ViewVC Help
View File | Revision Log | Show Annotations | Download File
/cvs/libev/ev.html
(Generate patch)

Comparing libev/ev.html (file contents):
Revision 1.63 by root, Thu Nov 29 20:05:59 2007 UTC vs.
Revision 1.66 by root, Fri Dec 7 19:15:39 2007 UTC

… …
4<head> 4<head>
5 <title>libev</title> 5 <title>libev</title>
6 <meta name="description" content="Pod documentation for libev" /> 6 <meta name="description" content="Pod documentation for libev" />
7 <meta name="inputfile" content="&lt;standard input&gt;" /> 7 <meta name="inputfile" content="&lt;standard input&gt;" />
8 <meta name="outputfile" content="&lt;standard output&gt;" /> 8 <meta name="outputfile" content="&lt;standard output&gt;" />
9 <meta name="created" content="Thu Nov 29 21:05:58 2007" /> 9 <meta name="created" content="Fri Dec 7 20:07:44 2007" />
10 <meta name="generator" content="Pod::Xhtml 1.57" /> 10 <meta name="generator" content="Pod::Xhtml 1.57" />
11<link rel="stylesheet" href="http://res.tst.eu/pod.css"/></head> 11<link rel="stylesheet" href="http://res.tst.eu/pod.css"/></head>
12<body> 12<body>
13<div class="pod"> 13<div class="pod">
14<!-- INDEX START --> 14<!-- INDEX START -->
… …
119</pre> 119</pre>
120 120
121</div> 121</div>
122<h1 id="DESCRIPTION">DESCRIPTION</h1> 122<h1 id="DESCRIPTION">DESCRIPTION</h1>
123<div id="DESCRIPTION_CONTENT"> 123<div id="DESCRIPTION_CONTENT">
124<p>The newest version of this document is also available as a html-formatted
125web page you might find easier to navigate when reading it for the first
126time: <a href="http://cvs.schmorp.de/libev/ev.html">http://cvs.schmorp.de/libev/ev.html</a>.</p>
124<p>Libev is an event loop: you register interest in certain events (such as a 127<p>Libev is an event loop: you register interest in certain events (such as a
125file descriptor being readable or a timeout occuring), and it will manage 128file descriptor being readable or a timeout occuring), and it will manage
126these event sources and provide your program with events.</p> 129these event sources and provide your program with events.</p>
127<p>To do this, it must take more or less complete control over your process 130<p>To do this, it must take more or less complete control over your process
128(or thread) by executing the <i>event loop</i> handler, and will then 131(or thread) by executing the <i>event loop</i> handler, and will then
… …
333 <p>Instead of calling <code>ev_default_fork</code> or <code>ev_loop_fork</code> manually after 336 <p>Instead of calling <code>ev_default_fork</code> or <code>ev_loop_fork</code> manually after
334a fork, you can also make libev check for a fork in each iteration by 337a fork, you can also make libev check for a fork in each iteration by
335enabling this flag.</p> 338enabling this flag.</p>
336 <p>This works by calling <code>getpid ()</code> on every iteration of the loop, 339 <p>This works by calling <code>getpid ()</code> on every iteration of the loop,
337and thus this might slow down your event loop if you do a lot of loop 340and thus this might slow down your event loop if you do a lot of loop
338iterations and little real work, but is usually not noticable (on my 341iterations and little real work, but is usually not noticeable (on my
339Linux system for example, <code>getpid</code> is actually a simple 5-insn sequence 342Linux system for example, <code>getpid</code> is actually a simple 5-insn sequence
340without a syscall and thus <i>very</i> fast, but my Linux system also has 343without a syscall and thus <i>very</i> fast, but my Linux system also has
341<code>pthread_atfork</code> which is even faster).</p> 344<code>pthread_atfork</code> which is even faster).</p>
342 <p>The big advantage of this flag is that you can forget about fork (and 345 <p>The big advantage of this flag is that you can forget about fork (and
