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

Comparing libev/ev.3 (file contents):
Revision 1.1 by root, Tue Nov 13 03:11:57 2007 UTC vs.
Revision 1.34 by root, Thu Nov 29 12:21:21 2007 UTC

… …
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 ""<STANDARD INPUT>" 1"
132.TH "<STANDARD INPUT>" 1 "2007-11-13" "perl v5.8.8" "User Contributed Perl Documentation" 132.TH "<STANDARD INPUT>" 1 "2007-11-29" "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"
142Libev is an event loop: you register interest in certain events (such as a 201Libev 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 202file descriptor being readable or a timeout occuring), and it will manage
144these event sources and provide your program with events. 203these event sources and provide your program with events.
… …
151watchers\fR, which are relatively small C structures you initialise with the 210watchers\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 211details of the event, and then hand it over to libev by \fIstarting\fR the
153watcher. 212watcher.
154.SH "FEATURES" 213.SH "FEATURES"
155.IX Header "FEATURES" 214.IX Header "FEATURES"
156Libev supports select, poll, the linux-specific epoll and the bsd-specific 215Libev 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 216BSD-specific \f(CW\*(C`kqueue\*(C'\fR and the Solaris-specific event port mechanisms
158timers with customised rescheduling, signal events, process status change 217for 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 218(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 219with customised rescheduling (\f(CW\*(C`ev_periodic\*(C'\fR), synchronous signals
161fast (see this benchmark comparing 220(\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). 221watchers dealing with the event loop mechanism itself (\f(CW\*(C`ev_idle\*(C'\fR,
222\&\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
223file watchers (\f(CW\*(C`ev_stat\*(C'\fR) and even limited support for fork events
224(\f(CW\*(C`ev_fork\*(C'\fR).
225.PP
226It also is quite fast (see this
227benchmark comparing it to libevent
228for example).
163.SH "CONVENTIONS" 229.SH "CONVENTIONS"
164.IX Header "CONVENTIONS" 230.IX Header "CONVENTIONS"
165Libev is very configurable. In this manual the default configuration 231Libev is very configurable. In this manual the default configuration will
166will be described, which supports multiple event loops. For more info 232be described, which supports multiple event loops. For more info about
167about various configuration options please have a look at the file 233various 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 234this manual. If libev was configured without support for multiple event
169support for multiple event loops, then all functions taking an initial 235loops, 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) 236(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" 237.SH "TIME REPRESENTATION"
173.IX Header "TIME REPRESENTATION" 238.IX Header "TIME REPRESENTATION"
174Libev represents time as a single floating point number, representing the 239Libev represents time as a single floating point number, representing the
175(fractional) number of seconds since the (\s-1POSIX\s0) epoch (somewhere near 240(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 241the 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 242called \f(CW\*(C`ev_tstamp\*(C'\fR, which is what you should use too. It usually aliases
178to the double type in C. 243to the \f(CW\*(C`double\*(C'\fR type in C, and when you need to do any calculations on
244it, you should treat it as such.
179.SH "GLOBAL FUNCTIONS" 245.SH "GLOBAL FUNCTIONS"
180.IX Header "GLOBAL FUNCTIONS" 246.IX Header "GLOBAL FUNCTIONS"
181These functions can be called anytime, even before initialising the 247These functions can be called anytime, even before initialising the
182library in any way. 248library in any way.
183.IP "ev_tstamp ev_time ()" 4 249.IP "ev_tstamp ev_time ()" 4
184.IX Item "ev_tstamp ev_time ()" 250.IX Item "ev_tstamp ev_time ()"
185Returns the current time as libev would use it. 251Returns the current time as libev would use it. Please note that the
252\&\f(CW\*(C`ev_now\*(C'\fR function is usually faster and also often returns the timestamp
253you actually want to know.
186.IP "int ev_version_major ()" 4 254.IP "int ev_version_major ()" 4
187.IX Item "int ev_version_major ()" 255.IX Item "int ev_version_major ()"
188.PD 0 256.PD 0
189.IP "int ev_version_minor ()" 4 257.IP "int ev_version_minor ()" 4
190.IX Item "int ev_version_minor ()" 258.IX Item "int ev_version_minor ()"
… …
197.Sp 265.Sp
198Usually, it's a good idea to terminate if the major versions mismatch, 266Usually, it's a good idea to terminate if the major versions mismatch,
199as this indicates an incompatible change. Minor versions are usually 267as this indicates an incompatible change. Minor versions are usually
200compatible to older versions, so a larger minor version alone is usually 268compatible to older versions, so a larger minor version alone is usually
201not a problem. 269not a problem.
270.Sp
271Example: Make sure we haven't accidentally been linked against the wrong
272version.
273.Sp
274.Vb 3
275\& assert (("libev version mismatch",
276\& ev_version_major () == EV_VERSION_MAJOR
277\& && ev_version_minor () >= EV_VERSION_MINOR));
278.Ve
279.IP "unsigned int ev_supported_backends ()" 4
280.IX Item "unsigned int ev_supported_backends ()"
281Return the set of all backends (i.e. their corresponding \f(CW\*(C`EV_BACKEND_*\*(C'\fR
282value) compiled into this binary of libev (independent of their
283availability on the system you are running on). See \f(CW\*(C`ev_default_loop\*(C'\fR for
284a description of the set values.
285.Sp
286Example: make sure we have the epoll method, because yeah this is cool and
287a must have and can we have a torrent of it please!!!11
288.Sp
289.Vb 2
290\& assert (("sorry, no epoll, no sex",
291\& ev_supported_backends () & EVBACKEND_EPOLL));
292.Ve
293.IP "unsigned int ev_recommended_backends ()" 4
294.IX Item "unsigned int ev_recommended_backends ()"
295Return the set of all backends compiled into this binary of libev and also
296recommended for this platform. This set is often smaller than the one
297returned by \f(CW\*(C`ev_supported_backends\*(C'\fR, as for example kqueue is broken on
298most BSDs and will not be autodetected unless you explicitly request it
299(assuming you know what you are doing). This is the set of backends that
300libev will probe for if you specify no backends explicitly.
301.IP "unsigned int ev_embeddable_backends ()" 4
302.IX Item "unsigned int ev_embeddable_backends ()"
303Returns the set of backends that are embeddable in other event loops. This
304is the theoretical, all\-platform, value. To find which backends
305might be supported on the current system, you would need to look at
306\&\f(CW\*(C`ev_embeddable_backends () & ev_supported_backends ()\*(C'\fR, likewise for
307recommended ones.
308.Sp
309See the description of \f(CW\*(C`ev_embed\*(C'\fR watchers for more info.
202.IP "ev_set_allocator (void *(*cb)(void *ptr, long size))" 4 310.IP "ev_set_allocator (void *(*cb)(void *ptr, long size))" 4
203.IX Item "ev_set_allocator (void *(*cb)(void *ptr, long size))" 311.IX Item "ev_set_allocator (void *(*cb)(void *ptr, long size))"
204Sets the allocation function to use (the prototype is similar to the 312Sets the allocation function to use (the prototype is similar \- the
205realloc C function, the semantics are identical). It is used to allocate 313semantics is identical \- to the realloc C function). It is used to
206and free memory (no surprises here). If it returns zero when memory 314allocate and free memory (no surprises here). If it returns zero when
207needs to be allocated, the library might abort or take some potentially 315memory needs to be allocated, the library might abort or take some
208destructive action. The default is your system realloc function. 316potentially destructive action. The default is your system realloc
317function.
209.Sp 318.Sp
210You could override this function in high-availability programs to, say, 319You could override this function in high-availability programs to, say,
211free some memory if it cannot allocate memory, to use a special allocator, 320free some memory if it cannot allocate memory, to use a special allocator,
212or even to sleep a while and retry until some memory is available. 321or even to sleep a while and retry until some memory is available.
322.Sp
323Example: Replace the libev allocator with one that waits a bit and then
324retries).
325.Sp
326.Vb 6
327\& static void *
328\& persistent_realloc (void *ptr, size_t size)
329\& {
330\& for (;;)
331\& {
332\& void *newptr = realloc (ptr, size);
333.Ve
334.Sp
335.Vb 2
336\& if (newptr)
337\& return newptr;
338.Ve
339.Sp
340.Vb 3
341\& sleep (60);
342\& }
343\& }
344.Ve
345.Sp
346.Vb 2
347\& ...
348\& ev_set_allocator (persistent_realloc);
349.Ve
213.IP "ev_set_syserr_cb (void (*cb)(const char *msg));" 4 350.IP "ev_set_syserr_cb (void (*cb)(const char *msg));" 4
214.IX Item "ev_set_syserr_cb (void (*cb)(const char *msg));" 351.IX Item "ev_set_syserr_cb (void (*cb)(const char *msg));"
215Set the callback function to call on a retryable syscall error (such 352Set the callback function to call on a retryable syscall error (such
216as failed select, poll, epoll_wait). The message is a printable string 353as failed select, poll, epoll_wait). The message is a printable string
217indicating the system call or subsystem causing the problem. If this 354indicating the system call or subsystem causing the problem. If this
218callback is set, then libev will expect it to remedy the sitution, no 355callback is set, then libev will expect it to remedy the sitution, no
219matter what, when it returns. That is, libev will generally retry the 356matter what, when it returns. That is, libev will generally retry the
220requested operation, or, if the condition doesn't go away, do bad stuff 357requested operation, or, if the condition doesn't go away, do bad stuff
221(such as abort). 358(such as abort).
359.Sp
360Example: This is basically the same thing that libev does internally, too.
361.Sp
362.Vb 6
363\& static void
364\& fatal_error (const char *msg)
365\& {
366\& perror (msg);
367\& abort ();
368\& }
369.Ve
370.Sp
371.Vb 2
372\& ...
373\& ev_set_syserr_cb (fatal_error);
374.Ve
222.SH "FUNCTIONS CONTROLLING THE EVENT LOOP" 375.SH "FUNCTIONS CONTROLLING THE EVENT LOOP"
223.IX Header "FUNCTIONS CONTROLLING THE EVENT LOOP" 376.IX Header "FUNCTIONS CONTROLLING THE EVENT LOOP"
224An event loop is described by a \f(CW\*(C`struct ev_loop *\*(C'\fR. The library knows two 377An event loop is described by a \f(CW\*(C`struct ev_loop *\*(C'\fR. The library knows two
225types of such loops, the \fIdefault\fR loop, which supports signals and child 378types of such loops, the \fIdefault\fR loop, which supports signals and child
226events, and dynamically created loops which do not. 379events, and dynamically created loops which do not.
… …
234.IP "struct ev_loop *ev_default_loop (unsigned int flags)" 4 387.IP "struct ev_loop *ev_default_loop (unsigned int flags)" 4
235.IX Item "struct ev_loop *ev_default_loop (unsigned int flags)" 388.IX Item "struct ev_loop *ev_default_loop (unsigned int flags)"
236This will initialise the default event loop if it hasn't been initialised 389This will initialise the default event loop if it hasn't been initialised
237yet and return it. If the default loop could not be initialised, returns 390yet and return it. If the default loop could not be initialised, returns
238false. If it already was initialised it simply returns it (and ignores the 391false. If it already was initialised it simply returns it (and ignores the
239flags). 392flags. If that is troubling you, check \f(CW\*(C`ev_backend ()\*(C'\fR afterwards).
240.Sp 393.Sp
241If you don't know what event loop to use, use the one returned from this 394If you don't know what event loop to use, use the one returned from this
242function. 395function.
243.Sp 396.Sp
244The flags argument can be used to specify special behaviour or specific 397The flags argument can be used to specify special behaviour or specific
245backends to use, and is usually specified as 0 (or \s-1EVFLAG_AUTO\s0). 398backends to use, and is usually specified as \f(CW0\fR (or \f(CW\*(C`EVFLAG_AUTO\*(C'\fR).
246.Sp 399.Sp
247It supports the following flags: 400The following flags are supported:
248.RS 4 401.RS 4
249.ie n .IP """EVFLAG_AUTO""" 4 402.ie n .IP """EVFLAG_AUTO""" 4
250.el .IP "\f(CWEVFLAG_AUTO\fR" 4 403.el .IP "\f(CWEVFLAG_AUTO\fR" 4
251.IX Item "EVFLAG_AUTO" 404.IX Item "EVFLAG_AUTO"
252The default flags value. Use this if you have no clue (it's the right 405The default flags value. Use this if you have no clue (it's the right
… …
258or setgid) then libev will \fInot\fR look at the environment variable 411or setgid) then libev will \fInot\fR look at the environment variable
259\&\f(CW\*(C`LIBEV_FLAGS\*(C'\fR. Otherwise (the default), this environment variable will 412\&\f(CW\*(C`LIBEV_FLAGS\*(C'\fR. Otherwise (the default), this environment variable will
260override the flags completely if it is found in the environment. This is 413override the flags completely if it is found in the environment. This is
261useful to try out specific backends to test their performance, or to work 414useful to try out specific backends to test their performance, or to work
262around bugs. 415around bugs.
263.ie n .IP """EVMETHOD_SELECT"" (portable select backend)" 4 416.ie n .IP """EVBACKEND_SELECT"" (value 1, portable select backend)" 4
264.el .IP "\f(CWEVMETHOD_SELECT\fR (portable select backend)" 4 417.el .IP "\f(CWEVBACKEND_SELECT\fR (value 1, portable select backend)" 4
265.IX Item "EVMETHOD_SELECT (portable select backend)" 418.IX Item "EVBACKEND_SELECT (value 1, portable select backend)"
266.PD 0 419This is your standard \fIselect\fR\|(2) backend. Not \fIcompletely\fR standard, as
420libev tries to roll its own fd_set with no limits on the number of fds,
421but if that fails, expect a fairly low limit on the number of fds when
422using this backend. It doesn't scale too well (O(highest_fd)), but its usually
423the fastest backend for a low number of fds.
267.ie n .IP """EVMETHOD_POLL"" (poll backend, available everywhere except on windows)" 4 424.ie n .IP """EVBACKEND_POLL"" (value 2, poll backend, available everywhere except on windows)" 4
268.el .IP "\f(CWEVMETHOD_POLL\fR (poll backend, available everywhere except on windows)" 4 425.el .IP "\f(CWEVBACKEND_POLL\fR (value 2, poll backend, available everywhere except on windows)" 4
269.IX Item "EVMETHOD_POLL (poll backend, available everywhere except on windows)" 426.IX Item "EVBACKEND_POLL (value 2, poll backend, available everywhere except on windows)"
427And this is your standard \fIpoll\fR\|(2) backend. It's more complicated than
428select, but handles sparse fds better and has no artificial limit on the
429number of fds you can use (except it will slow down considerably with a
430lot of inactive fds). It scales similarly to select, i.e. O(total_fds).
270.ie n .IP """EVMETHOD_EPOLL"" (linux only)" 4 431.ie n .IP """EVBACKEND_EPOLL"" (value 4, Linux)" 4
271.el .IP "\f(CWEVMETHOD_EPOLL\fR (linux only)" 4 432.el .IP "\f(CWEVBACKEND_EPOLL\fR (value 4, Linux)" 4
272.IX Item "EVMETHOD_EPOLL (linux only)" 433.IX Item "EVBACKEND_EPOLL (value 4, Linux)"
273.ie n .IP """EVMETHOD_KQUEUE"" (some bsds only)" 4 434For few fds, this backend is a bit little slower than poll and select,
274.el .IP "\f(CWEVMETHOD_KQUEUE\fR (some bsds only)" 4 435but it scales phenomenally better. While poll and select usually scale like
275.IX Item "EVMETHOD_KQUEUE (some bsds only)" 436O(total_fds) where n is the total number of fds (or the highest fd), epoll scales
437either O(1) or O(active_fds).
438.Sp
439While stopping and starting an I/O watcher in the same iteration will
440result in some caching, there is still a syscall per such incident
441(because the fd could point to a different file description now), so its
442best to avoid that. Also, \fIdup()\fRed file descriptors might not work very
443well if you register events for both fds.
