… | |
… | |
15 | |
15 | |
16 | aio_read $fh, 30000, 1024, $buffer, 0, sub { |
16 | aio_read $fh, 30000, 1024, $buffer, 0, sub { |
17 | $_[0] > 0 or die "read error: $!"; |
17 | $_[0] > 0 or die "read error: $!"; |
18 | }; |
18 | }; |
19 | |
19 | |
20 | # Event |
20 | # version 2+ has request and group objects |
|
|
21 | use IO::AIO 2; |
|
|
22 | |
|
|
23 | aioreq_pri 4; # give next request a very high priority |
|
|
24 | my $req = aio_unlink "/tmp/file", sub { }; |
|
|
25 | $req->cancel; # cancel request if still in queue |
|
|
26 | |
|
|
27 | my $grp = aio_group sub { print "all stats done\n" }; |
|
|
28 | add $grp aio_stat "..." for ...; |
|
|
29 | |
|
|
30 | # AnyEvent integration |
|
|
31 | open my $fh, "<&=" . IO::AIO::poll_fileno or die "$!"; |
|
|
32 | my $w = AnyEvent->io (fh => $fh, poll => 'r', cb => sub { IO::AIO::poll_cb }); |
|
|
33 | |
|
|
34 | # Event integration |
21 | Event->io (fd => IO::AIO::poll_fileno, |
35 | Event->io (fd => IO::AIO::poll_fileno, |
22 | poll => 'r', |
36 | poll => 'r', |
23 | cb => \&IO::AIO::poll_cb); |
37 | cb => \&IO::AIO::poll_cb); |
24 | |
38 | |
25 | # Glib/Gtk2 |
39 | # Glib/Gtk2 integration |
26 | add_watch Glib::IO IO::AIO::poll_fileno, |
40 | add_watch Glib::IO IO::AIO::poll_fileno, |
27 | in => sub { IO::AIO::poll_cb; 1 }; |
41 | in => sub { IO::AIO::poll_cb; 1 }; |
28 | |
42 | |
29 | # Tk |
43 | # Tk integration |
30 | Tk::Event::IO->fileevent (IO::AIO::poll_fileno, "", |
44 | Tk::Event::IO->fileevent (IO::AIO::poll_fileno, "", |
31 | readable => \&IO::AIO::poll_cb); |
45 | readable => \&IO::AIO::poll_cb); |
32 | |
46 | |
33 | # Danga::Socket |
47 | # Danga::Socket integration |
34 | Danga::Socket->AddOtherFds (IO::AIO::poll_fileno => |
48 | Danga::Socket->AddOtherFds (IO::AIO::poll_fileno => |
35 | \&IO::AIO::poll_cb); |
49 | \&IO::AIO::poll_cb); |
36 | |
50 | |
37 | |
|
|
38 | =head1 DESCRIPTION |
51 | =head1 DESCRIPTION |
39 | |
52 | |
40 | This module implements asynchronous I/O using whatever means your |
53 | This module implements asynchronous I/O using whatever means your |
41 | operating system supports. |
54 | operating system supports. |
42 | |
55 | |
43 | Currently, a number of threads are started that execute your read/writes |
56 | Currently, a number of threads are started that execute your read/writes |
44 | and signal their completion. You don't need thread support in your libc or |
57 | and signal their completion. You don't need thread support in perl, and |
45 | perl, and the threads created by this module will not be visible to the |
58 | the threads created by this module will not be visible to perl. In the |
46 | pthreads library. In the future, this module might make use of the native |
59 | future, this module might make use of the native aio functions available |
47 | aio functions available on many operating systems. However, they are often |
60 | on many operating systems. However, they are often not well-supported |
48 | not well-supported (Linux doesn't allow them on normal files currently, |
61 | (Linux doesn't allow them on normal files currently, for example), |
49 | for example), and they would only support aio_read and aio_write, so the |
62 | and they would only support aio_read and aio_write, so the remaining |
50 | remaining functionality would have to be implemented using threads anyway. |
63 | functionality would have to be implemented using threads anyway. |
51 | |
64 | |
52 | Although the module will work with in the presence of other threads, it is |
65 | Although the module will work with in the presence of other threads, |
53 | currently not reentrant, so use appropriate locking yourself, always call |
66 | it is currently not reentrant in any way, so use appropriate locking |
54 | C<poll_cb> from within the same thread, or never call C<poll_cb> (or other |
67 | yourself, always call C<poll_cb> from within the same thread, or never |
55 | C<aio_> functions) recursively. |
68 | call C<poll_cb> (or other C<aio_> functions) recursively. |
56 | |
69 | |
57 | =cut |
70 | =cut |
58 | |
71 | |
59 | package IO::AIO; |
72 | package IO::AIO; |
60 | |
73 | |
61 | no warnings; |
74 | no warnings; |
|
|
75 | use strict 'vars'; |
62 | |
76 | |
63 | use base 'Exporter'; |
77 | use base 'Exporter'; |
64 | |
78 | |
65 | use Fcntl (); |
|
|
66 | |
|
|
67 | BEGIN { |
79 | BEGIN { |
68 | $VERSION = 1.3; |
80 | our $VERSION = '2.0'; |
69 | |
81 | |
|
|
82 | our @AIO_REQ = qw(aio_sendfile aio_read aio_write aio_open aio_close aio_stat |
|
|
83 | aio_lstat aio_unlink aio_rmdir aio_readdir aio_scandir aio_symlink |
|
|
84 | aio_fsync aio_fdatasync aio_readahead aio_rename aio_link aio_move |
|
|
85 | aio_group aio_nop); |
|
|
86 | our @EXPORT = (@AIO_REQ, qw(aioreq_pri)); |
|
|
87 | our @EXPORT_OK = qw(poll_fileno poll_cb poll_wait flush |
|
|
88 | min_parallel max_parallel max_outstanding nreqs); |
70 | |
89 | |
71 | @EXPORT = qw(aio_read aio_write aio_open aio_close aio_stat aio_lstat aio_unlink |
90 | @IO::AIO::GRP::ISA = 'IO::AIO::REQ'; |
72 | aio_rmdir aio_symlink aio_fsync aio_fdatasync aio_readahead); |
|
|
73 | @EXPORT_OK = qw(poll_fileno poll_cb min_parallel max_parallel max_outstanding nreqs); |