343forget about forgetting to tell libev about forking) when you use this 346forget about forgetting to tell libev about forking) when you use this
… …
480 <dt>ev_loop_fork (loop)</dt> 483 <dt>ev_loop_fork (loop)</dt>
481 <dd> 484 <dd>
482 <p>Like <code>ev_default_fork</code>, but acts on an event loop created by 485 <p>Like <code>ev_default_fork</code>, but acts on an event loop created by
483<code>ev_loop_new</code>. Yes, you have to call this on every allocated event loop 486<code>ev_loop_new</code>. Yes, you have to call this on every allocated event loop
484after fork, and how you do this is entirely your own problem.</p> 487after fork, and how you do this is entirely your own problem.</p>
488 </dd>
489 <dt>unsigned int ev_loop_count (loop)</dt>
490 <dd>
491 <p>Returns the count of loop iterations for the loop, which is identical to
492the number of times libev did poll for new events. It starts at <code>0</code> and
493happily wraps around with enough iterations.</p>
494 <p>This value can sometimes be useful as a generation counter of sorts (it
495&quot;ticks&quot; the number of loop iterations), as it roughly corresponds with
496<code>ev_prepare</code> and <code>ev_check</code> calls.</p>
485 </dd> 497 </dd>
486 <dt>unsigned int ev_backend (loop)</dt> 498 <dt>unsigned int ev_backend (loop)</dt>
487 <dd> 499 <dd>
488 <p>Returns one of the <code>EVBACKEND_*</code> flags indicating the event backend in 500 <p>Returns one of the <code>EVBACKEND_*</code> flags indicating the event backend in
489use.</p> 501use.</p>
… …
769 <dt>ev_cb_set (ev_TYPE *watcher, callback)</dt> 781 <dt>ev_cb_set (ev_TYPE *watcher, callback)</dt>
770 <dd> 782 <dd>
771 <p>Change the callback. You can change the callback at virtually any time 783 <p>Change the callback. You can change the callback at virtually any time
772(modulo threads).</p> 784(modulo threads).</p>
773 </dd> 785 </dd>
786 <dt>ev_set_priority (ev_TYPE *watcher, priority)</dt>
787 <dt>int ev_priority (ev_TYPE *watcher)</dt>
788 <dd>
789 <p>Set and query the priority of the watcher. The priority is a small
790integer between <code>EV_MAXPRI</code> (default: <code>2</code>) and <code>EV_MINPRI</code>
791(default: <code>-2</code>). Pending watchers with higher priority will be invoked
792before watchers with lower priority, but priority will not keep watchers
793from being executed (except for <code>ev_idle</code> watchers).</p>
794 <p>This means that priorities are <i>only</i> used for ordering callback
795invocation after new events have been received. This is useful, for
796example, to reduce latency after idling, or more often, to bind two
797watchers on the same event and make sure one is called first.</p>
798 <p>If you need to suppress invocation when higher priority events are pending
799you need to look at <code>ev_idle</code> watchers, which provide this functionality.</p>
800 <p>The default priority used by watchers when no priority has been set is
801always <code>0</code>, which is supposed to not be too high and not be too low :).</p>
802 <p>Setting a priority outside the range of <code>EV_MINPRI</code> to <code>EV_MAXPRI</code> is
803fine, as long as you do not mind that the priority value you query might
804or might not have been adjusted to be within valid range.</p>
805 </dd>
774</dl> 806</dl>
775 807
776 808
777 809
778 810
… …
888this situation even with a relatively standard program structure. Thus 920this situation even with a relatively standard program structure. Thus
889it is best to always use non-blocking I/O: An extra <code>read</code>(2) returning 921it is best to always use non-blocking I/O: An extra <code>read</code>(2) returning
890<code>EAGAIN</code> is far preferable to a program hanging until some data arrives.</p> 922<code>EAGAIN</code> is far preferable to a program hanging until some data arrives.</p>