444.Sp
445Please note that epoll sometimes generates spurious notifications, so you
446need to use non-blocking I/O or other means to avoid blocking when no data
447(or space) is available.
448.ie n .IP """EVBACKEND_KQUEUE"" (value 8, most \s-1BSD\s0 clones)" 4
449.el .IP "\f(CWEVBACKEND_KQUEUE\fR (value 8, most \s-1BSD\s0 clones)" 4
450.IX Item "EVBACKEND_KQUEUE (value 8, most BSD clones)"
451Kqueue deserves special mention, as at the time of this writing, it
452was broken on all BSDs except NetBSD (usually it doesn't work with
453anything but sockets and pipes, except on Darwin, where of course its
454completely useless). For this reason its not being \*(L"autodetected\*(R"
455unless you explicitly specify it explicitly in the flags (i.e. using
456\&\f(CW\*(C`EVBACKEND_KQUEUE\*(C'\fR).
457.Sp
458It scales in the same way as the epoll backend, but the interface to the
459kernel is more efficient (which says nothing about its actual speed, of
460course). While starting and stopping an I/O watcher does not cause an
461extra syscall as with epoll, it still adds up to four event changes per
462incident, so its best to avoid that.
276.ie n .IP """EVMETHOD_DEVPOLL"" (solaris 8 only)" 4 463.ie n .IP """EVBACKEND_DEVPOLL"" (value 16, Solaris 8)" 4
277.el .IP "\f(CWEVMETHOD_DEVPOLL\fR (solaris 8 only)" 4 464.el .IP "\f(CWEVBACKEND_DEVPOLL\fR (value 16, Solaris 8)" 4
278.IX Item "EVMETHOD_DEVPOLL (solaris 8 only)" 465.IX Item "EVBACKEND_DEVPOLL (value 16, Solaris 8)"
466This is not implemented yet (and might never be).
279.ie n .IP """EVMETHOD_PORT"" (solaris 10 only)" 4 467.ie n .IP """EVBACKEND_PORT"" (value 32, Solaris 10)" 4
280.el .IP "\f(CWEVMETHOD_PORT\fR (solaris 10 only)" 4 468.el .IP "\f(CWEVBACKEND_PORT\fR (value 32, Solaris 10)" 4
281.IX Item "EVMETHOD_PORT (solaris 10 only)" 469.IX Item "EVBACKEND_PORT (value 32, Solaris 10)"
282.PD 470This uses the Solaris 10 port mechanism. As with everything on Solaris,
283If one or more of these are ored into the flags value, then only these 471it's really slow, but it still scales very well (O(active_fds)).
284backends will be tried (in the reverse order as given here). If one are 472.Sp
285specified, any backend will do. 473Please note that solaris ports can result in a lot of spurious
474notifications, so you need to use non-blocking I/O or other means to avoid
475blocking when no data (or space) is available.
476.ie n .IP """EVBACKEND_ALL""" 4
477.el .IP "\f(CWEVBACKEND_ALL\fR" 4
478.IX Item "EVBACKEND_ALL"
479Try all backends (even potentially broken ones that wouldn't be tried
480with \f(CW\*(C`EVFLAG_AUTO\*(C'\fR). Since this is a mask, you can do stuff such as
481\&\f(CW\*(C`EVBACKEND_ALL & ~EVBACKEND_KQUEUE\*(C'\fR.
286.RE 482.RE
287.RS 4 483.RS 4
484.Sp
485If one or more of these are ored into the flags value, then only these
486backends will be tried (in the reverse order as given here). If none are
487specified, most compiled-in backend will be tried, usually in reverse
488order of their flag values :)
489.Sp
490The most typical usage is like this:
491.Sp
492.Vb 2
493\& if (!ev_default_loop (0))
494\& fatal ("could not initialise libev, bad $LIBEV_FLAGS in environment?");
495.Ve
496.Sp
497Restrict libev to the select and poll backends, and do not allow
498environment settings to be taken into account:
499.Sp
500.Vb 1
501\& ev_default_loop (EVBACKEND_POLL | EVBACKEND_SELECT | EVFLAG_NOENV);
502.Ve
503.Sp
504Use whatever libev has to offer, but make sure that kqueue is used if
505available (warning, breaks stuff, best use only with your own private
506event loop and only if you know the \s-1OS\s0 supports your types of fds):
507.Sp
508.Vb 1
509\& ev_default_loop (ev_recommended_backends () | EVBACKEND_KQUEUE);
510.Ve
288.RE 511.RE
289.IP "struct ev_loop *ev_loop_new (unsigned int flags)" 4 512.IP "struct ev_loop *ev_loop_new (unsigned int flags)" 4
290.IX Item "struct ev_loop *ev_loop_new (unsigned int flags)" 513.IX Item "struct ev_loop *ev_loop_new (unsigned int flags)"
291Similar to \f(CW\*(C`ev_default_loop\*(C'\fR, but always creates a new event loop that is 514Similar to \f(CW\*(C`ev_default_loop\*(C'\fR, but always creates a new event loop that is
292always distinct from the default loop. Unlike the default loop, it cannot 515always distinct from the default loop. Unlike the default loop, it cannot
293handle signal and child watchers, and attempts to do so will be greeted by 516handle signal and child watchers, and attempts to do so will be greeted by
294undefined behaviour (or a failed assertion if assertions are enabled). 517undefined behaviour (or a failed assertion if assertions are enabled).
518.Sp
519Example: Try to create a event loop that uses epoll and nothing else.
520.Sp
521.Vb 3
522\& struct ev_loop *epoller = ev_loop_new (EVBACKEND_EPOLL | EVFLAG_NOENV);
523\& if (!epoller)
524\& fatal ("no epoll found here, maybe it hides under your chair");
525.Ve
295.IP "ev_default_destroy ()" 4 526.IP "ev_default_destroy ()" 4
296.IX Item "ev_default_destroy ()" 527.IX Item "ev_default_destroy ()"
297Destroys the default loop again (frees all memory and kernel state 528Destroys the default loop again (frees all memory and kernel state
298etc.). This stops all registered event watchers (by not touching them in 529etc.). None of the active event watchers will be stopped in the normal
299any way whatsoever, although you cannot rely on this :). 530sense, so e.g. \f(CW\*(C`ev_is_active\*(C'\fR might still return true. It is your
531responsibility to either stop all watchers cleanly yoursef \fIbefore\fR
532calling this function, or cope with the fact afterwards (which is usually
533the easiest thing, youc na just ignore the watchers and/or \f(CW\*(C`free ()\*(C'\fR them
534for example).
300.IP "ev_loop_destroy (loop)" 4 535.IP "ev_loop_destroy (loop)" 4
301.IX Item "ev_loop_destroy (loop)" 536.IX Item "ev_loop_destroy (loop)"
302Like \f(CW\*(C`ev_default_destroy\*(C'\fR, but destroys an event loop created by an 537Like \f(CW\*(C`ev_default_destroy\*(C'\fR, but destroys an event loop created by an
303earlier call to \f(CW\*(C`ev_loop_new\*(C'\fR. 538earlier call to \f(CW\*(C`ev_loop_new\*(C'\fR.
304.IP "ev_default_fork ()" 4 539.IP "ev_default_fork ()" 4
… …
306This function reinitialises the kernel state for backends that have 541This function reinitialises the kernel state for backends that have
307one. Despite the name, you can call it anytime, but it makes most sense 542one. Despite the name, you can call it anytime, but it makes most sense
308after forking, in either the parent or child process (or both, but that 543after forking, in either the parent or child process (or both, but that
309again makes little sense). 544again makes little sense).
310.Sp 545.Sp
311You \fImust\fR call this function after forking if and only if you want to 546You \fImust\fR call this function in the child process after forking if and
312use the event library in both processes. If you just fork+exec, you don't 547only if you want to use the event library in both processes. If you just
313have to call it. 548fork+exec, you don't have to call it.
314.Sp 549.Sp
315The function itself is quite fast and it's usually not a problem to call 550The function itself is quite fast and it's usually not a problem to call
316it just in case after a fork. To make this easy, the function will fit in 551it just in case after a fork. To make this easy, the function will fit in
317quite nicely into a call to \f(CW\*(C`pthread_atfork\*(C'\fR: 552quite nicely into a call to \f(CW\*(C`pthread_atfork\*(C'\fR:
318.Sp 553.Sp
319.Vb 1 554.Vb 1
320\& pthread_atfork (0, 0, ev_default_fork); 555\& pthread_atfork (0, 0, ev_default_fork);
321.Ve 556.Ve
557.Sp
558At the moment, \f(CW\*(C`EVBACKEND_SELECT\*(C'\fR and \f(CW\*(C`EVBACKEND_POLL\*(C'\fR are safe to use
559without calling this function, so if you force one of those backends you
560do not need to care.
322.IP "ev_loop_fork (loop)" 4 561.IP "ev_loop_fork (loop)" 4
323.IX Item "ev_loop_fork (loop)" 562.IX Item "ev_loop_fork (loop)"
324Like \f(CW\*(C`ev_default_fork\*(C'\fR, but acts on an event loop created by 563Like \f(CW\*(C`ev_default_fork\*(C'\fR, but acts on an event loop created by
325\&\f(CW\*(C`ev_loop_new\*(C'\fR. Yes, you have to call this on every allocated event loop 564\&\f(CW\*(C`ev_loop_new\*(C'\fR. Yes, you have to call this on every allocated event loop
326after fork, and how you do this is entirely your own problem. 565after fork, and how you do this is entirely your own problem.
327.IP "unsigned int ev_method (loop)" 4 566.IP "unsigned int ev_backend (loop)" 4
328.IX Item "unsigned int ev_method (loop)" 567.IX Item "unsigned int ev_backend (loop)"
329Returns one of the \f(CW\*(C`EVMETHOD_*\*(C'\fR flags indicating the event backend in 568Returns one of the \f(CW\*(C`EVBACKEND_*\*(C'\fR flags indicating the event backend in
330use. 569use.
331.IP "ev_tstamp ev_now (loop)" 4 570.IP "ev_tstamp ev_now (loop)" 4
332.IX Item "ev_tstamp ev_now (loop)" 571.IX Item "ev_tstamp ev_now (loop)"
333Returns the current \*(L"event loop time\*(R", which is the time the event loop 572Returns the current \*(L"event loop time\*(R", which is the time the event loop
334got events and started processing them. This timestamp does not change 573received events and started processing them. This timestamp does not
335as long as callbacks are being processed, and this is also the base time 574change as long as callbacks are being processed, and this is also the base
336used for relative timers. You can treat it as the timestamp of the event 575time used for relative timers. You can treat it as the timestamp of the
337occuring (or more correctly, the mainloop finding out about it). 576event occuring (or more correctly, libev finding out about it).
338.IP "ev_loop (loop, int flags)" 4 577.IP "ev_loop (loop, int flags)" 4
339.IX Item "ev_loop (loop, int flags)" 578.IX Item "ev_loop (loop, int flags)"
340Finally, this is it, the event handler. This function usually is called 579Finally, this is it, the event handler. This function usually is called
341after you initialised all your watchers and you want to start handling 580after you initialised all your watchers and you want to start handling
342events. 581events.
343.Sp 582.Sp
344If the flags argument is specified as 0, it will not return until either 583If the flags argument is specified as \f(CW0\fR, it will not return until
345no event watchers are active anymore or \f(CW\*(C`ev_unloop\*(C'\fR was called. 584either no event watchers are active anymore or \f(CW\*(C`ev_unloop\*(C'\fR was called.
585.Sp
586Please note that an explicit \f(CW\*(C`ev_unloop\*(C'\fR is usually better than
587relying on all watchers to be stopped when deciding when a program has
588finished (especially in interactive programs), but having a program that
589automatically loops as long as it has to and no longer by virtue of
590relying on its watchers stopping correctly is a thing of beauty.
346.Sp 591.Sp
347A flags value of \f(CW\*(C`EVLOOP_NONBLOCK\*(C'\fR will look for new events, will handle 592A flags value of \f(CW\*(C`EVLOOP_NONBLOCK\*(C'\fR will look for new events, will handle
348those events and any outstanding ones, but will not block your process in 593those events and any outstanding ones, but will not block your process in
349case there are no events and will return after one iteration of the loop. 594case there are no events and will return after one iteration of the loop.
350.Sp 595.Sp
351A flags value of \f(CW\*(C`EVLOOP_ONESHOT\*(C'\fR will look for new events (waiting if 596A flags value of \f(CW\*(C`EVLOOP_ONESHOT\*(C'\fR will look for new events (waiting if
352neccessary) and will handle those and any outstanding ones. It will block 597neccessary) and will handle those and any outstanding ones. It will block
353your process until at least one new event arrives, and will return after 598your process until at least one new event arrives, and will return after
354one iteration of the loop. 599one iteration of the loop. This is useful if you are waiting for some
600external event in conjunction with something not expressible using other
601libev watchers. However, a pair of \f(CW\*(C`ev_prepare\*(C'\fR/\f(CW\*(C`ev_check\*(C'\fR watchers is
602usually a better approach for this kind of thing.
355.Sp 603.Sp
356This flags value could be used to implement alternative looping 604Here are the gory details of what \f(CW\*(C`ev_loop\*(C'\fR does:
357constructs, but the \f(CW\*(C`prepare\*(C'\fR and \f(CW\*(C`check\*(C'\fR watchers provide a better and 605.Sp
358more generic mechanism. 606.Vb 18
607\& * If there are no active watchers (reference count is zero), return.
608\& - Queue prepare watchers and then call all outstanding watchers.
609\& - If we have been forked, recreate the kernel state.
610\& - Update the kernel state with all outstanding changes.
611\& - Update the "event loop time".
612\& - Calculate for how long to block.
613\& - Block the process, waiting for any events.
614\& - Queue all outstanding I/O (fd) events.
615\& - Update the "event loop time" and do time jump handling.
616\& - Queue all outstanding timers.
617\& - Queue all outstanding periodics.
618\& - If no events are pending now, queue all idle watchers.
619\& - Queue all check watchers.
620\& - Call all queued watchers in reverse order (i.e. check watchers first).
621\& Signals and child watchers are implemented as I/O watchers, and will
622\& be handled here by queueing them when their watcher gets executed.
623\& - If ev_unloop has been called or EVLOOP_ONESHOT or EVLOOP_NONBLOCK
624\& were used, return, otherwise continue with step *.
625.Ve
626.Sp
627Example: Queue some jobs and then loop until no events are outsanding
628anymore.
629.Sp
630.Vb 4
631\& ... queue jobs here, make sure they register event watchers as long
632\& ... as they still have work to do (even an idle watcher will do..)
633\& ev_loop (my_loop, 0);
634\& ... jobs done. yeah!
635.Ve
359.IP "ev_unloop (loop, how)" 4 636.IP "ev_unloop (loop, how)" 4
360.IX Item "ev_unloop (loop, how)" 637.IX Item "ev_unloop (loop, how)"
361Can be used to make a call to \f(CW\*(C`ev_loop\*(C'\fR return early (but only after it 638Can be used to make a call to \f(CW\*(C`ev_loop\*(C'\fR return early (but only after it
362has processed all outstanding events). The \f(CW\*(C`how\*(C'\fR argument must be either 639has processed all outstanding events). The \f(CW\*(C`how\*(C'\fR argument must be either
363\&\f(CW\*(C`EVUNLOOP_ONE\*(C'\fR, which will make the innermost \f(CW\*(C`ev_loop\*(C'\fR call return, or 640\&\f(CW\*(C`EVUNLOOP_ONE\*(C'\fR, which will make the innermost \f(CW\*(C`ev_loop\*(C'\fR call return, or
… …
376example, libev itself uses this for its internal signal pipe: It is not 653example, libev itself uses this for its internal signal pipe: It is not
377visible to the libev user and should not keep \f(CW\*(C`ev_loop\*(C'\fR from exiting if 654visible to the libev user and should not keep \f(CW\*(C`ev_loop\*(C'\fR from exiting if
378no event watchers registered by it are active. It is also an excellent 655no event watchers registered by it are active. It is also an excellent
379way to do this for generic recurring timers or from within third-party 656way to do this for generic recurring timers or from within third-party
380libraries. Just remember to \fIunref after start\fR and \fIref before stop\fR. 657libraries. Just remember to \fIunref after start\fR and \fIref before stop\fR.
