… | |
… | |
2 | IO::AIO - Asynchronous Input/Output |
2 | IO::AIO - Asynchronous Input/Output |
3 | |
3 | |
4 | SYNOPSIS |
4 | SYNOPSIS |
5 | use IO::AIO; |
5 | use IO::AIO; |
6 | |
6 | |
|
|
7 | aio_open "/etc/passwd", O_RDONLY, 0, sub { |
|
|
8 | my ($fh) = @_; |
|
|
9 | ... |
|
|
10 | }; |
|
|
11 | |
|
|
12 | aio_unlink "/tmp/file", sub { }; |
|
|
13 | |
|
|
14 | aio_read $fh, 30000, 1024, $buffer, 0, sub { |
|
|
15 | $_[0] > 0 or die "read error: $!"; |
|
|
16 | }; |
|
|
17 | |
|
|
18 | # Event |
|
|
19 | Event->io (fd => IO::AIO::poll_fileno, |
|
|
20 | poll => 'r', |
|
|
21 | cb => \&IO::AIO::poll_cb); |
|
|
22 | |
|
|
23 | # Glib/Gtk2 |
|
|
24 | add_watch Glib::IO IO::AIO::poll_fileno, |
|
|
25 | in => sub { IO::AIO::poll_cb; 1 }; |
|
|
26 | |
|
|
27 | # Tk |
|
|
28 | Tk::Event::IO->fileevent (IO::AIO::poll_fileno, "", |
|
|
29 | readable => \&IO::AIO::poll_cb); |
|
|
30 | |
|
|
31 | # Danga::Socket |
|
|
32 | Danga::Socket->AddOtherFds (IO::AIO::poll_fileno => |
|
|
33 | \&IO::AIO::poll_cb); |
|
|
34 | |
7 | DESCRIPTION |
35 | DESCRIPTION |
8 | This module implements asynchronous I/O using whatever means your |
36 | This module implements asynchronous I/O using whatever means your |
9 | operating system supports. Currently, it falls back to Linux::AIO if |
37 | operating system supports. |
10 | that module is available, or uses pthreads to emulato aio functionality. |
|
|
11 | |
38 | |
12 | Currently, in this module a number of threads are started that execute |
39 | Currently, a number of threads are started that execute your read/writes |
13 | your read/writes and signal their completion. You don't need thread |
40 | and signal their completion. You don't need thread support in your libc |
14 | support in your libc or perl, and the threads created by this module |
41 | or perl, and the threads created by this module will not be visible to |
15 | will not be visible to the pthreads library. |
42 | the pthreads library. In the future, this module might make use of the |
|
|
43 | native aio functions available on many operating systems. However, they |
|
|
44 | are often not well-supported (Linux doesn't allow them on normal files |
|
|
45 | currently, for example), and they would only support aio_read and |
|
|
46 | aio_write, so the remaining functionality would have to be implemented |
|
|
47 | using threads anyway. |
16 | |
48 | |
17 | Although the module will work with in the presence of other threads, it |
49 | Although the module will work with in the presence of other threads, it |
18 | is not reentrant, so use appropriate locking yourself. |
50 | is currently not reentrant, so use appropriate locking yourself, always |
|
|
51 | call "poll_cb" from within the same thread, or never call "poll_cb" (or |
|
|
52 | other "aio_" functions) recursively. |
19 | |
53 | |
20 | API NOTES |
54 | FUNCTIONS |
|
|
55 | AIO FUNCTIONS |
21 | All the "aio_*" calls are more or less thin wrappers around the syscall |
56 | All the "aio_*" calls are more or less thin wrappers around the syscall |
22 | with the same name (sans "aio_"). The arguments are similar or |
57 | with the same name (sans "aio_"). The arguments are similar or |
23 | identical, and they all accept an additional $callback argument which |
58 | identical, and they all accept an additional (and optional) $callback |
24 | must be a code reference. This code reference will get called with the |
59 | argument which must be a code reference. This code reference will get |
25 | syscall return code (e.g. most syscalls return -1 on error, unlike perl, |
60 | called with the syscall return code (e.g. most syscalls return -1 on |
