473Works like perl's C<chmod> function.
474
475
404=item aio_unlink $pathname, $callback->($status) 476=item aio_unlink $pathname, $callback->($status)
405 477
406Asynchronously unlink (delete) a file and call the callback with the 478Asynchronously unlink (delete) a file and call the callback with the
407result code. 479result code.
408 480
481
482=item aio_mknod $path, $mode, $dev, $callback->($status)
483
484[EXPERIMENTAL]
485
486Asynchronously create a device node (or fifo). See mknod(2).
487
488The only (POSIX-) portable way of calling this function is:
489
490 aio_mknod $path, IO::AIO::S_IFIFO | $mode, 0, sub { ...
491
492
409=item aio_link $srcpath, $dstpath, $callback->($status) 493=item aio_link $srcpath, $dstpath, $callback->($status)
410 494
411Asynchronously create a new link to the existing object at C<$srcpath> at 495Asynchronously create a new link to the existing object at C<$srcpath> at
412the path C<$dstpath> and call the callback with the result code. 496the path C<$dstpath> and call the callback with the result code.
413 497
498
414=item aio_symlink $srcpath, $dstpath, $callback->($status) 499=item aio_symlink $srcpath, $dstpath, $callback->($status)
415 500
416Asynchronously create a new symbolic link to the existing object at C<$srcpath> at 501Asynchronously create a new symbolic link to the existing object at C<$srcpath> at
417the path C<$dstpath> and call the callback with the result code. 502the path C<$dstpath> and call the callback with the result code.
418 503
504
505=item aio_readlink $path, $callback->($link)
506
507Asynchronously read the symlink specified by C<$path> and pass it to
508the callback. If an error occurs, nothing or undef gets passed to the
509callback.
510
511
419=item aio_rename $srcpath, $dstpath, $callback->($status) 512=item aio_rename $srcpath, $dstpath, $callback->($status)
420 513
421Asynchronously rename the object at C<$srcpath> to C<$dstpath>, just as 514Asynchronously rename the object at C<$srcpath> to C<$dstpath>, just as
422rename(2) and call the callback with the result code. 515rename(2) and call the callback with the result code.
423 516
517
518=item aio_mkdir $pathname, $mode, $callback->($status)
519
520Asynchronously mkdir (create) a directory and call the callback with
521the result code. C<$mode> will be modified by the umask at the time the
522request is executed, so do not change your umask.
523
524
424=item aio_rmdir $pathname, $callback->($status) 525=item aio_rmdir $pathname, $callback->($status)
425 526
426Asynchronously rmdir (delete) a directory and call the callback with the 527Asynchronously rmdir (delete) a directory and call the callback with the
427result code. 528result code.
529
428 530
429=item aio_readdir $pathname, $callback->($entries) 531=item aio_readdir $pathname, $callback->($entries)
430 532
431Unlike the POSIX call of the same name, C<aio_readdir> reads an entire 533Unlike the POSIX call of the same name, C<aio_readdir> reads an entire
432directory (i.e. opendir + readdir + closedir). The entries will not be 534directory (i.e. opendir + readdir + closedir). The entries will not be
433sorted, and will B<NOT> include the C<.> and C<..> entries. 535sorted, and will B<NOT> include the C<.> and C<..> entries.
434 536
435The callback a single argument which is either C<undef> or an array-ref 537The callback is passed a single argument which is either C<undef> or an
436with the filenames. 538array-ref with the filenames.
539
540
541=item aio_readdirx $pathname, $flags, $callback->($entries, $flags)
542
543Quite similar to C<aio_readdir>, but the C<$flags> argument allows to tune
544behaviour and output format. In case of an error, C<$entries> will be
545C<undef>.
546
547The flags are a combination of the following constants, ORed together (the
548flags will also be passed to the callback, possibly modified):
549
550=over 4
551
552=item IO::AIO::READDIR_DENTS
553
554When this flag is off, then the callback gets an arrayref with of names
555only (as with C<aio_readdir>), otherwise it gets an arrayref with
556C<[$name, $type, $inode]> arrayrefs, each describing a single directory
557entry in more detail.
558
559C<$name> is the name of the entry.
560
561C<$type> is one of the C<IO::AIO::DT_xxx> constants:
562
563C<IO::AIO::DT_UNKNOWN>, C<IO::AIO::DT_FIFO>, C<IO::AIO::DT_CHR>, C<IO::AIO::DT_DIR>,
564C<IO::AIO::DT_BLK>, C<IO::AIO::DT_REG>, C<IO::AIO::DT_LNK>, C<IO::AIO::DT_SOCK>,
565C<IO::AIO::DT_WHT>.
566
567C<IO::AIO::DT_UNKNOWN> means just that: readdir does not know. If you need to
568know, you have to run stat yourself. Also, for speed reasons, the C<$type>
569scalars are read-only: you can not modify them.
570
571C<$inode> is the inode number (which might not be exact on systems with 64
572bit inode numbers and 32 bit perls). This field has unspecified content on
573systems that do not deliver the inode information.
