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 1.65 by root, Mon Oct 23 14:49:51 2006 UTC vs.
Revision 1.224 by root, Sat Apr 7 00:50:33 2012 UTC

4 4
5=head1 SYNOPSIS 5=head1 SYNOPSIS
6 6
7 use IO::AIO; 7 use IO::AIO;
8 8
9 aio_open "/etc/passwd", O_RDONLY, 0, sub { 9 aio_open "/etc/passwd", IO::AIO::O_RDONLY, 0, sub {
10 my ($fh) = @_; 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
18 }; 19 };
19 20
20 # version 2+ has request and group objects 21 # version 2+ has request and group objects
21 use IO::AIO 2; 22 use IO::AIO 2;
22 23
24 aioreq_pri 4; # give next request a very high priority
23 my $req = aio_unlink "/tmp/file", sub { }; 25 my $req = aio_unlink "/tmp/file", sub { };
24 $req->cancel; # cancel request if still in queue 26 $req->cancel; # cancel request if still in queue
25 27
26 my $grp = aio_group sub { print "all stats done\n" }; 28 my $grp = aio_group sub { print "all stats done\n" };
27 add $grp aio_stat "..." for ...; 29 add $grp aio_stat "..." for ...;
28 30
29 # AnyEvent integration
30 open my $fh, "<&=" . IO::AIO::poll_fileno or die "$!";
31 my $w = AnyEvent->io (fh => $fh, poll => 'r', cb => sub { IO::AIO::poll_cb });
32
33 # Event integration
34 Event->io (fd => IO::AIO::poll_fileno,
35 poll => 'r',
36 cb => \&IO::AIO::poll_cb);
37
38 # Glib/Gtk2 integration
39 add_watch Glib::IO IO::AIO::poll_fileno,
40 in => sub { IO::AIO::poll_cb; 1 };
41
42 # Tk integration
43 Tk::Event::IO->fileevent (IO::AIO::poll_fileno, "",
44 readable => \&IO::AIO::poll_cb);
45
46 # Danga::Socket integration
47 Danga::Socket->AddOtherFds (IO::AIO::poll_fileno =>
48 \&IO::AIO::poll_cb);
49
50=head1 DESCRIPTION 31=head1 DESCRIPTION
51 32
52This module implements asynchronous I/O using whatever means your 33This module implements asynchronous I/O using whatever means your
53operating system supports. 34operating system supports. It is implemented as an interface to C<libeio>
35(L<http://software.schmorp.de/pkg/libeio.html>).
54 36
37Asynchronous means that operations that can normally block your program
38(e.g. reading from disk) will be done asynchronously: the operation
39will still block, but you can do something else in the meantime. This
40is extremely useful for programs that need to stay interactive even
41when doing heavy I/O (GUI programs, high performance network servers
42etc.), but can also be used to easily do operations in parallel that are
43normally done sequentially, e.g. stat'ing many files, which is much faster
44on a RAID volume or over NFS when you do a number of stat operations
45concurrently.
46
47While most of this works on all types of file descriptors (for
48example sockets), using these functions on file descriptors that
49support nonblocking operation (again, sockets, pipes etc.) is
50very inefficient. Use an event loop for that (such as the L<EV>
51module): IO::AIO will naturally fit into such an event loop itself.
52
55Currently, a number of threads are started that execute your read/writes 53In this version, a number of threads are started that execute your
56and signal their completion. You don't need thread support in your libc or 54requests and signal their completion. You don't need thread support
57perl, and the threads created by this module will not be visible to the 55in perl, and the threads created by this module will not be visible
58pthreads library. In the future, this module might make use of the native 56to perl. In the future, this module might make use of the native aio
59aio functions available on many operating systems. However, they are often 57functions available on many operating systems. However, they are often
60not well-supported (Linux doesn't allow them on normal files currently, 58not well-supported or restricted (GNU/Linux doesn't allow them on normal
61for example), and they would only support aio_read and aio_write, so the 59files currently, for example), and they would only support aio_read and
62remaining functionality would have to be implemented using threads anyway. 60aio_write, so the remaining functionality would have to be implemented
61using threads anyway.
63 62
64Although the module will work with in the presence of other threads, it is 63Although the module will work in the presence of other (Perl-) threads,
65currently not reentrant, so use appropriate locking yourself, always call 64it is currently not reentrant in any way, so use appropriate locking
66C<poll_cb> from within the same thread, or never call C<poll_cb> (or other 65yourself, always call C<poll_cb> from within the same thread, or never
67C<aio_> functions) recursively. 66call C<poll_cb> (or other C<aio_> functions) recursively.
67
68=head2 EXAMPLE
69
70This is a simple example that uses the EV module and loads
71F</etc/passwd> asynchronously:
72
73 use Fcntl;
74 use EV;
75 use IO::AIO;
76
77 # register the IO::AIO callback with EV
78 my $aio_w = EV::io IO::AIO::poll_fileno, EV::READ, \&IO::AIO::poll_cb;
79
80 # queue the request to open /etc/passwd
81 aio_open "/etc/passwd", IO::AIO::O_RDONLY, 0, sub {
82 my $fh = shift
83 or die "error while opening: $!";
84
85 # stat'ing filehandles is generally non-blocking
86 my $size = -s $fh;
87
88 # queue a request to read the file
89 my $contents;
90 aio_read $fh, 0, $size, $contents, 0, sub {
91 $_[0] == $size
92 or die "short read: $!";
93
94 close $fh;
95
96 # file contents now in $contents
97 print $contents;
98
99 # exit event loop and program
100 EV::unloop;
101 };
102 };
103
104 # possibly queue up other requests, or open GUI windows,
105 # check for sockets etc. etc.
106
107 # process events as long as there are some:
108 EV::loop;
109
110=head1 REQUEST ANATOMY AND LIFETIME
111
112Every C<aio_*> function creates a request. which is a C data structure not
113directly visible to Perl.
114
115If called in non-void context, every request function returns a Perl
116object representing the request. In void context, nothing is returned,
117which saves a bit of memory.
118
119The perl object is a fairly standard ref-to-hash object. The hash contents
120are not used by IO::AIO so you are free to store anything you like in it.
121
122During their existance, aio requests travel through the following states,
123in order:
124
125=over 4
126
127=item ready
128
129Immediately after a request is created it is put into the ready state,
130waiting for a thread to execute it.
131
132=item execute
133
134A thread has accepted the request for processing and is currently
135executing it (e.g. blocking in read).
136
137=item pending
138
139The request has been executed and is waiting for result processing.
140
141While request submission and execution is fully asynchronous, result
142processing is not and relies on the perl interpreter calling C<poll_cb>
143(or another function with the same effect).
144
145=item result
146
147The request results are processed synchronously by C<poll_cb>.
148
149The C<poll_cb> function will process all outstanding aio requests by
150calling their callbacks, freeing memory associated with them and managing
151any groups they are contained in.
152
153=item done
154
155Request has reached the end of its lifetime and holds no resources anymore
156(except possibly for the Perl object, but its connection to the actual
157aio request is severed and calling its methods will either do nothing or
158result in a runtime error).
159
160=back
68 161
69=cut 162=cut
70 163
71package IO::AIO; 164package IO::AIO;
72 165
73no warnings; 166use Carp ();
74use strict 'vars'; 167
168use common::sense;
75 169
76use base 'Exporter'; 170use base 'Exporter';
77 171
78BEGIN { 172BEGIN {
79 our $VERSION = '2.0'; 173 our $VERSION = '4.14';
80 174
81 our @EXPORT = qw(aio_sendfile aio_read aio_write aio_open aio_close aio_stat 175 our @AIO_REQ = qw(aio_sendfile aio_seek aio_read aio_write aio_open aio_close
82 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
83 aio_fsync aio_fdatasync aio_readahead aio_rename aio_link aio_move 177 aio_scandir aio_symlink aio_readlink aio_realpath aio_sync
178 aio_fsync aio_syncfs aio_fdatasync aio_sync_file_range aio_fallocate
179 aio_pathsync aio_readahead aio_fiemap
180 aio_rename aio_link aio_move aio_copy aio_group
181 aio_nop aio_mknod aio_load aio_rmtree aio_mkdir aio_chown
182 aio_chmod aio_utime aio_truncate
183 aio_msync aio_mtouch aio_mlock aio_mlockall
184 aio_statvfs
84 aio_group aio_nop); 185 aio_wd);
85 our @EXPORT_OK = qw(poll_fileno poll_cb min_parallel max_parallel max_outstanding nreqs); 186
187 our @EXPORT = (@AIO_REQ, qw(aioreq_pri aioreq_nice));
188 our @EXPORT_OK = qw(poll_fileno poll_cb poll_wait flush
189 min_parallel max_parallel max_idle idle_timeout
190 nreqs nready npending nthreads
191 max_poll_time max_poll_reqs
192 sendfile fadvise madvise
193 mmap munmap munlock munlockall);
194
195 push @AIO_REQ, qw(aio_busy); # not exported
86 196
87 @IO::AIO::GRP::ISA = 'IO::AIO::REQ'; 197 @IO::AIO::GRP::ISA = 'IO::AIO::REQ';
88 198
89 require XSLoader; 199 require XSLoader;
90 XSLoader::load ("IO::AIO", $VERSION); 200 XSLoader::load ("IO::AIO", $VERSION);
91} 201}
92 202
93=head1 FUNCTIONS 203=head1 FUNCTIONS
94 204
95=head2 AIO FUNCTIONS 205=head2 QUICK OVERVIEW
206
207This section simply lists the prototypes of the most important functions
208for quick reference. See the following sections for function-by-function
209documentation.
210
211 aio_wd $pathname, $callback->($wd)
212 aio_open $pathname, $flags, $mode, $callback->($fh)
213 aio_close $fh, $callback->($status)
214 aio_seek $fh,$offset,$whence, $callback->($offs)
215 aio_read $fh,$offset,$length, $data,$dataoffset, $callback->($retval)
216 aio_write $fh,$offset,$length, $data,$dataoffset, $callback->($retval)
217 aio_sendfile $out_fh, $in_fh, $in_offset, $length, $callback->($retval)
218 aio_readahead $fh,$offset,$length, $callback->($retval)
219 aio_stat $fh_or_path, $callback->($status)
220 aio_lstat $fh, $callback->($status)
221 aio_statvfs $fh_or_path, $callback->($statvfs)
222 aio_utime $fh_or_path, $atime, $mtime, $callback->($status)
223 aio_chown $fh_or_path, $uid, $gid, $callback->($status)
224 aio_chmod $fh_or_path, $mode, $callback->($status)
225 aio_truncate $fh_or_path, $offset, $callback->($status)
226 aio_unlink $pathname, $callback->($status)
227 aio_mknod $pathname, $mode, $dev, $callback->($status)
228 aio_link $srcpath, $dstpath, $callback->($status)
229 aio_symlink $srcpath, $dstpath, $callback->($status)
230 aio_readlink $pathname, $callback->($link)
231 aio_realpath $pathname, $callback->($link)
232 aio_rename $srcpath, $dstpath, $callback->($status)
233 aio_mkdir $pathname, $mode, $callback->($status)
234 aio_rmdir $pathname, $callback->($status)
235 aio_readdir $pathname, $callback->($entries)
236 aio_readdirx $pathname, $flags, $callback->($entries, $flags)
237 IO::AIO::READDIR_DENTS IO::AIO::READDIR_DIRS_FIRST
238 IO::AIO::READDIR_STAT_ORDER IO::AIO::READDIR_FOUND_UNKNOWN
239 aio_scandir $pathname, $maxreq, $callback->($dirs, $nondirs)
240 aio_load $pathname, $data, $callback->($status)
241 aio_copy $srcpath, $dstpath, $callback->($status)
242 aio_move $srcpath, $dstpath, $callback->($status)
243 aio_rmtree $pathname, $callback->($status)
244 aio_sync $callback->($status)
245 aio_syncfs $fh, $callback->($status)
246 aio_fsync $fh, $callback->($status)
247 aio_fdatasync $fh, $callback->($status)
248 aio_sync_file_range $fh, $offset, $nbytes, $flags, $callback->($status)
249 aio_pathsync $pathname, $callback->($status)
250 aio_msync $scalar, $offset = 0, $length = undef, flags = 0, $callback->($status)
251 aio_mtouch $scalar, $offset = 0, $length = undef, flags = 0, $callback->($status)
252 aio_mlock $scalar, $offset = 0, $length = undef, $callback->($status)
253 aio_mlockall $flags, $callback->($status)
254 aio_group $callback->(...)