|
|
74 | |
91 | |
75 | require XSLoader; |
92 | require XSLoader; |
76 | XSLoader::load IO::AIO, $VERSION; |
93 | XSLoader::load ("IO::AIO", $VERSION); |
77 | } |
94 | } |
78 | |
95 | |
79 | =head1 FUNCTIONS |
96 | =head1 FUNCTIONS |
80 | |
97 | |
81 | =head2 AIO FUNCTIONS |
98 | =head2 AIO FUNCTIONS |
… | |
… | |
89 | syscall has been executed asynchronously. |
106 | syscall has been executed asynchronously. |
90 | |
107 | |
91 | All functions expecting a filehandle keep a copy of the filehandle |
108 | All functions expecting a filehandle keep a copy of the filehandle |
92 | internally until the request has finished. |
109 | internally until the request has finished. |
93 | |
110 | |
|
|
111 | All requests return objects of type L<IO::AIO::REQ> that allow further |
|
|
112 | manipulation of those requests while they are in-flight. |
|
|
113 | |
94 | The pathnames you pass to these routines I<must> be absolute and |
114 | The pathnames you pass to these routines I<must> be absolute and |
95 | encoded in byte form. The reason for the former is that at the time the |
115 | encoded in byte form. The reason for the former is that at the time the |
96 | request is being executed, the current working directory could have |
116 | request is being executed, the current working directory could have |
97 | changed. Alternatively, you can make sure that you never change the |
117 | changed. Alternatively, you can make sure that you never change the |
98 | current working directory. |
118 | current working directory. |
… | |
… | |
104 | environment, d) use Glib::filename_from_unicode on unicode filenames or e) |
124 | environment, d) use Glib::filename_from_unicode on unicode filenames or e) |
105 | use something else. |
125 | use something else. |
106 | |
126 | |
107 | =over 4 |
127 | =over 4 |
108 | |
128 | |
|
|
129 | =item aioreq_pri $pri |
|
|
130 | |
|
|
131 | Sets the priority for the next aio request. The default priority |
|
|
132 | is C<0>, the minimum and maximum priorities are C<-4> and C<4>, |
|
|
133 | respectively. Requests with higher priority will be serviced first. |
|
|
134 | |
|
|
135 | The priority will be reset to C<0> after each call to one of the C<aio_> |
|
|
136 | functions. |
|
|
137 | |
109 | =item aio_open $pathname, $flags, $mode, $callback |
138 | =item aio_open $pathname, $flags, $mode, $callback->($fh) |
110 | |
139 | |
111 | Asynchronously open or create a file and call the callback with a newly |
140 | Asynchronously open or create a file and call the callback with a newly |
112 | created filehandle for the file. |
141 | created filehandle for the file. |
113 | |
142 | |
114 | The pathname passed to C<aio_open> must be absolute. See API NOTES, above, |
143 | The pathname passed to C<aio_open> must be absolute. See API NOTES, above, |
… | |
… | |
131 | } else { |
160 | } else { |
132 | die "open failed: $!\n"; |
161 | die "open failed: $!\n"; |
133 | } |
162 | } |
134 | }; |
163 | }; |
135 | |
164 | |
136 | =item aio_close $fh, $callback |
165 | =item aio_close $fh, $callback->($status) |
137 | |
166 | |
138 | Asynchronously close a file and call the callback with the result |
167 | Asynchronously close a file and call the callback with the result |
139 | code. I<WARNING:> although accepted, you should not pass in a perl |
168 | code. I<WARNING:> although accepted, you should not pass in a perl |
140 | filehandle here, as perl will likely close the file descriptor another |
169 | filehandle here, as perl will likely close the file descriptor another |
141 | time when the filehandle is destroyed. Normally, you can safely call perls |
170 | time when the filehandle is destroyed. Normally, you can safely call perls |
142 | C<close> or just let filehandles go out of scope. |
171 | C<close> or just let filehandles go out of scope. |
143 | |
172 | |
144 | This is supposed to be a bug in the API, so that might change. It's |
173 | This is supposed to be a bug in the API, so that might change. It's |
145 | therefore best to avoid this function. |
174 | therefore best to avoid this function. |
146 | |
175 | |
147 | =item aio_read $fh,$offset,$length, $data,$dataoffset,$callback |
176 | =item aio_read $fh,$offset,$length, $data,$dataoffset, $callback->($retval) |
148 | |
177 | |
149 | =item aio_write $fh,$offset,$length, $data,$dataoffset,$callback |
178 | =item aio_write $fh,$offset,$length, $data,$dataoffset, $callback->($retval) |
150 | |
179 | |
151 | Reads or writes C<length> bytes from the specified C<fh> and C<offset> |
180 | Reads or writes C<length> bytes from the specified C<fh> and C<offset> |
152 | into the scalar given by C<data> and offset C<dataoffset> and calls the |
181 | into the scalar given by C<data> and offset C<dataoffset> and calls the |
153 | callback without the actual number of bytes read (or -1 on error, just |
182 | callback without the actual number of bytes read (or -1 on error, just |
154 | like the syscall). |
183 | like the syscall). |
… | |
… | |
163 | aio_read $fh, 7, 15, $buffer, 0, sub { |
192 | aio_read $fh, 7, 15, $buffer, 0, sub { |
164 | $_[0] > 0 or die "read error: $!"; |
