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

Comparing libev/ev.html (file contents):
Revision 1.21 by root, Mon Nov 12 09:01:12 2007 UTC vs.
Revision 1.25 by root, Mon Nov 12 18:49:17 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="Mon Nov 12 10:01:12 2007" /> 9 <meta name="created" content="Mon Nov 12 19:49:15 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 -->
… …
35<li><a href="#code_ev_idle_code_when_you_ve_got_no"><code>ev_idle</code> - when you've got nothing better to do</a></li> 35<li><a href="#code_ev_idle_code_when_you_ve_got_no"><code>ev_idle</code> - when you've got nothing better to do</a></li>
36<li><a href="#code_ev_prepare_code_and_code_ev_che"><code>ev_prepare</code> and <code>ev_check</code> - customise your event loop</a></li> 36<li><a href="#code_ev_prepare_code_and_code_ev_che"><code>ev_prepare</code> and <code>ev_check</code> - customise your event loop</a></li>
37</ul> 37</ul>
38</li> 38</li>
39<li><a href="#OTHER_FUNCTIONS">OTHER FUNCTIONS</a></li> 39<li><a href="#OTHER_FUNCTIONS">OTHER FUNCTIONS</a></li>
40<li><a href="#LIBEVENT_EMULATION">LIBEVENT EMULATION</a></li>
41<li><a href="#C_SUPPORT">C++ SUPPORT</a></li>
40<li><a href="#AUTHOR">AUTHOR</a> 42<li><a href="#AUTHOR">AUTHOR</a>
41</li> 43</li>
42</ul><hr /> 44</ul><hr />
43<!-- INDEX END --> 45<!-- INDEX END -->
44 46
… …
440<p>I/O watchers check whether a file descriptor is readable or writable 442<p>I/O watchers check whether a file descriptor is readable or writable
441in each iteration of the event loop (This behaviour is called 443in each iteration of the event loop (This behaviour is called
442level-triggering because you keep receiving events as long as the 444level-triggering because you keep receiving events as long as the
443condition persists. Remember you can stop the watcher if you don't want to 445condition persists. Remember you can stop the watcher if you don't want to
444act on the event and neither want to receive future events).</p> 446act on the event and neither want to receive future events).</p>
445<p>In general you can register as many read and/or write event watchers oer 447<p>In general you can register as many read and/or write event watchers per
446fd as you want (as long as you don't confuse yourself). Setting all file 448fd as you want (as long as you don't confuse yourself). Setting all file
447descriptors to non-blocking mode is also usually a good idea (but not 449descriptors to non-blocking mode is also usually a good idea (but not
448required if you know what you are doing).</p> 450required if you know what you are doing).</p>
449<p>You have to be careful with dup'ed file descriptors, though. Some backends 451<p>You have to be careful with dup'ed file descriptors, though. Some backends
450(the linux epoll backend is a notable example) cannot handle dup'ed file 452(the linux epoll backend is a notable example) cannot handle dup'ed file
451descriptors correctly if you register interest in two or more fds pointing 453descriptors correctly if you register interest in two or more fds pointing
452to the same file/socket etc. description.</p> 454to the same file/socket etc. description (that is, they share the same
455underlying &quot;file open&quot;).</p>
453<p>If you must do this, then force the use of a known-to-be-good backend 456<p>If you must do this, then force the use of a known-to-be-good backend
454(at the time of this writing, this includes only EVMETHOD_SELECT and 457(at the time of this writing, this includes only EVMETHOD_SELECT and
455EVMETHOD_POLL).</p> 458EVMETHOD_POLL).</p>
456<dl> 459<dl>
457 <dt>ev_io_init (ev_io *, callback, int fd, int events)</dt> 460 <dt>ev_io_init (ev_io *, callback, int fd, int events)</dt>
… …
467<h2 id="code_ev_timer_code_relative_and_opti"><code>ev_timer</code> - relative and optionally recurring timeouts</h2> 470<h2 id="code_ev_timer_code_relative_and_opti"><code>ev_timer</code> - relative and optionally recurring timeouts</h2>
468<div id="code_ev_timer_code_relative_and_opti-2"> 471<div id="code_ev_timer_code_relative_and_opti-2">
469<p>Timer watchers are simple relative timers that generate an event after a 472<p>Timer watchers are simple relative timers that generate an event after a