26 | which usually delivers "false") as it's sole argument when the given |
61 | error, unlike perl, which usually delivers "false") as it's sole |
27 | syscall has been executed asynchronously. |
62 | argument when the given syscall has been executed asynchronously. |
28 | |
63 | |
29 | All functions that expect a filehandle will also accept a file |
64 | All functions that expect a filehandle will also accept a file |
30 | descriptor. |
65 | descriptor. |
31 | |
66 | |
32 | The filenames you pass to these routines *must* be absolute. The reason |
67 | The filenames you pass to these routines *must* be absolute. The reason |
33 | is that at the time the request is being executed, the current working |
68 | for this is that at the time the request is being executed, the current |
34 | directory could have changed. Alternatively, you can make sure that you |
69 | working directory could have changed. Alternatively, you can make sure |
35 | never change the current working directory. |
70 | that you never change the current working directory. |
36 | |
|
|
37 | IO::AIO::min_parallel $nthreads |
|
|
38 | Set the minimum number of AIO threads to $nthreads. The default is |
|
|
39 | 1, which means a single asynchronous operation can be done at one |
|
|
40 | time (the number of outstanding operations, however, is unlimited). |
|
|
41 | |
|
|
42 | It is recommended to keep the number of threads low, as some linux |
|
|
43 | kernel versions will scale negatively with the number of threads |
|
|
44 | (higher parallelity => MUCH higher latency). |
|
|
45 | |
|
|
46 | Under normal circumstances you don't need to call this function, as |
|
|
47 | this module automatically starts a single async thread. |
|
|
48 | |
|
|
49 | IO::AIO::max_parallel $nthreads |
|
|
50 | Sets the maximum number of AIO threads to $nthreads. If more than |
|
|
51 | the specified number of threads are currently running, kill them. |
|
|
52 | This function blocks until the limit is reached. |
|
|
53 | |
|
|
54 | This module automatically runs "max_parallel 0" at program end, to |
|
|
55 | ensure that all threads are killed and that there are no outstanding |
|
|
56 | requests. |
|
|
57 | |
|
|
58 | Under normal circumstances you don't need to call this function. |
|
|
59 | |
|
|
60 | $fileno = IO::AIO::poll_fileno |
|
|
61 | Return the *request result pipe filehandle*. This filehandle must be |
|
|
62 | polled for reading by some mechanism outside this module (e.g. Event |
|
|
63 | or select, see below). If the pipe becomes readable you have to call |
|
|
64 | "poll_cb" to check the results. |
|
|
65 | |
|
|
66 | See "poll_cb" for an example. |
|
|
67 | |
|
|
68 | IO::AIO::poll_cb |
|
|
69 | Process all outstanding events on the result pipe. You have to call |
|
|
70 | this regularly. Returns the number of events processed. Returns |
|
|
71 | immediately when no events are outstanding. |
|
|
72 | |
|
|
73 | You can use Event to multiplex, e.g.: |
|
|
74 | |
|
|
75 | Event->io (fd => IO::AIO::poll_fileno, |
|
|
76 | poll => 'r', async => 1, |
|
|
77 | cb => \&IO::AIO::poll_cb); |
|
|
78 | |
|
|
79 | IO::AIO::poll_wait |
|
|
80 | Wait till the result filehandle becomes ready for reading (simply |
|
|
81 | does a select on the filehandle. This is useful if you want to |
|
|
82 | synchronously wait for some requests to finish). |
|
|
83 | |
|
|
84 | See "nreqs" for an example. |
|
|
85 | |
|
|
86 | IO::AIO::nreqs |
|
|
87 | Returns the number of requests currently outstanding. |
|
|
88 | |
|
|
89 | Example: wait till there are no outstanding requests anymore: |
|
|
90 | |
|
|
91 | IO::AIO::poll_wait, IO::AIO::poll_cb |
|
