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 1.123 by root, Sat May 10 18:06:41 2008 UTC vs.
Revision 1.203 by root, Thu Jul 7 22:36:18 2011 UTC

… …
4 4
5=head1 SYNOPSIS 5=head1 SYNOPSIS
6 6
7 use IO::AIO; 7 use IO::AIO;
8 8
9 aio_open "/etc/passwd", O_RDONLY, 0, sub { 9 aio_open "/etc/passwd", IO::AIO::O_RDONLY, 0, sub {
10 my $fh = shift 10 my $fh = shift
11 or die "/etc/passwd: $!"; 11 or die "/etc/passwd: $!";
12 ... 12 ...
13 }; 13 };
14 14
… …
26 $req->cancel; # cancel request if still in queue 26 $req->cancel; # cancel request if still in queue
27 27
28 my $grp = aio_group sub { print "all stats done\n" }; 28 my $grp = aio_group sub { print "all stats done\n" };
29 add $grp aio_stat "..." for ...; 29 add $grp aio_stat "..." for ...;
30 30
31 # AnyEvent integration (EV, Event, Glib, Tk, urxvt, pureperl...)
32 open my $fh, "<&=" . IO::AIO::poll_fileno or die "$!";
33 my $w = AnyEvent->io (fh => $fh, poll => 'r', cb => sub { IO::AIO::poll_cb });
34
35 # EV integration
36 my $w = EV::io IO::AIO::poll_fileno, EV::READ, \&IO::AIO::poll_cb;
37
38 # Event integration
39 Event->io (fd => IO::AIO::poll_fileno,
40 poll => 'r',
41 cb => \&IO::AIO::poll_cb);
42
43 # Glib/Gtk2 integration
44 add_watch Glib::IO IO::AIO::poll_fileno,
45 in => sub { IO::AIO::poll_cb; 1 };
46
47 # Tk integration
48 Tk::Event::IO->fileevent (IO::AIO::poll_fileno, "",
49 readable => \&IO::AIO::poll_cb);
50
51 # Danga::Socket integration
52 Danga::Socket->AddOtherFds (IO::AIO::poll_fileno =>
53 \&IO::AIO::poll_cb);
54
55=head1 DESCRIPTION 31=head1 DESCRIPTION
56 32
57This module implements asynchronous I/O using whatever means your 33This module implements asynchronous I/O using whatever means your
58operating system supports. 34operating system supports. It is implemented as an interface to C<libeio>
35(L<http://software.schmorp.de/pkg/libeio.html>).
59 36
60Asynchronous means that operations that can normally block your program 37Asynchronous means that operations that can normally block your program
61(e.g. reading from disk) will be done asynchronously: the operation 38(e.g. reading from disk) will be done asynchronously: the operation
62will still block, but you can do something else in the meantime. This 39will still block, but you can do something else in the meantime. This
63is extremely useful for programs that need to stay interactive even 40is extremely useful for programs that need to stay interactive even
… …
67on a RAID volume or over NFS when you do a number of stat operations 44on a RAID volume or over NFS when you do a number of stat operations
68concurrently. 45concurrently.
69 46
70While most of this works on all types of file descriptors (for 47While most of this works on all types of file descriptors (for
71example sockets), using these functions on file descriptors that 48example sockets), using these functions on file descriptors that
72support nonblocking operation (again, sockets, pipes etc.) is very 49support nonblocking operation (again, sockets, pipes etc.) is
73inefficient. Use an event loop for that (such as the L<Event|Event> 50very inefficient. Use an event loop for that (such as the L<EV>
74module): IO::AIO will naturally fit into such an event loop itself. 51module): IO::AIO will naturally fit into such an event loop itself.
75 52
76In this version, a number of threads are started that execute your 53In this version, a number of threads are started that execute your
77requests and signal their completion. You don't need thread support 54requests and signal their completion. You don't need thread support
78in perl, and the threads created by this module will not be visible 55in perl, and the threads created by this module will not be visible
… …
88yourself, always call C<poll_cb> from within the same thread, or never 65yourself, always call C<poll_cb> from within the same thread, or never
89call C<poll_cb> (or other C<aio_> functions) recursively. 66call C<poll_cb> (or other C<aio_> functions) recursively.
90 67
91=head2 EXAMPLE 68=head2 EXAMPLE
92 69
93This is a simple example that uses the Event module and loads 70This is a simple example that uses the EV module and loads
94F</etc/passwd> asynchronously: 71F</etc/passwd> asynchronously:
95 72
96 use Fcntl; 73 use Fcntl;
97 use Event; 74 use EV;
98 use IO::AIO; 75 use IO::AIO;
99 76
100 # register the IO::AIO callback with Event 77 # register the IO::AIO callback with EV
101 Event->io (fd => IO::AIO::poll_fileno, 78 my $aio_w = EV::io IO::AIO::poll_fileno, EV::READ, \&IO::AIO::poll_cb;
102 poll => 'r',
103 cb => \&IO::AIO::poll_cb);
104 79
105 # queue the request to open /etc/passwd 80 # queue the request to open /etc/passwd
106 aio_open "/etc/passwd", O_RDONLY, 0, sub { 81 aio_open "/etc/passwd", IO::AIO::O_RDONLY, 0, sub {
107 my $fh = shift 82 my $fh = shift
108 or die "error while opening: $!"; 83 or die "error while opening: $!";
109 84
110 # stat'ing filehandles is generally non-blocking 85 # stat'ing filehandles is generally non-blocking
111 my $size = -s $fh; 86 my $size = -s $fh;
… …
120 95
121 # file contents now in $contents 96 # file contents now in $contents
122 print $contents; 97 print $contents;
123 98
124 # exit event loop and program 99 # exit event loop and program
125 Event::unloop; 100 EV::unloop;
126 }; 101 };
127 }; 102 };
128 103
129 # possibly queue up other requests, or open GUI windows, 104 # possibly queue up other requests, or open GUI windows,
130 # check for sockets etc. etc. 105 # check for sockets etc. etc.
131 106
132 # process events as long as there are some: 107 # process events as long as there are some:
133 Event::loop; 108 EV::loop;
134 109
135=head1 REQUEST ANATOMY AND LIFETIME 110=head1 REQUEST ANATOMY AND LIFETIME
136 111
137Every C<aio_*> function creates a request. which is a C data structure not 112Every C<aio_*> function creates a request. which is a C data structure not
138directly visible to Perl. 113directly visible to Perl.
… …
188 163
189package IO::AIO; 164package IO::AIO;
190 165
191use Carp (); 166use Carp ();
192 167
193no warnings; 168use common::sense;
194use strict 'vars';
195 169
196use base 'Exporter'; 170use base 'Exporter';
197 171
198BEGIN { 172BEGIN {
199 our $VERSION = '2.62'; 173 our $VERSION = '3.93';
200 174
201 our @AIO_REQ = qw(aio_sendfile aio_read aio_write aio_open aio_close 175 our @AIO_REQ = qw(aio_sendfile aio_read aio_write aio_open aio_close
202 aio_stat aio_lstat aio_unlink aio_rmdir aio_readdir 176 aio_stat aio_lstat aio_unlink aio_rmdir aio_readdir aio_readdirx
203 aio_scandir aio_symlink aio_readlink aio_sync aio_fsync 177 aio_scandir aio_symlink aio_readlink aio_realpath aio_sync aio_fsync
178 aio_fdatasync aio_sync_file_range aio_fallocate
204 aio_fdatasync aio_pathsync aio_readahead 179 aio_pathsync aio_readahead
205 aio_rename aio_link aio_move aio_copy aio_group 180 aio_rename aio_link aio_move aio_copy aio_group
206 aio_nop aio_mknod aio_load aio_rmtree aio_mkdir aio_chown 181 aio_nop aio_mknod aio_load aio_rmtree aio_mkdir aio_chown
207 aio_chmod aio_utime aio_truncate); 182 aio_chmod aio_utime aio_truncate
183 aio_msync aio_mtouch aio_mlock aio_mlockall
184 aio_statvfs);
208 185
209 our @EXPORT = (@AIO_REQ, qw(aioreq_pri aioreq_nice)); 186 our @EXPORT = (@AIO_REQ, qw(aioreq_pri aioreq_nice));
210 our @EXPORT_OK = qw(poll_fileno poll_cb poll_wait flush 187 our @EXPORT_OK = qw(poll_fileno poll_cb poll_wait flush
211 min_parallel max_parallel max_idle 188 min_parallel max_parallel max_idle idle_timeout
212 nreqs nready npending nthreads 189 nreqs nready npending nthreads
213 max_poll_time max_poll_reqs); 190 max_poll_time max_poll_reqs
191 sendfile fadvise madvise
192 mmap munmap munlock munlockall);
193
194 push @AIO_REQ, qw(aio_busy); # not exported
214 195
215 @IO::AIO::GRP::ISA = 'IO::AIO::REQ'; 196 @IO::AIO::GRP::ISA = 'IO::AIO::REQ';
216 197
217 require XSLoader; 198 require XSLoader;
218 XSLoader::load ("IO::AIO", $VERSION); 199 XSLoader::load ("IO::AIO", $VERSION);
219} 200}
220 201
221=head1 FUNCTIONS 202=head1 FUNCTIONS
203
204=head2 QUICK OVERVIEW
205
206This section simply lists the prototypes of the most important functions
207for quick reference. See the following sections for function-by-function
208documentation.
