… | |
… | |
3 | IO::AIO - Asynchronous Input/Output |
3 | IO::AIO - Asynchronous Input/Output |
4 | |
4 | |
5 | =head1 SYNOPSIS |
5 | =head1 SYNOPSIS |
6 | |
6 | |
7 | use IO::AIO; |
7 | use IO::AIO; |
|
|
8 | |
|
|
9 | aio_open "/etc/passwd", O_RDONLY, 0, sub { |
|
|
10 | my ($fh) = @_; |
|
|
11 | ... |
|
|
12 | }; |
|
|
13 | |
|
|
14 | aio_unlink "/tmp/file", sub { }; |
|
|
15 | |
|
|
16 | aio_read $fh, 30000, 1024, $buffer, 0, sub { |
|
|
17 | $_[0] > 0 or die "read error: $!"; |
|
|
18 | }; |
|
|
19 | |
|
|
20 | # version 2+ has request and group objects |
|
|
21 | use IO::AIO 2; |
|
|
22 | |
|
|
23 | my $req = aio_unlink "/tmp/file", sub { }; |
|
|
24 | $req->cancel; # cancel request if still in queue |
|
|
25 | |
|
|
26 | my $grp = aio_group sub { print "all stats done\n" }; |
|
|
27 | add $grp aio_stat "..." for ...; |
|
|
28 | |
|
|
29 | # AnyEvent integration |
|
|
30 | open my $fh, "<&=" . IO::AIO::poll_fileno or die "$!"; |
|
|
31 | my $w = AnyEvent->io (fh => $fh, poll => 'r', cb => sub { IO::AIO::poll_cb }); |
|
|
32 | |
|
|
33 | # Event integration |
|
|
34 | Event->io (fd => IO::AIO::poll_fileno, |
|
|
35 | poll => 'r', |
|
|
36 | cb => \&IO::AIO::poll_cb); |
|
|
37 | |
|
|
38 | # Glib/Gtk2 integration |
|
|
39 | add_watch Glib::IO IO::AIO::poll_fileno, |
|
|
40 | in => sub { IO::AIO::poll_cb; 1 }; |
|
|
41 | |
|
|
42 | # Tk integration |
|
|
43 | Tk::Event::IO->fileevent (IO::AIO::poll_fileno, "", |
|
|
44 | readable => \&IO::AIO::poll_cb); |
|
|
45 | |
|
|
46 | # Danga::Socket integration |
|
|
47 | Danga::Socket->AddOtherFds (IO::AIO::poll_fileno => |
|
|
48 | \&IO::AIO::poll_cb); |
8 | |
49 | |
9 | =head1 DESCRIPTION |
50 | =head1 DESCRIPTION |
10 | |
51 | |
11 | This module implements asynchronous I/O using whatever means your |
52 | This module implements asynchronous I/O using whatever means your |
12 | operating system supports. |
53 | operating system supports. |
13 | |
54 | |
14 | Currently, a number of threads are started that execute your read/writes |
55 | Currently, a number of threads are started that execute your read/writes |
15 | and signal their completion. You don't need thread support in your libc or |
56 | and signal their completion. You don't need thread support in perl, and |
16 | perl, and the threads created by this module will not be visible to the |
57 | the threads created by this module will not be visible to perl. In the |
17 | pthreads library. In the future, this module might make use of the native |
58 | future, this module might make use of the native aio functions available |
18 | aio functions available on many operating systems. However, they are often |
59 | on many operating systems. However, they are often not well-supported |
19 | not well-supported (Linux doesn't allow them on normal files currently, |
60 | (Linux doesn't allow them on normal files currently, for example), |
20 | for example), and they would only support aio_read and aio_write, so the |
61 | and they would only support aio_read and aio_write, so the remaining |
21 | remaining functionality would have to be implemented using threads anyway. |
62 | functionality would have to be implemented using threads anyway. |
22 | |
63 | |
23 | Although the module will work with in the presence of other threads, it is |
64 | Although the module will work with in the presence of other threads, it is |
24 | currently not reentrant, so use appropriate locking yourself. |
65 | currently not reentrant, so use appropriate locking yourself, always call |
|
|
66 | C<poll_cb> from within the same thread, or never call C<poll_cb> (or other |
|
|
67 | C<aio_> functions) recursively. |
25 | |
68 | |
26 | =head2 API NOTES |
69 | =cut |
|
|
70 | |
|
|
71 | package IO::AIO; |
|
|
72 | |
|
|
73 | no warnings; |
|
|
74 | use strict 'vars'; |
|
|
75 | |
|
|
76 | use base 'Exporter'; |
|
|
77 | |
|
|
78 | BEGIN { |
|
|
79 | our $VERSION = '2.0'; |
|
|
80 | |
|
|
81 | our @AIO_REQ = qw(aio_sendfile aio_read aio_write aio_open aio_close aio_stat |
|
|
82 | aio_lstat aio_unlink aio_rmdir aio_readdir aio_scandir aio_symlink |
|
|
83 | aio_fsync aio_fdatasync aio_readahead aio_rename aio_link aio_move |
|
|
84 | aio_group aio_nop); |
|
|
85 | our @EXPORT = (@AIO_REQ, qw(aioreq_pri)); |
|
|
86 | our @EXPORT_OK = qw(poll_fileno poll_cb poll_wait flush |
|
|
87 | min_parallel max_parallel max_outstanding nreqs); |
|
|
88 | |
|
|
89 | @IO::AIO::GRP::ISA = 'IO::AIO::REQ'; |
|
|
90 | |
|
|
91 | require XSLoader; |
|
|
92 | XSLoader::load ("IO::AIO", $VERSION); |
|
|
93 | } |
|
|
94 | |
|
|
95 | =head1 FUNCTIONS |
|
|
96 | |
|
|
97 | =head2 AIO FUNCTIONS |
27 | |
98 | |
28 | All the C<aio_*> calls are more or less thin wrappers around the syscall |
99 | All the C<aio_*> calls are more or less thin wrappers around the syscall |
29 | with the same name (sans C<aio_>). The arguments are similar or identical, |
100 | with the same name (sans C<aio_>). The arguments are similar or identical, |
30 | and they all accept an additional C<$callback> argument which must be |
101 | and they all accept an additional (and optional) C<$callback> argument |
31 | a code reference. This code reference will get called with the syscall |
102 | which must be a code reference. This code reference will get called with |
32 | return code (e.g. most syscalls return C<-1> on error, unlike perl, which |
103 | the syscall return code (e.g. most syscalls return C<-1> on error, unlike |
33 | usually delivers "false") as it's sole argument when the given syscall has |
104 | perl, which usually delivers "false") as it's sole argument when the given |
34 | been executed asynchronously. |
105 | syscall has been executed asynchronously. |
35 | |
106 | |
36 | All functions that expect a filehandle will also accept a file descriptor. |
107 | All functions expecting a filehandle keep a copy of the filehandle |
|
|
108 | internally until the request has finished. |
