ViewVC Help
View File | Revision Log | Show Annotations | Download File
/cvs/libev/ev.3
(Generate patch)

Comparing libev/ev.3 (file contents):
Revision 1.8 by root, Fri Nov 23 15:26:08 2007 UTC vs.
Revision 1.52 by root, Wed Dec 19 00:56:39 2007 UTC

… …
126. ds Ae AE 126. ds Ae AE
127.\} 127.\}
128.rm #[ #] #H #V #F C 128.rm #[ #] #H #V #F C
129.\" ======================================================================== 129.\" ========================================================================
130.\" 130.\"
131.IX Title ""<STANDARD INPUT>" 1" 131.IX Title "EV 1"
132.TH "<STANDARD INPUT>" 1 "2007-11-23" "perl v5.8.8" "User Contributed Perl Documentation" 132.TH EV 1 "2007-12-18" "perl v5.8.8" "User Contributed Perl Documentation"
133.SH "NAME" 133.SH "NAME"
134libev \- a high performance full\-featured event loop written in C 134libev \- a high performance full\-featured event loop written in C
135.SH "SYNOPSIS" 135.SH "SYNOPSIS"
136.IX Header "SYNOPSIS" 136.IX Header "SYNOPSIS"
137.Vb 1 137.Vb 1
138\& #include <ev.h> 138\& #include <ev.h>
139.Ve 139.Ve
140.SH "EXAMPLE PROGRAM"
141.IX Header "EXAMPLE PROGRAM"
142.Vb 1
143\& #include <ev.h>
144.Ve
145.PP
146.Vb 2
147\& ev_io stdin_watcher;
148\& ev_timer timeout_watcher;
149.Ve
150.PP
151.Vb 8
152\& /* called when data readable on stdin */
153\& static void
154\& stdin_cb (EV_P_ struct ev_io *w, int revents)
155\& {
156\& /* puts ("stdin ready"); */
157\& ev_io_stop (EV_A_ w); /* just a syntax example */
158\& ev_unloop (EV_A_ EVUNLOOP_ALL); /* leave all loop calls */
159\& }
160.Ve
161.PP
162.Vb 6
163\& static void
164\& timeout_cb (EV_P_ struct ev_timer *w, int revents)
165\& {
166\& /* puts ("timeout"); */
167\& ev_unloop (EV_A_ EVUNLOOP_ONE); /* leave one loop call */
168\& }
169.Ve
170.PP
171.Vb 4
172\& int
173\& main (void)
174\& {
175\& struct ev_loop *loop = ev_default_loop (0);
176.Ve
177.PP
178.Vb 3
179\& /* initialise an io watcher, then start it */
180\& ev_io_init (&stdin_watcher, stdin_cb, /*STDIN_FILENO*/ 0, EV_READ);
181\& ev_io_start (loop, &stdin_watcher);
182.Ve
183.PP
184.Vb 3
185\& /* simple non-repeating 5.5 second timeout */
186\& ev_timer_init (&timeout_watcher, timeout_cb, 5.5, 0.);
187\& ev_timer_start (loop, &timeout_watcher);
188.Ve
189.PP
190.Vb 2
191\& /* loop till timeout or data ready */
192\& ev_loop (loop, 0);
193.Ve
194.PP
195.Vb 2
196\& return 0;
197\& }
198.Ve
140.SH "DESCRIPTION" 199.SH "DESCRIPTION"
141.IX Header "DESCRIPTION" 200.IX Header "DESCRIPTION"
201The newest version of this document is also available as a html-formatted
202web page you might find easier to navigate when reading it for the first
203time: <http://cvs.schmorp.de/libev/ev.html>.
204.PP
142Libev is an event loop: you register interest in certain events (such as a 205Libev is an event loop: you register interest in certain events (such as a
143file descriptor being readable or a timeout occuring), and it will manage 206file descriptor being readable or a timeout occuring), and it will manage
144these event sources and provide your program with events. 207these event sources and provide your program with events.
145.PP 208.PP
146To do this, it must take more or less complete control over your process 209To do this, it must take more or less complete control over your process
… …
151watchers\fR, which are relatively small C structures you initialise with the 214watchers\fR, which are relatively small C structures you initialise with the
152details of the event, and then hand it over to libev by \fIstarting\fR the 215details of the event, and then hand it over to libev by \fIstarting\fR the
153watcher. 216watcher.
154.SH "FEATURES" 217.SH "FEATURES"
155.IX Header "FEATURES" 218.IX Header "FEATURES"
156Libev supports select, poll, the linux-specific epoll and the bsd-specific 219Libev supports \f(CW\*(C`select\*(C'\fR, \f(CW\*(C`poll\*(C'\fR, the Linux-specific \f(CW\*(C`epoll\*(C'\fR, the
157kqueue mechanisms for file descriptor events, relative timers, absolute 220BSD-specific \f(CW\*(C`kqueue\*(C'\fR and the Solaris-specific event port mechanisms
158timers with customised rescheduling, signal events, process status change 221for file descriptor events (\f(CW\*(C`ev_io\*(C'\fR), the Linux \f(CW\*(C`inotify\*(C'\fR interface
159events (related to \s-1SIGCHLD\s0), and event watchers dealing with the event 222(for \f(CW\*(C`ev_stat\*(C'\fR), relative timers (\f(CW\*(C`ev_timer\*(C'\fR), absolute timers
160loop mechanism itself (idle, prepare and check watchers). It also is quite 223with customised rescheduling (\f(CW\*(C`ev_periodic\*(C'\fR), synchronous signals
161fast (see this benchmark comparing 224(\f(CW\*(C`ev_signal\*(C'\fR), process status change events (\f(CW\*(C`ev_child\*(C'\fR), and event
162it to libevent for example). 225watchers dealing with the event loop mechanism itself (\f(CW\*(C`ev_idle\*(C'\fR,
226\&\f(CW\*(C`ev_embed\*(C'\fR, \f(CW\*(C`ev_prepare\*(C'\fR and \f(CW\*(C`ev_check\*(C'\fR watchers) as well as
227file watchers (\f(CW\*(C`ev_stat\*(C'\fR) and even limited support for fork events
228(\f(CW\*(C`ev_fork\*(C'\fR).
229.PP
230It also is quite fast (see this
231benchmark comparing it to libevent
232for example).
163.SH "CONVENTIONS" 233.SH "CONVENTIONS"
164.IX Header "CONVENTIONS" 234.IX Header "CONVENTIONS"
165Libev is very configurable. In this manual the default configuration 235Libev is very configurable. In this manual the default configuration will
166will be described, which supports multiple event loops. For more info 236be described, which supports multiple event loops. For more info about
167about various configuration options please have a look at the file 237various configuration options please have a look at \fB\s-1EMBED\s0\fR section in
168\&\fI\s-1README\s0.embed\fR in the libev distribution. If libev was configured without 238this manual. If libev was configured without support for multiple event
169support for multiple event loops, then all functions taking an initial 239loops, then all functions taking an initial argument of name \f(CW\*(C`loop\*(C'\fR
170argument of name \f(CW\*(C`loop\*(C'\fR (which is always of type \f(CW\*(C`struct ev_loop *\*(C'\fR) 240(which is always of type \f(CW\*(C`struct ev_loop *\*(C'\fR) will not have this argument.
171will not have this argument.
172.SH "TIME REPRESENTATION" 241.SH "TIME REPRESENTATION"
173.IX Header "TIME REPRESENTATION" 242.IX Header "TIME REPRESENTATION"
174Libev represents time as a single floating point number, representing the 243Libev represents time as a single floating point number, representing the
175(fractional) number of seconds since the (\s-1POSIX\s0) epoch (somewhere near 244(fractional) number of seconds since the (\s-1POSIX\s0) epoch (somewhere near
176the beginning of 1970, details are complicated, don't ask). This type is 245the beginning of 1970, details are complicated, don't ask). This type is
177called \f(CW\*(C`ev_tstamp\*(C'\fR, which is what you should use too. It usually aliases 246called \f(CW\*(C`ev_tstamp\*(C'\fR, which is what you should use too. It usually aliases
178to the double type in C. 247to the \f(CW\*(C`double\*(C'\fR type in C, and when you need to do any calculations on
248it, you should treat it as some floatingpoint value. Unlike the name
249component \f(CW\*(C`stamp\*(C'\fR might indicate, it is also used for time differences
250throughout libev.
179.SH "GLOBAL FUNCTIONS" 251.SH "GLOBAL FUNCTIONS"
180.IX Header "GLOBAL FUNCTIONS" 252.IX Header "GLOBAL FUNCTIONS"
181These functions can be called anytime, even before initialising the 253These functions can be called anytime, even before initialising the
182library in any way. 254library in any way.
183.IP "ev_tstamp ev_time ()" 4 255.IP "ev_tstamp ev_time ()" 4
… …
189.IX Item "int ev_version_major ()" 261.IX Item "int ev_version_major ()"
190.PD 0 262.PD 0
191.IP "int ev_version_minor ()" 4 263.IP "int ev_version_minor ()" 4
192.IX Item "int ev_version_minor ()" 264.IX Item "int ev_version_minor ()"
193.PD 265.PD
194You can find out the major and minor version numbers of the library 266You can find out the major and minor \s-1ABI\s0 version numbers of the library
195you linked against by calling the functions \f(CW\*(C`ev_version_major\*(C'\fR and 267you linked against by calling the functions \f(CW\*(C`ev_version_major\*(C'\fR and
196\&\f(CW\*(C`ev_version_minor\*(C'\fR. If you want, you can compare against the global 268\&\f(CW\*(C`ev_version_minor\*(C'\fR. If you want, you can compare against the global
197symbols \f(CW\*(C`EV_VERSION_MAJOR\*(C'\fR and \f(CW\*(C`EV_VERSION_MINOR\*(C'\fR, which specify the 269symbols \f(CW\*(C`EV_VERSION_MAJOR\*(C'\fR and \f(CW\*(C`EV_VERSION_MINOR\*(C'\fR, which specify the
198version of the library your program was compiled against. 270version of the library your program was compiled against.
199.Sp 271.Sp
272These version numbers refer to the \s-1ABI\s0 version of the library, not the
273release version.
274.Sp
200Usually, it's a good idea to terminate if the major versions mismatch, 275Usually, it's a good idea to terminate if the major versions mismatch,
201as this indicates an incompatible change. Minor versions are usually 276as this indicates an incompatible change. Minor versions are usually
202compatible to older versions, so a larger minor version alone is usually 277compatible to older versions, so a larger minor version alone is usually
203not a problem. 278not a problem.
279.Sp
280Example: Make sure we haven't accidentally been linked against the wrong
281version.
282.Sp
283.Vb 3
284\& assert (("libev version mismatch",
285\& ev_version_major () == EV_VERSION_MAJOR
286\& && ev_version_minor () >= EV_VERSION_MINOR));
287.Ve
204.IP "unsigned int ev_supported_backends ()" 4 288.IP "unsigned int ev_supported_backends ()" 4
205.IX Item "unsigned int ev_supported_backends ()" 289.IX Item "unsigned int ev_supported_backends ()"
206Return the set of all backends (i.e. their corresponding \f(CW\*(C`EV_BACKEND_*\*(C'\fR 290Return the set of all backends (i.e. their corresponding \f(CW\*(C`EV_BACKEND_*\*(C'\fR
207value) compiled into this binary of libev (independent of their 291value) compiled into this binary of libev (independent of their
208availability on the system you are running on). See \f(CW\*(C`ev_default_loop\*(C'\fR for 292availability on the system you are running on). See \f(CW\*(C`ev_default_loop\*(C'\fR for
209a description of the set values. 293a description of the set values.
294.Sp
295Example: make sure we have the epoll method, because yeah this is cool and
296a must have and can we have a torrent of it please!!!11
297.Sp
298.Vb 2
299\& assert (("sorry, no epoll, no sex",
300\& ev_supported_backends () & EVBACKEND_EPOLL));
301.Ve
210.IP "unsigned int ev_recommended_backends ()" 4 302.IP "unsigned int ev_recommended_backends ()" 4
211.IX Item "unsigned int ev_recommended_backends ()" 303.IX Item "unsigned int ev_recommended_backends ()"
212Return the set of all backends compiled into this binary of libev and also 304Return the set of all backends compiled into this binary of libev and also
213recommended for this platform. This set is often smaller than the one 305recommended for this platform. This set is often smaller than the one
214returned by \f(CW\*(C`ev_supported_backends\*(C'\fR, as for example kqueue is broken on 306returned by \f(CW\*(C`ev_supported_backends\*(C'\fR, as for example kqueue is broken on
215most BSDs and will not be autodetected unless you explicitly request it 307most BSDs and will not be autodetected unless you explicitly request it
216(assuming you know what you are doing). This is the set of backends that 308(assuming you know what you are doing). This is the set of backends that
217libev will probe for if you specify no backends explicitly. 309libev will probe for if you specify no backends explicitly.
310.IP "unsigned int ev_embeddable_backends ()" 4
311.IX Item "unsigned int ev_embeddable_backends ()"
312Returns the set of backends that are embeddable in other event loops. This
313is the theoretical, all\-platform, value. To find which backends
314might be supported on the current system, you would need to look at
315\&\f(CW\*(C`ev_embeddable_backends () & ev_supported_backends ()\*(C'\fR, likewise for
316recommended ones.
317.Sp
318See the description of \f(CW\*(C`ev_embed\*(C'\fR watchers for more info.
218.IP "ev_set_allocator (void *(*cb)(void *ptr, long size))" 4 319.IP "ev_set_allocator (void *(*cb)(void *ptr, long size))" 4
219.IX Item "ev_set_allocator (void *(*cb)(void *ptr, long size))" 320.IX Item "ev_set_allocator (void *(*cb)(void *ptr, long size))"
220Sets the allocation function to use (the prototype is similar to the 321Sets the allocation function to use (the prototype is similar \- the
221realloc C function, the semantics are identical). It is used to allocate 322semantics is identical \- to the realloc C function). It is used to
222and free memory (no surprises here). If it returns zero when memory 323allocate and free memory (no surprises here). If it returns zero when
223needs to be allocated, the library might abort or take some potentially 324memory needs to be allocated, the library might abort or take some
224destructive action. The default is your system realloc function. 325potentially destructive action. The default is your system realloc
326function.
225.Sp 327.Sp
226You could override this function in high-availability programs to, say, 328You could override this function in high-availability programs to, say,
227free some memory if it cannot allocate memory, to use a special allocator, 329free some memory if it cannot allocate memory, to use a special allocator,
228or even to sleep a while and retry until some memory is available. 330or even to sleep a while and retry until some memory is available.
331.Sp
332Example: Replace the libev allocator with one that waits a bit and then
333retries).
334.Sp
335.Vb 6
336\& static void *
337\& persistent_realloc (void *ptr, size_t size)
338\& {
339\& for (;;)
340\& {
341\& void *newptr = realloc (ptr, size);
342.Ve
343.Sp
344.Vb 2
345\& if (newptr)
346\& return newptr;
347.Ve
348.Sp
349.Vb 3
350\& sleep (60);
351\& }
352\& }
353.Ve
354.Sp
355.Vb 2
356\& ...
357\& ev_set_allocator (persistent_realloc);
358.Ve
229.IP "ev_set_syserr_cb (void (*cb)(const char *msg));" 4 359.IP "ev_set_syserr_cb (void (*cb)(const char *msg));" 4
230.IX Item "ev_set_syserr_cb (void (*cb)(const char *msg));" 360.IX Item "ev_set_syserr_cb (void (*cb)(const char *msg));"
231Set the callback function to call on a retryable syscall error (such 361Set the callback function to call on a retryable syscall error (such
232as failed select, poll, epoll_wait). The message is a printable string 362as failed select, poll, epoll_wait). The message is a printable string
233indicating the system call or subsystem causing the problem. If this 363indicating the system call or subsystem causing the problem. If this
234callback is set, then libev will expect it to remedy the sitution, no 364callback is set, then libev will expect it to remedy the sitution, no
235matter what, when it returns. That is, libev will generally retry the 365matter what, when it returns. That is, libev will generally retry the
236requested operation, or, if the condition doesn't go away, do bad stuff 366requested operation, or, if the condition doesn't go away, do bad stuff
237(such as abort). 367(such as abort).
368.Sp
369Example: This is basically the same thing that libev does internally, too.
370.Sp
371.Vb 6
372\& static void
373\& fatal_error (const char *msg)
374\& {
375\& perror (msg);
376\& abort ();
377\& }
378.Ve
379.Sp
380.Vb 2
381\& ...
382\& ev_set_syserr_cb (fatal_error);
383.Ve
238.SH "FUNCTIONS CONTROLLING THE EVENT LOOP" 384.SH "FUNCTIONS CONTROLLING THE EVENT LOOP"
239.IX Header "FUNCTIONS CONTROLLING THE EVENT LOOP" 385.IX Header "FUNCTIONS CONTROLLING THE EVENT LOOP"
240An event loop is described by a \f(CW\*(C`struct ev_loop *\*(C'\fR. The library knows two 386An event loop is described by a \f(CW\*(C`struct ev_loop *\*(C'\fR. The library knows two
241types of such loops, the \fIdefault\fR loop, which supports signals and child 387types of such loops, the \fIdefault\fR loop, which supports signals and child
242events, and dynamically created loops which do not. 388events, and dynamically created loops which do not.
… …
274or setgid) then libev will \fInot\fR look at the environment variable 420or setgid) then libev will \fInot\fR look at the environment variable
275\&\f(CW\*(C`LIBEV_FLAGS\*(C'\fR. Otherwise (the default), this environment variable will 421\&\f(CW\*(C`LIBEV_FLAGS\*(C'\fR. Otherwise (the default), this environment variable will
276override the flags completely if it is found in the environment. This is 422override the flags completely if it is found in the environment. This is
277useful to try out specific backends to test their performance, or to work 423useful to try out specific backends to test their performance, or to work
278around bugs. 424around bugs.
425.ie n .IP """EVFLAG_FORKCHECK""" 4
426.el .IP "\f(CWEVFLAG_FORKCHECK\fR" 4
427.IX Item "EVFLAG_FORKCHECK"
428Instead of calling \f(CW\*(C`ev_default_fork\*(C'\fR or \f(CW\*(C`ev_loop_fork\*(C'\fR manually after
429a fork, you can also make libev check for a fork in each iteration by
430enabling this flag.
431.Sp
432This works by calling \f(CW\*(C`getpid ()\*(C'\fR on every iteration of the loop,
433and thus this might slow down your event loop if you do a lot of loop
434iterations and little real work, but is usually not noticeable (on my
435Linux system for example, \f(CW\*(C`getpid\*(C'\fR is actually a simple 5\-insn sequence
436without a syscall and thus \fIvery\fR fast, but my Linux system also has
437\&\f(CW\*(C`pthread_atfork\*(C'\fR which is even faster).
