[ViewVC] Diff of: cvs/libev/ev.pod

Comparing libev/ev.pod (file contents):
Revision 1.210 by root, Thu Oct 30 08:09:30 2008 UTC vs.
Revision 1.273 by root, Tue Nov 24 14:46:59 2009 UTC

 =head2 EXAMPLE PROGRAM
    // a single header file is required
    #include <ev.h>
+   #include <stdio.h> // for puts
    // every watcher type has its own typedef'd struct
    // with the name ev_TYPE
    ev_io stdin_watcher;
    ev_timer timeout_watcher;
    int
    main (void)
    {
      // use the default event loop unless you have special needs
-     ev_loop *loop = ev_default_loop (0);
+     struct ev_loop *loop = ev_default_loop (0);
      // initialise an io watcher, then start it
      // this one will watch for stdin to become readable
      ev_io_init (&stdin_watcher, stdin_cb, /*STDIN_FILENO*/ 0, EV_READ);
      ev_io_start (loop, &stdin_watcher);
      // unloop was called, so exit
      return 0;
    }
-=head1 DESCRIPTION
+=head1 ABOUT THIS DOCUMENT
+This document documents the libev software package.
 The newest version of this document is also available as an html-formatted
 web page you might find easier to navigate when reading it for the first
 time: L<http://pod.tst.eu/http://cvs.schmorp.de/libev/ev.pod>.
+While this document tries to be as complete as possible in documenting
+libev, its usage and the rationale behind its design, it is not a tutorial
+on event-based programming, nor will it introduce event-based programming
+with libev.
+Familarity with event based programming techniques in general is assumed
+throughout this document.
+=head1 ABOUT LIBEV
 Libev is an event loop: you register interest in certain events (such as a
 file descriptor being readable or a timeout occurring), and it will manage
 these event sources and provide your program with events.
 =head2 FEATURES
 Libev supports C<select>, C<poll>, the Linux-specific C<epoll>, the
 BSD-specific C<kqueue> and the Solaris-specific event port mechanisms
 for file descriptor events (C<ev_io>), the Linux C<inotify> interface
-(for C<ev_stat>), relative timers (C<ev_timer>), absolute timers
+(for C<ev_stat>), Linux eventfd/signalfd (for faster and cleaner
-with customised rescheduling (C<ev_periodic>), synchronous signals
+inter-thread wakeup (C<ev_async>)/signal handling (C<ev_signal>)) relative
-(C<ev_signal>), process status change events (C<ev_child>), and event
+timers (C<ev_timer>), absolute timers with customised rescheduling
-watchers dealing with the event loop mechanism itself (C<ev_idle>,
+(C<ev_periodic>), synchronous signals (C<ev_signal>), process status
-C<ev_embed>, C<ev_prepare> and C<ev_check> watchers) as well as
+change events (C<ev_child>), and event watchers dealing with the event
-file watchers (C<ev_stat>) and even limited support for fork events
+loop mechanism itself (C<ev_idle>, C<ev_embed>, C<ev_prepare> and
-(C<ev_fork>).
+C<ev_check> watchers) as well as file watchers (C<ev_stat>) and even
+limited support for fork events (C<ev_fork>).
 It also is quite fast (see this
 L<benchmark|http://libev.schmorp.de/bench.html> comparing it to libevent
 for example).
 name C<loop> (which is always of type C<ev_loop *>) will not have
 this argument.
 =head2 TIME REPRESENTATION
-Libev represents time as a single floating point number, representing the
+Libev represents time as a single floating point number, representing
-(fractional) number of seconds since the (POSIX) epoch (somewhere near
+the (fractional) number of seconds since the (POSIX) epoch (somewhere
-the beginning of 1970, details are complicated, don't ask). This type is
+near the beginning of 1970, details are complicated, don't ask). This
-called C<ev_tstamp>, which is what you should use too. It usually aliases
+type is called C<ev_tstamp>, which is what you should use too. It usually
-to the C<double> type in C, and when you need to do any calculations on
+aliases to the C<double> type in C. When you need to do any calculations
-it, you should treat it as some floating point value. Unlike the name
+on it, you should treat it as some floating point value. Unlike the name
 component C<stamp> might indicate, it is also used for time differences
 throughout libev.
 =head1 ERROR HANDLING
 flag.
 This flag setting cannot be overridden or specified in the C<LIBEV_FLAGS>
 environment variable.
+=item C<EVFLAG_NOINOTIFY>
+When this flag is specified, then libev will not attempt to use the
+I<inotify> API for it's C<ev_stat> watchers. Apart from debugging and
+testing, this flag can be useful to conserve inotify file descriptors, as
+otherwise each loop using C<ev_stat> watchers consumes one inotify handle.
+=item C<EVFLAG_NOSIGFD>
+When this flag is specified, then libev will not attempt to use the
+I<signalfd> API for it's C<ev_signal> (and C<ev_child>) watchers. This is
+probably only useful to work around any bugs in libev. Consequently, this
+flag might go away once the signalfd functionality is considered stable,
+so it's useful mostly in environment variables and not in program code.
 =item C<EVBACKEND_SELECT>  (value 1, portable select backend)
 This is your standard select(2) backend. Not I<completely> standard, as
 libev tries to roll its own fd_set with no limits on the number of fds,
 but if that fails, expect a fairly low limit on the number of fds when
 This backend maps C<EV_READ> to C<POLLIN | POLLERR | POLLHUP>, and
 C<EV_WRITE> to C<POLLOUT | POLLERR | POLLHUP>.
 =item C<EVBACKEND_EPOLL>   (value 4, Linux)
+Use the linux-specific epoll(7) interface (for both pre- and post-2.6.9
+kernels).
 For few fds, this backend is a bit little slower than poll and select,
 but it scales phenomenally better. While poll and select usually scale
 like O(total_fds) where n is the total number of fds (or the highest fd),
 epoll scales either O(1) or O(active_fds).
 i.e. keep at least one watcher active per fd at all times. Stopping and
 starting a watcher (without re-setting it) also usually doesn't cause
 extra overhead. A fork can both result in spurious notifications as well
 as in libev having to destroy and recreate the epoll object, which can
 take considerable time and thus should be avoided.
+All this means that, in practice, C<EVBACKEND_SELECT> can be as fast or
+faster than epoll for maybe up to a hundred file descriptors, depending on
+the usage. So sad.
 While nominally embeddable in other event loops, this feature is broken in
 all kernel versions tested so far.
 This backend maps C<EV_READ> and C<EV_WRITE> in the same way as
 While nominally embeddable in other event loops, this doesn't work
 everywhere, so you might need to test for this. And since it is broken
 almost everywhere, you should only use it when you have a lot of sockets
 (for which it usually works), by embedding it into another event loop
-(e.g. C<EVBACKEND_SELECT> or C<EVBACKEND_POLL>) and, did I mention it,
+(e.g. C<EVBACKEND_SELECT> or C<EVBACKEND_POLL> (but C<poll> is of course
-using it only for sockets.
+also broken on OS X)) and, did I mention it, using it only for sockets.
 This backend maps C<EV_READ> into an C<EVFILT_READ> kevent with
 C<NOTE_EOF>, and C<EV_WRITE> into an C<EVFILT_WRITE> kevent with
 C<NOTE_EOF>.
 It is definitely not recommended to use this flag.
 =back
-If one or more of these are or'ed into the flags value, then only these
+If one or more of the backend flags are or'ed into the flags value,
-backends will be tried (in the reverse order as listed here). If none are
+then only these backends will be tried (in the reverse order as listed
-specified, all backends in C<ev_recommended_backends ()> will be tried.
+here). If none are specified, all backends in C<ev_recommended_backends
+()> will be tried.
 Example: This is the most typical usage.
    if (!ev_default_loop (0))
      fatal ("could not initialise libev, bad $LIBEV_FLAGS in environment?");
 as signal and child watchers) would need to be stopped manually.
 In general it is not advisable to call this function except in the
 rare occasion where you really need to free e.g. the signal handling
 pipe fds. If you need dynamically allocated loops it is better to use
-C<ev_loop_new> and C<ev_loop_destroy>).
+C<ev_loop_new> and C<ev_loop_destroy>.
 =item ev_loop_destroy (loop)
 Like C<ev_default_destroy>, but destroys an event loop created by an
 earlier call to C<ev_loop_new>.
 This value can sometimes be useful as a generation counter of sorts (it
 "ticks" the number of loop iterations), as it roughly corresponds with
 C<ev_prepare> and C<ev_check> calls.
+=item unsigned int ev_loop_depth (loop)
+Returns the number of times C<ev_loop> was entered minus the number of
+times C<ev_loop> was exited, in other words, the recursion depth.
+Outside C<ev_loop>, this number is zero. In a callback, this number is
+C<1>, unless C<ev_loop> was invoked recursively (or from another thread),
+in which case it is higher.
+Leaving C<ev_loop> abnormally (setjmp/longjmp, cancelling the thread
+etc.), doesn't count as exit.
