1 |
<?xml version="1.0" encoding="UTF-8"?> |
2 |
<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.1//EN" "http://www.w3.org/TR/xhtml11/DTD/xhtml11.dtd"> |
3 |
<html xmlns="http://www.w3.org/1999/xhtml" xml:lang="en"> |
4 |
<head> |
5 |
<title>libev</title> |
6 |
<meta name="description" content="Pod documentation for libev" /> |
7 |
<meta name="inputfile" content="<standard input>" /> |
8 |
<meta name="outputfile" content="<standard output>" /> |
9 |
<meta name="created" content="Mon Nov 12 09:02:48 2007" /> |
10 |
<meta name="generator" content="Pod::Xhtml 1.57" /> |
11 |
<link rel="stylesheet" href="http://res.tst.eu/pod.css"/></head> |
12 |
<body> |
13 |
<div class="pod"> |
14 |
<!-- INDEX START --> |
15 |
<h3 id="TOP">Index</h3> |
16 |
|
17 |
<ul><li><a href="#NAME">NAME</a></li> |
18 |
<li><a href="#SYNOPSIS">SYNOPSIS</a></li> |
19 |
<li><a href="#DESCRIPTION">DESCRIPTION</a></li> |
20 |
<li><a href="#FEATURES">FEATURES</a></li> |
21 |
<li><a href="#CONVENTIONS">CONVENTIONS</a></li> |
22 |
<li><a href="#TIME_AND_OTHER_GLOBAL_FUNCTIONS">TIME AND OTHER GLOBAL FUNCTIONS</a></li> |
23 |
<li><a href="#FUNCTIONS_CONTROLLING_THE_EVENT_LOOP">FUNCTIONS CONTROLLING THE EVENT LOOP</a></li> |
24 |
<li><a href="#ANATOMY_OF_A_WATCHER">ANATOMY OF A WATCHER</a> |
25 |
<ul><li><a href="#ASSOCIATING_CUSTOM_DATA_WITH_A_WATCH">ASSOCIATING CUSTOM DATA WITH A WATCHER</a></li> |
26 |
</ul> |
27 |
</li> |
28 |
<li><a href="#WATCHER_TYPES">WATCHER TYPES</a> |
29 |
<ul><li><a href="#struct_ev_io_is_my_file_descriptor_r">struct ev_io - is my file descriptor readable or writable</a></li> |
30 |
<li><a href="#struct_ev_timer_relative_and_optiona">struct ev_timer - relative and optionally recurring timeouts</a></li> |
31 |
<li><a href="#ev_periodic">ev_periodic</a></li> |
32 |
<li><a href="#ev_signal_signal_me_when_a_signal_ge">ev_signal - signal me when a signal gets signalled</a></li> |
33 |
<li><a href="#ev_child_wait_for_pid_status_changes">ev_child - wait for pid status changes</a></li> |
34 |
<li><a href="#ev_idle_when_you_ve_got_nothing_bett">ev_idle - when you've got nothing better to do</a></li> |
35 |
<li><a href="#prepare_and_check_your_hooks_into_th">prepare and check - your hooks into the event loop</a></li> |
36 |
</ul> |
37 |
</li> |
38 |
<li><a href="#OTHER_FUNCTIONS">OTHER FUNCTIONS</a></li> |
39 |
<li><a href="#AUTHOR">AUTHOR</a> |
40 |
</li> |
41 |
</ul><hr /> |
42 |
<!-- INDEX END --> |
43 |
|
44 |
<h1 id="NAME">NAME</h1><p><a href="#TOP" class="toplink">Top</a></p> |
45 |
<div id="NAME_CONTENT"> |
46 |
<p>libev - a high performance full-featured event loop written in C</p> |
47 |
|
48 |
</div> |
49 |
<h1 id="SYNOPSIS">SYNOPSIS</h1><p><a href="#TOP" class="toplink">Top</a></p> |
50 |
<div id="SYNOPSIS_CONTENT"> |
51 |
<pre> #include <ev.h> |
52 |
|
53 |
</pre> |
54 |
|
55 |
</div> |
56 |
<h1 id="DESCRIPTION">DESCRIPTION</h1><p><a href="#TOP" class="toplink">Top</a></p> |
57 |
<div id="DESCRIPTION_CONTENT"> |
58 |
<p>Libev is an event loop: you register interest in certain events (such as a |
59 |
file descriptor being readable or a timeout occuring), and it will manage |
60 |
these event sources and provide your program events.</p> |
61 |
<p>To do this, it must take more or less complete control over your process |
62 |
(or thread) by executing the <i>event loop</i> handler, and will then |
63 |
communicate events via a callback mechanism.</p> |
64 |
<p>You register interest in certain events by registering so-called <i>event |
65 |
watchers</i>, which are relatively small C structures you initialise with the |
66 |
details of the event, and then hand it over to libev by <i>starting</i> the |
67 |
watcher.</p> |
68 |
|
69 |
</div> |
70 |
<h1 id="FEATURES">FEATURES</h1><p><a href="#TOP" class="toplink">Top</a></p> |
71 |
<div id="FEATURES_CONTENT"> |
72 |
<p>Libev supports select, poll, the linux-specific epoll and the bsd-specific |
73 |
kqueue mechanisms for file descriptor events, relative timers, absolute |
74 |
timers with customised rescheduling, signal events, process status change |
75 |
events (related to SIGCHLD), and event watchers dealing with the event |
76 |
loop mechanism itself (idle, prepare and check watchers).</p> |
77 |
|
78 |
</div> |
79 |
<h1 id="CONVENTIONS">CONVENTIONS</h1><p><a href="#TOP" class="toplink">Top</a></p> |
80 |
<div id="CONVENTIONS_CONTENT"> |
81 |
<p>Libev is very configurable. In this manual the default configuration |
82 |
will be described, which supports multiple event loops. For more info |
83 |
about various configuraiton options please have a look at the file |
84 |
<cite>README.embed</cite> in the libev distribution. If libev was configured without |
85 |
support for multiple event loops, then all functions taking an initial |
86 |
argument of name <code>loop</code> (which is always of type <code>struct ev_loop *</code>) |
87 |