209
210 aio_open $pathname, $flags, $mode, $callback->($fh)
211 aio_close $fh, $callback->($status)
212 aio_read $fh,$offset,$length, $data,$dataoffset, $callback->($retval)
213 aio_write $fh,$offset,$length, $data,$dataoffset, $callback->($retval)
214 aio_sendfile $out_fh, $in_fh, $in_offset, $length, $callback->($retval)
215 aio_readahead $fh,$offset,$length, $callback->($retval)
216 aio_stat $fh_or_path, $callback->($status)
217 aio_lstat $fh, $callback->($status)
218 aio_statvfs $fh_or_path, $callback->($statvfs)
219 aio_utime $fh_or_path, $atime, $mtime, $callback->($status)
220 aio_chown $fh_or_path, $uid, $gid, $callback->($status)
221 aio_truncate $fh_or_path, $offset, $callback->($status)
222 aio_chmod $fh_or_path, $mode, $callback->($status)
223 aio_unlink $pathname, $callback->($status)
224 aio_mknod $path, $mode, $dev, $callback->($status)
225 aio_link $srcpath, $dstpath, $callback->($status)
226 aio_symlink $srcpath, $dstpath, $callback->($status)
227 aio_readlink $path, $callback->($link)
228 aio_realpath $path, $callback->($link)
229 aio_rename $srcpath, $dstpath, $callback->($status)
230 aio_mkdir $pathname, $mode, $callback->($status)
231 aio_rmdir $pathname, $callback->($status)
232 aio_readdir $pathname, $callback->($entries)
233 aio_readdirx $pathname, $flags, $callback->($entries, $flags)
234 IO::AIO::READDIR_DENTS IO::AIO::READDIR_DIRS_FIRST
235 IO::AIO::READDIR_STAT_ORDER IO::AIO::READDIR_FOUND_UNKNOWN
236 aio_load $path, $data, $callback->($status)
237 aio_copy $srcpath, $dstpath, $callback->($status)
238 aio_move $srcpath, $dstpath, $callback->($status)
239 aio_scandir $path, $maxreq, $callback->($dirs, $nondirs)
240 aio_rmtree $path, $callback->($status)
241 aio_sync $callback->($status)
242 aio_fsync $fh, $callback->($status)
243 aio_fdatasync $fh, $callback->($status)
244 aio_sync_file_range $fh, $offset, $nbytes, $flags, $callback->($status)
245 aio_pathsync $path, $callback->($status)
246 aio_msync $scalar, $offset = 0, $length = undef, flags = 0, $callback->($status)
247 aio_mtouch $scalar, $offset = 0, $length = undef, flags = 0, $callback->($status)
248 aio_mlock $scalar, $offset = 0, $length = undef, $callback->($status)
249 aio_mlockall $flags, $callback->($status)
250 aio_group $callback->(...)
251 aio_nop $callback->()
252
253 $prev_pri = aioreq_pri [$pri]
254 aioreq_nice $pri_adjust
255
256 IO::AIO::poll_wait
257 IO::AIO::poll_cb
258 IO::AIO::poll
259 IO::AIO::flush
260 IO::AIO::max_poll_reqs $nreqs
261 IO::AIO::max_poll_time $seconds
262 IO::AIO::min_parallel $nthreads
263 IO::AIO::max_parallel $nthreads
264 IO::AIO::max_idle $nthreads
265 IO::AIO::idle_timeout $seconds
266 IO::AIO::max_outstanding $maxreqs
267 IO::AIO::nreqs
268 IO::AIO::nready
269 IO::AIO::npending
270
271 IO::AIO::sendfile $ofh, $ifh, $offset, $count
272 IO::AIO::fadvise $fh, $offset, $len, $advice
273 IO::AIO::madvise $scalar, $offset, $length, $advice
274 IO::AIO::mprotect $scalar, $offset, $length, $protect
275 IO::AIO::munlock $scalar, $offset = 0, $length = undef
276 IO::AIO::munlockall
222 277
223=head2 AIO REQUEST FUNCTIONS 278=head2 AIO REQUEST FUNCTIONS
224 279
225All the C<aio_*> calls are more or less thin wrappers around the syscall 280All the C<aio_*> calls are more or less thin wrappers around the syscall
226with the same name (sans C<aio_>). The arguments are similar or identical, 281with the same name (sans C<aio_>). The arguments are similar or identical,
227and they all accept an additional (and optional) C<$callback> argument 282and they all accept an additional (and optional) C<$callback> argument
228which must be a code reference. This code reference will get called with 283which must be a code reference. This code reference will get called with
229the syscall return code (e.g. most syscalls return C<-1> on error, unlike 284the syscall return code (e.g. most syscalls return C<-1> on error, unlike
230perl, which usually delivers "false") as it's sole argument when the given 285perl, which usually delivers "false") as its sole argument after the given
231syscall has been executed asynchronously. 286syscall has been executed asynchronously.
232 287
233All functions expecting a filehandle keep a copy of the filehandle 288All functions expecting a filehandle keep a copy of the filehandle
234internally until the request has finished. 289internally until the request has finished.
235 290
… …
249your pathnames to the locale (or other) encoding in effect in the user 304your pathnames to the locale (or other) encoding in effect in the user
250environment, d) use Glib::filename_from_unicode on unicode filenames or e) 305environment, d) use Glib::filename_from_unicode on unicode filenames or e)
251use something else to ensure your scalar has the correct contents. 306use something else to ensure your scalar has the correct contents.
252 307
253This works, btw. independent of the internal UTF-8 bit, which IO::AIO 308This works, btw. independent of the internal UTF-8 bit, which IO::AIO
254handles correctly wether it is set or not. 309handles correctly whether it is set or not.
255 310
256=over 4 311=over 4
257 312
258=item $prev_pri = aioreq_pri [$pri] 313=item $prev_pri = aioreq_pri [$pri]
259 314
… …
306by the umask in effect then the request is being executed, so better never 361by the umask in effect then the request is being executed, so better never
307change the umask. 362change the umask.
308 363
309Example: 364Example:
310 365
311 aio_open "/etc/passwd", O_RDONLY, 0, sub { 366 aio_open "/etc/passwd", IO::AIO::O_RDONLY, 0, sub {
312 if ($_[0]) { 367 if ($_[0]) {
313 print "open successful, fh is $_[0]\n"; 368 print "open successful, fh is $_[0]\n";
314 ... 369 ...
315 } else { 370 } else {
316 die "open failed: $!\n"; 371 die "open failed: $!\n";
317 } 372 }
318 }; 373 };
319 374
375In addition to all the common open modes/flags (C<O_RDONLY>, C<O_WRONLY>,
376C<O_RDWR>, C<O_CREAT>, C<O_TRUNC>, C<O_EXCL> and C<O_APPEND>), the
377following POSIX and non-POSIX constants are available (missing ones on
378your system are, as usual, C<0>):
379
380C<O_ASYNC>, C<O_DIRECT>, C<O_NOATIME>, C<O_CLOEXEC>, C<O_NOCTTY>, C<O_NOFOLLOW>,
381C<O_NONBLOCK>, C<O_EXEC>, C<O_SEARCH>, C<O_DIRECTORY>, C<O_DSYNC>,
382C<O_RSYNC>, C<O_SYNC> and C<O_TTY_INIT>.
383
320 384
321=item aio_close $fh, $callback->($status) 385=item aio_close $fh, $callback->($status)
322 386
323Asynchronously close a file and call the callback with the result 387Asynchronously close a file and call the callback with the result
324code. 388code.
… …
337 401
338=item aio_read $fh,$offset,$length, $data,$dataoffset, $callback->($retval) 402=item aio_read $fh,$offset,$length, $data,$dataoffset, $callback->($retval)
339 403
340=item aio_write $fh,$offset,$length, $data,$dataoffset, $callback->($retval) 404=item aio_write $fh,$offset,$length, $data,$dataoffset, $callback->($retval)
341 405
342Reads or writes C<$length> bytes from the specified C<$fh> and C<$offset> 406Reads or writes C<$length> bytes from or to the specified C<$fh> and
343into the scalar given by C<$data> and offset C<$dataoffset> and calls the 407C<$offset> into the scalar given by C<$data> and offset C<$dataoffset>
344callback without the actual number of bytes read (or -1 on error, just 408and calls the callback without the actual number of bytes read (or -1 on
345like the syscall). 409error, just like the syscall).
410
411C<aio_read> will, like C<sysread>, shrink or grow the C<$data> scalar to
412offset plus the actual number of bytes read.