 =item unsigned int ev_backend (loop)
 Returns one of the C<EVBACKEND_*> flags indicating the event backend in
 use.
 This function is rarely useful, but when some event callback runs for a
 very long time without entering the event loop, updating libev's idea of
 the current time is a good idea.
-See also "The special problem of time updates" in the C<ev_timer> section.
+See also L<The special problem of time updates> in the C<ev_timer> section.
+=item ev_suspend (loop)
+=item ev_resume (loop)
+These two functions suspend and resume a loop, for use when the loop is
+not used for a while and timeouts should not be processed.
+A typical use case would be an interactive program such as a game:  When
+the user presses C<^Z> to suspend the game and resumes it an hour later it
+would be best to handle timeouts as if no time had actually passed while
+the program was suspended. This can be achieved by calling C<ev_suspend>
+in your C<SIGTSTP> handler, sending yourself a C<SIGSTOP> and calling
+C<ev_resume> directly afterwards to resume timer processing.
+Effectively, all C<ev_timer> watchers will be delayed by the time spend
+between C<ev_suspend> and C<ev_resume>, and all C<ev_periodic> watchers
+will be rescheduled (that is, they will lose any events that would have
+occured while suspended).
+After calling C<ev_suspend> you B<must not> call I<any> function on the
+given loop other than C<ev_resume>, and you B<must not> call C<ev_resume>
+without a previous call to C<ev_suspend>.
+Calling C<ev_suspend>/C<ev_resume> has the side effect of updating the
+event loop time (see C<ev_now_update>).
 =item ev_loop (loop, int flags)
 Finally, this is it, the event handler. This function usually is called
-after you initialised all your watchers and you want to start handling
+after you have initialised all your watchers and you want to start
-events.
+handling events.
 If the flags argument is specified as C<0>, it will not return until
 either no event watchers are active anymore or C<ev_unloop> was called.
 Please note that an explicit C<ev_unloop> is usually better than
 If you have a watcher you never unregister that should not keep C<ev_loop>
 from returning, call ev_unref() after starting, and ev_ref() before
 stopping it.
-As an example, libev itself uses this for its internal signal pipe: It is
+As an example, libev itself uses this for its internal signal pipe: It
-not visible to the libev user and should not keep C<ev_loop> from exiting
+is not visible to the libev user and should not keep C<ev_loop> from
-if no event watchers registered by it are active. It is also an excellent
+exiting if no event watchers registered by it are active. It is also an
-way to do this for generic recurring timers or from within third-party
+excellent way to do this for generic recurring timers or from within
-libraries. Just remember to I<unref after start> and I<ref before stop>
+third-party libraries. Just remember to I<unref after start> and I<ref
-(but only if the watcher wasn't active before, or was active before,
+before stop> (but only if the watcher wasn't active before, or was active
-respectively).
+before, respectively. Note also that libev might stop watchers itself
+(e.g. non-repeating timers) in which case you have to C<ev_ref>
+in the callback).
 Example: Create a signal watcher, but keep it from keeping C<ev_loop>
 running when nothing else is active.
    ev_signal exitsig;
 By setting a higher I<io collect interval> you allow libev to spend more
 time collecting I/O events, so you can handle more events per iteration,
 at the cost of increasing latency. Timeouts (both C<ev_periodic> and
 C<ev_timer>) will be not affected. Setting this to a non-null value will
-introduce an additional C<ev_sleep ()> call into most loop iterations.
+introduce an additional C<ev_sleep ()> call into most loop iterations. The
+sleep time ensures that libev will not poll for I/O events more often then
+once per this interval, on average.
 Likewise, by setting a higher I<timeout collect interval> you allow libev
 to spend more time collecting timeouts, at the expense of increased
 latency/jitter/inexactness (the watcher callback will be called
 later). C<ev_io> watchers will not be affected. Setting this to a non-null
 Many (busy) programs can usually benefit by setting the I/O collect
 interval to a value near C<0.1> or so, which is often enough for
 interactive servers (of course not for games), likewise for timeouts. It
 usually doesn't make much sense to set it to a lower value than C<0.01>,
-as this approaches the timing granularity of most systems.
+as this approaches the timing granularity of most systems. Note that if
+you do transactions with the outside world and you can't increase the
+parallelity, then this setting will limit your transaction rate (if you
+need to poll once per transaction and the I/O collect interval is 0.01,
+then you can't do more than 100 transations per second).
 Setting the I<timeout collect interval> can improve the opportunity for
 saving power, as the program will "bundle" timer callback invocations that
 are "near" in time together, by delaying some, thus reducing the number of
 times the process sleeps and wakes up again. Another useful technique to
 reduce iterations/wake-ups is to use C<ev_periodic> watchers and make sure
 they fire on, say, one-second boundaries only.
+Example: we only need 0.1s timeout granularity, and we wish not to poll
+more often than 100 times per second:
+   ev_set_timeout_collect_interval (EV_DEFAULT_UC_ 0.1);
+   ev_set_io_collect_interval (EV_DEFAULT_UC_ 0.01);
+=item ev_invoke_pending (loop)
+This call will simply invoke all pending watchers while resetting their
+pending state. Normally, C<ev_loop> does this automatically when required,
+but when overriding the invoke callback this call comes handy.
+=item int ev_pending_count (loop)
+Returns the number of pending watchers - zero indicates that no watchers
+are pending.
+=item ev_set_invoke_pending_cb (loop, void (*invoke_pending_cb)(EV_P))
+This overrides the invoke pending functionality of the loop: Instead of
+invoking all pending watchers when there are any, C<ev_loop> will call
+this callback instead. This is useful, for example, when you want to
+invoke the actual watchers inside another context (another thread etc.).
+If you want to reset the callback, use C<ev_invoke_pending> as new
+callback.
+=item ev_set_loop_release_cb (loop, void (*release)(EV_P), void (*acquire)(EV_P))
+Sometimes you want to share the same loop between multiple threads. This
+can be done relatively simply by putting mutex_lock/unlock calls around
+each call to a libev function.
+However, C<ev_loop> can run an indefinite time, so it is not feasible to
+wait for it to return. One way around this is to wake up the loop via
+C<ev_unloop> and C<av_async_send>, another way is to set these I<release>
+and I<acquire> callbacks on the loop.
+When set, then C<release> will be called just before the thread is
+suspended waiting for new events, and C<acquire> is called just
+afterwards.
+Ideally, C<release> will just call your mutex_unlock function, and
+C<acquire> will just call the mutex_lock function again.
+While event loop modifications are allowed between invocations of
+C<release> and C<acquire> (that's their only purpose after all), no
+modifications done will affect the event loop, i.e. adding watchers will
+have no effect on the set of file descriptors being watched, or the time
+waited. USe an C<ev_async> watcher to wake up C<ev_loop> when you want it
+to take note of any changes you made.
+In theory, threads executing C<ev_loop> will be async-cancel safe between
+invocations of C<release> and C<acquire>.
+See also the locking example in the C<THREADS> section later in this
+document.
+=item ev_set_userdata (loop, void *data)
+=item ev_userdata (loop)
+Set and retrieve a single C<void *> associated with a loop. When
+C<ev_set_userdata> has never been called, then C<ev_userdata> returns
+C<0.>
+These two functions can be used to associate arbitrary data with a loop,
+and are intended solely for the C<invoke_pending_cb>, C<release> and
+C<acquire> callbacks described above, but of course can be (ab-)used for
+any other purpose as well.
 =item ev_loop_verify (loop)
 This function only does something when C<EV_VERIFY> support has been
 compiled in, which is the default for non-minimal builds. It tries to go
 =item C<EV_ASYNC>
 The given async watcher has been asynchronously notified (see C<ev_async>).
+=item C<EV_CUSTOM>
+Not ever sent (or otherwise used) by libev itself, but can be freely used
+by libev users to signal watchers (e.g. via C<ev_feed_event>).
 =item C<EV_ERROR>
 An unspecified error has occurred, the watcher has been stopped. This might
 happen because the watcher could not be properly started because libev
 ran out of memory, a file descriptor was found to be closed or any other
 integer between C<EV_MAXPRI> (default: C<2>) and C<EV_MINPRI>
 (default: C<-2>). Pending watchers with higher priority will be invoked
 before watchers with lower priority, but priority will not keep watchers
 from being executed (except for C<ev_idle> watchers).
-This means that priorities are I<only> used for ordering callback
-invocation after new events have been received. This is useful, for
-example, to reduce latency after idling, or more often, to bind two
-watchers on the same event and make sure one is called first.
 If you need to suppress invocation when higher priority events are pending
 you need to look at C<ev_idle> watchers, which provide this functionality.
 You I<must not> change the priority of a watcher as long as it is active or
 pending.
-The default priority used by watchers when no priority has been set is
-always C<0>, which is supposed to not be too high and not be too low :).
 Setting a priority outside the range of C<EV_MINPRI> to C<EV_MAXPRI> is
 fine, as long as you do not mind that the priority value you query might
 or might not have been clamped to the valid range.