will not have this argument.</p> |
88 |
|
89 |
</div> |
90 |
<h1 id="TIME_AND_OTHER_GLOBAL_FUNCTIONS">TIME AND OTHER GLOBAL FUNCTIONS</h1><p><a href="#TOP" class="toplink">Top</a></p> |
91 |
<div id="TIME_AND_OTHER_GLOBAL_FUNCTIONS_CONT"> |
92 |
<p>Libev represents time as a single floating point number, representing the |
93 |
(fractional) number of seconds since the (POSIX) epoch (somewhere near |
94 |
the beginning of 1970, details are complicated, don't ask). This type is |
95 |
called <code>ev_tstamp</code>, which is what you should use too. It usually aliases |
96 |
to the double type in C.</p> |
97 |
<dl> |
98 |
<dt>ev_tstamp ev_time ()</dt> |
99 |
<dd> |
100 |
<p>Returns the current time as libev would use it.</p> |
101 |
</dd> |
102 |
<dt>int ev_version_major ()</dt> |
103 |
<dt>int ev_version_minor ()</dt> |
104 |
<dd> |
105 |
<p>You can find out the major and minor version numbers of the library |
106 |
you linked against by calling the functions <code>ev_version_major</code> and |
107 |
<code>ev_version_minor</code>. If you want, you can compare against the global |
108 |
symbols <code>EV_VERSION_MAJOR</code> and <code>EV_VERSION_MINOR</code>, which specify the |
109 |
version of the library your program was compiled against.</p> |
110 |
<p>Usually, its a good idea to terminate if the major versions mismatch, |
111 |
as this indicates an incompatible change. Minor versions are usually |
112 |
compatible to older versions, so a larger minor version alone is usually |
113 |
not a problem.</p> |
114 |
</dd> |
115 |
<dt>ev_set_allocator (void *(*cb)(void *ptr, long size))</dt> |
116 |
<dd> |
117 |
<p>Sets the allocation function to use (the prototype is similar to the |
118 |
realloc function). It is used to allocate and free memory (no surprises |
119 |
here). If it returns zero when memory needs to be allocated, the library |
120 |
might abort or take some potentially destructive action. The default is |
121 |
your system realloc function.</p> |
122 |
<p>You could override this function in high-availability programs to, say, |
123 |
free some memory if it cannot allocate memory, to use a special allocator, |
124 |
or even to sleep a while and retry until some memory is available.</p> |
125 |
</dd> |
126 |
<dt>ev_set_syserr_cb (void (*cb)(const char *msg));</dt> |
127 |
<dd> |
128 |
<p>Set the callback function to call on a retryable syscall error (such |
129 |
as failed select, poll, epoll_wait). The message is a printable string |
130 |
indicating the system call or subsystem causing the problem. If this |
131 |
callback is set, then libev will expect it to remedy the sitution, no |
132 |
matter what, when it returns. That is, libev will geenrally retry the |
133 |
requested operation, or, if the condition doesn't go away, do bad stuff |
134 |
(such as abort).</p> |
135 |
</dd> |
136 |
</dl> |
137 |
|
138 |
</div> |
139 |
<h1 id="FUNCTIONS_CONTROLLING_THE_EVENT_LOOP">FUNCTIONS CONTROLLING THE EVENT LOOP</h1><p><a href="#TOP" class="toplink">Top</a></p> |
140 |
<div id="FUNCTIONS_CONTROLLING_THE_EVENT_LOOP-2"> |
141 |
<p>An event loop is described by a <code>struct ev_loop *</code>. The library knows two |
142 |
types of such loops, the <i>default</i> loop, which supports signals and child |
143 |
events, and dynamically created loops which do not.</p> |
144 |
<p>If you use threads, a common model is to run the default event loop |
145 |
in your main thread (or in a separate thrad) and for each thread you |
146 |
create, you also create another event loop. Libev itself does no lockign |
147 |
whatsoever, so if you mix calls to different event loops, make sure you |
148 |
lock (this is usually a bad idea, though, even if done right).</p> |
149 |
<dl> |
150 |
<dt>struct ev_loop *ev_default_loop (unsigned int flags)</dt> |
151 |
<dd> |
152 |
<p>This will initialise the default event loop if it hasn't been initialised |
153 |
yet and return it. If the default loop could not be initialised, returns |
154 |
false. If it already was initialised it simply returns it (and ignores the |
155 |
flags).</p> |
156 |
<p>If you don't know what event loop to use, use the one returned from this |
157 |
function.</p> |
158 |
<p>The flags argument can be used to specify special behaviour or specific |
159 |
backends to use, and is usually specified as 0 (or EVFLAG_AUTO)</p> |
160 |
<p>It supports the following flags:</p> |
161 |
<p> |
162 |
<dl> |
163 |
<dt>EVFLAG_AUTO</dt> |
164 |
<dd> |
165 |
<p>The default flags value. Use this if you have no clue (its the right |
166 |
thing, believe me).</p> |
167 |
</dd> |
168 |
<dt>EVFLAG_NOENV</dt> |
169 |
<dd> |
170 |
<p>If this flag bit is ored into the flag value then libev will <i>not</i> look |
171 |
at the environment variable <code>LIBEV_FLAGS</code>. Otherwise (the default), this |
172 |
environment variable will override the flags completely. This is useful |