37 | |
109 | |
|
|
110 | All requests return objects of type L<IO::AIO::REQ> that allow further |
|
|
111 | manipulation of those requests while they are in-flight. |
|
|
112 | |
38 | The filenames you pass to these routines I<must> be absolute. The reason |
113 | The pathnames you pass to these routines I<must> be absolute and |
39 | is that at the time the request is being executed, the current working |
114 | encoded in byte form. The reason for the former is that at the time the |
40 | directory could have changed. Alternatively, you can make sure that you |
115 | request is being executed, the current working directory could have |
|
|
116 | changed. Alternatively, you can make sure that you never change the |
41 | never change the current working directory. |
117 | current working directory. |
|
|
118 | |
|
|
119 | To encode pathnames to byte form, either make sure you either: a) |
|
|
120 | always pass in filenames you got from outside (command line, readdir |
|
|
121 | etc.), b) are ASCII or ISO 8859-1, c) use the Encode module and encode |
|
|
122 | your pathnames to the locale (or other) encoding in effect in the user |
|
|
123 | environment, d) use Glib::filename_from_unicode on unicode filenames or e) |
|
|
124 | use something else. |
42 | |
125 | |
43 | =over 4 |
126 | =over 4 |
44 | |
127 | |
45 | =cut |
|
|
46 | |
|
|
47 | package IO::AIO; |
|
|
48 | |
|
|
49 | use base 'Exporter'; |
|
|
50 | |
|
|
51 | use Fcntl (); |
|
|
52 | |
|
|
53 | BEGIN { |
|
|
54 | $VERSION = 0.2; |
|
|
55 | |
|
|
56 | @EXPORT = qw(aio_read aio_write aio_open aio_close aio_stat aio_lstat aio_unlink |
|
|
57 | aio_fsync aio_fdatasync aio_readahead); |
|
|
58 | @EXPORT_OK = qw(poll_fileno poll_cb min_parallel max_parallel max_outstanding nreqs); |
|
|
59 | |
|
|
60 | require XSLoader; |
|
|
61 | XSLoader::load IO::AIO, $VERSION; |
|
|
62 | } |
|
|
63 | |
|
|
64 | =item IO::AIO::min_parallel $nthreads |
|
|
65 | |
|
|
66 | Set the minimum number of AIO threads to C<$nthreads>. The default is |
|
|
67 | C<1>, which means a single asynchronous operation can be done at one time |
|
|
68 | (the number of outstanding operations, however, is unlimited). |
|
|
69 | |
|
|
70 | It is recommended to keep the number of threads low, as some Linux |
|
|
71 | kernel versions will scale negatively with the number of threads (higher |
|
|
72 | parallelity => MUCH higher latency). With current Linux 2.6 versions, 4-32 |
|
|
73 | threads should be fine. |
|
|
74 | |
|
|
75 | Under normal circumstances you don't need to call this function, as this |
|
|
76 | module automatically starts some threads (the exact number might change, |
|
|
77 | and is currently 4). |
|
|
78 | |
|
|
79 | =item IO::AIO::max_parallel $nthreads |
|
|
80 | |
|
|
81 | Sets the maximum number of AIO threads to C<$nthreads>. If more than |
|
|
82 | the specified number of threads are currently running, kill them. This |
|
|
83 | function blocks until the limit is reached. |
|
|
84 | |
|
|
85 | This module automatically runs C<max_parallel 0> at program end, to ensure |
|
|
86 | that all threads are killed and that there are no outstanding requests. |
|
|
87 | |
|
|
88 | Under normal circumstances you don't need to call this function. |
|
|
89 | |
|
|
90 | =item $oldnreqs = IO::AIO::max_outstanding $nreqs |
|
|
91 | |
|
|
92 | Sets the maximum number of outstanding requests to C<$nreqs>. If you |
|
|
93 | try to queue up more than this number of requests, the caller will block until |
|
|
94 | some requests have been handled. |
|
|
95 | |
|
|
96 | The default is very large, so normally there is no practical limit. If you |
|
|
97 | queue up many requests in a loop it it often improves speed if you set |
|
|
98 | this to a relatively low number, such as C<100>. |
|
|
99 | |
|
|
100 | Under normal circumstances you don't need to call this function. |
|
|
101 | |
|
|
102 | =item $fileno = IO::AIO::poll_fileno |
|
|
103 | |
|
|
104 | Return the I<request result pipe filehandle>. This filehandle must be |
|
|
105 | polled for reading by some mechanism outside this module (e.g. Event |
|
|
106 | or select, see below). If the pipe becomes readable you have to call |
|
|
107 | C<poll_cb> to check the results. |
|
|
108 | |
|
|
109 | See C<poll_cb> for an example. |
|
|
110 | |
|
|
111 | =item IO::AIO::poll_cb |
|
|
112 | |
|
|
113 | Process all outstanding events on the result pipe. You have to call this |
|
|
114 | regularly. Returns the number of events processed. Returns immediately |
|
|
115 | when no events are outstanding. |
|
|
116 | |
|
|
117 | You can use Event to multiplex, e.g.: |
|
|
118 | |
|
|
119 | Event->io (fd => IO::AIO::poll_fileno, |
|
|
120 | poll => 'r', async => 1, |
|
|
121 | cb => \&IO::AIO::poll_cb); |
|
|
122 | |
|
|
123 | =item IO::AIO::poll_wait |
|
|
124 | |
|
|
125 | Wait till the result filehandle becomes ready for reading (simply does a |
|
|
126 | select on the filehandle. This is useful if you want to synchronously wait |
|
|
127 | for some requests to finish). |
|
|
128 | |
|
|
129 | See C<nreqs> for an example. |
|
|
130 | |
|
|
131 | =item IO::AIO::nreqs |
|
|
132 | |
|
|
133 | Returns the number of requests currently outstanding. |
|
|
134 | |
|
|
135 | Example: wait till there are no outstanding requests anymore: |
|
|
136 | |
|
|
137 | IO::AIO::poll_wait, IO::AIO::poll_cb |
|
|
138 | while IO::AIO::nreqs; |
|
|
139 | |
|
|
140 | =item aio_open $pathname, $flags, $mode, $callback |
128 | =item aio_open $pathname, $flags, $mode, $callback->($fh) |
141 | |
129 | |
142 | Asynchronously open or create a file and call the callback with a newly |
130 | Asynchronously open or create a file and call the callback with a newly |
143 | created filehandle for the file. |
131 | created filehandle for the file. |
144 | |
132 | |
145 | The pathname passed to C<aio_open> must be absolute. See API NOTES, above, |