+The default priority used by watchers when no priority has been set is
+always C<0>, which is supposed to not be too high and not be too low :).
+See L<WATCHER PRIORITY MODELS>, below, for a more thorough treatment of
+priorities.
 =item ev_invoke (loop, ev_TYPE *watcher, int revents)
 Invoke the C<watcher> with the given C<loop> and C<revents>. Neither
 C<loop> nor C<revents> need to be valid as long as the watcher callback
 returns its C<revents> bitset (as if its callback was invoked). If the
 watcher isn't pending it does nothing and returns C<0>.
 Sometimes it can be useful to "poll" a watcher instead of waiting for its
 callback to be invoked, which can be accomplished with this function.
+=item ev_feed_event (struct ev_loop *, watcher *, int revents)
+Feeds the given event set into the event loop, as if the specified event
+had happened for the specified watcher (which must be a pointer to an
+initialised but not necessarily started event watcher). Obviously you must
+not free the watcher as long as it has pending events.
+Stopping the watcher, letting libev invoke it, or calling
+C<ev_clear_pending> will clear the pending event, even if the watcher was
+not started in the first place.
+See also C<ev_feed_fd_event> and C<ev_feed_signal_event> for related
+functions that do not need a watcher.
 =back
 =head2 ASSOCIATING CUSTOM DATA WITH A WATCHER
    #include <stddef.h>
    static void
    t1_cb (EV_P_ ev_timer *w, int revents)
    {
-     struct my_biggy big = (struct my_biggy *
+     struct my_biggy big = (struct my_biggy *)
        (((char *)w) - offsetof (struct my_biggy, t1));
    }
    static void
    t2_cb (EV_P_ ev_timer *w, int revents)
    {
-     struct my_biggy big = (struct my_biggy *
+     struct my_biggy big = (struct my_biggy *)
        (((char *)w) - offsetof (struct my_biggy, t2));
    }
+=head2 WATCHER PRIORITY MODELS
+Many event loops support I<watcher priorities>, which are usually small
+integers that influence the ordering of event callback invocation
+between watchers in some way, all else being equal.
+In libev, Watcher priorities can be set using C<ev_set_priority>. See its
+description for the more technical details such as the actual priority
+range.
+There are two common ways how these these priorities are being interpreted
+by event loops:
+In the more common lock-out model, higher priorities "lock out" invocation
+of lower priority watchers, which means as long as higher priority
+watchers receive events, lower priority watchers are not being invoked.
+The less common only-for-ordering model uses priorities solely to order
+callback invocation within a single event loop iteration: Higher priority
+watchers are invoked before lower priority ones, but they all get invoked
+before polling for new events.
+Libev uses the second (only-for-ordering) model for all its watchers
+except for idle watchers (which use the lock-out model).
+The rationale behind this is that implementing the lock-out model for
+watchers is not well supported by most kernel interfaces, and most event
+libraries will just poll for the same events again and again as long as
+their callbacks have not been executed, which is very inefficient in the
+common case of one high-priority watcher locking out a mass of lower
+priority ones.
+Static (ordering) priorities are most useful when you have two or more
+watchers handling the same resource: a typical usage example is having an
+C<ev_io> watcher to receive data, and an associated C<ev_timer> to handle
+timeouts. Under load, data might be received while the program handles
+other jobs, but since timers normally get invoked first, the timeout
+handler will be executed before checking for data. In that case, giving
+the timer a lower priority than the I/O watcher ensures that I/O will be
+handled first even under adverse conditions (which is usually, but not
+always, what you want).
+Since idle watchers use the "lock-out" model, meaning that idle watchers
+will only be executed when no same or higher priority watchers have
+received events, they can be used to implement the "lock-out" model when
+required.
+For example, to emulate how many other event libraries handle priorities,
+you can associate an C<ev_idle> watcher to each such watcher, and in
+the normal watcher callback, you just start the idle watcher. The real
+processing is done in the idle watcher callback. This causes libev to
+continously poll and process kernel event data for the watcher, but when
+the lock-out case is known to be rare (which in turn is rare :), this is
+workable.
+Usually, however, the lock-out model implemented that way will perform
+miserably under the type of load it was designed to handle. In that case,
+it might be preferable to stop the real watcher before starting the
+idle watcher, so the kernel will not have to process the event in case
+the actual processing will be delayed for considerable time.
+Here is an example of an I/O watcher that should run at a strictly lower
+priority than the default, and which should only process data when no
+other events are pending:
+   ev_idle idle; // actual processing watcher
+   ev_io io;     // actual event watcher
+   static void
+   io_cb (EV_P_ ev_io *w, int revents)
+   {
+     // stop the I/O watcher, we received the event, but
+     // are not yet ready to handle it.
+     ev_io_stop (EV_A_ w);
+     // start the idle watcher to ahndle the actual event.
+     // it will not be executed as long as other watchers
+     // with the default priority are receiving events.
+     ev_idle_start (EV_A_ &idle);
+   }
+   static void
+   idle_cb (EV_P_ ev_idle *w, int revents)
+   {
+     // actual processing
+     read (STDIN_FILENO, ...);
+     // have to start the I/O watcher again, as
+     // we have handled the event
+     ev_io_start (EV_P_ &io);
+   }
+   // initialisation
+   ev_idle_init (&idle, idle_cb);
+   ev_io_init (&io, io_cb, STDIN_FILENO, EV_READ);
+   ev_io_start (EV_DEFAULT_ &io);
+In the "real" world, it might also be beneficial to start a timer, so that
+low-priority connections can not be locked out forever under load. This
+enables your program to keep a lower latency for important connections
+during short periods of high load, while not completely locking out less
+important ones.
 =head1 WATCHER TYPES
 This section describes each watcher in detail, but will not repeat
 descriptors to non-blocking mode is also usually a good idea (but not
 required if you know what you are doing).
 If you cannot use non-blocking mode, then force the use of a
 known-to-be-good backend (at the time of this writing, this includes only
-C<EVBACKEND_SELECT> and C<EVBACKEND_POLL>).
+C<EVBACKEND_SELECT> and C<EVBACKEND_POLL>). The same applies to file
+descriptors for which non-blocking operation makes no sense (such as
+files) - libev doesn't guarentee any specific behaviour in that case.
 Another thing you have to watch out for is that it is quite easy to
 receive "spurious" readiness notifications, that is your callback might
 be called with C<EV_READ> but a subsequent C<read>(2) will actually block
 because there is no data. Not only are some backends known to create a
 year, it will still time out after (roughly) one hour. "Roughly" because
 detecting time jumps is hard, and some inaccuracies are unavoidable (the
 monotonic clock option helps a lot here).
 The callback is guaranteed to be invoked only I<after> its timeout has
-passed, but if multiple timers become ready during the same loop iteration
+passed (not I<at>, so on systems with very low-resolution clocks this
-then order of execution is undefined.
+might introduce a small delay). If multiple timers become ready during the
+same loop iteration then the ones with earlier time-out values are invoked
+before ones of the same priority with later time-out values (but this is
+no longer true when a callback calls C<ev_loop> recursively).
 =head3 Be smart about timeouts
 Many real-world problems involve some kind of timeout, usually for error
 recovery. A typical example is an HTTP request - if the other side hangs,
 C<after> argument to C<ev_timer_set>, and only ever use the C<repeat>
 member and C<ev_timer_again>.
 At start:
-   ev_timer_init (timer, callback);
+   ev_init (timer, callback);
    timer->repeat = 60.;
    ev_timer_again (loop, timer);
 Each time there is some activity:
      else
        {
          // callback was invoked, but there was some activity, re-arm
          // the watcher to fire in last_activity + 60, which is
          // guaranteed to be in the future, so "again" is positive:
-         w->again = timeout - now;
+         w->repeat = timeout - now;
          ev_timer_again (EV_A_ w);
        }
    }
 To summarise the callback: first calculate the real timeout (defined
 To start the timer, simply initialise the watcher and set C<last_activity>
 to the current time (meaning we just have some activity :), then call the
 callback, which will "do the right thing" and start the timer:
-   ev_timer_init (timer, callback);
+   ev_init (timer, callback);
    last_activity = ev_now (loop);
    callback (loop, timer, EV_TIMEOUT);
 And when there is some activity, simply store the current time in
 C<last_activity>, no libev calls at all:
 If the event loop is suspended for a long time, you can also force an
 update of the time returned by C<ev_now ()> by calling C<ev_now_update
 ()>.
+=head3 The special problems of suspended animation
+When you leave the server world it is quite customary to hit machines that
+can suspend/hibernate - what happens to the clocks during such a suspend?
+Some quick tests made with a Linux 2.6.28 indicate that a suspend freezes
+all processes, while the clocks (C<times>, C<CLOCK_MONOTONIC>) continue
+to run until the system is suspended, but they will not advance while the
+system is suspended. That means, on resume, it will be as if the program
+was frozen for a few seconds, but the suspend time will not be counted
+towards C<ev_timer> when a monotonic clock source is used. The real time
+clock advanced as expected, but if it is used as sole clocksource, then a
+long suspend would be detected as a time jump by libev, and timers would
+be adjusted accordingly.