173 |
to try out specific backends to tets their performance, or to work around |
174 |
bugs.</p> |
175 |
</dd> |
176 |
<dt>EVMETHOD_SELECT portable select backend</dt> |
177 |
<dt>EVMETHOD_POLL poll backend (everywhere except windows)</dt> |
178 |
<dt>EVMETHOD_EPOLL linux only</dt> |
179 |
<dt>EVMETHOD_KQUEUE some bsds only</dt> |
180 |
<dt>EVMETHOD_DEVPOLL solaris 8 only</dt> |
181 |
<dt>EVMETHOD_PORT solaris 10 only</dt> |
182 |
<dd> |
183 |
<p>If one or more of these are ored into the flags value, then only these |
184 |
backends will be tried (in the reverse order as given here). If one are |
185 |
specified, any backend will do.</p> |
186 |
</dd> |
187 |
</dl> |
188 |
</p> |
189 |
</dd> |
190 |
<dt>struct ev_loop *ev_loop_new (unsigned int flags)</dt> |
191 |
<dd> |
192 |
<p>Similar to <code>ev_default_loop</code>, but always creates a new event loop that is |
193 |
always distinct from the default loop. Unlike the default loop, it cannot |
194 |
handle signal and child watchers, and attempts to do so will be greeted by |
195 |
undefined behaviour (or a failed assertion if assertions are enabled).</p> |
196 |
</dd> |
197 |
<dt>ev_default_destroy ()</dt> |
198 |
<dd> |
199 |
<p>Destroys the default loop again (frees all memory and kernel state |
200 |
etc.). This stops all registered event watchers (by not touching them in |
201 |
any way whatsoever, although you cnanot rely on this :).</p> |
202 |
</dd> |
203 |
<dt>ev_loop_destroy (loop)</dt> |
204 |
<dd> |
205 |
<p>Like <code>ev_default_destroy</code>, but destroys an event loop created by an |
206 |
earlier call to <code>ev_loop_new</code>.</p> |
207 |
</dd> |
208 |
<dt>ev_default_fork ()</dt> |
209 |
<dd> |
210 |
<p>This function reinitialises the kernel state for backends that have |
211 |
one. Despite the name, you can call it anytime, but it makes most sense |
212 |
after forking, in either the parent or child process (or both, but that |
213 |
again makes little sense).</p> |
214 |
<p>You <i>must</i> call this function after forking if and only if you want to |
215 |
use the event library in both processes. If you just fork+exec, you don't |
216 |
have to call it.</p> |
217 |
<p>The function itself is quite fast and its usually not a problem to call |
218 |
it just in case after a fork. To make this easy, the function will fit in |
219 |
quite nicely into a call to <code>pthread_atfork</code>:</p> |
220 |
<pre> pthread_atfork (0, 0, ev_default_fork); |
221 |
|
222 |
</pre> |
223 |
</dd> |
224 |
<dt>ev_loop_fork (loop)</dt> |
225 |
<dd> |
226 |
<p>Like <code>ev_default_fork</code>, but acts on an event loop created by |
227 |
<code>ev_loop_new</code>. Yes, you have to call this on every allocated event loop |
228 |
after fork, and how you do this is entirely your own problem.</p> |
229 |
</dd> |
230 |
<dt>unsigned int ev_method (loop)</dt> |
231 |
<dd> |
232 |
<p>Returns one of the <code>EVMETHOD_*</code> flags indicating the event backend in |
233 |
use.</p> |
234 |
</dd> |
235 |
<dt>ev_tstamp = ev_now (loop)</dt> |
236 |
<dd> |
237 |
<p>Returns the current "event loop time", which is the time the event loop |
238 |
got events and started processing them. This timestamp does not change |
239 |
as long as callbacks are being processed, and this is also the base time |
240 |
used for relative timers. You can treat it as the timestamp of the event |
241 |
occuring (or more correctly, the mainloop finding out about it).</p> |
242 |
</dd> |
243 |
<dt>ev_loop (loop, int flags)</dt> |
244 |
<dd> |
245 |
<p>Finally, this is it, the event handler. This function usually is called |
246 |
after you initialised all your watchers and you want to start handling |
247 |
events.</p> |
248 |
<p>If the flags argument is specified as 0, it will not return until either |
249 |
no event watchers are active anymore or <code>ev_unloop</code> was called.</p> |
250 |
<p>A flags value of <code>EVLOOP_NONBLOCK</code> will look for new events, will handle |
251 |
those events and any outstanding ones, but will not block your process in |
252 |
case there are no events.</p> |
253 |
<p>A flags value of <code>EVLOOP_ONESHOT</code> will look for new events (waiting if |
254 |
neccessary) and will handle those and any outstanding ones. It will block |
255 |
your process until at least one new event arrives.</p> |
256 |
<p>This flags value could be used to implement alternative looping |
257 |
constructs, but the <code>prepare</code> and <code>check</code> watchers provide a better and |
258 |
more generic mechanism.</p> |
259 |
</dd> |
260 |
<dt>ev_unloop (loop, how)</dt> |
261 |
<dd> |
262 |
<p>Can be used to make a call to <code>ev_loop</code> return early. The <code>how</code> argument |
263 |
must be either <code>EVUNLOOP_ONCE</code>, which will make the innermost <code>ev_loop</code> |
264 |