658.Sp
659Example: Create a signal watcher, but keep it from keeping \f(CW\*(C`ev_loop\*(C'\fR
660running when nothing else is active.
661.Sp
662.Vb 4
663\& struct ev_signal exitsig;
664\& ev_signal_init (&exitsig, sig_cb, SIGINT);
665\& ev_signal_start (loop, &exitsig);
666\& evf_unref (loop);
667.Ve
668.Sp
669Example: For some weird reason, unregister the above signal handler again.
670.Sp
671.Vb 2
672\& ev_ref (loop);
673\& ev_signal_stop (loop, &exitsig);
674.Ve
381.SH "ANATOMY OF A WATCHER" 675.SH "ANATOMY OF A WATCHER"
382.IX Header "ANATOMY OF A WATCHER" 676.IX Header "ANATOMY OF A WATCHER"
383A watcher is a structure that you create and register to record your 677A watcher is a structure that you create and register to record your
384interest in some event. For instance, if you want to wait for \s-1STDIN\s0 to 678interest in some event. For instance, if you want to wait for \s-1STDIN\s0 to
385become readable, you would create an \f(CW\*(C`ev_io\*(C'\fR watcher for that: 679become readable, you would create an \f(CW\*(C`ev_io\*(C'\fR watcher for that:
… …
421*)\*(C'\fR), and you can stop watching for events at any time by calling the 715*)\*(C'\fR), and you can stop watching for events at any time by calling the
422corresponding stop function (\f(CW\*(C`ev_<type>_stop (loop, watcher *)\*(C'\fR. 716corresponding stop function (\f(CW\*(C`ev_<type>_stop (loop, watcher *)\*(C'\fR.
423.PP 717.PP
424As long as your watcher is active (has been started but not stopped) you 718As long as your watcher is active (has been started but not stopped) you
425must not touch the values stored in it. Most specifically you must never 719must not touch the values stored in it. Most specifically you must never
426reinitialise it or call its set method. 720reinitialise it or call its \f(CW\*(C`set\*(C'\fR macro.
427.PP
428You can check whether an event is active by calling the \f(CW\*(C`ev_is_active
429(watcher *)\*(C'\fR macro. To see whether an event is outstanding (but the
430callback for it has not been called yet) you can use the \f(CW\*(C`ev_is_pending
431(watcher *)\*(C'\fR macro.
432.PP 721.PP
433Each and every callback receives the event loop pointer as first, the 722Each and every callback receives the event loop pointer as first, the
434registered watcher structure as second, and a bitset of received events as 723registered watcher structure as second, and a bitset of received events as
435third argument. 724third argument.
436.PP 725.PP
… …
461The signal specified in the \f(CW\*(C`ev_signal\*(C'\fR watcher has been received by a thread. 750The signal specified in the \f(CW\*(C`ev_signal\*(C'\fR watcher has been received by a thread.
462.ie n .IP """EV_CHILD""" 4 751.ie n .IP """EV_CHILD""" 4
463.el .IP "\f(CWEV_CHILD\fR" 4 752.el .IP "\f(CWEV_CHILD\fR" 4
464.IX Item "EV_CHILD" 753.IX Item "EV_CHILD"
465The pid specified in the \f(CW\*(C`ev_child\*(C'\fR watcher has received a status change. 754The pid specified in the \f(CW\*(C`ev_child\*(C'\fR watcher has received a status change.
755.ie n .IP """EV_STAT""" 4
756.el .IP "\f(CWEV_STAT\fR" 4
757.IX Item "EV_STAT"
758The path specified in the \f(CW\*(C`ev_stat\*(C'\fR watcher changed its attributes somehow.
466.ie n .IP """EV_IDLE""" 4 759.ie n .IP """EV_IDLE""" 4
467.el .IP "\f(CWEV_IDLE\fR" 4 760.el .IP "\f(CWEV_IDLE\fR" 4
468.IX Item "EV_IDLE" 761.IX Item "EV_IDLE"
469The \f(CW\*(C`ev_idle\*(C'\fR watcher has determined that you have nothing better to do. 762The \f(CW\*(C`ev_idle\*(C'\fR watcher has determined that you have nothing better to do.
470.ie n .IP """EV_PREPARE""" 4 763.ie n .IP """EV_PREPARE""" 4
… …
480\&\f(CW\*(C`ev_loop\*(C'\fR has gathered them, but before it invokes any callbacks for any 773\&\f(CW\*(C`ev_loop\*(C'\fR has gathered them, but before it invokes any callbacks for any
481received events. Callbacks of both watcher types can start and stop as 774received events. Callbacks of both watcher types can start and stop as
482many watchers as they want, and all of them will be taken into account 775many watchers as they want, and all of them will be taken into account
483(for example, a \f(CW\*(C`ev_prepare\*(C'\fR watcher might start an idle watcher to keep 776(for example, a \f(CW\*(C`ev_prepare\*(C'\fR watcher might start an idle watcher to keep
484\&\f(CW\*(C`ev_loop\*(C'\fR from blocking). 777\&\f(CW\*(C`ev_loop\*(C'\fR from blocking).
778.ie n .IP """EV_EMBED""" 4
779.el .IP "\f(CWEV_EMBED\fR" 4
780.IX Item "EV_EMBED"
781The embedded event loop specified in the \f(CW\*(C`ev_embed\*(C'\fR watcher needs attention.
782.ie n .IP """EV_FORK""" 4
783.el .IP "\f(CWEV_FORK\fR" 4
784.IX Item "EV_FORK"
785The event loop has been resumed in the child process after fork (see
786\&\f(CW\*(C`ev_fork\*(C'\fR).
485.ie n .IP """EV_ERROR""" 4 787.ie n .IP """EV_ERROR""" 4
486.el .IP "\f(CWEV_ERROR\fR" 4 788.el .IP "\f(CWEV_ERROR\fR" 4
487.IX Item "EV_ERROR" 789.IX Item "EV_ERROR"
488An unspecified error has occured, the watcher has been stopped. This might 790An unspecified error has occured, the watcher has been stopped. This might
489happen because the watcher could not be properly started because libev 791happen because the watcher could not be properly started because libev
… …
494Libev will usually signal a few \*(L"dummy\*(R" events together with an error, 796Libev will usually signal a few \*(L"dummy\*(R" events together with an error,
495for example it might indicate that a fd is readable or writable, and if 797for example it might indicate that a fd is readable or writable, and if
496your callbacks is well-written it can just attempt the operation and cope 798your callbacks is well-written it can just attempt the operation and cope
497with the error from \fIread()\fR or \fIwrite()\fR. This will not work in multithreaded 799with the error from \fIread()\fR or \fIwrite()\fR. This will not work in multithreaded
498programs, though, so beware. 800programs, though, so beware.
801.Sh "\s-1GENERIC\s0 \s-1WATCHER\s0 \s-1FUNCTIONS\s0"
802.IX Subsection "GENERIC WATCHER FUNCTIONS"
803In the following description, \f(CW\*(C`TYPE\*(C'\fR stands for the watcher type,
804e.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.
805.ie n .IP """ev_init"" (ev_TYPE *watcher, callback)" 4
806.el .IP "\f(CWev_init\fR (ev_TYPE *watcher, callback)" 4
807.IX Item "ev_init (ev_TYPE *watcher, callback)"
808This macro initialises the generic portion of a watcher. The contents
809of the watcher object can be arbitrary (so \f(CW\*(C`malloc\*(C'\fR will do). Only
810the generic parts of the watcher are initialised, you \fIneed\fR to call
811the type-specific \f(CW\*(C`ev_TYPE_set\*(C'\fR macro afterwards to initialise the
812type-specific parts. For each type there is also a \f(CW\*(C`ev_TYPE_init\*(C'\fR macro
813which rolls both calls into one.
814.Sp
815You can reinitialise a watcher at any time as long as it has been stopped
816(or never started) and there are no pending events outstanding.
817.Sp
818The callback is always of type \f(CW\*(C`void (*)(ev_loop *loop, ev_TYPE *watcher,
819int revents)\*(C'\fR.
820.ie n .IP """ev_TYPE_set"" (ev_TYPE *, [args])" 4
821.el .IP "\f(CWev_TYPE_set\fR (ev_TYPE *, [args])" 4
822.IX Item "ev_TYPE_set (ev_TYPE *, [args])"
823This macro initialises the type-specific parts of a watcher. You need to
824call \f(CW\*(C`ev_init\*(C'\fR at least once before you call this macro, but you can
825call \f(CW\*(C`ev_TYPE_set\*(C'\fR any number of times. You must not, however, call this
826macro on a watcher that is active (it can be pending, however, which is a
827difference to the \f(CW\*(C`ev_init\*(C'\fR macro).
828.Sp
829Although some watcher types do not have type-specific arguments
830(e.g. \f(CW\*(C`ev_prepare\*(C'\fR) you still need to call its \f(CW\*(C`set\*(C'\fR macro.
831.ie n .IP """ev_TYPE_init"" (ev_TYPE *watcher, callback, [args])" 4
832.el .IP "\f(CWev_TYPE_init\fR (ev_TYPE *watcher, callback, [args])" 4
833.IX Item "ev_TYPE_init (ev_TYPE *watcher, callback, [args])"
834This convinience macro rolls both \f(CW\*(C`ev_init\*(C'\fR and \f(CW\*(C`ev_TYPE_set\*(C'\fR macro
835calls into a single call. This is the most convinient method to initialise
836a watcher. The same limitations apply, of course.
837.ie n .IP """ev_TYPE_start"" (loop *, ev_TYPE *watcher)" 4
838.el .IP "\f(CWev_TYPE_start\fR (loop *, ev_TYPE *watcher)" 4
839.IX Item "ev_TYPE_start (loop *, ev_TYPE *watcher)"
840Starts (activates) the given watcher. Only active watchers will receive
841events. If the watcher is already active nothing will happen.
842.ie n .IP """ev_TYPE_stop"" (loop *, ev_TYPE *watcher)" 4
843.el .IP "\f(CWev_TYPE_stop\fR (loop *, ev_TYPE *watcher)" 4
844.IX Item "ev_TYPE_stop (loop *, ev_TYPE *watcher)"
845Stops the given watcher again (if active) and clears the pending
846status. It is possible that stopped watchers are pending (for example,
847non-repeating timers are being stopped when they become pending), but
848\&\f(CW\*(C`ev_TYPE_stop\*(C'\fR ensures that the watcher is neither active nor pending. If
849you want to free or reuse the memory used by the watcher it is therefore a
850good idea to always call its \f(CW\*(C`ev_TYPE_stop\*(C'\fR function.
851.IP "bool ev_is_active (ev_TYPE *watcher)" 4
852.IX Item "bool ev_is_active (ev_TYPE *watcher)"
853Returns a true value iff the watcher is active (i.e. it has been started
854and not yet been stopped). As long as a watcher is active you must not modify
855it.
856.IP "bool ev_is_pending (ev_TYPE *watcher)" 4
857.IX Item "bool ev_is_pending (ev_TYPE *watcher)"
858Returns a true value iff the watcher is pending, (i.e. it has outstanding
859events but its callback has not yet been invoked). As long as a watcher
860is pending (but not active) you must not call an init function on it (but
861\&\f(CW\*(C`ev_TYPE_set\*(C'\fR is safe) and you must make sure the watcher is available to
862libev (e.g. you cnanot \f(CW\*(C`free ()\*(C'\fR it).
863.IP "callback ev_cb (ev_TYPE *watcher)" 4
864.IX Item "callback ev_cb (ev_TYPE *watcher)"
865Returns the callback currently set on the watcher.
866.IP "ev_cb_set (ev_TYPE *watcher, callback)" 4
867.IX Item "ev_cb_set (ev_TYPE *watcher, callback)"
868Change the callback. You can change the callback at virtually any time
869(modulo threads).
499.Sh "\s-1ASSOCIATING\s0 \s-1CUSTOM\s0 \s-1DATA\s0 \s-1WITH\s0 A \s-1WATCHER\s0" 870.Sh "\s-1ASSOCIATING\s0 \s-1CUSTOM\s0 \s-1DATA\s0 \s-1WITH\s0 A \s-1WATCHER\s0"
500.IX Subsection "ASSOCIATING CUSTOM DATA WITH A WATCHER" 871.IX Subsection "ASSOCIATING CUSTOM DATA WITH A WATCHER"
501Each watcher has, by default, a member \f(CW\*(C`void *data\*(C'\fR that you can change 872Each watcher has, by default, a member \f(CW\*(C`void *data\*(C'\fR that you can change
502and read at any time, libev will completely ignore it. This can be used 873and read at any time, libev will completely ignore it. This can be used
503to associate arbitrary data with your watcher. If you need more data and 874to associate arbitrary data with your watcher. If you need more data and
… …
524\& struct my_io *w = (struct my_io *)w_; 895\& struct my_io *w = (struct my_io *)w_;
525\& ... 896\& ...
526\& } 897\& }
527.Ve 898.Ve
528.PP 899.PP
529More interesting and less C\-conformant ways of catsing your callback type 900More interesting and less C\-conformant ways of casting your callback type
530have been omitted.... 901instead have been omitted.
902.PP
903Another common scenario is having some data structure with multiple
904watchers:
905.PP
906.Vb 6
907\& struct my_biggy
908\& {
909\& int some_data;
910\& ev_timer t1;
911\& ev_timer t2;
912\& }
913.Ve
914.PP
915In this case getting the pointer to \f(CW\*(C`my_biggy\*(C'\fR is a bit more complicated,
916you need to use \f(CW\*(C`offsetof\*(C'\fR:
917.PP
918.Vb 1
919\& #include <stddef.h>
920.Ve
921.PP
922.Vb 6
923\& static void
924\& t1_cb (EV_P_ struct ev_timer *w, int revents)
925\& {
926\& struct my_biggy big = (struct my_biggy *
927\& (((char *)w) - offsetof (struct my_biggy, t1));
928\& }
929.Ve
930.PP
931.Vb 6
932\& static void
933\& t2_cb (EV_P_ struct ev_timer *w, int revents)
934\& {
935\& struct my_biggy big = (struct my_biggy *
936\& (((char *)w) - offsetof (struct my_biggy, t2));
937\& }
938.Ve
531.SH "WATCHER TYPES" 939.SH "WATCHER TYPES"
532.IX Header "WATCHER TYPES" 940.IX Header "WATCHER TYPES"
533This section describes each watcher in detail, but will not repeat 941This section describes each watcher in detail, but will not repeat
534information given in the last section. 942information given in the last section. Any initialisation/set macros,
943functions and members specific to the watcher type are explained.
944.PP
945Members are additionally marked with either \fI[read\-only]\fR, meaning that,
946while the watcher is active, you can look at the member and expect some
947sensible content, but you must not modify it (you can modify it while the
948watcher is stopped to your hearts content), or \fI[read\-write]\fR, which
949means you can expect it to have some sensible content while the watcher
950is active, but you can also modify it. Modifying it may not do something
951sensible or take immediate effect (or do anything at all), but libev will
952not crash or malfunction in any way.
535.ie n .Sh """ev_io"" \- is this file descriptor readable or writable" 953.ie n .Sh """ev_io"" \- is this file descriptor readable or writable?"
536.el .Sh "\f(CWev_io\fP \- is this file descriptor readable or writable" 954.el .Sh "\f(CWev_io\fP \- is this file descriptor readable or writable?"