574
575=item IO::AIO::READDIR_DIRS_FIRST
576
577When this flag is set, then the names will be returned in an order where
578likely directories come first. This is useful when you need to quickly
579find directories, or you want to find all directories while avoiding to
580stat() each entry.
581
582If the system returns type information in readdir, then this is used
583to find directories directly. Otherwise, likely directories are files
584beginning with ".", or otherwise files with no dots, of which files with
585short names are tried first.
586
587=item IO::AIO::READDIR_STAT_ORDER
588
589When this flag is set, then the names will be returned in an order
590suitable for stat()'ing each one. That is, when you plan to stat()
591all files in the given directory, then the returned order will likely
592be fastest.
593
594If both this flag and C<IO::AIO::READDIR_DIRS_FIRST> are specified, then
595the likely dirs come first, resulting in a less optimal stat order.
596
597=item IO::AIO::READDIR_FOUND_UNKNOWN
598
599This flag should not be set when calling C<aio_readdirx>. Instead, it
600is being set by C<aio_readdirx>, when any of the C<$type>'s found were
601C<IO::AIO::DT_UNKNOWN>. The absense of this flag therefore indicates that all
602C<$type>'s are known, which can be used to speed up some algorithms.
603
604=back
605
606
607=item aio_load $path, $data, $callback->($status)
608
609This is a composite request that tries to fully load the given file into
610memory. Status is the same as with aio_read.
611
612=cut
613
614sub aio_load($$;$) {
615 my ($path, undef, $cb) = @_;
616 my $data = \$_[1];
617
618 my $pri = aioreq_pri;
619 my $grp = aio_group $cb;
620
621 aioreq_pri $pri;
622 add $grp aio_open $path, O_RDONLY, 0, sub {
623 my $fh = shift
624 or return $grp->result (-1);
625
626 aioreq_pri $pri;
627 add $grp aio_read $fh, 0, (-s $fh), $$data, 0, sub {
628 $grp->result ($_[0]);
629 };
630 };
631
632 $grp
633}
634
635=item aio_copy $srcpath, $dstpath, $callback->($status)
636
637Try to copy the I<file> (directories not supported as either source or
638destination) from C<$srcpath> to C<$dstpath> and call the callback with
639the C<0> (error) or C<-1> ok.
640
641This is a composite request that creates the destination file with
642mode 0200 and copies the contents of the source file into it using
643C<aio_sendfile>, followed by restoring atime, mtime, access mode and
644uid/gid, in that order.
645
646If an error occurs, the partial destination file will be unlinked, if
647possible, except when setting atime, mtime, access mode and uid/gid, where
648errors are being ignored.
649
650=cut
651
652sub aio_copy($$;$) {
653 my ($src, $dst, $cb) = @_;
654
655 my $pri = aioreq_pri;
656 my $grp = aio_group $cb;
657
658 aioreq_pri $pri;
659 add $grp aio_open $src, O_RDONLY, 0, sub {
660 if (my $src_fh = $_[0]) {
661 my @stat = stat $src_fh; # hmm, might bock over nfs?
662
663 aioreq_pri $pri;
664 add $grp aio_open $dst, O_CREAT | O_WRONLY | O_TRUNC, 0200, sub {
665 if (my $dst_fh = $_[0]) {
666 aioreq_pri $pri;
667 add $grp aio_sendfile $dst_fh, $src_fh, 0, $stat[7], sub {
668 if ($_[0] == $stat[7]) {
669 $grp->result (0);
670 close $src_fh;
671
672 my $ch = sub {
673 aioreq_pri $pri;
674 add $grp aio_chmod $dst_fh, $stat[2] & 07777, sub {
675 aioreq_pri $pri;
676 add $grp aio_chown $dst_fh, $stat[4], $stat[5], sub {
677 aioreq_pri $pri;
678 add $grp aio_close $dst_fh;
679 }
680 };
681 };
682
683 aioreq_pri $pri;
684 add $grp aio_utime $dst_fh, $stat[8], $stat[9], sub {
685 if ($_[0] < 0 && $! == ENOSYS) {
686 aioreq_pri $pri;
687 add $grp aio_utime $dst, $stat[8], $stat[9], $ch;
688 } else {
689 $ch->();
690 }
691 };
692 } else {
693 $grp->result (-1);
694 close $src_fh;
695 close $dst_fh;
696
697 aioreq $pri;
698 add $grp aio_unlink $dst;
699 }
700 };
701 } else {
702 $grp->result (-1);
703 }
704 },
705
706 } else {
707 $grp->result (-1);
708 }
709 };
710
711 $grp
712}
713
714=item aio_move $srcpath, $dstpath, $callback->($status)
715
716Try to move the I<file> (directories not supported as either source or
717destination) from C<$srcpath> to C<$dstpath> and call the callback with
718the C<0> (error) or C<-1> ok.