438.Sp
439The big advantage of this flag is that you can forget about fork (and
440forget about forgetting to tell libev about forking) when you use this
441flag.
442.Sp
443This flag setting cannot be overriden or specified in the \f(CW\*(C`LIBEV_FLAGS\*(C'\fR
444environment variable.
279.ie n .IP """EVBACKEND_SELECT"" (value 1, portable select backend)" 4 445.ie n .IP """EVBACKEND_SELECT"" (value 1, portable select backend)" 4
280.el .IP "\f(CWEVBACKEND_SELECT\fR (value 1, portable select backend)" 4 446.el .IP "\f(CWEVBACKEND_SELECT\fR (value 1, portable select backend)" 4
281.IX Item "EVBACKEND_SELECT (value 1, portable select backend)" 447.IX Item "EVBACKEND_SELECT (value 1, portable select backend)"
282This is your standard \fIselect\fR\|(2) backend. Not \fIcompletely\fR standard, as 448This is your standard \fIselect\fR\|(2) backend. Not \fIcompletely\fR standard, as
283libev tries to roll its own fd_set with no limits on the number of fds, 449libev tries to roll its own fd_set with no limits on the number of fds,
… …
376.IX Item "struct ev_loop *ev_loop_new (unsigned int flags)" 542.IX Item "struct ev_loop *ev_loop_new (unsigned int flags)"
377Similar to \f(CW\*(C`ev_default_loop\*(C'\fR, but always creates a new event loop that is 543Similar to \f(CW\*(C`ev_default_loop\*(C'\fR, but always creates a new event loop that is
378always distinct from the default loop. Unlike the default loop, it cannot 544always distinct from the default loop. Unlike the default loop, it cannot
379handle signal and child watchers, and attempts to do so will be greeted by 545handle signal and child watchers, and attempts to do so will be greeted by
380undefined behaviour (or a failed assertion if assertions are enabled). 546undefined behaviour (or a failed assertion if assertions are enabled).
547.Sp
548Example: Try to create a event loop that uses epoll and nothing else.
549.Sp
550.Vb 3
551\& struct ev_loop *epoller = ev_loop_new (EVBACKEND_EPOLL | EVFLAG_NOENV);
552\& if (!epoller)
553\& fatal ("no epoll found here, maybe it hides under your chair");
554.Ve
381.IP "ev_default_destroy ()" 4 555.IP "ev_default_destroy ()" 4
382.IX Item "ev_default_destroy ()" 556.IX Item "ev_default_destroy ()"
383Destroys the default loop again (frees all memory and kernel state 557Destroys the default loop again (frees all memory and kernel state
384etc.). This stops all registered event watchers (by not touching them in 558etc.). None of the active event watchers will be stopped in the normal
385any way whatsoever, although you cannot rely on this :). 559sense, so e.g. \f(CW\*(C`ev_is_active\*(C'\fR might still return true. It is your
560responsibility to either stop all watchers cleanly yoursef \fIbefore\fR
561calling this function, or cope with the fact afterwards (which is usually
562the easiest thing, you can just ignore the watchers and/or \f(CW\*(C`free ()\*(C'\fR them
563for example).
564.Sp
565Not that certain global state, such as signal state, will not be freed by
566this function, and related watchers (such as signal and child watchers)
567would need to be stopped manually.
568.Sp
569In general it is not advisable to call this function except in the
570rare occasion where you really need to free e.g. the signal handling
571pipe fds. If you need dynamically allocated loops it is better to use
572\&\f(CW\*(C`ev_loop_new\*(C'\fR and \f(CW\*(C`ev_loop_destroy\*(C'\fR).
386.IP "ev_loop_destroy (loop)" 4 573.IP "ev_loop_destroy (loop)" 4
387.IX Item "ev_loop_destroy (loop)" 574.IX Item "ev_loop_destroy (loop)"
388Like \f(CW\*(C`ev_default_destroy\*(C'\fR, but destroys an event loop created by an 575Like \f(CW\*(C`ev_default_destroy\*(C'\fR, but destroys an event loop created by an
389earlier call to \f(CW\*(C`ev_loop_new\*(C'\fR. 576earlier call to \f(CW\*(C`ev_loop_new\*(C'\fR.
390.IP "ev_default_fork ()" 4 577.IP "ev_default_fork ()" 4
… …
412.IP "ev_loop_fork (loop)" 4 599.IP "ev_loop_fork (loop)" 4
413.IX Item "ev_loop_fork (loop)" 600.IX Item "ev_loop_fork (loop)"
414Like \f(CW\*(C`ev_default_fork\*(C'\fR, but acts on an event loop created by 601Like \f(CW\*(C`ev_default_fork\*(C'\fR, but acts on an event loop created by
415\&\f(CW\*(C`ev_loop_new\*(C'\fR. Yes, you have to call this on every allocated event loop 602\&\f(CW\*(C`ev_loop_new\*(C'\fR. Yes, you have to call this on every allocated event loop
416after fork, and how you do this is entirely your own problem. 603after fork, and how you do this is entirely your own problem.
604.IP "unsigned int ev_loop_count (loop)" 4
605.IX Item "unsigned int ev_loop_count (loop)"
606Returns the count of loop iterations for the loop, which is identical to
607the number of times libev did poll for new events. It starts at \f(CW0\fR and
608happily wraps around with enough iterations.
609.Sp
610This value can sometimes be useful as a generation counter of sorts (it
611\&\*(L"ticks\*(R" the number of loop iterations), as it roughly corresponds with
612\&\f(CW\*(C`ev_prepare\*(C'\fR and \f(CW\*(C`ev_check\*(C'\fR calls.
417.IP "unsigned int ev_backend (loop)" 4 613.IP "unsigned int ev_backend (loop)" 4
418.IX Item "unsigned int ev_backend (loop)" 614.IX Item "unsigned int ev_backend (loop)"
419Returns one of the \f(CW\*(C`EVBACKEND_*\*(C'\fR flags indicating the event backend in 615Returns one of the \f(CW\*(C`EVBACKEND_*\*(C'\fR flags indicating the event backend in
420use. 616use.
421.IP "ev_tstamp ev_now (loop)" 4 617.IP "ev_tstamp ev_now (loop)" 4
422.IX Item "ev_tstamp ev_now (loop)" 618.IX Item "ev_tstamp ev_now (loop)"
423Returns the current \*(L"event loop time\*(R", which is the time the event loop 619Returns the current \*(L"event loop time\*(R", which is the time the event loop
424got events and started processing them. This timestamp does not change 620received events and started processing them. This timestamp does not
425as long as callbacks are being processed, and this is also the base time 621change as long as callbacks are being processed, and this is also the base
426used for relative timers. You can treat it as the timestamp of the event 622time used for relative timers. You can treat it as the timestamp of the
427occuring (or more correctly, the mainloop finding out about it). 623event occuring (or more correctly, libev finding out about it).
428.IP "ev_loop (loop, int flags)" 4 624.IP "ev_loop (loop, int flags)" 4
429.IX Item "ev_loop (loop, int flags)" 625.IX Item "ev_loop (loop, int flags)"
430Finally, this is it, the event handler. This function usually is called 626Finally, this is it, the event handler. This function usually is called
431after you initialised all your watchers and you want to start handling 627after you initialised all your watchers and you want to start handling
432events. 628events.
433.Sp 629.Sp
434If the flags argument is specified as \f(CW0\fR, it will not return until 630If the flags argument is specified as \f(CW0\fR, it will not return until
435either no event watchers are active anymore or \f(CW\*(C`ev_unloop\*(C'\fR was called. 631either no event watchers are active anymore or \f(CW\*(C`ev_unloop\*(C'\fR was called.
632.Sp
633Please note that an explicit \f(CW\*(C`ev_unloop\*(C'\fR is usually better than
634relying on all watchers to be stopped when deciding when a program has
635finished (especially in interactive programs), but having a program that
636automatically loops as long as it has to and no longer by virtue of
637relying on its watchers stopping correctly is a thing of beauty.
436.Sp 638.Sp
437A flags value of \f(CW\*(C`EVLOOP_NONBLOCK\*(C'\fR will look for new events, will handle 639A flags value of \f(CW\*(C`EVLOOP_NONBLOCK\*(C'\fR will look for new events, will handle
438those events and any outstanding ones, but will not block your process in 640those events and any outstanding ones, but will not block your process in
439case there are no events and will return after one iteration of the loop. 641case there are no events and will return after one iteration of the loop.
440.Sp 642.Sp
… …
446libev watchers. However, a pair of \f(CW\*(C`ev_prepare\*(C'\fR/\f(CW\*(C`ev_check\*(C'\fR watchers is 648libev watchers. However, a pair of \f(CW\*(C`ev_prepare\*(C'\fR/\f(CW\*(C`ev_check\*(C'\fR watchers is
447usually a better approach for this kind of thing. 649usually a better approach for this kind of thing.
448.Sp 650.Sp
449Here are the gory details of what \f(CW\*(C`ev_loop\*(C'\fR does: 651Here are the gory details of what \f(CW\*(C`ev_loop\*(C'\fR does:
450.Sp 652.Sp
451.Vb 18 653.Vb 19
654\& - Before the first iteration, call any pending watchers.
452\& * If there are no active watchers (reference count is zero), return. 655\& * If there are no active watchers (reference count is zero), return.
453\& - Queue prepare watchers and then call all outstanding watchers. 656\& - Queue all prepare watchers and then call all outstanding watchers.
454\& - If we have been forked, recreate the kernel state. 657\& - If we have been forked, recreate the kernel state.
455\& - Update the kernel state with all outstanding changes. 658\& - Update the kernel state with all outstanding changes.
456\& - Update the "event loop time". 659\& - Update the "event loop time".
457\& - Calculate for how long to block. 660\& - Calculate for how long to block.
458\& - Block the process, waiting for any events. 661\& - Block the process, waiting for any events.
… …
465\& - Call all queued watchers in reverse order (i.e. check watchers first). 668\& - Call all queued watchers in reverse order (i.e. check watchers first).
466\& Signals and child watchers are implemented as I/O watchers, and will 669\& Signals and child watchers are implemented as I/O watchers, and will
467\& be handled here by queueing them when their watcher gets executed. 670\& be handled here by queueing them when their watcher gets executed.
468\& - If ev_unloop has been called or EVLOOP_ONESHOT or EVLOOP_NONBLOCK 671\& - If ev_unloop has been called or EVLOOP_ONESHOT or EVLOOP_NONBLOCK
469\& were used, return, otherwise continue with step *. 672\& were used, return, otherwise continue with step *.
673.Ve
674.Sp
675Example: Queue some jobs and then loop until no events are outsanding
676anymore.
677.Sp
678.Vb 4
679\& ... queue jobs here, make sure they register event watchers as long
680\& ... as they still have work to do (even an idle watcher will do..)
681\& ev_loop (my_loop, 0);
682\& ... jobs done. yeah!
470.Ve 683.Ve
471.IP "ev_unloop (loop, how)" 4 684.IP "ev_unloop (loop, how)" 4
472.IX Item "ev_unloop (loop, how)" 685.IX Item "ev_unloop (loop, how)"
473Can be used to make a call to \f(CW\*(C`ev_loop\*(C'\fR return early (but only after it 686Can be used to make a call to \f(CW\*(C`ev_loop\*(C'\fR return early (but only after it
474has processed all outstanding events). The \f(CW\*(C`how\*(C'\fR argument must be either 687has processed all outstanding events). The \f(CW\*(C`how\*(C'\fR argument must be either
… …
488example, libev itself uses this for its internal signal pipe: It is not 701example, libev itself uses this for its internal signal pipe: It is not
489visible to the libev user and should not keep \f(CW\*(C`ev_loop\*(C'\fR from exiting if 702visible to the libev user and should not keep \f(CW\*(C`ev_loop\*(C'\fR from exiting if
490no event watchers registered by it are active. It is also an excellent 703no event watchers registered by it are active. It is also an excellent
491way to do this for generic recurring timers or from within third-party 704way to do this for generic recurring timers or from within third-party
492libraries. Just remember to \fIunref after start\fR and \fIref before stop\fR. 705libraries. Just remember to \fIunref after start\fR and \fIref before stop\fR.
706.Sp
707Example: Create a signal watcher, but keep it from keeping \f(CW\*(C`ev_loop\*(C'\fR
708running when nothing else is active.
709.Sp
710.Vb 4
711\& struct ev_signal exitsig;
712\& ev_signal_init (&exitsig, sig_cb, SIGINT);
713\& ev_signal_start (loop, &exitsig);
714\& evf_unref (loop);
715.Ve
716.Sp
717Example: For some weird reason, unregister the above signal handler again.
718.Sp
719.Vb 2
720\& ev_ref (loop);
721\& ev_signal_stop (loop, &exitsig);
722.Ve
493.SH "ANATOMY OF A WATCHER" 723.SH "ANATOMY OF A WATCHER"
494.IX Header "ANATOMY OF A WATCHER" 724.IX Header "ANATOMY OF A WATCHER"
495A watcher is a structure that you create and register to record your 725A watcher is a structure that you create and register to record your
496interest in some event. For instance, if you want to wait for \s-1STDIN\s0 to 726interest in some event. For instance, if you want to wait for \s-1STDIN\s0 to
497become readable, you would create an \f(CW\*(C`ev_io\*(C'\fR watcher for that: 727become readable, you would create an \f(CW\*(C`ev_io\*(C'\fR watcher for that:
… …
533*)\*(C'\fR), and you can stop watching for events at any time by calling the 763*)\*(C'\fR), and you can stop watching for events at any time by calling the
534corresponding stop function (\f(CW\*(C`ev_<type>_stop (loop, watcher *)\*(C'\fR. 764corresponding stop function (\f(CW\*(C`ev_<type>_stop (loop, watcher *)\*(C'\fR.
535.PP 765.PP
536As long as your watcher is active (has been started but not stopped) you 766As long as your watcher is active (has been started but not stopped) you
537must not touch the values stored in it. Most specifically you must never 767must not touch the values stored in it. Most specifically you must never
538reinitialise it or call its set macro. 768reinitialise it or call its \f(CW\*(C`set\*(C'\fR macro.
539.PP
540You can check whether an event is active by calling the \f(CW\*(C`ev_is_active
541(watcher *)\*(C'\fR macro. To see whether an event is outstanding (but the
542callback for it has not been called yet) you can use the \f(CW\*(C`ev_is_pending
543(watcher *)\*(C'\fR macro.
544.PP 769.PP
545Each and every callback receives the event loop pointer as first, the 770Each and every callback receives the event loop pointer as first, the
546registered watcher structure as second, and a bitset of received events as 771registered watcher structure as second, and a bitset of received events as
547third argument. 772third argument.
548.PP 773.PP
… …
573The signal specified in the \f(CW\*(C`ev_signal\*(C'\fR watcher has been received by a thread. 798The signal specified in the \f(CW\*(C`ev_signal\*(C'\fR watcher has been received by a thread.
574.ie n .IP """EV_CHILD""" 4 799.ie n .IP """EV_CHILD""" 4
575.el .IP "\f(CWEV_CHILD\fR" 4 800.el .IP "\f(CWEV_CHILD\fR" 4
576.IX Item "EV_CHILD" 801.IX Item "EV_CHILD"
577The pid specified in the \f(CW\*(C`ev_child\*(C'\fR watcher has received a status change. 802The pid specified in the \f(CW\*(C`ev_child\*(C'\fR watcher has received a status change.
803.ie n .IP """EV_STAT""" 4
804.el .IP "\f(CWEV_STAT\fR" 4
805.IX Item "EV_STAT"
806The path specified in the \f(CW\*(C`ev_stat\*(C'\fR watcher changed its attributes somehow.
578.ie n .IP """EV_IDLE""" 4 807.ie n .IP """EV_IDLE""" 4
579.el .IP "\f(CWEV_IDLE\fR" 4 808.el .IP "\f(CWEV_IDLE\fR" 4
580.IX Item "EV_IDLE" 809.IX Item "EV_IDLE"
581The \f(CW\*(C`ev_idle\*(C'\fR watcher has determined that you have nothing better to do. 810The \f(CW\*(C`ev_idle\*(C'\fR watcher has determined that you have nothing better to do.
582.ie n .IP """EV_PREPARE""" 4 811.ie n .IP """EV_PREPARE""" 4
… …
592\&\f(CW\*(C`ev_loop\*(C'\fR has gathered them, but before it invokes any callbacks for any 821\&\f(CW\*(C`ev_loop\*(C'\fR has gathered them, but before it invokes any callbacks for any
593received events. Callbacks of both watcher types can start and stop as 822received events. Callbacks of both watcher types can start and stop as
594many watchers as they want, and all of them will be taken into account 823many watchers as they want, and all of them will be taken into account
595(for example, a \f(CW\*(C`ev_prepare\*(C'\fR watcher might start an idle watcher to keep 824(for example, a \f(CW\*(C`ev_prepare\*(C'\fR watcher might start an idle watcher to keep
596\&\f(CW\*(C`ev_loop\*(C'\fR from blocking). 825\&\f(CW\*(C`ev_loop\*(C'\fR from blocking).
826.ie n .IP """EV_EMBED""" 4
827.el .IP "\f(CWEV_EMBED\fR" 4
828.IX Item "EV_EMBED"
829The embedded event loop specified in the \f(CW\*(C`ev_embed\*(C'\fR watcher needs attention.
830.ie n .IP """EV_FORK""" 4
831.el .IP "\f(CWEV_FORK\fR" 4
832.IX Item "EV_FORK"
833The event loop has been resumed in the child process after fork (see
834\&\f(CW\*(C`ev_fork\*(C'\fR).
597.ie n .IP """EV_ERROR""" 4 835.ie n .IP """EV_ERROR""" 4
598.el .IP "\f(CWEV_ERROR\fR" 4 836.el .IP "\f(CWEV_ERROR\fR" 4
599.IX Item "EV_ERROR" 837.IX Item "EV_ERROR"
600An unspecified error has occured, the watcher has been stopped. This might 838An unspecified error has occured, the watcher has been stopped. This might
601happen because the watcher could not be properly started because libev 839happen because the watcher could not be properly started because libev
… …
606Libev will usually signal a few \*(L"dummy\*(R" events together with an error, 844Libev will usually signal a few \*(L"dummy\*(R" events together with an error,
607for example it might indicate that a fd is readable or writable, and if 845for example it might indicate that a fd is readable or writable, and if
608your callbacks is well-written it can just attempt the operation and cope 846your callbacks is well-written it can just attempt the operation and cope
609with the error from \fIread()\fR or \fIwrite()\fR. This will not work in multithreaded 847with the error from \fIread()\fR or \fIwrite()\fR. This will not work in multithreaded
610programs, though, so beware. 848programs, though, so beware.