193 | $_[0] > 0 or die "read error: $!"; |
165 | print "read $_[0] bytes: <$buffer>\n"; |
194 | print "read $_[0] bytes: <$buffer>\n"; |
166 | }; |
195 | }; |
167 | |
196 | |
|
|
197 | =item aio_move $srcpath, $dstpath, $callback->($status) |
|
|
198 | |
|
|
199 | [EXPERIMENTAL due to internal aio_group use] |
|
|
200 | |
|
|
201 | Try to move the I<file> (directories not supported as either source or |
|
|
202 | destination) from C<$srcpath> to C<$dstpath> and call the callback with |
|
|
203 | the C<0> (error) or C<-1> ok. |
|
|
204 | |
|
|
205 | This is a composite request that tries to rename(2) the file first. If |
|
|
206 | rename files with C<EXDEV>, it creates the destination file with mode 0200 |
|
|
207 | and copies the contents of the source file into it using C<aio_sendfile>, |
|
|
208 | followed by restoring atime, mtime, access mode and uid/gid, in that |
|
|
209 | order, and unlinking the C<$srcpath>. |
|
|
210 | |
|
|
211 | If an error occurs, the partial destination file will be unlinked, if |
|
|
212 | possible, except when setting atime, mtime, access mode and uid/gid, where |
|
|
213 | errors are being ignored. |
|
|
214 | |
|
|
215 | =cut |
|
|
216 | |
|
|
217 | sub aio_move($$$) { |
|
|
218 | my ($src, $dst, $cb) = @_; |
|
|
219 | |
|
|
220 | my $grp = aio_group $cb; |
|
|
221 | |
|
|
222 | add $grp aio_rename $src, $dst, sub { |
|
|
223 | if ($_[0] && $! == EXDEV) { |
|
|
224 | add $grp aio_open $src, O_RDONLY, 0, sub { |
|
|
225 | if (my $src_fh = $_[0]) { |
|
|
226 | my @stat = stat $src_fh; |
|
|
227 | |
|
|
228 | add $grp aio_open $dst, O_WRONLY, 0200, sub { |
|
|
229 | if (my $dst_fh = $_[0]) { |
|
|
230 | add $grp aio_sendfile $dst_fh, $src_fh, 0, $stat[7], sub { |
|
|
231 | close $src_fh; |
|
|
232 | |
|
|
233 | if ($_[0] == $stat[7]) { |
|
|
234 | utime $stat[8], $stat[9], $dst; |
|
|
235 | chmod $stat[2] & 07777, $dst_fh; |
|
|
236 | chown $stat[4], $stat[5], $dst_fh; |
|
|
237 | close $dst_fh; |
|
|
238 | |
|
|
239 | add $grp aio_unlink $src, sub { |
|
|
240 | $grp->result ($_[0]); |
|
|
241 | }; |
|
|
242 | } else { |
|
|
243 | my $errno = $!; |
|
|
244 | add $grp aio_unlink $dst, sub { |
|
|
245 | $! = $errno; |
|
|
246 | $grp->result (-1); |
|
|
247 | }; |
|
|
248 | } |
|
|
249 | }; |
|
|
250 | } else { |
|
|
251 | $grp->result (-1); |
|
|
252 | } |
|
|
253 | }, |
|
|
254 | |
|
|
255 | } else { |
|
|
256 | $grp->result (-1); |
|
|
257 | } |
|
|
258 | }; |
|
|
259 | } else { |
|
|
260 | $grp->result ($_[0]); |
|
|
261 | } |
|
|
262 | }; |
|
|
263 | |
|
|
264 | $grp |
|
|
265 | } |
|
|
266 | |
|
|
267 | =item aio_sendfile $out_fh, $in_fh, $in_offset, $length, $callback->($retval) |
|
|
268 | |
|
|
269 | Tries to copy C<$length> bytes from C<$in_fh> to C<$out_fh>. It starts |
|
|
270 | reading at byte offset C<$in_offset>, and starts writing at the current |
|
|
271 | file offset of C<$out_fh>. Because of that, it is not safe to issue more |
|
|
272 | than one C<aio_sendfile> per C<$out_fh>, as they will interfere with each |
|
|
273 | other. |
|
|
274 | |
|
|
275 | This call tries to make use of a native C<sendfile> syscall to provide |
|
|
276 | zero-copy operation. For this to work, C<$out_fh> should refer to a |
|
|
277 | socket, and C<$in_fh> should refer to mmap'able file. |
|
|
278 | |
|
|
279 | If the native sendfile call fails or is not implemented, it will be |
|
|
280 | emulated, so you can call C<aio_sendfile> on any type of filehandle |
|
|
281 | regardless of the limitations of the operating system. |
|
|
282 | |
|
|
283 | Please note, however, that C<aio_sendfile> can read more bytes from |
|
|
284 | C<$in_fh> than are written, and there is no way to find out how many |
|
|
285 | bytes have been read from C<aio_sendfile> alone, as C<aio_sendfile> only |
|
|
286 | provides the number of bytes written to C<$out_fh>. Only if the result |
|
|
287 | value equals C<$length> one can assume that C<$length> bytes have been |
|
|
288 | read. |
|
|
289 | |
168 | =item aio_readahead $fh,$offset,$length, $callback |
290 | =item aio_readahead $fh,$offset,$length, $callback->($retval) |
169 | |
291 | |
170 | C<aio_readahead> populates the page cache with data from a file so that |
292 | C<aio_readahead> populates the page cache with data from a file so that |
171 | subsequent reads from that file will not block on disk I/O. The C<$offset> |
293 | subsequent reads from that file will not block on disk I/O. The C<$offset> |
172 | argument specifies the starting point from which data is to be read and |
294 | argument specifies the starting point from which data is to be read and |
173 | C<$length> specifies the number of bytes to be read. I/O is performed in |
295 | C<$length> specifies the number of bytes to be read. I/O is performed in |
… | |
… | |
177 | file. The current file offset of the file is left unchanged. |
299 | file. The current file offset of the file is left unchanged. |
178 | |
300 | |
179 | If that syscall doesn't exist (likely if your OS isn't Linux) it will be |
301 | If that syscall doesn't exist (likely if your OS isn't Linux) it will be |
180 | emulated by simply reading the data, which would have a similar effect. |