719
720This is a composite request that tries to rename(2) the file first; if
721rename fails with C<EXDEV>, it copies the file with C<aio_copy> and, if
722that is successful, unlinks the C<$srcpath>.
723
724=cut
725
726sub aio_move($$;$) {
727 my ($src, $dst, $cb) = @_;
728
729 my $pri = aioreq_pri;
730 my $grp = aio_group $cb;
731
732 aioreq_pri $pri;
733 add $grp aio_rename $src, $dst, sub {
734 if ($_[0] && $! == EXDEV) {
735 aioreq_pri $pri;
736 add $grp aio_copy $src, $dst, sub {
737 $grp->result ($_[0]);
738
739 if (!$_[0]) {
740 aioreq_pri $pri;
741 add $grp aio_unlink $src;
742 }
743 };
744 } else {
745 $grp->result ($_[0]);
746 }
747 };
748
749 $grp
750}
437 751
438=item aio_scandir $path, $maxreq, $callback->($dirs, $nondirs) 752=item aio_scandir $path, $maxreq, $callback->($dirs, $nondirs)
439 753
440Scans a directory (similar to C<aio_readdir>) but additionally tries to 754Scans a directory (similar to C<aio_readdir>) but additionally tries to
441efficiently separate the entries of directory C<$path> into two sets of 755efficiently separate the entries of directory C<$path> into two sets of
… …
460 774
461Implementation notes. 775Implementation notes.
462 776
463The C<aio_readdir> cannot be avoided, but C<stat()>'ing every entry can. 777The C<aio_readdir> cannot be avoided, but C<stat()>'ing every entry can.
464 778
779If readdir returns file type information, then this is used directly to
780find directories.
781
465After reading the directory, the modification time, size etc. of the 782Otherwise, after reading the directory, the modification time, size etc.
466directory before and after the readdir is checked, and if they match (and 783of the directory before and after the readdir is checked, and if they
467isn't the current time), the link count will be used to decide how many 784match (and isn't the current time), the link count will be used to decide
468entries are directories (if >= 2). Otherwise, no knowledge of the number 785how many entries are directories (if >= 2). Otherwise, no knowledge of the
469of subdirectories will be assumed. 786number of subdirectories will be assumed.
470 787
471Then entries will be sorted into likely directories (everything without 788Then entries will be sorted into likely directories a non-initial dot
472a non-initial dot currently) and likely non-directories (everything 789currently) and likely non-directories (see C<aio_readdirx>). Then every
473else). Then every entry plus an appended C</.> will be C<stat>'ed, 790entry plus an appended C</.> will be C<stat>'ed, likely directories first,
474likely directories first. If that succeeds, it assumes that the entry 791in order of their inode numbers. If that succeeds, it assumes that the
475is a directory or a symlink to directory (which will be checked 792entry is a directory or a symlink to directory (which will be checked
476seperately). This is often faster than stat'ing the entry itself because 793seperately). This is often faster than stat'ing the entry itself because
477filesystems might detect the type of the entry without reading the inode 794filesystems might detect the type of the entry without reading the inode
478data (e.g. ext2fs filetype feature). 795data (e.g. ext2fs filetype feature), even on systems that cannot return
796the filetype information on readdir.
479 797
480If the known number of directories (link count - 2) has been reached, the 798If the known number of directories (link count - 2) has been reached, the
481rest of the entries is assumed to be non-directories. 799rest of the entries is assumed to be non-directories.
482 800
483This only works with certainty on POSIX (= UNIX) filesystems, which 801This only works with certainty on POSIX (= UNIX) filesystems, which
… …
487as those tend to return 0 or 1 as link counts, which disables the 805as those tend to return 0 or 1 as link counts, which disables the
488directory counting heuristic. 806directory counting heuristic.