255 aio_nop $callback->()
256
257 $prev_pri = aioreq_pri [$pri]
258 aioreq_nice $pri_adjust
259
260 IO::AIO::poll_wait
261 IO::AIO::poll_cb
262 IO::AIO::poll
263 IO::AIO::flush
264 IO::AIO::max_poll_reqs $nreqs
265 IO::AIO::max_poll_time $seconds
266 IO::AIO::min_parallel $nthreads
267 IO::AIO::max_parallel $nthreads
268 IO::AIO::max_idle $nthreads
269 IO::AIO::idle_timeout $seconds
270 IO::AIO::max_outstanding $maxreqs
271 IO::AIO::nreqs
272 IO::AIO::nready
273 IO::AIO::npending
274
275 IO::AIO::sendfile $ofh, $ifh, $offset, $count
276 IO::AIO::fadvise $fh, $offset, $len, $advice
277 IO::AIO::madvise $scalar, $offset, $length, $advice
278 IO::AIO::mprotect $scalar, $offset, $length, $protect
279 IO::AIO::munlock $scalar, $offset = 0, $length = undef
280 IO::AIO::munlockall
281
282=head2 API NOTES
96 283
97All the C<aio_*> calls are more or less thin wrappers around the syscall 284All the C<aio_*> calls are more or less thin wrappers around the syscall
98with the same name (sans C<aio_>). The arguments are similar or identical, 285with the same name (sans C<aio_>). The arguments are similar or identical,
99and they all accept an additional (and optional) C<$callback> argument 286and they all accept an additional (and optional) C<$callback> argument
100which must be a code reference. This code reference will get called with 287which must be a code reference. This code reference will be called after
101the syscall return code (e.g. most syscalls return C<-1> on error, unlike 288the syscall has been executed in an asynchronous fashion. The results
102perl, which usually delivers "false") as it's sole argument when the given 289of the request will be passed as arguments to the callback (and, if an
103syscall has been executed asynchronously. 290error occured, in C<$!>) - for most requests the syscall return code (e.g.
291most syscalls return C<-1> on error, unlike perl, which usually delivers
292"false").
293
294Some requests (such as C<aio_readdir>) pass the actual results and
295communicate failures by passing C<undef>.
104 296
105All functions expecting a filehandle keep a copy of the filehandle 297All functions expecting a filehandle keep a copy of the filehandle
106internally until the request has finished. 298internally until the request has finished.
107 299
108All requests return objects of type L<IO::AIO::REQ> that allow further 300All functions return request objects of type L<IO::AIO::REQ> that allow
109manipulation of those requests while they are in-flight. 301further manipulation of those requests while they are in-flight.
110 302
111The pathnames you pass to these routines I<must> be absolute and 303The pathnames you pass to these routines I<should> be absolute. The
112encoded in byte form. The reason for the former is that at the time the 304reason for this is that at the time the request is being executed, the
113request is being executed, the current working directory could have 305current working directory could have changed. Alternatively, you can
114changed. Alternatively, you can make sure that you never change the 306make sure that you never change the current working directory anywhere
115current working directory. 307in the program and then use relative paths. You can also take advantage
308of IO::AIOs working directory abstraction, that lets you specify paths
309relative to some previously-opened "working directory object" - see the
310description of the C<IO::AIO::WD> class later in this document.
116 311
117To encode pathnames to byte form, either make sure you either: a) 312To encode pathnames as octets, either make sure you either: a) always pass
118always pass in filenames you got from outside (command line, readdir 313in filenames you got from outside (command line, readdir etc.) without
119etc.), b) are ASCII or ISO 8859-1, c) use the Encode module and encode 314tinkering, b) are in your native filesystem encoding, c) use the Encode
120your pathnames to the locale (or other) encoding in effect in the user 315module and encode your pathnames to the locale (or other) encoding in
121environment, d) use Glib::filename_from_unicode on unicode filenames or e) 316effect in the user environment, d) use Glib::filename_from_unicode on
122use something else. 317unicode filenames or e) use something else to ensure your scalar has the
318correct contents.
319
320This works, btw. independent of the internal UTF-8 bit, which IO::AIO
321handles correctly whether it is set or not.
322
323=head2 AIO REQUEST FUNCTIONS
123 324
124=over 4 325=over 4
326
327=item $prev_pri = aioreq_pri [$pri]
328
329Returns the priority value that would be used for the next request and, if
330C<$pri> is given, sets the priority for the next aio request.
331
332The default priority is C<0>, the minimum and maximum priorities are C<-4>
333and C<4>, respectively. Requests with higher priority will be serviced
334first.
335
336The priority will be reset to C<0> after each call to one of the C<aio_*>
337functions.
338
339Example: open a file with low priority, then read something from it with
340higher priority so the read request is serviced before other low priority
341open requests (potentially spamming the cache):
342
343 aioreq_pri -3;
344 aio_open ..., sub {
345 return unless $_[0];
346
347 aioreq_pri -2;
348 aio_read $_[0], ..., sub {
349 ...
350 };
351 };
352
353
354=item aioreq_nice $pri_adjust
355
356Similar to C<aioreq_pri>, but subtracts the given value from the current
357priority, so the effect is cumulative.
358
125 359
126=item aio_open $pathname, $flags, $mode, $callback->($fh) 360=item aio_open $pathname, $flags, $mode, $callback->($fh)
127 361
128Asynchronously open or create a file and call the callback with a newly 362Asynchronously open or create a file and call the callback with a newly
129created filehandle for the file. 363created filehandle for the file.
135list. They are the same as used by C<sysopen>. 369list. They are the same as used by C<sysopen>.
136 370
137Likewise, C<$mode> specifies the mode of the newly created file, if it 371Likewise, C<$mode> specifies the mode of the newly created file, if it
138didn't exist and C<O_CREAT> has been given, just like perl's C<sysopen>, 372didn't exist and C<O_CREAT> has been given, just like perl's C<sysopen>,
139except that it is mandatory (i.e. use C<0> if you don't create new files, 373except that it is mandatory (i.e. use C<0> if you don't create new files,
140and C<0666> or C<0777> if you do). 374and C<0666> or C<0777> if you do). Note that the C<$mode> will be modified
375by the umask in effect then the request is being executed, so better never
376change the umask.
141 377
142Example: 378Example:
143 379
144 aio_open "/etc/passwd", O_RDONLY, 0, sub { 380 aio_open "/etc/passwd", IO::AIO::O_RDONLY, 0, sub {
145 if ($_[0]) { 381 if ($_[0]) {
146 print "open successful, fh is $_[0]\n"; 382 print "open successful, fh is $_[0]\n";
147 ... 383 ...
148 } else { 384 } else {
149 die "open failed: $!\n"; 385 die "open failed: $!\n";
150 } 386 }
151 }; 387 };
152 388
389In addition to all the common open modes/flags (C<O_RDONLY>, C<O_WRONLY>,
390C<O_RDWR>, C<O_CREAT>, C<O_TRUNC>, C<O_EXCL> and C<O_APPEND>), the
391following POSIX and non-POSIX constants are available (missing ones on
392your system are, as usual, C<0>):
393
394C<O_ASYNC>, C<O_DIRECT>, C<O_NOATIME>, C<O_CLOEXEC>, C<O_NOCTTY>, C<O_NOFOLLOW>,
395C<O_NONBLOCK>, C<O_EXEC>, C<O_SEARCH>, C<O_DIRECTORY>, C<O_DSYNC>,
396C<O_RSYNC>, C<O_SYNC> and C<O_TTY_INIT>.
397
398
153=item aio_close $fh, $callback->($status) 399=item aio_close $fh, $callback->($status)
154 400
155Asynchronously close a file and call the callback with the result 401Asynchronously close a file and call the callback with the result
156code. I<WARNING:> although accepted, you should not pass in a perl 402code.
157filehandle here, as perl will likely close the file descriptor another
158time when the filehandle is destroyed. Normally, you can safely call perls
159C<close> or just let filehandles go out of scope.
160 403
161This is supposed to be a bug in the API, so that might change. It's 404Unfortunately, you can't do this to perl. Perl I<insists> very strongly on
162therefore best to avoid this function. 405closing the file descriptor associated with the filehandle itself.
406
407Therefore, C<aio_close> will not close the filehandle - instead it will
408use dup2 to overwrite the file descriptor with the write-end of a pipe
409(the pipe fd will be created on demand and will be cached).
410
411Or in other words: the file descriptor will be closed, but it will not be
412free for reuse until the perl filehandle is closed.
413
414=cut
415
416=item aio_seek $fh, $offset, $whence, $callback->($offs)
417
418Seeks the filehandle to the new C<$offset>, similarly to perl's
419C<sysseek>. The C<$whence> can use the traditional values (C<0> for
420C<IO::AIO::SEEK_SET>, C<1> for C<IO::AIO::SEEK_CUR> or C<2> for
421C<IO::AIO::SEEK_END>).
422
423The resulting absolute offset will be passed to the callback, or C<-1> in
424case of an error.
425
426In theory, the C<$whence> constants could be different than the
427corresponding values from L<Fcntl>, but perl guarantees they are the same,
428so don't panic.
163 429
164=item aio_read $fh,$offset,$length, $data,$dataoffset, $callback->($retval) 430=item aio_read $fh,$offset,$length, $data,$dataoffset, $callback->($retval)
165 431
166=item aio_write $fh,$offset,$length, $data,$dataoffset, $callback->($retval) 432=item aio_write $fh,$offset,$length, $data,$dataoffset, $callback->($retval)
167 433
168Reads or writes C<length> bytes from the specified C<fh> and C<offset> 434Reads or writes C<$length> bytes from or to the specified C<$fh> and
169into the scalar given by C<data> and offset C<dataoffset> and calls the 435C<$offset> into the scalar given by C<$data> and offset C<$dataoffset>
170callback without the actual number of bytes read (or -1 on error, just 436and calls the callback without the actual number of bytes read (or -1 on
171like the syscall). 437error, just like the syscall).
438
439C<aio_read> will, like C<sysread>, shrink or grow the C<$data> scalar to
440offset plus the actual number of bytes read.
441
442If C<$offset> is undefined, then the current file descriptor offset will
443be used (and updated), otherwise the file descriptor offset will not be
444changed by these calls.
445
446If C<$length> is undefined in C<aio_write>, use the remaining length of
447C<$data>.
448
449If C<$dataoffset> is less than zero, it will be counted from the end of
450C<$data>.
172 451
173The C<$data> scalar I<MUST NOT> be modified in any way while the request 452The C<$data> scalar I<MUST NOT> be modified in any way while the request
174is outstanding. Modifying it can result in segfaults or WW3 (if the 453is outstanding. Modifying it can result in segfaults or World War III (if
175necessary/optional hardware is installed). 454the necessary/optional hardware is installed).
176 455
177Example: Read 15 bytes at offset 7 into scalar C<$buffer>, starting at 456Example: Read 15 bytes at offset 7 into scalar C<$buffer>, starting at
178offset C<0> within the scalar: 457offset C<0> within the scalar:
179 458
180 aio_read $fh, 7, 15, $buffer, 0, sub { 459 aio_read $fh, 7, 15, $buffer, 0, sub {
181 $_[0] > 0 or die "read error: $!"; 460 $_[0] > 0 or die "read error: $!";
182 print "read $_[0] bytes: <$buffer>\n"; 461 print "read $_[0] bytes: <$buffer>\n";
183 }; 462 };
184 463
185=item aio_move $srcpath, $dstpath, $callback->($status)
186
187[EXPERIMENTAL due to internal aio_group use]
188
189Try to move the I<file> (directories not supported as either source or
190destination) from C<$srcpath> to C<$dstpath> and call the callback with
191the C<0> (error) or C<-1> ok.
192
193This is a composite request that tries to rename(2) the file first. If
194rename files with C<EXDEV>, it creates the destination file with mode 0200
195and copies the contents of the source file into it using C<aio_sendfile>,
196followed by restoring atime, mtime, access mode and uid/gid, in that
197order, and unlinking the C<$srcpath>.
198
199If an error occurs, the partial destination file will be unlinked, if
200possible, except when setting atime, mtime, access mode and uid/gid, where
201errors are being ignored.