302 | emulated by simply reading the data, which would have a similar effect. |
181 | |
303 | |
182 | =item aio_stat $fh_or_path, $callback |
304 | =item aio_stat $fh_or_path, $callback->($status) |
183 | |
305 | |
184 | =item aio_lstat $fh, $callback |
306 | =item aio_lstat $fh, $callback->($status) |
185 | |
307 | |
186 | Works like perl's C<stat> or C<lstat> in void context. The callback will |
308 | Works like perl's C<stat> or C<lstat> in void context. The callback will |
187 | be called after the stat and the results will be available using C<stat _> |
309 | be called after the stat and the results will be available using C<stat _> |
188 | or C<-s _> etc... |
310 | or C<-s _> etc... |
189 | |
311 | |
… | |
… | |
199 | aio_stat "/etc/passwd", sub { |
321 | aio_stat "/etc/passwd", sub { |
200 | $_[0] and die "stat failed: $!"; |
322 | $_[0] and die "stat failed: $!"; |
201 | print "size is ", -s _, "\n"; |
323 | print "size is ", -s _, "\n"; |
202 | }; |
324 | }; |
203 | |
325 | |
204 | =item aio_unlink $pathname, $callback |
326 | =item aio_unlink $pathname, $callback->($status) |
205 | |
327 | |
206 | Asynchronously unlink (delete) a file and call the callback with the |
328 | Asynchronously unlink (delete) a file and call the callback with the |
207 | result code. |
329 | result code. |
208 | |
330 | |
|
|
331 | =item aio_link $srcpath, $dstpath, $callback->($status) |
|
|
332 | |
|
|
333 | Asynchronously create a new link to the existing object at C<$srcpath> at |
|
|
334 | the path C<$dstpath> and call the callback with the result code. |
|
|
335 | |
|
|
336 | =item aio_symlink $srcpath, $dstpath, $callback->($status) |
|
|
337 | |
|
|
338 | Asynchronously create a new symbolic link to the existing object at C<$srcpath> at |
|
|
339 | the path C<$dstpath> and call the callback with the result code. |
|
|
340 | |
|
|
341 | =item aio_rename $srcpath, $dstpath, $callback->($status) |
|
|
342 | |
|
|
343 | Asynchronously rename the object at C<$srcpath> to C<$dstpath>, just as |
|
|
344 | rename(2) and call the callback with the result code. |
|
|
345 | |
209 | =item aio_rmdir $pathname, $callback |
346 | =item aio_rmdir $pathname, $callback->($status) |
210 | |
347 | |
211 | Asynchronously rmdir (delete) a directory and call the callback with the |
348 | Asynchronously rmdir (delete) a directory and call the callback with the |
212 | result code. |
349 | result code. |
213 | |
350 | |
|
|
351 | =item aio_readdir $pathname, $callback->($entries) |
|
|
352 | |
|
|
353 | Unlike the POSIX call of the same name, C<aio_readdir> reads an entire |
|
|
354 | directory (i.e. opendir + readdir + closedir). The entries will not be |
|
|
355 | sorted, and will B<NOT> include the C<.> and C<..> entries. |
|
|
356 | |
|
|
357 | The callback a single argument which is either C<undef> or an array-ref |
|
|
358 | with the filenames. |
|
|
359 | |
|
|
360 | =item aio_scandir $path, $maxreq, $callback->($dirs, $nondirs) |
|
|
361 | |
|
|
362 | [EXPERIMENTAL due to internal aio_group use] |
|
|
363 | |
|
|
364 | Scans a directory (similar to C<aio_readdir>) but additionally tries to |
|
|
365 | separate the entries of directory C<$path> into two sets of names, ones |
|
|
366 | you can recurse into (directories or links to them), and ones you cannot |
|
|
367 | recurse into (everything else). |
|
|
368 | |
|
|
369 | C<aio_scandir> is a composite request that creates of many sub requests_ |
|
|
370 | C<$maxreq> specifies the maximum number of outstanding aio requests that |
|
|
371 | this function generates. If it is C<< <= 0 >>, then a suitable default |
|
|
372 | will be chosen (currently 6). |
|
|
373 | |
|
|
374 | On error, the callback is called without arguments, otherwise it receives |
|
|
375 | two array-refs with path-relative entry names. |
|
|
376 | |
|
|
377 | Example: |
|
|
378 | |
|
|
379 | aio_scandir $dir, 0, sub { |
|
|
380 | my ($dirs, $nondirs) = @_; |
|
|
381 | print "real directories: @$dirs\n"; |
|
|
382 | print "everything else: @$nondirs\n"; |
|
|
383 | }; |
|
|
384 | |
|
|
385 | Implementation notes. |
|
|
386 | |
|
|
387 | The C<aio_readdir> cannot be avoided, but C<stat()>'ing every entry can. |
|
|
388 | |
|
|
389 | After reading the directory, the modification time, size etc. of the |
|
|
390 | directory before and after the readdir is checked, and if they match (and |
|
|
391 | isn't the current time), the link count will be used to decide how many |
|
|
392 | entries are directories (if >= 2). Otherwise, no knowledge of the number |
|
|
393 | of subdirectories will be assumed. |
|
|
394 | |
|
|
395 | Then entries will be sorted into likely directories (everything without |
|
|
396 | a non-initial dot currently) and likely non-directories (everything |
|
|
397 | else). Then every entry plus an appended C</.> will be C<stat>'ed, |
|
|
398 | likely directories first. If that succeeds, it assumes that the entry |
|
|
399 | is a directory or a symlink to directory (which will be checked |
|
|
400 | seperately). This is often faster than stat'ing the entry itself because |
|
|
401 | filesystems might detect the type of the entry without reading the inode |
