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

Comparing libev/ev.3 (file contents):
Revision 1.6 by root, Fri Nov 23 05:14:58 2007 UTC vs.
Revision 1.35 by root, Thu Nov 29 17:28:13 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 "2007-11-23" "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
199.Sp 265.Sp
200Usually, 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,
201as this indicates an incompatible change. Minor versions are usually 267as this indicates an incompatible change. Minor versions are usually
202compatible to older versions, so a larger minor version alone is usually 268compatible to older versions, so a larger minor version alone is usually
203not 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
204.IP "unsigned int ev_supported_backends ()" 4 279.IP "unsigned int ev_supported_backends ()" 4
205.IX Item "unsigned int ev_supported_backends ()" 280.IX Item "unsigned int ev_supported_backends ()"
206Return the set of all backends (i.e. their corresponding \f(CW\*(C`EV_BACKEND_*\*(C'\fR 281Return the set of all backends (i.e. their corresponding \f(CW\*(C`EV_BACKEND_*\*(C'\fR
207value) compiled into this binary of libev (independent of their 282value) compiled into this binary of libev (independent of their
208availability on the system you are running on). See \f(CW\*(C`ev_default_loop\*(C'\fR for 283availability on the system you are running on). See \f(CW\*(C`ev_default_loop\*(C'\fR for
209a description of the set values. 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
210.IP "unsigned int ev_recommended_backends ()" 4 293.IP "unsigned int ev_recommended_backends ()" 4
211.IX Item "unsigned int ev_recommended_backends ()" 294.IX Item "unsigned int ev_recommended_backends ()"
212Return the set of all backends compiled into this binary of libev and also 295Return the set of all backends compiled into this binary of libev and also
213recommended for this platform. This set is often smaller than the one 296recommended for this platform. This set is often smaller than the one
214returned by \f(CW\*(C`ev_supported_backends\*(C'\fR, as for example kqueue is broken on 297returned by \f(CW\*(C`ev_supported_backends\*(C'\fR, as for example kqueue is broken on
215most BSDs and will not be autodetected unless you explicitly request it 298most BSDs and will not be autodetected unless you explicitly request it
216(assuming you know what you are doing). This is the set of backends that 299(assuming you know what you are doing). This is the set of backends that
217\&\f(CW\*(C`EVFLAG_AUTO\*(C'\fR will probe for. 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.
218.IP "ev_set_allocator (void *(*cb)(void *ptr, long size))" 4 310.IP "ev_set_allocator (void *(*cb)(void *ptr, long size))" 4
219.IX Item "ev_set_allocator (void *(*cb)(void *ptr, long size))" 311.IX Item "ev_set_allocator (void *(*cb)(void *ptr, long size))"
220Sets the allocation function to use (the prototype is similar to the 312Sets the allocation function to use (the prototype is similar \- the
221realloc C function, the semantics are identical). It is used to allocate 313semantics is identical \- to the realloc C function). It is used to
222and free memory (no surprises here). If it returns zero when memory 314allocate and free memory (no surprises here). If it returns zero when
223needs to be allocated, the library might abort or take some potentially 315memory needs to be allocated, the library might abort or take some
224destructive action. The default is your system realloc function. 316potentially destructive action. The default is your system realloc
317function.
225.Sp 318.Sp
226You could override this function in high-availability programs to, say, 319You could override this function in high-availability programs to, say,
227free 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,
228or 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
229.IP "ev_set_syserr_cb (void (*cb)(const char *msg));" 4 350.IP "ev_set_syserr_cb (void (*cb)(const char *msg));" 4
230.IX Item "ev_set_syserr_cb (void (*cb)(const char *msg));" 351.IX Item "ev_set_syserr_cb (void (*cb)(const char *msg));"
231Set the callback function to call on a retryable syscall error (such 352Set the callback function to call on a retryable syscall error (such
232as failed select, poll, epoll_wait). The message is a printable string 353as failed select, poll, epoll_wait). The message is a printable string
233indicating the system call or subsystem causing the problem. If this 354indicating the system call or subsystem causing the problem. If this
234callback 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
235matter what, when it returns. That is, libev will generally retry the 356matter what, when it returns. That is, libev will generally retry the
236requested 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
237(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
238.SH "FUNCTIONS CONTROLLING THE EVENT LOOP" 375.SH "FUNCTIONS CONTROLLING THE EVENT LOOP"
239.IX Header "FUNCTIONS CONTROLLING THE EVENT LOOP" 376.IX Header "FUNCTIONS CONTROLLING THE EVENT LOOP"
240An event loop is described by a \f(CW\*(C`struct ev_loop *\*(C'\fR. The library knows two 377An event loop is described by a \f(CW\*(C`struct ev_loop *\*(C'\fR. The library knows two
241types of such loops, the \fIdefault\fR loop, which supports signals and child 378types of such loops, the \fIdefault\fR loop, which supports signals and child
242events, and dynamically created loops which do not. 379events, and dynamically created loops which do not.