202
203=cut
204
205sub aio_move($$$) {
206 my ($src, $dst, $cb) = @_;
207
208 my $grp = aio_group $cb;
209
210 add $grp aio_rename $src, $dst, sub {
211 if ($_[0] && $! == EXDEV) {
212 add $grp aio_open $src, O_RDONLY, 0, sub {
213 if (my $src_fh = $_[0]) {
214 my @stat = stat $src_fh;
215
216 add $grp aio_open $dst, O_WRONLY, 0200, sub {
217 if (my $dst_fh = $_[0]) {
218 add $grp aio_sendfile $dst_fh, $src_fh, 0, $stat[7], sub {
219 close $src_fh;
220
221 if ($_[0] == $stat[7]) {
222 utime $stat[8], $stat[9], $dst;
223 chmod $stat[2] & 07777, $dst_fh;
224 chown $stat[4], $stat[5], $dst_fh;
225 close $dst_fh;
226
227 add $grp aio_unlink $src, sub {
228 $grp->result ($_[0]);
229 };
230 } else {
231 my $errno = $!;
232 add $grp aio_unlink $dst, sub {
233 $! = $errno;
234 $grp->result (-1);
235 };
236 }
237 };
238 } else {
239 $grp->result (-1);
240 }
241 },
242
243 } else {
244 $grp->result (-1);
245 }
246 };
247 } else {
248 $grp->result ($_[0]);
249 }
250 };
251
252 $grp
253}
254 464
255=item aio_sendfile $out_fh, $in_fh, $in_offset, $length, $callback->($retval) 465=item aio_sendfile $out_fh, $in_fh, $in_offset, $length, $callback->($retval)
256 466
257Tries to copy C<$length> bytes from C<$in_fh> to C<$out_fh>. It starts 467Tries to copy C<$length> bytes from C<$in_fh> to C<$out_fh>. It starts
258reading at byte offset C<$in_offset>, and starts writing at the current 468reading at byte offset C<$in_offset>, and starts writing at the current
259file offset of C<$out_fh>. Because of that, it is not safe to issue more 469file offset of C<$out_fh>. Because of that, it is not safe to issue more
260than one C<aio_sendfile> per C<$out_fh>, as they will interfere with each 470than one C<aio_sendfile> per C<$out_fh>, as they will interfere with each
261other. 471other. The same C<$in_fh> works fine though, as this function does not
472move or use the file offset of C<$in_fh>.
262 473
474Please note that C<aio_sendfile> can read more bytes from C<$in_fh> than
475are written, and there is no way to find out how many more bytes have been
476read from C<aio_sendfile> alone, as C<aio_sendfile> only provides the
477number of bytes written to C<$out_fh>. Only if the result value equals
478C<$length> one can assume that C<$length> bytes have been read.
479
480Unlike with other C<aio_> functions, it makes a lot of sense to use
481C<aio_sendfile> on non-blocking sockets, as long as one end (typically
482the C<$in_fh>) is a file - the file I/O will then be asynchronous, while
483the socket I/O will be non-blocking. Note, however, that you can run
484into a trap where C<aio_sendfile> reads some data with readahead, then
485fails to write all data, and when the socket is ready the next time, the
486data in the cache is already lost, forcing C<aio_sendfile> to again hit
487the disk. Explicit C<aio_read> + C<aio_write> let's you better control
488resource usage.
489
263This call tries to make use of a native C<sendfile> syscall to provide 490This call tries to make use of a native C<sendfile>-like syscall to
264zero-copy operation. For this to work, C<$out_fh> should refer to a 491provide zero-copy operation. For this to work, C<$out_fh> should refer to
265socket, and C<$in_fh> should refer to mmap'able file. 492a socket, and C<$in_fh> should refer to an mmap'able file.
266 493
267If the native sendfile call fails or is not implemented, it will be 494If a native sendfile cannot be found or it fails with C<ENOSYS>,
268emulated, so you can call C<aio_sendfile> on any type of filehandle 495C<EINVAL>, C<ENOTSUP>, C<EOPNOTSUPP>, C<EAFNOSUPPORT>, C<EPROTOTYPE> or
496C<ENOTSOCK>, it will be emulated, so you can call C<aio_sendfile> on any
269regardless of the limitations of the operating system. 497type of filehandle regardless of the limitations of the operating system.
270 498
271Please note, however, that C<aio_sendfile> can read more bytes from 499As native sendfile syscalls (as practically any non-POSIX interface hacked
272C<$in_fh> than are written, and there is no way to find out how many 500together in a hurry to improve benchmark numbers) tend to be rather buggy
273bytes have been read from C<aio_sendfile> alone, as C<aio_sendfile> only 501on many systems, this implementation tries to work around some known bugs
274provides the number of bytes written to C<$out_fh>. Only if the result 502in Linux and FreeBSD kernels (probably others, too), but that might fail,
275value equals C<$length> one can assume that C<$length> bytes have been 503so you really really should check the return value of C<aio_sendfile> -
276read. 504fewre bytes than expected might have been transferred.
505
277 506
278=item aio_readahead $fh,$offset,$length, $callback->($retval) 507=item aio_readahead $fh,$offset,$length, $callback->($retval)
279 508
280C<aio_readahead> populates the page cache with data from a file so that 509C<aio_readahead> populates the page cache with data from a file so that
281subsequent reads from that file will not block on disk I/O. The C<$offset> 510subsequent reads from that file will not block on disk I/O. The C<$offset>
287file. The current file offset of the file is left unchanged. 516file. The current file offset of the file is left unchanged.
288 517
289If that syscall doesn't exist (likely if your OS isn't Linux) it will be 518If that syscall doesn't exist (likely if your OS isn't Linux) it will be
290emulated by simply reading the data, which would have a similar effect. 519emulated by simply reading the data, which would have a similar effect.
291 520
521
292=item aio_stat $fh_or_path, $callback->($status) 522=item aio_stat $fh_or_path, $callback->($status)
293 523
294=item aio_lstat $fh, $callback->($status) 524=item aio_lstat $fh, $callback->($status)
295 525
296Works like perl's C<stat> or C<lstat> in void context. The callback will 526Works like perl's C<stat> or C<lstat> in void context. The callback will
301for an explanation. 531for an explanation.
302 532
303Currently, the stats are always 64-bit-stats, i.e. instead of returning an 533Currently, the stats are always 64-bit-stats, i.e. instead of returning an
304error when stat'ing a large file, the results will be silently truncated 534error when stat'ing a large file, the results will be silently truncated
305unless perl itself is compiled with large file support. 535unless perl itself is compiled with large file support.
536
537To help interpret the mode and dev/rdev stat values, IO::AIO offers the
538following constants and functions (if not implemented, the constants will
539be C<0> and the functions will either C<croak> or fall back on traditional
540behaviour).
541
542C<S_IFMT>, C<S_IFIFO>, C<S_IFCHR>, C<S_IFBLK>, C<S_IFLNK>, C<S_IFREG>,
543C<S_IFDIR>, C<S_IFWHT>, C<S_IFSOCK>, C<IO::AIO::major $dev_t>,
544C<IO::AIO::minor $dev_t>, C<IO::AIO::makedev $major, $minor>.
306 545
307Example: Print the length of F</etc/passwd>: 546Example: Print the length of F</etc/passwd>:
308 547
309 aio_stat "/etc/passwd", sub { 548 aio_stat "/etc/passwd", sub {
310 $_[0] and die "stat failed: $!"; 549 $_[0] and die "stat failed: $!";
311 print "size is ", -s _, "\n"; 550 print "size is ", -s _, "\n";
312 }; 551 };
313 552
553
554=item aio_statvfs $fh_or_path, $callback->($statvfs)
555
556Works like the POSIX C<statvfs> or C<fstatvfs> syscalls, depending on
557whether a file handle or path was passed.
558
559On success, the callback is passed a hash reference with the following
560members: C<bsize>, C<frsize>, C<blocks>, C<bfree>, C<bavail>, C<files>,
561C<ffree>, C<favail>, C<fsid>, C<flag> and C<namemax>. On failure, C<undef>
562is passed.
563
564The following POSIX IO::AIO::ST_* constants are defined: C<ST_RDONLY> and
565C<ST_NOSUID>.
566
567The following non-POSIX IO::AIO::ST_* flag masks are defined to
568their correct value when available, or to C<0> on systems that do
569not support them: C<ST_NODEV>, C<ST_NOEXEC>, C<ST_SYNCHRONOUS>,
570C<ST_MANDLOCK>, C<ST_WRITE>, C<ST_APPEND>, C<ST_IMMUTABLE>, C<ST_NOATIME>,
571C<ST_NODIRATIME> and C<ST_RELATIME>.
572
573Example: stat C</wd> and dump out the data if successful.
574
575 aio_statvfs "/wd", sub {
576 my $f = $_[0]
577 or die "statvfs: $!";
578
579 use Data::Dumper;
580 say Dumper $f;
581 };
582
583 # result:
584 {
585 bsize => 1024,
586 bfree => 4333064312,
587 blocks => 10253828096,
588 files => 2050765568,
589 flag => 4096,
590 favail => 2042092649,
591 bavail => 4333064312,
592 ffree => 2042092649,
593 namemax => 255,
594 frsize => 1024,
595 fsid => 1810
596 }
597
598
599=item aio_utime $fh_or_path, $atime, $mtime, $callback->($status)
600
601Works like perl's C<utime> function (including the special case of $atime
602and $mtime being undef). Fractional times are supported if the underlying
603syscalls support them.
604
605When called with a pathname, uses utimes(2) if available, otherwise
606utime(2). If called on a file descriptor, uses futimes(2) if available,
607otherwise returns ENOSYS, so this is not portable.
608
609Examples:
610
611 # set atime and mtime to current time (basically touch(1)):
612 aio_utime "path", undef, undef;
613 # set atime to current time and mtime to beginning of the epoch:
614 aio_utime "path", time, undef; # undef==0
615
616
617=item aio_chown $fh_or_path, $uid, $gid, $callback->($status)
618
619Works like perl's C<chown> function, except that C<undef> for either $uid
620or $gid is being interpreted as "do not change" (but -1 can also be used).
621
622Examples:
623
624 # same as "chown root path" in the shell:
625 aio_chown "path", 0, -1;
626 # same as above:
627 aio_chown "path", 0, undef;
628
629
630=item aio_truncate $fh_or_path, $offset, $callback->($status)
631
632Works like truncate(2) or ftruncate(2).
633
634
635=item aio_chmod $fh_or_path, $mode, $callback->($status)
636
637Works like perl's C<chmod> function.
638
639
314=item aio_unlink $pathname, $callback->($status) 640=item aio_unlink $pathname, $callback->($status)
315 641
316Asynchronously unlink (delete) a file and call the callback with the 642Asynchronously unlink (delete) a file and call the callback with the
317result code. 643result code.
318 644
645
646=item aio_mknod $pathname, $mode, $dev, $callback->($status)
647
648[EXPERIMENTAL]
649
650Asynchronously create a device node (or fifo). See mknod(2).
651
652The only (POSIX-) portable way of calling this function is:
653
654 aio_mknod $pathname, IO::AIO::S_IFIFO | $mode, 0, sub { ...
655
656See C<aio_stat> for info about some potentially helpful extra constants
657and functions.
658
319=item aio_link $srcpath, $dstpath, $callback->($status) 659=item aio_link $srcpath, $dstpath, $callback->($status)
320 660
321Asynchronously create a new link to the existing object at C<$srcpath> at 661Asynchronously create a new link to the existing object at C<$srcpath> at
322the path C<$dstpath> and call the callback with the result code. 662the path C<$dstpath> and call the callback with the result code.
323 663
664
324=item aio_symlink $srcpath, $dstpath, $callback->($status) 665=item aio_symlink $srcpath, $dstpath, $callback->($status)
325 666
326Asynchronously create a new symbolic link to the existing object at C<$srcpath> at 667Asynchronously create a new symbolic link to the existing object at C<$srcpath> at
327the path C<$dstpath> and call the callback with the result code. 668the path C<$dstpath> and call the callback with the result code.
328 669
670
671=item aio_readlink $pathname, $callback->($link)
672
673Asynchronously read the symlink specified by C<$path> and pass it to
674the callback. If an error occurs, nothing or undef gets passed to the
675callback.
676
677
678=item aio_realpath $pathname, $callback->($path)
679
680Asynchronously make the path absolute and resolve any symlinks in
681C<$path>. The resulting path only consists of directories (Same as
682L<Cwd::realpath>).
683
684This request can be used to get the absolute path of the current working
685directory by passing it a path of F<.> (a single dot).
686
687
329=item aio_rename $srcpath, $dstpath, $callback->($status) 688=item aio_rename $srcpath, $dstpath, $callback->($status)
330 689
331Asynchronously rename the object at C<$srcpath> to C<$dstpath>, just as 690Asynchronously rename the object at C<$srcpath> to C<$dstpath>, just as
332rename(2) and call the callback with the result code. 691rename(2) and call the callback with the result code.
333 692
693
694=item aio_mkdir $pathname, $mode, $callback->($status)
695
696Asynchronously mkdir (create) a directory and call the callback with
697the result code. C<$mode> will be modified by the umask at the time the
698request is executed, so do not change your umask.
699
700
334=item aio_rmdir $pathname, $callback->($status) 701=item aio_rmdir $pathname, $callback->($status)
335 702
336Asynchronously rmdir (delete) a directory and call the callback with the 703Asynchronously rmdir (delete) a directory and call the callback with the
337result code. 704result code.