133 | The pathname passed to C<aio_open> must be absolute. See API NOTES, above, |
146 | for an explanation. |
134 | for an explanation. |
147 | |
135 | |
148 | The C<$mode> argument is a bitmask. See the C<Fcntl> module for a |
136 | The C<$flags> argument is a bitmask. See the C<Fcntl> module for a |
149 | list. They are the same as used in C<sysopen>. |
137 | list. They are the same as used by C<sysopen>. |
|
|
138 | |
|
|
139 | Likewise, C<$mode> specifies the mode of the newly created file, if it |
|
|
140 | didn't exist and C<O_CREAT> has been given, just like perl's C<sysopen>, |
|
|
141 | except that it is mandatory (i.e. use C<0> if you don't create new files, |
|
|
142 | and C<0666> or C<0777> if you do). |
150 | |
143 | |
151 | Example: |
144 | Example: |
152 | |
145 | |
153 | aio_open "/etc/passwd", O_RDONLY, 0, sub { |
146 | aio_open "/etc/passwd", O_RDONLY, 0, sub { |
154 | if ($_[0]) { |
147 | if ($_[0]) { |
… | |
… | |
157 | } else { |
150 | } else { |
158 | die "open failed: $!\n"; |
151 | die "open failed: $!\n"; |
159 | } |
152 | } |
160 | }; |
153 | }; |
161 | |
154 | |
162 | =item aio_close $fh, $callback |
155 | =item aio_close $fh, $callback->($status) |
163 | |
156 | |
164 | Asynchronously close a file and call the callback with the result |
157 | Asynchronously close a file and call the callback with the result |
165 | code. I<WARNING:> although accepted, you should not pass in a perl |
158 | code. I<WARNING:> although accepted, you should not pass in a perl |
166 | filehandle here, as perl will likely close the file descriptor itself when |
159 | filehandle here, as perl will likely close the file descriptor another |
167 | the filehandle is destroyed. Normally, you can safely call perls C<close> |
160 | time when the filehandle is destroyed. Normally, you can safely call perls |
168 | or just let filehandles go out of scope. |
161 | C<close> or just let filehandles go out of scope. |
169 | |
162 | |
|
|
163 | This is supposed to be a bug in the API, so that might change. It's |
|
|
164 | therefore best to avoid this function. |
|
|
165 | |
170 | =item aio_read $fh,$offset,$length, $data,$dataoffset,$callback |
166 | =item aio_read $fh,$offset,$length, $data,$dataoffset, $callback->($retval) |
171 | |
167 | |
172 | =item aio_write $fh,$offset,$length, $data,$dataoffset,$callback |
168 | =item aio_write $fh,$offset,$length, $data,$dataoffset, $callback->($retval) |
173 | |
169 | |
174 | Reads or writes C<length> bytes from the specified C<fh> and C<offset> |
170 | Reads or writes C<length> bytes from the specified C<fh> and C<offset> |
175 | into the scalar given by C<data> and offset C<dataoffset> and calls the |
171 | into the scalar given by C<data> and offset C<dataoffset> and calls the |
176 | callback without the actual number of bytes read (or -1 on error, just |
172 | callback without the actual number of bytes read (or -1 on error, just |
177 | like the syscall). |
173 | like the syscall). |
178 | |
174 | |
|
|
175 | The C<$data> scalar I<MUST NOT> be modified in any way while the request |
|
|
176 | is outstanding. Modifying it can result in segfaults or WW3 (if the |
|
|
177 | necessary/optional hardware is installed). |
|
|
178 | |
179 | Example: Read 15 bytes at offset 7 into scalar C<$buffer>, strating at |
179 | Example: Read 15 bytes at offset 7 into scalar C<$buffer>, starting at |
180 | offset C<0> within the scalar: |
180 | offset C<0> within the scalar: |
181 | |
181 | |
182 | aio_read $fh, 7, 15, $buffer, 0, sub { |
182 | aio_read $fh, 7, 15, $buffer, 0, sub { |
183 | $_[0] >= 0 or die "read error: $!"; |
183 | $_[0] > 0 or die "read error: $!"; |
184 | print "read <$buffer>\n"; |
184 | print "read $_[0] bytes: <$buffer>\n"; |
185 | }; |
185 | }; |
186 | |
186 | |
|
|
187 | =item aio_move $srcpath, $dstpath, $callback->($status) |
|
|
188 | |
|
|
189 | [EXPERIMENTAL due to internal aio_group use] |
|
|
190 | |
|
|
191 | Try to move the I<file> (directories not supported as either source or |
|
|
192 | destination) from C<$srcpath> to C<$dstpath> and call the callback with |
|
|
193 | the C<0> (error) or C<-1> ok. |
|
|
194 | |
|
|
195 | This is a composite request that tries to rename(2) the file first. If |
|
|
196 | rename files with C<EXDEV>, it creates the destination file with mode 0200 |
|
|
197 | and copies the contents of the source file into it using C<aio_sendfile>, |
|
|
198 | followed by restoring atime, mtime, access mode and uid/gid, in that |
|
|
199 | order, and unlinking the C<$srcpath>. |
|
|
200 | |
|
|
201 | If an error occurs, the partial destination file will be unlinked, if |
|
|
202 | possible, except when setting atime, mtime, access mode and uid/gid, where |
|
|
203 | errors are being ignored. |
|
|
204 | |
|
|
205 | =cut |
|
|
206 | |
|
|
207 | sub aio_move($$$) { |
|
|
208 | my ($src, $dst, $cb) = @_; |
|
|
209 | |
|
|
210 | my $grp = aio_group $cb; |
|
|
211 | |
|
|
212 | add $grp aio_rename $src, $dst, sub { |
|
|
213 | if ($_[0] && $! == EXDEV) { |
|
|
214 | add $grp aio_open $src, O_RDONLY, 0, sub { |
|
|
215 | if (my $src_fh = $_[0]) { |
|
|
216 | my @stat = stat $src_fh; |
|
|
217 | |
|
|
218 | add $grp aio_open $dst, O_WRONLY, 0200, sub { |
|
|
219 | if (my $dst_fh = $_[0]) { |
|
|
220 | add $grp aio_sendfile $dst_fh, $src_fh, 0, $stat[7], sub { |
|
|
221 | close $src_fh; |
|
|
222 | |
|
|
223 | if ($_[0] == $stat[7]) { |
|
|
224 | utime $stat[8], $stat[9], $dst; |
|
|
225 | chmod $stat[2] & 07777, $dst_fh; |
|
|
226 | chown $stat[4], $stat[5], $dst_fh; |
|
|
227 | close $dst_fh; |
|
|
228 | |
|
|
229 | add $grp aio_unlink $src, sub { |
|
|
230 | $grp->result ($_[0]); |
|
|
231 | }; |
|
|
232 | } else { |
|
|
233 | my $errno = $!; |
|
|
234 | add $grp aio_unlink $dst, sub { |
|
|
235 | $! = $errno; |
|
|
236 | $grp->result (-1); |
|
|
237 | }; |
|
|
238 | } |
|
|
239 | }; |
|
|
240 | } else { |
|
|
241 | $grp->result (-1); |