537.IX Subsection "ev_io - is this file descriptor readable or writable" 955.IX Subsection "ev_io - is this file descriptor readable or writable?"
538I/O watchers check whether a file descriptor is readable or writable 956I/O watchers check whether a file descriptor is readable or writable
539in each iteration of the event loop (This behaviour is called 957in each iteration of the event loop, or, more precisely, when reading
540level-triggering because you keep receiving events as long as the 958would not block the process and writing would at least be able to write
541condition persists. Remember you can stop the watcher if you don't want to 959some data. This behaviour is called level-triggering because you keep
542act on the event and neither want to receive future events). 960receiving events as long as the condition persists. Remember you can stop
961the watcher if you don't want to act on the event and neither want to
962receive future events.
543.PP 963.PP
544In general you can register as many read and/or write event watchers per 964In general you can register as many read and/or write event watchers per
545fd as you want (as long as you don't confuse yourself). Setting all file 965fd as you want (as long as you don't confuse yourself). Setting all file
546descriptors to non-blocking mode is also usually a good idea (but not 966descriptors to non-blocking mode is also usually a good idea (but not
547required if you know what you are doing). 967required if you know what you are doing).
548.PP 968.PP
549You have to be careful with dup'ed file descriptors, though. Some backends 969You have to be careful with dup'ed file descriptors, though. Some backends
550(the linux epoll backend is a notable example) cannot handle dup'ed file 970(the linux epoll backend is a notable example) cannot handle dup'ed file
551descriptors correctly if you register interest in two or more fds pointing 971descriptors correctly if you register interest in two or more fds pointing
552to the same underlying file/socket etc. description (that is, they share 972to the same underlying file/socket/etc. description (that is, they share
553the same underlying \*(L"file open\*(R"). 973the same underlying \*(L"file open\*(R").
554.PP 974.PP
555If you must do this, then force the use of a known-to-be-good backend 975If you must do this, then force the use of a known-to-be-good backend
556(at the time of this writing, this includes only \s-1EVMETHOD_SELECT\s0 and 976(at the time of this writing, this includes only \f(CW\*(C`EVBACKEND_SELECT\*(C'\fR and
557\&\s-1EVMETHOD_POLL\s0). 977\&\f(CW\*(C`EVBACKEND_POLL\*(C'\fR).
978.PP
979Another thing you have to watch out for is that it is quite easy to
980receive \*(L"spurious\*(R" readyness notifications, that is your callback might
981be called with \f(CW\*(C`EV_READ\*(C'\fR but a subsequent \f(CW\*(C`read\*(C'\fR(2) will actually block
982because there is no data. Not only are some backends known to create a
983lot of those (for example solaris ports), it is very easy to get into
984this situation even with a relatively standard program structure. Thus
985it is best to always use non-blocking I/O: An extra \f(CW\*(C`read\*(C'\fR(2) returning
986\&\f(CW\*(C`EAGAIN\*(C'\fR is far preferable to a program hanging until some data arrives.
987.PP
988If you cannot run the fd in non-blocking mode (for example you should not
989play around with an Xlib connection), then you have to seperately re-test
990wether a file descriptor is really ready with a known-to-be good interface
991such as poll (fortunately in our Xlib example, Xlib already does this on
992its own, so its quite safe to use).
558.IP "ev_io_init (ev_io *, callback, int fd, int events)" 4 993.IP "ev_io_init (ev_io *, callback, int fd, int events)" 4
559.IX Item "ev_io_init (ev_io *, callback, int fd, int events)" 994.IX Item "ev_io_init (ev_io *, callback, int fd, int events)"
560.PD 0 995.PD 0
561.IP "ev_io_set (ev_io *, int fd, int events)" 4 996.IP "ev_io_set (ev_io *, int fd, int events)" 4
562.IX Item "ev_io_set (ev_io *, int fd, int events)" 997.IX Item "ev_io_set (ev_io *, int fd, int events)"
563.PD 998.PD
564Configures an \f(CW\*(C`ev_io\*(C'\fR watcher. The fd is the file descriptor to rceeive 999Configures an \f(CW\*(C`ev_io\*(C'\fR watcher. The \f(CW\*(C`fd\*(C'\fR is the file descriptor to
565events 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 | 1000rceeive events for and events is either \f(CW\*(C`EV_READ\*(C'\fR, \f(CW\*(C`EV_WRITE\*(C'\fR or
566EV_WRITE\*(C'\fR to receive the given events. 1001\&\f(CW\*(C`EV_READ | EV_WRITE\*(C'\fR to receive the given events.
1002.IP "int fd [read\-only]" 4
1003.IX Item "int fd [read-only]"
1004The file descriptor being watched.
1005.IP "int events [read\-only]" 4
1006.IX Item "int events [read-only]"
1007The events being watched.
1008.PP
1009Example: Call \f(CW\*(C`stdin_readable_cb\*(C'\fR when \s-1STDIN_FILENO\s0 has become, well
1010readable, but only once. Since it is likely line\-buffered, you could
1011attempt to read a whole line in the callback.
1012.PP
1013.Vb 6
1014\& static void
1015\& stdin_readable_cb (struct ev_loop *loop, struct ev_io *w, int revents)
1016\& {
1017\& ev_io_stop (loop, w);
1018\& .. read from stdin here (or from w->fd) and haqndle any I/O errors
1019\& }
1020.Ve
1021.PP
1022.Vb 6
1023\& ...
1024\& struct ev_loop *loop = ev_default_init (0);
1025\& struct ev_io stdin_readable;
1026\& ev_io_init (&stdin_readable, stdin_readable_cb, STDIN_FILENO, EV_READ);
1027\& ev_io_start (loop, &stdin_readable);
1028\& ev_loop (loop, 0);
1029.Ve
567.ie n .Sh """ev_timer"" \- relative and optionally recurring timeouts" 1030.ie n .Sh """ev_timer"" \- relative and optionally repeating timeouts"
568.el .Sh "\f(CWev_timer\fP \- relative and optionally recurring timeouts" 1031.el .Sh "\f(CWev_timer\fP \- relative and optionally repeating timeouts"
569.IX Subsection "ev_timer - relative and optionally recurring timeouts" 1032.IX Subsection "ev_timer - relative and optionally repeating timeouts"
570Timer watchers are simple relative timers that generate an event after a 1033Timer watchers are simple relative timers that generate an event after a
571given time, and optionally repeating in regular intervals after that. 1034given time, and optionally repeating in regular intervals after that.
572.PP 1035.PP
573The timers are based on real time, that is, if you register an event that 1036The timers are based on real time, that is, if you register an event that
574times out after an hour and you reset your system clock to last years 1037times out after an hour and you reset your system clock to last years
575time, it will still time out after (roughly) and hour. \*(L"Roughly\*(R" because 1038time, it will still time out after (roughly) and hour. \*(L"Roughly\*(R" because
576detecting time jumps is hard, and soem inaccuracies are unavoidable (the 1039detecting time jumps is hard, and some inaccuracies are unavoidable (the
577monotonic clock option helps a lot here). 1040monotonic clock option helps a lot here).
578.PP 1041.PP
579The relative timeouts are calculated relative to the \f(CW\*(C`ev_now ()\*(C'\fR 1042The relative timeouts are calculated relative to the \f(CW\*(C`ev_now ()\*(C'\fR
580time. This is usually the right thing as this timestamp refers to the time 1043time. This is usually the right thing as this timestamp refers to the time
581of the event triggering whatever timeout you are modifying/starting. If 1044of the event triggering whatever timeout you are modifying/starting. If
582you suspect event processing to be delayed and you *need* to base the timeout 1045you suspect event processing to be delayed and you \fIneed\fR to base the timeout
583on the current time, use something like this to adjust for this: 1046on the current time, use something like this to adjust for this:
584.PP 1047.PP
585.Vb 1 1048.Vb 1
586\& ev_timer_set (&timer, after + ev_now () - ev_time (), 0.); 1049\& ev_timer_set (&timer, after + ev_now () - ev_time (), 0.);
587.Ve 1050.Ve
1051.PP
1052The callback is guarenteed to be invoked only when its timeout has passed,
1053but if multiple timers become ready during the same loop iteration then
1054order of execution is undefined.
588.IP "ev_timer_init (ev_timer *, callback, ev_tstamp after, ev_tstamp repeat)" 4 1055.IP "ev_timer_init (ev_timer *, callback, ev_tstamp after, ev_tstamp repeat)" 4
589.IX Item "ev_timer_init (ev_timer *, callback, ev_tstamp after, ev_tstamp repeat)" 1056.IX Item "ev_timer_init (ev_timer *, callback, ev_tstamp after, ev_tstamp repeat)"
590.PD 0 1057.PD 0
591.IP "ev_timer_set (ev_timer *, ev_tstamp after, ev_tstamp repeat)" 4 1058.IP "ev_timer_set (ev_timer *, ev_tstamp after, ev_tstamp repeat)" 4
592.IX Item "ev_timer_set (ev_timer *, ev_tstamp after, ev_tstamp repeat)" 1059.IX Item "ev_timer_set (ev_timer *, ev_tstamp after, ev_tstamp repeat)"
… …
604.IP "ev_timer_again (loop)" 4 1071.IP "ev_timer_again (loop)" 4
605.IX Item "ev_timer_again (loop)" 1072.IX Item "ev_timer_again (loop)"
606This will act as if the timer timed out and restart it again if it is 1073This will act as if the timer timed out and restart it again if it is
607repeating. The exact semantics are: 1074repeating. The exact semantics are:
608.Sp 1075.Sp
1076If the timer is pending, its pending status is cleared.
1077.Sp
609If the timer is started but nonrepeating, stop it. 1078If the timer is started but nonrepeating, stop it (as if it timed out).
610.Sp 1079.Sp
611If the timer is repeating, either start it if necessary (with the repeat 1080If the timer is repeating, either start it if necessary (with the
612value), or reset the running timer to the repeat value. 1081\&\f(CW\*(C`repeat\*(C'\fR value), or reset the running timer to the \f(CW\*(C`repeat\*(C'\fR value.
613.Sp 1082.Sp
614This sounds a bit complicated, but here is a useful and typical 1083This sounds a bit complicated, but here is a useful and typical
615example: Imagine you have a tcp connection and you want a so-called idle 1084example: Imagine you have a tcp connection and you want a so-called idle
616timeout, that is, you want to be called when there have been, say, 60 1085timeout, that is, you want to be called when there have been, say, 60
617seconds of inactivity on the socket. The easiest way to do this is to 1086seconds of inactivity on the socket. The easiest way to do this is to
618configure an \f(CW\*(C`ev_timer\*(C'\fR with after=repeat=60 and calling ev_timer_again each 1087configure an \f(CW\*(C`ev_timer\*(C'\fR with a \f(CW\*(C`repeat\*(C'\fR value of \f(CW60\fR and then call
619time you successfully read or write some data. If you go into an idle 1088\&\f(CW\*(C`ev_timer_again\*(C'\fR each time you successfully read or write some data. If
620state where you do not expect data to travel on the socket, you can stop 1089you go into an idle state where you do not expect data to travel on the
621the timer, and again will automatically restart it if need be. 1090socket, you can \f(CW\*(C`ev_timer_stop\*(C'\fR the timer, and \f(CW\*(C`ev_timer_again\*(C'\fR will
1091automatically restart it if need be.
1092.Sp
1093That means you can ignore the \f(CW\*(C`after\*(C'\fR value and \f(CW\*(C`ev_timer_start\*(C'\fR
1094altogether and only ever use the \f(CW\*(C`repeat\*(C'\fR value and \f(CW\*(C`ev_timer_again\*(C'\fR:
1095.Sp
1096.Vb 8
1097\& ev_timer_init (timer, callback, 0., 5.);
1098\& ev_timer_again (loop, timer);
1099\& ...
1100\& timer->again = 17.;
1101\& ev_timer_again (loop, timer);
1102\& ...
1103\& timer->again = 10.;
1104\& ev_timer_again (loop, timer);
1105.Ve
1106.Sp
1107This is more slightly efficient then stopping/starting the timer each time
1108you want to modify its timeout value.
1109.IP "ev_tstamp repeat [read\-write]" 4
1110.IX Item "ev_tstamp repeat [read-write]"
1111The current \f(CW\*(C`repeat\*(C'\fR value. Will be used each time the watcher times out
1112or \f(CW\*(C`ev_timer_again\*(C'\fR is called and determines the next timeout (if any),
1113which is also when any modifications are taken into account.
1114.PP
1115Example: Create a timer that fires after 60 seconds.
1116.PP
1117.Vb 5
1118\& static void
1119\& one_minute_cb (struct ev_loop *loop, struct ev_timer *w, int revents)
1120\& {
1121\& .. one minute over, w is actually stopped right here
1122\& }
1123.Ve
1124.PP
1125.Vb 3
1126\& struct ev_timer mytimer;
1127\& ev_timer_init (&mytimer, one_minute_cb, 60., 0.);
1128\& ev_timer_start (loop, &mytimer);
1129.Ve
1130.PP
1131Example: Create a timeout timer that times out after 10 seconds of
1132inactivity.
1133.PP
1134.Vb 5
1135\& static void
1136\& timeout_cb (struct ev_loop *loop, struct ev_timer *w, int revents)
1137\& {
1138\& .. ten seconds without any activity
1139\& }
1140.Ve
1141.PP
1142.Vb 4
1143\& struct ev_timer mytimer;
1144\& ev_timer_init (&mytimer, timeout_cb, 0., 10.); /* note, only repeat used */
1145\& ev_timer_again (&mytimer); /* start timer */
1146\& ev_loop (loop, 0);
1147.Ve
1148.PP
1149.Vb 3
1150\& // and in some piece of code that gets executed on any "activity":
1151\& // reset the timeout to start ticking again at 10 seconds
1152\& ev_timer_again (&mytimer);
1153.Ve
622.ie n .Sh """ev_periodic"" \- to cron or not to cron" 1154.ie n .Sh """ev_periodic"" \- to cron or not to cron?"
623.el .Sh "\f(CWev_periodic\fP \- to cron or not to cron" 1155.el .Sh "\f(CWev_periodic\fP \- to cron or not to cron?"
624.IX Subsection "ev_periodic - to cron or not to cron" 1156.IX Subsection "ev_periodic - to cron or not to cron?"
625Periodic watchers are also timers of a kind, but they are very versatile 1157Periodic watchers are also timers of a kind, but they are very versatile
626(and unfortunately a bit complex). 1158(and unfortunately a bit complex).
627.PP 1159.PP
628Unlike \f(CW\*(C`ev_timer\*(C'\fR's, they are not based on real time (or relative time) 1160Unlike \f(CW\*(C`ev_timer\*(C'\fR's, they are not based on real time (or relative time)
629but on wallclock time (absolute time). You can tell a periodic watcher 1161but on wallclock time (absolute time). You can tell a periodic watcher
630to trigger \*(L"at\*(R" some specific point in time. For example, if you tell a 1162to trigger \*(L"at\*(R" some specific point in time. For example, if you tell a
631periodic watcher to trigger in 10 seconds (by specifiying e.g. c<ev_now () 1163periodic watcher to trigger in 10 seconds (by specifiying e.g. \f(CW\*(C`ev_now ()
632+ 10.>) and then reset your system clock to the last year, then it will 1164+ 10.\*(C'\fR) and then reset your system clock to the last year, then it will
633take a year to trigger the event (unlike an \f(CW\*(C`ev_timer\*(C'\fR, which would trigger 1165take a year to trigger the event (unlike an \f(CW\*(C`ev_timer\*(C'\fR, which would trigger
634roughly 10 seconds later and of course not if you reset your system time 1166roughly 10 seconds later and of course not if you reset your system time
635again). 1167again).
636.PP 1168.PP
637They can also be used to implement vastly more complex timers, such as 1169They can also be used to implement vastly more complex timers, such as
638triggering an event on eahc midnight, local time. 1170triggering an event on eahc midnight, local time.