705
338 706
339=item aio_readdir $pathname, $callback->($entries) 707=item aio_readdir $pathname, $callback->($entries)
340 708
341Unlike the POSIX call of the same name, C<aio_readdir> reads an entire 709Unlike the POSIX call of the same name, C<aio_readdir> reads an entire
342directory (i.e. opendir + readdir + closedir). The entries will not be 710directory (i.e. opendir + readdir + closedir). The entries will not be
343sorted, and will B<NOT> include the C<.> and C<..> entries. 711sorted, and will B<NOT> include the C<.> and C<..> entries.
344 712
345The callback a single argument which is either C<undef> or an array-ref 713The callback is passed a single argument which is either C<undef> or an
346with the filenames. 714array-ref with the filenames.
347 715
716
717=item aio_readdirx $pathname, $flags, $callback->($entries, $flags)
718
719Quite similar to C<aio_readdir>, but the C<$flags> argument allows one to
720tune behaviour and output format. In case of an error, C<$entries> will be
721C<undef>.
722
723The flags are a combination of the following constants, ORed together (the
724flags will also be passed to the callback, possibly modified):
725
726=over 4
727
728=item IO::AIO::READDIR_DENTS
729
730When this flag is off, then the callback gets an arrayref consisting of
731names only (as with C<aio_readdir>), otherwise it gets an arrayref with
732C<[$name, $type, $inode]> arrayrefs, each describing a single directory
733entry in more detail.
734
735C<$name> is the name of the entry.
736
737C<$type> is one of the C<IO::AIO::DT_xxx> constants:
738
739C<IO::AIO::DT_UNKNOWN>, C<IO::AIO::DT_FIFO>, C<IO::AIO::DT_CHR>, C<IO::AIO::DT_DIR>,
740C<IO::AIO::DT_BLK>, C<IO::AIO::DT_REG>, C<IO::AIO::DT_LNK>, C<IO::AIO::DT_SOCK>,
741C<IO::AIO::DT_WHT>.
742
743C<IO::AIO::DT_UNKNOWN> means just that: readdir does not know. If you need to
744know, you have to run stat yourself. Also, for speed reasons, the C<$type>
745scalars are read-only: you can not modify them.
746
747C<$inode> is the inode number (which might not be exact on systems with 64
748bit inode numbers and 32 bit perls). This field has unspecified content on
749systems that do not deliver the inode information.
750
751=item IO::AIO::READDIR_DIRS_FIRST
752
753When this flag is set, then the names will be returned in an order where
754likely directories come first, in optimal stat order. This is useful when
755you need to quickly find directories, or you want to find all directories
756while avoiding to stat() each entry.
757
758If the system returns type information in readdir, then this is used
759to find directories directly. Otherwise, likely directories are names
760beginning with ".", or otherwise names with no dots, of which names with
761short names are tried first.
762
763=item IO::AIO::READDIR_STAT_ORDER
764
765When this flag is set, then the names will be returned in an order
766suitable for stat()'ing each one. That is, when you plan to stat()
767all files in the given directory, then the returned order will likely
768be fastest.
769
770If both this flag and C<IO::AIO::READDIR_DIRS_FIRST> are specified, then
771the likely dirs come first, resulting in a less optimal stat order.
772
773=item IO::AIO::READDIR_FOUND_UNKNOWN
774
775This flag should not be set when calling C<aio_readdirx>. Instead, it
776is being set by C<aio_readdirx>, when any of the C<$type>'s found were
777C<IO::AIO::DT_UNKNOWN>. The absence of this flag therefore indicates that all
778C<$type>'s are known, which can be used to speed up some algorithms.
779
780=back
781
782
783=item aio_load $pathname, $data, $callback->($status)
784
785This is a composite request that tries to fully load the given file into
786memory. Status is the same as with aio_read.
787
788=cut
789
790sub aio_load($$;$) {
791 my ($path, undef, $cb) = @_;
792 my $data = \$_[1];
793
794 my $pri = aioreq_pri;
795 my $grp = aio_group $cb;
796
797 aioreq_pri $pri;
798 add $grp aio_open $path, O_RDONLY, 0, sub {
799 my $fh = shift
800 or return $grp->result (-1);
801
802 aioreq_pri $pri;
803 add $grp aio_read $fh, 0, (-s $fh), $$data, 0, sub {
804 $grp->result ($_[0]);
805 };
806 };
807
808 $grp
809}
810
811=item aio_copy $srcpath, $dstpath, $callback->($status)
812
813Try to copy the I<file> (directories not supported as either source or
814destination) from C<$srcpath> to C<$dstpath> and call the callback with
815a status of C<0> (ok) or C<-1> (error, see C<$!>).
816
817This is a composite request that creates the destination file with
818mode 0200 and copies the contents of the source file into it using
819C<aio_sendfile>, followed by restoring atime, mtime, access mode and
820uid/gid, in that order.
821
822If an error occurs, the partial destination file will be unlinked, if
823possible, except when setting atime, mtime, access mode and uid/gid, where
824errors are being ignored.
825
826=cut
827
828sub aio_copy($$;$) {
829 my ($src, $dst, $cb) = @_;
830
831 my $pri = aioreq_pri;
832 my $grp = aio_group $cb;
833
834 aioreq_pri $pri;
835 add $grp aio_open $src, O_RDONLY, 0, sub {
836 if (my $src_fh = $_[0]) {
837 my @stat = stat $src_fh; # hmm, might block over nfs?
838
839 aioreq_pri $pri;
840 add $grp aio_open $dst, O_CREAT | O_WRONLY | O_TRUNC, 0200, sub {
841 if (my $dst_fh = $_[0]) {
842 aioreq_pri $pri;
843 add $grp aio_sendfile $dst_fh, $src_fh, 0, $stat[7], sub {
844 if ($_[0] == $stat[7]) {
845 $grp->result (0);
846 close $src_fh;
847
848 my $ch = sub {
849 aioreq_pri $pri;
850 add $grp aio_chmod $dst_fh, $stat[2] & 07777, sub {
851 aioreq_pri $pri;
852 add $grp aio_chown $dst_fh, $stat[4], $stat[5], sub {
853 aioreq_pri $pri;
854 add $grp aio_close $dst_fh;
855 }
856 };
857 };
858
859 aioreq_pri $pri;
860 add $grp aio_utime $dst_fh, $stat[8], $stat[9], sub {
861 if ($_[0] < 0 && $! == ENOSYS) {
862 aioreq_pri $pri;
863 add $grp aio_utime $dst, $stat[8], $stat[9], $ch;
864 } else {
865 $ch->();
866 }
867 };
868 } else {
869 $grp->result (-1);
870 close $src_fh;
871 close $dst_fh;
872
873 aioreq $pri;
874 add $grp aio_unlink $dst;
875 }
876 };
877 } else {
878 $grp->result (-1);
879 }
880 },
881
882 } else {
883 $grp->result (-1);
884 }
885 };
886
887 $grp
888}
889
890=item aio_move $srcpath, $dstpath, $callback->($status)
891
892Try to move the I<file> (directories not supported as either source or
893destination) from C<$srcpath> to C<$dstpath> and call the callback with
894a status of C<0> (ok) or C<-1> (error, see C<$!>).
895
896This is a composite request that tries to rename(2) the file first; if
897rename fails with C<EXDEV>, it copies the file with C<aio_copy> and, if
898that is successful, unlinks the C<$srcpath>.
899
900=cut
901
902sub aio_move($$;$) {
903 my ($src, $dst, $cb) = @_;
904
905 my $pri = aioreq_pri;
906 my $grp = aio_group $cb;
907
908 aioreq_pri $pri;
909 add $grp aio_rename $src, $dst, sub {
910 if ($_[0] && $! == EXDEV) {
911 aioreq_pri $pri;
912 add $grp aio_copy $src, $dst, sub {
913 $grp->result ($_[0]);
914
915 unless ($_[0]) {
916 aioreq_pri $pri;
917 add $grp aio_unlink $src;
918 }
919 };
920 } else {
921 $grp->result ($_[0]);
922 }
923 };
924
925 $grp
926}
927
348=item aio_scandir $path, $maxreq, $callback->($dirs, $nondirs) 928=item aio_scandir $pathname, $maxreq, $callback->($dirs, $nondirs)
349
350[EXPERIMENTAL due to internal aio_group use]
351 929
352Scans a directory (similar to C<aio_readdir>) but additionally tries to 930Scans a directory (similar to C<aio_readdir>) but additionally tries to
353separate the entries of directory C<$path> into two sets of names, ones 931efficiently separate the entries of directory C<$path> into two sets of
354you can recurse into (directories or links to them), and ones you cannot 932names, directories you can recurse into (directories), and ones you cannot
355recurse into (everything else). 933recurse into (everything else, including symlinks to directories).
356 934
357C<aio_scandir> is a composite request that creates of many sub requests_ 935C<aio_scandir> is a composite request that creates of many sub requests_
358C<$maxreq> specifies the maximum number of outstanding aio requests that 936C<$maxreq> specifies the maximum number of outstanding aio requests that
359this function generates. If it is C<< <= 0 >>, then a suitable default 937this function generates. If it is C<< <= 0 >>, then a suitable default
360will be chosen (currently 6). 938will be chosen (currently 4).
361 939
362On error, the callback is called without arguments, otherwise it receives 940On error, the callback is called without arguments, otherwise it receives
363two array-refs with path-relative entry names. 941two array-refs with path-relative entry names.
364 942
365Example: 943Example:
372 950
373Implementation notes. 951Implementation notes.
374 952
375The C<aio_readdir> cannot be avoided, but C<stat()>'ing every entry can. 953The C<aio_readdir> cannot be avoided, but C<stat()>'ing every entry can.
376 954
955If readdir returns file type information, then this is used directly to
956find directories.
957
377After reading the directory, the modification time, size etc. of the 958Otherwise, after reading the directory, the modification time, size etc.
378directory before and after the readdir is checked, and if they match (and 959of the directory before and after the readdir is checked, and if they
379isn't the current time), the link count will be used to decide how many 960match (and isn't the current time), the link count will be used to decide
380entries are directories (if >= 2). Otherwise, no knowledge of the number 961how many entries are directories (if >= 2). Otherwise, no knowledge of the
381of subdirectories will be assumed. 962number of subdirectories will be assumed.
382 963
383Then entries will be sorted into likely directories (everything without 964Then entries will be sorted into likely directories a non-initial dot
384a non-initial dot currently) and likely non-directories (everything 965currently) and likely non-directories (see C<aio_readdirx>). Then every
385else). Then every entry plus an appended C</.> will be C<stat>'ed, 966entry plus an appended C</.> will be C<stat>'ed, likely directories first,
386likely directories first. If that succeeds, it assumes that the entry 967in order of their inode numbers. If that succeeds, it assumes that the
387is a directory or a symlink to directory (which will be checked 968entry is a directory or a symlink to directory (which will be checked
388seperately). This is often faster than stat'ing the entry itself because 969separately). This is often faster than stat'ing the entry itself because
389filesystems might detect the type of the entry without reading the inode 970filesystems might detect the type of the entry without reading the inode
390data (e.g. ext2fs filetype feature). 971data (e.g. ext2fs filetype feature), even on systems that cannot return
972the filetype information on readdir.
391 973
392If the known number of directories (link count - 2) has been reached, the 974If the known number of directories (link count - 2) has been reached, the
393rest of the entries is assumed to be non-directories. 975rest of the entries is assumed to be non-directories.
394 976
395This only works with certainty on POSIX (= UNIX) filesystems, which 977This only works with certainty on POSIX (= UNIX) filesystems, which
399as those tend to return 0 or 1 as link counts, which disables the 981as those tend to return 0 or 1 as link counts, which disables the
400directory counting heuristic. 982directory counting heuristic.