346 413
347If C<$offset> is undefined, then the current file descriptor offset will 414If C<$offset> is undefined, then the current file descriptor offset will
348be used (and updated), otherwise the file descriptor offset will not be 415be used (and updated), otherwise the file descriptor offset will not be
349changed by these calls. 416changed by these calls.
350 417
351If C<$length> is undefined in C<aio_write>, use the remaining length of C<$data>. 418If C<$length> is undefined in C<aio_write>, use the remaining length of
419C<$data>.
352 420
353If C<$dataoffset> is less than zero, it will be counted from the end of 421If C<$dataoffset> is less than zero, it will be counted from the end of
354C<$data>. 422C<$data>.
355 423
356The C<$data> scalar I<MUST NOT> be modified in any way while the request 424The C<$data> scalar I<MUST NOT> be modified in any way while the request
… …
370 438
371Tries to copy C<$length> bytes from C<$in_fh> to C<$out_fh>. It starts 439Tries to copy C<$length> bytes from C<$in_fh> to C<$out_fh>. It starts
372reading at byte offset C<$in_offset>, and starts writing at the current 440reading at byte offset C<$in_offset>, and starts writing at the current
373file offset of C<$out_fh>. Because of that, it is not safe to issue more 441file offset of C<$out_fh>. Because of that, it is not safe to issue more
374than one C<aio_sendfile> per C<$out_fh>, as they will interfere with each 442than one C<aio_sendfile> per C<$out_fh>, as they will interfere with each
375other. 443other. The same C<$in_fh> works fine though, as this function does not
444move or use the file offset of C<$in_fh>.
376 445
446Please note that C<aio_sendfile> can read more bytes from C<$in_fh> than
447are written, and there is no way to find out how many more bytes have been
448read from C<aio_sendfile> alone, as C<aio_sendfile> only provides the
449number of bytes written to C<$out_fh>. Only if the result value equals
450C<$length> one can assume that C<$length> bytes have been read.
451
452Unlike with other C<aio_> functions, it makes a lot of sense to use
453C<aio_sendfile> on non-blocking sockets, as long as one end (typically
454the C<$in_fh>) is a file - the file I/O will then be asynchronous, while
455the socket I/O will be non-blocking. Note, however, that you can run
456into a trap where C<aio_sendfile> reads some data with readahead, then
457fails to write all data, and when the socket is ready the next time, the
458data in the cache is already lost, forcing C<aio_sendfile> to again hit
459the disk. Explicit C<aio_read> + C<aio_write> let's you better control
460resource usage.
461
377This call tries to make use of a native C<sendfile> syscall to provide 462This call tries to make use of a native C<sendfile>-like syscall to
378zero-copy operation. For this to work, C<$out_fh> should refer to a 463provide zero-copy operation. For this to work, C<$out_fh> should refer to
379socket, and C<$in_fh> should refer to mmap'able file. 464a socket, and C<$in_fh> should refer to an mmap'able file.
380 465
381If the native sendfile call fails or is not implemented, it will be 466If a native sendfile cannot be found or it fails with C<ENOSYS>,
382emulated, so you can call C<aio_sendfile> on any type of filehandle 467C<EINVAL>, C<ENOTSUP>, C<EOPNOTSUPP>, C<EAFNOSUPPORT>, C<EPROTOTYPE> or
468C<ENOTSOCK>, it will be emulated, so you can call C<aio_sendfile> on any
383regardless of the limitations of the operating system. 469type of filehandle regardless of the limitations of the operating system.
384 470
385Please note, however, that C<aio_sendfile> can read more bytes from 471As native sendfile syscalls (as practically any non-POSIX interface hacked
386C<$in_fh> than are written, and there is no way to find out how many 472together in a hurry to improve benchmark numbers) tend to be rather buggy
387bytes have been read from C<aio_sendfile> alone, as C<aio_sendfile> only 473on many systems, this implementation tries to work around some known bugs
388provides the number of bytes written to C<$out_fh>. Only if the result 474in Linux and FreeBSD kernels (probably others, too), but that might fail,
389value equals C<$length> one can assume that C<$length> bytes have been 475so you really really should check the return value of C<aio_sendfile> -
390read. 476fewre bytes than expected might have been transferred.
391 477
392 478
393=item aio_readahead $fh,$offset,$length, $callback->($retval) 479=item aio_readahead $fh,$offset,$length, $callback->($retval)
394 480
395C<aio_readahead> populates the page cache with data from a file so that 481C<aio_readahead> populates the page cache with data from a file so that
… …
418 504
419Currently, the stats are always 64-bit-stats, i.e. instead of returning an 505Currently, the stats are always 64-bit-stats, i.e. instead of returning an
420error when stat'ing a large file, the results will be silently truncated 506error when stat'ing a large file, the results will be silently truncated
421unless perl itself is compiled with large file support. 507unless perl itself is compiled with large file support.
422 508
509To help interpret the mode and dev/rdev stat values, IO::AIO offers the
510following constants and functions (if not implemented, the constants will
511be C<0> and the functions will either C<croak> or fall back on traditional
512behaviour).
513
514C<S_IFMT>, C<S_IFIFO>, C<S_IFCHR>, C<S_IFBLK>, C<S_IFLNK>, C<S_IFREG>,
515C<S_IFDIR>, C<S_IFWHT>, C<S_IFSOCK>, C<IO::AIO::major $dev_t>,
516C<IO::AIO::minor $dev_t>, C<IO::AIO::makedev $major, $minor>.
517
423Example: Print the length of F</etc/passwd>: 518Example: Print the length of F</etc/passwd>:
424 519
425 aio_stat "/etc/passwd", sub { 520 aio_stat "/etc/passwd", sub {
426 $_[0] and die "stat failed: $!"; 521 $_[0] and die "stat failed: $!";
427 print "size is ", -s _, "\n"; 522 print "size is ", -s _, "\n";
428 }; 523 };
429 524
430 525
526=item aio_statvfs $fh_or_path, $callback->($statvfs)
527
528Works like the POSIX C<statvfs> or C<fstatvfs> syscalls, depending on
529whether a file handle or path was passed.
530
531On success, the callback is passed a hash reference with the following
532members: C<bsize>, C<frsize>, C<blocks>, C<bfree>, C<bavail>, C<files>,
533C<ffree>, C<favail>, C<fsid>, C<flag> and C<namemax>. On failure, C<undef>
534is passed.
535
536The following POSIX IO::AIO::ST_* constants are defined: C<ST_RDONLY> and
537C<ST_NOSUID>.
538
539The following non-POSIX IO::AIO::ST_* flag masks are defined to
540their correct value when available, or to C<0> on systems that do
541not support them: C<ST_NODEV>, C<ST_NOEXEC>, C<ST_SYNCHRONOUS>,
542C<ST_MANDLOCK>, C<ST_WRITE>, C<ST_APPEND>, C<ST_IMMUTABLE>, C<ST_NOATIME>,
543C<ST_NODIRATIME> and C<ST_RELATIME>.
544
545Example: stat C</wd> and dump out the data if successful.
546
547 aio_statvfs "/wd", sub {
548 my $f = $_[0]
549 or die "statvfs: $!";
550
551 use Data::Dumper;
552 say Dumper $f;
553 };
554
555 # result:
556 {
557 bsize => 1024,
558 bfree => 4333064312,
559 blocks => 10253828096,
560 files => 2050765568,
561 flag => 4096,
562 favail => 2042092649,
563 bavail => 4333064312,
564 ffree => 2042092649,
565 namemax => 255,
566 frsize => 1024,
567 fsid => 1810
568 }
569
570
431=item aio_utime $fh_or_path, $atime, $mtime, $callback->($status) 571=item aio_utime $fh_or_path, $atime, $mtime, $callback->($status)
432 572
433Works like perl's C<utime> function (including the special case of $atime 573Works like perl's C<utime> function (including the special case of $atime
434and $mtime being undef). Fractional times are supported if the underlying 574and $mtime being undef). Fractional times are supported if the underlying
435syscalls support them. 575syscalls support them.
… …
483 623
484The only (POSIX-) portable way of calling this function is: 624The only (POSIX-) portable way of calling this function is:
485 625
486 aio_mknod $path, IO::AIO::S_IFIFO | $mode, 0, sub { ... 626 aio_mknod $path, IO::AIO::S_IFIFO | $mode, 0, sub { ...
487 627
628See C<aio_stat> for info about some potentially helpful extra constants
629and functions.
488 630
489=item aio_link $srcpath, $dstpath, $callback->($status) 631=item aio_link $srcpath, $dstpath, $callback->($status)
490 632
491Asynchronously create a new link to the existing object at C<$srcpath> at 633Asynchronously create a new link to the existing object at C<$srcpath> at
492the path C<$dstpath> and call the callback with the result code. 634the path C<$dstpath> and call the callback with the result code.