470given time, and optionally repeating in regular intervals after that.</p> 473given time, and optionally repeating in regular intervals after that.</p>
471<p>The timers are based on real time, that is, if you register an event that 474<p>The timers are based on real time, that is, if you register an event that
472times out after an hour and youreset your system clock to last years 475times out after an hour and you reset your system clock to last years
473time, it will still time out after (roughly) and hour. &quot;Roughly&quot; because 476time, it will still time out after (roughly) and hour. &quot;Roughly&quot; because
474detecting time jumps is hard, and soem inaccuracies are unavoidable (the 477detecting time jumps is hard, and soem inaccuracies are unavoidable (the
475monotonic clock option helps a lot here).</p> 478monotonic clock option helps a lot here).</p>
476<p>The relative timeouts are calculated relative to the <code>ev_now ()</code> 479<p>The relative timeouts are calculated relative to the <code>ev_now ()</code>
477time. This is usually the right thing as this timestamp refers to the time 480time. This is usually the right thing as this timestamp refers to the time
478of the event triggering whatever timeout you are modifying/starting. If 481of the event triggering whatever timeout you are modifying/starting. If
479you suspect event processing to be delayed and you *need* to base the timeout 482you suspect event processing to be delayed and you *need* to base the timeout
480ion the current time, use something like this to adjust for this:</p> 483on the current time, use something like this to adjust for this:</p>
481<pre> ev_timer_set (&amp;timer, after + ev_now () - ev_time (), 0.); 484<pre> ev_timer_set (&amp;timer, after + ev_now () - ev_time (), 0.);
482 485
483</pre> 486</pre>
484<dl> 487<dl>
485 <dt>ev_timer_init (ev_timer *, callback, ev_tstamp after, ev_tstamp repeat)</dt> 488 <dt>ev_timer_init (ev_timer *, callback, ev_tstamp after, ev_tstamp repeat)</dt>
… …
490timer will automatically be configured to trigger again <code>repeat</code> seconds 493timer will automatically be configured to trigger again <code>repeat</code> seconds
491later, again, and again, until stopped manually.</p> 494later, again, and again, until stopped manually.</p>
492 <p>The timer itself will do a best-effort at avoiding drift, that is, if you 495 <p>The timer itself will do a best-effort at avoiding drift, that is, if you
493configure a timer to trigger every 10 seconds, then it will trigger at 496configure a timer to trigger every 10 seconds, then it will trigger at
494exactly 10 second intervals. If, however, your program cannot keep up with 497exactly 10 second intervals. If, however, your program cannot keep up with
495the timer (ecause it takes longer than those 10 seconds to do stuff) the 498the timer (because it takes longer than those 10 seconds to do stuff) the
496timer will not fire more than once per event loop iteration.</p> 499timer will not fire more than once per event loop iteration.</p>
497 </dd> 500 </dd>
498 <dt>ev_timer_again (loop)</dt> 501 <dt>ev_timer_again (loop)</dt>
499 <dd> 502 <dd>
500 <p>This will act as if the timer timed out and restart it again if it is 503 <p>This will act as if the timer timed out and restart it again if it is
… …
586 <p>It must return the next time to trigger, based on the passed time value 589 <p>It must return the next time to trigger, based on the passed time value
587(that is, the lowest time value larger than to the second argument). It 590(that is, the lowest time value larger than to the second argument). It
588will usually be called just before the callback will be triggered, but 591will usually be called just before the callback will be triggered, but
589might be called at other times, too.</p> 592might be called at other times, too.</p>
590 <p>NOTE: <i>This callback must always return a time that is later than the 593 <p>NOTE: <i>This callback must always return a time that is later than the
591passed <code>now</code> value</i>. Not even <code>now</code> itself will do, it must be larger.</p> 594passed <code>now</code> value</i>. Not even <code>now</code> itself will do, it <i>must</i> be larger.</p>