call return, or <code>EVUNLOOP_ALL</code>, which will make all nested <code>ev_loop</code> |
265 |
calls return.</p> |
266 |
</dd> |
267 |
<dt>ev_ref (loop)</dt> |
268 |
<dt>ev_unref (loop)</dt> |
269 |
<dd> |
270 |
<p>Ref/unref can be used to add or remove a refcount on the event loop: Every |
271 |
watcher keeps one reference. If you have a long-runing watcher you never |
272 |
unregister that should not keep ev_loop from running, ev_unref() after |
273 |
starting, and ev_ref() before stopping it. Libev itself uses this for |
274 |
example for its internal signal pipe: It is not visible to you as a user |
275 |
and should not keep <code>ev_loop</code> from exiting if the work is done. It is |
276 |
also an excellent way to do this for generic recurring timers or from |
277 |
within third-party libraries. Just remember to unref after start and ref |
278 |
before stop.</p> |
279 |
</dd> |
280 |
</dl> |
281 |
|
282 |
</div> |
283 |
<h1 id="ANATOMY_OF_A_WATCHER">ANATOMY OF A WATCHER</h1><p><a href="#TOP" class="toplink">Top</a></p> |
284 |
<div id="ANATOMY_OF_A_WATCHER_CONTENT"> |
285 |
<p>A watcher is a structure that you create and register to record your |
286 |
interest in some event. For instance, if you want to wait for STDIN to |
287 |
become readable, you would create an ev_io watcher for that:</p> |
288 |
<pre> static void my_cb (struct ev_loop *loop, struct ev_io *w, int revents) |
289 |
{ |
290 |
ev_io_stop (w); |
291 |
ev_unloop (loop, EVUNLOOP_ALL); |
292 |
} |
293 |
|
294 |
struct ev_loop *loop = ev_default_loop (0); |
295 |
struct ev_io stdin_watcher; |
296 |
ev_init (&stdin_watcher, my_cb); |
297 |
ev_io_set (&stdin_watcher, STDIN_FILENO, EV_READ); |
298 |
ev_io_start (loop, &stdin_watcher); |
299 |
ev_loop (loop, 0); |
300 |
|
301 |
</pre> |
302 |
<p>As you can see, you are responsible for allocating the memory for your |
303 |
watcher structures (and it is usually a bad idea to do this on the stack, |
304 |
although this can sometimes be quite valid).</p> |
305 |
<p>Each watcher structure must be initialised by a call to <code>ev_init |
306 |
(watcher *, callback)</code>, which expects a callback to be provided. This |
307 |
callback gets invoked each time the event occurs (or, in the case of io |
308 |
watchers, each time the event loop detects that the file descriptor given |
309 |
is readable and/or writable).</p> |
310 |
<p>Each watcher type has its own <code>ev_<type>_set (watcher *, ...)</code> macro |
311 |
with arguments specific to this watcher type. There is also a macro |
312 |
to combine initialisation and setting in one call: <code>ev_<type>_init |
313 |
(watcher *, callback, ...)</code>.</p> |
314 |
<p>To make the watcher actually watch out for events, you have to start it |
315 |
with a watcher-specific start function (<code>ev_<type>_start (loop, watcher |
316 |
*)</code>), and you can stop watching for events at any time by calling the |
317 |
corresponding stop function (<code>ev_<type>_stop (loop, watcher *)</code>.</p> |
318 |
<p>As long as your watcher is active (has been started but not stopped) you |
319 |
must not touch the values stored in it. Most specifically you must never |
320 |
reinitialise it or call its set method.</p> |
321 |
<p>You cna check wether an event is active by calling the <code>ev_is_active |
322 |
(watcher *)</code> macro. To see wether an event is outstanding (but the |
323 |
callback for it has not been called yet) you cna use the <code>ev_is_pending |
324 |
(watcher *)</code> macro.</p> |
325 |
<p>Each and every callback receives the event loop pointer as first, the |
326 |
registered watcher structure as second, and a bitset of received events as |
327 |
third argument.</p> |
328 |
<p>The rceeived events usually include a single bit per event type received |
329 |
(you can receive multiple events at the same time). The possible bit masks |
330 |
are:</p> |
331 |
<dl> |
332 |
<dt>EV_READ</dt> |
333 |
<dt>EV_WRITE</dt> |
334 |
<dd> |
335 |
<p>The file descriptor in the ev_io watcher has become readable and/or |
336 |
writable.</p> |
337 |
</dd> |
338 |
<dt>EV_TIMEOUT</dt> |
339 |
<dd> |
340 |
<p>The ev_timer watcher has timed out.</p> |
341 |
</dd> |
342 |
<dt>EV_PERIODIC</dt> |
343 |
<dd> |
344 |
<p>The ev_periodic watcher has timed out.</p> |
345 |
</dd> |
346 |
<dt>EV_SIGNAL</dt> |
347 |
<dd> |
348 |
<p>The signal specified in the ev_signal watcher has been received by a thread.</p> |
349 |
</dd> |
350 |
<dt>EV_CHILD</dt> |
351 |
<dd> |
352 |
<p>The pid specified in the ev_child watcher has received a status change.</p> |
353 |
</dd> |
354 |
<dt>EV_IDLE</dt> |
355 |
<dd> |
356 |
<p>The ev_idle watcher has determined that you have nothing better to do.</p> |
357 |
</dd> |
358 |
<dt>EV_PREPARE</dt> |
359 |
<dt>EV_CHECK</dt> |
360 |
<dd> |
361 |
<p>All ev_prepare watchers are invoked just <i>before</i> <code>ev_loop</code> starts |