401 983
402=cut 984=cut
403 985
404sub aio_scandir($$$) { 986sub aio_scandir($$;$) {
405 my ($path, $maxreq, $cb) = @_; 987 my ($path, $maxreq, $cb) = @_;
406 988
989 my $pri = aioreq_pri;
990
407 my $grp = aio_group $cb; 991 my $grp = aio_group $cb;
408 992
409 $maxreq = 6 if $maxreq <= 0; 993 $maxreq = 4 if $maxreq <= 0;
410 994
411 # stat once 995 # get a wd object
996 aioreq_pri $pri;
412 add $grp aio_stat $path, sub { 997 add $grp aio_wd $path, sub {
998 $_[0]
999 or return $grp->result ();
1000
1001 my $wd = [shift, "."];
1002
1003 # stat once
1004 aioreq_pri $pri;
1005 add $grp aio_stat $wd, sub {
413 return $grp->result () if $_[0]; 1006 return $grp->result () if $_[0];
414 my $now = time; 1007 my $now = time;
415 my $hash1 = join ":", (stat _)[0,1,3,7,9]; 1008 my $hash1 = join ":", (stat _)[0,1,3,7,9];
416 1009
417 # read the directory entries 1010 # read the directory entries
418 add $grp aio_readdir $path, sub { 1011 aioreq_pri $pri;
1012 add $grp aio_readdirx $wd, READDIR_DIRS_FIRST, sub {
419 my $entries = shift 1013 my $entries = shift
420 or return $grp->result (); 1014 or return $grp->result ();
421 1015
422 # stat the dir another time 1016 # stat the dir another time
1017 aioreq_pri $pri;
423 add $grp aio_stat $path, sub { 1018 add $grp aio_stat $wd, sub {
424 my $hash2 = join ":", (stat _)[0,1,3,7,9]; 1019 my $hash2 = join ":", (stat _)[0,1,3,7,9];
425 1020
426 my $ndirs; 1021 my $ndirs;
427 1022
428 # take the slow route if anything looks fishy 1023 # take the slow route if anything looks fishy
429 if ($hash1 ne $hash2 or (stat _)[9] == $now) { 1024 if ($hash1 ne $hash2 or (stat _)[9] == $now) {
430 $ndirs = -1; 1025 $ndirs = -1;
431 } else { 1026 } else {
432 # if nlink == 2, we are finished 1027 # if nlink == 2, we are finished
433 # on non-posix-fs's, we rely on nlink < 2 1028 # for non-posix-fs's, we rely on nlink < 2
434 $ndirs = (stat _)[3] - 2 1029 $ndirs = (stat _)[3] - 2
435 or return $grp->result ([], $entries); 1030 or return $grp->result ([], $entries);
436 } 1031 }
437 1032
438 # sort into likely dirs and likely nondirs
439 # dirs == files without ".", short entries first
440 $entries = [map $_->[0],
441 sort { $b->[1] cmp $a->[1] }
442 map [$_, sprintf "%s%04d", (/.\./ ? "1" : "0"), length],
443 @$entries];
444
445 my (@dirs, @nondirs); 1033 my (@dirs, @nondirs);
446 1034
447 my ($statcb, $schedcb);
448 my $nreq = 0;
449
450 my $statgrp = add $grp aio_group; 1035 my $statgrp = add $grp aio_group sub {
451
452 $schedcb = sub {
453 if (@$entries) {
454 if ($nreq < $maxreq) {
455 my $ent = pop @$entries;
456 $nreq++;
457 add $statgrp aio_stat "$path/$ent/.", sub { $statcb->($_[0], $ent) };
458 }
459 } elsif (!$nreq) {
460 # finished
461 $statgrp->cancel;
462 undef $statcb;
463 undef $schedcb;
464 $grp->result (\@dirs, \@nondirs); 1036 $grp->result (\@dirs, \@nondirs);
465 } 1037 };
1038
1039 limit $statgrp $maxreq;
1040 feed $statgrp sub {
1041 return unless @$entries;
1042 my $entry = shift @$entries;
1043
1044 aioreq_pri $pri;
1045 $wd->[1] = "$entry/.";
1046 add $statgrp aio_stat $wd, sub {
1047 if ($_[0] < 0) {
1048 push @nondirs, $entry;
1049 } else {
1050 # need to check for real directory
1051 aioreq_pri $pri;
1052 $wd->[1] = $entry;
1053 add $statgrp aio_lstat $wd, sub {
1054 if (-d _) {
1055 push @dirs, $entry;
1056
1057 unless (--$ndirs) {
1058 push @nondirs, @$entries;
1059 feed $statgrp;
1060 }
1061 } else {
1062 push @nondirs, $entry;
1063 }
1064 }
1065 }
1066 };
1067 };
466 }; 1068 };
467 $statcb = sub {
468 my ($status, $entry) = @_;
469
470 if ($status < 0) {
471 $nreq--;
472 push @nondirs, $entry;
473 &$schedcb;
474 } else {
475 # need to check for real directory
476 add $grp aio_lstat "$path/$entry", sub {
477 $nreq--;
478
479 if (-d _) {
480 push @dirs, $entry;
481
482 if (!--$ndirs) {
483 push @nondirs, @$entries;
484 $entries = [];
485 }
486 } else {
487 push @nondirs, $entry;
488 }
489
490 &$schedcb;
491 }
492 }
493 };
494
495 &$schedcb while @$entries && $nreq < $maxreq;
496 }; 1069 };
497 }; 1070 };
498 }; 1071 };
499 1072
500 $grp 1073 $grp
501} 1074}
502 1075
1076=item aio_rmtree $pathname, $callback->($status)
1077
1078Delete a directory tree starting (and including) C<$path>, return the
1079status of the final C<rmdir> only. This is a composite request that
1080uses C<aio_scandir> to recurse into and rmdir directories, and unlink
1081everything else.
1082
1083=cut
1084
1085sub aio_rmtree;
1086sub aio_rmtree($;$) {
1087 my ($path, $cb) = @_;
1088
1089 my $pri = aioreq_pri;
1090 my $grp = aio_group $cb;
1091
1092 aioreq_pri $pri;
1093 add $grp aio_scandir $path, 0, sub {
1094 my ($dirs, $nondirs) = @_;
1095
1096 my $dirgrp = aio_group sub {
1097 add $grp aio_rmdir $path, sub {
1098 $grp->result ($_[0]);
1099 };
1100 };
1101
1102 (aioreq_pri $pri), add $dirgrp aio_rmtree "$path/$_" for @$dirs;
1103 (aioreq_pri $pri), add $dirgrp aio_unlink "$path/$_" for @$nondirs;
1104
1105 add $grp $dirgrp;
1106 };
1107
1108 $grp
1109}
1110
1111=item aio_sync $callback->($status)
1112
1113Asynchronously call sync and call the callback when finished.
1114
503=item aio_fsync $fh, $callback->($status) 1115=item aio_fsync $fh, $callback->($status)
504 1116
505Asynchronously call fsync on the given filehandle and call the callback 1117Asynchronously call fsync on the given filehandle and call the callback
506with the fsync result code. 1118with the fsync result code.
507 1119
511callback with the fdatasync result code. 1123callback with the fdatasync result code.
512 1124
513If this call isn't available because your OS lacks it or it couldn't be 1125If this call isn't available because your OS lacks it or it couldn't be
514detected, it will be emulated by calling C<fsync> instead. 1126detected, it will be emulated by calling C<fsync> instead.
515 1127
1128=item aio_syncfs $fh, $callback->($status)
1129
1130Asynchronously call the syncfs syscall to sync the filesystem associated
1131to the given filehandle and call the callback with the syncfs result
1132code. If syncfs is not available, calls sync(), but returns C<-1> and sets
1133errno to C<ENOSYS> nevertheless.
1134
1135=item aio_sync_file_range $fh, $offset, $nbytes, $flags, $callback->($status)
1136
1137Sync the data portion of the file specified by C<$offset> and C<$length>
1138to disk (but NOT the metadata), by calling the Linux-specific
1139sync_file_range call. If sync_file_range is not available or it returns
1140ENOSYS, then fdatasync or fsync is being substituted.
1141
1142C<$flags> can be a combination of C<IO::AIO::SYNC_FILE_RANGE_WAIT_BEFORE>,
1143C<IO::AIO::SYNC_FILE_RANGE_WRITE> and
1144C<IO::AIO::SYNC_FILE_RANGE_WAIT_AFTER>: refer to the sync_file_range
1145manpage for details.
1146
1147=item aio_pathsync $pathname, $callback->($status)
1148
1149This request tries to open, fsync and close the given path. This is a
1150composite request intended to sync directories after directory operations
1151(E.g. rename). This might not work on all operating systems or have any
1152specific effect, but usually it makes sure that directory changes get
1153written to disc. It works for anything that can be opened for read-only,
1154not just directories.
1155
1156Future versions of this function might fall back to other methods when
1157C<fsync> on the directory fails (such as calling C<sync>).
1158
1159Passes C<0> when everything went ok, and C<-1> on error.
1160
1161=cut
1162
1163sub aio_pathsync($;$) {
1164 my ($path, $cb) = @_;
1165
1166 my $pri = aioreq_pri;
1167 my $grp = aio_group $cb;
1168
1169 aioreq_pri $pri;
1170 add $grp aio_open $path, O_RDONLY, 0, sub {
1171 my ($fh) = @_;
1172 if ($fh) {
1173 aioreq_pri $pri;
1174 add $grp aio_fsync $fh, sub {
1175 $grp->result ($_[0]);
1176
1177 aioreq_pri $pri;
1178 add $grp aio_close $fh;
1179 };
1180 } else {
1181 $grp->result (-1);
1182 }
1183 };
1184
1185 $grp
1186}
1187
1188=item aio_msync $scalar, $offset = 0, $length = undef, flags = 0, $callback->($status)
1189
1190This is a rather advanced IO::AIO call, which only works on mmap(2)ed
1191scalars (see the C<IO::AIO::mmap> function, although it also works on data
1192scalars managed by the L<Sys::Mmap> or L<Mmap> modules, note that the
1193scalar must only be modified in-place while an aio operation is pending on
1194it).
1195
1196It calls the C<msync> function of your OS, if available, with the memory
1197area starting at C<$offset> in the string and ending C<$length> bytes
1198later. If C<$length> is negative, counts from the end, and if C<$length>
1199is C<undef>, then it goes till the end of the string. The flags can be
1200a combination of C<IO::AIO::MS_ASYNC>, C<IO::AIO::MS_INVALIDATE> and
1201C<IO::AIO::MS_SYNC>.
1202
1203=item aio_mtouch $scalar, $offset = 0, $length = undef, flags = 0, $callback->($status)
1204
1205This is a rather advanced IO::AIO call, which works best on mmap(2)ed
1206scalars.
1207
1208It touches (reads or writes) all memory pages in the specified
1209range inside the scalar. All caveats and parameters are the same
1210as for C<aio_msync>, above, except for flags, which must be either
1211C<0> (which reads all pages and ensures they are instantiated) or
1212C<IO::AIO::MT_MODIFY>, which modifies the memory page s(by reading and
1213writing an octet from it, which dirties the page).
1214
1215=item aio_mlock $scalar, $offset = 0, $length = undef, $callback->($status)
1216
1217This is a rather advanced IO::AIO call, which works best on mmap(2)ed
1218scalars.
1219
1220It reads in all the pages of the underlying storage into memory (if any)
1221and locks them, so they are not getting swapped/paged out or removed.
1222
1223If C<$length> is undefined, then the scalar will be locked till the end.
1224
1225On systems that do not implement C<mlock>, this function returns C<-1>
1226and sets errno to C<ENOSYS>.
1227
1228Note that the corresponding C<munlock> is synchronous and is
1229documented under L<MISCELLANEOUS FUNCTIONS>.
1230
1231Example: open a file, mmap and mlock it - both will be undone when
1232C<$data> gets destroyed.
1233
1234 open my $fh, "<", $path or die "$path: $!";
1235 my $data;
1236 IO::AIO::mmap $data, -s $fh, IO::AIO::PROT_READ, IO::AIO::MAP_SHARED, $fh;
1237 aio_mlock $data; # mlock in background
1238
1239=item aio_mlockall $flags, $callback->($status)
1240
1241Calls the C<mlockall> function with the given C<$flags> (a combination of
1242C<IO::AIO::MCL_CURRENT> and C<IO::AIO::MCL_FUTURE>).
1243
1244On systems that do not implement C<mlockall>, this function returns C<-1>
1245and sets errno to C<ENOSYS>.
1246
1247Note that the corresponding C<munlockall> is synchronous and is
1248documented under L<MISCELLANEOUS FUNCTIONS>.
1249
1250Example: asynchronously lock all current and future pages into memory.
1251
1252 aio_mlockall IO::AIO::MCL_FUTURE;
1253
1254=item aio_fiemap $fh, $start, $length, $flags, $count, $cb->(\@extents)
1255
1256Queries the extents of the given file (by calling the Linux FIEMAP ioctl,
1257see L<http://cvs.schmorp.de/IO-AIO/doc/fiemap.txt> for details). If the
1258C<ioctl> is not available on your OS, then this rquiest will fail with
1259C<ENOSYS>.
1260
1261C<$start> is the starting offset to query extents for, C<$length> is the
1262size of the range to query - if it is C<undef>, then the whole file will
1263be queried.
1264
1265C<$flags> is a combination of flags (C<IO::AIO::FIEMAP_FLAG_SYNC> or
1266C<IO::AIO::FIEMAP_FLAG_XATTR> - C<IO::AIO::FIEMAP_FLAGS_COMPAT> is also
1267exported), and is normally C<0> or C<IO::AIO::FIEMAP_FLAG_SYNC> to query
1268the data portion.