849.Sh "\s-1GENERIC\s0 \s-1WATCHER\s0 \s-1FUNCTIONS\s0"
850.IX Subsection "GENERIC WATCHER FUNCTIONS"
851In the following description, \f(CW\*(C`TYPE\*(C'\fR stands for the watcher type,
852e.g. \f(CW\*(C`timer\*(C'\fR for \f(CW\*(C`ev_timer\*(C'\fR watchers and \f(CW\*(C`io\*(C'\fR for \f(CW\*(C`ev_io\*(C'\fR watchers.
853.ie n .IP """ev_init"" (ev_TYPE *watcher, callback)" 4
854.el .IP "\f(CWev_init\fR (ev_TYPE *watcher, callback)" 4
855.IX Item "ev_init (ev_TYPE *watcher, callback)"
856This macro initialises the generic portion of a watcher. The contents
857of the watcher object can be arbitrary (so \f(CW\*(C`malloc\*(C'\fR will do). Only
858the generic parts of the watcher are initialised, you \fIneed\fR to call
859the type-specific \f(CW\*(C`ev_TYPE_set\*(C'\fR macro afterwards to initialise the
860type-specific parts. For each type there is also a \f(CW\*(C`ev_TYPE_init\*(C'\fR macro
861which rolls both calls into one.
862.Sp
863You can reinitialise a watcher at any time as long as it has been stopped
864(or never started) and there are no pending events outstanding.
865.Sp
866The callback is always of type \f(CW\*(C`void (*)(ev_loop *loop, ev_TYPE *watcher,
867int revents)\*(C'\fR.
868.ie n .IP """ev_TYPE_set"" (ev_TYPE *, [args])" 4
869.el .IP "\f(CWev_TYPE_set\fR (ev_TYPE *, [args])" 4
870.IX Item "ev_TYPE_set (ev_TYPE *, [args])"
871This macro initialises the type-specific parts of a watcher. You need to
872call \f(CW\*(C`ev_init\*(C'\fR at least once before you call this macro, but you can
873call \f(CW\*(C`ev_TYPE_set\*(C'\fR any number of times. You must not, however, call this
874macro on a watcher that is active (it can be pending, however, which is a
875difference to the \f(CW\*(C`ev_init\*(C'\fR macro).
876.Sp
877Although some watcher types do not have type-specific arguments
878(e.g. \f(CW\*(C`ev_prepare\*(C'\fR) you still need to call its \f(CW\*(C`set\*(C'\fR macro.
879.ie n .IP """ev_TYPE_init"" (ev_TYPE *watcher, callback, [args])" 4
880.el .IP "\f(CWev_TYPE_init\fR (ev_TYPE *watcher, callback, [args])" 4
881.IX Item "ev_TYPE_init (ev_TYPE *watcher, callback, [args])"
882This convinience macro rolls both \f(CW\*(C`ev_init\*(C'\fR and \f(CW\*(C`ev_TYPE_set\*(C'\fR macro
883calls into a single call. This is the most convinient method to initialise
884a watcher. The same limitations apply, of course.
885.ie n .IP """ev_TYPE_start"" (loop *, ev_TYPE *watcher)" 4
886.el .IP "\f(CWev_TYPE_start\fR (loop *, ev_TYPE *watcher)" 4
887.IX Item "ev_TYPE_start (loop *, ev_TYPE *watcher)"
888Starts (activates) the given watcher. Only active watchers will receive
889events. If the watcher is already active nothing will happen.
890.ie n .IP """ev_TYPE_stop"" (loop *, ev_TYPE *watcher)" 4
891.el .IP "\f(CWev_TYPE_stop\fR (loop *, ev_TYPE *watcher)" 4
892.IX Item "ev_TYPE_stop (loop *, ev_TYPE *watcher)"
893Stops the given watcher again (if active) and clears the pending
894status. It is possible that stopped watchers are pending (for example,
895non-repeating timers are being stopped when they become pending), but
896\&\f(CW\*(C`ev_TYPE_stop\*(C'\fR ensures that the watcher is neither active nor pending. If
897you want to free or reuse the memory used by the watcher it is therefore a
898good idea to always call its \f(CW\*(C`ev_TYPE_stop\*(C'\fR function.
899.IP "bool ev_is_active (ev_TYPE *watcher)" 4
900.IX Item "bool ev_is_active (ev_TYPE *watcher)"
901Returns a true value iff the watcher is active (i.e. it has been started
902and not yet been stopped). As long as a watcher is active you must not modify
903it.
904.IP "bool ev_is_pending (ev_TYPE *watcher)" 4
905.IX Item "bool ev_is_pending (ev_TYPE *watcher)"
906Returns a true value iff the watcher is pending, (i.e. it has outstanding
907events but its callback has not yet been invoked). As long as a watcher
908is pending (but not active) you must not call an init function on it (but
909\&\f(CW\*(C`ev_TYPE_set\*(C'\fR is safe), you must not change its priority, and you must
910make sure the watcher is available to libev (e.g. you cannot \f(CW\*(C`free ()\*(C'\fR
911it).
912.IP "callback ev_cb (ev_TYPE *watcher)" 4
913.IX Item "callback ev_cb (ev_TYPE *watcher)"
914Returns the callback currently set on the watcher.
915.IP "ev_cb_set (ev_TYPE *watcher, callback)" 4
916.IX Item "ev_cb_set (ev_TYPE *watcher, callback)"
917Change the callback. You can change the callback at virtually any time
918(modulo threads).
919.IP "ev_set_priority (ev_TYPE *watcher, priority)" 4
920.IX Item "ev_set_priority (ev_TYPE *watcher, priority)"
921.PD 0
922.IP "int ev_priority (ev_TYPE *watcher)" 4
923.IX Item "int ev_priority (ev_TYPE *watcher)"
924.PD
925Set and query the priority of the watcher. The priority is a small
926integer between \f(CW\*(C`EV_MAXPRI\*(C'\fR (default: \f(CW2\fR) and \f(CW\*(C`EV_MINPRI\*(C'\fR
927(default: \f(CW\*(C`\-2\*(C'\fR). Pending watchers with higher priority will be invoked
928before watchers with lower priority, but priority will not keep watchers
929from being executed (except for \f(CW\*(C`ev_idle\*(C'\fR watchers).
930.Sp
931This means that priorities are \fIonly\fR used for ordering callback
932invocation after new events have been received. This is useful, for
933example, to reduce latency after idling, or more often, to bind two
934watchers on the same event and make sure one is called first.
935.Sp
936If you need to suppress invocation when higher priority events are pending
937you need to look at \f(CW\*(C`ev_idle\*(C'\fR watchers, which provide this functionality.
938.Sp
939You \fImust not\fR change the priority of a watcher as long as it is active or
940pending.
941.Sp
942The default priority used by watchers when no priority has been set is
943always \f(CW0\fR, which is supposed to not be too high and not be too low :).
944.Sp
945Setting a priority outside the range of \f(CW\*(C`EV_MINPRI\*(C'\fR to \f(CW\*(C`EV_MAXPRI\*(C'\fR is
946fine, as long as you do not mind that the priority value you query might
947or might not have been adjusted to be within valid range.
948.IP "ev_invoke (loop, ev_TYPE *watcher, int revents)" 4
949.IX Item "ev_invoke (loop, ev_TYPE *watcher, int revents)"
950Invoke the \f(CW\*(C`watcher\*(C'\fR with the given \f(CW\*(C`loop\*(C'\fR and \f(CW\*(C`revents\*(C'\fR. Neither
951\&\f(CW\*(C`loop\*(C'\fR nor \f(CW\*(C`revents\*(C'\fR need to be valid as long as the watcher callback
952can deal with that fact.
953.IP "int ev_clear_pending (loop, ev_TYPE *watcher)" 4
954.IX Item "int ev_clear_pending (loop, ev_TYPE *watcher)"
955If the watcher is pending, this function returns clears its pending status
956and returns its \f(CW\*(C`revents\*(C'\fR bitset (as if its callback was invoked). If the
957watcher isn't pending it does nothing and returns \f(CW0\fR.
611.Sh "\s-1ASSOCIATING\s0 \s-1CUSTOM\s0 \s-1DATA\s0 \s-1WITH\s0 A \s-1WATCHER\s0" 958.Sh "\s-1ASSOCIATING\s0 \s-1CUSTOM\s0 \s-1DATA\s0 \s-1WITH\s0 A \s-1WATCHER\s0"
612.IX Subsection "ASSOCIATING CUSTOM DATA WITH A WATCHER" 959.IX Subsection "ASSOCIATING CUSTOM DATA WITH A WATCHER"
613Each watcher has, by default, a member \f(CW\*(C`void *data\*(C'\fR that you can change 960Each watcher has, by default, a member \f(CW\*(C`void *data\*(C'\fR that you can change
614and read at any time, libev will completely ignore it. This can be used 961and read at any time, libev will completely ignore it. This can be used
615to associate arbitrary data with your watcher. If you need more data and 962to associate arbitrary data with your watcher. If you need more data and
… …
636\& struct my_io *w = (struct my_io *)w_; 983\& struct my_io *w = (struct my_io *)w_;
637\& ... 984\& ...
638\& } 985\& }
639.Ve 986.Ve
640.PP 987.PP
641More interesting and less C\-conformant ways of catsing your callback type 988More interesting and less C\-conformant ways of casting your callback type
642have been omitted.... 989instead have been omitted.
990.PP
991Another common scenario is having some data structure with multiple
992watchers:
993.PP
994.Vb 6
995\& struct my_biggy
996\& {
997\& int some_data;
998\& ev_timer t1;
999\& ev_timer t2;
1000\& }
1001.Ve
1002.PP
1003In this case getting the pointer to \f(CW\*(C`my_biggy\*(C'\fR is a bit more complicated,
1004you need to use \f(CW\*(C`offsetof\*(C'\fR:
1005.PP
1006.Vb 1
1007\& #include <stddef.h>
1008.Ve
1009.PP
1010.Vb 6
1011\& static void
1012\& t1_cb (EV_P_ struct ev_timer *w, int revents)
1013\& {
1014\& struct my_biggy big = (struct my_biggy *
1015\& (((char *)w) - offsetof (struct my_biggy, t1));
1016\& }
1017.Ve
1018.PP
1019.Vb 6
1020\& static void
1021\& t2_cb (EV_P_ struct ev_timer *w, int revents)
1022\& {
1023\& struct my_biggy big = (struct my_biggy *
1024\& (((char *)w) - offsetof (struct my_biggy, t2));
1025\& }
1026.Ve
643.SH "WATCHER TYPES" 1027.SH "WATCHER TYPES"
644.IX Header "WATCHER TYPES" 1028.IX Header "WATCHER TYPES"
645This section describes each watcher in detail, but will not repeat 1029This section describes each watcher in detail, but will not repeat
646information given in the last section. 1030information given in the last section. Any initialisation/set macros,
1031functions and members specific to the watcher type are explained.
1032.PP
1033Members are additionally marked with either \fI[read\-only]\fR, meaning that,
1034while the watcher is active, you can look at the member and expect some
1035sensible content, but you must not modify it (you can modify it while the
1036watcher is stopped to your hearts content), or \fI[read\-write]\fR, which
1037means you can expect it to have some sensible content while the watcher
1038is active, but you can also modify it. Modifying it may not do something
1039sensible or take immediate effect (or do anything at all), but libev will
1040not crash or malfunction in any way.
647.ie n .Sh """ev_io"" \- is this file descriptor readable or writable" 1041.ie n .Sh """ev_io"" \- is this file descriptor readable or writable?"
648.el .Sh "\f(CWev_io\fP \- is this file descriptor readable or writable" 1042.el .Sh "\f(CWev_io\fP \- is this file descriptor readable or writable?"
649.IX Subsection "ev_io - is this file descriptor readable or writable" 1043.IX Subsection "ev_io - is this file descriptor readable or writable?"
650I/O watchers check whether a file descriptor is readable or writable 1044I/O watchers check whether a file descriptor is readable or writable
651in each iteration of the event loop (This behaviour is called 1045in each iteration of the event loop, or, more precisely, when reading
652level-triggering because you keep receiving events as long as the 1046would not block the process and writing would at least be able to write
653condition persists. Remember you can stop the watcher if you don't want to 1047some data. This behaviour is called level-triggering because you keep
654act on the event and neither want to receive future events). 1048receiving events as long as the condition persists. Remember you can stop
1049the watcher if you don't want to act on the event and neither want to
1050receive future events.
655.PP 1051.PP
656In general you can register as many read and/or write event watchers per 1052In general you can register as many read and/or write event watchers per
657fd as you want (as long as you don't confuse yourself). Setting all file 1053fd as you want (as long as you don't confuse yourself). Setting all file
658descriptors to non-blocking mode is also usually a good idea (but not 1054descriptors to non-blocking mode is also usually a good idea (but not
659required if you know what you are doing). 1055required if you know what you are doing).
660.PP 1056.PP
661You have to be careful with dup'ed file descriptors, though. Some backends 1057You have to be careful with dup'ed file descriptors, though. Some backends
662(the linux epoll backend is a notable example) cannot handle dup'ed file 1058(the linux epoll backend is a notable example) cannot handle dup'ed file
663descriptors correctly if you register interest in two or more fds pointing 1059descriptors correctly if you register interest in two or more fds pointing
664to the same underlying file/socket etc. description (that is, they share 1060to the same underlying file/socket/etc. description (that is, they share
665the same underlying \*(L"file open\*(R"). 1061the same underlying \*(L"file open\*(R").
666.PP 1062.PP
667If you must do this, then force the use of a known-to-be-good backend 1063If you must do this, then force the use of a known-to-be-good backend
668(at the time of this writing, this includes only \f(CW\*(C`EVBACKEND_SELECT\*(C'\fR and 1064(at the time of this writing, this includes only \f(CW\*(C`EVBACKEND_SELECT\*(C'\fR and
669\&\f(CW\*(C`EVBACKEND_POLL\*(C'\fR). 1065\&\f(CW\*(C`EVBACKEND_POLL\*(C'\fR).
1066.PP
1067Another thing you have to watch out for is that it is quite easy to
1068receive \*(L"spurious\*(R" readyness notifications, that is your callback might
1069be called with \f(CW\*(C`EV_READ\*(C'\fR but a subsequent \f(CW\*(C`read\*(C'\fR(2) will actually block
1070because there is no data. Not only are some backends known to create a
1071lot of those (for example solaris ports), it is very easy to get into
1072this situation even with a relatively standard program structure. Thus
1073it is best to always use non-blocking I/O: An extra \f(CW\*(C`read\*(C'\fR(2) returning
1074\&\f(CW\*(C`EAGAIN\*(C'\fR is far preferable to a program hanging until some data arrives.
1075.PP
1076If you cannot run the fd in non-blocking mode (for example you should not
1077play around with an Xlib connection), then you have to seperately re-test
1078whether a file descriptor is really ready with a known-to-be good interface
1079such as poll (fortunately in our Xlib example, Xlib already does this on
1080its own, so its quite safe to use).
1081.PP
1082\fIThe special problem of disappearing file descriptors\fR
1083.IX Subsection "The special problem of disappearing file descriptors"
1084.PP
1085Some backends (e.g kqueue, epoll) need to be told about closing a file
1086descriptor (either by calling \f(CW\*(C`close\*(C'\fR explicitly or by any other means,
1087such as \f(CW\*(C`dup\*(C'\fR). The reason is that you register interest in some file
1088descriptor, but when it goes away, the operating system will silently drop
1089this interest. If another file descriptor with the same number then is
1090registered with libev, there is no efficient way to see that this is, in
1091fact, a different file descriptor.
1092.PP
1093To avoid having to explicitly tell libev about such cases, libev follows
1094the following policy: Each time \f(CW\*(C`ev_io_set\*(C'\fR is being called, libev
1095will assume that this is potentially a new file descriptor, otherwise
1096it is assumed that the file descriptor stays the same. That means that
1097you \fIhave\fR to call \f(CW\*(C`ev_io_set\*(C'\fR (or \f(CW\*(C`ev_io_init\*(C'\fR) when you change the
1098descriptor even if the file descriptor number itself did not change.
1099.PP
1100This is how one would do it normally anyway, the important point is that
1101the libev application should not optimise around libev but should leave
1102optimisations to libev.
1103.PP
1104\fIWatcher-Specific Functions\fR
1105.IX Subsection "Watcher-Specific Functions"
670.IP "ev_io_init (ev_io *, callback, int fd, int events)" 4 1106.IP "ev_io_init (ev_io *, callback, int fd, int events)" 4
671.IX Item "ev_io_init (ev_io *, callback, int fd, int events)" 1107.IX Item "ev_io_init (ev_io *, callback, int fd, int events)"
672.PD 0 1108.PD 0
673.IP "ev_io_set (ev_io *, int fd, int events)" 4 1109.IP "ev_io_set (ev_io *, int fd, int events)" 4
674.IX Item "ev_io_set (ev_io *, int fd, int events)" 1110.IX Item "ev_io_set (ev_io *, int fd, int events)"
675.PD 1111.PD
676Configures an \f(CW\*(C`ev_io\*(C'\fR watcher. The fd is the file descriptor to rceeive 1112Configures an \f(CW\*(C`ev_io\*(C'\fR watcher. The \f(CW\*(C`fd\*(C'\fR is the file descriptor to
677events for and events is either \f(CW\*(C`EV_READ\*(C'\fR, \f(CW\*(C`EV_WRITE\*(C'\fR or \f(CW\*(C`EV_READ | 1113rceeive events for and events is either \f(CW\*(C`EV_READ\*(C'\fR, \f(CW\*(C`EV_WRITE\*(C'\fR or
678EV_WRITE\*(C'\fR to receive the given events. 1114\&\f(CW\*(C`EV_READ | EV_WRITE\*(C'\fR to receive the given events.
679.Sp 1115.IP "int fd [read\-only]" 4
680Please note that most of the more scalable backend mechanisms (for example 1116.IX Item "int fd [read-only]"
681epoll and solaris ports) can result in spurious readyness notifications 1117The file descriptor being watched.
682for file descriptors, so you practically need to use non-blocking I/O (and 1118.IP "int events [read\-only]" 4
683treat callback invocation as hint only), or retest separately with a safe 1119.IX Item "int events [read-only]"
684interface before doing I/O (XLib can do this), or force the use of either 1120The events being watched.
685\&\f(CW\*(C`EVBACKEND_SELECT\*(C'\fR or \f(CW\*(C`EVBACKEND_POLL\*(C'\fR, which don't suffer from this 1121.PP
686problem. Also note that it is quite easy to have your callback invoked 1122Example: Call \f(CW\*(C`stdin_readable_cb\*(C'\fR when \s-1STDIN_FILENO\s0 has become, well
687when the readyness condition is no longer valid even when employing 1123readable, but only once. Since it is likely line\-buffered, you could
688typical ways of handling events, so its a good idea to use non-blocking 1124attempt to read a whole line in the callback.