362 |
to gather new events, and all ev_check watchers are invoked just after |
363 |
<code>ev_loop</code> has gathered them, but before it invokes any callbacks for any |
364 |
received events. Callbacks of both watcher types can start and stop as |
365 |
many watchers as they want, and all of them will be taken into account |
366 |
(for example, a ev_prepare watcher might start an idle watcher to keep |
367 |
<code>ev_loop</code> from blocking).</p> |
368 |
</dd> |
369 |
<dt>EV_ERROR</dt> |
370 |
<dd> |
371 |
<p>An unspecified error has occured, the watcher has been stopped. This might |
372 |
happen because the watcher could not be properly started because libev |
373 |
ran out of memory, a file descriptor was found to be closed or any other |
374 |
problem. You best act on it by reporting the problem and somehow coping |
375 |
with the watcher being stopped.</p> |
376 |
<p>Libev will usually signal a few "dummy" events together with an error, |
377 |
for example it might indicate that a fd is readable or writable, and if |
378 |
your callbacks is well-written it can just attempt the operation and cope |
379 |
with the error from read() or write(). This will not work in multithreaded |
380 |
programs, though, so beware.</p> |
381 |
</dd> |
382 |
</dl> |
383 |
|
384 |
</div> |
385 |
<h2 id="ASSOCIATING_CUSTOM_DATA_WITH_A_WATCH">ASSOCIATING CUSTOM DATA WITH A WATCHER</h2> |
386 |
<div id="ASSOCIATING_CUSTOM_DATA_WITH_A_WATCH-2"> |
387 |
<p>Each watcher has, by default, a member <code>void *data</code> that you can change |
388 |
and read at any time, libev will completely ignore it. This cna be used |
389 |
to associate arbitrary data with your watcher. If you need more data and |
390 |
don't want to allocate memory and store a pointer to it in that data |
391 |
member, you can also "subclass" the watcher type and provide your own |
392 |
data:</p> |
393 |
<pre> struct my_io |
394 |
{ |
395 |
struct ev_io io; |
396 |
int otherfd; |
397 |
void *somedata; |
398 |
struct whatever *mostinteresting; |
399 |
} |
400 |
|
401 |
</pre> |
402 |
<p>And since your callback will be called with a pointer to the watcher, you |
403 |
can cast it back to your own type:</p> |
404 |
<pre> static void my_cb (struct ev_loop *loop, struct ev_io *w_, int revents) |
405 |
{ |
406 |
struct my_io *w = (struct my_io *)w_; |
407 |
... |
408 |
} |
409 |
|
410 |
</pre> |
411 |
<p>More interesting and less C-conformant ways of catsing your callback type |
412 |
have been omitted....</p> |
413 |
|
414 |
|
415 |
|
416 |
|
417 |
|
418 |
</div> |
419 |
<h1 id="WATCHER_TYPES">WATCHER TYPES</h1><p><a href="#TOP" class="toplink">Top</a></p> |
420 |
<div id="WATCHER_TYPES_CONTENT"> |
421 |
<p>This section describes each watcher in detail, but will not repeat |
422 |
information given in the last section.</p> |
423 |
|
424 |
</div> |
425 |
<h2 id="struct_ev_io_is_my_file_descriptor_r">struct ev_io - is my file descriptor readable or writable</h2> |
426 |
<div id="struct_ev_io_is_my_file_descriptor_r-2"> |
427 |
<p>I/O watchers check wether a file descriptor is readable or writable |
428 |
in each iteration of the event loop (This behaviour is called |
429 |
level-triggering because you keep receiving events as long as the |
430 |
condition persists. Remember you cna stop the watcher if you don't want to |
431 |
act on the event and neither want to receive future events).</p> |
432 |
<dl> |
433 |
<dt>ev_io_init (ev_io *, callback, int fd, int events)</dt> |
434 |
<dt>ev_io_set (ev_io *, int fd, int events)</dt> |
435 |
<dd> |
436 |
<p>Configures an ev_io watcher. The fd is the file descriptor to rceeive |
437 |
events for and events is either <code>EV_READ</code>, <code>EV_WRITE</code> or <code>EV_READ | |
438 |
EV_WRITE</code> to receive the given events.</p> |
439 |
</dd> |
440 |
</dl> |
441 |
|
442 |
</div> |
443 |
<h2 id="struct_ev_timer_relative_and_optiona">struct ev_timer - relative and optionally recurring timeouts</h2> |
444 |
<div id="struct_ev_timer_relative_and_optiona-2"> |
445 |
<p>Timer watchers are simple relative timers that generate an event after a |
446 |
given time, and optionally repeating in regular intervals after that.</p> |
447 |
<p>The timers are based on real time, that is, if you register an event that |
448 |
times out after an hour and youreset your system clock to last years |
449 |
time, it will still time out after (roughly) and hour. "Roughly" because |
450 |
detecting time jumps is hard, and soem inaccuracies are unavoidable (the |
451 |
monotonic clock option helps a lot here).</p> |
452 |
<dl> |
453 |
<dt>ev_timer_init (ev_timer *, callback, ev_tstamp after, ev_tstamp repeat)</dt> |
454 |
<dt>ev_timer_set (ev_timer *, ev_tstamp after, ev_tstamp repeat)</dt> |
455 |
<dd> |
456 |
<p>Configure the timer to trigger after <code>after</code> seconds. If <code>repeat</code> is |