+I would not be surprised to see different behaviour in different between
+operating systems, OS versions or even different hardware.
+The other form of suspend (job control, or sending a SIGSTOP) will see a
+time jump in the monotonic clocks and the realtime clock. If the program
+is suspended for a very long time, and monotonic clock sources are in use,
+then you can expect C<ev_timer>s to expire as the full suspension time
+will be counted towards the timers. When no monotonic clock source is in
+use, then libev will again assume a timejump and adjust accordingly.
+It might be beneficial for this latter case to call C<ev_suspend>
+and C<ev_resume> in code that handles C<SIGTSTP>, to at least get
+deterministic behaviour in this case (you can do nothing against
+C<SIGSTOP>).
 =head3 Watcher-Specific Functions and Data Members
 =over 4
 =item ev_timer_init (ev_timer *, callback, ev_tstamp after, ev_tstamp repeat)
 If the timer is started but non-repeating, stop it (as if it timed out).
 If the timer is repeating, either start it if necessary (with the
 C<repeat> value), or reset the running timer to the C<repeat> value.
-This sounds a bit complicated, see "Be smart about timeouts", above, for a
+This sounds a bit complicated, see L<Be smart about timeouts>, above, for a
 usage example.
+=item ev_timer_remaining (loop, ev_timer *)
+Returns the remaining time until a timer fires. If the timer is active,
+then this time is relative to the current event loop time, otherwise it's
+the timeout value currently configured.
+That is, after an C<ev_timer_set (w, 5, 7)>, C<ev_timer_remaining> returns
+C<5>. When the timer is started and one second passes, C<ev_timer_remain>
+will return C<4>. When the timer expires and is restarted, it will return
+roughly C<7> (likely slightly less as callback invocation takes some time,
+too), and so on.
 =item ev_tstamp repeat [read-write]
 The current C<repeat> value. Will be used each time the watcher times out
 or C<ev_timer_again> is called, and determines the next timeout (if any),
 =head2 C<ev_periodic> - to cron or not to cron?
 Periodic watchers are also timers of a kind, but they are very versatile
 (and unfortunately a bit complex).
-Unlike C<ev_timer>'s, they are not based on real time (or relative time)
+Unlike C<ev_timer>, periodic watchers are not based on real time (or
-but on wall clock time (absolute time). You can tell a periodic watcher
+relative time, the physical time that passes) but on wall clock time
-to trigger after some specific point in time. For example, if you tell a
+(absolute time, the thing you can read on your calender or clock). The
-periodic watcher to trigger in 10 seconds (by specifying e.g. C<ev_now ()
+difference is that wall clock time can run faster or slower than real
-+ 10.>, that is, an absolute time not a delay) and then reset your system
+time, and time jumps are not uncommon (e.g. when you adjust your
-clock to January of the previous year, then it will take more than year
+wrist-watch).
-to trigger the event (unlike an C<ev_timer>, which would still trigger
-roughly 10 seconds later as it uses a relative timeout).
+You can tell a periodic watcher to trigger after some specific point
+in time: for example, if you tell a periodic watcher to trigger "in 10
+seconds" (by specifying e.g. C<ev_now () + 10.>, that is, an absolute time
+not a delay) and then reset your system clock to January of the previous
+year, then it will take a year or more to trigger the event (unlike an
+C<ev_timer>, which would still trigger roughly 10 seconds after starting
+it, as it uses a relative timeout).
-C<ev_periodic>s can also be used to implement vastly more complex timers,
+C<ev_periodic> watchers can also be used to implement vastly more complex
-such as triggering an event on each "midnight, local time", or other
+timers, such as triggering an event on each "midnight, local time", or
-complicated rules.
+other complicated rules. This cannot be done with C<ev_timer> watchers, as
+those cannot react to time jumps.
 As with timers, the callback is guaranteed to be invoked only when the
-time (C<at>) has passed, but if multiple periodic timers become ready
+point in time where it is supposed to trigger has passed. If multiple
-during the same loop iteration, then order of execution is undefined.
+timers become ready during the same loop iteration then the ones with
+earlier time-out values are invoked before ones with later time-out values
+(but this is no longer true when a callback calls C<ev_loop> recursively).
 =head3 Watcher-Specific Functions and Data Members
 =over 4
-=item ev_periodic_init (ev_periodic *, callback, ev_tstamp at, ev_tstamp interval, reschedule_cb)
+=item ev_periodic_init (ev_periodic *, callback, ev_tstamp offset, ev_tstamp interval, reschedule_cb)
-=item ev_periodic_set (ev_periodic *, ev_tstamp after, ev_tstamp repeat, reschedule_cb)
+=item ev_periodic_set (ev_periodic *, ev_tstamp offset, ev_tstamp interval, reschedule_cb)
-Lots of arguments, lets sort it out... There are basically three modes of
+Lots of arguments, let's sort it out... There are basically three modes of
 operation, and we will explain them from simplest to most complex:
 =over 4
-=item * absolute timer (at = time, interval = reschedule_cb = 0)
+=item * absolute timer (offset = absolute time, interval = 0, reschedule_cb = 0)
 In this configuration the watcher triggers an event after the wall clock
-time C<at> has passed. It will not repeat and will not adjust when a time
+time C<offset> has passed. It will not repeat and will not adjust when a
-jump occurs, that is, if it is to be run at January 1st 2011 then it will
+time jump occurs, that is, if it is to be run at January 1st 2011 then it
-only run when the system clock reaches or surpasses this time.
+will be stopped and invoked when the system clock reaches or surpasses
+this point in time.
-=item * repeating interval timer (at = offset, interval > 0, reschedule_cb = 0)
+=item * repeating interval timer (offset = offset within interval, interval > 0, reschedule_cb = 0)
 In this mode the watcher will always be scheduled to time out at the next
-C<at + N * interval> time (for some integer N, which can also be negative)
+C<offset + N * interval> time (for some integer N, which can also be
-and then repeat, regardless of any time jumps.
+negative) and then repeat, regardless of any time jumps. The C<offset>
+argument is merely an offset into the C<interval> periods.
 This can be used to create timers that do not drift with respect to the
-system clock, for example, here is a C<ev_periodic> that triggers each
+system clock, for example, here is an C<ev_periodic> that triggers each
-hour, on the hour:
+hour, on the hour (with respect to UTC):
    ev_periodic_set (&periodic, 0., 3600., 0);
 This doesn't mean there will always be 3600 seconds in between triggers,
 but only that the callback will be called when the system time shows a
 full hour (UTC), or more correctly, when the system time is evenly divisible
 by 3600.
 Another way to think about it (for the mathematically inclined) is that
 C<ev_periodic> will try to run the callback in this mode at the next possible
-time where C<time = at (mod interval)>, regardless of any time jumps.
+time where C<time = offset (mod interval)>, regardless of any time jumps.
-For numerical stability it is preferable that the C<at> value is near
+For numerical stability it is preferable that the C<offset> value is near
 C<ev_now ()> (the current time), but there is no range requirement for
 this value, and in fact is often specified as zero.
 Note also that there is an upper limit to how often a timer can fire (CPU
 speed for example), so if C<interval> is very small then timing stability
 will of course deteriorate. Libev itself tries to be exact to be about one
 millisecond (if the OS supports it and the machine is fast enough).
-=item * manual reschedule mode (at and interval ignored, reschedule_cb = callback)
+=item * manual reschedule mode (offset ignored, interval ignored, reschedule_cb = callback)
-In this mode the values for C<interval> and C<at> are both being
+In this mode the values for C<interval> and C<offset> are both being
 ignored. Instead, each time the periodic watcher gets scheduled, the
 reschedule callback will be called with the watcher as first, and the
 current time as second argument.
-NOTE: I<This callback MUST NOT stop or destroy any periodic watcher,
+NOTE: I<This callback MUST NOT stop or destroy any periodic watcher, ever,
-ever, or make ANY event loop modifications whatsoever>.
+or make ANY other event loop modifications whatsoever, unless explicitly
+allowed by documentation here>.
 If you need to stop it, return C<now + 1e30> (or so, fudge fudge) and stop
 it afterwards (e.g. by starting an C<ev_prepare> watcher, which is the
 only event loop modification you are allowed to do).
 a different time than the last time it was called (e.g. in a crond like
 program when the crontabs have changed).
 =item ev_tstamp ev_periodic_at (ev_periodic *)
-When active, returns the absolute time that the watcher is supposed to
+When active, returns the absolute time that the watcher is supposed
-trigger next.
+to trigger next. This is not the same as the C<offset> argument to
+C<ev_periodic_set>, but indeed works even in interval and manual
+rescheduling modes.
 =item ev_tstamp offset [read-write]
 When repeating, this contains the offset value, otherwise this is the
-absolute point in time (the C<at> value passed to C<ev_periodic_set>).