689I/O unconditionally. 1125.PP
1126.Vb 6
1127\& static void
1128\& stdin_readable_cb (struct ev_loop *loop, struct ev_io *w, int revents)
1129\& {
1130\& ev_io_stop (loop, w);
1131\& .. read from stdin here (or from w->fd) and haqndle any I/O errors
1132\& }
1133.Ve
1134.PP
1135.Vb 6
1136\& ...
1137\& struct ev_loop *loop = ev_default_init (0);
1138\& struct ev_io stdin_readable;
1139\& ev_io_init (&stdin_readable, stdin_readable_cb, STDIN_FILENO, EV_READ);
1140\& ev_io_start (loop, &stdin_readable);
1141\& ev_loop (loop, 0);
1142.Ve
690.ie n .Sh """ev_timer"" \- relative and optionally recurring timeouts" 1143.ie n .Sh """ev_timer"" \- relative and optionally repeating timeouts"
691.el .Sh "\f(CWev_timer\fP \- relative and optionally recurring timeouts" 1144.el .Sh "\f(CWev_timer\fP \- relative and optionally repeating timeouts"
692.IX Subsection "ev_timer - relative and optionally recurring timeouts" 1145.IX Subsection "ev_timer - relative and optionally repeating timeouts"
693Timer watchers are simple relative timers that generate an event after a 1146Timer watchers are simple relative timers that generate an event after a
694given time, and optionally repeating in regular intervals after that. 1147given time, and optionally repeating in regular intervals after that.
695.PP 1148.PP
696The timers are based on real time, that is, if you register an event that 1149The timers are based on real time, that is, if you register an event that
697times out after an hour and you reset your system clock to last years 1150times out after an hour and you reset your system clock to last years
… …
710.Ve 1163.Ve
711.PP 1164.PP
712The callback is guarenteed to be invoked only when its timeout has passed, 1165The callback is guarenteed to be invoked only when its timeout has passed,
713but if multiple timers become ready during the same loop iteration then 1166but if multiple timers become ready during the same loop iteration then
714order of execution is undefined. 1167order of execution is undefined.
1168.PP
1169\fIWatcher-Specific Functions and Data Members\fR
1170.IX Subsection "Watcher-Specific Functions and Data Members"
715.IP "ev_timer_init (ev_timer *, callback, ev_tstamp after, ev_tstamp repeat)" 4 1171.IP "ev_timer_init (ev_timer *, callback, ev_tstamp after, ev_tstamp repeat)" 4
716.IX Item "ev_timer_init (ev_timer *, callback, ev_tstamp after, ev_tstamp repeat)" 1172.IX Item "ev_timer_init (ev_timer *, callback, ev_tstamp after, ev_tstamp repeat)"
717.PD 0 1173.PD 0
718.IP "ev_timer_set (ev_timer *, ev_tstamp after, ev_tstamp repeat)" 4 1174.IP "ev_timer_set (ev_timer *, ev_tstamp after, ev_tstamp repeat)" 4
719.IX Item "ev_timer_set (ev_timer *, ev_tstamp after, ev_tstamp repeat)" 1175.IX Item "ev_timer_set (ev_timer *, ev_tstamp after, ev_tstamp repeat)"
… …
731.IP "ev_timer_again (loop)" 4 1187.IP "ev_timer_again (loop)" 4
732.IX Item "ev_timer_again (loop)" 1188.IX Item "ev_timer_again (loop)"
733This will act as if the timer timed out and restart it again if it is 1189This will act as if the timer timed out and restart it again if it is
734repeating. The exact semantics are: 1190repeating. The exact semantics are:
735.Sp 1191.Sp
1192If the timer is pending, its pending status is cleared.
1193.Sp
736If the timer is started but nonrepeating, stop it. 1194If the timer is started but nonrepeating, stop it (as if it timed out).
737.Sp 1195.Sp
738If the timer is repeating, either start it if necessary (with the repeat 1196If the timer is repeating, either start it if necessary (with the
739value), or reset the running timer to the repeat value. 1197\&\f(CW\*(C`repeat\*(C'\fR value), or reset the running timer to the \f(CW\*(C`repeat\*(C'\fR value.
740.Sp 1198.Sp
741This sounds a bit complicated, but here is a useful and typical 1199This sounds a bit complicated, but here is a useful and typical
742example: Imagine you have a tcp connection and you want a so-called idle 1200example: Imagine you have a tcp connection and you want a so-called idle
743timeout, that is, you want to be called when there have been, say, 60 1201timeout, that is, you want to be called when there have been, say, 60
744seconds of inactivity on the socket. The easiest way to do this is to 1202seconds of inactivity on the socket. The easiest way to do this is to
745configure an \f(CW\*(C`ev_timer\*(C'\fR with after=repeat=60 and calling ev_timer_again each 1203configure an \f(CW\*(C`ev_timer\*(C'\fR with a \f(CW\*(C`repeat\*(C'\fR value of \f(CW60\fR and then call
746time you successfully read or write some data. If you go into an idle 1204\&\f(CW\*(C`ev_timer_again\*(C'\fR each time you successfully read or write some data. If
747state where you do not expect data to travel on the socket, you can stop 1205you go into an idle state where you do not expect data to travel on the
748the timer, and again will automatically restart it if need be. 1206socket, you can \f(CW\*(C`ev_timer_stop\*(C'\fR the timer, and \f(CW\*(C`ev_timer_again\*(C'\fR will
1207automatically restart it if need be.
1208.Sp
1209That means you can ignore the \f(CW\*(C`after\*(C'\fR value and \f(CW\*(C`ev_timer_start\*(C'\fR
1210altogether and only ever use the \f(CW\*(C`repeat\*(C'\fR value and \f(CW\*(C`ev_timer_again\*(C'\fR:
1211.Sp
1212.Vb 8
1213\& ev_timer_init (timer, callback, 0., 5.);
1214\& ev_timer_again (loop, timer);
1215\& ...
1216\& timer->again = 17.;
1217\& ev_timer_again (loop, timer);
1218\& ...
1219\& timer->again = 10.;
1220\& ev_timer_again (loop, timer);
1221.Ve
1222.Sp
1223This is more slightly efficient then stopping/starting the timer each time
1224you want to modify its timeout value.
1225.IP "ev_tstamp repeat [read\-write]" 4
1226.IX Item "ev_tstamp repeat [read-write]"
1227The current \f(CW\*(C`repeat\*(C'\fR value. Will be used each time the watcher times out
1228or \f(CW\*(C`ev_timer_again\*(C'\fR is called and determines the next timeout (if any),
1229which is also when any modifications are taken into account.
1230.PP
1231Example: Create a timer that fires after 60 seconds.
1232.PP
1233.Vb 5
1234\& static void
1235\& one_minute_cb (struct ev_loop *loop, struct ev_timer *w, int revents)
1236\& {
1237\& .. one minute over, w is actually stopped right here
1238\& }
1239.Ve
1240.PP
1241.Vb 3
1242\& struct ev_timer mytimer;
1243\& ev_timer_init (&mytimer, one_minute_cb, 60., 0.);
1244\& ev_timer_start (loop, &mytimer);
1245.Ve
1246.PP
1247Example: Create a timeout timer that times out after 10 seconds of
1248inactivity.
1249.PP
1250.Vb 5
1251\& static void
1252\& timeout_cb (struct ev_loop *loop, struct ev_timer *w, int revents)
1253\& {
1254\& .. ten seconds without any activity
1255\& }
1256.Ve
1257.PP
1258.Vb 4
1259\& struct ev_timer mytimer;
1260\& ev_timer_init (&mytimer, timeout_cb, 0., 10.); /* note, only repeat used */
1261\& ev_timer_again (&mytimer); /* start timer */
1262\& ev_loop (loop, 0);
1263.Ve
1264.PP
1265.Vb 3
1266\& // and in some piece of code that gets executed on any "activity":
1267\& // reset the timeout to start ticking again at 10 seconds
1268\& ev_timer_again (&mytimer);
1269.Ve
749.ie n .Sh """ev_periodic"" \- to cron or not to cron" 1270.ie n .Sh """ev_periodic"" \- to cron or not to cron?"
750.el .Sh "\f(CWev_periodic\fP \- to cron or not to cron" 1271.el .Sh "\f(CWev_periodic\fP \- to cron or not to cron?"
751.IX Subsection "ev_periodic - to cron or not to cron" 1272.IX Subsection "ev_periodic - to cron or not to cron?"
752Periodic watchers are also timers of a kind, but they are very versatile 1273Periodic watchers are also timers of a kind, but they are very versatile
753(and unfortunately a bit complex). 1274(and unfortunately a bit complex).
754.PP 1275.PP
755Unlike \f(CW\*(C`ev_timer\*(C'\fR's, they are not based on real time (or relative time) 1276Unlike \f(CW\*(C`ev_timer\*(C'\fR's, they are not based on real time (or relative time)
756but on wallclock time (absolute time). You can tell a periodic watcher 1277but on wallclock time (absolute time). You can tell a periodic watcher
757to trigger \*(L"at\*(R" some specific point in time. For example, if you tell a 1278to trigger \*(L"at\*(R" some specific point in time. For example, if you tell a
758periodic watcher to trigger in 10 seconds (by specifiying e.g. c<ev_now () 1279periodic watcher to trigger in 10 seconds (by specifiying e.g. \f(CW\*(C`ev_now ()
759+ 10.>) and then reset your system clock to the last year, then it will 1280+ 10.\*(C'\fR) and then reset your system clock to the last year, then it will
760take a year to trigger the event (unlike an \f(CW\*(C`ev_timer\*(C'\fR, which would trigger 1281take a year to trigger the event (unlike an \f(CW\*(C`ev_timer\*(C'\fR, which would trigger
761roughly 10 seconds later and of course not if you reset your system time 1282roughly 10 seconds later).
762again).
763.PP 1283.PP
764They can also be used to implement vastly more complex timers, such as 1284They can also be used to implement vastly more complex timers, such as
765triggering an event on eahc midnight, local time. 1285triggering an event on each midnight, local time or other, complicated,
1286rules.
766.PP 1287.PP
767As with timers, the callback is guarenteed to be invoked only when the 1288As with timers, the callback is guarenteed to be invoked only when the
768time (\f(CW\*(C`at\*(C'\fR) has been passed, but if multiple periodic timers become ready 1289time (\f(CW\*(C`at\*(C'\fR) has been passed, but if multiple periodic timers become ready
769during the same loop iteration then order of execution is undefined. 1290during the same loop iteration then order of execution is undefined.
1291.PP
1292\fIWatcher-Specific Functions and Data Members\fR
1293.IX Subsection "Watcher-Specific Functions and Data Members"
770.IP "ev_periodic_init (ev_periodic *, callback, ev_tstamp at, ev_tstamp interval, reschedule_cb)" 4 1294.IP "ev_periodic_init (ev_periodic *, callback, ev_tstamp at, ev_tstamp interval, reschedule_cb)" 4
771.IX Item "ev_periodic_init (ev_periodic *, callback, ev_tstamp at, ev_tstamp interval, reschedule_cb)" 1295.IX Item "ev_periodic_init (ev_periodic *, callback, ev_tstamp at, ev_tstamp interval, reschedule_cb)"
772.PD 0 1296.PD 0
773.IP "ev_periodic_set (ev_periodic *, ev_tstamp after, ev_tstamp repeat, reschedule_cb)" 4 1297.IP "ev_periodic_set (ev_periodic *, ev_tstamp after, ev_tstamp repeat, reschedule_cb)" 4
774.IX Item "ev_periodic_set (ev_periodic *, ev_tstamp after, ev_tstamp repeat, reschedule_cb)" 1298.IX Item "ev_periodic_set (ev_periodic *, ev_tstamp after, ev_tstamp repeat, reschedule_cb)"
775.PD 1299.PD
776Lots of arguments, lets sort it out... There are basically three modes of 1300Lots of arguments, lets sort it out... There are basically three modes of
777operation, and we will explain them from simplest to complex: 1301operation, and we will explain them from simplest to complex:
778.RS 4 1302.RS 4
779.IP "* absolute timer (interval = reschedule_cb = 0)" 4 1303.IP "* absolute timer (at = time, interval = reschedule_cb = 0)" 4
780.IX Item "absolute timer (interval = reschedule_cb = 0)" 1304.IX Item "absolute timer (at = time, interval = reschedule_cb = 0)"
781In this configuration the watcher triggers an event at the wallclock time 1305In this configuration the watcher triggers an event at the wallclock time
782\&\f(CW\*(C`at\*(C'\fR and doesn't repeat. It will not adjust when a time jump occurs, 1306\&\f(CW\*(C`at\*(C'\fR and doesn't repeat. It will not adjust when a time jump occurs,
783that is, if it is to be run at January 1st 2011 then it will run when the 1307that is, if it is to be run at January 1st 2011 then it will run when the
784system time reaches or surpasses this time. 1308system time reaches or surpasses this time.
785.IP "* non-repeating interval timer (interval > 0, reschedule_cb = 0)" 4 1309.IP "* non-repeating interval timer (at = offset, interval > 0, reschedule_cb = 0)" 4
786.IX Item "non-repeating interval timer (interval > 0, reschedule_cb = 0)" 1310.IX Item "non-repeating interval timer (at = offset, interval > 0, reschedule_cb = 0)"
787In this mode the watcher will always be scheduled to time out at the next 1311In this mode the watcher will always be scheduled to time out at the next
788\&\f(CW\*(C`at + N * interval\*(C'\fR time (for some integer N) and then repeat, regardless 1312\&\f(CW\*(C`at + N * interval\*(C'\fR time (for some integer N, which can also be negative)
789of any time jumps. 1313and then repeat, regardless of any time jumps.
790.Sp 1314.Sp
791This can be used to create timers that do not drift with respect to system 1315This can be used to create timers that do not drift with respect to system
792time: 1316time:
793.Sp 1317.Sp
794.Vb 1 1318.Vb 1
… …
801by 3600. 1325by 3600.
802.Sp 1326.Sp
803Another way to think about it (for the mathematically inclined) is that 1327Another way to think about it (for the mathematically inclined) is that
804\&\f(CW\*(C`ev_periodic\*(C'\fR will try to run the callback in this mode at the next possible 1328\&\f(CW\*(C`ev_periodic\*(C'\fR will try to run the callback in this mode at the next possible
805time where \f(CW\*(C`time = at (mod interval)\*(C'\fR, regardless of any time jumps. 1329time where \f(CW\*(C`time = at (mod interval)\*(C'\fR, regardless of any time jumps.
1330.Sp
1331For numerical stability it is preferable that the \f(CW\*(C`at\*(C'\fR value is near
1332\&\f(CW\*(C`ev_now ()\*(C'\fR (the current time), but there is no range requirement for
1333this value.
806.IP "* manual reschedule mode (reschedule_cb = callback)" 4 1334.IP "* manual reschedule mode (at and interval ignored, reschedule_cb = callback)" 4
807.IX Item "manual reschedule mode (reschedule_cb = callback)" 1335.IX Item "manual reschedule mode (at and interval ignored, reschedule_cb = callback)"
808In this mode the values for \f(CW\*(C`interval\*(C'\fR and \f(CW\*(C`at\*(C'\fR are both being 1336In this mode the values for \f(CW\*(C`interval\*(C'\fR and \f(CW\*(C`at\*(C'\fR are both being
809ignored. Instead, each time the periodic watcher gets scheduled, the 1337ignored. Instead, each time the periodic watcher gets scheduled, the
810reschedule callback will be called with the watcher as first, and the 1338reschedule callback will be called with the watcher as first, and the
811current time as second argument. 1339current time as second argument.
812.Sp 1340.Sp
813\&\s-1NOTE:\s0 \fIThis callback \s-1MUST\s0 \s-1NOT\s0 stop or destroy any periodic watcher, 1341\&\s-1NOTE:\s0 \fIThis callback \s-1MUST\s0 \s-1NOT\s0 stop or destroy any periodic watcher,
814ever, or make any event loop modifications\fR. If you need to stop it, 1342ever, or make any event loop modifications\fR. If you need to stop it,
815return \f(CW\*(C`now + 1e30\*(C'\fR (or so, fudge fudge) and stop it afterwards (e.g. by 1343return \f(CW\*(C`now + 1e30\*(C'\fR (or so, fudge fudge) and stop it afterwards (e.g. by
816starting a prepare watcher). 1344starting an \f(CW\*(C`ev_prepare\*(C'\fR watcher, which is legal).
817.Sp 1345.Sp
818Its prototype is \f(CW\*(C`ev_tstamp (*reschedule_cb)(struct ev_periodic *w, 1346Its prototype is \f(CW\*(C`ev_tstamp (*reschedule_cb)(struct ev_periodic *w,
819ev_tstamp now)\*(C'\fR, e.g.: 1347ev_tstamp now)\*(C'\fR, e.g.:
820.Sp 1348.Sp
821.Vb 4 1349.Vb 4
… …
845.IX Item "ev_periodic_again (loop, ev_periodic *)" 1373.IX Item "ev_periodic_again (loop, ev_periodic *)"
846Simply stops and restarts the periodic watcher again. This is only useful 1374Simply stops and restarts the periodic watcher again. This is only useful
847when you changed some parameters or the reschedule callback would return 1375when you changed some parameters or the reschedule callback would return
848a different time than the last time it was called (e.g. in a crond like 1376a different time than the last time it was called (e.g. in a crond like
849program when the crontabs have changed). 1377program when the crontabs have changed).
1378.IP "ev_tstamp offset [read\-write]" 4
1379.IX Item "ev_tstamp offset [read-write]"
1380When repeating, this contains the offset value, otherwise this is the
1381absolute point in time (the \f(CW\*(C`at\*(C'\fR value passed to \f(CW\*(C`ev_periodic_set\*(C'\fR).
1382.Sp
1383Can be modified any time, but changes only take effect when the periodic
1384timer fires or \f(CW\*(C`ev_periodic_again\*(C'\fR is being called.
1385.IP "ev_tstamp interval [read\-write]" 4
1386.IX Item "ev_tstamp interval [read-write]"
1387The current interval value. Can be modified any time, but changes only
1388take effect when the periodic timer fires or \f(CW\*(C`ev_periodic_again\*(C'\fR is being
1389called.
1390.IP "ev_tstamp (*reschedule_cb)(struct ev_periodic *w, ev_tstamp now) [read\-write]" 4
1391.IX Item "ev_tstamp (*reschedule_cb)(struct ev_periodic *w, ev_tstamp now) [read-write]"
1392The current reschedule callback, or \f(CW0\fR, if this functionality is
1393switched off. Can be changed any time, but changes only take effect when
1394the periodic timer fires or \f(CW\*(C`ev_periodic_again\*(C'\fR is being called.