… …
503Asynchronously read the symlink specified by C<$path> and pass it to 645Asynchronously read the symlink specified by C<$path> and pass it to
504the callback. If an error occurs, nothing or undef gets passed to the 646the callback. If an error occurs, nothing or undef gets passed to the
505callback. 647callback.
506 648
507 649
650=item aio_realpath $path, $callback->($path)
651
652Asynchronously make the path absolute and resolve any symlinks in
653C<$path>. The resulting path only consists of directories (Same as
654L<Cwd::realpath>).
655
656This request can be used to get the absolute path of the current working
657directory by passing it a path of F<.> (a single dot).
658
659
508=item aio_rename $srcpath, $dstpath, $callback->($status) 660=item aio_rename $srcpath, $dstpath, $callback->($status)
509 661
510Asynchronously rename the object at C<$srcpath> to C<$dstpath>, just as 662Asynchronously rename the object at C<$srcpath> to C<$dstpath>, just as
511rename(2) and call the callback with the result code. 663rename(2) and call the callback with the result code.
512 664
… …
528 680
529Unlike the POSIX call of the same name, C<aio_readdir> reads an entire 681Unlike the POSIX call of the same name, C<aio_readdir> reads an entire
530directory (i.e. opendir + readdir + closedir). The entries will not be 682directory (i.e. opendir + readdir + closedir). The entries will not be
531sorted, and will B<NOT> include the C<.> and C<..> entries. 683sorted, and will B<NOT> include the C<.> and C<..> entries.
532 684
533The callback a single argument which is either C<undef> or an array-ref 685The callback is passed a single argument which is either C<undef> or an
534with the filenames. 686array-ref with the filenames.
687
688
689=item aio_readdirx $pathname, $flags, $callback->($entries, $flags)
690
691Quite similar to C<aio_readdir>, but the C<$flags> argument allows to tune
692behaviour and output format. In case of an error, C<$entries> will be
693C<undef>.
694
695The flags are a combination of the following constants, ORed together (the
696flags will also be passed to the callback, possibly modified):
697
698=over 4
699
700=item IO::AIO::READDIR_DENTS
701
702When this flag is off, then the callback gets an arrayref consisting of
703names only (as with C<aio_readdir>), otherwise it gets an arrayref with
704C<[$name, $type, $inode]> arrayrefs, each describing a single directory
705entry in more detail.
706
707C<$name> is the name of the entry.
708
709C<$type> is one of the C<IO::AIO::DT_xxx> constants:
710
711C<IO::AIO::DT_UNKNOWN>, C<IO::AIO::DT_FIFO>, C<IO::AIO::DT_CHR>, C<IO::AIO::DT_DIR>,
712C<IO::AIO::DT_BLK>, C<IO::AIO::DT_REG>, C<IO::AIO::DT_LNK>, C<IO::AIO::DT_SOCK>,
713C<IO::AIO::DT_WHT>.
714
715C<IO::AIO::DT_UNKNOWN> means just that: readdir does not know. If you need to
716know, you have to run stat yourself. Also, for speed reasons, the C<$type>
717scalars are read-only: you can not modify them.
718
719C<$inode> is the inode number (which might not be exact on systems with 64
720bit inode numbers and 32 bit perls). This field has unspecified content on
721systems that do not deliver the inode information.
722
723=item IO::AIO::READDIR_DIRS_FIRST
724
725When this flag is set, then the names will be returned in an order where
726likely directories come first, in optimal stat order. This is useful when
727you need to quickly find directories, or you want to find all directories
728while avoiding to stat() each entry.
729
730If the system returns type information in readdir, then this is used
731to find directories directly. Otherwise, likely directories are names
732beginning with ".", or otherwise names with no dots, of which names with
733short names are tried first.
734
735=item IO::AIO::READDIR_STAT_ORDER
736
737When this flag is set, then the names will be returned in an order
738suitable for stat()'ing each one. That is, when you plan to stat()
739all files in the given directory, then the returned order will likely
740be fastest.
741
742If both this flag and C<IO::AIO::READDIR_DIRS_FIRST> are specified, then
743the likely dirs come first, resulting in a less optimal stat order.
744
745=item IO::AIO::READDIR_FOUND_UNKNOWN
746
747This flag should not be set when calling C<aio_readdirx>. Instead, it
748is being set by C<aio_readdirx>, when any of the C<$type>'s found were
749C<IO::AIO::DT_UNKNOWN>. The absense of this flag therefore indicates that all
750C<$type>'s are known, which can be used to speed up some algorithms.
751
752=back
535 753
536 754
537=item aio_load $path, $data, $callback->($status) 755=item aio_load $path, $data, $callback->($status)
538 756
539This is a composite request that tries to fully load the given file into 757This is a composite request that tries to fully load the given file into
… …
564 782
565=item aio_copy $srcpath, $dstpath, $callback->($status) 783=item aio_copy $srcpath, $dstpath, $callback->($status)
566 784
567Try to copy the I<file> (directories not supported as either source or 785Try to copy the I<file> (directories not supported as either source or
568destination) from C<$srcpath> to C<$dstpath> and call the callback with 786destination) from C<$srcpath> to C<$dstpath> and call the callback with
569the C<0> (error) or C<-1> ok. 787a status of C<0> (ok) or C<-1> (error, see C<$!>).
570 788
571This is a composite request that it creates the destination file with 789This is a composite request that creates the destination file with
572mode 0200 and copies the contents of the source file into it using 790mode 0200 and copies the contents of the source file into it using
573C<aio_sendfile>, followed by restoring atime, mtime, access mode and 791C<aio_sendfile>, followed by restoring atime, mtime, access mode and
574uid/gid, in that order. 792uid/gid, in that order.
575 793
576If an error occurs, the partial destination file will be unlinked, if 794If an error occurs, the partial destination file will be unlinked, if
… …
586 my $grp = aio_group $cb; 804 my $grp = aio_group $cb;
587 805
588 aioreq_pri $pri; 806 aioreq_pri $pri;
589 add $grp aio_open $src, O_RDONLY, 0, sub { 807 add $grp aio_open $src, O_RDONLY, 0, sub {
590 if (my $src_fh = $_[0]) { 808 if (my $src_fh = $_[0]) {
591 my @stat = stat $src_fh; 809 my @stat = stat $src_fh; # hmm, might block over nfs?
592 810
593 aioreq_pri $pri; 811 aioreq_pri $pri;
594 add $grp aio_open $dst, O_CREAT | O_WRONLY | O_TRUNC, 0200, sub { 812 add $grp aio_open $dst, O_CREAT | O_WRONLY | O_TRUNC, 0200, sub {
595 if (my $dst_fh = $_[0]) { 813 if (my $dst_fh = $_[0]) {
596 aioreq_pri $pri; 814 aioreq_pri $pri;
597 add $grp aio_sendfile $dst_fh, $src_fh, 0, $stat[7], sub { 815 add $grp aio_sendfile $dst_fh, $src_fh, 0, $stat[7], sub {
598 if ($_[0] == $stat[7]) { 816 if ($_[0] == $stat[7]) {
599 $grp->result (0); 817 $grp->result (0);
600 close $src_fh; 818 close $src_fh;
601 819
602 # those should not normally block. should. should. 820 my $ch = sub {
603 utime $stat[8], $stat[9], $dst; 821 aioreq_pri $pri;
604 chmod $stat[2] & 07777, $dst_fh; 822 add $grp aio_chmod $dst_fh, $stat[2] & 07777, sub {
605 chown $stat[4], $stat[5], $dst_fh; 823 aioreq_pri $pri;
824 add $grp aio_chown $dst_fh, $stat[4], $stat[5], sub {
825 aioreq_pri $pri;
826 add $grp aio_close $dst_fh;
827 }
828 };
829 };
606 830
607 aioreq_pri $pri; 831 aioreq_pri $pri;
608 add $grp aio_close $dst_fh; 832 add $grp aio_utime $dst_fh, $stat[8], $stat[9], sub {
833 if ($_[0] < 0 && $! == ENOSYS) {
834 aioreq_pri $pri;
835 add $grp aio_utime $dst, $stat[8], $stat[9], $ch;
836 } else {
837 $ch->();
838 }
839 };
609 } else { 840 } else {
610 $grp->result (-1); 841 $grp->result (-1);
611 close $src_fh; 842 close $src_fh;
612 close $dst_fh; 843 close $dst_fh;
613 844
… …
630 861
631=item aio_move $srcpath, $dstpath, $callback->($status) 862=item aio_move $srcpath, $dstpath, $callback->($status)
632 863
633Try to move the I<file> (directories not supported as either source or 864Try to move the I<file> (directories not supported as either source or
634destination) from C<$srcpath> to C<$dstpath> and call the callback with 865destination) from C<$srcpath> to C<$dstpath> and call the callback with
635the C<0> (error) or C<-1> ok. 866a status of C<0> (ok) or C<-1> (error, see C<$!>).