489 807
490=cut 808=cut
491 809
492sub aio_scandir($$$) { 810sub aio_scandir($$;$) {
493 my ($path, $maxreq, $cb) = @_; 811 my ($path, $maxreq, $cb) = @_;
494 812
495 my $pri = aioreq_pri; 813 my $pri = aioreq_pri;
496 814
497 my $grp = aio_group $cb; 815 my $grp = aio_group $cb;
… …
505 my $now = time; 823 my $now = time;
506 my $hash1 = join ":", (stat _)[0,1,3,7,9]; 824 my $hash1 = join ":", (stat _)[0,1,3,7,9];
507 825
508 # read the directory entries 826 # read the directory entries
509 aioreq_pri $pri; 827 aioreq_pri $pri;
510 add $grp aio_readdir $path, sub { 828 add $grp aio_readdirx $path, READDIR_DIRS_FIRST, sub {
511 my $entries = shift 829 my $entries = shift
512 or return $grp->result (); 830 or return $grp->result ();
513 831
514 # stat the dir another time 832 # stat the dir another time
515 aioreq_pri $pri; 833 aioreq_pri $pri;
… …
521 # take the slow route if anything looks fishy 839 # take the slow route if anything looks fishy
522 if ($hash1 ne $hash2 or (stat _)[9] == $now) { 840 if ($hash1 ne $hash2 or (stat _)[9] == $now) {
523 $ndirs = -1; 841 $ndirs = -1;
524 } else { 842 } else {
525 # if nlink == 2, we are finished 843 # if nlink == 2, we are finished
526 # on non-posix-fs's, we rely on nlink < 2 844 # for non-posix-fs's, we rely on nlink < 2
527 $ndirs = (stat _)[3] - 2 845 $ndirs = (stat _)[3] - 2
528 or return $grp->result ([], $entries); 846 or return $grp->result ([], $entries);
529 } 847 }
530 848
531 # sort into likely dirs and likely nondirs
532 # dirs == files without ".", short entries first
533 $entries = [map $_->[0],
534 sort { $b->[1] cmp $a->[1] }
535 map [$_, sprintf "%s%04d", (/.\./ ? "1" : "0"), length],
536 @$entries];
537
538 my (@dirs, @nondirs); 849 my (@dirs, @nondirs);
539 850
540 my $statgrp = add $grp aio_group sub { 851 my $statgrp = add $grp aio_group sub {
541 $grp->result (\@dirs, \@nondirs); 852 $grp->result (\@dirs, \@nondirs);
542 }; 853 };
543 854
544 limit $statgrp $maxreq; 855 limit $statgrp $maxreq;
545 feed $statgrp sub { 856 feed $statgrp sub {
546 return unless @$entries; 857 return unless @$entries;
547 my $entry = pop @$entries; 858 my $entry = shift @$entries;
548 859
549 aioreq_pri $pri; 860 aioreq_pri $pri;
550 add $statgrp aio_stat "$path/$entry/.", sub { 861 add $statgrp aio_stat "$path/$entry/.", sub {
551 if ($_[0] < 0) { 862 if ($_[0] < 0) {
552 push @nondirs, $entry; 863 push @nondirs, $entry;
… …
573 }; 884 };
574 885
575 $grp 886 $grp
576} 887}
577 888
889=item aio_rmtree $path, $callback->($status)
890
891Delete a directory tree starting (and including) C<$path>, return the
892status of the final C<rmdir> only. This is a composite request that
893uses C<aio_scandir> to recurse into and rmdir directories, and unlink
894everything else.
895
896=cut
897
898sub aio_rmtree;
899sub aio_rmtree($;$) {
900 my ($path, $cb) = @_;
901
902 my $pri = aioreq_pri;
903 my $grp = aio_group $cb;
904
905 aioreq_pri $pri;
906 add $grp aio_scandir $path, 0, sub {
907 my ($dirs, $nondirs) = @_;
908
909 my $dirgrp = aio_group sub {
910 add $grp aio_rmdir $path, sub {
911 $grp->result ($_[0]);
912 };
913 };
914
915 (aioreq_pri $pri), add $dirgrp aio_rmtree "$path/$_" for @$dirs;
916 (aioreq_pri $pri), add $dirgrp aio_unlink "$path/$_" for @$nondirs;
917
918 add $grp $dirgrp;
919 };
920
921 $grp
922}
923
924=item aio_sync $callback->($status)
925
926Asynchronously call sync and call the callback when finished.
927
578=item aio_fsync $fh, $callback->($status) 928=item aio_fsync $fh, $callback->($status)
579 929
580Asynchronously call fsync on the given filehandle and call the callback 930Asynchronously call fsync on the given filehandle and call the callback
581with the fsync result code. 931with the fsync result code.
582 932
… …
585Asynchronously call fdatasync on the given filehandle and call the 935Asynchronously call fdatasync on the given filehandle and call the
586callback with the fdatasync result code. 936callback with the fdatasync result code.
587 937
588If this call isn't available because your OS lacks it or it couldn't be 938If this call isn't available because your OS lacks it or it couldn't be
589detected, it will be emulated by calling C<fsync> instead. 939detected, it will be emulated by calling C<fsync> instead.
940
941=item aio_sync_file_range $fh, $offset, $nbytes, $flags, $callback->($status)
942
943Sync the data portion of the file specified by C<$offset> and C<$length>
944to disk (but NOT the metadata), by calling the Linux-specific
945sync_file_range call. If sync_file_range is not available or it returns
946ENOSYS, then fdatasync or fsync is being substituted.
947
948C<$flags> can be a combination of C<IO::AIO::SYNC_FILE_RANGE_WAIT_BEFORE>,
949C<IO::AIO::SYNC_FILE_RANGE_WRITE> and
950C<IO::AIO::SYNC_FILE_RANGE_WAIT_AFTER>: refer to the sync_file_range
951manpage for details.
952
953=item aio_pathsync $path, $callback->($status)
954
955This request tries to open, fsync and close the given path. This is a
956composite request intended to sync directories after directory operations
957(E.g. rename). This might not work on all operating systems or have any
958specific effect, but usually it makes sure that directory changes get
959written to disc. It works for anything that can be opened for read-only,
960not just directories.