891<p>If you cannot run the fd in non-blocking mode (for example you should not 923<p>If you cannot run the fd in non-blocking mode (for example you should not
892play around with an Xlib connection), then you have to seperately re-test 924play around with an Xlib connection), then you have to seperately re-test
893wether a file descriptor is really ready with a known-to-be good interface 925whether a file descriptor is really ready with a known-to-be good interface
894such as poll (fortunately in our Xlib example, Xlib already does this on 926such as poll (fortunately in our Xlib example, Xlib already does this on
895its own, so its quite safe to use).</p> 927its own, so its quite safe to use).</p>
896<dl> 928<dl>
897 <dt>ev_io_init (ev_io *, callback, int fd, int events)</dt> 929 <dt>ev_io_init (ev_io *, callback, int fd, int events)</dt>
898 <dt>ev_io_set (ev_io *, int fd, int events)</dt> 930 <dt>ev_io_set (ev_io *, int fd, int events)</dt>
… …
1358</pre> 1390</pre>
1359 1391
1360</div> 1392</div>
1361<h2 id="code_ev_idle_code_when_you_ve_got_no"><code>ev_idle</code> - when you've got nothing better to do...</h2> 1393<h2 id="code_ev_idle_code_when_you_ve_got_no"><code>ev_idle</code> - when you've got nothing better to do...</h2>
1362<div id="code_ev_idle_code_when_you_ve_got_no-2"> 1394<div id="code_ev_idle_code_when_you_ve_got_no-2">
1363<p>Idle watchers trigger events when there are no other events are pending 1395<p>Idle watchers trigger events when no other events of the same or higher
1364(prepare, check and other idle watchers do not count). That is, as long 1396priority are pending (prepare, check and other idle watchers do not
1365as your process is busy handling sockets or timeouts (or even signals, 1397count).</p>
1366imagine) it will not be triggered. But when your process is idle all idle 1398<p>That is, as long as your process is busy handling sockets or timeouts
1367watchers are being called again and again, once per event loop iteration - 1399(or even signals, imagine) of the same or higher priority it will not be
1400triggered. But when your process is idle (or only lower-priority watchers
1401are pending), the idle watchers are being called once per event loop
1368until stopped, that is, or your process receives more events and becomes 1402iteration - until stopped, that is, or your process receives more events
1369busy.</p> 1403and becomes busy again with higher priority stuff.</p>
1370<p>The most noteworthy effect is that as long as any idle watchers are 1404<p>The most noteworthy effect is that as long as any idle watchers are
1371active, the process will not block when waiting for new events.</p> 1405active, the process will not block when waiting for new events.</p>
1372<p>Apart from keeping your process non-blocking (which is a useful 1406<p>Apart from keeping your process non-blocking (which is a useful
1373effect on its own sometimes), idle watchers are a good place to do 1407effect on its own sometimes), idle watchers are a good place to do
1374&quot;pseudo-background processing&quot;, or delay processing stuff to after the 1408&quot;pseudo-background processing&quot;, or delay processing stuff to after the
… …
1464 1498
1465 // create io watchers for each fd and a timer before blocking 1499 // create io watchers for each fd and a timer before blocking
1466 static void 1500 static void
1467 adns_prepare_cb (ev_loop *loop, ev_prepare *w, int revents) 1501 adns_prepare_cb (ev_loop *loop, ev_prepare *w, int revents)
1468 { 1502 {
1469 int timeout = 3600000;truct pollfd fds [nfd]; 1503 int timeout = 3600000;
1504 struct pollfd fds [nfd];
1470 // actual code will need to loop here and realloc etc. 1505 // actual code will need to loop here and realloc etc.