636 867
637This is a composite request that tries to rename(2) the file first. If 868This is a composite request that tries to rename(2) the file first; if
638rename files with C<EXDEV>, it copies the file with C<aio_copy> and, if 869rename fails with C<EXDEV>, it copies the file with C<aio_copy> and, if
639that is successful, unlinking the C<$srcpath>. 870that is successful, unlinks the C<$srcpath>.
640 871
641=cut 872=cut
642 873
643sub aio_move($$;$) { 874sub aio_move($$;$) {
644 my ($src, $dst, $cb) = @_; 875 my ($src, $dst, $cb) = @_;
… …
651 if ($_[0] && $! == EXDEV) { 882 if ($_[0] && $! == EXDEV) {
652 aioreq_pri $pri; 883 aioreq_pri $pri;
653 add $grp aio_copy $src, $dst, sub { 884 add $grp aio_copy $src, $dst, sub {
654 $grp->result ($_[0]); 885 $grp->result ($_[0]);
655 886
656 if (!$_[0]) { 887 unless ($_[0]) {
657 aioreq_pri $pri; 888 aioreq_pri $pri;
658 add $grp aio_unlink $src; 889 add $grp aio_unlink $src;
659 } 890 }
660 }; 891 };
661 } else { 892 } else {
… …
691 922
692Implementation notes. 923Implementation notes.
693 924
694The C<aio_readdir> cannot be avoided, but C<stat()>'ing every entry can. 925The C<aio_readdir> cannot be avoided, but C<stat()>'ing every entry can.
695 926
927If readdir returns file type information, then this is used directly to
928find directories.
929
696After reading the directory, the modification time, size etc. of the 930Otherwise, after reading the directory, the modification time, size etc.
697directory before and after the readdir is checked, and if they match (and 931of the directory before and after the readdir is checked, and if they
698isn't the current time), the link count will be used to decide how many 932match (and isn't the current time), the link count will be used to decide
699entries are directories (if >= 2). Otherwise, no knowledge of the number 933how many entries are directories (if >= 2). Otherwise, no knowledge of the
700of subdirectories will be assumed. 934number of subdirectories will be assumed.
701 935
702Then entries will be sorted into likely directories (everything without 936Then entries will be sorted into likely directories a non-initial dot
703a non-initial dot currently) and likely non-directories (everything 937currently) and likely non-directories (see C<aio_readdirx>). Then every
704else). Then every entry plus an appended C</.> will be C<stat>'ed, 938entry plus an appended C</.> will be C<stat>'ed, likely directories first,
705likely directories first. If that succeeds, it assumes that the entry 939in order of their inode numbers. If that succeeds, it assumes that the
706is a directory or a symlink to directory (which will be checked 940entry is a directory or a symlink to directory (which will be checked
707seperately). This is often faster than stat'ing the entry itself because 941seperately). This is often faster than stat'ing the entry itself because
708filesystems might detect the type of the entry without reading the inode 942filesystems might detect the type of the entry without reading the inode
709data (e.g. ext2fs filetype feature). 943data (e.g. ext2fs filetype feature), even on systems that cannot return
944the filetype information on readdir.
710 945
711If the known number of directories (link count - 2) has been reached, the 946If the known number of directories (link count - 2) has been reached, the
712rest of the entries is assumed to be non-directories. 947rest of the entries is assumed to be non-directories.
713 948
714This only works with certainty on POSIX (= UNIX) filesystems, which 949This only works with certainty on POSIX (= UNIX) filesystems, which
… …
736 my $now = time; 971 my $now = time;
737 my $hash1 = join ":", (stat _)[0,1,3,7,9]; 972 my $hash1 = join ":", (stat _)[0,1,3,7,9];
738 973
739 # read the directory entries 974 # read the directory entries
740 aioreq_pri $pri; 975 aioreq_pri $pri;
741 add $grp aio_readdir $path, sub { 976 add $grp aio_readdirx $path, READDIR_DIRS_FIRST, sub {
742 my $entries = shift 977 my $entries = shift
743 or return $grp->result (); 978 or return $grp->result ();
744 979
745 # stat the dir another time 980 # stat the dir another time
746 aioreq_pri $pri; 981 aioreq_pri $pri;
… …
752 # take the slow route if anything looks fishy 987 # take the slow route if anything looks fishy
753 if ($hash1 ne $hash2 or (stat _)[9] == $now) { 988 if ($hash1 ne $hash2 or (stat _)[9] == $now) {
754 $ndirs = -1; 989 $ndirs = -1;
755 } else { 990 } else {
756 # if nlink == 2, we are finished 991 # if nlink == 2, we are finished
757 # on non-posix-fs's, we rely on nlink < 2 992 # for non-posix-fs's, we rely on nlink < 2
758 $ndirs = (stat _)[3] - 2 993 $ndirs = (stat _)[3] - 2
759 or return $grp->result ([], $entries); 994 or return $grp->result ([], $entries);
760 } 995 }
761 996
762 # sort into likely dirs and likely nondirs
763 # dirs == files without ".", short entries first
764 $entries = [map $_->[0],
765 sort { $b->[1] cmp $a->[1] }
766 map [$_, sprintf "%s%04d", (/.\./ ? "1" : "0"), length],
767 @$entries];
768
769 my (@dirs, @nondirs); 997 my (@dirs, @nondirs);
770 998
771 my $statgrp = add $grp aio_group sub { 999 my $statgrp = add $grp aio_group sub {
772 $grp->result (\@dirs, \@nondirs); 1000 $grp->result (\@dirs, \@nondirs);
773 }; 1001 };
774 1002
775 limit $statgrp $maxreq; 1003 limit $statgrp $maxreq;
776 feed $statgrp sub { 1004 feed $statgrp sub {
777 return unless @$entries; 1005 return unless @$entries;
778 my $entry = pop @$entries; 1006 my $entry = shift @$entries;
779 1007
780 aioreq_pri $pri; 1008 aioreq_pri $pri;
781 add $statgrp aio_stat "$path/$entry/.", sub { 1009 add $statgrp aio_stat "$path/$entry/.", sub {
782 if ($_[0] < 0) { 1010 if ($_[0] < 0) {
783 push @nondirs, $entry; 1011 push @nondirs, $entry;
… …
856callback with the fdatasync result code. 1084callback with the fdatasync result code.
857 1085
858If this call isn't available because your OS lacks it or it couldn't be 1086If this call isn't available because your OS lacks it or it couldn't be
859detected, it will be emulated by calling C<fsync> instead. 1087detected, it will be emulated by calling C<fsync> instead.
860 1088
1089=item aio_sync_file_range $fh, $offset, $nbytes, $flags, $callback->($status)
1090
1091Sync the data portion of the file specified by C<$offset> and C<$length>
1092to disk (but NOT the metadata), by calling the Linux-specific
1093sync_file_range call. If sync_file_range is not available or it returns
1094ENOSYS, then fdatasync or fsync is being substituted.
1095
1096C<$flags> can be a combination of C<IO::AIO::SYNC_FILE_RANGE_WAIT_BEFORE>,
1097C<IO::AIO::SYNC_FILE_RANGE_WRITE> and
1098C<IO::AIO::SYNC_FILE_RANGE_WAIT_AFTER>: refer to the sync_file_range
1099manpage for details.
1100
861=item aio_pathsync $path, $callback->($status) 1101=item aio_pathsync $path, $callback->($status)
862 1102
863This request tries to open, fsync and close the given path. This is a 1103This request tries to open, fsync and close the given path. This is a
864composite request intended tosync directories after directory operations 1104composite request intended to sync directories after directory operations
865(E.g. rename). This might not work on all operating systems or have any 1105(E.g. rename). This might not work on all operating systems or have any
866specific effect, but usually it makes sure that directory changes get 1106specific effect, but usually it makes sure that directory changes get
867written to disc. It works for anything that can be opened for read-only, 1107written to disc. It works for anything that can be opened for read-only,
868not just directories. 1108not just directories.
1109
1110Future versions of this function might fall back to other methods when
1111C<fsync> on the directory fails (such as calling C<sync>).
869 1112
870Passes C<0> when everything went ok, and C<-1> on error. 1113Passes C<0> when everything went ok, and C<-1> on error.
871 1114
872=cut 1115=cut
873 1116
… …
894 }; 1137 };
895 1138
896 $grp 1139 $grp
897} 1140}
898 1141
1142=item aio_msync $scalar, $offset = 0, $length = undef, flags = 0, $callback->($status)
1143
1144This is a rather advanced IO::AIO call, which only works on mmap(2)ed
1145scalars (see the C<IO::AIO::mmap> function, although it also works on data
1146scalars managed by the L<Sys::Mmap> or L<Mmap> modules, note that the
1147scalar must only be modified in-place while an aio operation is pending on
1148it).
1149
1150It calls the C<msync> function of your OS, if available, with the memory
1151area starting at C<$offset> in the string and ending C<$length> bytes
1152later. If C<$length> is negative, counts from the end, and if C<$length>
1153is C<undef>, then it goes till the end of the string. The flags can be
1154a combination of C<IO::AIO::MS_ASYNC>, C<IO::AIO::MS_INVALIDATE> and
1155C<IO::AIO::MS_SYNC>.