|
|
242 | } |
|
|
243 | }, |
|
|
244 | |
|
|
245 | } else { |
|
|
246 | $grp->result (-1); |
|
|
247 | } |
|
|
248 | }; |
|
|
249 | } else { |
|
|
250 | $grp->result ($_[0]); |
|
|
251 | } |
|
|
252 | }; |
|
|
253 | |
|
|
254 | $grp |
|
|
255 | } |
|
|
256 | |
|
|
257 | =item aio_sendfile $out_fh, $in_fh, $in_offset, $length, $callback->($retval) |
|
|
258 | |
|
|
259 | Tries to copy C<$length> bytes from C<$in_fh> to C<$out_fh>. It starts |
|
|
260 | reading at byte offset C<$in_offset>, and starts writing at the current |
|
|
261 | file offset of C<$out_fh>. Because of that, it is not safe to issue more |
|
|
262 | than one C<aio_sendfile> per C<$out_fh>, as they will interfere with each |
|
|
263 | other. |
|
|
264 | |
|
|
265 | This call tries to make use of a native C<sendfile> syscall to provide |
|
|
266 | zero-copy operation. For this to work, C<$out_fh> should refer to a |
|
|
267 | socket, and C<$in_fh> should refer to mmap'able file. |
|
|
268 | |
|
|
269 | If the native sendfile call fails or is not implemented, it will be |
|
|
270 | emulated, so you can call C<aio_sendfile> on any type of filehandle |
|
|
271 | regardless of the limitations of the operating system. |
|
|
272 | |
|
|
273 | Please note, however, that C<aio_sendfile> can read more bytes from |
|
|
274 | C<$in_fh> than are written, and there is no way to find out how many |
|
|
275 | bytes have been read from C<aio_sendfile> alone, as C<aio_sendfile> only |
|
|
276 | provides the number of bytes written to C<$out_fh>. Only if the result |
|
|
277 | value equals C<$length> one can assume that C<$length> bytes have been |
|
|
278 | read. |
|
|
279 | |
187 | =item aio_readahead $fh,$offset,$length, $callback |
280 | =item aio_readahead $fh,$offset,$length, $callback->($retval) |
188 | |
281 | |
189 | Asynchronously reads the specified byte range into the page cache, using |
|
|
190 | the C<readahead> syscall. If that syscall doesn't exist the status will be |
|
|
191 | C<-1> and C<$!> is set to ENOSYS. |
|
|
192 | |
|
|
193 | readahead() populates the page cache with data from a file so that |
282 | C<aio_readahead> populates the page cache with data from a file so that |
194 | subsequent reads from that file will not block on disk I/O. The C<$offset> |
283 | subsequent reads from that file will not block on disk I/O. The C<$offset> |
195 | argument specifies the starting point from which data is to be read and |
284 | argument specifies the starting point from which data is to be read and |
196 | C<$length> specifies the number of bytes to be read. I/O is performed in |
285 | C<$length> specifies the number of bytes to be read. I/O is performed in |
197 | whole pages, so that offset is effectively rounded down to a page boundary |
286 | whole pages, so that offset is effectively rounded down to a page boundary |
198 | and bytes are read up to the next page boundary greater than or equal to |
287 | and bytes are read up to the next page boundary greater than or equal to |
199 | (off-set+length). aio_readahead() does not read beyond the end of the |
288 | (off-set+length). C<aio_readahead> does not read beyond the end of the |
200 | file. The current file offset of the file is left unchanged. |
289 | file. The current file offset of the file is left unchanged. |
201 | |
290 | |
|
|
291 | If that syscall doesn't exist (likely if your OS isn't Linux) it will be |
|
|
292 | emulated by simply reading the data, which would have a similar effect. |
|
|
293 | |
202 | =item aio_stat $fh_or_path, $callback |
294 | =item aio_stat $fh_or_path, $callback->($status) |
203 | |
295 | |
204 | =item aio_lstat $fh, $callback |
296 | =item aio_lstat $fh, $callback->($status) |
205 | |
297 | |
206 | Works like perl's C<stat> or C<lstat> in void context. The callback will |
298 | Works like perl's C<stat> or C<lstat> in void context. The callback will |
207 | be called after the stat and the results will be available using C<stat _> |
299 | be called after the stat and the results will be available using C<stat _> |
208 | or C<-s _> etc... |
300 | or C<-s _> etc... |
209 | |
301 | |
… | |
… | |
219 | aio_stat "/etc/passwd", sub { |
311 | aio_stat "/etc/passwd", sub { |
220 | $_[0] and die "stat failed: $!"; |
312 | $_[0] and die "stat failed: $!"; |
221 | print "size is ", -s _, "\n"; |
313 | print "size is ", -s _, "\n"; |
222 | }; |
314 | }; |
223 | |
315 | |
224 | =item aio_unlink $pathname, $callback |
316 | =item aio_unlink $pathname, $callback->($status) |
225 | |
317 | |
226 | Asynchronously unlink (delete) a file and call the callback with the |
318 | Asynchronously unlink (delete) a file and call the callback with the |
227 | result code. |
319 | result code. |
228 | |
320 | |
|
|
321 | =item aio_link $srcpath, $dstpath, $callback->($status) |
|
|
322 | |
|
|
323 | Asynchronously create a new link to the existing object at C<$srcpath> at |
|
|
324 | the path C<$dstpath> and call the callback with the result code. |
|
|
325 | |
|
|
326 | =item aio_symlink $srcpath, $dstpath, $callback->($status) |
|
|
327 | |
|
|
328 | Asynchronously create a new symbolic link to the existing object at C<$srcpath> at |
|
|
329 | the path C<$dstpath> and call the callback with the result code. |
|
|
330 | |
|
|
331 | =item aio_rename $srcpath, $dstpath, $callback->($status) |
|
|
332 | |
|
|
333 | Asynchronously rename the object at C<$srcpath> to C<$dstpath>, just as |
|
|
334 | rename(2) and call the callback with the result code. |
|
|
335 | |
|
|
336 | =item aio_rmdir $pathname, $callback->($status) |
|
|
337 | |
|
|
338 | Asynchronously rmdir (delete) a directory and call the callback with the |
|
|
339 | result code. |
|
|
340 | |
|
|
341 | =item aio_readdir $pathname, $callback->($entries) |
|
|
342 | |
|
|
343 | Unlike the POSIX call of the same name, C<aio_readdir> reads an entire |
|
|
344 | directory (i.e. opendir + readdir + closedir). The entries will not be |