1171.PP
1172As with timers, the callback is guarenteed to be invoked only when the
1173time (\f(CW\*(C`at\*(C'\fR) has been passed, but if multiple periodic timers become ready
1174during the same loop iteration then order of execution is undefined.
639.IP "ev_periodic_init (ev_periodic *, callback, ev_tstamp at, ev_tstamp interval, reschedule_cb)" 4 1175.IP "ev_periodic_init (ev_periodic *, callback, ev_tstamp at, ev_tstamp interval, reschedule_cb)" 4
640.IX Item "ev_periodic_init (ev_periodic *, callback, ev_tstamp at, ev_tstamp interval, reschedule_cb)" 1176.IX Item "ev_periodic_init (ev_periodic *, callback, ev_tstamp at, ev_tstamp interval, reschedule_cb)"
641.PD 0 1177.PD 0
642.IP "ev_periodic_set (ev_periodic *, ev_tstamp after, ev_tstamp repeat, reschedule_cb)" 4 1178.IP "ev_periodic_set (ev_periodic *, ev_tstamp after, ev_tstamp repeat, reschedule_cb)" 4
643.IX Item "ev_periodic_set (ev_periodic *, ev_tstamp after, ev_tstamp repeat, reschedule_cb)" 1179.IX Item "ev_periodic_set (ev_periodic *, ev_tstamp after, ev_tstamp repeat, reschedule_cb)"
… …
714.IX Item "ev_periodic_again (loop, ev_periodic *)" 1250.IX Item "ev_periodic_again (loop, ev_periodic *)"
715Simply stops and restarts the periodic watcher again. This is only useful 1251Simply stops and restarts the periodic watcher again. This is only useful
716when you changed some parameters or the reschedule callback would return 1252when you changed some parameters or the reschedule callback would return
717a different time than the last time it was called (e.g. in a crond like 1253a different time than the last time it was called (e.g. in a crond like
718program when the crontabs have changed). 1254program when the crontabs have changed).
1255.IP "ev_tstamp interval [read\-write]" 4
1256.IX Item "ev_tstamp interval [read-write]"
1257The current interval value. Can be modified any time, but changes only
1258take effect when the periodic timer fires or \f(CW\*(C`ev_periodic_again\*(C'\fR is being
1259called.
1260.IP "ev_tstamp (*reschedule_cb)(struct ev_periodic *w, ev_tstamp now) [read\-write]" 4
1261.IX Item "ev_tstamp (*reschedule_cb)(struct ev_periodic *w, ev_tstamp now) [read-write]"
1262The current reschedule callback, or \f(CW0\fR, if this functionality is
1263switched off. Can be changed any time, but changes only take effect when
1264the periodic timer fires or \f(CW\*(C`ev_periodic_again\*(C'\fR is being called.
1265.PP
1266Example: Call a callback every hour, or, more precisely, whenever the
1267system clock is divisible by 3600. The callback invocation times have
1268potentially a lot of jittering, but good long-term stability.
1269.PP
1270.Vb 5
1271\& static void
1272\& clock_cb (struct ev_loop *loop, struct ev_io *w, int revents)
1273\& {
1274\& ... its now a full hour (UTC, or TAI or whatever your clock follows)
1275\& }
1276.Ve
1277.PP
1278.Vb 3
1279\& struct ev_periodic hourly_tick;
1280\& ev_periodic_init (&hourly_tick, clock_cb, 0., 3600., 0);
1281\& ev_periodic_start (loop, &hourly_tick);
1282.Ve
1283.PP
1284Example: The same as above, but use a reschedule callback to do it:
1285.PP
1286.Vb 1
1287\& #include <math.h>
1288.Ve
1289.PP
1290.Vb 5
1291\& static ev_tstamp
1292\& my_scheduler_cb (struct ev_periodic *w, ev_tstamp now)
1293\& {
1294\& return fmod (now, 3600.) + 3600.;
1295\& }
1296.Ve
1297.PP
1298.Vb 1
1299\& ev_periodic_init (&hourly_tick, clock_cb, 0., 0., my_scheduler_cb);
1300.Ve
1301.PP
1302Example: Call a callback every hour, starting now:
1303.PP
1304.Vb 4
1305\& struct ev_periodic hourly_tick;
1306\& ev_periodic_init (&hourly_tick, clock_cb,
1307\& fmod (ev_now (loop), 3600.), 3600., 0);
1308\& ev_periodic_start (loop, &hourly_tick);
1309.Ve
719.ie n .Sh """ev_signal"" \- signal me when a signal gets signalled" 1310.ie n .Sh """ev_signal"" \- signal me when a signal gets signalled!"
720.el .Sh "\f(CWev_signal\fP \- signal me when a signal gets signalled" 1311.el .Sh "\f(CWev_signal\fP \- signal me when a signal gets signalled!"
721.IX Subsection "ev_signal - signal me when a signal gets signalled" 1312.IX Subsection "ev_signal - signal me when a signal gets signalled!"
722Signal watchers will trigger an event when the process receives a specific 1313Signal watchers will trigger an event when the process receives a specific
723signal one or more times. Even though signals are very asynchronous, libev 1314signal one or more times. Even though signals are very asynchronous, libev
724will try it's best to deliver signals synchronously, i.e. as part of the 1315will try it's best to deliver signals synchronously, i.e. as part of the
725normal event processing, like any other event. 1316normal event processing, like any other event.
726.PP 1317.PP
… …
736.IP "ev_signal_set (ev_signal *, int signum)" 4 1327.IP "ev_signal_set (ev_signal *, int signum)" 4
737.IX Item "ev_signal_set (ev_signal *, int signum)" 1328.IX Item "ev_signal_set (ev_signal *, int signum)"
738.PD 1329.PD
739Configures the watcher to trigger on the given signal number (usually one 1330Configures the watcher to trigger on the given signal number (usually one
740of the \f(CW\*(C`SIGxxx\*(C'\fR constants). 1331of the \f(CW\*(C`SIGxxx\*(C'\fR constants).
1332.IP "int signum [read\-only]" 4
1333.IX Item "int signum [read-only]"
1334The signal the watcher watches out for.
741.ie n .Sh """ev_child"" \- wait for pid status changes" 1335.ie n .Sh """ev_child"" \- watch out for process status changes"
742.el .Sh "\f(CWev_child\fP \- wait for pid status changes" 1336.el .Sh "\f(CWev_child\fP \- watch out for process status changes"
743.IX Subsection "ev_child - wait for pid status changes" 1337.IX Subsection "ev_child - watch out for process status changes"
744Child watchers trigger when your process receives a \s-1SIGCHLD\s0 in response to 1338Child watchers trigger when your process receives a \s-1SIGCHLD\s0 in response to
745some child status changes (most typically when a child of yours dies). 1339some child status changes (most typically when a child of yours dies).
746.IP "ev_child_init (ev_child *, callback, int pid)" 4 1340.IP "ev_child_init (ev_child *, callback, int pid)" 4
747.IX Item "ev_child_init (ev_child *, callback, int pid)" 1341.IX Item "ev_child_init (ev_child *, callback, int pid)"
748.PD 0 1342.PD 0
… …
753\&\fIany\fR process if \f(CW\*(C`pid\*(C'\fR is specified as \f(CW0\fR). The callback can look 1347\&\fIany\fR process if \f(CW\*(C`pid\*(C'\fR is specified as \f(CW0\fR). The callback can look
754at the \f(CW\*(C`rstatus\*(C'\fR member of the \f(CW\*(C`ev_child\*(C'\fR watcher structure to see 1348at the \f(CW\*(C`rstatus\*(C'\fR member of the \f(CW\*(C`ev_child\*(C'\fR watcher structure to see
755the status word (use the macros from \f(CW\*(C`sys/wait.h\*(C'\fR and see your systems 1349the status word (use the macros from \f(CW\*(C`sys/wait.h\*(C'\fR and see your systems
756\&\f(CW\*(C`waitpid\*(C'\fR documentation). The \f(CW\*(C`rpid\*(C'\fR member contains the pid of the 1350\&\f(CW\*(C`waitpid\*(C'\fR documentation). The \f(CW\*(C`rpid\*(C'\fR member contains the pid of the
757process causing the status change. 1351process causing the status change.
1352.IP "int pid [read\-only]" 4
1353.IX Item "int pid [read-only]"
1354The process id this watcher watches out for, or \f(CW0\fR, meaning any process id.
1355.IP "int rpid [read\-write]" 4
1356.IX Item "int rpid [read-write]"
1357The process id that detected a status change.
1358.IP "int rstatus [read\-write]" 4
1359.IX Item "int rstatus [read-write]"
1360The process exit/trace status caused by \f(CW\*(C`rpid\*(C'\fR (see your systems
1361\&\f(CW\*(C`waitpid\*(C'\fR and \f(CW\*(C`sys/wait.h\*(C'\fR documentation for details).
1362.PP
1363Example: Try to exit cleanly on \s-1SIGINT\s0 and \s-1SIGTERM\s0.
1364.PP
1365.Vb 5
1366\& static void
1367\& sigint_cb (struct ev_loop *loop, struct ev_signal *w, int revents)
1368\& {
1369\& ev_unloop (loop, EVUNLOOP_ALL);
1370\& }
1371.Ve
1372.PP
1373.Vb 3
1374\& struct ev_signal signal_watcher;
1375\& ev_signal_init (&signal_watcher, sigint_cb, SIGINT);
1376\& ev_signal_start (loop, &sigint_cb);
1377.Ve
1378.ie n .Sh """ev_stat"" \- did the file attributes just change?"
1379.el .Sh "\f(CWev_stat\fP \- did the file attributes just change?"
1380.IX Subsection "ev_stat - did the file attributes just change?"
1381This watches a filesystem path for attribute changes. That is, it calls
1382\&\f(CW\*(C`stat\*(C'\fR regularly (or when the \s-1OS\s0 says it changed) and sees if it changed
1383compared to the last time, invoking the callback if it did.
1384.PP
1385The path does not need to exist: changing from \*(L"path exists\*(R" to \*(L"path does
1386not exist\*(R" is a status change like any other. The condition \*(L"path does
1387not exist\*(R" is signified by the \f(CW\*(C`st_nlink\*(C'\fR field being zero (which is
1388otherwise always forced to be at least one) and all the other fields of
1389the stat buffer having unspecified contents.
1390.PP
1391The path \fIshould\fR be absolute and \fImust not\fR end in a slash. If it is
1392relative and your working directory changes, the behaviour is undefined.
1393.PP
1394Since there is no standard to do this, the portable implementation simply
1395calls \f(CW\*(C`stat (2)\*(C'\fR regularly on the path to see if it changed somehow. You
1396can specify a recommended polling interval for this case. If you specify
1397a polling interval of \f(CW0\fR (highly recommended!) then a \fIsuitable,
1398unspecified default\fR value will be used (which you can expect to be around
1399five seconds, although this might change dynamically). Libev will also
1400impose a minimum interval which is currently around \f(CW0.1\fR, but thats
1401usually overkill.
1402.PP
1403This watcher type is not meant for massive numbers of stat watchers,
1404as even with OS-supported change notifications, this can be
1405resource\-intensive.
1406.PP
1407At the time of this writing, only the Linux inotify interface is
1408implemented (implementing kqueue support is left as an exercise for the
1409reader). Inotify will be used to give hints only and should not change the
1410semantics of \f(CW\*(C`ev_stat\*(C'\fR watchers, which means that libev sometimes needs
1411to fall back to regular polling again even with inotify, but changes are
1412usually detected immediately, and if the file exists there will be no
1413polling.
1414.IP "ev_stat_init (ev_stat *, callback, const char *path, ev_tstamp interval)" 4
1415.IX Item "ev_stat_init (ev_stat *, callback, const char *path, ev_tstamp interval)"
1416.PD 0
1417.IP "ev_stat_set (ev_stat *, const char *path, ev_tstamp interval)" 4
1418.IX Item "ev_stat_set (ev_stat *, const char *path, ev_tstamp interval)"
1419.PD
1420Configures the watcher to wait for status changes of the given
1421\&\f(CW\*(C`path\*(C'\fR. The \f(CW\*(C`interval\*(C'\fR is a hint on how quickly a change is expected to
1422be detected and should normally be specified as \f(CW0\fR to let libev choose
1423a suitable value. The memory pointed to by \f(CW\*(C`path\*(C'\fR must point to the same
1424path for as long as the watcher is active.
1425.Sp
1426The callback will be receive \f(CW\*(C`EV_STAT\*(C'\fR when a change was detected,
1427relative to the attributes at the time the watcher was started (or the
1428last change was detected).
1429.IP "ev_stat_stat (ev_stat *)" 4
1430.IX Item "ev_stat_stat (ev_stat *)"
1431Updates the stat buffer immediately with new values. If you change the
1432watched path in your callback, you could call this fucntion to avoid
1433detecting this change (while introducing a race condition). Can also be
1434useful simply to find out the new values.
1435.IP "ev_statdata attr [read\-only]" 4
1436.IX Item "ev_statdata attr [read-only]"
1437The most-recently detected attributes of the file. Although the type is of
1438\&\f(CW\*(C`ev_statdata\*(C'\fR, this is usually the (or one of the) \f(CW\*(C`struct stat\*(C'\fR types
1439suitable for your system. If the \f(CW\*(C`st_nlink\*(C'\fR member is \f(CW0\fR, then there
1440was some error while \f(CW\*(C`stat\*(C'\fRing the file.
1441.IP "ev_statdata prev [read\-only]" 4
1442.IX Item "ev_statdata prev [read-only]"
1443The previous attributes of the file. The callback gets invoked whenever
1444\&\f(CW\*(C`prev\*(C'\fR != \f(CW\*(C`attr\*(C'\fR.
1445.IP "ev_tstamp interval [read\-only]" 4
1446.IX Item "ev_tstamp interval [read-only]"
1447The specified interval.
1448.IP "const char *path [read\-only]" 4
1449.IX Item "const char *path [read-only]"
1450The filesystem path that is being watched.
1451.PP
1452Example: Watch \f(CW\*(C`/etc/passwd\*(C'\fR for attribute changes.
1453.PP
1454.Vb 15
1455\& static void
1456\& passwd_cb (struct ev_loop *loop, ev_stat *w, int revents)
1457\& {
1458\& /* /etc/passwd changed in some way */
1459\& if (w->attr.st_nlink)
1460\& {
1461\& printf ("passwd current size %ld\en", (long)w->attr.st_size);
1462\& printf ("passwd current atime %ld\en", (long)w->attr.st_mtime);
1463\& printf ("passwd current mtime %ld\en", (long)w->attr.st_mtime);
1464\& }
1465\& else
1466\& /* you shalt not abuse printf for puts */
1467\& puts ("wow, /etc/passwd is not there, expect problems. "
1468\& "if this is windows, they already arrived\en");
1469\& }
1470.Ve
1471.PP
1472.Vb 2
1473\& ...
1474\& ev_stat passwd;
1475.Ve
1476.PP
1477.Vb 2
1478\& ev_stat_init (&passwd, passwd_cb, "/etc/passwd");
1479\& ev_stat_start (loop, &passwd);
1480.Ve
758.ie n .Sh """ev_idle"" \- when you've got nothing better to do" 1481.ie n .Sh """ev_idle"" \- when you've got nothing better to do..."
759.el .Sh "\f(CWev_idle\fP \- when you've got nothing better to do" 1482.el .Sh "\f(CWev_idle\fP \- when you've got nothing better to do..."
760.IX Subsection "ev_idle - when you've got nothing better to do" 1483.IX Subsection "ev_idle - when you've got nothing better to do..."