|
|
402 | data (e.g. ext2fs filetype feature). |
|
|
403 | |
|
|
404 | If the known number of directories (link count - 2) has been reached, the |
|
|
405 | rest of the entries is assumed to be non-directories. |
|
|
406 | |
|
|
407 | This only works with certainty on POSIX (= UNIX) filesystems, which |
|
|
408 | fortunately are the vast majority of filesystems around. |
|
|
409 | |
|
|
410 | It will also likely work on non-POSIX filesystems with reduced efficiency |
|
|
411 | as those tend to return 0 or 1 as link counts, which disables the |
|
|
412 | directory counting heuristic. |
|
|
413 | |
|
|
414 | =cut |
|
|
415 | |
|
|
416 | sub aio_scandir($$$) { |
|
|
417 | my ($path, $maxreq, $cb) = @_; |
|
|
418 | |
|
|
419 | my $grp = aio_group $cb; |
|
|
420 | |
|
|
421 | $maxreq = 6 if $maxreq <= 0; |
|
|
422 | |
|
|
423 | # stat once |
|
|
424 | add $grp aio_stat $path, sub { |
|
|
425 | return $grp->result () if $_[0]; |
|
|
426 | my $now = time; |
|
|
427 | my $hash1 = join ":", (stat _)[0,1,3,7,9]; |
|
|
428 | |
|
|
429 | # read the directory entries |
|
|
430 | add $grp aio_readdir $path, sub { |
|
|
431 | my $entries = shift |
|
|
432 | or return $grp->result (); |
|
|
433 | |
|
|
434 | # stat the dir another time |
|
|
435 | add $grp aio_stat $path, sub { |
|
|
436 | my $hash2 = join ":", (stat _)[0,1,3,7,9]; |
|
|
437 | |
|
|
438 | my $ndirs; |
|
|
439 | |
|
|
440 | # take the slow route if anything looks fishy |
|
|
441 | if ($hash1 ne $hash2 or (stat _)[9] == $now) { |
|
|
442 | $ndirs = -1; |
|
|
443 | } else { |
|
|
444 | # if nlink == 2, we are finished |
|
|
445 | # on non-posix-fs's, we rely on nlink < 2 |
|
|
446 | $ndirs = (stat _)[3] - 2 |
|
|
447 | or return $grp->result ([], $entries); |
|
|
448 | } |
|
|
449 | |
|
|
450 | # sort into likely dirs and likely nondirs |
|
|
451 | # dirs == files without ".", short entries first |
|
|
452 | $entries = [map $_->[0], |
|
|
453 | sort { $b->[1] cmp $a->[1] } |
|
|
454 | map [$_, sprintf "%s%04d", (/.\./ ? "1" : "0"), length], |
|
|
455 | @$entries]; |
|
|
456 | |
|
|
457 | my (@dirs, @nondirs); |
|
|
458 | |
|
|
459 | my ($statcb, $schedcb); |
|
|
460 | my $nreq = 0; |
|
|
461 | |
|
|
462 | my $statgrp = add $grp aio_group; |
|
|
463 | |
|
|
464 | $schedcb = sub { |
|
|
465 | if (@$entries) { |
|
|
466 | if ($nreq < $maxreq) { |
|
|
467 | my $ent = pop @$entries; |
|
|
468 | $nreq++; |
|
|
469 | add $statgrp aio_stat "$path/$ent/.", sub { $statcb->($_[0], $ent) }; |
|
|
470 | } |
|
|
471 | } elsif (!$nreq) { |
|
|
472 | # finished |
|
|
473 | $statgrp->cancel; |
|
|
474 | undef $statcb; |
|
|
475 | undef $schedcb; |
|
|
476 | $grp->result (\@dirs, \@nondirs); |
|
|
477 | } |
|
|
478 | }; |
|
|
479 | $statcb = sub { |
|
|
480 | my ($status, $entry) = @_; |
|
|
481 | |
|
|
482 | if ($status < 0) { |
|
|
483 | $nreq--; |
|
|
484 | push @nondirs, $entry; |
|
|
485 | &$schedcb; |
|
|
486 | } else { |
|
|
487 | # need to check for real directory |
|
|
488 | add $grp aio_lstat "$path/$entry", sub { |
|
|
489 | $nreq--; |
|
|
490 | |
|
|
491 | if (-d _) { |
|
|
492 | push @dirs, $entry; |
|
|
493 | |
|
|
494 | if (!--$ndirs) { |
|
|
495 | push @nondirs, @$entries; |
|
|
496 | $entries = []; |
|
|
497 | } |
|
|
498 | } else { |
|
|
499 | push @nondirs, $entry; |
|
|
500 | } |
|
|
501 | |
|
|
502 | &$schedcb; |
|
|
503 | } |
|
|
504 | } |
|
|
505 | }; |
|
|
506 | |
|
|
507 | &$schedcb while @$entries && $nreq < $maxreq; |
|
|
508 | }; |
|
|
509 | }; |
|
|
510 | }; |
|
|
511 | |
|
|
512 | $grp |
|
|
513 | } |
|
|
514 | |
214 | =item aio_fsync $fh, $callback |
515 | =item aio_fsync $fh, $callback->($status) |
215 | |
516 | |
216 | Asynchronously call fsync on the given filehandle and call the callback |
517 | Asynchronously call fsync on the given filehandle and call the callback |
217 | with the fsync result code. |
518 | with the fsync result code. |
218 | |
519 | |
219 | =item aio_fdatasync $fh, $callback |
520 | =item aio_fdatasync $fh, $callback->($status) |
220 | |
521 | |
221 | Asynchronously call fdatasync on the given filehandle and call the |
522 | Asynchronously call fdatasync on the given filehandle and call the |
222 | callback with the fdatasync result code. |
523 | callback with the fdatasync result code. |
223 | |
524 | |
224 | If this call isn't available because your OS lacks it or it couldn't be |
525 | If this call isn't available because your OS lacks it or it couldn't be |
225 | detected, it will be emulated by calling C<fsync> instead. |
526 | detected, it will be emulated by calling C<fsync> instead. |
|
|
527 | |
|
|
528 | =item aio_group $callback->(...) |
|
|
529 | |
|
|
530 | [EXPERIMENTAL] |
|
|
531 | |
|
|
532 | This is a very special aio request: Instead of doing something, it is a |
|
|
533 | container for other aio requests, which is useful if you want to bundle |
|
|
534 | many requests into a single, composite, request. |
|
|
535 | |
|
|
536 | Returns an object of class L<IO::AIO::GRP>. See its documentation below |
|
|
537 | for more info. |
|
|
538 | |
|
|
539 | Example: |
|
|
540 | |
|
|
541 | my $grp = aio_group sub { |
|
|