1395.IP "ev_tstamp at [read\-only]" 4
1396.IX Item "ev_tstamp at [read-only]"
1397When active, contains the absolute time that the watcher is supposed to
1398trigger next.
1399.PP
1400Example: Call a callback every hour, or, more precisely, whenever the
1401system clock is divisible by 3600. The callback invocation times have
1402potentially a lot of jittering, but good long-term stability.
1403.PP
1404.Vb 5
1405\& static void
1406\& clock_cb (struct ev_loop *loop, struct ev_io *w, int revents)
1407\& {
1408\& ... its now a full hour (UTC, or TAI or whatever your clock follows)
1409\& }
1410.Ve
1411.PP
1412.Vb 3
1413\& struct ev_periodic hourly_tick;
1414\& ev_periodic_init (&hourly_tick, clock_cb, 0., 3600., 0);
1415\& ev_periodic_start (loop, &hourly_tick);
1416.Ve
1417.PP
1418Example: The same as above, but use a reschedule callback to do it:
1419.PP
1420.Vb 1
1421\& #include <math.h>
1422.Ve
1423.PP
1424.Vb 5
1425\& static ev_tstamp
1426\& my_scheduler_cb (struct ev_periodic *w, ev_tstamp now)
1427\& {
1428\& return fmod (now, 3600.) + 3600.;
1429\& }
1430.Ve
1431.PP
1432.Vb 1
1433\& ev_periodic_init (&hourly_tick, clock_cb, 0., 0., my_scheduler_cb);
1434.Ve
1435.PP
1436Example: Call a callback every hour, starting now:
1437.PP
1438.Vb 4
1439\& struct ev_periodic hourly_tick;
1440\& ev_periodic_init (&hourly_tick, clock_cb,
1441\& fmod (ev_now (loop), 3600.), 3600., 0);
1442\& ev_periodic_start (loop, &hourly_tick);
1443.Ve
850.ie n .Sh """ev_signal"" \- signal me when a signal gets signalled" 1444.ie n .Sh """ev_signal"" \- signal me when a signal gets signalled!"
851.el .Sh "\f(CWev_signal\fP \- signal me when a signal gets signalled" 1445.el .Sh "\f(CWev_signal\fP \- signal me when a signal gets signalled!"
852.IX Subsection "ev_signal - signal me when a signal gets signalled" 1446.IX Subsection "ev_signal - signal me when a signal gets signalled!"
853Signal watchers will trigger an event when the process receives a specific 1447Signal watchers will trigger an event when the process receives a specific
854signal one or more times. Even though signals are very asynchronous, libev 1448signal one or more times. Even though signals are very asynchronous, libev
855will try it's best to deliver signals synchronously, i.e. as part of the 1449will try it's best to deliver signals synchronously, i.e. as part of the
856normal event processing, like any other event. 1450normal event processing, like any other event.
857.PP 1451.PP
… …
859first watcher gets started will libev actually register a signal watcher 1453first watcher gets started will libev actually register a signal watcher
860with the kernel (thus it coexists with your own signal handlers as long 1454with the kernel (thus it coexists with your own signal handlers as long
861as you don't register any with libev). Similarly, when the last signal 1455as you don't register any with libev). Similarly, when the last signal
862watcher for a signal is stopped libev will reset the signal handler to 1456watcher for a signal is stopped libev will reset the signal handler to
863\&\s-1SIG_DFL\s0 (regardless of what it was set to before). 1457\&\s-1SIG_DFL\s0 (regardless of what it was set to before).
1458.PP
1459\fIWatcher-Specific Functions and Data Members\fR
1460.IX Subsection "Watcher-Specific Functions and Data Members"
864.IP "ev_signal_init (ev_signal *, callback, int signum)" 4 1461.IP "ev_signal_init (ev_signal *, callback, int signum)" 4
865.IX Item "ev_signal_init (ev_signal *, callback, int signum)" 1462.IX Item "ev_signal_init (ev_signal *, callback, int signum)"
866.PD 0 1463.PD 0
867.IP "ev_signal_set (ev_signal *, int signum)" 4 1464.IP "ev_signal_set (ev_signal *, int signum)" 4
868.IX Item "ev_signal_set (ev_signal *, int signum)" 1465.IX Item "ev_signal_set (ev_signal *, int signum)"
869.PD 1466.PD
870Configures the watcher to trigger on the given signal number (usually one 1467Configures the watcher to trigger on the given signal number (usually one
871of the \f(CW\*(C`SIGxxx\*(C'\fR constants). 1468of the \f(CW\*(C`SIGxxx\*(C'\fR constants).
1469.IP "int signum [read\-only]" 4
1470.IX Item "int signum [read-only]"
1471The signal the watcher watches out for.
872.ie n .Sh """ev_child"" \- wait for pid status changes" 1472.ie n .Sh """ev_child"" \- watch out for process status changes"
873.el .Sh "\f(CWev_child\fP \- wait for pid status changes" 1473.el .Sh "\f(CWev_child\fP \- watch out for process status changes"
874.IX Subsection "ev_child - wait for pid status changes" 1474.IX Subsection "ev_child - watch out for process status changes"
875Child watchers trigger when your process receives a \s-1SIGCHLD\s0 in response to 1475Child watchers trigger when your process receives a \s-1SIGCHLD\s0 in response to
876some child status changes (most typically when a child of yours dies). 1476some child status changes (most typically when a child of yours dies).
1477.PP
1478\fIWatcher-Specific Functions and Data Members\fR
1479.IX Subsection "Watcher-Specific Functions and Data Members"
877.IP "ev_child_init (ev_child *, callback, int pid)" 4 1480.IP "ev_child_init (ev_child *, callback, int pid)" 4
878.IX Item "ev_child_init (ev_child *, callback, int pid)" 1481.IX Item "ev_child_init (ev_child *, callback, int pid)"
879.PD 0 1482.PD 0
880.IP "ev_child_set (ev_child *, int pid)" 4 1483.IP "ev_child_set (ev_child *, int pid)" 4
881.IX Item "ev_child_set (ev_child *, int pid)" 1484.IX Item "ev_child_set (ev_child *, int pid)"
… …
884\&\fIany\fR process if \f(CW\*(C`pid\*(C'\fR is specified as \f(CW0\fR). The callback can look 1487\&\fIany\fR process if \f(CW\*(C`pid\*(C'\fR is specified as \f(CW0\fR). The callback can look
885at the \f(CW\*(C`rstatus\*(C'\fR member of the \f(CW\*(C`ev_child\*(C'\fR watcher structure to see 1488at the \f(CW\*(C`rstatus\*(C'\fR member of the \f(CW\*(C`ev_child\*(C'\fR watcher structure to see
886the status word (use the macros from \f(CW\*(C`sys/wait.h\*(C'\fR and see your systems 1489the status word (use the macros from \f(CW\*(C`sys/wait.h\*(C'\fR and see your systems
887\&\f(CW\*(C`waitpid\*(C'\fR documentation). The \f(CW\*(C`rpid\*(C'\fR member contains the pid of the 1490\&\f(CW\*(C`waitpid\*(C'\fR documentation). The \f(CW\*(C`rpid\*(C'\fR member contains the pid of the
888process causing the status change. 1491process causing the status change.
1492.IP "int pid [read\-only]" 4
1493.IX Item "int pid [read-only]"
1494The process id this watcher watches out for, or \f(CW0\fR, meaning any process id.
1495.IP "int rpid [read\-write]" 4
1496.IX Item "int rpid [read-write]"
1497The process id that detected a status change.
1498.IP "int rstatus [read\-write]" 4
1499.IX Item "int rstatus [read-write]"
1500The process exit/trace status caused by \f(CW\*(C`rpid\*(C'\fR (see your systems
1501\&\f(CW\*(C`waitpid\*(C'\fR and \f(CW\*(C`sys/wait.h\*(C'\fR documentation for details).
1502.PP
1503Example: Try to exit cleanly on \s-1SIGINT\s0 and \s-1SIGTERM\s0.
1504.PP
1505.Vb 5
1506\& static void
1507\& sigint_cb (struct ev_loop *loop, struct ev_signal *w, int revents)
1508\& {
1509\& ev_unloop (loop, EVUNLOOP_ALL);
1510\& }
1511.Ve
1512.PP
1513.Vb 3
1514\& struct ev_signal signal_watcher;
1515\& ev_signal_init (&signal_watcher, sigint_cb, SIGINT);
1516\& ev_signal_start (loop, &sigint_cb);
1517.Ve
1518.ie n .Sh """ev_stat"" \- did the file attributes just change?"
1519.el .Sh "\f(CWev_stat\fP \- did the file attributes just change?"
1520.IX Subsection "ev_stat - did the file attributes just change?"
1521This watches a filesystem path for attribute changes. That is, it calls
1522\&\f(CW\*(C`stat\*(C'\fR regularly (or when the \s-1OS\s0 says it changed) and sees if it changed
1523compared to the last time, invoking the callback if it did.
1524.PP
1525The path does not need to exist: changing from \*(L"path exists\*(R" to \*(L"path does
1526not exist\*(R" is a status change like any other. The condition \*(L"path does
1527not exist\*(R" is signified by the \f(CW\*(C`st_nlink\*(C'\fR field being zero (which is
1528otherwise always forced to be at least one) and all the other fields of
1529the stat buffer having unspecified contents.
1530.PP
1531The path \fIshould\fR be absolute and \fImust not\fR end in a slash. If it is
1532relative and your working directory changes, the behaviour is undefined.
1533.PP
1534Since there is no standard to do this, the portable implementation simply
1535calls \f(CW\*(C`stat (2)\*(C'\fR regularly on the path to see if it changed somehow. You
1536can specify a recommended polling interval for this case. If you specify
1537a polling interval of \f(CW0\fR (highly recommended!) then a \fIsuitable,
1538unspecified default\fR value will be used (which you can expect to be around
1539five seconds, although this might change dynamically). Libev will also
1540impose a minimum interval which is currently around \f(CW0.1\fR, but thats
1541usually overkill.
1542.PP
1543This watcher type is not meant for massive numbers of stat watchers,
1544as even with OS-supported change notifications, this can be
1545resource\-intensive.
1546.PP
1547At the time of this writing, only the Linux inotify interface is
1548implemented (implementing kqueue support is left as an exercise for the
1549reader). Inotify will be used to give hints only and should not change the
1550semantics of \f(CW\*(C`ev_stat\*(C'\fR watchers, which means that libev sometimes needs
1551to fall back to regular polling again even with inotify, but changes are
1552usually detected immediately, and if the file exists there will be no
1553polling.
1554.PP
1555\fIWatcher-Specific Functions and Data Members\fR
1556.IX Subsection "Watcher-Specific Functions and Data Members"
1557.IP "ev_stat_init (ev_stat *, callback, const char *path, ev_tstamp interval)" 4
1558.IX Item "ev_stat_init (ev_stat *, callback, const char *path, ev_tstamp interval)"
1559.PD 0
1560.IP "ev_stat_set (ev_stat *, const char *path, ev_tstamp interval)" 4
1561.IX Item "ev_stat_set (ev_stat *, const char *path, ev_tstamp interval)"
1562.PD
1563Configures the watcher to wait for status changes of the given
1564\&\f(CW\*(C`path\*(C'\fR. The \f(CW\*(C`interval\*(C'\fR is a hint on how quickly a change is expected to
1565be detected and should normally be specified as \f(CW0\fR to let libev choose
1566a suitable value. The memory pointed to by \f(CW\*(C`path\*(C'\fR must point to the same
1567path for as long as the watcher is active.
1568.Sp
1569The callback will be receive \f(CW\*(C`EV_STAT\*(C'\fR when a change was detected,
1570relative to the attributes at the time the watcher was started (or the
1571last change was detected).
1572.IP "ev_stat_stat (ev_stat *)" 4
1573.IX Item "ev_stat_stat (ev_stat *)"
1574Updates the stat buffer immediately with new values. If you change the
1575watched path in your callback, you could call this fucntion to avoid
1576detecting this change (while introducing a race condition). Can also be
1577useful simply to find out the new values.
1578.IP "ev_statdata attr [read\-only]" 4
1579.IX Item "ev_statdata attr [read-only]"
1580The most-recently detected attributes of the file. Although the type is of
1581\&\f(CW\*(C`ev_statdata\*(C'\fR, this is usually the (or one of the) \f(CW\*(C`struct stat\*(C'\fR types
1582suitable for your system. If the \f(CW\*(C`st_nlink\*(C'\fR member is \f(CW0\fR, then there
1583was some error while \f(CW\*(C`stat\*(C'\fRing the file.
1584.IP "ev_statdata prev [read\-only]" 4
1585.IX Item "ev_statdata prev [read-only]"
1586The previous attributes of the file. The callback gets invoked whenever
1587\&\f(CW\*(C`prev\*(C'\fR != \f(CW\*(C`attr\*(C'\fR.
1588.IP "ev_tstamp interval [read\-only]" 4
1589.IX Item "ev_tstamp interval [read-only]"
1590The specified interval.
1591.IP "const char *path [read\-only]" 4
1592.IX Item "const char *path [read-only]"
1593The filesystem path that is being watched.
1594.PP
1595Example: Watch \f(CW\*(C`/etc/passwd\*(C'\fR for attribute changes.
1596.PP
1597.Vb 15
1598\& static void
1599\& passwd_cb (struct ev_loop *loop, ev_stat *w, int revents)
1600\& {
1601\& /* /etc/passwd changed in some way */
1602\& if (w->attr.st_nlink)
1603\& {
1604\& printf ("passwd current size %ld\en", (long)w->attr.st_size);
1605\& printf ("passwd current atime %ld\en", (long)w->attr.st_mtime);
1606\& printf ("passwd current mtime %ld\en", (long)w->attr.st_mtime);
1607\& }
1608\& else
1609\& /* you shalt not abuse printf for puts */
1610\& puts ("wow, /etc/passwd is not there, expect problems. "
1611\& "if this is windows, they already arrived\en");
1612\& }
1613.Ve
1614.PP
1615.Vb 2
1616\& ...
1617\& ev_stat passwd;
1618.Ve
1619.PP
1620.Vb 2
1621\& ev_stat_init (&passwd, passwd_cb, "/etc/passwd");
1622\& ev_stat_start (loop, &passwd);
1623.Ve
889.ie n .Sh """ev_idle"" \- when you've got nothing better to do" 1624.ie n .Sh """ev_idle"" \- when you've got nothing better to do..."
890.el .Sh "\f(CWev_idle\fP \- when you've got nothing better to do" 1625.el .Sh "\f(CWev_idle\fP \- when you've got nothing better to do..."
891.IX Subsection "ev_idle - when you've got nothing better to do" 1626.IX Subsection "ev_idle - when you've got nothing better to do..."
892Idle watchers trigger events when there are no other events are pending 1627Idle watchers trigger events when no other events of the same or higher
893(prepare, check and other idle watchers do not count). That is, as long 1628priority are pending (prepare, check and other idle watchers do not
894as your process is busy handling sockets or timeouts (or even signals, 1629count).
895imagine) it will not be triggered. But when your process is idle all idle 1630.PP
896watchers are being called again and again, once per event loop iteration \- 1631That is, as long as your process is busy handling sockets or timeouts
1632(or even signals, imagine) of the same or higher priority it will not be
1633triggered. But when your process is idle (or only lower-priority watchers
1634are pending), the idle watchers are being called once per event loop
897until stopped, that is, or your process receives more events and becomes 1635iteration \- until stopped, that is, or your process receives more events
898busy. 1636and becomes busy again with higher priority stuff.
899.PP 1637.PP
900The most noteworthy effect is that as long as any idle watchers are 1638The most noteworthy effect is that as long as any idle watchers are
901active, the process will not block when waiting for new events. 1639active, the process will not block when waiting for new events.
902.PP 1640.PP
903Apart from keeping your process non-blocking (which is a useful 1641Apart from keeping your process non-blocking (which is a useful
904effect on its own sometimes), idle watchers are a good place to do 1642effect on its own sometimes), idle watchers are a good place to do
905\&\*(L"pseudo\-background processing\*(R", or delay processing stuff to after the 1643\&\*(L"pseudo\-background processing\*(R", or delay processing stuff to after the
906event loop has handled all outstanding events. 1644event loop has handled all outstanding events.
1645.PP
1646\fIWatcher-Specific Functions and Data Members\fR
1647.IX Subsection "Watcher-Specific Functions and Data Members"
907.IP "ev_idle_init (ev_signal *, callback)" 4 1648.IP "ev_idle_init (ev_signal *, callback)" 4
908.IX Item "ev_idle_init (ev_signal *, callback)" 1649.IX Item "ev_idle_init (ev_signal *, callback)"
909Initialises and configures the idle watcher \- it has no parameters of any 1650Initialises and configures the idle watcher \- it has no parameters of any
910kind. There is a \f(CW\*(C`ev_idle_set\*(C'\fR macro, but using it is utterly pointless, 1651kind. There is a \f(CW\*(C`ev_idle_set\*(C'\fR macro, but using it is utterly pointless,
911believe me. 1652believe me.
1653.PP
1654Example: Dynamically allocate an \f(CW\*(C`ev_idle\*(C'\fR watcher, start it, and in the
1655callback, free it. Also, use no error checking, as usual.
1656.PP
1657.Vb 7
1658\& static void
1659\& idle_cb (struct ev_loop *loop, struct ev_idle *w, int revents)
1660\& {
1661\& free (w);
1662\& // now do something you wanted to do when the program has
1663\& // no longer asnything immediate to do.
1664\& }
1665.Ve
1666.PP
1667.Vb 3
1668\& struct ev_idle *idle_watcher = malloc (sizeof (struct ev_idle));
1669\& ev_idle_init (idle_watcher, idle_cb);
1670\& ev_idle_start (loop, idle_cb);
1671.Ve
912.ie n .Sh """ev_prepare""\fP and \f(CW""ev_check"" \- customise your event loop" 1672.ie n .Sh """ev_prepare""\fP and \f(CW""ev_check"" \- customise your event loop!"
913.el .Sh "\f(CWev_prepare\fP and \f(CWev_check\fP \- customise your event loop" 1673.el .Sh "\f(CWev_prepare\fP and \f(CWev_check\fP \- customise your event loop!"
914.IX Subsection "ev_prepare and ev_check - customise your event loop" 1674.IX Subsection "ev_prepare and ev_check - customise your event loop!"
915Prepare and check watchers are usually (but not always) used in tandem: 1675Prepare and check watchers are usually (but not always) used in tandem:
916prepare watchers get invoked before the process blocks and check watchers 1676prepare watchers get invoked before the process blocks and check watchers
917afterwards. 1677afterwards.