+absolute point in time (the C<offset> value passed to C<ev_periodic_set>,
+although libev might modify this value for better numerical stability).
 Can be modified any time, but changes only take effect when the periodic
 timer fires or C<ev_periodic_again> is being called.
 =item ev_tstamp interval [read-write]
 Signal watchers will trigger an event when the process receives a specific
 signal one or more times. Even though signals are very asynchronous, libev
 will try it's best to deliver signals synchronously, i.e. as part of the
 normal event processing, like any other event.
-If you want signals asynchronously, just use C<sigaction> as you would
+If you want signals to be delivered truly asynchronously, just use
-do without libev and forget about sharing the signal. You can even use
+C<sigaction> as you would do without libev and forget about sharing
-C<ev_async> from a signal handler to synchronously wake up an event loop.
+the signal. You can even use C<ev_async> from a signal handler to
+synchronously wake up an event loop.
-You can configure as many watchers as you like per signal. Only when the
+You can configure as many watchers as you like for the same signal, but
+only within the same loop, i.e. you can watch for C<SIGINT> in your
+default loop and for C<SIGIO> in another loop, but you cannot watch for
+C<SIGINT> in both the default loop and another loop at the same time. At
+the moment, C<SIGCHLD> is permanently tied to the default loop.
-first watcher gets started will libev actually register a signal handler
+When the first watcher gets started will libev actually register something
 with the kernel (thus it coexists with your own signal handlers as long as
-you don't register any with libev for the same signal). Similarly, when
+you don't register any with libev for the same signal).
-the last signal watcher for a signal is stopped, libev will reset the
-signal handler to SIG_DFL (regardless of what it was set to before).
 If possible and supported, libev will install its handlers with
-C<SA_RESTART> behaviour enabled, so system calls should not be unduly
+C<SA_RESTART> (or equivalent) behaviour enabled, so system calls should
-interrupted. If you have a problem with system calls getting interrupted by
+not be unduly interrupted. If you have a problem with system calls getting
-signals you can block all signals in an C<ev_check> watcher and unblock
+interrupted by signals you can block all signals in an C<ev_check> watcher
-them in an C<ev_prepare> watcher.
+and unblock them in an C<ev_prepare> watcher.
+=head3 The special problem of inheritance over execve
+Both the signal mask (C<sigprocmask>) and the signal disposition
+(C<sigaction>) are unspecified after starting a signal watcher (and after
+stopping it again), that is, libev might or might not block the signal,
+and might or might not set or restore the installed signal handler.
+While this does not matter for the signal disposition (libev never
+sets signals to C<SIG_IGN>, so handlers will be reset to C<SIG_DFL> on
+C<execve>), this matters for the signal mask: many programs do not expect
+certain signals to be blocked.
+This means that before calling C<exec> (from the child) you should reset
+the signal mask to whatever "default" you expect (all clear is a good
+choice usually).
+The simplest way to ensure that the signal mask is reset in the child is
+to install a fork handler with C<pthread_atfork> that resets it. That will
+catch fork calls done by libraries (such as the libc) as well.
+In current versions of libev, you can also ensure that the signal mask is
+not blocking any signals (except temporarily, so thread users watch out)
+by specifying the C<EVFLAG_NOSIGFD> when creating the event loop. This
+is not guaranteed for future versions, however.
 =head3 Watcher-Specific Functions and Data Members
 =over 4
 some child status changes (most typically when a child of yours dies or
 exits). It is permissible to install a child watcher I<after> the child
 has been forked (which implies it might have already exited), as long
 as the event loop isn't entered (or is continued from a watcher), i.e.,
 forking and then immediately registering a watcher for the child is fine,
-but forking and registering a watcher a few event loop iterations later is
+but forking and registering a watcher a few event loop iterations later or
-not.
+in the next callback invocation is not.
 Only the default event loop is capable of handling signals, and therefore
 you can only register child watchers in the default event loop.
+Due to some design glitches inside libev, child watchers will always be
+handled at maximum priority (their priority is set to C<EV_MAXPRI> by
+libev)
 =head3 Process Interaction
 Libev grabs C<SIGCHLD> as soon as the default event loop is
-initialised. This is necessary to guarantee proper behaviour even if
+initialised. This is necessary to guarantee proper behaviour even if the
-the first child watcher is started after the child exits. The occurrence
+first child watcher is started after the child exits. The occurrence
 of C<SIGCHLD> is recorded asynchronously, but child reaping is done
 synchronously as part of the event loop processing. Libev always reaps all
 children, even ones not watched.
 =head3 Overriding the Built-In Processing
 =head3 Stopping the Child Watcher
 Currently, the child watcher never gets stopped, even when the
 child terminates, so normally one needs to stop the watcher in the
 callback. Future versions of libev might stop the watcher automatically
-when a child exit is detected.
+when a child exit is detected (calling C<ev_child_stop> twice is not a
+problem).
 =head3 Watcher-Specific Functions and Data Members
 =over 4
 C<stat> on that path in regular intervals (or when the OS says it changed)
 and sees if it changed compared to the last time, invoking the callback if
 it did.
 The path does not need to exist: changing from "path exists" to "path does
-not exist" is a status change like any other. The condition "path does
+not exist" is a status change like any other. The condition "path does not
-not exist" is signified by the C<st_nlink> field being zero (which is
+exist" (or more correctly "path cannot be stat'ed") is signified by the
-otherwise always forced to be at least one) and all the other fields of
+C<st_nlink> field being zero (which is otherwise always forced to be at
-the stat buffer having unspecified contents.
+least one) and all the other fields of the stat buffer having unspecified
+contents.
 The path I<must not> end in a slash or contain special components such as
 C<.> or C<..>. The path I<should> be absolute: If it is relative and
 your working directory changes, then the behaviour is undefined.
 This watcher type is not meant for massive numbers of stat watchers,
 as even with OS-supported change notifications, this can be
 resource-intensive.
 At the time of this writing, the only OS-specific interface implemented
-is the Linux inotify interface (implementing kqueue support is left as
+is the Linux inotify interface (implementing kqueue support is left as an
-an exercise for the reader. Note, however, that the author sees no way
+exercise for the reader. Note, however, that the author sees no way of
-of implementing C<ev_stat> semantics with kqueue).
+implementing C<ev_stat> semantics with kqueue, except as a hint).
 =head3 ABI Issues (Largefile Support)
 Libev by default (unless the user overrides this) uses the default
 compilation environment, which means that on systems with large file
 to exchange stat structures with application programs compiled using the
 default compilation environment.
 =head3 Inotify and Kqueue
-When C<inotify (7)> support has been compiled into libev (generally
+When C<inotify (7)> support has been compiled into libev and present at
-only available with Linux 2.6.25 or above due to bugs in earlier
+runtime, it will be used to speed up change detection where possible. The
-implementations) and present at runtime, it will be used to speed up
+inotify descriptor will be created lazily when the first C<ev_stat>
-change detection where possible. The inotify descriptor will be created
+watcher is being started.
-lazily when the first C<ev_stat> watcher is being started.
 Inotify presence does not change the semantics of C<ev_stat> watchers
 except that changes might be detected earlier, and in some cases, to avoid
 making regular C<stat> calls. Even in the presence of inotify support
 there are many cases where libev has to resort to regular C<stat> polling,
-but as long as the path exists, libev usually gets away without polling.
+but as long as kernel 2.6.25 or newer is used (2.6.24 and older have too
+many bugs), the path exists (i.e. stat succeeds), and the path resides on
+a local filesystem (libev currently assumes only ext2/3, jfs, reiserfs and
+xfs are fully working) libev usually gets away without polling.
 There is no support for kqueue, as apparently it cannot be used to
 implement this functionality, due to the requirement of having a file
 descriptor open on the object at all times, and detecting renames, unlinks
 etc. is difficult.
+=head3 C<stat ()> is a synchronous operation
+Libev doesn't normally do any kind of I/O itself, and so is not blocking
+the process. The exception are C<ev_stat> watchers - those call C<stat
+()>, which is a synchronous operation.
+For local paths, this usually doesn't matter: unless the system is very
+busy or the intervals between stat's are large, a stat call will be fast,
+as the path data is usually in memory already (except when starting the
+watcher).
+For networked file systems, calling C<stat ()> can block an indefinite
+time due to network issues, and even under good conditions, a stat call
+often takes multiple milliseconds.
+Therefore, it is best to avoid using C<ev_stat> watchers on networked
+paths, although this is fully supported by libev.
 =head3 The special problem of stat time resolution
 The C<stat ()> system call only supports full-second resolution portably,
 and even on systems where the resolution is higher, most file systems
 =head3 Watcher-Specific Functions and Data Members
 =over 4
-=item ev_idle_init (ev_signal *, callback)
+=item ev_idle_init (ev_idle *, callback)
 Initialises and configures the idle watcher - it has no parameters of any
 kind. There is a C<ev_idle_set> macro, but using it is utterly pointless,
 believe me.