1156
1157=item aio_mtouch $scalar, $offset = 0, $length = undef, flags = 0, $callback->($status)
1158
1159This is a rather advanced IO::AIO call, which works best on mmap(2)ed
1160scalars.
1161
1162It touches (reads or writes) all memory pages in the specified
1163range inside the scalar. All caveats and parameters are the same
1164as for C<aio_msync>, above, except for flags, which must be either
1165C<0> (which reads all pages and ensures they are instantiated) or
1166C<IO::AIO::MT_MODIFY>, which modifies the memory page s(by reading and
1167writing an octet from it, which dirties the page).
1168
1169=item aio_mlock $scalar, $offset = 0, $length = undef, $callback->($status)
1170
1171This is a rather advanced IO::AIO call, which works best on mmap(2)ed
1172scalars.
1173
1174It reads in all the pages of the underlying storage into memory (if any)
1175and locks them, so they are not getting swapped/paged out or removed.
1176
1177If C<$length> is undefined, then the scalar will be locked till the end.
1178
1179On systems that do not implement C<mlock>, this function returns C<-1>
1180and sets errno to C<ENOSYS>.
1181
1182Note that the corresponding C<munlock> is synchronous and is
1183documented under L<MISCELLANEOUS FUNCTIONS>.
1184
1185Example: open a file, mmap and mlock it - both will be undone when
1186C<$data> gets destroyed.
1187
1188 open my $fh, "<", $path or die "$path: $!";
1189 my $data;
1190 IO::AIO::mmap $data, -s $fh, IO::AIO::PROT_READ, IO::AIO::MAP_SHARED, $fh;
1191 aio_mlock $data; # mlock in background
1192
1193=item aio_mlockall $flags, $callback->($status)
1194
1195Calls the C<mlockall> function with the given C<$flags> (a combination of
1196C<IO::AIO::MCL_CURRENT> and C<IO::AIO::MCL_FUTURE>).
1197
1198On systems that do not implement C<mlockall>, this function returns C<-1>
1199and sets errno to C<ENOSYS>.
1200
1201Note that the corresponding C<munlockall> is synchronous and is
1202documented under L<MISCELLANEOUS FUNCTIONS>.
1203
1204Example: asynchronously lock all current and future pages into memory.
1205
1206 aio_mlockall IO::AIO::MCL_FUTURE;
1207
899=item aio_group $callback->(...) 1208=item aio_group $callback->(...)
900 1209
901This is a very special aio request: Instead of doing something, it is a 1210This is a very special aio request: Instead of doing something, it is a
902container for other aio requests, which is useful if you want to bundle 1211container for other aio requests, which is useful if you want to bundle
903many requests into a single, composite, request with a definite callback 1212many requests into a single, composite, request with a definite callback
… …
952=item cancel $req 1261=item cancel $req
953 1262
954Cancels the request, if possible. Has the effect of skipping execution 1263Cancels the request, if possible. Has the effect of skipping execution
955when entering the B<execute> state and skipping calling the callback when 1264when entering the B<execute> state and skipping calling the callback when
956entering the the B<result> state, but will leave the request otherwise 1265entering the the B<result> state, but will leave the request otherwise
957untouched. That means that requests that currently execute will not be 1266untouched (with the exception of readdir). That means that requests that
958stopped and resources held by the request will not be freed prematurely. 1267currently execute will not be stopped and resources held by the request
1268will not be freed prematurely.
959 1269
960=item cb $req $callback->(...) 1270=item cb $req $callback->(...)
961 1271
962Replace (or simply set) the callback registered to the request. 1272Replace (or simply set) the callback registered to the request.
963 1273
… …
1014Their lifetime, simplified, looks like this: when they are empty, they 1324Their lifetime, simplified, looks like this: when they are empty, they
1015will finish very quickly. If they contain only requests that are in the 1325will finish very quickly. If they contain only requests that are in the
1016C<done> state, they will also finish. Otherwise they will continue to 1326C<done> state, they will also finish. Otherwise they will continue to
1017exist. 1327exist.
1018 1328
1019That means after creating a group you have some time to add requests. And 1329That means after creating a group you have some time to add requests
1020in the callbacks of those requests, you can add further requests to the 1330(precisely before the callback has been invoked, which is only done within
1021group. And only when all those requests have finished will the the group 1331the C<poll_cb>). And in the callbacks of those requests, you can add
1022itself finish. 1332further requests to the group. And only when all those requests have
1333finished will the the group itself finish.
1023 1334
1024=over 4 1335=over 4
1025 1336
1026=item add $grp ... 1337=item add $grp ...
1027 1338
… …
1035 1346
1036=item $grp->cancel_subs 1347=item $grp->cancel_subs
1037 1348
1038Cancel all subrequests and clears any feeder, but not the group request 1349Cancel all subrequests and clears any feeder, but not the group request
1039itself. Useful when you queued a lot of events but got a result early. 1350itself. Useful when you queued a lot of events but got a result early.
1351
1352The group request will finish normally (you cannot add requests to the
1353group).
1040 1354
1041=item $grp->result (...) 1355=item $grp->result (...)
1042 1356
1043Set the result value(s) that will be passed to the group callback when all 1357Set the result value(s) that will be passed to the group callback when all
1044subrequests have finished and set the groups errno to the current value 1358subrequests have finished and set the groups errno to the current value
… …
1060=item feed $grp $callback->($grp) 1374=item feed $grp $callback->($grp)
1061 1375
1062Sets a feeder/generator on this group: every group can have an attached 1376Sets a feeder/generator on this group: every group can have an attached
1063generator that generates requests if idle. The idea behind this is that, 1377generator that generates requests if idle. The idea behind this is that,
1064although you could just queue as many requests as you want in a group, 1378although you could just queue as many requests as you want in a group,
1065this might starve other requests for a potentially long time. For 1379this might starve other requests for a potentially long time. For example,
1066example, C<aio_scandir> might generate hundreds of thousands C<aio_stat> 1380C<aio_scandir> might generate hundreds of thousands C<aio_stat> requests,
1067requests, delaying any later requests for a long time. 1381delaying any later requests for a long time.
1068 1382
1069To avoid this, and allow incremental generation of requests, you can 1383To avoid this, and allow incremental generation of requests, you can
1070instead a group and set a feeder on it that generates those requests. The 1384instead a group and set a feeder on it that generates those requests. The
1071feed callback will be called whenever there are few enough (see C<limit>, 1385feed callback will be called whenever there are few enough (see C<limit>,
1072below) requests active in the group itself and is expected to queue more 1386below) requests active in the group itself and is expected to queue more
… …
1076not impose any limits). 1390not impose any limits).
1077 1391
1078If the feed does not queue more requests when called, it will be 1392If the feed does not queue more requests when called, it will be
1079automatically removed from the group. 1393automatically removed from the group.
1080 1394
1081If the feed limit is C<0>, it will be set to C<2> automatically. 1395If the feed limit is C<0> when this method is called, it will be set to
1396C<2> automatically.
1082 1397
1083Example: 1398Example:
1084 1399
1085 # stat all files in @files, but only ever use four aio requests concurrently: 1400 # stat all files in @files, but only ever use four aio requests concurrently:
1086 1401
… …
1098Sets the feeder limit for the group: The feeder will be called whenever 1413Sets the feeder limit for the group: The feeder will be called whenever
1099the group contains less than this many requests. 1414the group contains less than this many requests.
1100 1415
1101Setting the limit to C<0> will pause the feeding process. 1416Setting the limit to C<0> will pause the feeding process.
1102 1417
1418The default value for the limit is C<0>, but note that setting a feeder
1419automatically bumps it up to C<2>.
1420
1103=back 1421=back
1104 1422
1105=head2 SUPPORT FUNCTIONS 1423=head2 SUPPORT FUNCTIONS
1106 1424
1107=head3 EVENT PROCESSING AND EVENT LOOP INTEGRATION 1425=head3 EVENT PROCESSING AND EVENT LOOP INTEGRATION
… …
1109=over 4 1427=over 4
1110 1428
1111=item $fileno = IO::AIO::poll_fileno 1429=item $fileno = IO::AIO::poll_fileno
1112 1430
1113Return the I<request result pipe file descriptor>. This filehandle must be 1431Return the I<request result pipe file descriptor>. This filehandle must be
1114polled for reading by some mechanism outside this module (e.g. Event or 1432polled for reading by some mechanism outside this module (e.g. EV, Glib,
1115select, see below or the SYNOPSIS). If the pipe becomes readable you have 1433select and so on, see below or the SYNOPSIS). If the pipe becomes readable
1116to call C<poll_cb> to check the results. 1434you have to call C<poll_cb> to check the results.
1117 1435
1118See C<poll_cb> for an example. 1436See C<poll_cb> for an example.