961
962Passes C<0> when everything went ok, and C<-1> on error.
963
964=cut
965
966sub aio_pathsync($;$) {
967 my ($path, $cb) = @_;
968
969 my $pri = aioreq_pri;
970 my $grp = aio_group $cb;
971
972 aioreq_pri $pri;
973 add $grp aio_open $path, O_RDONLY, 0, sub {
974 my ($fh) = @_;
975 if ($fh) {
976 aioreq_pri $pri;
977 add $grp aio_fsync $fh, sub {
978 $grp->result ($_[0]);
979
980 aioreq_pri $pri;
981 add $grp aio_close $fh;
982 };
983 } else {
984 $grp->result (-1);
985 }
986 };
987
988 $grp
989}
590 990
591=item aio_group $callback->(...) 991=item aio_group $callback->(...)
592 992
593This is a very special aio request: Instead of doing something, it is a 993This is a very special aio request: Instead of doing something, it is a
594container for other aio requests, which is useful if you want to bundle 994container for other aio requests, which is useful if you want to bundle
… …
644=item cancel $req 1044=item cancel $req
645 1045
646Cancels the request, if possible. Has the effect of skipping execution 1046Cancels the request, if possible. Has the effect of skipping execution
647when entering the B<execute> state and skipping calling the callback when 1047when entering the B<execute> state and skipping calling the callback when
648entering the the B<result> state, but will leave the request otherwise 1048entering the the B<result> state, but will leave the request otherwise
649untouched. That means that requests that currently execute will not be 1049untouched (with the exception of readdir). That means that requests that
650stopped and resources held by the request will not be freed prematurely. 1050currently execute will not be stopped and resources held by the request
1051will not be freed prematurely.
651 1052
652=item cb $req $callback->(...) 1053=item cb $req $callback->(...)
653 1054
654Replace (or simply set) the callback registered to the request. 1055Replace (or simply set) the callback registered to the request.
655 1056
… …
706Their lifetime, simplified, looks like this: when they are empty, they 1107Their lifetime, simplified, looks like this: when they are empty, they
707will finish very quickly. If they contain only requests that are in the 1108will finish very quickly. If they contain only requests that are in the
708C<done> state, they will also finish. Otherwise they will continue to 1109C<done> state, they will also finish. Otherwise they will continue to
709exist. 1110exist.
710 1111
711That means after creating a group you have some time to add requests. And 1112That means after creating a group you have some time to add requests
712in the callbacks of those requests, you can add further requests to the 1113(precisely before the callback has been invoked, which is only done within
713group. And only when all those requests have finished will the the group 1114the C<poll_cb>). And in the callbacks of those requests, you can add
714itself finish. 1115further requests to the group. And only when all those requests have
1116finished will the the group itself finish.
715 1117
716=over 4 1118=over 4
717 1119
718=item add $grp ... 1120=item add $grp ...
719 1121
… …
731itself. Useful when you queued a lot of events but got a result early. 1133itself. Useful when you queued a lot of events but got a result early.
732 1134
733=item $grp->result (...) 1135=item $grp->result (...)
734 1136
735Set the result value(s) that will be passed to the group callback when all 1137Set the result value(s) that will be passed to the group callback when all
736subrequests have finished and set thre groups errno to the current value 1138subrequests have finished and set the groups errno to the current value
737of errno (just like calling C<errno> without an error number). By default, 1139of errno (just like calling C<errno> without an error number). By default,
738no argument will be passed and errno is zero. 1140no argument will be passed and errno is zero.
739 1141
740=item $grp->errno ([$errno]) 1142=item $grp->errno ([$errno])
741 1143
… …
752=item feed $grp $callback->($grp) 1154=item feed $grp $callback->($grp)
753 1155
754Sets a feeder/generator on this group: every group can have an attached 1156Sets a feeder/generator on this group: every group can have an attached
755generator that generates requests if idle. The idea behind this is that, 1157generator that generates requests if idle. The idea behind this is that,
756although you could just queue as many requests as you want in a group, 1158although you could just queue as many requests as you want in a group,
757this might starve other requests for a potentially long time. For 1159this might starve other requests for a potentially long time. For example,
758example, C<aio_scandir> might generate hundreds of thousands C<aio_stat> 1160C<aio_scandir> might generate hundreds of thousands C<aio_stat> requests,
759requests, delaying any later requests for a long time. 1161delaying any later requests for a long time.
760 1162
761To avoid this, and allow incremental generation of requests, you can 1163To avoid this, and allow incremental generation of requests, you can
762instead a group and set a feeder on it that generates those requests. The 1164instead a group and set a feeder on it that generates those requests. The
763feed callback will be called whenever there are few enough (see C<limit>, 1165feed callback will be called whenever there are few enough (see C<limit>,
764below) requests active in the group itself and is expected to queue more 1166below) requests active in the group itself and is expected to queue more
… …
768not impose any limits). 1170not impose any limits).