1269
1270C<$count> is the maximum number of extent records to return. If it is
1271C<undef>, then IO::AIO queries all extents of the file. As a very special
1272case, if it is C<0>, then the callback receives the number of extents
1273instead of the extents themselves.
1274
1275If an error occurs, the callback receives no arguments. The special
1276C<errno> value C<IO::AIO::EBADR> is available to test for flag errors.
1277
1278Otherwise, the callback receives an array reference with extent
1279structures. Each extent structure is an array reference itself, with the
1280following members:
1281
1282 [$logical, $physical, $length, $flags]
1283
1284Flags is any combination of the following flag values (typically either C<0>
1285or C<IO::AIO::FIEMAP_EXTENT_LAST>):
1286
1287C<IO::AIO::FIEMAP_EXTENT_LAST>, C<IO::AIO::FIEMAP_EXTENT_UNKNOWN>,
1288C<IO::AIO::FIEMAP_EXTENT_DELALLOC>, C<IO::AIO::FIEMAP_EXTENT_ENCODED>,
1289C<IO::AIO::FIEMAP_EXTENT_DATA_ENCRYPTED>, C<IO::AIO::FIEMAP_EXTENT_NOT_ALIGNED>,
1290C<IO::AIO::FIEMAP_EXTENT_DATA_INLINE>, C<IO::AIO::FIEMAP_EXTENT_DATA_TAIL>,
1291C<IO::AIO::FIEMAP_EXTENT_UNWRITTEN>, C<IO::AIO::FIEMAP_EXTENT_MERGED> or
1292C<IO::AIO::FIEMAP_EXTENT_SHARED>.
1293
516=item aio_group $callback->(...) 1294=item aio_group $callback->(...)
517
518[EXPERIMENTAL]
519 1295
520This is a very special aio request: Instead of doing something, it is a 1296This is a very special aio request: Instead of doing something, it is a
521container for other aio requests, which is useful if you want to bundle 1297container for other aio requests, which is useful if you want to bundle
522many requests into a single, composite, request. 1298many requests into a single, composite, request with a definite callback
1299and the ability to cancel the whole request with its subrequests.
523 1300
524Returns an object of class L<IO::AIO::GRP>. See its documentation below 1301Returns an object of class L<IO::AIO::GRP>. See its documentation below
525for more info. 1302for more info.
526 1303
527Example: 1304Example:
546phase and still requires a worker thread. Thus, the callback will not 1323phase and still requires a worker thread. Thus, the callback will not
547be executed immediately but only after other requests in the queue have 1324be executed immediately but only after other requests in the queue have
548entered their execution phase. This can be used to measure request 1325entered their execution phase. This can be used to measure request
549latency. 1326latency.
550 1327
551=item IO::AIO::aio_sleep $fractional_seconds, $callback->() *NOT EXPORTED* 1328=item IO::AIO::aio_busy $fractional_seconds, $callback->() *NOT EXPORTED*
552 1329
553Mainly used for debugging and benchmarking, this aio request puts one of 1330Mainly used for debugging and benchmarking, this aio request puts one of
554the request workers to sleep for the given time. 1331the request workers to sleep for the given time.
555 1332
556While it is theoretically handy to have simple I/O scheduling requests 1333While it is theoretically handy to have simple I/O scheduling requests
557like sleep and file handle readable/writable, the overhead this creates 1334like sleep and file handle readable/writable, the overhead this creates is
558is immense, so do not use this function except to put your application 1335immense (it blocks a thread for a long time) so do not use this function
559under artificial I/O pressure. 1336except to put your application under artificial I/O pressure.
560 1337
561=back 1338=back
1339
1340
1341=head2 IO::AIO::WD - multiple working directories
1342
1343Your process only has one current working directory, which is used by all
1344threads. This makes it hard to use relative paths (some other component
1345could call C<chdir> at any time, and it is hard to control when the path
1346will be used by IO::AIO).
1347
1348One solution for this is to always use absolute paths. This usually works,
1349but can be quite slow (the kernel has to walk the whole path on every
1350access), and can also be a hassle to implement.
1351
1352Newer POSIX systems have a number of functions (openat, fdopendir,
1353futimensat and so on) that make it possible to specify working directories
1354per operation.
1355
1356For portability, and because the clowns who "designed", or shall I write,
1357perpetrated this new interface were obviously half-drunk, this abstraction
1358cannot be perfect, though.
1359
1360IO::AIO allows you to convert directory paths into a so-called IO::AIO::WD
1361object. This object stores the canonicalised, absolute version of the
1362path, and on systems that allow it, also a directory file descriptor.
1363
1364Everywhere where a pathname is accepted by IO::AIO (e.g. in C<aio_stat>
1365or C<aio_unlink>), one can specify an array reference with an IO::AIO::WD
1366object and a pathname instead (or the IO::AIO::WD object alone, which
1367gets interpreted as C<[$wd, "."]>). If the pathname is absolute, the
1368IO::AIO::WD object is ignored, otherwise the pathname is resolved relative
1369to that IO::AIO::WD object.
1370
1371For example, to get a wd object for F</etc> and then stat F<passwd>
1372inside, you would write:
1373
1374 aio_wd "/etc", sub {
1375 my $etcdir = shift;
1376
1377 # although $etcdir can be undef on error, there is generally no reason
1378 # to check for errors here, as aio_stat will fail with ENOENT
1379 # when $etcdir is undef.
1380
1381 aio_stat [$etcdir, "passwd"], sub {
1382 # yay
1383 };
1384 };
1385
1386That C<aio_wd> is a request and not a normal function shows that creating
1387an IO::AIO::WD object is itself a potentially blocking operation, which is
1388why it is done asynchronously.
1389
1390To stat the directory obtained with C<aio_wd> above, one could write
1391either of the following three request calls:
1392
1393 aio_lstat "/etc" , sub { ... # pathname as normal string
1394 aio_lstat [$wd, "."], sub { ... # "." relative to $wd (i.e. $wd itself)
1395 aio_lstat $wd , sub { ... # shorthand for the previous
1396
1397As with normal pathnames, IO::AIO keeps a copy of the working directory
1398object and the pathname string, so you could write the following without
1399causing any issues due to C<$path> getting reused:
1400
1401 my $path = [$wd, undef];
1402
1403 for my $name (qw(abc def ghi)) {
1404 $path->[1] = $name;
1405 aio_stat $path, sub {
1406 # ...
1407 };
1408 }
1409
1410There are some caveats: when directories get renamed (or deleted), the
1411pathname string doesn't change, so will point to the new directory (or
1412nowhere at all), while the directory fd, if available on the system,
1413will still point to the original directory. Most functions accepting a
1414pathname will use the directory fd on newer systems, and the string on
1415older systems. Some functions (such as realpath) will always rely on the
1416string form of the pathname.
1417
1418So this fucntionality is mainly useful to get some protection against
1419C<chdir>, to easily get an absolute path out of a relative path for future
1420reference, and to speed up doing many operations in the same directory
1421(e.g. when stat'ing all files in a directory).
1422
1423The following functions implement this working directory abstraction:
1424
1425=over 4
1426
1427=item aio_wd $pathname, $callback->($wd)
1428
1429Asynchonously canonicalise the given pathname and convert it to an
1430IO::AIO::WD object representing it. If possible and supported on the
1431system, also open a directory fd to speed up pathname resolution relative
1432to this working directory.
1433
1434If something goes wrong, then C<undef> is passwd to the callback instead
1435of a working directory object and C<$!> is set appropriately. Since
1436passing C<undef> as working directory component of a pathname fails the
1437request with C<ENOENT>, there is often no need for error checking in the
1438C<aio_wd> callback, as future requests using the value will fail in the
1439expected way.
1440
1441If this call isn't available because your OS lacks it or it couldn't be
1442detected, it will be emulated by calling C<fsync> instead.
1443
1444=item IO::AIO::CWD
1445
1446This is a compiletime constant (object) that represents the process
1447current working directory.
1448
1449Specifying this object as working directory object for a pathname is as
1450if the pathname would be specified directly, without a directory object,
1451e.g., these calls are functionally identical:
1452
1453 aio_stat "somefile", sub { ... };
1454 aio_stat [IO::AIO::CWD, "somefile"], sub { ... };
1455
1456=back
1457
562 1458
563=head2 IO::AIO::REQ CLASS 1459=head2 IO::AIO::REQ CLASS
564 1460
565All non-aggregate C<aio_*> functions return an object of this class when 1461All non-aggregate C<aio_*> functions return an object of this class when
566called in non-void context. 1462called in non-void context.
567
568A request always moves through the following five states in its lifetime,
569in order: B<ready> (request has been created, but has not been executed
570yet), B<execute> (request is currently being executed), B<pending>
571(request has been executed but callback has not been called yet),
572B<result> (results are being processed synchronously, includes calling the
573callback) and B<done> (request has reached the end of its lifetime and
574holds no resources anymore).
575 1463
576=over 4 1464=over 4
577 1465
578=item cancel $req 1466=item cancel $req
579 1467
580Cancels the request, if possible. Has the effect of skipping execution 1468Cancels the request, if possible. Has the effect of skipping execution
581when entering the B<execute> state and skipping calling the callback when 1469when entering the B<execute> state and skipping calling the callback when
582entering the the B<result> state, but will leave the request otherwise 1470entering the the B<result> state, but will leave the request otherwise
583untouched. That means that requests that currently execute will not be 1471untouched (with the exception of readdir). That means that requests that
584stopped and resources held by the request will not be freed prematurely. 1472currently execute will not be stopped and resources held by the request
1473will not be freed prematurely.
585 1474
586=item cb $req $callback->(...) 1475=item cb $req $callback->(...)
587 1476
588Replace (or simply set) the callback registered to the request. 1477Replace (or simply set) the callback registered to the request.
589 1478
633=item * They can also can also be added to other IO::AIO::GRP objects. 1522=item * They can also can also be added to other IO::AIO::GRP objects.
634 1523
635=item * You must not add requests to a group from within the group callback (or 1524=item * You must not add requests to a group from within the group callback (or
636any later time). 1525any later time).
637 1526
638=item * This does not harmonise well with C<max_outstanding>, so best do
639not combine C<aio_group> with it. Groups and feeders are recommended for
640this kind of concurrency-limiting.
641
642=back 1527=back
643 1528
644Their lifetime, simplified, looks like this: when they are empty, they 1529Their lifetime, simplified, looks like this: when they are empty, they
645will finish very quickly. If they contain only requests that are in the 1530will finish very quickly. If they contain only requests that are in the
646C<done> state, they will also finish. Otherwise they will continue to 1531C<done> state, they will also finish. Otherwise they will continue to
647exist. 1532exist.
648 1533
649That means after creating a group you have some time to add requests. And 1534That means after creating a group you have some time to add requests
650in the callbacks of those requests, you can add further requests to the 1535(precisely before the callback has been invoked, which is only done within
651group. And only when all those requests have finished will the the group 1536the C<poll_cb>). And in the callbacks of those requests, you can add
652itself finish. 1537further requests to the group. And only when all those requests have
1538finished will the the group itself finish.
653 1539
654=over 4 1540=over 4
655 1541
656=item add $grp ... 1542=item add $grp ...
657 1543
661be added, including other groups, as long as you do not create circular 1547be added, including other groups, as long as you do not create circular
662dependencies. 1548dependencies.
663 1549
664Returns all its arguments. 1550Returns all its arguments.
665 1551
1552=item $grp->cancel_subs
1553
1554Cancel all subrequests and clears any feeder, but not the group request
1555itself. Useful when you queued a lot of events but got a result early.
1556
1557The group request will finish normally (you cannot add requests to the
1558group).
1559
666=item $grp->result (...) 1560=item $grp->result (...)
667 1561
668Set the result value(s) that will be passed to the group callback when all 1562Set the result value(s) that will be passed to the group callback when all
669subrequests have finished. By default, no argument will be passed. 1563subrequests have finished and set the groups errno to the current value
1564of errno (just like calling C<errno> without an error number). By default,
1565no argument will be passed and errno is zero.
1566
1567=item $grp->errno ([$errno])
1568
1569Sets the group errno value to C<$errno>, or the current value of errno
1570when the argument is missing.
1571
1572Every aio request has an associated errno value that is restored when
1573the callback is invoked. This method lets you change this value from its
1574default (0).
1575
1576Calling C<result> will also set errno, so make sure you either set C<$!>
1577before the call to C<result>, or call c<errno> after it.