542 | print "all stats done\n"; |
|
|
543 | }; |
|
|
544 | |
|
|
545 | add $grp |
|
|
546 | (aio_stat ...), |
|
|
547 | (aio_stat ...), |
|
|
548 | ...; |
|
|
549 | |
|
|
550 | =item aio_nop $callback->() |
|
|
551 | |
|
|
552 | This is a special request - it does nothing in itself and is only used for |
|
|
553 | side effects, such as when you want to add a dummy request to a group so |
|
|
554 | that finishing the requests in the group depends on executing the given |
|
|
555 | code. |
|
|
556 | |
|
|
557 | While this request does nothing, it still goes through the execution |
|
|
558 | phase and still requires a worker thread. Thus, the callback will not |
|
|
559 | be executed immediately but only after other requests in the queue have |
|
|
560 | entered their execution phase. This can be used to measure request |
|
|
561 | latency. |
|
|
562 | |
|
|
563 | =item IO::AIO::aio_sleep $fractional_seconds, $callback->() *NOT EXPORTED* |
|
|
564 | |
|
|
565 | Mainly used for debugging and benchmarking, this aio request puts one of |
|
|
566 | the request workers to sleep for the given time. |
|
|
567 | |
|
|
568 | While it is theoretically handy to have simple I/O scheduling requests |
|
|
569 | like sleep and file handle readable/writable, the overhead this creates |
|
|
570 | is immense, so do not use this function except to put your application |
|
|
571 | under artificial I/O pressure. |
|
|
572 | |
|
|
573 | =back |
|
|
574 | |
|
|
575 | =head2 IO::AIO::REQ CLASS |
|
|
576 | |
|
|
577 | All non-aggregate C<aio_*> functions return an object of this class when |
|
|
578 | called in non-void context. |
|
|
579 | |
|
|
580 | A request always moves through the following five states in its lifetime, |
|
|
581 | in order: B<ready> (request has been created, but has not been executed |
|
|
582 | yet), B<execute> (request is currently being executed), B<pending> |
|
|
583 | (request has been executed but callback has not been called yet), |
|
|
584 | B<result> (results are being processed synchronously, includes calling the |
|
|
585 | callback) and B<done> (request has reached the end of its lifetime and |
|
|
586 | holds no resources anymore). |
|
|
587 | |
|
|
588 | =over 4 |
|
|
589 | |
|
|
590 | =item cancel $req |
|
|
591 | |
|
|
592 | Cancels the request, if possible. Has the effect of skipping execution |
|
|
593 | when entering the B<execute> state and skipping calling the callback when |
|
|
594 | entering the the B<result> state, but will leave the request otherwise |
|
|
595 | untouched. That means that requests that currently execute will not be |
|
|
596 | stopped and resources held by the request will not be freed prematurely. |
|
|
597 | |
|
|
598 | =item cb $req $callback->(...) |
|
|
599 | |
|
|
600 | Replace (or simply set) the callback registered to the request. |
|
|
601 | |
|
|
602 | =back |
|
|
603 | |
|
|
604 | =head2 IO::AIO::GRP CLASS |
|
|
605 | |
|
|
606 | This class is a subclass of L<IO::AIO::REQ>, so all its methods apply to |
|
|
607 | objects of this class, too. |
|
|
608 | |
|
|
609 | A IO::AIO::GRP object is a special request that can contain multiple other |
|
|
610 | aio requests. |
|
|
611 | |
|
|
612 | You create one by calling the C<aio_group> constructing function with a |
|
|
613 | callback that will be called when all contained requests have entered the |
|
|
614 | C<done> state: |
|
|
615 | |
|
|
616 | my $grp = aio_group sub { |
|
|
617 | print "all requests are done\n"; |
|
|
618 | }; |
|
|
619 | |
|
|
620 | You add requests by calling the C<add> method with one or more |
|
|
621 | C<IO::AIO::REQ> objects: |
|
|
622 | |
|
|
623 | $grp->add (aio_unlink "..."); |
|
|
624 | |
|
|
625 | add $grp aio_stat "...", sub { |
|
|
626 | $_[0] or return $grp->result ("error"); |
|
|
627 | |
|
|
628 | # add another request dynamically, if first succeeded |
|
|
629 | add $grp aio_open "...", sub { |
|
|
630 | $grp->result ("ok"); |
|
|
631 | }; |
|
|
632 | }; |
|
|
633 | |
|
|
634 | This makes it very easy to create composite requests (see the source of |
|
|
635 | C<aio_move> for an application) that work and feel like simple requests. |
|
|
636 | |
|
|
637 | =over 4 |
|
|
638 | |
|
|
639 | =item * The IO::AIO::GRP objects will be cleaned up during calls to |
|
|
640 | C<IO::AIO::poll_cb>, just like any other request. |
|
|
641 | |
|
|
642 | =item * They can be canceled like any other request. Canceling will cancel not |
|
|
643 | only the request itself, but also all requests it contains. |
|
|
644 | |
|
|
645 | =item * They can also can also be added to other IO::AIO::GRP objects. |
|
|
646 | |
|
|
647 | =item * You must not add requests to a group from within the group callback (or |
|
|
648 | any later time). |
|
|
649 | |
|
|
650 | =item * This does not harmonise well with C<max_outstanding>, so best do |
|
|
651 | not combine C<aio_group> with it. Groups and feeders are recommended for |
|
|
652 | this kind of concurrency-limiting. |
|
|
653 | |
|
|
654 | =back |
|
|
655 | |
|
|
656 | Their lifetime, simplified, looks like this: when they are empty, they |
|
|
657 | will finish very quickly. If they contain only requests that are in the |