761Idle watchers trigger events when there are no other events are pending 1484Idle watchers trigger events when there are no other events are pending
762(prepare, check and other idle watchers do not count). That is, as long 1485(prepare, check and other idle watchers do not count). That is, as long
763as your process is busy handling sockets or timeouts (or even signals, 1486as your process is busy handling sockets or timeouts (or even signals,
764imagine) it will not be triggered. But when your process is idle all idle 1487imagine) it will not be triggered. But when your process is idle all idle
765watchers are being called again and again, once per event loop iteration \- 1488watchers are being called again and again, once per event loop iteration \-
… …
776.IP "ev_idle_init (ev_signal *, callback)" 4 1499.IP "ev_idle_init (ev_signal *, callback)" 4
777.IX Item "ev_idle_init (ev_signal *, callback)" 1500.IX Item "ev_idle_init (ev_signal *, callback)"
778Initialises and configures the idle watcher \- it has no parameters of any 1501Initialises and configures the idle watcher \- it has no parameters of any
779kind. There is a \f(CW\*(C`ev_idle_set\*(C'\fR macro, but using it is utterly pointless, 1502kind. There is a \f(CW\*(C`ev_idle_set\*(C'\fR macro, but using it is utterly pointless,
780believe me. 1503believe me.
1504.PP
1505Example: Dynamically allocate an \f(CW\*(C`ev_idle\*(C'\fR watcher, start it, and in the
1506callback, free it. Also, use no error checking, as usual.
1507.PP
1508.Vb 7
1509\& static void
1510\& idle_cb (struct ev_loop *loop, struct ev_idle *w, int revents)
1511\& {
1512\& free (w);
1513\& // now do something you wanted to do when the program has
1514\& // no longer asnything immediate to do.
1515\& }
1516.Ve
1517.PP
1518.Vb 3
1519\& struct ev_idle *idle_watcher = malloc (sizeof (struct ev_idle));
1520\& ev_idle_init (idle_watcher, idle_cb);
1521\& ev_idle_start (loop, idle_cb);
1522.Ve
781.ie n .Sh """ev_prepare""\fP and \f(CW""ev_check"" \- customise your event loop" 1523.ie n .Sh """ev_prepare""\fP and \f(CW""ev_check"" \- customise your event loop!"
782.el .Sh "\f(CWev_prepare\fP and \f(CWev_check\fP \- customise your event loop" 1524.el .Sh "\f(CWev_prepare\fP and \f(CWev_check\fP \- customise your event loop!"
783.IX Subsection "ev_prepare and ev_check - customise your event loop" 1525.IX Subsection "ev_prepare and ev_check - customise your event loop!"
784Prepare and check watchers are usually (but not always) used in tandem: 1526Prepare and check watchers are usually (but not always) used in tandem:
785prepare watchers get invoked before the process blocks and check watchers 1527prepare watchers get invoked before the process blocks and check watchers
786afterwards. 1528afterwards.
787.PP 1529.PP
1530You \fImust not\fR call \f(CW\*(C`ev_loop\*(C'\fR or similar functions that enter
1531the current event loop from either \f(CW\*(C`ev_prepare\*(C'\fR or \f(CW\*(C`ev_check\*(C'\fR
1532watchers. Other loops than the current one are fine, however. The
1533rationale behind this is that you do not need to check for recursion in
1534those watchers, i.e. the sequence will always be \f(CW\*(C`ev_prepare\*(C'\fR, blocking,
1535\&\f(CW\*(C`ev_check\*(C'\fR so if you have one watcher of each kind they will always be
1536called in pairs bracketing the blocking call.
1537.PP
788Their main purpose is to integrate other event mechanisms into libev. This 1538Their main purpose is to integrate other event mechanisms into libev and
789could be used, for example, to track variable changes, implement your own 1539their use is somewhat advanced. This could be used, for example, to track
790watchers, integrate net-snmp or a coroutine library and lots more. 1540variable changes, implement your own watchers, integrate net-snmp or a
1541coroutine library and lots more. They are also occasionally useful if
1542you cache some data and want to flush it before blocking (for example,
1543in X programs you might want to do an \f(CW\*(C`XFlush ()\*(C'\fR in an \f(CW\*(C`ev_prepare\*(C'\fR
1544watcher).
791.PP 1545.PP
792This is done by examining in each prepare call which file descriptors need 1546This is done by examining in each prepare call which file descriptors need
793to be watched by the other library, registering \f(CW\*(C`ev_io\*(C'\fR watchers for 1547to be watched by the other library, registering \f(CW\*(C`ev_io\*(C'\fR watchers for
794them and starting an \f(CW\*(C`ev_timer\*(C'\fR watcher for any timeouts (many libraries 1548them and starting an \f(CW\*(C`ev_timer\*(C'\fR watcher for any timeouts (many libraries
795provide just this functionality). Then, in the check watcher you check for 1549provide just this functionality). Then, in the check watcher you check for
… …
813.IX Item "ev_check_init (ev_check *, callback)" 1567.IX Item "ev_check_init (ev_check *, callback)"
814.PD 1568.PD
815Initialises and configures the prepare or check watcher \- they have no 1569Initialises and configures the prepare or check watcher \- they have no
816parameters of any kind. There are \f(CW\*(C`ev_prepare_set\*(C'\fR and \f(CW\*(C`ev_check_set\*(C'\fR 1570parameters of any kind. There are \f(CW\*(C`ev_prepare_set\*(C'\fR and \f(CW\*(C`ev_check_set\*(C'\fR
817macros, but using them is utterly, utterly and completely pointless. 1571macros, but using them is utterly, utterly and completely pointless.
1572.PP
1573Example: To include a library such as adns, you would add \s-1IO\s0 watchers
1574and a timeout watcher in a prepare handler, as required by libadns, and
1575in a check watcher, destroy them and call into libadns. What follows is
1576pseudo-code only of course:
1577.PP
1578.Vb 2
1579\& static ev_io iow [nfd];
1580\& static ev_timer tw;
1581.Ve
1582.PP
1583.Vb 9
1584\& static void
1585\& io_cb (ev_loop *loop, ev_io *w, int revents)
1586\& {
1587\& // set the relevant poll flags
1588\& // could also call adns_processreadable etc. here
1589\& struct pollfd *fd = (struct pollfd *)w->data;
1590\& if (revents & EV_READ ) fd->revents |= fd->events & POLLIN;
1591\& if (revents & EV_WRITE) fd->revents |= fd->events & POLLOUT;
1592\& }
1593.Ve
1594.PP
1595.Vb 7
1596\& // create io watchers for each fd and a timer before blocking
1597\& static void
1598\& adns_prepare_cb (ev_loop *loop, ev_prepare *w, int revents)
1599\& {
1600\& int timeout = 3600000;truct pollfd fds [nfd];
1601\& // actual code will need to loop here and realloc etc.
1602\& adns_beforepoll (ads, fds, &nfd, &timeout, timeval_from (ev_time ()));
1603.Ve
1604.PP
1605.Vb 3
1606\& /* the callback is illegal, but won't be called as we stop during check */
1607\& ev_timer_init (&tw, 0, timeout * 1e-3);
1608\& ev_timer_start (loop, &tw);
1609.Ve
1610.PP
1611.Vb 6
1612\& // create on ev_io per pollfd
1613\& for (int i = 0; i < nfd; ++i)
1614\& {
1615\& ev_io_init (iow + i, io_cb, fds [i].fd,
1616\& ((fds [i].events & POLLIN ? EV_READ : 0)
1617\& | (fds [i].events & POLLOUT ? EV_WRITE : 0)));
1618.Ve
1619.PP
1620.Vb 5
1621\& fds [i].revents = 0;
1622\& iow [i].data = fds + i;
1623\& ev_io_start (loop, iow + i);
1624\& }
1625\& }
1626.Ve
1627.PP
1628.Vb 5
1629\& // stop all watchers after blocking
1630\& static void
1631\& adns_check_cb (ev_loop *loop, ev_check *w, int revents)
1632\& {
1633\& ev_timer_stop (loop, &tw);
1634.Ve
1635.PP
1636.Vb 2
1637\& for (int i = 0; i < nfd; ++i)
1638\& ev_io_stop (loop, iow + i);
1639.Ve
1640.PP
1641.Vb 2
1642\& adns_afterpoll (adns, fds, nfd, timeval_from (ev_now (loop));
1643\& }
1644.Ve
1645.ie n .Sh """ev_embed"" \- when one backend isn't enough..."
1646.el .Sh "\f(CWev_embed\fP \- when one backend isn't enough..."
1647.IX Subsection "ev_embed - when one backend isn't enough..."
1648This is a rather advanced watcher type that lets you embed one event loop
1649into another (currently only \f(CW\*(C`ev_io\*(C'\fR events are supported in the embedded
1650loop, other types of watchers might be handled in a delayed or incorrect
1651fashion and must not be used).
1652.PP
1653There are primarily two reasons you would want that: work around bugs and
1654prioritise I/O.
1655.PP
1656As an example for a bug workaround, the kqueue backend might only support
1657sockets on some platform, so it is unusable as generic backend, but you
1658still want to make use of it because you have many sockets and it scales
1659so nicely. In this case, you would create a kqueue-based loop and embed it
1660into your default loop (which might use e.g. poll). Overall operation will
1661be a bit slower because first libev has to poll and then call kevent, but
1662at least you can use both at what they are best.
1663.PP
1664As for prioritising I/O: rarely you have the case where some fds have
1665to be watched and handled very quickly (with low latency), and even
1666priorities and idle watchers might have too much overhead. In this case
1667you would put all the high priority stuff in one loop and all the rest in
1668a second one, and embed the second one in the first.
1669.PP
1670As long as the watcher is active, the callback will be invoked every time
1671there might be events pending in the embedded loop. The callback must then
1672call \f(CW\*(C`ev_embed_sweep (mainloop, watcher)\*(C'\fR to make a single sweep and invoke
1673their callbacks (you could also start an idle watcher to give the embedded
1674loop strictly lower priority for example). You can also set the callback
1675to \f(CW0\fR, in which case the embed watcher will automatically execute the
1676embedded loop sweep.
1677.PP
1678As long as the watcher is started it will automatically handle events. The
1679callback will be invoked whenever some events have been handled. You can
1680set the callback to \f(CW0\fR to avoid having to specify one if you are not
1681interested in that.
1682.PP
1683Also, there have not currently been made special provisions for forking:
1684when you fork, you not only have to call \f(CW\*(C`ev_loop_fork\*(C'\fR on both loops,
1685but you will also have to stop and restart any \f(CW\*(C`ev_embed\*(C'\fR watchers
1686yourself.
1687.PP
1688Unfortunately, not all backends are embeddable, only the ones returned by
1689\&\f(CW\*(C`ev_embeddable_backends\*(C'\fR are, which, unfortunately, does not include any
1690portable one.
1691.PP
1692So when you want to use this feature you will always have to be prepared
1693that you cannot get an embeddable loop. The recommended way to get around
1694this is to have a separate variables for your embeddable loop, try to
1695create it, and if that fails, use the normal loop for everything:
1696.PP
1697.Vb 3
1698\& struct ev_loop *loop_hi = ev_default_init (0);
1699\& struct ev_loop *loop_lo = 0;
1700\& struct ev_embed embed;
1701.Ve
1702.PP
1703.Vb 5
1704\& // see if there is a chance of getting one that works
1705\& // (remember that a flags value of 0 means autodetection)
1706\& loop_lo = ev_embeddable_backends () & ev_recommended_backends ()
1707\& ? ev_loop_new (ev_embeddable_backends () & ev_recommended_backends ())
1708\& : 0;
1709.Ve
1710.PP
1711.Vb 8
1712\& // if we got one, then embed it, otherwise default to loop_hi
1713\& if (loop_lo)
1714\& {
1715\& ev_embed_init (&embed, 0, loop_lo);
1716\& ev_embed_start (loop_hi, &embed);
1717\& }
1718\& else
1719\& loop_lo = loop_hi;
1720.Ve
1721.IP "ev_embed_init (ev_embed *, callback, struct ev_loop *embedded_loop)" 4
1722.IX Item "ev_embed_init (ev_embed *, callback, struct ev_loop *embedded_loop)"
1723.PD 0
1724.IP "ev_embed_set (ev_embed *, callback, struct ev_loop *embedded_loop)" 4
1725.IX Item "ev_embed_set (ev_embed *, callback, struct ev_loop *embedded_loop)"
1726.PD
1727Configures the watcher to embed the given loop, which must be
1728embeddable. If the callback is \f(CW0\fR, then \f(CW\*(C`ev_embed_sweep\*(C'\fR will be
1729invoked automatically, otherwise it is the responsibility of the callback
1730to invoke it (it will continue to be called until the sweep has been done,
1731if you do not want thta, you need to temporarily stop the embed watcher).
1732.IP "ev_embed_sweep (loop, ev_embed *)" 4
1733.IX Item "ev_embed_sweep (loop, ev_embed *)"
1734Make a single, non-blocking sweep over the embedded loop. This works
1735similarly to \f(CW\*(C`ev_loop (embedded_loop, EVLOOP_NONBLOCK)\*(C'\fR, but in the most
1736apropriate way for embedded loops.
1737.IP "struct ev_loop *loop [read\-only]" 4
1738.IX Item "struct ev_loop *loop [read-only]"
1739The embedded event loop.
1740.ie n .Sh """ev_fork"" \- the audacity to resume the event loop after a fork"
1741.el .Sh "\f(CWev_fork\fP \- the audacity to resume the event loop after a fork"
1742.IX Subsection "ev_fork - the audacity to resume the event loop after a fork"
1743Fork watchers are called when a \f(CW\*(C`fork ()\*(C'\fR was detected (usually because
1744whoever is a good citizen cared to tell libev about it by calling
1745\&\f(CW\*(C`ev_default_fork\*(C'\fR or \f(CW\*(C`ev_loop_fork\*(C'\fR). The invocation is done before the
1746event loop blocks next and before \f(CW\*(C`ev_check\*(C'\fR watchers are being called,
1747and only in the child after the fork. If whoever good citizen calling
1748\&\f(CW\*(C`ev_default_fork\*(C'\fR cheats and calls it in the wrong process, the fork
1749handlers will be invoked, too, of course.
1750.IP "ev_fork_init (ev_signal *, callback)" 4
1751.IX Item "ev_fork_init (ev_signal *, callback)"
1752Initialises and configures the fork watcher \- it has no parameters of any
1753kind. There is a \f(CW\*(C`ev_fork_set\*(C'\fR macro, but using it is utterly pointless,
1754believe me.
818.SH "OTHER FUNCTIONS" 1755.SH "OTHER FUNCTIONS"
819.IX Header "OTHER FUNCTIONS" 1756.IX Header "OTHER FUNCTIONS"
820There are some other functions of possible interest. Described. Here. Now. 1757There are some other functions of possible interest. Described. Here. Now.
821.IP "ev_once (loop, int fd, int events, ev_tstamp timeout, callback)" 4 1758.IP "ev_once (loop, int fd, int events, ev_tstamp timeout, callback)" 4
822.IX Item "ev_once (loop, int fd, int events, ev_tstamp timeout, callback)" 1759.IX Item "ev_once (loop, int fd, int events, ev_tstamp timeout, callback)"
… …
851.Ve 1788.Ve
852.Sp 1789.Sp
853.Vb 1 1790.Vb 1
854\& ev_once (STDIN_FILENO, EV_READ, 10., stdin_ready, 0); 1791\& ev_once (STDIN_FILENO, EV_READ, 10., stdin_ready, 0);
855.Ve 1792.Ve
856.IP "ev_feed_event (loop, watcher, int events)" 4 1793.IP "ev_feed_event (ev_loop *, watcher *, int revents)" 4
857.IX Item "ev_feed_event (loop, watcher, int events)" 1794.IX Item "ev_feed_event (ev_loop *, watcher *, int revents)"
858Feeds the given event set into the event loop, as if the specified event 1795Feeds the given event set into the event loop, as if the specified event
859had happened for the specified watcher (which must be a pointer to an 1796had happened for the specified watcher (which must be a pointer to an
860initialised but not necessarily started event watcher). 1797initialised but not necessarily started event watcher).