1471 adns_beforepoll (ads, fds, &amp;nfd, &amp;timeout, timeval_from (ev_time ())); 1506 adns_beforepoll (ads, fds, &amp;nfd, &amp;timeout, timeval_from (ev_time ()));
1472 1507
1473 /* the callback is illegal, but won't be called as we stop during check */ 1508 /* the callback is illegal, but won't be called as we stop during check */
1474 ev_timer_init (&amp;tw, 0, timeout * 1e-3); 1509 ev_timer_init (&amp;tw, 0, timeout * 1e-3);
… …
1800 1835
1801</div> 1836</div>
1802<h1 id="MACRO_MAGIC">MACRO MAGIC</h1> 1837<h1 id="MACRO_MAGIC">MACRO MAGIC</h1>
1803<div id="MACRO_MAGIC_CONTENT"> 1838<div id="MACRO_MAGIC_CONTENT">
1804<p>Libev can be compiled with a variety of options, the most fundemantal is 1839<p>Libev can be compiled with a variety of options, the most fundemantal is
1805<code>EV_MULTIPLICITY</code>. This option determines wether (most) functions and 1840<code>EV_MULTIPLICITY</code>. This option determines whether (most) functions and
1806callbacks have an initial <code>struct ev_loop *</code> argument.</p> 1841callbacks have an initial <code>struct ev_loop *</code> argument.</p>
1807<p>To make it easier to write programs that cope with either variant, the 1842<p>To make it easier to write programs that cope with either variant, the
1808following macros are defined:</p> 1843following macros are defined:</p>
1809<dl> 1844<dl>
1810 <dt><code>EV_A</code>, <code>EV_A_</code></dt> 1845 <dt><code>EV_A</code>, <code>EV_A_</code></dt>
… …
1840 <p>Similar to the other two macros, this gives you the value of the default 1875 <p>Similar to the other two macros, this gives you the value of the default
1841loop, if multiple loops are supported (&quot;ev loop default&quot;).</p> 1876loop, if multiple loops are supported (&quot;ev loop default&quot;).</p>
1842 </dd> 1877 </dd>
1843</dl> 1878</dl>
1844<p>Example: Declare and initialise a check watcher, utilising the above 1879<p>Example: Declare and initialise a check watcher, utilising the above
1845macros so it will work regardless of wether multiple loops are supported 1880macros so it will work regardless of whether multiple loops are supported
1846or not.</p> 1881or not.</p>
1847<pre> static void 1882<pre> static void
1848 check_cb (EV_P_ ev_timer *w, int revents) 1883 check_cb (EV_P_ ev_timer *w, int revents)
1849 { 1884 {
1850 ev_check_stop (EV_A_ w); 1885 ev_check_stop (EV_A_ w);
… …
2078will have the <code>struct ev_loop *</code> as first argument, and you can create 2113will have the <code>struct ev_loop *</code> as first argument, and you can create
2079additional independent event loops. Otherwise there will be no support 2114additional independent event loops. Otherwise there will be no support
2080for multiple event loops and there is no first event loop pointer 2115for multiple event loops and there is no first event loop pointer
2081argument. Instead, all functions act on the single default loop.</p> 2116argument. Instead, all functions act on the single default loop.</p>
2082 </dd> 2117 </dd>
2118 <dt>EV_MINPRI</dt>
2119 <dt>EV_MAXPRI</dt>
2120 <dd>
2121 <p>The range of allowed priorities. <code>EV_MINPRI</code> must be smaller or equal to
2122<code>EV_MAXPRI</code>, but otherwise there are no non-obvious limitations. You can
2123provide for more priorities by overriding those symbols (usually defined
2124to be <code>-2</code> and <code>2</code>, respectively).</p>
2125 <p>When doing priority-based operations, libev usually has to linearly search
2126all the priorities, so having many of them (hundreds) uses a lot of space
2127and time, so using the defaults of five priorities (-2 .. +2) is usually
2128fine.</p>
2129 <p>If your embedding app does not need any priorities, defining these both to
2130<code>0</code> will save some memory and cpu.</p>
2131 </dd>
2083 <dt>EV_PERIODIC_ENABLE</dt> 2132 <dt>EV_PERIODIC_ENABLE</dt>