|
|
658 | C<done> state, they will also finish. Otherwise they will continue to |
|
|
659 | exist. |
|
|
660 | |
|
|
661 | That means after creating a group you have some time to add requests. And |
|
|
662 | in the callbacks of those requests, you can add further requests to the |
|
|
663 | group. And only when all those requests have finished will the the group |
|
|
664 | itself finish. |
|
|
665 | |
|
|
666 | =over 4 |
|
|
667 | |
|
|
668 | =item add $grp ... |
|
|
669 | |
|
|
670 | =item $grp->add (...) |
|
|
671 | |
|
|
672 | Add one or more requests to the group. Any type of L<IO::AIO::REQ> can |
|
|
673 | be added, including other groups, as long as you do not create circular |
|
|
674 | dependencies. |
|
|
675 | |
|
|
676 | Returns all its arguments. |
|
|
677 | |
|
|
678 | =item $grp->result (...) |
|
|
679 | |
|
|
680 | Set the result value(s) that will be passed to the group callback when all |
|
|
681 | subrequests have finished. By default, no argument will be passed. |
|
|
682 | |
|
|
683 | =item feed $grp $callback->($grp) |
|
|
684 | |
|
|
685 | [VERY EXPERIMENTAL] |
|
|
686 | |
|
|
687 | Sets a feeder/generator on this group: every group can have an attached |
|
|
688 | generator that generates requests if idle. The idea behind this is that, |
|
|
689 | although you could just queue as many requests as you want in a group, |
|
|
690 | this might starve other requests for a potentially long time. For |
|
|
691 | example, C<aio_scandir> might generate hundreds of thousands C<aio_stat> |
|
|
692 | requests, delaying any later requests for a long time. |
|
|
693 | |
|
|
694 | To avoid this, and allow incremental generation of requests, you can |
|
|
695 | instead a group and set a feeder on it that generates those requests. The |
|
|
696 | feed callback will be called whenever there are few enough (see C<limit>, |
|
|
697 | below) requests active in the group itself and is expected to queue more |
|
|
698 | requests. |
|
|
699 | |
|
|
700 | The feed callback can queue as many requests as it likes (i.e. C<add> does |
|
|
701 | not impose any limits). |
|
|
702 | |
|
|
703 | If the feed does not queue more requests when called, it will be |
|
|
704 | automatically removed from the group. |
|
|
705 | |
|
|
706 | If the feed limit is C<0>, it will be set to C<2> automatically. |
|
|
707 | |
|
|
708 | Example: |
|
|
709 | |
|
|
710 | # stat all files in @files, but only ever use four aio requests concurrently: |
|
|
711 | |
|
|
712 | my $grp = aio_group sub { print "finished\n" }; |
|
|
713 | limit $grp 4; |
|
|
714 | feed $grp sub { |
|
|
715 | my $file = pop @files |
|
|
716 | or return; |
|
|
717 | |
|
|
718 | add $grp aio_stat $file, sub { ... }; |
|
|
719 | }; |
|
|
720 | |
|
|
721 | =item limit $grp $num |
|
|
722 | |
|
|
723 | Sets the feeder limit for the group: The feeder will be called whenever |
|
|
724 | the group contains less than this many requests. |
|
|
725 | |
|
|
726 | Setting the limit to C<0> will pause the feeding process. |
226 | |
727 | |
227 | =back |
728 | =back |
228 | |
729 | |
229 | =head2 SUPPORT FUNCTIONS |
730 | =head2 SUPPORT FUNCTIONS |
230 | |
731 | |
… | |
… | |
288 | IO::AIO::poll_wait, IO::AIO::poll_cb |
789 | IO::AIO::poll_wait, IO::AIO::poll_cb |
289 | if IO::AIO::nreqs; |
790 | if IO::AIO::nreqs; |
290 | |
791 | |
291 | =item IO::AIO::min_parallel $nthreads |
792 | =item IO::AIO::min_parallel $nthreads |
292 | |
793 | |
293 | Set the minimum number of AIO threads to C<$nthreads>. The default is |
794 | Set the minimum number of AIO threads to C<$nthreads>. The current |
294 | C<1>, which means a single asynchronous operation can be done at one time |
795 | default is C<8>, which means eight asynchronous operations can execute |
295 | (the number of outstanding operations, however, is unlimited). |
796 | concurrently at any one time (the number of outstanding requests, |
|
|
797 | however, is unlimited). |
296 | |
798 | |
|
|
799 | IO::AIO starts threads only on demand, when an AIO request is queued and |
|
|
800 | no free thread exists. |
|
|
801 | |
297 | It is recommended to keep the number of threads low, as some Linux |
802 | It is recommended to keep the number of threads relatively low, as some |
298 | kernel versions will scale negatively with the number of threads (higher |
803 | Linux kernel versions will scale negatively with the number of threads |
299 | parallelity => MUCH higher latency). With current Linux 2.6 versions, 4-32 |
804 | (higher parallelity => MUCH higher latency). With current Linux 2.6 |
300 | threads should be fine. |
805 | versions, 4-32 threads should be fine. |
301 | |
806 | |
302 | Under normal circumstances you don't need to call this function, as this |
807 | Under most circumstances you don't need to call this function, as the |
303 | module automatically starts some threads (the exact number might change, |
808 | module selects a default that is suitable for low to moderate load. |
304 | and is currently 4). |
|
|
305 | |
809 | |
306 | =item IO::AIO::max_parallel $nthreads |
810 | =item IO::AIO::max_parallel $nthreads |
307 | |
811 | |