918.PP 1678.PP
1679You \fImust not\fR call \f(CW\*(C`ev_loop\*(C'\fR or similar functions that enter
1680the current event loop from either \f(CW\*(C`ev_prepare\*(C'\fR or \f(CW\*(C`ev_check\*(C'\fR
1681watchers. Other loops than the current one are fine, however. The
1682rationale behind this is that you do not need to check for recursion in
1683those watchers, i.e. the sequence will always be \f(CW\*(C`ev_prepare\*(C'\fR, blocking,
1684\&\f(CW\*(C`ev_check\*(C'\fR so if you have one watcher of each kind they will always be
1685called in pairs bracketing the blocking call.
1686.PP
919Their main purpose is to integrate other event mechanisms into libev. This 1687Their main purpose is to integrate other event mechanisms into libev and
920could be used, for example, to track variable changes, implement your own 1688their use is somewhat advanced. This could be used, for example, to track
921watchers, integrate net-snmp or a coroutine library and lots more. 1689variable changes, implement your own watchers, integrate net-snmp or a
1690coroutine library and lots more. They are also occasionally useful if
1691you cache some data and want to flush it before blocking (for example,
1692in X programs you might want to do an \f(CW\*(C`XFlush ()\*(C'\fR in an \f(CW\*(C`ev_prepare\*(C'\fR
1693watcher).
922.PP 1694.PP
923This is done by examining in each prepare call which file descriptors need 1695This is done by examining in each prepare call which file descriptors need
924to be watched by the other library, registering \f(CW\*(C`ev_io\*(C'\fR watchers for 1696to be watched by the other library, registering \f(CW\*(C`ev_io\*(C'\fR watchers for
925them and starting an \f(CW\*(C`ev_timer\*(C'\fR watcher for any timeouts (many libraries 1697them and starting an \f(CW\*(C`ev_timer\*(C'\fR watcher for any timeouts (many libraries
926provide just this functionality). Then, in the check watcher you check for 1698provide just this functionality). Then, in the check watcher you check for
… …
935are ready to run (it's actually more complicated: it only runs coroutines 1707are ready to run (it's actually more complicated: it only runs coroutines
936with priority higher than or equal to the event loop and one coroutine 1708with priority higher than or equal to the event loop and one coroutine
937of lower priority, but only once, using idle watchers to keep the event 1709of lower priority, but only once, using idle watchers to keep the event
938loop from blocking if lower-priority coroutines are active, thus mapping 1710loop from blocking if lower-priority coroutines are active, thus mapping
939low-priority coroutines to idle/background tasks). 1711low-priority coroutines to idle/background tasks).
1712.PP
1713It is recommended to give \f(CW\*(C`ev_check\*(C'\fR watchers highest (\f(CW\*(C`EV_MAXPRI\*(C'\fR)
1714priority, to ensure that they are being run before any other watchers
1715after the poll. Also, \f(CW\*(C`ev_check\*(C'\fR watchers (and \f(CW\*(C`ev_prepare\*(C'\fR watchers,
1716too) should not activate (\*(L"feed\*(R") events into libev. While libev fully
1717supports this, they will be called before other \f(CW\*(C`ev_check\*(C'\fR watchers did
1718their job. As \f(CW\*(C`ev_check\*(C'\fR watchers are often used to embed other event
1719loops those other event loops might be in an unusable state until their
1720\&\f(CW\*(C`ev_check\*(C'\fR watcher ran (always remind yourself to coexist peacefully with
1721others).
1722.PP
1723\fIWatcher-Specific Functions and Data Members\fR
1724.IX Subsection "Watcher-Specific Functions and Data Members"
940.IP "ev_prepare_init (ev_prepare *, callback)" 4 1725.IP "ev_prepare_init (ev_prepare *, callback)" 4
941.IX Item "ev_prepare_init (ev_prepare *, callback)" 1726.IX Item "ev_prepare_init (ev_prepare *, callback)"
942.PD 0 1727.PD 0
943.IP "ev_check_init (ev_check *, callback)" 4 1728.IP "ev_check_init (ev_check *, callback)" 4
944.IX Item "ev_check_init (ev_check *, callback)" 1729.IX Item "ev_check_init (ev_check *, callback)"
945.PD 1730.PD
946Initialises and configures the prepare or check watcher \- they have no 1731Initialises and configures the prepare or check watcher \- they have no
947parameters of any kind. There are \f(CW\*(C`ev_prepare_set\*(C'\fR and \f(CW\*(C`ev_check_set\*(C'\fR 1732parameters of any kind. There are \f(CW\*(C`ev_prepare_set\*(C'\fR and \f(CW\*(C`ev_check_set\*(C'\fR
948macros, but using them is utterly, utterly and completely pointless. 1733macros, but using them is utterly, utterly and completely pointless.
1734.PP
1735There are a number of principal ways to embed other event loops or modules
1736into libev. Here are some ideas on how to include libadns into libev
1737(there is a Perl module named \f(CW\*(C`EV::ADNS\*(C'\fR that does this, which you could
1738use for an actually working example. Another Perl module named \f(CW\*(C`EV::Glib\*(C'\fR
1739embeds a Glib main context into libev, and finally, \f(CW\*(C`Glib::EV\*(C'\fR embeds \s-1EV\s0
1740into the Glib event loop).
1741.PP
1742Method 1: Add \s-1IO\s0 watchers and a timeout watcher in a prepare handler,
1743and in a check watcher, destroy them and call into libadns. What follows
1744is pseudo-code only of course. This requires you to either use a low
1745priority for the check watcher or use \f(CW\*(C`ev_clear_pending\*(C'\fR explicitly, as
1746the callbacks for the IO/timeout watchers might not have been called yet.
1747.PP
1748.Vb 2
1749\& static ev_io iow [nfd];
1750\& static ev_timer tw;
1751.Ve
1752.PP
1753.Vb 4
1754\& static void
1755\& io_cb (ev_loop *loop, ev_io *w, int revents)
1756\& {
1757\& }
1758.Ve
1759.PP
1760.Vb 8
1761\& // create io watchers for each fd and a timer before blocking
1762\& static void
1763\& adns_prepare_cb (ev_loop *loop, ev_prepare *w, int revents)
1764\& {
1765\& int timeout = 3600000;
1766\& struct pollfd fds [nfd];
1767\& // actual code will need to loop here and realloc etc.
1768\& adns_beforepoll (ads, fds, &nfd, &timeout, timeval_from (ev_time ()));
1769.Ve
1770.PP
1771.Vb 3
1772\& /* the callback is illegal, but won't be called as we stop during check */
1773\& ev_timer_init (&tw, 0, timeout * 1e-3);
1774\& ev_timer_start (loop, &tw);
1775.Ve
1776.PP
1777.Vb 6
1778\& // create one ev_io per pollfd
1779\& for (int i = 0; i < nfd; ++i)
1780\& {
1781\& ev_io_init (iow + i, io_cb, fds [i].fd,
1782\& ((fds [i].events & POLLIN ? EV_READ : 0)
1783\& | (fds [i].events & POLLOUT ? EV_WRITE : 0)));
1784.Ve
1785.PP
1786.Vb 4
1787\& fds [i].revents = 0;
1788\& ev_io_start (loop, iow + i);
1789\& }
1790\& }
1791.Ve
1792.PP
1793.Vb 5
1794\& // stop all watchers after blocking
1795\& static void
1796\& adns_check_cb (ev_loop *loop, ev_check *w, int revents)
1797\& {
1798\& ev_timer_stop (loop, &tw);
1799.Ve
1800.PP
1801.Vb 8
1802\& for (int i = 0; i < nfd; ++i)
1803\& {
1804\& // set the relevant poll flags
1805\& // could also call adns_processreadable etc. here
1806\& struct pollfd *fd = fds + i;
1807\& int revents = ev_clear_pending (iow + i);
1808\& if (revents & EV_READ ) fd->revents |= fd->events & POLLIN;
1809\& if (revents & EV_WRITE) fd->revents |= fd->events & POLLOUT;
1810.Ve
1811.PP
1812.Vb 3
1813\& // now stop the watcher
1814\& ev_io_stop (loop, iow + i);
1815\& }
1816.Ve
1817.PP
1818.Vb 2
1819\& adns_afterpoll (adns, fds, nfd, timeval_from (ev_now (loop));
1820\& }
1821.Ve
1822.PP
1823Method 2: This would be just like method 1, but you run \f(CW\*(C`adns_afterpoll\*(C'\fR
1824in the prepare watcher and would dispose of the check watcher.
1825.PP
1826Method 3: If the module to be embedded supports explicit event
1827notification (adns does), you can also make use of the actual watcher
1828callbacks, and only destroy/create the watchers in the prepare watcher.
1829.PP
1830.Vb 5
1831\& static void
1832\& timer_cb (EV_P_ ev_timer *w, int revents)
1833\& {
1834\& adns_state ads = (adns_state)w->data;
1835\& update_now (EV_A);
1836.Ve
1837.PP
1838.Vb 2
1839\& adns_processtimeouts (ads, &tv_now);
1840\& }
1841.Ve
1842.PP
1843.Vb 5
1844\& static void
1845\& io_cb (EV_P_ ev_io *w, int revents)
1846\& {
1847\& adns_state ads = (adns_state)w->data;
1848\& update_now (EV_A);
1849.Ve
1850.PP
1851.Vb 3
1852\& if (revents & EV_READ ) adns_processreadable (ads, w->fd, &tv_now);
1853\& if (revents & EV_WRITE) adns_processwriteable (ads, w->fd, &tv_now);
1854\& }
1855.Ve
1856.PP
1857.Vb 1
1858\& // do not ever call adns_afterpoll
1859.Ve
1860.PP
1861Method 4: Do not use a prepare or check watcher because the module you
1862want to embed is too inflexible to support it. Instead, youc na override
1863their poll function. The drawback with this solution is that the main
1864loop is now no longer controllable by \s-1EV\s0. The \f(CW\*(C`Glib::EV\*(C'\fR module does
1865this.
1866.PP
1867.Vb 4
1868\& static gint
1869\& event_poll_func (GPollFD *fds, guint nfds, gint timeout)
1870\& {
1871\& int got_events = 0;
1872.Ve
1873.PP
1874.Vb 2
1875\& for (n = 0; n < nfds; ++n)
1876\& // create/start io watcher that sets the relevant bits in fds[n] and increment got_events
1877.Ve
1878.PP
1879.Vb 2
1880\& if (timeout >= 0)
1881\& // create/start timer
1882.Ve
1883.PP
1884.Vb 2
1885\& // poll
1886\& ev_loop (EV_A_ 0);
1887.Ve
1888.PP
1889.Vb 3
1890\& // stop timer again
1891\& if (timeout >= 0)
1892\& ev_timer_stop (EV_A_ &to);
1893.Ve
1894.PP
1895.Vb 3
1896\& // stop io watchers again - their callbacks should have set
1897\& for (n = 0; n < nfds; ++n)
1898\& ev_io_stop (EV_A_ iow [n]);
1899.Ve
1900.PP
1901.Vb 2
1902\& return got_events;
1903\& }
1904.Ve
1905.ie n .Sh """ev_embed"" \- when one backend isn't enough..."
1906.el .Sh "\f(CWev_embed\fP \- when one backend isn't enough..."
1907.IX Subsection "ev_embed - when one backend isn't enough..."
1908This is a rather advanced watcher type that lets you embed one event loop
1909into another (currently only \f(CW\*(C`ev_io\*(C'\fR events are supported in the embedded
1910loop, other types of watchers might be handled in a delayed or incorrect
1911fashion and must not be used).
1912.PP
1913There are primarily two reasons you would want that: work around bugs and
1914prioritise I/O.
1915.PP
1916As an example for a bug workaround, the kqueue backend might only support
1917sockets on some platform, so it is unusable as generic backend, but you
1918still want to make use of it because you have many sockets and it scales
1919so nicely. In this case, you would create a kqueue-based loop and embed it
1920into your default loop (which might use e.g. poll). Overall operation will
1921be a bit slower because first libev has to poll and then call kevent, but
1922at least you can use both at what they are best.
1923.PP
1924As for prioritising I/O: rarely you have the case where some fds have
1925to be watched and handled very quickly (with low latency), and even
1926priorities and idle watchers might have too much overhead. In this case
1927you would put all the high priority stuff in one loop and all the rest in
1928a second one, and embed the second one in the first.
1929.PP
1930As long as the watcher is active, the callback will be invoked every time
1931there might be events pending in the embedded loop. The callback must then
1932call \f(CW\*(C`ev_embed_sweep (mainloop, watcher)\*(C'\fR to make a single sweep and invoke
1933their callbacks (you could also start an idle watcher to give the embedded
1934loop strictly lower priority for example). You can also set the callback
1935to \f(CW0\fR, in which case the embed watcher will automatically execute the
1936embedded loop sweep.
1937.PP
1938As long as the watcher is started it will automatically handle events. The
1939callback will be invoked whenever some events have been handled. You can
1940set the callback to \f(CW0\fR to avoid having to specify one if you are not
1941interested in that.
1942.PP
1943Also, there have not currently been made special provisions for forking:
1944when you fork, you not only have to call \f(CW\*(C`ev_loop_fork\*(C'\fR on both loops,
1945but you will also have to stop and restart any \f(CW\*(C`ev_embed\*(C'\fR watchers
1946yourself.
1947.PP
1948Unfortunately, not all backends are embeddable, only the ones returned by
1949\&\f(CW\*(C`ev_embeddable_backends\*(C'\fR are, which, unfortunately, does not include any
1950portable one.
1951.PP
1952So when you want to use this feature you will always have to be prepared
1953that you cannot get an embeddable loop. The recommended way to get around
1954this is to have a separate variables for your embeddable loop, try to
1955create it, and if that fails, use the normal loop for everything:
1956.PP
1957.Vb 3
1958\& struct ev_loop *loop_hi = ev_default_init (0);
1959\& struct ev_loop *loop_lo = 0;
1960\& struct ev_embed embed;
1961.Ve
1962.PP
1963.Vb 5
1964\& // see if there is a chance of getting one that works
1965\& // (remember that a flags value of 0 means autodetection)
1966\& loop_lo = ev_embeddable_backends () & ev_recommended_backends ()
1967\& ? ev_loop_new (ev_embeddable_backends () & ev_recommended_backends ())
1968\& : 0;
1969.Ve
1970.PP
1971.Vb 8
1972\& // if we got one, then embed it, otherwise default to loop_hi
1973\& if (loop_lo)
1974\& {
1975\& ev_embed_init (&embed, 0, loop_lo);
1976\& ev_embed_start (loop_hi, &embed);
1977\& }
1978\& else
1979\& loop_lo = loop_hi;
1980.Ve
1981.PP
1982\fIWatcher-Specific Functions and Data Members\fR
1983.IX Subsection "Watcher-Specific Functions and Data Members"
1984.IP "ev_embed_init (ev_embed *, callback, struct ev_loop *embedded_loop)" 4
1985.IX Item "ev_embed_init (ev_embed *, callback, struct ev_loop *embedded_loop)"
1986.PD 0
1987.IP "ev_embed_set (ev_embed *, callback, struct ev_loop *embedded_loop)" 4
1988.IX Item "ev_embed_set (ev_embed *, callback, struct ev_loop *embedded_loop)"
1989.PD
1990Configures the watcher to embed the given loop, which must be
1991embeddable. If the callback is \f(CW0\fR, then \f(CW\*(C`ev_embed_sweep\*(C'\fR will be
1992invoked automatically, otherwise it is the responsibility of the callback
1993to invoke it (it will continue to be called until the sweep has been done,
1994if you do not want thta, you need to temporarily stop the embed watcher).
1995.IP "ev_embed_sweep (loop, ev_embed *)" 4
1996.IX Item "ev_embed_sweep (loop, ev_embed *)"
1997Make a single, non-blocking sweep over the embedded loop. This works
1998similarly to \f(CW\*(C`ev_loop (embedded_loop, EVLOOP_NONBLOCK)\*(C'\fR, but in the most
1999apropriate way for embedded loops.
2000.IP "struct ev_loop *loop [read\-only]" 4
2001.IX Item "struct ev_loop *loop [read-only]"
2002The embedded event loop.
2003.ie n .Sh """ev_fork"" \- the audacity to resume the event loop after a fork"
2004.el .Sh "\f(CWev_fork\fP \- the audacity to resume the event loop after a fork"
2005.IX Subsection "ev_fork - the audacity to resume the event loop after a fork"
2006Fork watchers are called when a \f(CW\*(C`fork ()\*(C'\fR was detected (usually because
2007whoever is a good citizen cared to tell libev about it by calling
2008\&\f(CW\*(C`ev_default_fork\*(C'\fR or \f(CW\*(C`ev_loop_fork\*(C'\fR). The invocation is done before the
2009event loop blocks next and before \f(CW\*(C`ev_check\*(C'\fR watchers are being called,
2010and only in the child after the fork. If whoever good citizen calling
2011\&\f(CW\*(C`ev_default_fork\*(C'\fR cheats and calls it in the wrong process, the fork
2012handlers will be invoked, too, of course.
2013.PP
2014\fIWatcher-Specific Functions and Data Members\fR
2015.IX Subsection "Watcher-Specific Functions and Data Members"
2016.IP "ev_fork_init (ev_signal *, callback)" 4
2017.IX Item "ev_fork_init (ev_signal *, callback)"
2018Initialises and configures the fork watcher \- it has no parameters of any
2019kind. There is a \f(CW\*(C`ev_fork_set\*(C'\fR macro, but using it is utterly pointless,
2020believe me.
949.SH "OTHER FUNCTIONS" 2021.SH "OTHER FUNCTIONS"
950.IX Header "OTHER FUNCTIONS" 2022.IX Header "OTHER FUNCTIONS"
951There are some other functions of possible interest. Described. Here. Now. 2023There are some other functions of possible interest. Described. Here. Now.
952.IP "ev_once (loop, int fd, int events, ev_tstamp timeout, callback)" 4 2024.IP "ev_once (loop, int fd, int events, ev_tstamp timeout, callback)" 4
953.IX Item "ev_once (loop, int fd, int events, ev_tstamp timeout, callback)" 2025.IX Item "ev_once (loop, int fd, int events, ev_tstamp timeout, callback)"
… …
982.Ve 2054.Ve
983.Sp 2055.Sp
984.Vb 1 2056.Vb 1
985\& ev_once (STDIN_FILENO, EV_READ, 10., stdin_ready, 0); 2057\& ev_once (STDIN_FILENO, EV_READ, 10., stdin_ready, 0);
986.Ve 2058.Ve
987.IP "ev_feed_event (loop, watcher, int events)" 4 2059.IP "ev_feed_event (ev_loop *, watcher *, int revents)" 4
988.IX Item "ev_feed_event (loop, watcher, int events)" 2060.IX Item "ev_feed_event (ev_loop *, watcher *, int revents)"
989Feeds the given event set into the event loop, as if the specified event 2061Feeds the given event set into the event loop, as if the specified event
990had happened for the specified watcher (which must be a pointer to an 2062had happened for the specified watcher (which must be a pointer to an
991initialised but not necessarily started event watcher). 2063initialised but not necessarily started event watcher).