|
92 | while IO::AIO::nreqs; |
|
|
93 | |
71 | |
94 | aio_open $pathname, $flags, $mode, $callback |
72 | aio_open $pathname, $flags, $mode, $callback |
95 | Asynchronously open or create a file and call the callback with the |
73 | Asynchronously open or create a file and call the callback with a |
96 | filedescriptor (NOT a perl filehandle, sorry for that, but watch |
74 | newly created filehandle for the file. |
97 | out, this might change in the future). |
|
|
98 | |
75 | |
|
|
76 | The pathname passed to "aio_open" must be absolute. See API NOTES, |
|
|
77 | above, for an explanation. |
|
|
78 | |
99 | The $mode argument is a bitmask. See the "Fcntl" module for a list. |
79 | The $flags argument is a bitmask. See the "Fcntl" module for a list. |
100 | They are the same as used in "sysopen". |
80 | They are the same as used by "sysopen". |
|
|
81 | |
|
|
82 | Likewise, $mode specifies the mode of the newly created file, if it |
|
|
83 | didn't exist and "O_CREAT" has been given, just like perl's |
|
|
84 | "sysopen", except that it is mandatory (i.e. use 0 if you don't |
|
|
85 | create new files, and 0666 or 0777 if you do). |
101 | |
86 | |
102 | Example: |
87 | Example: |
103 | |
88 | |
104 | aio_open "/etc/passwd", O_RDONLY, 0, sub { |
89 | aio_open "/etc/passwd", O_RDONLY, 0, sub { |
105 | if ($_[0] >= 0) { |
90 | if ($_[0]) { |
106 | open my $fh, "<&$_[0]"; # create a copy for perl |
|
|
107 | aio_close $_[0], sub { }; # close the aio handle |
|
|
108 | print "open successful, fh is $fh\n"; |
91 | print "open successful, fh is $_[0]\n"; |
109 | ... |
92 | ... |
110 | } else { |
93 | } else { |
111 | die "open failed: $!\n"; |
94 | die "open failed: $!\n"; |
112 | } |
95 | } |
113 | }; |
96 | }; |
114 | |
97 | |
115 | aio_close $fh, $callback |
98 | aio_close $fh, $callback |
116 | Asynchronously close a file and call the callback with the result |
99 | Asynchronously close a file and call the callback with the result |
117 | code. |
100 | code. *WARNING:* although accepted, you should not pass in a perl |
|
|
101 | filehandle here, as perl will likely close the file descriptor |
|
|
102 | another time when the filehandle is destroyed. Normally, you can |
|
|
103 | safely call perls "close" or just let filehandles go out of scope. |
|
|
104 | |
|
|
105 | This is supposed to be a bug in the API, so that might change. It's |
|
|
106 | therefore best to avoid this function. |
118 | |
107 | |
119 | aio_read $fh,$offset,$length, $data,$dataoffset,$callback |
108 | aio_read $fh,$offset,$length, $data,$dataoffset,$callback |
120 | aio_write $fh,$offset,$length, $data,$dataoffset,$callback |
109 | aio_write $fh,$offset,$length, $data,$dataoffset,$callback |
121 | Reads or writes "length" bytes from the specified "fh" and "offset" |
110 | Reads or writes "length" bytes from the specified "fh" and "offset" |
122 | into the scalar given by "data" and offset "dataoffset" and calls |
111 | into the scalar given by "data" and offset "dataoffset" and calls |
123 | the callback without the actual number of bytes read (or -1 on |
112 | the callback without the actual number of bytes read (or -1 on |
124 | error, just like the syscall). |
113 | error, just like the syscall). |
125 | |
114 | |
126 | Example: Read 15 bytes at offset 7 into scalar $buffer, strating at |
115 | Example: Read 15 bytes at offset 7 into scalar $buffer, starting at |
127 | offset 0 within the scalar: |
116 | offset 0 within the scalar: |
128 | |
117 | |
129 | aio_read $fh, 7, 15, $buffer, 0, sub { |
118 | aio_read $fh, 7, 15, $buffer, 0, sub { |
130 | $_[0] >= 0 or die "read error: $!"; |
119 | $_[0] > 0 or die "read error: $!"; |