|
|
345 | sorted, and will B<NOT> include the C<.> and C<..> entries. |
|
|
346 | |
|
|
347 | The callback a single argument which is either C<undef> or an array-ref |
|
|
348 | with the filenames. |
|
|
349 | |
|
|
350 | =item aio_scandir $path, $maxreq, $callback->($dirs, $nondirs) |
|
|
351 | |
|
|
352 | [EXPERIMENTAL due to internal aio_group use] |
|
|
353 | |
|
|
354 | Scans a directory (similar to C<aio_readdir>) but additionally tries to |
|
|
355 | separate the entries of directory C<$path> into two sets of names, ones |
|
|
356 | you can recurse into (directories or links to them), and ones you cannot |
|
|
357 | recurse into (everything else). |
|
|
358 | |
|
|
359 | C<aio_scandir> is a composite request that creates of many sub requests_ |
|
|
360 | C<$maxreq> specifies the maximum number of outstanding aio requests that |
|
|
361 | this function generates. If it is C<< <= 0 >>, then a suitable default |
|
|
362 | will be chosen (currently 6). |
|
|
363 | |
|
|
364 | On error, the callback is called without arguments, otherwise it receives |
|
|
365 | two array-refs with path-relative entry names. |
|
|
366 | |
|
|
367 | Example: |
|
|
368 | |
|
|
369 | aio_scandir $dir, 0, sub { |
|
|
370 | my ($dirs, $nondirs) = @_; |
|
|
371 | print "real directories: @$dirs\n"; |
|
|
372 | print "everything else: @$nondirs\n"; |
|
|
373 | }; |
|
|
374 | |
|
|
375 | Implementation notes. |
|
|
376 | |
|
|
377 | The C<aio_readdir> cannot be avoided, but C<stat()>'ing every entry can. |
|
|
378 | |
|
|
379 | After reading the directory, the modification time, size etc. of the |
|
|
380 | directory before and after the readdir is checked, and if they match (and |
|
|
381 | isn't the current time), the link count will be used to decide how many |
|
|
382 | entries are directories (if >= 2). Otherwise, no knowledge of the number |
|
|
383 | of subdirectories will be assumed. |
|
|
384 | |
|
|
385 | Then entries will be sorted into likely directories (everything without |
|
|
386 | a non-initial dot currently) and likely non-directories (everything |
|
|
387 | else). Then every entry plus an appended C</.> will be C<stat>'ed, |
|
|
388 | likely directories first. If that succeeds, it assumes that the entry |
|
|
389 | is a directory or a symlink to directory (which will be checked |
|
|
390 | seperately). This is often faster than stat'ing the entry itself because |
|
|
391 | filesystems might detect the type of the entry without reading the inode |
|
|
392 | data (e.g. ext2fs filetype feature). |
|
|
393 | |
|
|
394 | If the known number of directories (link count - 2) has been reached, the |
|
|
395 | rest of the entries is assumed to be non-directories. |
|
|
396 | |
|
|
397 | This only works with certainty on POSIX (= UNIX) filesystems, which |
|
|
398 | fortunately are the vast majority of filesystems around. |
|
|
399 | |
|
|
400 | It will also likely work on non-POSIX filesystems with reduced efficiency |
|
|
401 | as those tend to return 0 or 1 as link counts, which disables the |
|
|
402 | directory counting heuristic. |
|
|
403 | |
|
|
404 | =cut |
|
|
405 | |
|
|
406 | sub aio_scandir($$$) { |
|
|
407 | my ($path, $maxreq, $cb) = @_; |
|
|
408 | |
|
|
409 | my $grp = aio_group $cb; |
|
|
410 | |
|
|
411 | $maxreq = 6 if $maxreq <= 0; |
|
|
412 | |
|
|
413 | # stat once |
|
|
414 | add $grp aio_stat $path, sub { |
|
|
415 | return $grp->result () if $_[0]; |
|
|
416 | my $now = time; |
|
|
417 | my $hash1 = join ":", (stat _)[0,1,3,7,9]; |
|
|
418 | |
|
|
419 | # read the directory entries |
|
|
420 | add $grp aio_readdir $path, sub { |
|
|
421 | my $entries = shift |
|
|
422 | or return $grp->result (); |
|
|
423 | |
|
|
424 | # stat the dir another time |
|
|
425 | add $grp aio_stat $path, sub { |
|
|
426 | my $hash2 = join ":", (stat _)[0,1,3,7,9]; |
|
|
427 | |
|
|
428 | my $ndirs; |
|
|
429 | |
|
|
430 | # take the slow route if anything looks fishy |
|
|
431 | if ($hash1 ne $hash2 or (stat _)[9] == $now) { |
|
|
432 | $ndirs = -1; |
|
|
433 | } else { |
|
|
434 | # if nlink == 2, we are finished |
|
|
435 | # on non-posix-fs's, we rely on nlink < 2 |
|
|
436 | $ndirs = (stat _)[3] - 2 |
|
|
437 | or return $grp->result ([], $entries); |
|
|
438 | } |
|
|
439 | |
|
|
440 | # sort into likely dirs and likely nondirs |
|
|
441 | # dirs == files without ".", short entries first |
|
|
442 | $entries = [map $_->[0], |
|
|
443 | sort { $b->[1] cmp $a->[1] } |
|
|
444 | map [$_, sprintf "%s%04d", (/.\./ ? "1" : "0"), length], |
|
|
445 | @$entries]; |
|
|
446 | |
|
|
447 | my (@dirs, @nondirs); |
|
|
448 | |
|
|
449 | my ($statcb, $schedcb); |
|
|
450 | my $nreq = 0; |
|
|
451 | |
|
|
452 | my $statgrp = add $grp aio_group; |
|
|
453 | |
|
|
454 | $schedcb = sub { |
|
|
455 | if (@$entries) { |
|
|
456 | if ($nreq < $maxreq) { |
|
|
457 | my $ent = pop @$entries; |
|
|
458 | $nreq++; |
|
|
459 | add $statgrp aio_stat "$path/$ent/.", sub { $statcb->($_[0], $ent) }; |
|
|
460 | } |
|
|
461 | } elsif (!$nreq) { |
|
|
462 | # finished |
|
|
463 | $statgrp->cancel; |
|
|
464 | undef $statcb; |
|
|
465 | undef $schedcb; |
|
|
466 | $grp->result (\@dirs, \@nondirs); |
|
|
467 | } |
|
|
468 | }; |
|
|
469 | $statcb = sub { |
|
|
470 | my ($status, $entry) = @_; |
|
|
471 | |
|
|
472 | if ($status < 0) { |
|
|
473 | $nreq--; |
|
|
474 | push @nondirs, $entry; |
|
|
475 | &$schedcb; |
|
|
476 | } else { |
|
|
477 | # need to check for real directory |
|
|
478 | add $grp aio_lstat "$path/$entry", sub { |
|
|
479 | $nreq--; |
|
|
480 | |
|
|
481 | if (-d _) { |
|
|
482 | push @dirs, $entry; |
|
|
483 | |
|
|
484 | if (!--$ndirs) { |
|
|
485 | push @nondirs, @$entries; |
|
|
486 | $entries = []; |
|
|
487 | } |
|
|
488 | } else { |
|
|
489 | push @nondirs, $entry; |
|
|
490 | } |
|
|
491 | |
|
|
492 | &$schedcb; |
|
|
493 | } |
|
|
494 | } |
|
|
495 | }; |
|
|
496 | |
|
|