457 |
<code>0.</code>, then it will automatically be stopped. If it is positive, then the |
458 |
timer will automatically be configured to trigger again <code>repeat</code> seconds |
459 |
later, again, and again, until stopped manually.</p> |
460 |
<p>The timer itself will do a best-effort at avoiding drift, that is, if you |
461 |
configure a timer to trigger every 10 seconds, then it will trigger at |
462 |
exactly 10 second intervals. If, however, your program cannot keep up with |
463 |
the timer (ecause it takes longer than those 10 seconds to do stuff) the |
464 |
timer will not fire more than once per event loop iteration.</p> |
465 |
</dd> |
466 |
<dt>ev_timer_again (loop)</dt> |
467 |
<dd> |
468 |
<p>This will act as if the timer timed out and restart it again if it is |
469 |
repeating. The exact semantics are:</p> |
470 |
<p>If the timer is started but nonrepeating, stop it.</p> |
471 |
<p>If the timer is repeating, either start it if necessary (with the repeat |
472 |
value), or reset the running timer to the repeat value.</p> |
473 |
<p>This sounds a bit complicated, but here is a useful and typical |
474 |
example: Imagine you have a tcp connection and you want a so-called idle |
475 |
timeout, that is, you want to be called when there have been, say, 60 |
476 |
seconds of inactivity on the socket. The easiest way to do this is to |
477 |
configure an ev_timer with after=repeat=60 and calling ev_timer_again each |
478 |
time you successfully read or write some data. If you go into an idle |
479 |
state where you do not expect data to travel on the socket, you can stop |
480 |
the timer, and again will automatically restart it if need be.</p> |
481 |
</dd> |
482 |
</dl> |
483 |
|
484 |
</div> |
485 |
<h2 id="ev_periodic">ev_periodic</h2> |
486 |
<div id="ev_periodic_CONTENT"> |
487 |
<p>Periodic watchers are also timers of a kind, but they are very versatile |
488 |
(and unfortunately a bit complex).</p> |
489 |
<p>Unlike ev_timer's, they are not based on real time (or relative time) |
490 |
but on wallclock time (absolute time). You can tell a periodic watcher |
491 |
to trigger "at" some specific point in time. For example, if you tell a |
492 |
periodic watcher to trigger in 10 seconds (by specifiying e.g. c<ev_now () |
493 |
+ 10.>) and then reset your system clock to the last year, then it will |
494 |
take a year to trigger the event (unlike an ev_timer, which would trigger |
495 |
roughly 10 seconds later and of course not if you reset your system time |
496 |
again).</p> |
497 |
<p>They can also be used to implement vastly more complex timers, such as |
498 |
triggering an event on eahc midnight, local time.</p> |
499 |
<dl> |
500 |
<dt>ev_periodic_init (ev_periodic *, callback, ev_tstamp at, ev_tstamp interval, reschedule_cb)</dt> |
501 |
<dt>ev_periodic_set (ev_periodic *, ev_tstamp after, ev_tstamp repeat, reschedule_cb)</dt> |
502 |
<dd> |
503 |
<p>Lots of arguments, lets sort it out... There are basically three modes of |
504 |
operation, and we will explain them from simplest to complex:</p> |
505 |
|
506 |
|
507 |
|
508 |
|
509 |
<p> |
510 |
<dl> |
511 |
<dt>* absolute timer (interval = reschedule_cb = 0)</dt> |
512 |
<dd> |
513 |
<p>In this configuration the watcher triggers an event at the wallclock time |
514 |
<code>at</code> and doesn't repeat. It will not adjust when a time jump occurs, |
515 |
that is, if it is to be run at January 1st 2011 then it will run when the |
516 |
system time reaches or surpasses this time.</p> |
517 |
</dd> |
518 |
<dt>* non-repeating interval timer (interval > 0, reschedule_cb = 0)</dt> |
519 |
<dd> |
520 |
<p>In this mode the watcher will always be scheduled to time out at the next |
521 |
<code>at + N * interval</code> time (for some integer N) and then repeat, regardless |
522 |
of any time jumps.</p> |
523 |
<p>This can be used to create timers that do not drift with respect to system |
524 |
time:</p> |
525 |
<pre> ev_periodic_set (&periodic, 0., 3600., 0); |
526 |
|
527 |
</pre> |
528 |
<p>This doesn't mean there will always be 3600 seconds in between triggers, |
529 |
but only that the the callback will be called when the system time shows a |
530 |
full hour (UTC), or more correct, when the system time is evenly divisible |
531 |
by 3600.</p> |
532 |
<p>Another way to think about it (for the mathematically inclined) is that |
533 |
ev_periodic will try to run the callback in this mode at the next possible |
534 |
time where <code>time = at (mod interval)</code>, regardless of any time jumps.</p> |
535 |
</dd> |
536 |
<dt>* manual reschedule mode (reschedule_cb = callback)</dt> |
537 |
<dd> |
538 |
<p>In this mode the values for <code>interval</code> and <code>at</code> are both being |
539 |
ignored. Instead, each time the periodic watcher gets scheduled, the |
540 |