131 | print "read <$buffer>\n"; |
120 | print "read $_[0] bytes: <$buffer>\n"; |
132 | }; |
121 | }; |
133 | |
122 | |
134 | aio_readahead $fh,$offset,$length, $callback |
123 | aio_readahead $fh,$offset,$length, $callback |
135 | Asynchronously reads the specified byte range into the page cache, |
124 | Asynchronously reads the specified byte range into the page cache, |
136 | using the "readahead" syscall. |
125 | using the "readahead" syscall. If that syscall doesn't exist (likely |
|
|
126 | if your OS isn't Linux) the status will be -1 and $! is set to |
|
|
127 | "ENOSYS". |
137 | |
128 | |
138 | readahead() populates the page cache with data from a file so that |
129 | "aio_readahead" populates the page cache with data from a file so |
139 | subsequent reads from that file will not block on disk I/O. The |
130 | that subsequent reads from that file will not block on disk I/O. The |
140 | $offset argument specifies the starting point from which data is to |
131 | $offset argument specifies the starting point from which data is to |
141 | be read and $length specifies the number of bytes to be read. I/O is |
132 | be read and $length specifies the number of bytes to be read. I/O is |
142 | performed in whole pages, so that offset is effectively rounded down |
133 | performed in whole pages, so that offset is effectively rounded down |
143 | to a page boundary and bytes are read up to the next page boundary |
134 | to a page boundary and bytes are read up to the next page boundary |
144 | greater than or equal to (off-set+length). aio_readahead() does not |
135 | greater than or equal to (off-set+length). "aio_readahead" does not |
145 | read beyond the end of the file. The current file offset of the file |
136 | read beyond the end of the file. The current file offset of the file |
146 | is left unchanged. |
137 | is left unchanged. |
147 | |
138 | |
148 | aio_stat $fh_or_path, $callback |
139 | aio_stat $fh_or_path, $callback |
149 | aio_lstat $fh, $callback |
140 | aio_lstat $fh, $callback |
… | |
… | |
174 | Asynchronously call fsync on the given filehandle and call the |
165 | Asynchronously call fsync on the given filehandle and call the |
175 | callback with the fsync result code. |
166 | callback with the fsync result code. |
176 | |
167 | |
177 | aio_fdatasync $fh, $callback |
168 | aio_fdatasync $fh, $callback |
178 | Asynchronously call fdatasync on the given filehandle and call the |
169 | Asynchronously call fdatasync on the given filehandle and call the |
179 | callback with the fdatasync result code. |
170 | callback with the fdatasync result code. Might set $! to "ENOSYS" if |
|
|
171 | "fdatasync" is not available. |
180 | |
172 | |
181 | BUGS |
173 | SUPPORT FUNCTIONS |
182 | This module has been extensively tested in a large and very busy |
174 | $fileno = IO::AIO::poll_fileno |
183 | webserver for many years now. |
175 | Return the *request result pipe file descriptor*. This filehandle |
|
|
176 | must be polled for reading by some mechanism outside this module |
|
|
177 | (e.g. Event or select, see below or the SYNOPSIS). If the pipe |
|
|
178 | becomes readable you have to call "poll_cb" to check the results. |
184 | |
179 | |
185 | - aio_open gives a fd, but all other functions expect a perl filehandle. |
180 | See "poll_cb" for an example. |
|
|
181 | |
|
|
182 | IO::AIO::poll_cb |
|
|
183 | Process all outstanding events on the result pipe. You have to call |
|
|
184 | this regularly. Returns the number of events processed. Returns |
|
|
185 | immediately when no events are outstanding. |
|
|
186 | |
|
|