670 1578
671=item feed $grp $callback->($grp) 1579=item feed $grp $callback->($grp)
672
673[VERY EXPERIMENTAL]
674 1580
675Sets a feeder/generator on this group: every group can have an attached 1581Sets a feeder/generator on this group: every group can have an attached
676generator that generates requests if idle. The idea behind this is that, 1582generator that generates requests if idle. The idea behind this is that,
677although you could just queue as many requests as you want in a group, 1583although you could just queue as many requests as you want in a group,
678this might starve other requests for a potentially long time. For 1584this might starve other requests for a potentially long time. For example,
679example, C<aio_scandir> might generate hundreds of thousands C<aio_stat> 1585C<aio_scandir> might generate hundreds of thousands of C<aio_stat>
680requests, delaying any later requests for a long time. 1586requests, delaying any later requests for a long time.
681 1587
682To avoid this, and allow incremental generation of requests, you can 1588To avoid this, and allow incremental generation of requests, you can
683instead a group and set a feeder on it that generates those requests. The 1589instead a group and set a feeder on it that generates those requests. The
684feed callback will be called whenever there are few enough (see C<feed_limit>, 1590feed callback will be called whenever there are few enough (see C<limit>,
685below) requests active in the group itself and is expected to queue more 1591below) requests active in the group itself and is expected to queue more
686requests. 1592requests.
687 1593
688The feed can queue as many requests as it likes (i.e. C<add> does not 1594The feed callback can queue as many requests as it likes (i.e. C<add> does
689impose any limits). 1595not impose any limits).
690 1596
691If the feed does not queue more requests when called, it will be 1597If the feed does not queue more requests when called, it will be
692automatically removed from the group. 1598automatically removed from the group.
693 1599
694If the feed limit is C<0>, it will be set to C<2> automatically. 1600If the feed limit is C<0> when this method is called, it will be set to
1601C<2> automatically.
695 1602
696Example: 1603Example:
697 1604
698 # stat all files in @files, but only ever use four aio requests concurrently: 1605 # stat all files in @files, but only ever use four aio requests concurrently:
699 1606
700 my $grp = aio_group sub { print "finished\n" }; 1607 my $grp = aio_group sub { print "finished\n" };
701 feed_limit $grp 4; 1608 limit $grp 4;
702 feed $grp sub { 1609 feed $grp sub {
703 my $file = pop @files 1610 my $file = pop @files
704 or return; 1611 or return;
705 1612
706 add $grp aio_stat $file, sub { ... }; 1613 add $grp aio_stat $file, sub { ... };
707 }; 1614 };
708 1615
709=item feed_limit $grp $num 1616=item limit $grp $num
710 1617
711Sets the feeder limit for the group: The feeder will be called whenever 1618Sets the feeder limit for the group: The feeder will be called whenever
712the group contains less than this many requests. 1619the group contains less than this many requests.
713 1620
714Setting the limit to C<0> will pause the feeding process. 1621Setting the limit to C<0> will pause the feeding process.
715 1622
1623The default value for the limit is C<0>, but note that setting a feeder
1624automatically bumps it up to C<2>.
1625
716=back 1626=back
717 1627
718=head2 SUPPORT FUNCTIONS 1628=head2 SUPPORT FUNCTIONS
719 1629
1630=head3 EVENT PROCESSING AND EVENT LOOP INTEGRATION
1631
720=over 4 1632=over 4
721 1633
722=item $fileno = IO::AIO::poll_fileno 1634=item $fileno = IO::AIO::poll_fileno
723 1635
724Return the I<request result pipe file descriptor>. This filehandle must be 1636Return the I<request result pipe file descriptor>. This filehandle must be
725polled for reading by some mechanism outside this module (e.g. Event or 1637polled for reading by some mechanism outside this module (e.g. EV, Glib,
726select, see below or the SYNOPSIS). If the pipe becomes readable you have 1638select and so on, see below or the SYNOPSIS). If the pipe becomes readable
727to call C<poll_cb> to check the results. 1639you have to call C<poll_cb> to check the results.
728 1640
729See C<poll_cb> for an example. 1641See C<poll_cb> for an example.
730 1642
731=item IO::AIO::poll_cb 1643=item IO::AIO::poll_cb
732 1644
733Process all outstanding events on the result pipe. You have to call this 1645Process some outstanding events on the result pipe. You have to call
734regularly. Returns the number of events processed. Returns immediately 1646this regularly. Returns C<0> if all events could be processed (or there
735when no events are outstanding. 1647were no events to process), or C<-1> if it returned earlier for whatever
1648reason. Returns immediately when no events are outstanding. The amount of
1649events processed depends on the settings of C<IO::AIO::max_poll_req> and
1650C<IO::AIO::max_poll_time>.
1651
1652If not all requests were processed for whatever reason, the filehandle
1653will still be ready when C<poll_cb> returns, so normally you don't have to
1654do anything special to have it called later.
1655
1656Apart from calling C<IO::AIO::poll_cb> when the event filehandle becomes
1657ready, it can be beneficial to call this function from loops which submit
1658a lot of requests, to make sure the results get processed when they become
1659available and not just when the loop is finished and the event loop takes
1660over again. This function returns very fast when there are no outstanding
1661requests.
736 1662
737Example: Install an Event watcher that automatically calls 1663Example: Install an Event watcher that automatically calls
738IO::AIO::poll_cb with high priority: 1664IO::AIO::poll_cb with high priority (more examples can be found in the
1665SYNOPSIS section, at the top of this document):
739 1666
740 Event->io (fd => IO::AIO::poll_fileno, 1667 Event->io (fd => IO::AIO::poll_fileno,
741 poll => 'r', async => 1, 1668 poll => 'r', async => 1,
742 cb => \&IO::AIO::poll_cb); 1669 cb => \&IO::AIO::poll_cb);
743 1670
744=item IO::AIO::poll_wait 1671=item IO::AIO::poll_wait
745 1672
1673If there are any outstanding requests and none of them in the result
746Wait till the result filehandle becomes ready for reading (simply does a 1674phase, wait till the result filehandle becomes ready for reading (simply
747C<select> on the filehandle. This is useful if you want to synchronously wait 1675does a C<select> on the filehandle. This is useful if you want to
748for some requests to finish). 1676synchronously wait for some requests to finish).
749 1677
750See C<nreqs> for an example. 1678See C<nreqs> for an example.
751 1679
1680=item IO::AIO::poll
1681
1682Waits until some requests have been handled.
1683
1684Returns the number of requests processed, but is otherwise strictly
1685equivalent to:
1686
1687 IO::AIO::poll_wait, IO::AIO::poll_cb
1688
752=item IO::AIO::nreqs 1689=item IO::AIO::flush
753 1690
754Returns the number of requests currently outstanding (i.e. for which their 1691Wait till all outstanding AIO requests have been handled.
755callback has not been invoked yet).
756 1692
757Example: wait till there are no outstanding requests anymore: 1693Strictly equivalent to:
758 1694
759 IO::AIO::poll_wait, IO::AIO::poll_cb 1695 IO::AIO::poll_wait, IO::AIO::poll_cb
760 while IO::AIO::nreqs; 1696 while IO::AIO::nreqs;
761 1697
762=item IO::AIO::flush 1698=item IO::AIO::max_poll_reqs $nreqs
763 1699
764Wait till all outstanding AIO requests have been handled. 1700=item IO::AIO::max_poll_time $seconds
765 1701
766Strictly equivalent to: 1702These set the maximum number of requests (default C<0>, meaning infinity)
1703that are being processed by C<IO::AIO::poll_cb> in one call, respectively
1704the maximum amount of time (default C<0>, meaning infinity) spent in
1705C<IO::AIO::poll_cb> to process requests (more correctly the mininum amount
1706of time C<poll_cb> is allowed to use).
767 1707
768 IO::AIO::poll_wait, IO::AIO::poll_cb 1708Setting C<max_poll_time> to a non-zero value creates an overhead of one
769 while IO::AIO::nreqs; 1709syscall per request processed, which is not normally a problem unless your
1710callbacks are really really fast or your OS is really really slow (I am
1711not mentioning Solaris here). Using C<max_poll_reqs> incurs no overhead.
770 1712
771=item IO::AIO::poll 1713Setting these is useful if you want to ensure some level of
1714interactiveness when perl is not fast enough to process all requests in
1715time.
772 1716
773Waits until some requests have been handled. 1717For interactive programs, values such as C<0.01> to C<0.1> should be fine.
774 1718
775Strictly equivalent to: 1719Example: Install an Event watcher that automatically calls
1720IO::AIO::poll_cb with low priority, to ensure that other parts of the
1721program get the CPU sometimes even under high AIO load.
776 1722
777 IO::AIO::poll_wait, IO::AIO::poll_cb 1723 # try not to spend much more than 0.1s in poll_cb
778 if IO::AIO::nreqs; 1724 IO::AIO::max_poll_time 0.1;
1725
1726 # use a low priority so other tasks have priority
1727 Event->io (fd => IO::AIO::poll_fileno,
1728 poll => 'r', nice => 1,
1729 cb => &IO::AIO::poll_cb);
1730
1731=back
1732
1733=head3 CONTROLLING THE NUMBER OF THREADS
1734
1735=over
779 1736
780=item IO::AIO::min_parallel $nthreads 1737=item IO::AIO::min_parallel $nthreads
781 1738
782Set the minimum number of AIO threads to C<$nthreads>. The current 1739Set the minimum number of AIO threads to C<$nthreads>. The current
783default is C<8>, which means eight asynchronous operations can execute 1740default is C<8>, which means eight asynchronous operations can execute
784concurrently at any one time (the number of outstanding requests, 1741concurrently at any one time (the number of outstanding requests,
785however, is unlimited). 1742however, is unlimited).
786 1743
787IO::AIO starts threads only on demand, when an AIO request is queued and 1744IO::AIO starts threads only on demand, when an AIO request is queued and
788no free thread exists. 1745no free thread exists. Please note that queueing up a hundred requests can
1746create demand for a hundred threads, even if it turns out that everything
1747is in the cache and could have been processed faster by a single thread.
789 1748
790It is recommended to keep the number of threads relatively low, as some 1749It is recommended to keep the number of threads relatively low, as some
791Linux kernel versions will scale negatively with the number of threads 1750Linux kernel versions will scale negatively with the number of threads
792(higher parallelity => MUCH higher latency). With current Linux 2.6 1751(higher parallelity => MUCH higher latency). With current Linux 2.6
793versions, 4-32 threads should be fine. 1752versions, 4-32 threads should be fine.
807This module automatically runs C<max_parallel 0> at program end, to ensure 1766This module automatically runs C<max_parallel 0> at program end, to ensure
808that all threads are killed and that there are no outstanding requests. 1767that all threads are killed and that there are no outstanding requests.
809 1768
810Under normal circumstances you don't need to call this function. 1769Under normal circumstances you don't need to call this function.
811 1770
1771=item IO::AIO::max_idle $nthreads
1772
1773Limit the number of threads (default: 4) that are allowed to idle
1774(i.e., threads that did not get a request to process within the idle
1775timeout (default: 10 seconds). That means if a thread becomes idle while
1776C<$nthreads> other threads are also idle, it will free its resources and
1777exit.
1778
1779This is useful when you allow a large number of threads (e.g. 100 or 1000)
1780to allow for extremely high load situations, but want to free resources
1781under normal circumstances (1000 threads can easily consume 30MB of RAM).
1782
1783The default is probably ok in most situations, especially if thread
1784creation is fast. If thread creation is very slow on your system you might
1785want to use larger values.
1786
1787=item IO::AIO::idle_timeout $seconds
1788
1789Sets the minimum idle timeout (default 10) after which worker threads are
1790allowed to exit. SEe C<IO::AIO::max_idle>.
1791
812=item $oldnreqs = IO::AIO::max_outstanding $nreqs 1792=item IO::AIO::max_outstanding $maxreqs
813 1793
814[DEPRECATED]
815
816Sets the maximum number of outstanding requests to C<$nreqs>. If you 1794Sets the maximum number of outstanding requests to C<$nreqs>. If
817try to queue up more than this number of requests, the caller will block until 1795you do queue up more than this number of requests, the next call to
818some requests have been handled. 1796C<IO::AIO::poll_cb> (and other functions calling C<poll_cb>, such as
1797C<IO::AIO::flush> or C<IO::AIO::poll>) will block until the limit is no
1798longer exceeded.
819 1799
820The default is very large, so normally there is no practical limit. If you 1800In other words, this setting does not enforce a queue limit, but can be
821queue up many requests in a loop it often improves speed if you set 1801used to make poll functions block if the limit is exceeded.
822this to a relatively low number, such as C<100>.
823 1802
824This function does not work well together with C<aio_group>'s, and their 1803This is a very bad function to use in interactive programs because it
825feeder interface is better suited to limiting concurrency, so do not use 1804blocks, and a bad way to reduce concurrency because it is inexact: Better
826this function. 1805use an C<aio_group> together with a feed callback.