769 1171
770If the feed does not queue more requests when called, it will be 1172If the feed does not queue more requests when called, it will be
771automatically removed from the group. 1173automatically removed from the group.
772 1174
773If the feed limit is C<0>, it will be set to C<2> automatically. 1175If the feed limit is C<0> when this method is called, it will be set to
1176C<2> automatically.
774 1177
775Example: 1178Example:
776 1179
777 # stat all files in @files, but only ever use four aio requests concurrently: 1180 # stat all files in @files, but only ever use four aio requests concurrently:
778 1181
… …
790Sets the feeder limit for the group: The feeder will be called whenever 1193Sets the feeder limit for the group: The feeder will be called whenever
791the group contains less than this many requests. 1194the group contains less than this many requests.
792 1195
793Setting the limit to C<0> will pause the feeding process. 1196Setting the limit to C<0> will pause the feeding process.
794 1197
1198The default value for the limit is C<0>, but note that setting a feeder
1199automatically bumps it up to C<2>.
1200
795=back 1201=back
796 1202
797=head2 SUPPORT FUNCTIONS 1203=head2 SUPPORT FUNCTIONS
798 1204
1205=head3 EVENT PROCESSING AND EVENT LOOP INTEGRATION
1206
799=over 4 1207=over 4
800 1208
801=item $fileno = IO::AIO::poll_fileno 1209=item $fileno = IO::AIO::poll_fileno
802 1210
803Return the I<request result pipe file descriptor>. This filehandle must be 1211Return the I<request result pipe file descriptor>. This filehandle must be
804polled for reading by some mechanism outside this module (e.g. Event or 1212polled for reading by some mechanism outside this module (e.g. EV, Glib,
805select, see below or the SYNOPSIS). If the pipe becomes readable you have 1213select and so on, see below or the SYNOPSIS). If the pipe becomes readable
806to call C<poll_cb> to check the results. 1214you have to call C<poll_cb> to check the results.
807 1215
808See C<poll_cb> for an example. 1216See C<poll_cb> for an example.
809 1217
810=item IO::AIO::poll_cb 1218=item IO::AIO::poll_cb
811 1219
812Process all outstanding events on the result pipe. You have to call this 1220Process some outstanding events on the result pipe. You have to call this
813regularly. Returns the number of events processed. Returns immediately 1221regularly. Returns C<0> if all events could be processed, or C<-1> if it
814when no events are outstanding. 1222returned earlier for whatever reason. Returns immediately when no events
1223are outstanding. The amount of events processed depends on the settings of
1224C<IO::AIO::max_poll_req> and C<IO::AIO::max_poll_time>.
815 1225
816If not all requests were processed for whatever reason, the filehandle 1226If not all requests were processed for whatever reason, the filehandle
817will still be ready when C<poll_cb> returns. 1227will still be ready when C<poll_cb> returns, so normally you don't have to
1228do anything special to have it called later.
818 1229
819Example: Install an Event watcher that automatically calls 1230Example: Install an Event watcher that automatically calls
820IO::AIO::poll_cb with high priority: 1231IO::AIO::poll_cb with high priority (more examples can be found in the
1232SYNOPSIS section, at the top of this document):
821 1233
822 Event->io (fd => IO::AIO::poll_fileno, 1234 Event->io (fd => IO::AIO::poll_fileno,
823 poll => 'r', async => 1, 1235 poll => 'r', async => 1,
824 cb => \&IO::AIO::poll_cb); 1236 cb => \&IO::AIO::poll_cb);
825 1237
826=item IO::AIO::poll_some $max_requests 1238=item IO::AIO::max_poll_reqs $nreqs
827 1239
828Similar to C<poll_cb>, but only processes up to C<$max_requests> requests 1240=item IO::AIO::max_poll_time $seconds
829at a time.
830 1241
831Useful if you want to ensure some level of interactiveness when perl is 1242These set the maximum number of requests (default C<0>, meaning infinity)
832not fast enough to process all requests in time. 1243that are being processed by C<IO::AIO::poll_cb> in one call, respectively
1244the maximum amount of time (default C<0>, meaning infinity) spent in
1245C<IO::AIO::poll_cb> to process requests (more correctly the mininum amount
1246of time C<poll_cb> is allowed to use).
1247
1248Setting C<max_poll_time> to a non-zero value creates an overhead of one
1249syscall per request processed, which is not normally a problem unless your
1250callbacks are really really fast or your OS is really really slow (I am
1251not mentioning Solaris here). Using C<max_poll_reqs> incurs no overhead.
1252
1253Setting these is useful if you want to ensure some level of
1254interactiveness when perl is not fast enough to process all requests in
1255time.
1256
1257For interactive programs, values such as C<0.01> to C<0.1> should be fine.
833 1258
834Example: Install an Event watcher that automatically calls 1259Example: Install an Event watcher that automatically calls
835IO::AIO::poll_some with low priority, to ensure that other parts of the 1260IO::AIO::poll_cb with low priority, to ensure that other parts of the
836program get the CPU sometimes even under high AIO load. 1261program get the CPU sometimes even under high AIO load.