187 | Example: Install an Event watcher that automatically calls |
|
|
188 | IO::AIO::poll_cb with high priority: |
|
|
189 | |
|
|
190 | Event->io (fd => IO::AIO::poll_fileno, |
|
|
191 | poll => 'r', async => 1, |
|
|
192 | cb => \&IO::AIO::poll_cb); |
|
|
193 | |
|
|
194 | IO::AIO::poll_wait |
|
|
195 | Wait till the result filehandle becomes ready for reading (simply |
|
|
196 | does a "select" on the filehandle. This is useful if you want to |
|
|
197 | synchronously wait for some requests to finish). |
|
|
198 | |
|
|
199 | See "nreqs" for an example. |
|
|
200 | |
|
|
201 | IO::AIO::nreqs |
|
|
202 | Returns the number of requests currently outstanding (i.e. for which |
|
|
203 | their callback has not been invoked yet). |
|
|
204 | |
|
|
205 | Example: wait till there are no outstanding requests anymore: |
|
|
206 | |
|
|
207 | IO::AIO::poll_wait, IO::AIO::poll_cb |
|
|
208 | while IO::AIO::nreqs; |
|
|
209 | |
|
|
210 | IO::AIO::flush |
|
|
211 | Wait till all outstanding AIO requests have been handled. |
|
|
212 | |
|
|
213 | Strictly equivalent to: |
|
|
214 | |
|
|
215 | IO::AIO::poll_wait, IO::AIO::poll_cb |
|
|
216 | while IO::AIO::nreqs; |
|
|
217 | |
|
|
218 | IO::AIO::poll |
|
|
219 | Waits until some requests have been handled. |
|
|
220 | |
|
|
221 | Strictly equivalent to: |
|
|
222 | |
|
|
223 | IO::AIO::poll_wait, IO::AIO::poll_cb |
|
|
224 | if IO::AIO::nreqs; |
|
|
225 | |
|
|
226 | IO::AIO::min_parallel $nthreads |
|
|
227 | Set the minimum number of AIO threads to $nthreads. The default is |
|
|
228 | 1, which means a single asynchronous operation can be done at one |
|
|
229 | time (the number of outstanding operations, however, is unlimited). |
|
|
230 | |
|
|
231 | It is recommended to keep the number of threads low, as some Linux |
|
|
232 | kernel versions will scale negatively with the number of threads |
|
|
233 | (higher parallelity => MUCH higher latency). With current Linux 2.6 |
|
|
234 | versions, 4-32 threads should be fine. |
|
|
235 | |
|
|
236 | Under normal circumstances you don't need to call this function, as |
|
|
237 | this module automatically starts some threads (the exact number |
|
|
238 | might change, and is currently 4). |
|
|
239 | |
|
|
240 | IO::AIO::max_parallel $nthreads |
|
|
241 | Sets the maximum number of AIO threads to $nthreads. If more than |
|
|
242 | the specified number of threads are currently running, kill them. |
|
|
243 | This function blocks until the limit is reached. |
|
|
244 | |
|
|
245 | This module automatically runs "max_parallel 0" at program end, to |
|
|
246 | ensure that all threads are killed and that there are no outstanding |
|
|
247 | requests. |
|
|
248 | |
|
|
249 | Under normal circumstances you don't need to call this function. |
|
|
250 | |
|
|
251 | $oldnreqs = IO::AIO::max_outstanding $nreqs |
|
|
252 | Sets the maximum number of outstanding requests to $nreqs. If you |
|
|
253 | try to queue up more than this number of requests, the caller will |
|
|
254 | block until some requests have been handled. |
|
|
255 | |
|
|
256 | The default is very large, so normally there is no practical limit. |
|
|
257 | If you queue up many requests in a loop it it often improves speed |
|
|
258 | if you set this to a relatively low number, such as 100. |
|
|
259 | |
|
|
260 | Under normal circumstances you don't need to call this function. |
186 | |
261 | |
187 | SEE ALSO |
262 | SEE ALSO |
188 | Coro. |
263 | Coro, Linux::AIO. |
189 | |
264 | |
190 | AUTHOR |
265 | AUTHOR |
191 | Marc Lehmann <schmorp@schmorp.de> |
266 | Marc Lehmann <schmorp@schmorp.de> |
192 | http://home.schmorp.de/ |
267 | http://home.schmorp.de/ |
193 | |
268 | |