      // no longer anything immediate to do.
    }
    ev_idle *idle_watcher = malloc (sizeof (ev_idle));
    ev_idle_init (idle_watcher, idle_cb);
-   ev_idle_start (loop, idle_cb);
+   ev_idle_start (loop, idle_watcher);
 =head2 C<ev_prepare> and C<ev_check> - customise your event loop!
 Prepare and check watchers are usually (but not always) used in pairs:
      struct pollfd fds [nfd];
      // actual code will need to loop here and realloc etc.
      adns_beforepoll (ads, fds, &nfd, &timeout, timeval_from (ev_time ()));
      /* the callback is illegal, but won't be called as we stop during check */
-     ev_timer_init (&tw, 0, timeout * 1e-3);
+     ev_timer_init (&tw, 0, timeout * 1e-3, 0.);
      ev_timer_start (loop, &tw);
      // create one ev_io per pollfd
      for (int i = 0; i < nfd; ++i)
        {
 some fds have to be watched and handled very quickly (with low latency),
 and even priorities and idle watchers might have too much overhead. In
 this case you would put all the high priority stuff in one loop and all
 the rest in a second one, and embed the second one in the first.
-As long as the watcher is active, the callback will be invoked every time
+As long as the watcher is active, the callback will be invoked every
-there might be events pending in the embedded loop. The callback must then
+time there might be events pending in the embedded loop. The callback
-call C<ev_embed_sweep (mainloop, watcher)> to make a single sweep and invoke
+must then call C<ev_embed_sweep (mainloop, watcher)> to make a single
-their callbacks (you could also start an idle watcher to give the embedded
+sweep and invoke their callbacks (the callback doesn't need to invoke the
-loop strictly lower priority for example). You can also set the callback
+C<ev_embed_sweep> function directly, it could also start an idle watcher
-to C<0>, in which case the embed watcher will automatically execute the
+to give the embedded loop strictly lower priority for example).
-embedded loop sweep.
-As long as the watcher is started it will automatically handle events. The
+You can also set the callback to C<0>, in which case the embed watcher
-callback will be invoked whenever some events have been handled. You can
+will automatically execute the embedded loop sweep whenever necessary.
-set the callback to C<0> to avoid having to specify one if you are not
-interested in that.
-Also, there have not currently been made special provisions for forking:
+Fork detection will be handled transparently while the C<ev_embed> watcher
-when you fork, you not only have to call C<ev_loop_fork> on both loops,
+is active, i.e., the embedded loop will automatically be forked when the
-but you will also have to stop and restart any C<ev_embed> watchers
+embedding loop forks. In other cases, the user is responsible for calling
-yourself - but you can use a fork watcher to handle this automatically,
+C<ev_loop_fork> on the embedded loop.
-and future versions of libev might do just that.
 Unfortunately, not all backends are embeddable: only the ones returned by
 C<ev_embeddable_backends> are, which, unfortunately, does not include any
 portable one.
 event loop blocks next and before C<ev_check> watchers are being called,
 and only in the child after the fork. If whoever good citizen calling
 C<ev_default_fork> cheats and calls it in the wrong process, the fork
 handlers will be invoked, too, of course.
+=head3 The special problem of life after fork - how is it possible?
+Most uses of C<fork()> consist of forking, then some simple calls to ste
+up/change the process environment, followed by a call to C<exec()>. This
+sequence should be handled by libev without any problems.
+This changes when the application actually wants to do event handling
+in the child, or both parent in child, in effect "continuing" after the
+fork.
+The default mode of operation (for libev, with application help to detect
+forks) is to duplicate all the state in the child, as would be expected
+when I<either> the parent I<or> the child process continues.
+When both processes want to continue using libev, then this is usually the
+wrong result. In that case, usually one process (typically the parent) is
+supposed to continue with all watchers in place as before, while the other
+process typically wants to start fresh, i.e. without any active watchers.
+The cleanest and most efficient way to achieve that with libev is to
+simply create a new event loop, which of course will be "empty", and
+use that for new watchers. This has the advantage of not touching more
+memory than necessary, and thus avoiding the copy-on-write, and the
+disadvantage of having to use multiple event loops (which do not support
+signal watchers).
+When this is not possible, or you want to use the default loop for
+other reasons, then in the process that wants to start "fresh", call
+C<ev_default_destroy ()> followed by C<ev_default_loop (...)>. Destroying
+the default loop will "orphan" (not stop) all registered watchers, so you
+have to be careful not to execute code that modifies those watchers. Note
+also that in that case, you have to re-register any signal watchers.
 =head3 Watcher-Specific Functions and Data Members
 =over 4
 =item ev_fork_init (ev_signal *, callback)
 an C<EV_ASYNC> event on the watcher into the event loop. Unlike
 C<ev_feed_event>, this call is safe to do from other threads, signal or
 similar contexts (see the discussion of C<EV_ATOMIC_T> in the embedding
 section below on what exactly this means).
+Note that, as with other watchers in libev, multiple events might get
+compressed into a single callback invocation (another way to look at this
+is that C<ev_async> watchers are level-triggered, set on C<ev_async_send>,
+reset when the event loop detects that).
-This call incurs the overhead of a system call only once per loop iteration,
+This call incurs the overhead of a system call only once per event loop
-so while the overhead might be noticeable, it doesn't apply to repeated
+iteration, so while the overhead might be noticeable, it doesn't apply to
-calls to C<ev_async_send>.
+repeated calls to C<ev_async_send> for the same event loop.
 =item bool = ev_async_pending (ev_async *)
 Returns a non-zero value when C<ev_async_send> has been called on the
 watcher but the event has not yet been processed (or even noted) by the
 C<ev_async_send> sets a flag in the watcher and wakes up the loop. When
 the loop iterates next and checks for the watcher to have become active,
 it will reset the flag again. C<ev_async_pending> can be used to very
 quickly check whether invoking the loop might be a good idea.
-Not that this does I<not> check whether the watcher itself is pending, only
+Not that this does I<not> check whether the watcher itself is pending,
-whether it has been requested to make this watcher pending.
+only whether it has been requested to make this watcher pending: there
+is a time window between the event loop checking and resetting the async
+notification, and the callback being invoked.
 =back
 =head1 OTHER FUNCTIONS
        /* doh, nothing entered */;
    }
    ev_once (STDIN_FILENO, EV_READ, 10., stdin_ready, 0);
-=item ev_feed_event (struct ev_loop *, watcher *, int revents)
-Feeds the given event set into the event loop, as if the specified event
-had happened for the specified watcher (which must be a pointer to an
-initialised but not necessarily started event watcher).
 =item ev_feed_fd_event (struct ev_loop *, int fd, int revents)
 Feed an event on the given fd, as if a file descriptor backend detected
 the given events it.
    }
    myclass obj;
    ev::io iow;
    iow.set <myclass, &myclass::io_cb> (&obj);
+=item w->set (object *)
+This is an B<experimental> feature that might go away in a future version.
+This is a variation of a method callback - leaving out the method to call
+will default the method to C<operator ()>, which makes it possible to use
+functor objects without having to manually specify the C<operator ()> all
+the time. Incidentally, you can then also leave out the template argument
+list.
+The C<operator ()> method prototype must be C<void operator ()(watcher &w,
+int revents)>.
+See the method-C<set> above for more details.
+Example: use a functor object as callback.
+   struct myfunctor
+   {
+     void operator() (ev::io &w, int revents)
+     {
+       ...
+     }
+   }
+   myfunctor f;
+   ev::io w;
+   w.set (&f);
 =item w->set<function> (void *data = 0)
 Also sets a callback, but uses a static method or plain function as
 callback. The optional C<data> argument will be stored in the watcher's
 L<http://software.schmorp.de/pkg/EV>.
 =item Python
 Python bindings can be found at L<http://code.google.com/p/pyev/>. It
-seems to be quite complete and well-documented. Note, however, that the
+seems to be quite complete and well-documented.
-patch they require for libev is outright dangerous as it breaks the ABI
-for everybody else, and therefore, should never be applied in an installed
-libev (if python requires an incompatible ABI then it needs to embed
-libev).
 =item Ruby
 Tony Arcieri has written a ruby extension that offers access to a subset
 of the libev API and adds file handle abstractions, asynchronous DNS and
 more on top of it. It can be found via gem servers. Its homepage is at
 L<http://rev.rubyforge.org/>.
+Roger Pack reports that using the link order C<-lws2_32 -lmsvcrt-ruby-190>
+makes rev work even on mingw.
+=item Haskell
+A haskell binding to libev is available at
+L<http://hackage.haskell.org/cgi-bin/hackage-scripts/package/hlibev>.
 =item D
 Leandro Lucarella has written a D language binding (F<ev.d>) for libev, to
 be found at L<http://proj.llucax.com.ar/wiki/evd>.
 =item Ocaml
 Erkki Seppala has written Ocaml bindings for libev, to be found at
 L<http://modeemi.cs.tut.fi/~flux/software/ocaml-ev/>.