reschedule callback will be called with the watcher as first, and the |
541 |
current time as second argument.</p> |
542 |
<p>NOTE: <i>This callback MUST NOT stop or destroy the periodic or any other |
543 |
periodic watcher, ever, or make any event loop modificstions</i>. If you need |
544 |
to stop it, return 1e30 (or so, fudge fudge) and stop it afterwards.</p> |
545 |
<p>Its prototype is c<ev_tstamp (*reschedule_cb)(struct ev_periodic *w, |
546 |
ev_tstamp now)>, e.g.:</p> |
547 |
<pre> static ev_tstamp my_rescheduler (struct ev_periodic *w, ev_tstamp now) |
548 |
{ |
549 |
return now + 60.; |
550 |
} |
551 |
|
552 |
</pre> |
553 |
<p>It must return the next time to trigger, based on the passed time value |
554 |
(that is, the lowest time value larger than to the second argument). It |
555 |
will usually be called just before the callback will be triggered, but |
556 |
might be called at other times, too.</p> |
557 |
<p>This can be used to create very complex timers, such as a timer that |
558 |
triggers on each midnight, local time. To do this, you would calculate the |
559 |
next midnight after <code>now</code> and return the timestamp value for this. How you do this |
560 |
is, again, up to you (but it is not trivial).</p> |
561 |
</dd> |
562 |
</dl> |
563 |
</p> |
564 |
</dd> |
565 |
<dt>ev_periodic_again (loop, ev_periodic *)</dt> |
566 |
<dd> |
567 |
<p>Simply stops and restarts the periodic watcher again. This is only useful |
568 |
when you changed some parameters or the reschedule callback would return |
569 |
a different time than the last time it was called (e.g. in a crond like |
570 |
program when the crontabs have changed).</p> |
571 |
</dd> |
572 |
</dl> |
573 |
|
574 |
</div> |
575 |
<h2 id="ev_signal_signal_me_when_a_signal_ge">ev_signal - signal me when a signal gets signalled</h2> |
576 |
<div id="ev_signal_signal_me_when_a_signal_ge-2"> |
577 |
<p>Signal watchers will trigger an event when the process receives a specific |
578 |
signal one or more times. Even though signals are very asynchronous, libev |
579 |
will try its best to deliver signals synchronously, i.e. as part of the |
580 |
normal event processing, like any other event.</p> |
581 |
<p>You cna configure as many watchers as you like per signal. Only when the |
582 |
first watcher gets started will libev actually register a signal watcher |
583 |
with the kernel (thus it coexists with your own signal handlers as long |
584 |
as you don't register any with libev). Similarly, when the last signal |
585 |
watcher for a signal is stopped libev will reset the signal handler to |
586 |
SIG_DFL (regardless of what it was set to before).</p> |
587 |
<dl> |
588 |
<dt>ev_signal_init (ev_signal *, callback, int signum)</dt> |
589 |
<dt>ev_signal_set (ev_signal *, int signum)</dt> |
590 |
<dd> |
591 |
<p>Configures the watcher to trigger on the given signal number (usually one |
592 |
of the <code>SIGxxx</code> constants).</p> |
593 |
</dd> |
594 |
</dl> |
595 |
|
596 |
</div> |
597 |
<h2 id="ev_child_wait_for_pid_status_changes">ev_child - wait for pid status changes</h2> |
598 |
<div id="ev_child_wait_for_pid_status_changes-2"> |
599 |
<p>Child watchers trigger when your process receives a SIGCHLD in response to |
600 |
some child status changes (most typically when a child of yours dies).</p> |
601 |
<dl> |
602 |
<dt>ev_child_init (ev_child *, callback, int pid)</dt> |
603 |
<dt>ev_child_set (ev_child *, int pid)</dt> |
604 |
<dd> |
605 |
<p>Configures the watcher to wait for status changes of process <code>pid</code> (or |
606 |
<i>any</i> process if <code>pid</code> is specified as <code>0</code>). The callback can look |
607 |
at the <code>rstatus</code> member of the <code>ev_child</code> watcher structure to see |
608 |
the status word (use the macros from <code>sys/wait.h</code>). The <code>rpid</code> member |
609 |
contains the pid of the process causing the status change.</p> |
610 |
</dd> |
611 |
</dl> |
612 |
|
613 |
</div> |
614 |
<h2 id="ev_idle_when_you_ve_got_nothing_bett">ev_idle - when you've got nothing better to do</h2> |
615 |
<div id="ev_idle_when_you_ve_got_nothing_bett-2"> |
616 |
<p>Idle watchers trigger events when there are no other I/O or timer (or |
617 |
periodic) events pending. That is, as long as your process is busy |
618 |
handling sockets or timeouts it will not be called. But when your process |
619 |
is idle all idle watchers are being called again and again - until |
620 |
stopped, that is, or your process receives more events.</p> |
621 |
<p>The most noteworthy effect is that as long as any idle watchers are |
622 |
active, the process will not block when waiting for new events.</p> |
623 |
<p>Apart from keeping your process non-blocking (which is a useful |
624 |
effect on its own sometimes), idle watchers are a good place to do |
625 |
"pseudo-background processing", or delay processing stuff to after the |