2084 <dd> 2133 <dd>
2085 <p>If undefined or defined to be <code>1</code>, then periodic timers are supported. If 2134 <p>If undefined or defined to be <code>1</code>, then periodic timers are supported. If
2135defined to be <code>0</code>, then they are not. Disabling them saves a few kB of
2136code.</p>
2137 </dd>
2138 <dt>EV_IDLE_ENABLE</dt>
2139 <dd>
2140 <p>If undefined or defined to be <code>1</code>, then idle watchers are supported. If
2086defined to be <code>0</code>, then they are not. Disabling them saves a few kB of 2141defined to be <code>0</code>, then they are not. Disabling them saves a few kB of
2087code.</p> 2142code.</p>
2088 </dd> 2143 </dd>
2089 <dt>EV_EMBED_ENABLE</dt> 2144 <dt>EV_EMBED_ENABLE</dt>
2090 <dd> 2145 <dd>
… …
2187libev will be explained. For complexity discussions about backends see the 2242libev will be explained. For complexity discussions about backends see the
2188documentation for <code>ev_default_init</code>.</p> 2243documentation for <code>ev_default_init</code>.</p>
2189 <p> 2244 <p>
2190 <dl> 2245 <dl>
2191 <dt>Starting and stopping timer/periodic watchers: O(log skipped_other_timers)</dt> 2246 <dt>Starting and stopping timer/periodic watchers: O(log skipped_other_timers)</dt>
2247 <dd>
2248 <p>This means that, when you have a watcher that triggers in one hour and
2249there are 100 watchers that would trigger before that then inserting will
2250have to skip those 100 watchers.</p>
2251 </dd>
2192 <dt>Changing timer/periodic watchers (by autorepeat, again): O(log skipped_other_timers)</dt> 2252 <dt>Changing timer/periodic watchers (by autorepeat, again): O(log skipped_other_timers)</dt>
2253 <dd>
2254 <p>That means that for changing a timer costs less than removing/adding them
2255as only the relative motion in the event queue has to be paid for.</p>
2256 </dd>
2193 <dt>Starting io/check/prepare/idle/signal/child watchers: O(1)</dt> 2257 <dt>Starting io/check/prepare/idle/signal/child watchers: O(1)</dt>
2258 <dd>
2259 <p>These just add the watcher into an array or at the head of a list. If
2260the array needs to be extended libev needs to realloc and move the whole
2261array, but this happen asymptotically less and less with more watchers,
2262thus amortised O(1).</p>
2263 </dd>
2194 <dt>Stopping check/prepare/idle watchers: O(1)</dt> 2264 <dt>Stopping check/prepare/idle watchers: O(1)</dt>
2195 <dt>Stopping an io/signal/child watcher: O(number_of_watchers_for_this_(fd/signal/pid % EV_PID_HASHSIZE))</dt> 2265 <dt>Stopping an io/signal/child watcher: O(number_of_watchers_for_this_(fd/signal/pid % EV_PID_HASHSIZE))</dt>
2266 <dd>
2267 <p>These watchers are stored in lists then need to be walked to find the
2268correct watcher to remove. The lists are usually short (you don't usually
2269have many watchers waiting for the same fd or signal).</p>
2270 </dd>
2196 <dt>Finding the next timer per loop iteration: O(1)</dt> 2271 <dt>Finding the next timer per loop iteration: O(1)</dt>
2197 <dt>Each change on a file descriptor per loop iteration: O(number_of_watchers_for_this_fd)</dt> 2272 <dt>Each change on a file descriptor per loop iteration: O(number_of_watchers_for_this_fd)</dt>
2273 <dd>
2274 <p>A change means an I/O watcher gets started or stopped, which requires
2275libev to recalculate its status (and possibly tell the kernel).</p>
2276 </dd>
2198 <dt>Activating one watcher: O(1)</dt> 2277 <dt>Activating one watcher: O(1)</dt>
2278 <dt>Priority handling: O(number_of_priorities)</dt>
2279 <dd>
2280 <p>Priorities are implemented by allocating some space for each
2281priority. When doing priority-based operations, libev usually has to
2282linearly search all the priorities.</p>
2283 </dd>
2199 </dl> 2284 </dl>
2200 </p> 2285 </p>
2201 2286
2202 2287
2203 2288

Diff Legend

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