992.IP "ev_feed_fd_event (loop, int fd, int revents)" 4 2064.IP "ev_feed_fd_event (ev_loop *, int fd, int revents)" 4
993.IX Item "ev_feed_fd_event (loop, int fd, int revents)" 2065.IX Item "ev_feed_fd_event (ev_loop *, int fd, int revents)"
994Feed an event on the given fd, as if a file descriptor backend detected 2066Feed an event on the given fd, as if a file descriptor backend detected
995the given events it. 2067the given events it.
996.IP "ev_feed_signal_event (loop, int signum)" 4 2068.IP "ev_feed_signal_event (ev_loop *loop, int signum)" 4
997.IX Item "ev_feed_signal_event (loop, int signum)" 2069.IX Item "ev_feed_signal_event (ev_loop *loop, int signum)"
998Feed an event as if the given signal occured (loop must be the default loop!). 2070Feed an event as if the given signal occured (\f(CW\*(C`loop\*(C'\fR must be the default
2071loop!).
999.SH "LIBEVENT EMULATION" 2072.SH "LIBEVENT EMULATION"
1000.IX Header "LIBEVENT EMULATION" 2073.IX Header "LIBEVENT EMULATION"
1001Libev offers a compatibility emulation layer for libevent. It cannot 2074Libev offers a compatibility emulation layer for libevent. It cannot
1002emulate the internals of libevent, so here are some usage hints: 2075emulate the internals of libevent, so here are some usage hints:
1003.IP "* Use it by including <event.h>, as usual." 4 2076.IP "* Use it by including <event.h>, as usual." 4
… …
1014.IP "* The libev emulation is \fInot\fR \s-1ABI\s0 compatible to libevent, you need to use the libev header file and library." 4 2087.IP "* The libev emulation is \fInot\fR \s-1ABI\s0 compatible to libevent, you need to use the libev header file and library." 4
1015.IX Item "The libev emulation is not ABI compatible to libevent, you need to use the libev header file and library." 2088.IX Item "The libev emulation is not ABI compatible to libevent, you need to use the libev header file and library."
1016.PD 2089.PD
1017.SH "\*(C+ SUPPORT" 2090.SH "\*(C+ SUPPORT"
1018.IX Header " SUPPORT" 2091.IX Header " SUPPORT"
1019\&\s-1TBD\s0. 2092Libev comes with some simplistic wrapper classes for \*(C+ that mainly allow
2093you to use some convinience methods to start/stop watchers and also change
2094the callback model to a model using method callbacks on objects.
2095.PP
2096To use it,
2097.PP
2098.Vb 1
2099\& #include <ev++.h>
2100.Ve
2101.PP
2102This automatically includes \fIev.h\fR and puts all of its definitions (many
2103of them macros) into the global namespace. All \*(C+ specific things are
2104put into the \f(CW\*(C`ev\*(C'\fR namespace. It should support all the same embedding
2105options as \fIev.h\fR, most notably \f(CW\*(C`EV_MULTIPLICITY\*(C'\fR.
2106.PP
2107Care has been taken to keep the overhead low. The only data member the \*(C+
2108classes add (compared to plain C\-style watchers) is the event loop pointer
2109that the watcher is associated with (or no additional members at all if
2110you disable \f(CW\*(C`EV_MULTIPLICITY\*(C'\fR when embedding libev).
2111.PP
2112Currently, functions, and static and non-static member functions can be
2113used as callbacks. Other types should be easy to add as long as they only
2114need one additional pointer for context. If you need support for other
2115types of functors please contact the author (preferably after implementing
2116it).
2117.PP
2118Here is a list of things available in the \f(CW\*(C`ev\*(C'\fR namespace:
2119.ie n .IP """ev::READ""\fR, \f(CW""ev::WRITE"" etc." 4
2120.el .IP "\f(CWev::READ\fR, \f(CWev::WRITE\fR etc." 4
2121.IX Item "ev::READ, ev::WRITE etc."
2122These are just enum values with the same values as the \f(CW\*(C`EV_READ\*(C'\fR etc.
2123macros from \fIev.h\fR.
2124.ie n .IP """ev::tstamp""\fR, \f(CW""ev::now""" 4
2125.el .IP "\f(CWev::tstamp\fR, \f(CWev::now\fR" 4
2126.IX Item "ev::tstamp, ev::now"
2127Aliases to the same types/functions as with the \f(CW\*(C`ev_\*(C'\fR prefix.
2128.ie n .IP """ev::io""\fR, \f(CW""ev::timer""\fR, \f(CW""ev::periodic""\fR, \f(CW""ev::idle""\fR, \f(CW""ev::sig"" etc." 4
2129.el .IP "\f(CWev::io\fR, \f(CWev::timer\fR, \f(CWev::periodic\fR, \f(CWev::idle\fR, \f(CWev::sig\fR etc." 4
2130.IX Item "ev::io, ev::timer, ev::periodic, ev::idle, ev::sig etc."
2131For each \f(CW\*(C`ev_TYPE\*(C'\fR watcher in \fIev.h\fR there is a corresponding class of
2132the same name in the \f(CW\*(C`ev\*(C'\fR namespace, with the exception of \f(CW\*(C`ev_signal\*(C'\fR
2133which is called \f(CW\*(C`ev::sig\*(C'\fR to avoid clashes with the \f(CW\*(C`signal\*(C'\fR macro
2134defines by many implementations.
2135.Sp
2136All of those classes have these methods:
2137.RS 4
2138.IP "ev::TYPE::TYPE ()" 4
2139.IX Item "ev::TYPE::TYPE ()"
2140.PD 0
2141.IP "ev::TYPE::TYPE (struct ev_loop *)" 4
2142.IX Item "ev::TYPE::TYPE (struct ev_loop *)"
2143.IP "ev::TYPE::~TYPE" 4
2144.IX Item "ev::TYPE::~TYPE"
2145.PD
2146The constructor (optionally) takes an event loop to associate the watcher
2147with. If it is omitted, it will use \f(CW\*(C`EV_DEFAULT\*(C'\fR.
2148.Sp
2149The constructor calls \f(CW\*(C`ev_init\*(C'\fR for you, which means you have to call the
2150\&\f(CW\*(C`set\*(C'\fR method before starting it.
2151.Sp
2152It will not set a callback, however: You have to call the templated \f(CW\*(C`set\*(C'\fR
2153method to set a callback before you can start the watcher.
2154.Sp
2155(The reason why you have to use a method is a limitation in \*(C+ which does
2156not allow explicit template arguments for constructors).
2157.Sp
2158The destructor automatically stops the watcher if it is active.
2159.IP "w\->set<class, &class::method> (object *)" 4
2160.IX Item "w->set<class, &class::method> (object *)"
2161This method sets the callback method to call. The method has to have a
2162signature of \f(CW\*(C`void (*)(ev_TYPE &, int)\*(C'\fR, it receives the watcher as
2163first argument and the \f(CW\*(C`revents\*(C'\fR as second. The object must be given as
2164parameter and is stored in the \f(CW\*(C`data\*(C'\fR member of the watcher.
2165.Sp
2166This method synthesizes efficient thunking code to call your method from
2167the C callback that libev requires. If your compiler can inline your
2168callback (i.e. it is visible to it at the place of the \f(CW\*(C`set\*(C'\fR call and
2169your compiler is good :), then the method will be fully inlined into the
2170thunking function, making it as fast as a direct C callback.
2171.Sp
2172Example: simple class declaration and watcher initialisation
2173.Sp
2174.Vb 4
2175\& struct myclass
2176\& {
2177\& void io_cb (ev::io &w, int revents) { }
2178\& }
2179.Ve
2180.Sp
2181.Vb 3
2182\& myclass obj;
2183\& ev::io iow;
2184\& iow.set <myclass, &myclass::io_cb> (&obj);
2185.Ve
2186.IP "w\->set<function> (void *data = 0)" 4
2187.IX Item "w->set<function> (void *data = 0)"
2188Also sets a callback, but uses a static method or plain function as
2189callback. The optional \f(CW\*(C`data\*(C'\fR argument will be stored in the watcher's
2190\&\f(CW\*(C`data\*(C'\fR member and is free for you to use.
2191.Sp
2192The prototype of the \f(CW\*(C`function\*(C'\fR must be \f(CW\*(C`void (*)(ev::TYPE &w, int)\*(C'\fR.
2193.Sp
2194See the method\-\f(CW\*(C`set\*(C'\fR above for more details.
2195.Sp
2196Example:
2197.Sp
2198.Vb 2
2199\& static void io_cb (ev::io &w, int revents) { }
2200\& iow.set <io_cb> ();
2201.Ve
2202.IP "w\->set (struct ev_loop *)" 4
2203.IX Item "w->set (struct ev_loop *)"
2204Associates a different \f(CW\*(C`struct ev_loop\*(C'\fR with this watcher. You can only
2205do this when the watcher is inactive (and not pending either).
2206.IP "w\->set ([args])" 4
2207.IX Item "w->set ([args])"
2208Basically the same as \f(CW\*(C`ev_TYPE_set\*(C'\fR, with the same args. Must be
2209called at least once. Unlike the C counterpart, an active watcher gets
2210automatically stopped and restarted when reconfiguring it with this
2211method.
2212.IP "w\->start ()" 4
2213.IX Item "w->start ()"
2214Starts the watcher. Note that there is no \f(CW\*(C`loop\*(C'\fR argument, as the
2215constructor already stores the event loop.
2216.IP "w\->stop ()" 4
2217.IX Item "w->stop ()"
2218Stops the watcher if it is active. Again, no \f(CW\*(C`loop\*(C'\fR argument.
2219.ie n .IP "w\->again () (""ev::timer""\fR, \f(CW""ev::periodic"" only)" 4
2220.el .IP "w\->again () (\f(CWev::timer\fR, \f(CWev::periodic\fR only)" 4
2221.IX Item "w->again () (ev::timer, ev::periodic only)"
2222For \f(CW\*(C`ev::timer\*(C'\fR and \f(CW\*(C`ev::periodic\*(C'\fR, this invokes the corresponding
2223\&\f(CW\*(C`ev_TYPE_again\*(C'\fR function.
2224.ie n .IP "w\->sweep () (""ev::embed"" only)" 4
2225.el .IP "w\->sweep () (\f(CWev::embed\fR only)" 4
2226.IX Item "w->sweep () (ev::embed only)"
2227Invokes \f(CW\*(C`ev_embed_sweep\*(C'\fR.
2228.ie n .IP "w\->update () (""ev::stat"" only)" 4
2229.el .IP "w\->update () (\f(CWev::stat\fR only)" 4
2230.IX Item "w->update () (ev::stat only)"
2231Invokes \f(CW\*(C`ev_stat_stat\*(C'\fR.
2232.RE
2233.RS 4
2234.RE
2235.PP
2236Example: Define a class with an \s-1IO\s0 and idle watcher, start one of them in
2237the constructor.
2238.PP
2239.Vb 4
2240\& class myclass
2241\& {
2242\& ev_io io; void io_cb (ev::io &w, int revents);
2243\& ev_idle idle void idle_cb (ev::idle &w, int revents);
2244.Ve
2245.PP
2246.Vb 2
2247\& myclass ();
2248\& }
2249.Ve
2250.PP
2251.Vb 4
2252\& myclass::myclass (int fd)
2253\& {
2254\& io .set <myclass, &myclass::io_cb > (this);
2255\& idle.set <myclass, &myclass::idle_cb> (this);
2256.Ve
2257.PP
2258.Vb 2
2259\& io.start (fd, ev::READ);
2260\& }
2261.Ve
2262.SH "MACRO MAGIC"
2263.IX Header "MACRO MAGIC"
2264Libev can be compiled with a variety of options, the most fundamantal
2265of which is \f(CW\*(C`EV_MULTIPLICITY\*(C'\fR. This option determines whether (most)
2266functions and callbacks have an initial \f(CW\*(C`struct ev_loop *\*(C'\fR argument.
2267.PP
2268To make it easier to write programs that cope with either variant, the
2269following macros are defined:
2270.ie n .IP """EV_A""\fR, \f(CW""EV_A_""" 4
2271.el .IP "\f(CWEV_A\fR, \f(CWEV_A_\fR" 4
2272.IX Item "EV_A, EV_A_"
2273This provides the loop \fIargument\fR for functions, if one is required (\*(L"ev
2274loop argument\*(R"). The \f(CW\*(C`EV_A\*(C'\fR form is used when this is the sole argument,
2275\&\f(CW\*(C`EV_A_\*(C'\fR is used when other arguments are following. Example:
2276.Sp
2277.Vb 3
2278\& ev_unref (EV_A);
2279\& ev_timer_add (EV_A_ watcher);
2280\& ev_loop (EV_A_ 0);
2281.Ve
2282.Sp
2283It assumes the variable \f(CW\*(C`loop\*(C'\fR of type \f(CW\*(C`struct ev_loop *\*(C'\fR is in scope,
2284which is often provided by the following macro.
2285.ie n .IP """EV_P""\fR, \f(CW""EV_P_""" 4
2286.el .IP "\f(CWEV_P\fR, \f(CWEV_P_\fR" 4
2287.IX Item "EV_P, EV_P_"
2288This provides the loop \fIparameter\fR for functions, if one is required (\*(L"ev
2289loop parameter\*(R"). The \f(CW\*(C`EV_P\*(C'\fR form is used when this is the sole parameter,
2290\&\f(CW\*(C`EV_P_\*(C'\fR is used when other parameters are following. Example:
2291.Sp
2292.Vb 2
2293\& // this is how ev_unref is being declared
2294\& static void ev_unref (EV_P);
2295.Ve
2296.Sp
2297.Vb 2
2298\& // this is how you can declare your typical callback
2299\& static void cb (EV_P_ ev_timer *w, int revents)
2300.Ve
2301.Sp
2302It declares a parameter \f(CW\*(C`loop\*(C'\fR of type \f(CW\*(C`struct ev_loop *\*(C'\fR, quite
2303suitable for use with \f(CW\*(C`EV_A\*(C'\fR.
2304.ie n .IP """EV_DEFAULT""\fR, \f(CW""EV_DEFAULT_""" 4
2305.el .IP "\f(CWEV_DEFAULT\fR, \f(CWEV_DEFAULT_\fR" 4
2306.IX Item "EV_DEFAULT, EV_DEFAULT_"
2307Similar to the other two macros, this gives you the value of the default
2308loop, if multiple loops are supported (\*(L"ev loop default\*(R").
2309.PP
2310Example: Declare and initialise a check watcher, utilising the above
2311macros so it will work regardless of whether multiple loops are supported
2312or not.
2313.PP
2314.Vb 5
2315\& static void
2316\& check_cb (EV_P_ ev_timer *w, int revents)
2317\& {
2318\& ev_check_stop (EV_A_ w);
2319\& }
2320.Ve
2321.PP
2322.Vb 4
2323\& ev_check check;
2324\& ev_check_init (&check, check_cb);
2325\& ev_check_start (EV_DEFAULT_ &check);
2326\& ev_loop (EV_DEFAULT_ 0);
2327.Ve
2328.SH "EMBEDDING"
2329.IX Header "EMBEDDING"
2330Libev can (and often is) directly embedded into host
2331applications. Examples of applications that embed it include the Deliantra
2332Game Server, the \s-1EV\s0 perl module, the \s-1GNU\s0 Virtual Private Ethernet (gvpe)
2333and rxvt\-unicode.
2334.PP
2335The goal is to enable you to just copy the neecssary files into your
2336source directory without having to change even a single line in them, so
2337you can easily upgrade by simply copying (or having a checked-out copy of
2338libev somewhere in your source tree).
2339.Sh "\s-1FILESETS\s0"
2340.IX Subsection "FILESETS"
2341Depending on what features you need you need to include one or more sets of files
2342in your app.
2343.PP
2344\fI\s-1CORE\s0 \s-1EVENT\s0 \s-1LOOP\s0\fR
2345.IX Subsection "CORE EVENT LOOP"
2346.PP
2347To include only the libev core (all the \f(CW\*(C`ev_*\*(C'\fR functions), with manual
2348configuration (no autoconf):
2349.PP
2350.Vb 2
2351\& #define EV_STANDALONE 1
2352\& #include "ev.c"
2353.Ve
2354.PP
2355This will automatically include \fIev.h\fR, too, and should be done in a
2356single C source file only to provide the function implementations. To use
2357it, do the same for \fIev.h\fR in all files wishing to use this \s-1API\s0 (best
2358done by writing a wrapper around \fIev.h\fR that you can include instead and
2359where you can put other configuration options):
2360.PP
2361.Vb 2
2362\& #define EV_STANDALONE 1
2363\& #include "ev.h"
2364.Ve
2365.PP
2366Both header files and implementation files can be compiled with a \*(C+
2367compiler (at least, thats a stated goal, and breakage will be treated
2368as a bug).
2369.PP
2370You need the following files in your source tree, or in a directory
2371in your include path (e.g. in libev/ when using \-Ilibev):
2372.PP
2373.Vb 4
2374\& ev.h
2375\& ev.c
2376\& ev_vars.h
2377\& ev_wrap.h
2378.Ve
2379.PP
2380.Vb 1
2381\& ev_win32.c required on win32 platforms only
2382.Ve
2383.PP
2384.Vb 5
2385\& ev_select.c only when select backend is enabled (which is enabled by default)
2386\& ev_poll.c only when poll backend is enabled (disabled by default)
2387\& ev_epoll.c only when the epoll backend is enabled (disabled by default)
2388\& ev_kqueue.c only when the kqueue backend is enabled (disabled by default)
2389\& ev_port.c only when the solaris port backend is enabled (disabled by default)
2390.Ve
2391.PP
2392\&\fIev.c\fR includes the backend files directly when enabled, so you only need
2393to compile this single file.
2394.PP
2395\fI\s-1LIBEVENT\s0 \s-1COMPATIBILITY\s0 \s-1API\s0\fR
2396.IX Subsection "LIBEVENT COMPATIBILITY API"
2397.PP
2398To include the libevent compatibility \s-1API\s0, also include:
2399.PP
2400.Vb 1
2401\& #include "event.c"
2402.Ve
2403.PP
2404in the file including \fIev.c\fR, and:
2405.PP
2406.Vb 1
2407\& #include "event.h"
2408.Ve
2409.PP
2410in the files that want to use the libevent \s-1API\s0. This also includes \fIev.h\fR.