+=item Lua
+Brian Maher has written a partial interface to libev
+for lua (only C<ev_io> and C<ev_timer>), to be found at
+L<http://github.com/brimworks/lua-ev>.
 =back
 =head1 MACRO MAGIC
 keeps libev from including F<config.h>, and it also defines dummy
 implementations for some libevent functions (such as logging, which is not
 supported). It will also not define any of the structs usually found in
 F<event.h> that are not directly supported by the libev core alone.
+In standalone mode, libev will still try to automatically deduce the
+configuration, but has to be more conservative.
 =item EV_USE_MONOTONIC
 If defined to be C<1>, libev will try to detect the availability of the
-monotonic clock option at both compile time and runtime. Otherwise no use
+monotonic clock option at both compile time and runtime. Otherwise no
-of the monotonic clock option will be attempted. If you enable this, you
+use of the monotonic clock option will be attempted. If you enable this,
-usually have to link against librt or something similar. Enabling it when
+you usually have to link against librt or something similar. Enabling it
-the functionality isn't available is safe, though, although you have
+when the functionality isn't available is safe, though, although you have
 to make sure you link against any libraries where the C<clock_gettime>
-function is hiding in (often F<-lrt>).
+function is hiding in (often F<-lrt>). See also C<EV_USE_CLOCK_SYSCALL>.
 =item EV_USE_REALTIME
 If defined to be C<1>, libev will try to detect the availability of the
-real-time clock option at compile time (and assume its availability at
+real-time clock option at compile time (and assume its availability
-runtime if successful). Otherwise no use of the real-time clock option will
+at runtime if successful). Otherwise no use of the real-time clock
-be attempted. This effectively replaces C<gettimeofday> by C<clock_get
+option will be attempted. This effectively replaces C<gettimeofday>
-(CLOCK_REALTIME, ...)> and will not normally affect correctness. See the
+by C<clock_get (CLOCK_REALTIME, ...)> and will not normally affect
-note about libraries in the description of C<EV_USE_MONOTONIC>, though.
+correctness. See the note about libraries in the description of
+C<EV_USE_MONOTONIC>, though. Defaults to the opposite value of
+C<EV_USE_CLOCK_SYSCALL>.
+=item EV_USE_CLOCK_SYSCALL
+If defined to be C<1>, libev will try to use a direct syscall instead
+of calling the system-provided C<clock_gettime> function. This option
+exists because on GNU/Linux, C<clock_gettime> is in C<librt>, but C<librt>
+unconditionally pulls in C<libpthread>, slowing down single-threaded
+programs needlessly. Using a direct syscall is slightly slower (in
+theory), because no optimised vdso implementation can be used, but avoids
+the pthread dependency. Defaults to C<1> on GNU/Linux with glibc 2.x or
+higher, as it simplifies linking (no need for C<-lrt>).
 =item EV_USE_NANOSLEEP
 If defined to be C<1>, libev will assume that C<nanosleep ()> is available
 and will use it for delays. Otherwise it will use C<select ()>.
 =item EV_SELECT_USE_FD_SET
 If defined to C<1>, then the select backend will use the system C<fd_set>
 structure. This is useful if libev doesn't compile due to a missing
-C<NFDBITS> or C<fd_mask> definition or it mis-guesses the bitset layout on
+C<NFDBITS> or C<fd_mask> definition or it mis-guesses the bitset layout
-exotic systems. This usually limits the range of file descriptors to some
+on exotic systems. This usually limits the range of file descriptors to
-low limit such as 1024 or might have other limitations (winsocket only
+some low limit such as 1024 or might have other limitations (winsocket
-allows 64 sockets). The C<FD_SETSIZE> macro, set before compilation, might
+only allows 64 sockets). The C<FD_SETSIZE> macro, set before compilation,
-influence the size of the C<fd_set> used.
+configures the maximum size of the C<fd_set>.
 =item EV_SELECT_IS_WINSOCKET
 When defined to C<1>, the select backend will assume that
 select/socket/connect etc. don't understand file descriptors but
 be used is the winsock select). This means that it will call
 C<_get_osfhandle> on the fd to convert it to an OS handle. Otherwise,
 it is assumed that all these functions actually work on fds, even
 on win32. Should not be defined on non-win32 platforms.
-=item EV_FD_TO_WIN32_HANDLE
+=item EV_FD_TO_WIN32_HANDLE(fd)
 If C<EV_SELECT_IS_WINSOCKET> is enabled, then libev needs a way to map
 file descriptors to socket handles. When not defining this symbol (the
 default), then libev will call C<_get_osfhandle>, which is usually
 correct. In some cases, programs use their own file descriptor management,
 in which case they can provide this function to map fds to socket handles.
+=item EV_WIN32_HANDLE_TO_FD(handle)
+If C<EV_SELECT_IS_WINSOCKET> then libev maps handles to file descriptors
+using the standard C<_open_osfhandle> function. For programs implementing
+their own fd to handle mapping, overwriting this function makes it easier
+to do so. This can be done by defining this macro to an appropriate value.
+=item EV_WIN32_CLOSE_FD(fd)
+If programs implement their own fd to handle mapping on win32, then this
+macro can be used to override the C<close> function, useful to unregister
+file descriptors again. Note that the replacement function has to close
+the underlying OS handle.
 =item EV_USE_POLL
 If defined to be C<1>, libev will compile in support for the C<poll>(2)
 backend. Otherwise it will be enabled on non-win32 platforms. It
 defined to be C<0>, then they are not.
 =item EV_MINIMAL
 If you need to shave off some kilobytes of code at the expense of some
-speed, define this symbol to C<1>. Currently this is used to override some
+speed (but with the full API), define this symbol to C<1>. Currently this
-inlining decisions, saves roughly 30% code size on amd64. It also selects a
+is used to override some inlining decisions, saves roughly 30% code size
-much smaller 2-heap for timer management over the default 4-heap.
+on amd64. It also selects a much smaller 2-heap for timer management over
+the default 4-heap.
+You can save even more by disabling watcher types you do not need
+and setting C<EV_MAXPRI> == C<EV_MINPRI>. Also, disabling C<assert>
+(C<-DNDEBUG>) will usually reduce code size a lot.
+Defining C<EV_MINIMAL> to C<2> will additionally reduce the core API to
+provide a bare-bones event library. See C<ev.h> for details on what parts
+of the API are still available, and do not complain if this subset changes
+over time.
+=item EV_NSIG
+The highest supported signal number, +1 (or, the number of
+signals): Normally, libev tries to deduce the maximum number of signals
+automatically, but sometimes this fails, in which case it can be
+specified. Also, using a lower number than detected (C<32> should be
+good for about any system in existance) can save some memory, as libev
+statically allocates some 12-24 bytes per signal number.
 =item EV_PID_HASHSIZE
 C<ev_child> watchers use a small hash table to distribute workload by
 pid. The default size is C<16> (or C<1> with C<EV_MINIMAL>), usually more
 default loop and triggering an C<ev_async> watcher from the default loop
 watcher callback into the event loop interested in the signal.
 =back
+=head4 THREAD LOCKING EXAMPLE
+Here is a fictitious example of how to run an event loop in a different
+thread than where callbacks are being invoked and watchers are
+created/added/removed.
+For a real-world example, see the C<EV::Loop::Async> perl module,
+which uses exactly this technique (which is suited for many high-level
+languages).
+The example uses a pthread mutex to protect the loop data, a condition
+variable to wait for callback invocations, an async watcher to notify the
+event loop thread and an unspecified mechanism to wake up the main thread.
+First, you need to associate some data with the event loop:
+   typedef struct {
+     mutex_t lock; /* global loop lock */
+     ev_async async_w;
+     thread_t tid;
+     cond_t invoke_cv;
+   } userdata;
+   void prepare_loop (EV_P)
+   {
+      // for simplicity, we use a static userdata struct.
+      static userdata u;
+      ev_async_init (&u->async_w, async_cb);
+      ev_async_start (EV_A_ &u->async_w);
+      pthread_mutex_init (&u->lock, 0);
+      pthread_cond_init (&u->invoke_cv, 0);
+      // now associate this with the loop
+      ev_set_userdata (EV_A_ u);
+      ev_set_invoke_pending_cb (EV_A_ l_invoke);
+      ev_set_loop_release_cb (EV_A_ l_release, l_acquire);
+      // then create the thread running ev_loop
+      pthread_create (&u->tid, 0, l_run, EV_A);
+   }
+The callback for the C<ev_async> watcher does nothing: the watcher is used
+solely to wake up the event loop so it takes notice of any new watchers
+that might have been added:
+   static void
+   async_cb (EV_P_ ev_async *w, int revents)
+   {
+      // just used for the side effects
+   }
+The C<l_release> and C<l_acquire> callbacks simply unlock/lock the mutex
+protecting the loop data, respectively.