861.IP "ev_feed_fd_event (loop, int fd, int revents)" 4 1798.IP "ev_feed_fd_event (ev_loop *, int fd, int revents)" 4
862.IX Item "ev_feed_fd_event (loop, int fd, int revents)" 1799.IX Item "ev_feed_fd_event (ev_loop *, int fd, int revents)"
863Feed an event on the given fd, as if a file descriptor backend detected 1800Feed an event on the given fd, as if a file descriptor backend detected
864the given events it. 1801the given events it.
865.IP "ev_feed_signal_event (loop, int signum)" 4 1802.IP "ev_feed_signal_event (ev_loop *loop, int signum)" 4
866.IX Item "ev_feed_signal_event (loop, int signum)" 1803.IX Item "ev_feed_signal_event (ev_loop *loop, int signum)"
867Feed an event as if the given signal occured (loop must be the default loop!). 1804Feed an event as if the given signal occured (\f(CW\*(C`loop\*(C'\fR must be the default
1805loop!).
868.SH "LIBEVENT EMULATION" 1806.SH "LIBEVENT EMULATION"
869.IX Header "LIBEVENT EMULATION" 1807.IX Header "LIBEVENT EMULATION"
870Libev offers a compatibility emulation layer for libevent. It cannot 1808Libev offers a compatibility emulation layer for libevent. It cannot
871emulate the internals of libevent, so here are some usage hints: 1809emulate the internals of libevent, so here are some usage hints:
872.IP "* Use it by including <event.h>, as usual." 4 1810.IP "* Use it by including <event.h>, as usual." 4
… …
883.IP "* The libev emulation is \fInot\fR \s-1ABI\s0 compatible to libevent, you need to use the libev header file and library." 4 1821.IP "* The libev emulation is \fInot\fR \s-1ABI\s0 compatible to libevent, you need to use the libev header file and library." 4
884.IX Item "The libev emulation is not ABI compatible to libevent, you need to use the libev header file and library." 1822.IX Item "The libev emulation is not ABI compatible to libevent, you need to use the libev header file and library."
885.PD 1823.PD
886.SH "\*(C+ SUPPORT" 1824.SH "\*(C+ SUPPORT"
887.IX Header " SUPPORT" 1825.IX Header " SUPPORT"
888\&\s-1TBD\s0. 1826Libev comes with some simplistic wrapper classes for \*(C+ that mainly allow
1827you to use some convinience methods to start/stop watchers and also change
1828the callback model to a model using method callbacks on objects.
1829.PP
1830To use it,
1831.PP
1832.Vb 1
1833\& #include <ev++.h>
1834.Ve
1835.PP
1836(it is not installed by default). This automatically includes \fIev.h\fR
1837and puts all of its definitions (many of them macros) into the global
1838namespace. All \*(C+ specific things are put into the \f(CW\*(C`ev\*(C'\fR namespace.
1839.PP
1840It should support all the same embedding options as \fIev.h\fR, most notably
1841\&\f(CW\*(C`EV_MULTIPLICITY\*(C'\fR.
1842.PP
1843Here is a list of things available in the \f(CW\*(C`ev\*(C'\fR namespace:
1844.ie n .IP """ev::READ""\fR, \f(CW""ev::WRITE"" etc." 4
1845.el .IP "\f(CWev::READ\fR, \f(CWev::WRITE\fR etc." 4
1846.IX Item "ev::READ, ev::WRITE etc."
1847These are just enum values with the same values as the \f(CW\*(C`EV_READ\*(C'\fR etc.
1848macros from \fIev.h\fR.
1849.ie n .IP """ev::tstamp""\fR, \f(CW""ev::now""" 4
1850.el .IP "\f(CWev::tstamp\fR, \f(CWev::now\fR" 4
1851.IX Item "ev::tstamp, ev::now"
1852Aliases to the same types/functions as with the \f(CW\*(C`ev_\*(C'\fR prefix.
1853.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
1854.el .IP "\f(CWev::io\fR, \f(CWev::timer\fR, \f(CWev::periodic\fR, \f(CWev::idle\fR, \f(CWev::sig\fR etc." 4
1855.IX Item "ev::io, ev::timer, ev::periodic, ev::idle, ev::sig etc."
1856For each \f(CW\*(C`ev_TYPE\*(C'\fR watcher in \fIev.h\fR there is a corresponding class of
1857the same name in the \f(CW\*(C`ev\*(C'\fR namespace, with the exception of \f(CW\*(C`ev_signal\*(C'\fR
1858which is called \f(CW\*(C`ev::sig\*(C'\fR to avoid clashes with the \f(CW\*(C`signal\*(C'\fR macro
1859defines by many implementations.
1860.Sp
1861All of those classes have these methods:
1862.RS 4
1863.IP "ev::TYPE::TYPE (object *, object::method *)" 4
1864.IX Item "ev::TYPE::TYPE (object *, object::method *)"
1865.PD 0
1866.IP "ev::TYPE::TYPE (object *, object::method *, struct ev_loop *)" 4
1867.IX Item "ev::TYPE::TYPE (object *, object::method *, struct ev_loop *)"
1868.IP "ev::TYPE::~TYPE" 4
1869.IX Item "ev::TYPE::~TYPE"
1870.PD
1871The constructor takes a pointer to an object and a method pointer to
1872the event handler callback to call in this class. The constructor calls
1873\&\f(CW\*(C`ev_init\*(C'\fR for you, which means you have to call the \f(CW\*(C`set\*(C'\fR method
1874before starting it. If you do not specify a loop then the constructor
1875automatically associates the default loop with this watcher.
1876.Sp
1877The destructor automatically stops the watcher if it is active.
1878.IP "w\->set (struct ev_loop *)" 4
1879.IX Item "w->set (struct ev_loop *)"
1880Associates a different \f(CW\*(C`struct ev_loop\*(C'\fR with this watcher. You can only
1881do this when the watcher is inactive (and not pending either).
1882.IP "w\->set ([args])" 4
1883.IX Item "w->set ([args])"
1884Basically the same as \f(CW\*(C`ev_TYPE_set\*(C'\fR, with the same args. Must be
1885called at least once. Unlike the C counterpart, an active watcher gets
1886automatically stopped and restarted.
1887.IP "w\->start ()" 4
1888.IX Item "w->start ()"
1889Starts the watcher. Note that there is no \f(CW\*(C`loop\*(C'\fR argument as the
1890constructor already takes the loop.
1891.IP "w\->stop ()" 4
1892.IX Item "w->stop ()"
1893Stops the watcher if it is active. Again, no \f(CW\*(C`loop\*(C'\fR argument.
1894.ie n .IP "w\->again () ""ev::timer""\fR, \f(CW""ev::periodic"" only" 4
1895.el .IP "w\->again () \f(CWev::timer\fR, \f(CWev::periodic\fR only" 4
1896.IX Item "w->again () ev::timer, ev::periodic only"
1897For \f(CW\*(C`ev::timer\*(C'\fR and \f(CW\*(C`ev::periodic\*(C'\fR, this invokes the corresponding
1898\&\f(CW\*(C`ev_TYPE_again\*(C'\fR function.
1899.ie n .IP "w\->sweep () ""ev::embed"" only" 4
1900.el .IP "w\->sweep () \f(CWev::embed\fR only" 4
1901.IX Item "w->sweep () ev::embed only"
1902Invokes \f(CW\*(C`ev_embed_sweep\*(C'\fR.
1903.ie n .IP "w\->update () ""ev::stat"" only" 4
1904.el .IP "w\->update () \f(CWev::stat\fR only" 4
1905.IX Item "w->update () ev::stat only"
1906Invokes \f(CW\*(C`ev_stat_stat\*(C'\fR.
1907.RE
1908.RS 4
1909.RE
1910.PP
1911Example: Define a class with an \s-1IO\s0 and idle watcher, start one of them in
1912the constructor.
1913.PP
1914.Vb 4
1915\& class myclass
1916\& {
1917\& ev_io io; void io_cb (ev::io &w, int revents);
1918\& ev_idle idle void idle_cb (ev::idle &w, int revents);
1919.Ve
1920.PP
1921.Vb 2
1922\& myclass ();
1923\& }
1924.Ve
1925.PP
1926.Vb 6
1927\& myclass::myclass (int fd)
1928\& : io (this, &myclass::io_cb),
1929\& idle (this, &myclass::idle_cb)
1930\& {
1931\& io.start (fd, ev::READ);
1932\& }
1933.Ve
1934.SH "MACRO MAGIC"
1935.IX Header "MACRO MAGIC"
1936Libev can be compiled with a variety of options, the most fundemantal is
1937\&\f(CW\*(C`EV_MULTIPLICITY\*(C'\fR. This option determines wether (most) functions and
1938callbacks have an initial \f(CW\*(C`struct ev_loop *\*(C'\fR argument.
1939.PP
1940To make it easier to write programs that cope with either variant, the
1941following macros are defined:
1942.ie n .IP """EV_A""\fR, \f(CW""EV_A_""" 4
1943.el .IP "\f(CWEV_A\fR, \f(CWEV_A_\fR" 4
1944.IX Item "EV_A, EV_A_"
1945This provides the loop \fIargument\fR for functions, if one is required (\*(L"ev
1946loop argument\*(R"). The \f(CW\*(C`EV_A\*(C'\fR form is used when this is the sole argument,
1947\&\f(CW\*(C`EV_A_\*(C'\fR is used when other arguments are following. Example:
1948.Sp
1949.Vb 3
1950\& ev_unref (EV_A);
1951\& ev_timer_add (EV_A_ watcher);
1952\& ev_loop (EV_A_ 0);
1953.Ve
1954.Sp
1955It assumes the variable \f(CW\*(C`loop\*(C'\fR of type \f(CW\*(C`struct ev_loop *\*(C'\fR is in scope,
1956which is often provided by the following macro.
1957.ie n .IP """EV_P""\fR, \f(CW""EV_P_""" 4
1958.el .IP "\f(CWEV_P\fR, \f(CWEV_P_\fR" 4
1959.IX Item "EV_P, EV_P_"
1960This provides the loop \fIparameter\fR for functions, if one is required (\*(L"ev
1961loop parameter\*(R"). The \f(CW\*(C`EV_P\*(C'\fR form is used when this is the sole parameter,
1962\&\f(CW\*(C`EV_P_\*(C'\fR is used when other parameters are following. Example:
1963.Sp
1964.Vb 2
1965\& // this is how ev_unref is being declared
1966\& static void ev_unref (EV_P);
1967.Ve
1968.Sp
1969.Vb 2
1970\& // this is how you can declare your typical callback
1971\& static void cb (EV_P_ ev_timer *w, int revents)
1972.Ve
1973.Sp
1974It declares a parameter \f(CW\*(C`loop\*(C'\fR of type \f(CW\*(C`struct ev_loop *\*(C'\fR, quite
1975suitable for use with \f(CW\*(C`EV_A\*(C'\fR.
1976.ie n .IP """EV_DEFAULT""\fR, \f(CW""EV_DEFAULT_""" 4
1977.el .IP "\f(CWEV_DEFAULT\fR, \f(CWEV_DEFAULT_\fR" 4
1978.IX Item "EV_DEFAULT, EV_DEFAULT_"
1979Similar to the other two macros, this gives you the value of the default
1980loop, if multiple loops are supported (\*(L"ev loop default\*(R").
1981.PP
1982Example: Declare and initialise a check watcher, working regardless of
1983wether multiple loops are supported or not.
1984.PP
1985.Vb 5
1986\& static void
1987\& check_cb (EV_P_ ev_timer *w, int revents)
1988\& {
1989\& ev_check_stop (EV_A_ w);
1990\& }
1991.Ve
1992.PP
1993.Vb 4
1994\& ev_check check;
1995\& ev_check_init (&check, check_cb);
1996\& ev_check_start (EV_DEFAULT_ &check);
1997\& ev_loop (EV_DEFAULT_ 0);
1998.Ve
1999.SH "EMBEDDING"
2000.IX Header "EMBEDDING"
2001Libev can (and often is) directly embedded into host
2002applications. Examples of applications that embed it include the Deliantra
2003Game Server, the \s-1EV\s0 perl module, the \s-1GNU\s0 Virtual Private Ethernet (gvpe)
2004and rxvt\-unicode.
2005.PP
2006The goal is to enable you to just copy the neecssary files into your
2007source directory without having to change even a single line in them, so
2008you can easily upgrade by simply copying (or having a checked-out copy of
2009libev somewhere in your source tree).
2010.Sh "\s-1FILESETS\s0"
2011.IX Subsection "FILESETS"
2012Depending on what features you need you need to include one or more sets of files
2013in your app.
2014.PP
2015\fI\s-1CORE\s0 \s-1EVENT\s0 \s-1LOOP\s0\fR
2016.IX Subsection "CORE EVENT LOOP"
2017.PP
2018To include only the libev core (all the \f(CW\*(C`ev_*\*(C'\fR functions), with manual
2019configuration (no autoconf):
2020.PP
2021.Vb 2
2022\& #define EV_STANDALONE 1
2023\& #include "ev.c"
2024.Ve
2025.PP
2026This will automatically include \fIev.h\fR, too, and should be done in a
2027single C source file only to provide the function implementations. To use
2028it, do the same for \fIev.h\fR in all files wishing to use this \s-1API\s0 (best
2029done by writing a wrapper around \fIev.h\fR that you can include instead and
2030where you can put other configuration options):
2031.PP
2032.Vb 2
2033\& #define EV_STANDALONE 1
2034\& #include "ev.h"
2035.Ve
2036.PP
2037Both header files and implementation files can be compiled with a \*(C+
2038compiler (at least, thats a stated goal, and breakage will be treated
2039as a bug).
2040.PP
2041You need the following files in your source tree, or in a directory
2042in your include path (e.g. in libev/ when using \-Ilibev):
2043.PP
2044.Vb 4
2045\& ev.h
2046\& ev.c
2047\& ev_vars.h
2048\& ev_wrap.h
2049.Ve
2050.PP
2051.Vb 1
2052\& ev_win32.c required on win32 platforms only
2053.Ve
2054.PP
2055.Vb 5
2056\& ev_select.c only when select backend is enabled (which is by default)
2057\& ev_poll.c only when poll backend is enabled (disabled by default)
2058\& ev_epoll.c only when the epoll backend is enabled (disabled by default)
2059\& ev_kqueue.c only when the kqueue backend is enabled (disabled by default)
2060\& ev_port.c only when the solaris port backend is enabled (disabled by default)
2061.Ve
2062.PP
2063\&\fIev.c\fR includes the backend files directly when enabled, so you only need
2064to compile this single file.
2065.PP
2066\fI\s-1LIBEVENT\s0 \s-1COMPATIBILITY\s0 \s-1API\s0\fR
2067.IX Subsection "LIBEVENT COMPATIBILITY API"
2068.PP
2069To include the libevent compatibility \s-1API\s0, also include:
2070.PP
2071.Vb 1
2072\& #include "event.c"
2073.Ve
2074.PP
2075in the file including \fIev.c\fR, and:
2076.PP
2077.Vb 1
2078\& #include "event.h"
2079.Ve
2080.PP
2081in the files that want to use the libevent \s-1API\s0. This also includes \fIev.h\fR.
2082.PP
2083You need the following additional files for this:
2084.PP
2085.Vb 2
2086\& event.h
2087\& event.c
2088.Ve
2089.PP
2090\fI\s-1AUTOCONF\s0 \s-1SUPPORT\s0\fR
2091.IX Subsection "AUTOCONF SUPPORT"
2092.PP
2093Instead of using \f(CW\*(C`EV_STANDALONE=1\*(C'\fR and providing your config in
2094whatever way you want, you can also \f(CW\*(C`m4_include([libev.m4])\*(C'\fR in your
2095\&\fIconfigure.ac\fR and leave \f(CW\*(C`EV_STANDALONE\*(C'\fR undefined. \fIev.c\fR will then
2096include \fIconfig.h\fR and configure itself accordingly.