2411.PP
2412You need the following additional files for this:
2413.PP
2414.Vb 2
2415\& event.h
2416\& event.c
2417.Ve
2418.PP
2419\fI\s-1AUTOCONF\s0 \s-1SUPPORT\s0\fR
2420.IX Subsection "AUTOCONF SUPPORT"
2421.PP
2422Instead of using \f(CW\*(C`EV_STANDALONE=1\*(C'\fR and providing your config in
2423whatever way you want, you can also \f(CW\*(C`m4_include([libev.m4])\*(C'\fR in your
2424\&\fIconfigure.ac\fR and leave \f(CW\*(C`EV_STANDALONE\*(C'\fR undefined. \fIev.c\fR will then
2425include \fIconfig.h\fR and configure itself accordingly.
2426.PP
2427For this of course you need the m4 file:
2428.PP
2429.Vb 1
2430\& libev.m4
2431.Ve
2432.Sh "\s-1PREPROCESSOR\s0 \s-1SYMBOLS/MACROS\s0"
2433.IX Subsection "PREPROCESSOR SYMBOLS/MACROS"
2434Libev can be configured via a variety of preprocessor symbols you have to define
2435before including any of its files. The default is not to build for multiplicity
2436and only include the select backend.
2437.IP "\s-1EV_STANDALONE\s0" 4
2438.IX Item "EV_STANDALONE"
2439Must always be \f(CW1\fR if you do not use autoconf configuration, which
2440keeps libev from including \fIconfig.h\fR, and it also defines dummy
2441implementations for some libevent functions (such as logging, which is not
2442supported). It will also not define any of the structs usually found in
2443\&\fIevent.h\fR that are not directly supported by the libev core alone.
2444.IP "\s-1EV_USE_MONOTONIC\s0" 4
2445.IX Item "EV_USE_MONOTONIC"
2446If defined to be \f(CW1\fR, libev will try to detect the availability of the
2447monotonic clock option at both compiletime and runtime. Otherwise no use
2448of the monotonic clock option will be attempted. If you enable this, you
2449usually have to link against librt or something similar. Enabling it when
2450the functionality isn't available is safe, though, althoguh you have
2451to make sure you link against any libraries where the \f(CW\*(C`clock_gettime\*(C'\fR
2452function is hiding in (often \fI\-lrt\fR).
2453.IP "\s-1EV_USE_REALTIME\s0" 4
2454.IX Item "EV_USE_REALTIME"
2455If defined to be \f(CW1\fR, libev will try to detect the availability of the
2456realtime clock option at compiletime (and assume its availability at
2457runtime if successful). Otherwise no use of the realtime clock option will
2458be attempted. This effectively replaces \f(CW\*(C`gettimeofday\*(C'\fR by \f(CW\*(C`clock_get
2459(CLOCK_REALTIME, ...)\*(C'\fR and will not normally affect correctness. See tzhe note about libraries
2460in the description of \f(CW\*(C`EV_USE_MONOTONIC\*(C'\fR, though.
2461.IP "\s-1EV_USE_SELECT\s0" 4
2462.IX Item "EV_USE_SELECT"
2463If undefined or defined to be \f(CW1\fR, libev will compile in support for the
2464\&\f(CW\*(C`select\*(C'\fR(2) backend. No attempt at autodetection will be done: if no
2465other method takes over, select will be it. Otherwise the select backend
2466will not be compiled in.
2467.IP "\s-1EV_SELECT_USE_FD_SET\s0" 4
2468.IX Item "EV_SELECT_USE_FD_SET"
2469If defined to \f(CW1\fR, then the select backend will use the system \f(CW\*(C`fd_set\*(C'\fR
2470structure. This is useful if libev doesn't compile due to a missing
2471\&\f(CW\*(C`NFDBITS\*(C'\fR or \f(CW\*(C`fd_mask\*(C'\fR definition or it misguesses the bitset layout on
2472exotic systems. This usually limits the range of file descriptors to some
2473low limit such as 1024 or might have other limitations (winsocket only
2474allows 64 sockets). The \f(CW\*(C`FD_SETSIZE\*(C'\fR macro, set before compilation, might
2475influence the size of the \f(CW\*(C`fd_set\*(C'\fR used.
2476.IP "\s-1EV_SELECT_IS_WINSOCKET\s0" 4
2477.IX Item "EV_SELECT_IS_WINSOCKET"
2478When defined to \f(CW1\fR, the select backend will assume that
2479select/socket/connect etc. don't understand file descriptors but
2480wants osf handles on win32 (this is the case when the select to
2481be used is the winsock select). This means that it will call
2482\&\f(CW\*(C`_get_osfhandle\*(C'\fR on the fd to convert it to an \s-1OS\s0 handle. Otherwise,
2483it is assumed that all these functions actually work on fds, even
2484on win32. Should not be defined on non\-win32 platforms.
2485.IP "\s-1EV_USE_POLL\s0" 4
2486.IX Item "EV_USE_POLL"
2487If defined to be \f(CW1\fR, libev will compile in support for the \f(CW\*(C`poll\*(C'\fR(2)
2488backend. Otherwise it will be enabled on non\-win32 platforms. It
2489takes precedence over select.
2490.IP "\s-1EV_USE_EPOLL\s0" 4
2491.IX Item "EV_USE_EPOLL"
2492If defined to be \f(CW1\fR, libev will compile in support for the Linux
2493\&\f(CW\*(C`epoll\*(C'\fR(7) backend. Its availability will be detected at runtime,
2494otherwise another method will be used as fallback. This is the
2495preferred backend for GNU/Linux systems.
2496.IP "\s-1EV_USE_KQUEUE\s0" 4
2497.IX Item "EV_USE_KQUEUE"
2498If defined to be \f(CW1\fR, libev will compile in support for the \s-1BSD\s0 style
2499\&\f(CW\*(C`kqueue\*(C'\fR(2) backend. Its actual availability will be detected at runtime,
2500otherwise another method will be used as fallback. This is the preferred
2501backend for \s-1BSD\s0 and BSD-like systems, although on most BSDs kqueue only
2502supports some types of fds correctly (the only platform we found that
2503supports ptys for example was NetBSD), so kqueue might be compiled in, but
2504not be used unless explicitly requested. The best way to use it is to find
2505out whether kqueue supports your type of fd properly and use an embedded
2506kqueue loop.
2507.IP "\s-1EV_USE_PORT\s0" 4
2508.IX Item "EV_USE_PORT"
2509If defined to be \f(CW1\fR, libev will compile in support for the Solaris
251010 port style backend. Its availability will be detected at runtime,
2511otherwise another method will be used as fallback. This is the preferred
2512backend for Solaris 10 systems.
2513.IP "\s-1EV_USE_DEVPOLL\s0" 4
2514.IX Item "EV_USE_DEVPOLL"
2515reserved for future expansion, works like the \s-1USE\s0 symbols above.
2516.IP "\s-1EV_USE_INOTIFY\s0" 4
2517.IX Item "EV_USE_INOTIFY"
2518If defined to be \f(CW1\fR, libev will compile in support for the Linux inotify
2519interface to speed up \f(CW\*(C`ev_stat\*(C'\fR watchers. Its actual availability will
2520be detected at runtime.
2521.IP "\s-1EV_H\s0" 4
2522.IX Item "EV_H"
2523The name of the \fIev.h\fR header file used to include it. The default if
2524undefined is \f(CW\*(C`<ev.h>\*(C'\fR in \fIevent.h\fR and \f(CW"ev.h"\fR in \fIev.c\fR. This
2525can be used to virtually rename the \fIev.h\fR header file in case of conflicts.
2526.IP "\s-1EV_CONFIG_H\s0" 4
2527.IX Item "EV_CONFIG_H"
2528If \f(CW\*(C`EV_STANDALONE\*(C'\fR isn't \f(CW1\fR, this variable can be used to override
2529\&\fIev.c\fR's idea of where to find the \fIconfig.h\fR file, similarly to
2530\&\f(CW\*(C`EV_H\*(C'\fR, above.
2531.IP "\s-1EV_EVENT_H\s0" 4
2532.IX Item "EV_EVENT_H"
2533Similarly to \f(CW\*(C`EV_H\*(C'\fR, this macro can be used to override \fIevent.c\fR's idea
2534of how the \fIevent.h\fR header can be found.
2535.IP "\s-1EV_PROTOTYPES\s0" 4
2536.IX Item "EV_PROTOTYPES"
2537If defined to be \f(CW0\fR, then \fIev.h\fR will not define any function
2538prototypes, but still define all the structs and other symbols. This is
2539occasionally useful if you want to provide your own wrapper functions
2540around libev functions.
2541.IP "\s-1EV_MULTIPLICITY\s0" 4
2542.IX Item "EV_MULTIPLICITY"
2543If undefined or defined to \f(CW1\fR, then all event-loop-specific functions
2544will have the \f(CW\*(C`struct ev_loop *\*(C'\fR as first argument, and you can create
2545additional independent event loops. Otherwise there will be no support
2546for multiple event loops and there is no first event loop pointer
2547argument. Instead, all functions act on the single default loop.
2548.IP "\s-1EV_MINPRI\s0" 4
2549.IX Item "EV_MINPRI"
2550.PD 0
2551.IP "\s-1EV_MAXPRI\s0" 4
2552.IX Item "EV_MAXPRI"
2553.PD
2554The range of allowed priorities. \f(CW\*(C`EV_MINPRI\*(C'\fR must be smaller or equal to
2555\&\f(CW\*(C`EV_MAXPRI\*(C'\fR, but otherwise there are no non-obvious limitations. You can
2556provide for more priorities by overriding those symbols (usually defined
2557to be \f(CW\*(C`\-2\*(C'\fR and \f(CW2\fR, respectively).
2558.Sp
2559When doing priority-based operations, libev usually has to linearly search
2560all the priorities, so having many of them (hundreds) uses a lot of space
2561and time, so using the defaults of five priorities (\-2 .. +2) is usually
2562fine.
2563.Sp
2564If your embedding app does not need any priorities, defining these both to
2565\&\f(CW0\fR will save some memory and cpu.
2566.IP "\s-1EV_PERIODIC_ENABLE\s0" 4
2567.IX Item "EV_PERIODIC_ENABLE"
2568If undefined or defined to be \f(CW1\fR, then periodic timers are supported. If
2569defined to be \f(CW0\fR, then they are not. Disabling them saves a few kB of
2570code.
2571.IP "\s-1EV_IDLE_ENABLE\s0" 4
2572.IX Item "EV_IDLE_ENABLE"
2573If undefined or defined to be \f(CW1\fR, then idle watchers are supported. If
2574defined to be \f(CW0\fR, then they are not. Disabling them saves a few kB of
2575code.
2576.IP "\s-1EV_EMBED_ENABLE\s0" 4
2577.IX Item "EV_EMBED_ENABLE"
2578If undefined or defined to be \f(CW1\fR, then embed watchers are supported. If
2579defined to be \f(CW0\fR, then they are not.
2580.IP "\s-1EV_STAT_ENABLE\s0" 4
2581.IX Item "EV_STAT_ENABLE"
2582If undefined or defined to be \f(CW1\fR, then stat watchers are supported. If
2583defined to be \f(CW0\fR, then they are not.
2584.IP "\s-1EV_FORK_ENABLE\s0" 4
2585.IX Item "EV_FORK_ENABLE"
2586If undefined or defined to be \f(CW1\fR, then fork watchers are supported. If
2587defined to be \f(CW0\fR, then they are not.
2588.IP "\s-1EV_MINIMAL\s0" 4
2589.IX Item "EV_MINIMAL"
2590If you need to shave off some kilobytes of code at the expense of some
2591speed, define this symbol to \f(CW1\fR. Currently only used for gcc to override
2592some inlining decisions, saves roughly 30% codesize of amd64.
2593.IP "\s-1EV_PID_HASHSIZE\s0" 4
2594.IX Item "EV_PID_HASHSIZE"
2595\&\f(CW\*(C`ev_child\*(C'\fR watchers use a small hash table to distribute workload by
2596pid. The default size is \f(CW16\fR (or \f(CW1\fR with \f(CW\*(C`EV_MINIMAL\*(C'\fR), usually more
2597than enough. If you need to manage thousands of children you might want to
2598increase this value (\fImust\fR be a power of two).
2599.IP "\s-1EV_INOTIFY_HASHSIZE\s0" 4
2600.IX Item "EV_INOTIFY_HASHSIZE"
2601\&\f(CW\*(C`ev_staz\*(C'\fR watchers use a small hash table to distribute workload by
2602inotify watch id. The default size is \f(CW16\fR (or \f(CW1\fR with \f(CW\*(C`EV_MINIMAL\*(C'\fR),
2603usually more than enough. If you need to manage thousands of \f(CW\*(C`ev_stat\*(C'\fR
2604watchers you might want to increase this value (\fImust\fR be a power of
2605two).
2606.IP "\s-1EV_COMMON\s0" 4
2607.IX Item "EV_COMMON"
2608By default, all watchers have a \f(CW\*(C`void *data\*(C'\fR member. By redefining
2609this macro to a something else you can include more and other types of
2610members. You have to define it each time you include one of the files,
2611though, and it must be identical each time.
2612.Sp
2613For example, the perl \s-1EV\s0 module uses something like this:
2614.Sp
2615.Vb 3
2616\& #define EV_COMMON \e
2617\& SV *self; /* contains this struct */ \e
2618\& SV *cb_sv, *fh /* note no trailing ";" */
2619.Ve
2620.IP "\s-1EV_CB_DECLARE\s0 (type)" 4
2621.IX Item "EV_CB_DECLARE (type)"
2622.PD 0
2623.IP "\s-1EV_CB_INVOKE\s0 (watcher, revents)" 4
2624.IX Item "EV_CB_INVOKE (watcher, revents)"
2625.IP "ev_set_cb (ev, cb)" 4
2626.IX Item "ev_set_cb (ev, cb)"
2627.PD
2628Can be used to change the callback member declaration in each watcher,
2629and the way callbacks are invoked and set. Must expand to a struct member
2630definition and a statement, respectively. See the \fIev.v\fR header file for
2631their default definitions. One possible use for overriding these is to
2632avoid the \f(CW\*(C`struct ev_loop *\*(C'\fR as first argument in all cases, or to use
2633method calls instead of plain function calls in \*(C+.
2634.Sh "\s-1EXAMPLES\s0"
2635.IX Subsection "EXAMPLES"
2636For a real-world example of a program the includes libev
2637verbatim, you can have a look at the \s-1EV\s0 perl module
2638(<http://software.schmorp.de/pkg/EV.html>). It has the libev files in
2639the \fIlibev/\fR subdirectory and includes them in the \fI\s-1EV/EVAPI\s0.h\fR (public
2640interface) and \fI\s-1EV\s0.xs\fR (implementation) files. Only the \fI\s-1EV\s0.xs\fR file
2641will be compiled. It is pretty complex because it provides its own header
2642file.
2643.Sp
2644The usage in rxvt-unicode is simpler. It has a \fIev_cpp.h\fR header file
2645that everybody includes and which overrides some configure choices:
2646.Sp
2647.Vb 9
2648\& #define EV_MINIMAL 1
2649\& #define EV_USE_POLL 0
2650\& #define EV_MULTIPLICITY 0
2651\& #define EV_PERIODIC_ENABLE 0
2652\& #define EV_STAT_ENABLE 0
2653\& #define EV_FORK_ENABLE 0
2654\& #define EV_CONFIG_H <config.h>
2655\& #define EV_MINPRI 0
2656\& #define EV_MAXPRI 0
2657.Ve
2658.Sp
2659.Vb 1
2660\& #include "ev++.h"
2661.Ve
2662.Sp
2663And a \fIev_cpp.C\fR implementation file that contains libev proper and is compiled:
2664.Sp
2665.Vb 2
2666\& #include "ev_cpp.h"
2667\& #include "ev.c"
2668.Ve
2669.SH "COMPLEXITIES"
2670.IX Header "COMPLEXITIES"
2671In this section the complexities of (many of) the algorithms used inside
2672libev will be explained. For complexity discussions about backends see the
2673documentation for \f(CW\*(C`ev_default_init\*(C'\fR.
2674.Sp
2675All of the following are about amortised time: If an array needs to be
2676extended, libev needs to realloc and move the whole array, but this
2677happens asymptotically never with higher number of elements, so O(1) might
2678mean it might do a lengthy realloc operation in rare cases, but on average
2679it is much faster and asymptotically approaches constant time.
2680.RS 4
2681.IP "Starting and stopping timer/periodic watchers: O(log skipped_other_timers)" 4
2682.IX Item "Starting and stopping timer/periodic watchers: O(log skipped_other_timers)"
2683This means that, when you have a watcher that triggers in one hour and
2684there are 100 watchers that would trigger before that then inserting will
2685have to skip those 100 watchers.
2686.IP "Changing timer/periodic watchers (by autorepeat, again): O(log skipped_other_timers)" 4
2687.IX Item "Changing timer/periodic watchers (by autorepeat, again): O(log skipped_other_timers)"
2688That means that for changing a timer costs less than removing/adding them
2689as only the relative motion in the event queue has to be paid for.
2690.IP "Starting io/check/prepare/idle/signal/child watchers: O(1)" 4
2691.IX Item "Starting io/check/prepare/idle/signal/child watchers: O(1)"
2692These just add the watcher into an array or at the head of a list.
2693=item Stopping check/prepare/idle watchers: O(1)
2694.IP "Stopping an io/signal/child watcher: O(number_of_watchers_for_this_(fd/signal/pid % \s-1EV_PID_HASHSIZE\s0))" 4
2695.IX Item "Stopping an io/signal/child watcher: O(number_of_watchers_for_this_(fd/signal/pid % EV_PID_HASHSIZE))"
2696These watchers are stored in lists then need to be walked to find the
2697correct watcher to remove. The lists are usually short (you don't usually
2698have many watchers waiting for the same fd or signal).
2699.IP "Finding the next timer per loop iteration: O(1)" 4
2700.IX Item "Finding the next timer per loop iteration: O(1)"
2701.PD 0
2702.IP "Each change on a file descriptor per loop iteration: O(number_of_watchers_for_this_fd)" 4
2703.IX Item "Each change on a file descriptor per loop iteration: O(number_of_watchers_for_this_fd)"
2704.PD
2705A change means an I/O watcher gets started or stopped, which requires
2706libev to recalculate its status (and possibly tell the kernel).
2707.IP "Activating one watcher: O(1)" 4
2708.IX Item "Activating one watcher: O(1)"
2709.PD 0
2710.IP "Priority handling: O(number_of_priorities)" 4
2711.IX Item "Priority handling: O(number_of_priorities)"
2712.PD
2713Priorities are implemented by allocating some space for each
2714priority. When doing priority-based operations, libev usually has to
2715linearly search all the priorities.
2716.RE
2717.RS 4
1020.SH "AUTHOR" 2718.SH "AUTHOR"
1021.IX Header "AUTHOR" 2719.IX Header "AUTHOR"
1022Marc Lehmann <libev@schmorp.de>. 2720Marc Lehmann <libev@schmorp.de>.

Diff Legend

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