497 | &$schedcb while @$entries && $nreq < $maxreq; |
|
|
498 | }; |
|
|
499 | }; |
|
|
500 | }; |
|
|
501 | |
|
|
502 | $grp |
|
|
503 | } |
|
|
504 | |
229 | =item aio_fsync $fh, $callback |
505 | =item aio_fsync $fh, $callback->($status) |
230 | |
506 | |
231 | Asynchronously call fsync on the given filehandle and call the callback |
507 | Asynchronously call fsync on the given filehandle and call the callback |
232 | with the fsync result code. |
508 | with the fsync result code. |
233 | |
509 | |
234 | =item aio_fdatasync $fh, $callback |
510 | =item aio_fdatasync $fh, $callback->($status) |
235 | |
511 | |
236 | Asynchronously call fdatasync on the given filehandle and call the |
512 | Asynchronously call fdatasync on the given filehandle and call the |
237 | callback with the fdatasync result code. |
513 | callback with the fdatasync result code. |
|
|
514 | |
|
|
515 | If this call isn't available because your OS lacks it or it couldn't be |
|
|
516 | detected, it will be emulated by calling C<fsync> instead. |
|
|
517 | |
|
|
518 | =item aio_group $callback->(...) |
|
|
519 | |
|
|
520 | [EXPERIMENTAL] |
|
|
521 | |
|
|
522 | This is a very special aio request: Instead of doing something, it is a |
|
|
523 | container for other aio requests, which is useful if you want to bundle |
|
|
524 | many requests into a single, composite, request. |
|
|
525 | |
|
|
526 | Returns an object of class L<IO::AIO::GRP>. See its documentation below |
|
|
527 | for more info. |
|
|
528 | |
|
|
529 | Example: |
|
|
530 | |
|
|
531 | my $grp = aio_group sub { |
|
|
532 | print "all stats done\n"; |
|
|
533 | }; |
|
|
534 | |
|
|
535 | add $grp |
|
|
536 | (aio_stat ...), |
|
|
537 | (aio_stat ...), |
|
|
538 | ...; |
|
|
539 | |
|
|
540 | =item aio_nop $callback->() |
|
|
541 | |
|
|
542 | This is a special request - it does nothing in itself and is only used for |
|
|
543 | side effects, such as when you want to add a dummy request to a group so |
|
|
544 | that finishing the requests in the group depends on executing the given |
|
|
545 | code. |
|
|
546 | |
|
|
547 | While this request does nothing, it still goes through the execution |
|
|
548 | phase and still requires a worker thread. Thus, the callback will not |
|
|
549 | be executed immediately but only after other requests in the queue have |
|
|
550 | entered their execution phase. This can be used to measure request |
|
|
551 | latency. |
|
|
552 | |
|
|
553 | =item IO::AIO::aio_sleep $fractional_seconds, $callback->() *NOT EXPORTED* |
|
|
554 | |
|
|
555 | Mainly used for debugging and benchmarking, this aio request puts one of |
|
|
556 | the request workers to sleep for the given time. |
|
|
557 | |
|
|
558 | While it is theoretically handy to have simple I/O scheduling requests |
|
|
559 | like sleep and file handle readable/writable, the overhead this creates |
|
|
560 | is immense, so do not use this function except to put your application |
|
|
561 | under artificial I/O pressure. |
|
|
562 | |
|
|
563 | =back |
|
|
564 | |
|
|
565 | =head2 IO::AIO::REQ CLASS |
|
|
566 | |
|
|
567 | All non-aggregate C<aio_*> functions return an object of this class when |
|
|
568 | called in non-void context. |
|
|
569 | |
|
|
570 | A request always moves through the following five states in its lifetime, |
|
|
571 | in order: B<ready> (request has been created, but has not been executed |
|
|
572 | yet), B<execute> (request is currently being executed), B<pending> |
|
|
573 | (request has been executed but callback has not been called yet), |
|
|
574 | B<result> (results are being processed synchronously, includes calling the |
|
|
575 | callback) and B<done> (request has reached the end of its lifetime and |
|
|
576 | holds no resources anymore). |
|
|
577 | |
|
|
578 | =over 4 |
|
|
579 | |
|
|
580 | =item cancel $req |
|
|
581 | |
|
|
582 | Cancels the request, if possible. Has the effect of skipping execution |
|
|
583 | when entering the B<execute> state and skipping calling the callback when |
|
|
584 | entering the the B<result> state, but will leave the request otherwise |
|
|
585 | untouched. That means that requests that currently execute will not be |
|
|
586 | stopped and resources held by the request will not be freed prematurely. |
|
|
587 | |
|
|
588 | =item cb $req $callback->(...) |
|
|
589 | |
|
|
590 | Replace (or simply set) the callback registered to the request. |
|
|
591 | |
|
|
592 | =back |
|
|
593 | |
|
|
594 | =head2 IO::AIO::GRP CLASS |
|
|
595 | |
|
|
596 | This class is a subclass of L<IO::AIO::REQ>, so all its methods apply to |
|
|
597 | objects of this class, too. |
|
|
598 | |
|
|
599 | A IO::AIO::GRP object is a special request that can contain multiple other |
|
|
600 | aio requests. |
|
|
601 | |
|
|
602 | You create one by calling the C<aio_group> constructing function with a |
|
|
603 | callback that will be called when all contained requests have entered the |
|
|
604 | C<done> state: |
|
|
605 | |
|
|
606 | my $grp = aio_group sub { |
|
|
607 | print "all requests are done\n"; |
|
|
608 | }; |
|
|
609 | |
|
|
610 | You add requests by calling the C<add> method with one or more |
|
|
611 | C<IO::AIO::REQ> objects: |
|
|
612 | |
|
|
613 | $grp->add (aio_unlink "..."); |
|
|
614 | |
|
|
615 | add $grp aio_stat "...", sub { |
|
|
616 | $_[0] or return $grp->result ("error"); |
|
|
617 | |
|
|
618 | # add another request dynamically, if first succeeded |
|
|
619 | add $grp aio_open "...", sub { |
|
|
620 | $grp->result ("ok"); |
|
|
621 | }; |
|
|
622 | }; |
|
|
623 | |
|
|
624 | This makes it very easy to create composite requests (see the source of |
|
|
625 | C<aio_move> for an application) that work and feel like simple requests. |
|
|
626 | |
|
|
627 | =over 4 |
|
|
628 | |
|
|
629 | =item * The IO::AIO::GRP objects will be cleaned up during calls to |
|
|
630 | C<IO::AIO::poll_cb>, just like any other request. |