837 1262
1263 # try not to spend much more than 0.1s in poll_cb
1264 IO::AIO::max_poll_time 0.1;
1265
1266 # use a low priority so other tasks have priority
838 Event->io (fd => IO::AIO::poll_fileno, 1267 Event->io (fd => IO::AIO::poll_fileno,
839 poll => 'r', nice => 1, 1268 poll => 'r', nice => 1,
840 cb => sub { IO::AIO::poll_some 256 }); 1269 cb => &IO::AIO::poll_cb);
841 1270
842=item IO::AIO::poll_wait 1271=item IO::AIO::poll_wait
843 1272
1273If there are any outstanding requests and none of them in the result
844Wait till the result filehandle becomes ready for reading (simply does a 1274phase, wait till the result filehandle becomes ready for reading (simply
845C<select> on the filehandle. This is useful if you want to synchronously wait 1275does a C<select> on the filehandle. This is useful if you want to
846for some requests to finish). 1276synchronously wait for some requests to finish).
847 1277
848See C<nreqs> for an example. 1278See C<nreqs> for an example.
849 1279
1280=item IO::AIO::poll
1281
1282Waits until some requests have been handled.
1283
1284Returns the number of requests processed, but is otherwise strictly
1285equivalent to:
1286
1287 IO::AIO::poll_wait, IO::AIO::poll_cb
1288
850=item IO::AIO::nreqs 1289=item IO::AIO::flush
851 1290
852Returns the number of requests currently in the ready, execute or pending 1291Wait till all outstanding AIO requests have been handled.
853states (i.e. for which their callback has not been invoked yet).
854 1292
855Example: wait till there are no outstanding requests anymore: 1293Strictly equivalent to:
856 1294
857 IO::AIO::poll_wait, IO::AIO::poll_cb 1295 IO::AIO::poll_wait, IO::AIO::poll_cb
858 while IO::AIO::nreqs; 1296 while IO::AIO::nreqs;
859 1297
860=item IO::AIO::nready 1298=back
861 1299
862Returns the number of requests currently in the ready state (not yet 1300=head3 CONTROLLING THE NUMBER OF THREADS
863executed).
864 1301
865=item IO::AIO::npending 1302=over
866
867Returns the number of requests currently in the pending state (executed,
868but not yet processed by poll_cb).
869
870=item IO::AIO::flush
871
872Wait till all outstanding AIO requests have been handled.
873
874Strictly equivalent to:
875
876 IO::AIO::poll_wait, IO::AIO::poll_cb
877 while IO::AIO::nreqs;
878
879=item IO::AIO::poll
880
881Waits until some requests have been handled.
882
883Strictly equivalent to:
884
885 IO::AIO::poll_wait, IO::AIO::poll_cb
886 if IO::AIO::nreqs;
887 1303
888=item IO::AIO::min_parallel $nthreads 1304=item IO::AIO::min_parallel $nthreads
889 1305
890Set the minimum number of AIO threads to C<$nthreads>. The current 1306Set the minimum number of AIO threads to C<$nthreads>. The current
891default is C<8>, which means eight asynchronous operations can execute 1307default is C<8>, which means eight asynchronous operations can execute
892concurrently at any one time (the number of outstanding requests, 1308concurrently at any one time (the number of outstanding requests,
893however, is unlimited). 1309however, is unlimited).
894 1310
895IO::AIO starts threads only on demand, when an AIO request is queued and 1311IO::AIO starts threads only on demand, when an AIO request is queued and
896no free thread exists. 1312no free thread exists. Please note that queueing up a hundred requests can
1313create demand for a hundred threads, even if it turns out that everything
1314is in the cache and could have been processed faster by a single thread.
897 1315
898It is recommended to keep the number of threads relatively low, as some 1316It is recommended to keep the number of threads relatively low, as some
899Linux kernel versions will scale negatively with the number of threads 1317Linux kernel versions will scale negatively with the number of threads
900(higher parallelity => MUCH higher latency). With current Linux 2.6 1318(higher parallelity => MUCH higher latency). With current Linux 2.6
901versions, 4-32 threads should be fine. 1319versions, 4-32 threads should be fine.
… …
915This module automatically runs C<max_parallel 0> at program end, to ensure 1333This module automatically runs C<max_parallel 0> at program end, to ensure
916that all threads are killed and that there are no outstanding requests. 1334that all threads are killed and that there are no outstanding requests.
917 1335
918Under normal circumstances you don't need to call this function. 1336Under normal circumstances you don't need to call this function.
919 1337
1338=item IO::AIO::max_idle $nthreads
1339
1340Limit the number of threads (default: 4) that are allowed to idle (i.e.,
1341threads that did not get a request to process within 10 seconds). That
1342means if a thread becomes idle while C<$nthreads> other threads are also
1343idle, it will free its resources and exit.