308 | Sets the maximum number of AIO threads to C<$nthreads>. If more than |
812 | Sets the maximum number of AIO threads to C<$nthreads>. If more than the |
309 | the specified number of threads are currently running, kill them. This |
813 | specified number of threads are currently running, this function kills |
310 | function blocks until the limit is reached. |
814 | them. This function blocks until the limit is reached. |
|
|
815 | |
|
|
816 | While C<$nthreads> are zero, aio requests get queued but not executed |
|
|
817 | until the number of threads has been increased again. |
311 | |
818 | |
312 | This module automatically runs C<max_parallel 0> at program end, to ensure |
819 | This module automatically runs C<max_parallel 0> at program end, to ensure |
313 | that all threads are killed and that there are no outstanding requests. |
820 | that all threads are killed and that there are no outstanding requests. |
314 | |
821 | |
315 | Under normal circumstances you don't need to call this function. |
822 | Under normal circumstances you don't need to call this function. |
316 | |
823 | |
317 | =item $oldnreqs = IO::AIO::max_outstanding $nreqs |
824 | =item $oldnreqs = IO::AIO::max_outstanding $nreqs |
|
|
825 | |
|
|
826 | [DEPRECATED] |
318 | |
827 | |
319 | Sets the maximum number of outstanding requests to C<$nreqs>. If you |
828 | Sets the maximum number of outstanding requests to C<$nreqs>. If you |
320 | try to queue up more than this number of requests, the caller will block until |
829 | try to queue up more than this number of requests, the caller will block until |
321 | some requests have been handled. |
830 | some requests have been handled. |
322 | |
831 | |
323 | The default is very large, so normally there is no practical limit. If you |
832 | The default is very large, so normally there is no practical limit. If you |
324 | queue up many requests in a loop it it often improves speed if you set |
833 | queue up many requests in a loop it often improves speed if you set |
325 | this to a relatively low number, such as C<100>. |
834 | this to a relatively low number, such as C<100>. |
|
|
835 | |
|
|
836 | This function does not work well together with C<aio_group>'s, and their |
|
|
837 | feeder interface is better suited to limiting concurrency, so do not use |
|
|
838 | this function. |
326 | |
839 | |
327 | Under normal circumstances you don't need to call this function. |
840 | Under normal circumstances you don't need to call this function. |
328 | |
841 | |
329 | =back |
842 | =back |
330 | |
843 | |
… | |
… | |
344 | or return undef; |
857 | or return undef; |
345 | |
858 | |
346 | *$sym |
859 | *$sym |
347 | } |
860 | } |
348 | |
861 | |
349 | min_parallel 4; |
862 | min_parallel 8; |
350 | |
863 | |
351 | END { |
864 | END { |
352 | max_parallel 0; |
865 | max_parallel 0; |
353 | } |
866 | } |
354 | |
867 | |
355 | 1; |
868 | 1; |
356 | |
869 | |
357 | =head2 FORK BEHAVIOUR |
870 | =head2 FORK BEHAVIOUR |
358 | |
871 | |
|
|
872 | This module should do "the right thing" when the process using it forks: |
|
|
873 | |
359 | Before the fork IO::AIO enters a quiescent state where no requests can be |
874 | Before the fork, IO::AIO enters a quiescent state where no requests |
360 | added in other threads and no results will be processed. After the fork |
875 | can be added in other threads and no results will be processed. After |
361 | the parent simply leaves the quiescent state and continues request/result |
876 | the fork the parent simply leaves the quiescent state and continues |
362 | processing, while the child clears the request/result queue and starts the |
877 | request/result processing, while the child clears the request/result |
363 | same number of threads as were in use by the parent. |
878 | queue (so the requests started before the fork will only be handled in |
|
|
879 | the parent). Threads will be started on demand until the limit ste in the |
|
|
880 | parent process has been reached again. |
|
|
881 | |
|
|
882 | In short: the parent will, after a short pause, continue as if fork had |
|
|
883 | not been called, while the child will act as if IO::AIO has not been used |
|
|
884 | yet. |
|
|
885 | |
|
|
886 | =head2 MEMORY USAGE |
|
|
887 | |
|
|
888 | Each aio request uses - depending on your architecture - around 128 bytes |
|
|
889 | of memory. In addition, stat requests need a stat buffer (possibly a few |
|
|
890 | hundred bytes). Perl scalars and other data passed into aio requests will |
|
|
891 | also be locked. |
|
|
892 | |
|
|
893 | This is now awfully much, so queuing lots of requests is not usually a |
|
|
894 | problem. |
|
|
895 | |
|
|
896 | Each thread needs a stack area which is usually around 16k, sometimes much |
|
|
897 | larger, depending on the OS. |
364 | |
898 | |
365 | =head1 SEE ALSO |
899 | =head1 SEE ALSO |
366 | |
900 | |
367 | L<Coro>, L<Linux::AIO>. |
901 | L<Coro::AIO>. |
368 | |
902 | |
369 | =head1 AUTHOR |
903 | =head1 AUTHOR |
370 | |
904 | |
371 | Marc Lehmann <schmorp@schmorp.de> |
905 | Marc Lehmann <schmorp@schmorp.de> |
372 | http://home.schmorp.de/ |
906 | http://home.schmorp.de/ |