|
|
631 | |
|
|
632 | =item * They can be canceled like any other request. Canceling will cancel not |
|
|
633 | only the request itself, but also all requests it contains. |
|
|
634 | |
|
|
635 | =item * They can also can also be added to other IO::AIO::GRP objects. |
|
|
636 | |
|
|
637 | =item * You must not add requests to a group from within the group callback (or |
|
|
638 | any later time). |
|
|
639 | |
|
|
640 | =item * This does not harmonise well with C<max_outstanding>, so best do |
|
|
641 | not combine C<aio_group> with it. Groups and feeders are recommended for |
|
|
642 | this kind of concurrency-limiting. |
|
|
643 | |
|
|
644 | =back |
|
|
645 | |
|
|
646 | Their lifetime, simplified, looks like this: when they are empty, they |
|
|
647 | will finish very quickly. If they contain only requests that are in the |
|
|
648 | C<done> state, they will also finish. Otherwise they will continue to |
|
|
649 | exist. |
|
|
650 | |
|
|
651 | That means after creating a group you have some time to add requests. And |
|
|
652 | in the callbacks of those requests, you can add further requests to the |
|
|
653 | group. And only when all those requests have finished will the the group |
|
|
654 | itself finish. |
|
|
655 | |
|
|
656 | =over 4 |
|
|
657 | |
|
|
658 | =item add $grp ... |
|
|
659 | |
|
|
660 | =item $grp->add (...) |
|
|
661 | |
|
|
662 | Add one or more requests to the group. Any type of L<IO::AIO::REQ> can |
|
|
663 | be added, including other groups, as long as you do not create circular |
|
|
664 | dependencies. |
|
|
665 | |
|
|
666 | Returns all its arguments. |
|
|
667 | |
|
|
668 | =item $grp->result (...) |
|
|
669 | |
|
|
670 | Set the result value(s) that will be passed to the group callback when all |
|
|
671 | subrequests have finished. By default, no argument will be passed. |
|
|
672 | |
|
|
673 | =item feed $grp $callback->($grp) |
|
|
674 | |
|
|
675 | [VERY EXPERIMENTAL] |
|
|
676 | |
|
|
677 | Sets a feeder/generator on this group: every group can have an attached |
|
|
678 | generator that generates requests if idle. The idea behind this is that, |
|
|
679 | although you could just queue as many requests as you want in a group, |
|
|
680 | this might starve other requests for a potentially long time. For |
|
|
681 | example, C<aio_scandir> might generate hundreds of thousands C<aio_stat> |
|
|
682 | requests, delaying any later requests for a long time. |
|
|
683 | |
|
|
684 | To avoid this, and allow incremental generation of requests, you can |
|
|
685 | instead a group and set a feeder on it that generates those requests. The |
|
|
686 | feed callback will be called whenever there are few enough (see C<feed_limit>, |
|
|
687 | below) requests active in the group itself and is expected to queue more |
|
|
688 | requests. |
|
|
689 | |
|
|
690 | The feed can queue as many requests as it likes (i.e. C<add> does not |
|
|
691 | impose any limits). |
|
|
692 | |
|
|
693 | If the feed does not queue more requests when called, it will be |
|
|
694 | automatically removed from the group. |
|
|
695 | |
|
|
696 | If the feed limit is C<0>, it will be set to C<2> automatically. |
|
|
697 | |
|
|
698 | Example: |
|
|
699 | |
|
|
700 | # stat all files in @files, but only ever use four aio requests concurrently: |
|
|
701 | |
|
|
702 | my $grp = aio_group sub { print "finished\n" }; |
|
|
703 | feed_limit $grp 4; |
|
|
704 | feed $grp sub { |
|
|
705 | my $file = pop @files |
|
|
706 | or return; |
|
|
707 | |
|
|
708 | add $grp aio_stat $file, sub { ... }; |
|
|
709 | }; |
|
|
710 | |
|
|
711 | =item feed_limit $grp $num |
|
|
712 | |
|
|
713 | Sets the feeder limit for the group: The feeder will be called whenever |
|
|
714 | the group contains less than this many requests. |
|
|
715 | |
|
|
716 | Setting the limit to C<0> will pause the feeding process. |
|
|
717 | |
|
|
718 | =back |
|
|
719 | |
|
|
720 | =head2 SUPPORT FUNCTIONS |
|
|
721 | |
|
|
722 | =over 4 |
|
|
723 | |
|
|
724 | =item $fileno = IO::AIO::poll_fileno |
|
|
725 | |
|
|
726 | Return the I<request result pipe file descriptor>. This filehandle must be |
|
|
727 | polled for reading by some mechanism outside this module (e.g. Event or |
|
|
728 | select, see below or the SYNOPSIS). If the pipe becomes readable you have |
|
|
729 | to call C<poll_cb> to check the results. |
|
|
730 | |
|
|
731 | See C<poll_cb> for an example. |
|
|
732 | |
|
|
733 | =item IO::AIO::poll_cb |
|
|
734 | |
|
|
735 | Process all outstanding events on the result pipe. You have to call this |
|
|
736 | regularly. Returns the number of events processed. Returns immediately |
|
|
737 | when no events are outstanding. |
|
|
738 | |
|
|
739 | Example: Install an Event watcher that automatically calls |
|
|
740 | IO::AIO::poll_cb with high priority: |
|
|
741 | |
|
|
742 | Event->io (fd => IO::AIO::poll_fileno, |
|
|
743 | poll => 'r', async => 1, |
|
|
744 | cb => \&IO::AIO::poll_cb); |
|
|
745 | |
|
|
746 | =item IO::AIO::poll_wait |
|
|
747 | |
|
|
748 | Wait till the result filehandle becomes ready for reading (simply does a |
|
|
749 | C<select> on the filehandle. This is useful if you want to synchronously wait |
|
|
750 | for some requests to finish). |
|
|
751 | |
|
|
752 | See C<nreqs> for an example. |
|
|
753 | |
|
|
754 | =item IO::AIO::nreqs |
|
|
755 | |
|
|
756 | Returns the number of requests currently outstanding (i.e. for which their |
|
|
757 | callback has not been invoked yet). |
|
|
758 | |
|
|
759 | Example: wait till there are no outstanding requests anymore: |
|
|
760 | |
|
|
761 | IO::AIO::poll_wait, IO::AIO::poll_cb |
|
|
762 | while IO::AIO::nreqs; |
|
|
763 | |
|
|
764 | =item IO::AIO::flush |
|
|
765 | |
|
|
766 | Wait till all outstanding AIO requests have been handled. |
|
|
767 | |
|
|
768 | Strictly equivalent to: |
|
|
769 | |
|
|
770 | IO::AIO::poll_wait, IO::AIO::poll_cb |
|
|