1119 1437
1120=item IO::AIO::poll_cb 1438=item IO::AIO::poll_cb
1121 1439
1122Process some outstanding events on the result pipe. You have to call this 1440Process some outstanding events on the result pipe. You have to call
1123regularly. Returns the number of events processed. Returns immediately 1441this regularly. Returns C<0> if all events could be processed (or there
1124when no events are outstanding. The amount of events processed depends on 1442were no events to process), or C<-1> if it returned earlier for whatever
1125the settings of C<IO::AIO::max_poll_req> and C<IO::AIO::max_poll_time>. 1443reason. Returns immediately when no events are outstanding. The amount of
1444events processed depends on the settings of C<IO::AIO::max_poll_req> and
1445C<IO::AIO::max_poll_time>.
1126 1446
1127If not all requests were processed for whatever reason, the filehandle 1447If not all requests were processed for whatever reason, the filehandle
1128will still be ready when C<poll_cb> returns. 1448will still be ready when C<poll_cb> returns, so normally you don't have to
1449do anything special to have it called later.
1450
1451Apart from calling C<IO::AIO::poll_cb> when the event filehandle becomes
1452ready, it can be beneficial to call this function from loops which submit
1453a lot of requests, to make sure the results get processed when they become
1454available and not just when the loop is finished and the event loop takes
1455over again. This function returns very fast when there are no outstanding
1456requests.
1129 1457
1130Example: Install an Event watcher that automatically calls 1458Example: Install an Event watcher that automatically calls
1131IO::AIO::poll_cb with high priority: 1459IO::AIO::poll_cb with high priority (more examples can be found in the
1460SYNOPSIS section, at the top of this document):
1132 1461
1133 Event->io (fd => IO::AIO::poll_fileno, 1462 Event->io (fd => IO::AIO::poll_fileno,
1134 poll => 'r', async => 1, 1463 poll => 'r', async => 1,
1135 cb => \&IO::AIO::poll_cb); 1464 cb => \&IO::AIO::poll_cb);
1465
1466=item IO::AIO::poll_wait
1467
1468If there are any outstanding requests and none of them in the result
1469phase, wait till the result filehandle becomes ready for reading (simply
1470does a C<select> on the filehandle. This is useful if you want to
1471synchronously wait for some requests to finish).
1472
1473See C<nreqs> for an example.
1474
1475=item IO::AIO::poll
1476
1477Waits until some requests have been handled.
1478
1479Returns the number of requests processed, but is otherwise strictly
1480equivalent to:
1481
1482 IO::AIO::poll_wait, IO::AIO::poll_cb
1483
1484=item IO::AIO::flush
1485
1486Wait till all outstanding AIO requests have been handled.
1487
1488Strictly equivalent to:
1489
1490 IO::AIO::poll_wait, IO::AIO::poll_cb
1491 while IO::AIO::nreqs;
1136 1492
1137=item IO::AIO::max_poll_reqs $nreqs 1493=item IO::AIO::max_poll_reqs $nreqs
1138 1494
1139=item IO::AIO::max_poll_time $seconds 1495=item IO::AIO::max_poll_time $seconds
1140 1496
… …
1165 # use a low priority so other tasks have priority 1521 # use a low priority so other tasks have priority
1166 Event->io (fd => IO::AIO::poll_fileno, 1522 Event->io (fd => IO::AIO::poll_fileno,
1167 poll => 'r', nice => 1, 1523 poll => 'r', nice => 1,
1168 cb => &IO::AIO::poll_cb); 1524 cb => &IO::AIO::poll_cb);
1169 1525
1170=item IO::AIO::poll_wait
1171
1172If there are any outstanding requests and none of them in the result
1173phase, wait till the result filehandle becomes ready for reading (simply
1174does a C<select> on the filehandle. This is useful if you want to
1175synchronously wait for some requests to finish).
1176
1177See C<nreqs> for an example.
1178
1179=item IO::AIO::poll
1180
1181Waits until some requests have been handled.
1182
1183Returns the number of requests processed, but is otherwise strictly
1184equivalent to:
1185
1186 IO::AIO::poll_wait, IO::AIO::poll_cb
1187
1188=item IO::AIO::flush
1189
1190Wait till all outstanding AIO requests have been handled.
1191
1192Strictly equivalent to:
1193
1194 IO::AIO::poll_wait, IO::AIO::poll_cb
1195 while IO::AIO::nreqs;
1196
1197=back 1526=back
1198 1527
1199=head3 CONTROLLING THE NUMBER OF THREADS 1528=head3 CONTROLLING THE NUMBER OF THREADS
1200 1529
1201=over 1530=over
… …
1234 1563
1235Under normal circumstances you don't need to call this function. 1564Under normal circumstances you don't need to call this function.
1236 1565
1237=item IO::AIO::max_idle $nthreads 1566=item IO::AIO::max_idle $nthreads
1238 1567
1239Limit the number of threads (default: 4) that are allowed to idle (i.e., 1568Limit the number of threads (default: 4) that are allowed to idle
1240threads that did not get a request to process within 10 seconds). That 1569(i.e., threads that did not get a request to process within the idle
1241means if a thread becomes idle while C<$nthreads> other threads are also 1570timeout (default: 10 seconds). That means if a thread becomes idle while
1242idle, it will free its resources and exit. 1571C<$nthreads> other threads are also idle, it will free its resources and
1572exit.
1243 1573
1244This is useful when you allow a large number of threads (e.g. 100 or 1000) 1574This is useful when you allow a large number of threads (e.g. 100 or 1000)
1245to allow for extremely high load situations, but want to free resources 1575to allow for extremely high load situations, but want to free resources
1246under normal circumstances (1000 threads can easily consume 30MB of RAM). 1576under normal circumstances (1000 threads can easily consume 30MB of RAM).
1247 1577
1248The default is probably ok in most situations, especially if thread 1578The default is probably ok in most situations, especially if thread
1249creation is fast. If thread creation is very slow on your system you might 1579creation is fast. If thread creation is very slow on your system you might
1250want to use larger values. 1580want to use larger values.
1251 1581
1582=item IO::AIO::idle_timeout $seconds
1583
1584Sets the minimum idle timeout (default 10) after which worker threads are
1585allowed to exit. SEe C<IO::AIO::max_idle>.
1586
1252=item IO::AIO::max_outstanding $maxreqs 1587=item IO::AIO::max_outstanding $maxreqs
1588
1589Sets the maximum number of outstanding requests to C<$nreqs>. If
1590you do queue up more than this number of requests, the next call to
1591C<IO::AIO::poll_cb> (and other functions calling C<poll_cb>, such as
1592C<IO::AIO::flush> or C<IO::AIO::poll>) will block until the limit is no
1593longer exceeded.
1594
1595In other words, this setting does not enforce a queue limit, but can be
1596used to make poll functions block if the limit is exceeded.
1253 1597
1254This is a very bad function to use in interactive programs because it 1598This is a very bad function to use in interactive programs because it
1255blocks, and a bad way to reduce concurrency because it is inexact: Better 1599blocks, and a bad way to reduce concurrency because it is inexact: Better
1256use an C<aio_group> together with a feed callback. 1600use an C<aio_group> together with a feed callback.
1257 1601
1258Sets the maximum number of outstanding requests to C<$nreqs>. If you 1602It's main use is in scripts without an event loop - when you want to stat
1259do queue up more than this number of requests, the next call to the 1603a lot of files, you can write somehting like this:
1260C<poll_cb> (and C<poll_some> and other functions calling C<poll_cb>)
1261function will block until the limit is no longer exceeded.
1262 1604
1263The default value is very large, so there is no practical limit on the 1605 IO::AIO::max_outstanding 32;
1264number of outstanding requests.
1265 1606
1266You can still queue as many requests as you want. Therefore, 1607 for my $path (...) {
1267C<max_outstanding> is mainly useful in simple scripts (with low values) or 1608 aio_stat $path , ...;
1268as a stop gap to shield against fatal memory overflow (with large values). 1609 IO::AIO::poll_cb;
1610 }
1611
1612 IO::AIO::flush;
1613
1614The call to C<poll_cb> inside the loop will normally return instantly, but
1615as soon as more thna C<32> reqeusts are in-flight, it will block until
1616some requests have been handled. This keeps the loop from pushing a large
1617number of C<aio_stat> requests onto the queue.
1618
1619The default value for C<max_outstanding> is very large, so there is no
1620practical limit on the number of outstanding requests.
1269 1621
1270=back 1622=back
1271 1623
1272=head3 STATISTICAL INFORMATION 1624=head3 STATISTICAL INFORMATION
1273 1625
… …
1293Returns the number of requests currently in the pending state (executed, 1645Returns the number of requests currently in the pending state (executed,
1294but not yet processed by poll_cb). 1646but not yet processed by poll_cb).
1295 1647
1296=back 1648=back
1297 1649
1650=head3 MISCELLANEOUS FUNCTIONS
1651
1652IO::AIO implements some functions that might be useful, but are not
1653asynchronous.