+   static void
+   l_release (EV_P)
+   {
+     userdata *u = ev_userdata (EV_A);
+     pthread_mutex_unlock (&u->lock);
+   }
+   static void
+   l_acquire (EV_P)
+   {
+     userdata *u = ev_userdata (EV_A);
+     pthread_mutex_lock (&u->lock);
+   }
+The event loop thread first acquires the mutex, and then jumps straight
+into C<ev_loop>:
+   void *
+   l_run (void *thr_arg)
+   {
+     struct ev_loop *loop = (struct ev_loop *)thr_arg;
+     l_acquire (EV_A);
+     pthread_setcanceltype (PTHREAD_CANCEL_ASYNCHRONOUS, 0);
+     ev_loop (EV_A_ 0);
+     l_release (EV_A);
+     return 0;
+   }
+Instead of invoking all pending watchers, the C<l_invoke> callback will
+signal the main thread via some unspecified mechanism (signals? pipe
+writes? C<Async::Interrupt>?) and then waits until all pending watchers
+have been called (in a while loop because a) spurious wakeups are possible
+and b) skipping inter-thread-communication when there are no pending
+watchers is very beneficial):
+   static void
+   l_invoke (EV_P)
+   {
+     userdata *u = ev_userdata (EV_A);
+     while (ev_pending_count (EV_A))
+       {
+         wake_up_other_thread_in_some_magic_or_not_so_magic_way ();
+         pthread_cond_wait (&u->invoke_cv, &u->lock);
+       }
+   }
+Now, whenever the main thread gets told to invoke pending watchers, it
+will grab the lock, call C<ev_invoke_pending> and then signal the loop
+thread to continue:
+   static void
+   real_invoke_pending (EV_P)
+   {
+     userdata *u = ev_userdata (EV_A);
+     pthread_mutex_lock (&u->lock);
+     ev_invoke_pending (EV_A);
+     pthread_cond_signal (&u->invoke_cv);
+     pthread_mutex_unlock (&u->lock);
+   }
+Whenever you want to start/stop a watcher or do other modifications to an
+event loop, you will now have to lock:
+   ev_timer timeout_watcher;
+   userdata *u = ev_userdata (EV_A);
+   ev_timer_init (&timeout_watcher, timeout_cb, 5.5, 0.);
+   pthread_mutex_lock (&u->lock);
+   ev_timer_start (EV_A_ &timeout_watcher);
+   ev_async_send (EV_A_ &u->async_w);
+   pthread_mutex_unlock (&u->lock);
+Note that sending the C<ev_async> watcher is required because otherwise
+an event loop currently blocking in the kernel will have no knowledge
+about the newly added timer. By waking up the loop it will pick up any new
+watchers in the next event loop iteration.
 =head3 COROUTINES
 Libev is very accommodating to coroutines ("cooperative threads"):
 libev fully supports nesting calls to its functions from different
 coroutines (e.g. you can call C<ev_loop> on the same loop from two
-different coroutines, and switch freely between both coroutines running the
+different coroutines, and switch freely between both coroutines running
-loop, as long as you don't confuse yourself). The only exception is that
+the loop, as long as you don't confuse yourself). The only exception is
-you must not do this from C<ev_periodic> reschedule callbacks.
+that you must not do this from C<ev_periodic> reschedule callbacks.
 Care has been taken to ensure that libev does not keep local state inside
 C<ev_loop>, and other calls do not usually allow for coroutine switches as
 they do not call any callbacks.
 way (note also that glib is the slowest event library known to man).
 There is no supported compilation method available on windows except
 embedding it into other applications.
+Sensible signal handling is officially unsupported by Microsoft - libev
+tries its best, but under most conditions, signals will simply not work.
 Not a libev limitation but worth mentioning: windows apparently doesn't
 accept large writes: instead of resulting in a partial write, windows will
 either accept everything or return C<ENOBUFS> if the buffer is too large,
 so make sure you only write small amounts into your sockets (less than a
 megabyte seems safe, but this apparently depends on the amount of memory
 the abysmal performance of winsockets, using a large number of sockets
 is not recommended (and not reasonable). If your program needs to use
 more than a hundred or so sockets, then likely it needs to use a totally
 different implementation for windows, as libev offers the POSIX readiness
 notification model, which cannot be implemented efficiently on windows
-(Microsoft monopoly games).
+(due to Microsoft monopoly games).
 A typical way to use libev under windows is to embed it (see the embedding
 section for details) and use the following F<evwrap.h> header file instead
 of F<ev.h>:
 Early versions of winsocket's select only supported waiting for a maximum
 of C<64> handles (probably owning to the fact that all windows kernels
 can only wait for C<64> things at the same time internally; Microsoft
 recommends spawning a chain of threads and wait for 63 handles and the
-previous thread in each. Great).
+previous thread in each. Sounds great!).
 Newer versions support more handles, but you need to define C<FD_SETSIZE>
 to some high number (e.g. C<2048>) before compiling the winsocket select
-call (which might be in libev or elsewhere, for example, perl does its own
+call (which might be in libev or elsewhere, for example, perl and many
-select emulation on windows).
+other interpreters do their own select emulation on windows).
 Another limit is the number of file descriptors in the Microsoft runtime
-libraries, which by default is C<64> (there must be a hidden I<64> fetish
+libraries, which by default is C<64> (there must be a hidden I<64>
-or something like this inside Microsoft). You can increase this by calling
+fetish or something like this inside Microsoft). You can increase this
-C<_setmaxstdio>, which can increase this limit to C<2048> (another
+by calling C<_setmaxstdio>, which can increase this limit to C<2048>
-arbitrary limit), but is broken in many versions of the Microsoft runtime
+(another arbitrary limit), but is broken in many versions of the Microsoft
-libraries.
-This might get you to about C<512> or C<2048> sockets (depending on
+runtime libraries. This might get you to about C<512> or C<2048> sockets
-windows version and/or the phase of the moon). To get more, you need to
+(depending on windows version and/or the phase of the moon). To get more,
-wrap all I/O functions and provide your own fd management, but the cost of
+you need to wrap all I/O functions and provide your own fd management, but
-calling select (O(n²)) will likely make this unworkable.
+the cost of calling select (O(n²)) will likely make this unworkable.
 =back
 =head2 PORTABILITY REQUIREMENTS
 =item C<double> must hold a time value in seconds with enough accuracy
 The type C<double> is used to represent timestamps. It is required to
 have at least 51 bits of mantissa (and 9 bits of exponent), which is good
 enough for at least into the year 4000. This requirement is fulfilled by
-implementations implementing IEEE 754 (basically all existing ones).
+implementations implementing IEEE 754, which is basically all existing
+ones. With IEEE 754 doubles, you get microsecond accuracy until at least
+2200.
 =back
 If you know of other additional requirements drop me a note.
 involves iterating over all running async watchers or all signal numbers.
 =back
+=head1 GLOSSARY
+=over 4
+=item active
+A watcher is active as long as it has been started (has been attached to
+an event loop) but not yet stopped (disassociated from the event loop).
+=item application
+In this document, an application is whatever is using libev.
+=item callback
+The address of a function that is called when some event has been
+detected. Callbacks are being passed the event loop, the watcher that
+received the event, and the actual event bitset.
+=item callback invocation
+The act of calling the callback associated with a watcher.
+=item event
+A change of state of some external event, such as data now being available
+for reading on a file descriptor, time having passed or simply not having
+any other events happening anymore.
+In libev, events are represented as single bits (such as C<EV_READ> or
+C<EV_TIMEOUT>).
+=item event library
+A software package implementing an event model and loop.
+=item event loop
+An entity that handles and processes external events and converts them
+into callback invocations.
+=item event model
+The model used to describe how an event loop handles and processes
+watchers and events.
+=item pending
+A watcher is pending as soon as the corresponding event has been detected,
+and stops being pending as soon as the watcher will be invoked or its
+pending status is explicitly cleared by the application.
+A watcher can be pending, but not active. Stopping a watcher also clears
+its pending status.
+=item real time
+The physical time that is observed. It is apparently strictly monotonic :)
+=item wall-clock time
+The time and date as shown on clocks. Unlike real time, it can actually
+be wrong and jump forwards and backwards, e.g. when the you adjust your
+clock.
+=item watcher
+A data structure that describes interest in certain events. Watchers need
+to be started (attached to an event loop) before they can receive events.
+=item watcher invocation
+The act of calling the callback associated with a watcher.
+=back
 =head1 AUTHOR
 Marc Lehmann <libev@schmorp.de>, with repeated corrections by Mikael Magnusson.

Diff Legend

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

Comparing libev/ev.pod (file contents): Revision 1.210 by root, Thu Oct 30 08:09:30 2008 UTC vs. Revision 1.273 by root, Tue Nov 24 14:46:59 2009 UTC

Diff Legend

Comparing libev/ev.pod (file contents):
Revision 1.210 by root, Thu Oct 30 08:09:30 2008 UTC vs.
Revision 1.273 by root, Tue Nov 24 14:46:59 2009 UTC