771 | while IO::AIO::nreqs; |
|
|
772 | |
|
|
773 | =item IO::AIO::poll |
|
|
774 | |
|
|
775 | Waits until some requests have been handled. |
|
|
776 | |
|
|
777 | Strictly equivalent to: |
|
|
778 | |
|
|
779 | IO::AIO::poll_wait, IO::AIO::poll_cb |
|
|
780 | if IO::AIO::nreqs; |
|
|
781 | |
|
|
782 | =item IO::AIO::min_parallel $nthreads |
|
|
783 | |
|
|
784 | Set the minimum number of AIO threads to C<$nthreads>. The current |
|
|
785 | default is C<8>, which means eight asynchronous operations can execute |
|
|
786 | concurrently at any one time (the number of outstanding requests, |
|
|
787 | however, is unlimited). |
|
|
788 | |
|
|
789 | IO::AIO starts threads only on demand, when an AIO request is queued and |
|
|
790 | no free thread exists. |
|
|
791 | |
|
|
792 | It is recommended to keep the number of threads relatively low, as some |
|
|
793 | Linux kernel versions will scale negatively with the number of threads |
|
|
794 | (higher parallelity => MUCH higher latency). With current Linux 2.6 |
|
|
795 | versions, 4-32 threads should be fine. |
|
|
796 | |
|
|
797 | Under most circumstances you don't need to call this function, as the |
|
|
798 | module selects a default that is suitable for low to moderate load. |
|
|
799 | |
|
|
800 | =item IO::AIO::max_parallel $nthreads |
|
|
801 | |
|
|
802 | Sets the maximum number of AIO threads to C<$nthreads>. If more than the |
|
|
803 | specified number of threads are currently running, this function kills |
|
|
804 | them. This function blocks until the limit is reached. |
|
|
805 | |
|
|
806 | While C<$nthreads> are zero, aio requests get queued but not executed |
|
|
807 | until the number of threads has been increased again. |
|
|
808 | |
|
|
809 | This module automatically runs C<max_parallel 0> at program end, to ensure |
|
|
810 | that all threads are killed and that there are no outstanding requests. |
|
|
811 | |
|
|
812 | Under normal circumstances you don't need to call this function. |
|
|
813 | |
|
|
814 | =item $oldnreqs = IO::AIO::max_outstanding $nreqs |
|
|
815 | |
|
|
816 | [DEPRECATED] |
|
|
817 | |
|
|
818 | Sets the maximum number of outstanding requests to C<$nreqs>. If you |
|
|
819 | try to queue up more than this number of requests, the caller will block until |
|
|
820 | some requests have been handled. |
|
|
821 | |
|
|
822 | The default is very large, so normally there is no practical limit. If you |
|
|
823 | queue up many requests in a loop it often improves speed if you set |
|
|
824 | this to a relatively low number, such as C<100>. |
|
|
825 | |
|
|
826 | This function does not work well together with C<aio_group>'s, and their |
|
|
827 | feeder interface is better suited to limiting concurrency, so do not use |
|
|
828 | this function. |
|
|
829 | |
|
|
830 | Under normal circumstances you don't need to call this function. |
|
|
831 | |
|
|
832 | =back |
238 | |
833 | |
239 | =cut |
834 | =cut |
240 | |
835 | |
241 | # support function to convert a fd into a perl filehandle |
836 | # support function to convert a fd into a perl filehandle |
242 | sub _fd2fh { |
837 | sub _fd2fh { |
243 | return undef if $_[0] < 0; |
838 | return undef if $_[0] < 0; |
244 | |
839 | |
245 | # try to be perl5.6-compatible |
840 | # try to generate nice filehandles |
246 | local *AIO_FH; |
841 | my $sym = "IO::AIO::fd#$_[0]"; |
247 | open AIO_FH, "+<&=$_[0]" |
842 | local *$sym; |
|
|
843 | |
|
|
844 | open *$sym, "+<&=$_[0]" # usually works under any unix |
|
|
845 | or open *$sym, "<&=$_[0]" # cygwin needs this |
|
|
846 | or open *$sym, ">&=$_[0]" # or this |
248 | or return undef; |
847 | or return undef; |
249 | |
848 | |
250 | *AIO_FH |
849 | *$sym |
251 | } |
850 | } |
252 | |
851 | |
253 | min_parallel 4; |
852 | min_parallel 8; |
254 | |
853 | |
255 | END { |
854 | END { |
256 | max_parallel 0; |
855 | max_parallel 0; |
257 | } |
856 | } |
258 | |
857 | |
259 | 1; |
858 | 1; |
260 | |
859 | |
261 | =back |
860 | =head2 FORK BEHAVIOUR |
262 | |
861 | |
263 | =head1 BUGS |
862 | This module should do "the right thing" when the process using it forks: |
264 | |
863 | |
265 | - could be optimized to use more semaphores instead of filehandles. |
864 | Before the fork, IO::AIO enters a quiescent state where no requests |
|
|
865 | can be added in other threads and no results will be processed. After |
|
|
866 | the fork the parent simply leaves the quiescent state and continues |
|
|
867 | request/result processing, while the child clears the request/result |
|
|
868 | queue (so the requests started before the fork will only be handled in |
|
|
869 | the parent). Threads will be started on demand until the limit ste in the |
|
|
870 | parent process has been reached again. |
|
|
871 | |
|
|
872 | In short: the parent will, after a short pause, continue as if fork had |
|
|
873 | not been called, while the child will act as if IO::AIO has not been used |
|
|
874 | yet. |
|
|
875 | |
|
|
876 | =head2 MEMORY USAGE |
|
|
877 | |
|
|
878 | Each aio request uses - depending on your architecture - around 128 bytes |
|
|
879 | of memory. In addition, stat requests need a stat buffer (possibly a few |
|
|
880 | hundred bytes). Perl scalars and other data passed into aio requests will |
|
|
881 | also be locked. |
|
|
882 | |
|
|
883 | This is now awfully much, so queuing lots of requests is not usually a |
|
|
884 | problem. |
|
|
885 | |
|
|
886 | Each thread needs a stack area which is usually around 16k, sometimes much |
|
|
887 | larger, depending on the OS. |
266 | |
888 | |
267 | =head1 SEE ALSO |
889 | =head1 SEE ALSO |
268 | |
890 | |
269 | L<Coro>, L<Linux::AIO>. |
891 | L<Coro>, L<Linux::AIO> (obsolete). |
270 | |
892 | |
271 | =head1 AUTHOR |
893 | =head1 AUTHOR |
272 | |
894 | |
273 | Marc Lehmann <schmorp@schmorp.de> |
895 | Marc Lehmann <schmorp@schmorp.de> |
274 | http://home.schmorp.de/ |
896 | http://home.schmorp.de/ |