1344
1345This is useful when you allow a large number of threads (e.g. 100 or 1000)
1346to allow for extremely high load situations, but want to free resources
1347under normal circumstances (1000 threads can easily consume 30MB of RAM).
1348
1349The default is probably ok in most situations, especially if thread
1350creation is fast. If thread creation is very slow on your system you might
1351want to use larger values.
1352
920=item $oldmaxreqs = IO::AIO::max_outstanding $maxreqs 1353=item IO::AIO::max_outstanding $maxreqs
921 1354
922This is a very bad function to use in interactive programs because it 1355This is a very bad function to use in interactive programs because it
923blocks, and a bad way to reduce concurrency because it is inexact: Better 1356blocks, and a bad way to reduce concurrency because it is inexact: Better
924use an C<aio_group> together with a feed callback. 1357use an C<aio_group> together with a feed callback.
925 1358
926Sets the maximum number of outstanding requests to C<$nreqs>. If you 1359Sets the maximum number of outstanding requests to C<$nreqs>. If you
927to queue up more than this number of requests, the next call to the 1360do queue up more than this number of requests, the next call to the
928C<poll_cb> (and C<poll_some> and other functions calling C<poll_cb>) 1361C<poll_cb> (and C<poll_some> and other functions calling C<poll_cb>)
929function will block until the limit is no longer exceeded. 1362function will block until the limit is no longer exceeded.
930 1363
931The default value is very large, so there is no practical limit on the 1364The default value is very large, so there is no practical limit on the
932number of outstanding requests. 1365number of outstanding requests.
933 1366
934You can still queue as many requests as you want. Therefore, 1367You can still queue as many requests as you want. Therefore,
935C<max_oustsanding> is mainly useful in simple scripts (with low values) or 1368C<max_outstanding> is mainly useful in simple scripts (with low values) or
936as a stop gap to shield against fatal memory overflow (with large values). 1369as a stop gap to shield against fatal memory overflow (with large values).
937 1370
938=back 1371=back
939 1372
1373=head3 STATISTICAL INFORMATION
1374
1375=over
1376
1377=item IO::AIO::nreqs
1378
1379Returns the number of requests currently in the ready, execute or pending
1380states (i.e. for which their callback has not been invoked yet).
1381
1382Example: wait till there are no outstanding requests anymore:
1383
1384 IO::AIO::poll_wait, IO::AIO::poll_cb
1385 while IO::AIO::nreqs;
1386
1387=item IO::AIO::nready
1388
1389Returns the number of requests currently in the ready state (not yet
1390executed).
1391
1392=item IO::AIO::npending
1393
1394Returns the number of requests currently in the pending state (executed,
1395but not yet processed by poll_cb).
1396
1397=back
1398
940=cut 1399=cut
941 1400
942# support function to convert a fd into a perl filehandle
943sub _fd2fh {
944 return undef if $_[0] < 0;
945
946 # try to generate nice filehandles
947 my $sym = "IO::AIO::fd#$_[0]";
948 local *$sym;
949
950 open *$sym, "+<&=$_[0]" # usually works under any unix
951 or open *$sym, "<&=$_[0]" # cygwin needs this
952 or open *$sym, ">&=$_[0]" # or this
953 or return undef;
954
955 *$sym
956}
957
958min_parallel 8; 1401min_parallel 8;
1402
1403END { flush }
959 1404
9601; 14051;
961 1406
962=head2 FORK BEHAVIOUR 1407=head2 FORK BEHAVIOUR
963 1408
… …
983bytes of memory. In addition, stat requests need a stat buffer (possibly 1428bytes of memory. In addition, stat requests need a stat buffer (possibly
984a few hundred bytes), readdir requires a result buffer and so on. Perl 1429a few hundred bytes), readdir requires a result buffer and so on. Perl
985scalars and other data passed into aio requests will also be locked and 1430scalars and other data passed into aio requests will also be locked and
986will consume memory till the request has entered the done state. 1431will consume memory till the request has entered the done state.
987 1432
988This is now awfully much, so queuing lots of requests is not usually a 1433This is not awfully much, so queuing lots of requests is not usually a
989problem. 1434problem.
990 1435
991Per-thread usage: 1436Per-thread usage:
992 1437
993In the execution phase, some aio requests require more memory for 1438In the execution phase, some aio requests require more memory for
… …
998 1443
999Known bugs will be fixed in the next release. 1444Known bugs will be fixed in the next release.
1000 1445
1001=head1 SEE ALSO 1446=head1 SEE ALSO
1002 1447
1003L<Coro::AIO>. 1448L<AnyEvent::AIO> for easy integration into event loops, L<Coro::AIO> for a
1449more natural syntax.
1004 1450
1005=head1 AUTHOR 1451=head1 AUTHOR
1006 1452
1007 Marc Lehmann <schmorp@schmorp.de> 1453 Marc Lehmann <schmorp@schmorp.de>
1008 http://home.schmorp.de/ 1454 http://home.schmorp.de/

Diff Legend

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