592 <p>This can be used to create very complex timers, such as a timer that 595 <p>This can be used to create very complex timers, such as a timer that
593triggers on each midnight, local time. To do this, you would calculate the 596triggers on each midnight, local time. To do this, you would calculate the
594next midnight after <code>now</code> and return the timestamp value for this. How you do this 597next midnight after <code>now</code> and return the timestamp value for this. How
595is, again, up to you (but it is not trivial).</p> 598you do this is, again, up to you (but it is not trivial, which is the main
599reason I omitted it as an example).</p>
596 </dd> 600 </dd>
597 </dl> 601 </dl>
598 </p> 602 </p>
599 </dd> 603 </dd>
600 <dt>ev_periodic_again (loop, ev_periodic *)</dt> 604 <dt>ev_periodic_again (loop, ev_periodic *)</dt>
… …
673 677
674</div> 678</div>
675<h2 id="code_ev_prepare_code_and_code_ev_che"><code>ev_prepare</code> and <code>ev_check</code> - customise your event loop</h2> 679<h2 id="code_ev_prepare_code_and_code_ev_che"><code>ev_prepare</code> and <code>ev_check</code> - customise your event loop</h2>
676<div id="code_ev_prepare_code_and_code_ev_che-2"> 680<div id="code_ev_prepare_code_and_code_ev_che-2">
677<p>Prepare and check watchers are usually (but not always) used in tandem: 681<p>Prepare and check watchers are usually (but not always) used in tandem:
678Prepare watchers get invoked before the process blocks and check watchers 682prepare watchers get invoked before the process blocks and check watchers
679afterwards.</p> 683afterwards.</p>
680<p>Their main purpose is to integrate other event mechanisms into libev. This 684<p>Their main purpose is to integrate other event mechanisms into libev. This
681could be used, for example, to track variable changes, implement your own 685could be used, for example, to track variable changes, implement your own
682watchers, integrate net-snmp or a coroutine library and lots more.</p> 686watchers, integrate net-snmp or a coroutine library and lots more.</p>
683<p>This is done by examining in each prepare call which file descriptors need 687<p>This is done by examining in each prepare call which file descriptors need
684to be watched by the other library, registering <code>ev_io</code> watchers for 688to be watched by the other library, registering <code>ev_io</code> watchers for
685them and starting an <code>ev_timer</code> watcher for any timeouts (many libraries 689them and starting an <code>ev_timer</code> watcher for any timeouts (many libraries
686provide just this functionality). Then, in the check watcher you check for 690provide just this functionality). Then, in the check watcher you check for
687any events that occured (by checking the pending status of all watchers 691any events that occured (by checking the pending status of all watchers
688and stopping them) and call back into the library. The I/O and timer 692and stopping them) and call back into the library. The I/O and timer
689callbacks will never actually be called (but must be valid neverthelles, 693callbacks will never actually be called (but must be valid nevertheless,
690because you never know, you know?).</p> 694because you never know, you know?).</p>
691<p>As another example, the Perl Coro module uses these hooks to integrate 695<p>As another example, the Perl Coro module uses these hooks to integrate
692coroutines into libev programs, by yielding to other active coroutines 696coroutines into libev programs, by yielding to other active coroutines
693during each prepare and only letting the process block if no coroutines 697during each prepare and only letting the process block if no coroutines
694are ready to run (its actually more complicated, it only runs coroutines 698are ready to run (it's actually more complicated: it only runs coroutines
695with priority higher than the event loop and one lower priority once, 699with priority higher than or equal to the event loop and one coroutine
696using idle watchers to keep the event loop from blocking if lower-priority 700of lower priority, but only once, using idle watchers to keep the event
697coroutines exist, thus mapping low-priority coroutines to idle/background 701loop from blocking if lower-priority coroutines are active, thus mapping