2097.PP
2098For this of course you need the m4 file:
2099.PP
2100.Vb 1
2101\& libev.m4
2102.Ve
2103.Sh "\s-1PREPROCESSOR\s0 \s-1SYMBOLS/MACROS\s0"
2104.IX Subsection "PREPROCESSOR SYMBOLS/MACROS"
2105Libev can be configured via a variety of preprocessor symbols you have to define
2106before including any of its files. The default is not to build for multiplicity
2107and only include the select backend.
2108.IP "\s-1EV_STANDALONE\s0" 4
2109.IX Item "EV_STANDALONE"
2110Must always be \f(CW1\fR if you do not use autoconf configuration, which
2111keeps libev from including \fIconfig.h\fR, and it also defines dummy
2112implementations for some libevent functions (such as logging, which is not
2113supported). It will also not define any of the structs usually found in
2114\&\fIevent.h\fR that are not directly supported by the libev core alone.
2115.IP "\s-1EV_USE_MONOTONIC\s0" 4
2116.IX Item "EV_USE_MONOTONIC"
2117If defined to be \f(CW1\fR, libev will try to detect the availability of the
2118monotonic clock option at both compiletime and runtime. Otherwise no use
2119of the monotonic clock option will be attempted. If you enable this, you
2120usually have to link against librt or something similar. Enabling it when
2121the functionality isn't available is safe, though, althoguh you have
2122to make sure you link against any libraries where the \f(CW\*(C`clock_gettime\*(C'\fR
2123function is hiding in (often \fI\-lrt\fR).
2124.IP "\s-1EV_USE_REALTIME\s0" 4
2125.IX Item "EV_USE_REALTIME"
2126If defined to be \f(CW1\fR, libev will try to detect the availability of the
2127realtime clock option at compiletime (and assume its availability at
2128runtime if successful). Otherwise no use of the realtime clock option will
2129be attempted. This effectively replaces \f(CW\*(C`gettimeofday\*(C'\fR by \f(CW\*(C`clock_get
2130(CLOCK_REALTIME, ...)\*(C'\fR and will not normally affect correctness. See tzhe note about libraries
2131in the description of \f(CW\*(C`EV_USE_MONOTONIC\*(C'\fR, though.
2132.IP "\s-1EV_USE_SELECT\s0" 4
2133.IX Item "EV_USE_SELECT"
2134If undefined or defined to be \f(CW1\fR, libev will compile in support for the
2135\&\f(CW\*(C`select\*(C'\fR(2) backend. No attempt at autodetection will be done: if no
2136other method takes over, select will be it. Otherwise the select backend
2137will not be compiled in.
2138.IP "\s-1EV_SELECT_USE_FD_SET\s0" 4
2139.IX Item "EV_SELECT_USE_FD_SET"
2140If defined to \f(CW1\fR, then the select backend will use the system \f(CW\*(C`fd_set\*(C'\fR
2141structure. This is useful if libev doesn't compile due to a missing
2142\&\f(CW\*(C`NFDBITS\*(C'\fR or \f(CW\*(C`fd_mask\*(C'\fR definition or it misguesses the bitset layout on
2143exotic systems. This usually limits the range of file descriptors to some
2144low limit such as 1024 or might have other limitations (winsocket only
2145allows 64 sockets). The \f(CW\*(C`FD_SETSIZE\*(C'\fR macro, set before compilation, might
2146influence the size of the \f(CW\*(C`fd_set\*(C'\fR used.
2147.IP "\s-1EV_SELECT_IS_WINSOCKET\s0" 4
2148.IX Item "EV_SELECT_IS_WINSOCKET"
2149When defined to \f(CW1\fR, the select backend will assume that
2150select/socket/connect etc. don't understand file descriptors but
2151wants osf handles on win32 (this is the case when the select to
2152be used is the winsock select). This means that it will call
2153\&\f(CW\*(C`_get_osfhandle\*(C'\fR on the fd to convert it to an \s-1OS\s0 handle. Otherwise,
2154it is assumed that all these functions actually work on fds, even
2155on win32. Should not be defined on non\-win32 platforms.
2156.IP "\s-1EV_USE_POLL\s0" 4
2157.IX Item "EV_USE_POLL"
2158If defined to be \f(CW1\fR, libev will compile in support for the \f(CW\*(C`poll\*(C'\fR(2)
2159backend. Otherwise it will be enabled on non\-win32 platforms. It
2160takes precedence over select.
2161.IP "\s-1EV_USE_EPOLL\s0" 4
2162.IX Item "EV_USE_EPOLL"
2163If defined to be \f(CW1\fR, libev will compile in support for the Linux
2164\&\f(CW\*(C`epoll\*(C'\fR(7) backend. Its availability will be detected at runtime,
2165otherwise another method will be used as fallback. This is the
2166preferred backend for GNU/Linux systems.
2167.IP "\s-1EV_USE_KQUEUE\s0" 4
2168.IX Item "EV_USE_KQUEUE"
2169If defined to be \f(CW1\fR, libev will compile in support for the \s-1BSD\s0 style
2170\&\f(CW\*(C`kqueue\*(C'\fR(2) backend. Its actual availability will be detected at runtime,
2171otherwise another method will be used as fallback. This is the preferred
2172backend for \s-1BSD\s0 and BSD-like systems, although on most BSDs kqueue only
2173supports some types of fds correctly (the only platform we found that
2174supports ptys for example was NetBSD), so kqueue might be compiled in, but
2175not be used unless explicitly requested. The best way to use it is to find
2176out whether kqueue supports your type of fd properly and use an embedded
2177kqueue loop.
2178.IP "\s-1EV_USE_PORT\s0" 4
2179.IX Item "EV_USE_PORT"
2180If defined to be \f(CW1\fR, libev will compile in support for the Solaris
218110 port style backend. Its availability will be detected at runtime,
2182otherwise another method will be used as fallback. This is the preferred
2183backend for Solaris 10 systems.
2184.IP "\s-1EV_USE_DEVPOLL\s0" 4
2185.IX Item "EV_USE_DEVPOLL"
2186reserved for future expansion, works like the \s-1USE\s0 symbols above.
2187.IP "\s-1EV_USE_INOTIFY\s0" 4
2188.IX Item "EV_USE_INOTIFY"
2189If defined to be \f(CW1\fR, libev will compile in support for the Linux inotify
2190interface to speed up \f(CW\*(C`ev_stat\*(C'\fR watchers. Its actual availability will
2191be detected at runtime.
2192.IP "\s-1EV_H\s0" 4
2193.IX Item "EV_H"
2194The name of the \fIev.h\fR header file used to include it. The default if
2195undefined is \f(CW\*(C`<ev.h>\*(C'\fR in \fIevent.h\fR and \f(CW"ev.h"\fR in \fIev.c\fR. This
2196can be used to virtually rename the \fIev.h\fR header file in case of conflicts.
2197.IP "\s-1EV_CONFIG_H\s0" 4
2198.IX Item "EV_CONFIG_H"
2199If \f(CW\*(C`EV_STANDALONE\*(C'\fR isn't \f(CW1\fR, this variable can be used to override
2200\&\fIev.c\fR's idea of where to find the \fIconfig.h\fR file, similarly to
2201\&\f(CW\*(C`EV_H\*(C'\fR, above.
2202.IP "\s-1EV_EVENT_H\s0" 4
2203.IX Item "EV_EVENT_H"
2204Similarly to \f(CW\*(C`EV_H\*(C'\fR, this macro can be used to override \fIevent.c\fR's idea
2205of how the \fIevent.h\fR header can be found.
2206.IP "\s-1EV_PROTOTYPES\s0" 4
2207.IX Item "EV_PROTOTYPES"
2208If defined to be \f(CW0\fR, then \fIev.h\fR will not define any function
2209prototypes, but still define all the structs and other symbols. This is
2210occasionally useful if you want to provide your own wrapper functions
2211around libev functions.
2212.IP "\s-1EV_MULTIPLICITY\s0" 4
2213.IX Item "EV_MULTIPLICITY"
2214If undefined or defined to \f(CW1\fR, then all event-loop-specific functions
2215will have the \f(CW\*(C`struct ev_loop *\*(C'\fR as first argument, and you can create
2216additional independent event loops. Otherwise there will be no support
2217for multiple event loops and there is no first event loop pointer
2218argument. Instead, all functions act on the single default loop.
2219.IP "\s-1EV_PERIODIC_ENABLE\s0" 4
2220.IX Item "EV_PERIODIC_ENABLE"
2221If undefined or defined to be \f(CW1\fR, then periodic timers are supported. If
2222defined to be \f(CW0\fR, then they are not. Disabling them saves a few kB of
2223code.
2224.IP "\s-1EV_EMBED_ENABLE\s0" 4
2225.IX Item "EV_EMBED_ENABLE"
2226If undefined or defined to be \f(CW1\fR, then embed watchers are supported. If
2227defined to be \f(CW0\fR, then they are not.
2228.IP "\s-1EV_STAT_ENABLE\s0" 4
2229.IX Item "EV_STAT_ENABLE"
2230If undefined or defined to be \f(CW1\fR, then stat watchers are supported. If
2231defined to be \f(CW0\fR, then they are not.
2232.IP "\s-1EV_FORK_ENABLE\s0" 4
2233.IX Item "EV_FORK_ENABLE"
2234If undefined or defined to be \f(CW1\fR, then fork watchers are supported. If
2235defined to be \f(CW0\fR, then they are not.
2236.IP "\s-1EV_MINIMAL\s0" 4
2237.IX Item "EV_MINIMAL"
2238If you need to shave off some kilobytes of code at the expense of some
2239speed, define this symbol to \f(CW1\fR. Currently only used for gcc to override
2240some inlining decisions, saves roughly 30% codesize of amd64.
2241.IP "\s-1EV_PID_HASHSIZE\s0" 4
2242.IX Item "EV_PID_HASHSIZE"
2243\&\f(CW\*(C`ev_child\*(C'\fR watchers use a small hash table to distribute workload by
2244pid. The default size is \f(CW16\fR (or \f(CW1\fR with \f(CW\*(C`EV_MINIMAL\*(C'\fR), usually more
2245than enough. If you need to manage thousands of children you might want to
2246increase this value (\fImust\fR be a power of two).
2247.IP "\s-1EV_INOTIFY_HASHSIZE\s0" 4
2248.IX Item "EV_INOTIFY_HASHSIZE"
2249\&\f(CW\*(C`ev_staz\*(C'\fR watchers use a small hash table to distribute workload by
2250inotify watch id. The default size is \f(CW16\fR (or \f(CW1\fR with \f(CW\*(C`EV_MINIMAL\*(C'\fR),
2251usually more than enough. If you need to manage thousands of \f(CW\*(C`ev_stat\*(C'\fR
2252watchers you might want to increase this value (\fImust\fR be a power of
2253two).
2254.IP "\s-1EV_COMMON\s0" 4
2255.IX Item "EV_COMMON"
2256By default, all watchers have a \f(CW\*(C`void *data\*(C'\fR member. By redefining
2257this macro to a something else you can include more and other types of
2258members. You have to define it each time you include one of the files,
2259though, and it must be identical each time.
2260.Sp
2261For example, the perl \s-1EV\s0 module uses something like this:
2262.Sp
2263.Vb 3
2264\& #define EV_COMMON \e
2265\& SV *self; /* contains this struct */ \e
2266\& SV *cb_sv, *fh /* note no trailing ";" */
2267.Ve
2268.IP "\s-1EV_CB_DECLARE\s0 (type)" 4
2269.IX Item "EV_CB_DECLARE (type)"
2270.PD 0
2271.IP "\s-1EV_CB_INVOKE\s0 (watcher, revents)" 4
2272.IX Item "EV_CB_INVOKE (watcher, revents)"
2273.IP "ev_set_cb (ev, cb)" 4
2274.IX Item "ev_set_cb (ev, cb)"
2275.PD
2276Can be used to change the callback member declaration in each watcher,
2277and the way callbacks are invoked and set. Must expand to a struct member
2278definition and a statement, respectively. See the \fIev.v\fR header file for
2279their default definitions. One possible use for overriding these is to
2280avoid the \f(CW\*(C`struct ev_loop *\*(C'\fR as first argument in all cases, or to use
2281method calls instead of plain function calls in \*(C+.
2282.Sh "\s-1EXAMPLES\s0"
2283.IX Subsection "EXAMPLES"
2284For a real-world example of a program the includes libev
2285verbatim, you can have a look at the \s-1EV\s0 perl module
2286(<http://software.schmorp.de/pkg/EV.html>). It has the libev files in
2287the \fIlibev/\fR subdirectory and includes them in the \fI\s-1EV/EVAPI\s0.h\fR (public
2288interface) and \fI\s-1EV\s0.xs\fR (implementation) files. Only the \fI\s-1EV\s0.xs\fR file
2289will be compiled. It is pretty complex because it provides its own header
2290file.
2291.Sp
2292The usage in rxvt-unicode is simpler. It has a \fIev_cpp.h\fR header file
2293that everybody includes and which overrides some autoconf choices:
2294.Sp
2295.Vb 4
2296\& #define EV_USE_POLL 0
2297\& #define EV_MULTIPLICITY 0
2298\& #define EV_PERIODICS 0
2299\& #define EV_CONFIG_H <config.h>
2300.Ve
2301.Sp
2302.Vb 1
2303\& #include "ev++.h"
2304.Ve
2305.Sp
2306And a \fIev_cpp.C\fR implementation file that contains libev proper and is compiled:
2307.Sp
2308.Vb 2
2309\& #include "ev_cpp.h"
2310\& #include "ev.c"
2311.Ve
2312.SH "COMPLEXITIES"
2313.IX Header "COMPLEXITIES"
2314In this section the complexities of (many of) the algorithms used inside
2315libev will be explained. For complexity discussions about backends see the
2316documentation for \f(CW\*(C`ev_default_init\*(C'\fR.
2317.RS 4
2318.IP "Starting and stopping timer/periodic watchers: O(log skipped_other_timers)" 4
2319.IX Item "Starting and stopping timer/periodic watchers: O(log skipped_other_timers)"
2320.PD 0
2321.IP "Changing timer/periodic watchers (by autorepeat, again): O(log skipped_other_timers)" 4
2322.IX Item "Changing timer/periodic watchers (by autorepeat, again): O(log skipped_other_timers)"
2323.IP "Starting io/check/prepare/idle/signal/child watchers: O(1)" 4
2324.IX Item "Starting io/check/prepare/idle/signal/child watchers: O(1)"
2325.IP "Stopping check/prepare/idle watchers: O(1)" 4
2326.IX Item "Stopping check/prepare/idle watchers: O(1)"
2327.IP "Stopping an io/signal/child watcher: O(number_of_watchers_for_this_(fd/signal/pid % \s-1EV_PID_HASHSIZE\s0))" 4
2328.IX Item "Stopping an io/signal/child watcher: O(number_of_watchers_for_this_(fd/signal/pid % EV_PID_HASHSIZE))"
2329.IP "Finding the next timer per loop iteration: O(1)" 4
2330.IX Item "Finding the next timer per loop iteration: O(1)"
2331.IP "Each change on a file descriptor per loop iteration: O(number_of_watchers_for_this_fd)" 4
2332.IX Item "Each change on a file descriptor per loop iteration: O(number_of_watchers_for_this_fd)"
2333.IP "Activating one watcher: O(1)" 4
2334.IX Item "Activating one watcher: O(1)"
2335.RE
2336.RS 4
2337.PD
889.SH "AUTHOR" 2338.SH "AUTHOR"
890.IX Header "AUTHOR" 2339.IX Header "AUTHOR"
891Marc Lehmann <libev@schmorp.de>. 2340Marc Lehmann <libev@schmorp.de>.

Diff Legend

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