626 |
event loop has handled all outstanding events.</p> |
627 |
<dl> |
628 |
<dt>ev_idle_init (ev_signal *, callback)</dt> |
629 |
<dd> |
630 |
<p>Initialises and configures the idle watcher - it has no parameters of any |
631 |
kind. There is a <code>ev_idle_set</code> macro, but using it is utterly pointless, |
632 |
believe me.</p> |
633 |
</dd> |
634 |
</dl> |
635 |
|
636 |
</div> |
637 |
<h2 id="prepare_and_check_your_hooks_into_th">prepare and check - your hooks into the event loop</h2> |
638 |
<div id="prepare_and_check_your_hooks_into_th-2"> |
639 |
<p>Prepare and check watchers usually (but not always) are used in |
640 |
tandom. Prepare watchers get invoked before the process blocks and check |
641 |
watchers afterwards.</p> |
642 |
<p>Their main purpose is to integrate other event mechanisms into libev. This |
643 |
could be used, for example, to track variable changes, implement your own |
644 |
watchers, integrate net-snmp or a coroutine library and lots more.</p> |
645 |
<p>This is done by examining in each prepare call which file descriptors need |
646 |
to be watched by the other library, registering ev_io watchers for them |
647 |
and starting an ev_timer watcher for any timeouts (many libraries provide |
648 |
just this functionality). Then, in the check watcher you check for any |
649 |
events that occured (by making your callbacks set soem flags for example) |
650 |
and call back into the library.</p> |
651 |
<p>As another example, the perl Coro module uses these hooks to integrate |
652 |
coroutines into libev programs, by yielding to other active coroutines |
653 |
during each prepare and only letting the process block if no coroutines |
654 |
are ready to run.</p> |
655 |
<dl> |
656 |
<dt>ev_prepare_init (ev_prepare *, callback)</dt> |
657 |
<dt>ev_check_init (ev_check *, callback)</dt> |
658 |
<dd> |
659 |
<p>Initialises and configures the prepare or check watcher - they have no |
660 |
parameters of any kind. There are <code>ev_prepare_set</code> and <code>ev_check_set</code> |
661 |
macros, but using them is utterly, utterly pointless.</p> |
662 |
</dd> |
663 |
</dl> |
664 |
|
665 |
</div> |
666 |
<h1 id="OTHER_FUNCTIONS">OTHER FUNCTIONS</h1><p><a href="#TOP" class="toplink">Top</a></p> |
667 |
<div id="OTHER_FUNCTIONS_CONTENT"> |
668 |
<p>There are some other fucntions of possible interest. Described. Here. Now.</p> |
669 |
<dl> |
670 |
<dt>ev_once (loop, int fd, int events, ev_tstamp timeout, callback)</dt> |
671 |
<dd> |
672 |
<p>This function combines a simple timer and an I/O watcher, calls your |
673 |
callback on whichever event happens first and automatically stop both |
674 |
watchers. This is useful if you want to wait for a single event on an fd |
675 |
or timeout without havign to allocate/configure/start/stop/free one or |
676 |
more watchers yourself.</p> |
677 |
<p>If <code>fd</code> is less than 0, then no I/O watcher will be started and events is |
678 |
ignored. Otherwise, an ev_io watcher for the given <code>fd</code> and <code>events</code> set |
679 |
will be craeted and started.</p> |
680 |
<p>If <code>timeout</code> is less than 0, then no timeout watcher will be |
681 |
started. Otherwise an ev_timer watcher with after = <code>timeout</code> (and repeat |
682 |
= 0) will be started.</p> |
683 |
<p>The callback has the type <code>void (*cb)(int revents, void *arg)</code> and |
684 |
gets passed an events set (normally a combination of EV_ERROR, EV_READ, |
685 |
EV_WRITE or EV_TIMEOUT) and the <code>arg</code> value passed to <code>ev_once</code>:</p> |
686 |
<pre> static void stdin_ready (int revents, void *arg) |
687 |
{ |
688 |
if (revents & EV_TIMEOUT) |
689 |
/* doh, nothing entered */ |
690 |
else if (revents & EV_READ) |
691 |
/* stdin might have data for us, joy! */ |
692 |
} |
693 |
|
694 |
ev_once (STDIN_FILENO, EV_READm 10., stdin_ready, 0); |
695 |
|
696 |
</pre> |
697 |
</dd> |
698 |
<dt>ev_feed_event (loop, watcher, int events)</dt> |
699 |
<dd> |
700 |
<p>Feeds the given event set into the event loop, as if the specified event |
701 |
has happened for the specified watcher (which must be a pointer to an |
702 |
initialised but not necessarily active event watcher).</p> |
703 |
</dd> |
704 |
<dt>ev_feed_fd_event (loop, int fd, int revents)</dt> |
705 |
<dd> |
706 |
<p>Feed an event on the given fd, as if a file descriptor backend detected it.</p> |
707 |
</dd> |
708 |
<dt>ev_feed_signal_event (loop, int signum)</dt> |
709 |
<dd> |
710 |
<p>Feed an event as if the given signal occured (loop must be the default loop!).</p> |
711 |
</dd> |
712 |
</dl> |
713 |
|
714 |
</div> |
715 |
<h1 id="AUTHOR">AUTHOR</h1><p><a href="#TOP" class="toplink">Top</a></p> |
716 |
<div id="AUTHOR_CONTENT"> |
717 |
<p>Marc Lehmann <libev@schmorp.de>.</p> |
718 |
|
719 |
</div> |
720 |
</div></body> |
721 |
</html> |