698tasks).</p> 702low-priority coroutines to idle/background tasks).</p>
699<dl> 703<dl>
700 <dt>ev_prepare_init (ev_prepare *, callback)</dt> 704 <dt>ev_prepare_init (ev_prepare *, callback)</dt>
701 <dt>ev_check_init (ev_check *, callback)</dt> 705 <dt>ev_check_init (ev_check *, callback)</dt>
702 <dd> 706 <dd>
703 <p>Initialises and configures the prepare or check watcher - they have no 707 <p>Initialises and configures the prepare or check watcher - they have no
… …
714 <dt>ev_once (loop, int fd, int events, ev_tstamp timeout, callback)</dt> 718 <dt>ev_once (loop, int fd, int events, ev_tstamp timeout, callback)</dt>
715 <dd> 719 <dd>
716 <p>This function combines a simple timer and an I/O watcher, calls your 720 <p>This function combines a simple timer and an I/O watcher, calls your
717callback on whichever event happens first and automatically stop both 721callback on whichever event happens first and automatically stop both
718watchers. This is useful if you want to wait for a single event on an fd 722watchers. This is useful if you want to wait for a single event on an fd
719or timeout without havign to allocate/configure/start/stop/free one or 723or timeout without having to allocate/configure/start/stop/free one or
720more watchers yourself.</p> 724more watchers yourself.</p>
721 <p>If <code>fd</code> is less than 0, then no I/O watcher will be started and events 725 <p>If <code>fd</code> is less than 0, then no I/O watcher will be started and events
722is being ignored. Otherwise, an <code>ev_io</code> watcher for the given <code>fd</code> and 726is being ignored. Otherwise, an <code>ev_io</code> watcher for the given <code>fd</code> and
723<code>events</code> set will be craeted and started.</p> 727<code>events</code> set will be craeted and started.</p>
724 <p>If <code>timeout</code> is less than 0, then no timeout watcher will be 728 <p>If <code>timeout</code> is less than 0, then no timeout watcher will be
725started. Otherwise an <code>ev_timer</code> watcher with after = <code>timeout</code> (and 729started. Otherwise an <code>ev_timer</code> watcher with after = <code>timeout</code> (and
726repeat = 0) will be started. While <code>0</code> is a valid timeout, it is of 730repeat = 0) will be started. While <code>0</code> is a valid timeout, it is of
727dubious value.</p> 731dubious value.</p>
728 <p>The callback has the type <code>void (*cb)(int revents, void *arg)</code> and gets 732 <p>The callback has the type <code>void (*cb)(int revents, void *arg)</code> and gets
729passed an events set like normal event callbacks (with a combination of 733passed an <code>revents</code> set like normal event callbacks (a combination of
730<code>EV_ERROR</code>, <code>EV_READ</code>, <code>EV_WRITE</code> or <code>EV_TIMEOUT</code>) and the <code>arg</code> 734<code>EV_ERROR</code>, <code>EV_READ</code>, <code>EV_WRITE</code> or <code>EV_TIMEOUT</code>) and the <code>arg</code>
731value passed to <code>ev_once</code>:</p> 735value passed to <code>ev_once</code>:</p>
732<pre> static void stdin_ready (int revents, void *arg) 736<pre> static void stdin_ready (int revents, void *arg)
733 { 737 {
734 if (revents &amp; EV_TIMEOUT) 738 if (revents &amp; EV_TIMEOUT)
… …
757 <p>Feed an event as if the given signal occured (loop must be the default loop!).</p> 761 <p>Feed an event as if the given signal occured (loop must be the default loop!).</p>
758 </dd> 762 </dd>
759</dl> 763</dl>
760 764
761</div> 765</div>
766<h1 id="LIBEVENT_EMULATION">LIBEVENT EMULATION</h1><p><a href="#TOP" class="toplink">Top</a></p>
767<div id="LIBEVENT_EMULATION_CONTENT">
768<p>TBD.</p>
769
770</div>
771<h1 id="C_SUPPORT">C++ SUPPORT</h1><p><a href="#TOP" class="toplink">Top</a></p>
772<div id="C_SUPPORT_CONTENT">
773<p>TBD.</p>
774
775</div>
762<h1 id="AUTHOR">AUTHOR</h1><p><a href="#TOP" class="toplink">Top</a></p> 776<h1 id="AUTHOR">AUTHOR</h1><p><a href="#TOP" class="toplink">Top</a></p>
763<div id="AUTHOR_CONTENT"> 777<div id="AUTHOR_CONTENT">
764<p>Marc Lehmann &lt;libev@schmorp.de&gt;.</p> 778<p>Marc Lehmann &lt;libev@schmorp.de&gt;.</p>
765 779
766</div> 780</div>

Diff Legend

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