1654
1655=over 4
1656
1657=item IO::AIO::sendfile $ofh, $ifh, $offset, $count
1658
1659Calls the C<eio_sendfile_sync> function, which is like C<aio_sendfile>,
1660but is blocking (this makes most sense if you know the input data is
1661likely cached already and the output filehandle is set to non-blocking
1662operations).
1663
1664Returns the number of bytes copied, or C<-1> on error.
1665
1666=item IO::AIO::fadvise $fh, $offset, $len, $advice
1667
1668Simply calls the C<posix_fadvise> function (see its
1669manpage for details). The following advice constants are
1670avaiable: C<IO::AIO::FADV_NORMAL>, C<IO::AIO::FADV_SEQUENTIAL>,
1671C<IO::AIO::FADV_RANDOM>, C<IO::AIO::FADV_NOREUSE>,
1672C<IO::AIO::FADV_WILLNEED>, C<IO::AIO::FADV_DONTNEED>.
1673
1674On systems that do not implement C<posix_fadvise>, this function returns
1675ENOSYS, otherwise the return value of C<posix_fadvise>.
1676
1677=item IO::AIO::madvise $scalar, $offset, $len, $advice
1678
1679Simply calls the C<posix_madvise> function (see its
1680manpage for details). The following advice constants are
1681avaiable: C<IO::AIO::MADV_NORMAL>, C<IO::AIO::MADV_SEQUENTIAL>,
1682C<IO::AIO::MADV_RANDOM>, C<IO::AIO::MADV_WILLNEED>, C<IO::AIO::MADV_DONTNEED>.
1683
1684On systems that do not implement C<posix_madvise>, this function returns
1685ENOSYS, otherwise the return value of C<posix_madvise>.
1686
1687=item IO::AIO::mprotect $scalar, $offset, $len, $protect
1688
1689Simply calls the C<mprotect> function on the preferably AIO::mmap'ed
1690$scalar (see its manpage for details). The following protect
1691constants are avaiable: C<IO::AIO::PROT_NONE>, C<IO::AIO::PROT_READ>,
1692C<IO::AIO::PROT_WRITE>, C<IO::AIO::PROT_EXEC>.
1693
1694On systems that do not implement C<mprotect>, this function returns
1695ENOSYS, otherwise the return value of C<mprotect>.
1696
1697=item IO::AIO::mmap $scalar, $length, $prot, $flags, $fh[, $offset]
1698
1699Memory-maps a file (or anonymous memory range) and attaches it to the
1700given C<$scalar>, which will act like a string scalar.
1701
1702The only operations allowed on the scalar are C<substr>/C<vec> that don't
1703change the string length, and most read-only operations such as copying it
1704or searching it with regexes and so on.
1705
1706Anything else is unsafe and will, at best, result in memory leaks.
1707
1708The memory map associated with the C<$scalar> is automatically removed
1709when the C<$scalar> is destroyed, or when the C<IO::AIO::mmap> or
1710C<IO::AIO::munmap> functions are called.
1711
1712This calls the C<mmap>(2) function internally. See your system's manual
1713page for details on the C<$length>, C<$prot> and C<$flags> parameters.
1714
1715The C<$length> must be larger than zero and smaller than the actual
1716filesize.
1717
1718C<$prot> is a combination of C<IO::AIO::PROT_NONE>, C<IO::AIO::PROT_EXEC>,
1719C<IO::AIO::PROT_READ> and/or C<IO::AIO::PROT_WRITE>,
1720
1721C<$flags> can be a combination of C<IO::AIO::MAP_SHARED> or
1722C<IO::AIO::MAP_PRIVATE>, or a number of system-specific flags (when
1723not available, the are defined as 0): C<IO::AIO::MAP_ANONYMOUS>
1724(which is set to C<MAP_ANON> if your system only provides this
1725constant), C<IO::AIO::MAP_HUGETLB>, C<IO::AIO::MAP_LOCKED>,
1726C<IO::AIO::MAP_NORESERVE>, C<IO::AIO::MAP_POPULATE> or
1727C<IO::AIO::MAP_NONBLOCK>
1728
1729If C<$fh> is C<undef>, then a file descriptor of C<-1> is passed.
1730
1731C<$offset> is the offset from the start of the file - it generally must be
1732a multiple of C<IO::AIO::PAGESIZE> and defaults to C<0>.
1733
1734Example:
1735
1736 use Digest::MD5;
1737 use IO::AIO;
1738
1739 open my $fh, "<verybigfile"
1740 or die "$!";
1741
1742 IO::AIO::mmap my $data, -s $fh, IO::AIO::PROT_READ, IO::AIO::MAP_SHARED, $fh
1743 or die "verybigfile: $!";
1744
1745 my $fast_md5 = md5 $data;
1746
1747=item IO::AIO::munmap $scalar
1748
1749Removes a previous mmap and undefines the C<$scalar>.
1750
1751=item IO::AIO::munlock $scalar, $offset = 0, $length = undef
1752
1753Calls the C<munlock> function, undoing the effects of a previous
1754C<aio_mlock> call (see its description for details).
1755
1756=item IO::AIO::munlockall
1757
1758Calls the C<munlockall> function.
1759
1760On systems that do not implement C<munlockall>, this function returns
1761ENOSYS, otherwise the return value of C<munlockall>.
1762
1763=back
1764
1298=cut 1765=cut
1299 1766
1300min_parallel 8; 1767min_parallel 8;
1301 1768
1302END { flush } 1769END { flush }
1303 1770
13041; 17711;
1305 1772
1773=head1 EVENT LOOP INTEGRATION
1774
1775It is recommended to use L<AnyEvent::AIO> to integrate IO::AIO
1776automatically into many event loops:
1777
1778 # AnyEvent integration (EV, Event, Glib, Tk, POE, urxvt, pureperl...)
1779 use AnyEvent::AIO;
1780
1781You can also integrate IO::AIO manually into many event loops, here are
1782some examples of how to do this:
1783
1784 # EV integration
1785 my $aio_w = EV::io IO::AIO::poll_fileno, EV::READ, \&IO::AIO::poll_cb;
1786
1787 # Event integration
1788 Event->io (fd => IO::AIO::poll_fileno,
1789 poll => 'r',
1790 cb => \&IO::AIO::poll_cb);
1791
1792 # Glib/Gtk2 integration
1793 add_watch Glib::IO IO::AIO::poll_fileno,
1794 in => sub { IO::AIO::poll_cb; 1 };
1795
1796 # Tk integration
1797 Tk::Event::IO->fileevent (IO::AIO::poll_fileno, "",
1798 readable => \&IO::AIO::poll_cb);
1799
1800 # Danga::Socket integration
1801 Danga::Socket->AddOtherFds (IO::AIO::poll_fileno =>
1802 \&IO::AIO::poll_cb);
1803
1306=head2 FORK BEHAVIOUR 1804=head2 FORK BEHAVIOUR
1307 1805
1308This module should do "the right thing" when the process using it forks: 1806Usage of pthreads in a program changes the semantics of fork
1807considerably. Specifically, only async-safe functions can be called after
1808fork. Perl doesn't know about this, so in general, you cannot call fork
1809with defined behaviour in perl. IO::AIO uses pthreads, so this applies,
1810but many other extensions and (for inexplicable reasons) perl itself often
1811is linked against pthreads, so this limitation applies.
1309 1812
1310Before the fork, IO::AIO enters a quiescent state where no requests 1813Some operating systems have extensions that allow safe use of fork, and
1311can be added in other threads and no results will be processed. After 1814this module should do "the right thing" on those, and tries on others. At
1312the fork the parent simply leaves the quiescent state and continues 1815the time of this writing (2011) only GNU/Linux supports these extensions
1313request/result processing, while the child frees the request/result queue 1816to POSIX.
1314(so that the requests started before the fork will only be handled in the
1315parent). Threads will be started on demand until the limit set in the
1316parent process has been reached again.
1317
1318In short: the parent will, after a short pause, continue as if fork had
1319not been called, while the child will act as if IO::AIO has not been used
1320yet.
1321 1817
1322=head2 MEMORY USAGE 1818=head2 MEMORY USAGE
1323 1819
1324Per-request usage: 1820Per-request usage:
1325 1821
… …
1342 1838
1343Known bugs will be fixed in the next release. 1839Known bugs will be fixed in the next release.
1344 1840
1345=head1 SEE ALSO 1841=head1 SEE ALSO
1346 1842
1347L<Coro::AIO>. 1843L<AnyEvent::AIO> for easy integration into event loops, L<Coro::AIO> for a
1844more natural syntax.
1348 1845
1349=head1 AUTHOR 1846=head1 AUTHOR
1350 1847
1351 Marc Lehmann <schmorp@schmorp.de> 1848 Marc Lehmann <schmorp@schmorp.de>
1352 http://home.schmorp.de/ 1849 http://home.schmorp.de/

Diff Legend

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