827 1806
828Under normal circumstances you don't need to call this function. 1807It's main use is in scripts without an event loop - when you want to stat
1808a lot of files, you can write somehting like this:
1809
1810 IO::AIO::max_outstanding 32;
1811
1812 for my $path (...) {
1813 aio_stat $path , ...;
1814 IO::AIO::poll_cb;
1815 }
1816
1817 IO::AIO::flush;
1818
1819The call to C<poll_cb> inside the loop will normally return instantly, but
1820as soon as more thna C<32> reqeusts are in-flight, it will block until
1821some requests have been handled. This keeps the loop from pushing a large
1822number of C<aio_stat> requests onto the queue.
1823
1824The default value for C<max_outstanding> is very large, so there is no
1825practical limit on the number of outstanding requests.
829 1826
830=back 1827=back
831 1828
1829=head3 STATISTICAL INFORMATION
1830
1831=over
1832
1833=item IO::AIO::nreqs
1834
1835Returns the number of requests currently in the ready, execute or pending
1836states (i.e. for which their callback has not been invoked yet).
1837
1838Example: wait till there are no outstanding requests anymore:
1839
1840 IO::AIO::poll_wait, IO::AIO::poll_cb
1841 while IO::AIO::nreqs;
1842
1843=item IO::AIO::nready
1844
1845Returns the number of requests currently in the ready state (not yet
1846executed).
1847
1848=item IO::AIO::npending
1849
1850Returns the number of requests currently in the pending state (executed,
1851but not yet processed by poll_cb).
1852
1853=back
1854
1855=head3 MISCELLANEOUS FUNCTIONS
1856
1857IO::AIO implements some functions that might be useful, but are not
1858asynchronous.
1859
1860=over 4
1861
1862=item IO::AIO::sendfile $ofh, $ifh, $offset, $count
1863
1864Calls the C<eio_sendfile_sync> function, which is like C<aio_sendfile>,
1865but is blocking (this makes most sense if you know the input data is
1866likely cached already and the output filehandle is set to non-blocking
1867operations).
1868
1869Returns the number of bytes copied, or C<-1> on error.
1870
1871=item IO::AIO::fadvise $fh, $offset, $len, $advice
1872
1873Simply calls the C<posix_fadvise> function (see its
1874manpage for details). The following advice constants are
1875available: C<IO::AIO::FADV_NORMAL>, C<IO::AIO::FADV_SEQUENTIAL>,
1876C<IO::AIO::FADV_RANDOM>, C<IO::AIO::FADV_NOREUSE>,
1877C<IO::AIO::FADV_WILLNEED>, C<IO::AIO::FADV_DONTNEED>.
1878
1879On systems that do not implement C<posix_fadvise>, this function returns
1880ENOSYS, otherwise the return value of C<posix_fadvise>.
1881
1882=item IO::AIO::madvise $scalar, $offset, $len, $advice
1883
1884Simply calls the C<posix_madvise> function (see its
1885manpage for details). The following advice constants are
1886available: C<IO::AIO::MADV_NORMAL>, C<IO::AIO::MADV_SEQUENTIAL>,
1887C<IO::AIO::MADV_RANDOM>, C<IO::AIO::MADV_WILLNEED>, C<IO::AIO::MADV_DONTNEED>.
1888
1889On systems that do not implement C<posix_madvise>, this function returns
1890ENOSYS, otherwise the return value of C<posix_madvise>.
1891
1892=item IO::AIO::mprotect $scalar, $offset, $len, $protect
1893
1894Simply calls the C<mprotect> function on the preferably AIO::mmap'ed
1895$scalar (see its manpage for details). The following protect
1896constants are available: C<IO::AIO::PROT_NONE>, C<IO::AIO::PROT_READ>,
1897C<IO::AIO::PROT_WRITE>, C<IO::AIO::PROT_EXEC>.
1898
1899On systems that do not implement C<mprotect>, this function returns
1900ENOSYS, otherwise the return value of C<mprotect>.
1901
1902=item IO::AIO::mmap $scalar, $length, $prot, $flags, $fh[, $offset]
1903
1904Memory-maps a file (or anonymous memory range) and attaches it to the
1905given C<$scalar>, which will act like a string scalar.
1906
1907The only operations allowed on the scalar are C<substr>/C<vec> that don't
1908change the string length, and most read-only operations such as copying it
1909or searching it with regexes and so on.
1910
1911Anything else is unsafe and will, at best, result in memory leaks.
1912
1913The memory map associated with the C<$scalar> is automatically removed
1914when the C<$scalar> is destroyed, or when the C<IO::AIO::mmap> or
1915C<IO::AIO::munmap> functions are called.
1916
1917This calls the C<mmap>(2) function internally. See your system's manual
1918page for details on the C<$length>, C<$prot> and C<$flags> parameters.
1919
1920The C<$length> must be larger than zero and smaller than the actual
1921filesize.
1922
1923C<$prot> is a combination of C<IO::AIO::PROT_NONE>, C<IO::AIO::PROT_EXEC>,
1924C<IO::AIO::PROT_READ> and/or C<IO::AIO::PROT_WRITE>,
1925
1926C<$flags> can be a combination of C<IO::AIO::MAP_SHARED> or
1927C<IO::AIO::MAP_PRIVATE>, or a number of system-specific flags (when
1928not available, the are defined as 0): C<IO::AIO::MAP_ANONYMOUS>
1929(which is set to C<MAP_ANON> if your system only provides this
1930constant), C<IO::AIO::MAP_HUGETLB>, C<IO::AIO::MAP_LOCKED>,
1931C<IO::AIO::MAP_NORESERVE>, C<IO::AIO::MAP_POPULATE> or
1932C<IO::AIO::MAP_NONBLOCK>
1933
1934If C<$fh> is C<undef>, then a file descriptor of C<-1> is passed.
1935
1936C<$offset> is the offset from the start of the file - it generally must be
1937a multiple of C<IO::AIO::PAGESIZE> and defaults to C<0>.
1938
1939Example:
1940
1941 use Digest::MD5;
1942 use IO::AIO;
1943
1944 open my $fh, "<verybigfile"
1945 or die "$!";
1946
1947 IO::AIO::mmap my $data, -s $fh, IO::AIO::PROT_READ, IO::AIO::MAP_SHARED, $fh
1948 or die "verybigfile: $!";
1949
1950 my $fast_md5 = md5 $data;
1951
1952=item IO::AIO::munmap $scalar
1953
1954Removes a previous mmap and undefines the C<$scalar>.
1955
1956=item IO::AIO::munlock $scalar, $offset = 0, $length = undef
1957
1958Calls the C<munlock> function, undoing the effects of a previous
1959C<aio_mlock> call (see its description for details).
1960
1961=item IO::AIO::munlockall
1962
1963Calls the C<munlockall> function.
1964
1965On systems that do not implement C<munlockall>, this function returns
1966ENOSYS, otherwise the return value of C<munlockall>.
1967
1968=back
1969
832=cut 1970=cut
833 1971
834# support function to convert a fd into a perl filehandle
835sub _fd2fh {
836 return undef if $_[0] < 0;
837
838 # try to generate nice filehandles
839 my $sym = "IO::AIO::fd#$_[0]";
840 local *$sym;
841
842 open *$sym, "+<&=$_[0]" # usually works under any unix
843 or open *$sym, "<&=$_[0]" # cygwin needs this
844 or open *$sym, ">&=$_[0]" # or this
845 or return undef;
846
847 *$sym
848}
849
850min_parallel 8; 1972min_parallel 8;
851 1973
852END { 1974END { flush }
853 max_parallel 0;
854}
855 1975
8561; 19761;
857 1977
1978=head1 EVENT LOOP INTEGRATION
1979
1980It is recommended to use L<AnyEvent::AIO> to integrate IO::AIO
1981automatically into many event loops:
1982
1983 # AnyEvent integration (EV, Event, Glib, Tk, POE, urxvt, pureperl...)
1984 use AnyEvent::AIO;
1985
1986You can also integrate IO::AIO manually into many event loops, here are
1987some examples of how to do this:
1988
1989 # EV integration
1990 my $aio_w = EV::io IO::AIO::poll_fileno, EV::READ, \&IO::AIO::poll_cb;
1991
1992 # Event integration
1993 Event->io (fd => IO::AIO::poll_fileno,
1994 poll => 'r',
1995 cb => \&IO::AIO::poll_cb);
1996
1997 # Glib/Gtk2 integration
1998 add_watch Glib::IO IO::AIO::poll_fileno,
1999 in => sub { IO::AIO::poll_cb; 1 };
2000
2001 # Tk integration
2002 Tk::Event::IO->fileevent (IO::AIO::poll_fileno, "",
2003 readable => \&IO::AIO::poll_cb);
2004
2005 # Danga::Socket integration
2006 Danga::Socket->AddOtherFds (IO::AIO::poll_fileno =>
2007 \&IO::AIO::poll_cb);
2008
858=head2 FORK BEHAVIOUR 2009=head2 FORK BEHAVIOUR
859 2010
860This module should do "the right thing" when the process using it forks: 2011Usage of pthreads in a program changes the semantics of fork
2012considerably. Specifically, only async-safe functions can be called after
2013fork. Perl doesn't know about this, so in general, you cannot call fork
2014with defined behaviour in perl if pthreads are involved. IO::AIO uses
2015pthreads, so this applies, but many other extensions and (for inexplicable
2016reasons) perl itself often is linked against pthreads, so this limitation
2017applies to quite a lot of perls.
861 2018
862Before the fork, IO::AIO enters a quiescent state where no requests 2019This module no longer tries to fight your OS, or POSIX. That means IO::AIO
863can be added in other threads and no results will be processed. After 2020only works in the process that loaded it. Forking is fully supported, but
864the fork the parent simply leaves the quiescent state and continues 2021using IO::AIO in the child is not.
865request/result processing, while the child clears the request/result
866queue (so the requests started before the fork will only be handled in
867the parent). Threads will be started on demand until the limit ste in the
868parent process has been reached again.
869 2022
870In short: the parent will, after a short pause, continue as if fork had 2023You might get around by not I<using> IO::AIO before (or after)
871not been called, while the child will act as if IO::AIO has not been used 2024forking. You could also try to call the L<IO::AIO::reinit> function in the
872yet. 2025child:
2026
2027=over 4
2028
2029=item IO::AIO::reinit
2030
2031Abandons all current requests and I/O threads and simply reinitialises all
2032data structures. This is not an operation supported by any standards, but
2033happens to work on GNU/Linux and some newer BSD systems.
2034
2035The only reasonable use for this function is to call it after forking, if
2036C<IO::AIO> was used in the parent. Calling it while IO::AIO is active in
2037the process will result in undefined behaviour. Calling it at any time
2038will also result in any undefined (by POSIX) behaviour.
2039
2040=back
873 2041
874=head2 MEMORY USAGE 2042=head2 MEMORY USAGE
875 2043
2044Per-request usage:
2045
876Each aio request uses - depending on your architecture - around 128 bytes 2046Each aio request uses - depending on your architecture - around 100-200
877of memory. In addition, stat requests need a stat buffer (possibly a few 2047bytes of memory. In addition, stat requests need a stat buffer (possibly
878hundred bytes). Perl scalars and other data passed into aio requests will 2048a few hundred bytes), readdir requires a result buffer and so on. Perl
879also be locked. 2049scalars and other data passed into aio requests will also be locked and
2050will consume memory till the request has entered the done state.
880 2051
881This is now awfully much, so queuing lots of requests is not usually a 2052This is not awfully much, so queuing lots of requests is not usually a
882problem. 2053problem.
883 2054
884Each thread needs a stack area which is usually around 16k, sometimes much 2055Per-thread usage:
885larger, depending on the OS. 2056
2057In the execution phase, some aio requests require more memory for
2058temporary buffers, and each thread requires a stack and other data
2059structures (usually around 16k-128k, depending on the OS).
2060
2061=head1 KNOWN BUGS
2062
2063Known bugs will be fixed in the next release.
886 2064
887=head1 SEE ALSO 2065=head1 SEE ALSO
888 2066
889L<Coro>, L<Linux::AIO> (obsolete). 2067L<AnyEvent::AIO> for easy integration into event loops, L<Coro::AIO> for a
2068more natural syntax.
890 2069
891=head1 AUTHOR 2070=head1 AUTHOR
892 2071
893 Marc Lehmann <schmorp@schmorp.de> 2072 Marc Lehmann <schmorp@schmorp.de>
894 http://home.schmorp.de/ 2073 http://home.schmorp.de/

Diff Legend

Removed lines
+ Added lines
< Changed lines
> Changed lines