256.Sp 393.Sp
257If 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
258function. 395function.
259.Sp 396.Sp
260The flags argument can be used to specify special behaviour or specific 397The flags argument can be used to specify special behaviour or specific
261backends to use, and is usually specified as \f(CW0\fR (or \s-1EVFLAG_AUTO\s0). 398backends to use, and is usually specified as \f(CW0\fR (or \f(CW\*(C`EVFLAG_AUTO\*(C'\fR).
262.Sp 399.Sp
263It supports the following flags: 400The following flags are supported:
264.RS 4 401.RS 4
265.ie n .IP """EVFLAG_AUTO""" 4 402.ie n .IP """EVFLAG_AUTO""" 4
266.el .IP "\f(CWEVFLAG_AUTO\fR" 4 403.el .IP "\f(CWEVFLAG_AUTO\fR" 4
267.IX Item "EVFLAG_AUTO" 404.IX Item "EVFLAG_AUTO"
268The 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
274or setgid) then libev will \fInot\fR look at the environment variable 411or setgid) then libev will \fInot\fR look at the environment variable
275\&\f(CW\*(C`LIBEV_FLAGS\*(C'\fR. Otherwise (the default), this environment variable will 412\&\f(CW\*(C`LIBEV_FLAGS\*(C'\fR. Otherwise (the default), this environment variable will
276override the flags completely if it is found in the environment. This is 413override the flags completely if it is found in the environment. This is
277useful to try out specific backends to test their performance, or to work 414useful to try out specific backends to test their performance, or to work
278around bugs. 415around bugs.
416.ie n .IP """EVFLAG_FORKCHECK""" 4
417.el .IP "\f(CWEVFLAG_FORKCHECK\fR" 4
418.IX Item "EVFLAG_FORKCHECK"
419Instead of calling \f(CW\*(C`ev_default_fork\*(C'\fR or \f(CW\*(C`ev_loop_fork\*(C'\fR manually after
420a fork, you can also make libev check for a fork in each iteration by
421enabling this flag.
422.Sp
423This works by calling \f(CW\*(C`getpid ()\*(C'\fR on every iteration of the loop,
424and thus this might slow down your event loop if you do a lot of loop
425iterations and little real work, but is usually not noticable (on my
426Linux system for example, \f(CW\*(C`getpid\*(C'\fR is actually a simple 5\-insn sequence
427without a syscall and thus \fIvery\fR fast, but my Linux system also has
428\&\f(CW\*(C`pthread_atfork\*(C'\fR which is even faster).
429.Sp
430The big advantage of this flag is that you can forget about fork (and
431forget about forgetting to tell libev about forking) when you use this
432flag.
433.Sp
434This flag setting cannot be overriden or specified in the \f(CW\*(C`LIBEV_FLAGS\*(C'\fR
435environment variable.
279.ie n .IP """EVBACKEND_SELECT"" (value 1, portable select backend)" 4 436.ie n .IP """EVBACKEND_SELECT"" (value 1, portable select backend)" 4
280.el .IP "\f(CWEVBACKEND_SELECT\fR (value 1, portable select backend)" 4 437.el .IP "\f(CWEVBACKEND_SELECT\fR (value 1, portable select backend)" 4
281.IX Item "EVBACKEND_SELECT (value 1, portable select backend)" 438.IX Item "EVBACKEND_SELECT (value 1, portable select backend)"
282This is your standard \fIselect\fR\|(2) backend. Not \fIcompletely\fR standard, as 439This is your standard \fIselect\fR\|(2) backend. Not \fIcompletely\fR standard, as
283libev tries to roll its own fd_set with no limits on the number of fds, 440libev tries to roll its own fd_set with no limits on the number of fds,
302While stopping and starting an I/O watcher in the same iteration will 459While stopping and starting an I/O watcher in the same iteration will
303result in some caching, there is still a syscall per such incident 460result in some caching, there is still a syscall per such incident
304(because the fd could point to a different file description now), so its 461(because the fd could point to a different file description now), so its
305best to avoid that. Also, \fIdup()\fRed file descriptors might not work very 462best to avoid that. Also, \fIdup()\fRed file descriptors might not work very
306well if you register events for both fds. 463well if you register events for both fds.
464.Sp
465Please note that epoll sometimes generates spurious notifications, so you
466need to use non-blocking I/O or other means to avoid blocking when no data
467(or space) is available.
307.ie n .IP """EVBACKEND_KQUEUE"" (value 8, most \s-1BSD\s0 clones)" 4 468.ie n .IP """EVBACKEND_KQUEUE"" (value 8, most \s-1BSD\s0 clones)" 4
308.el .IP "\f(CWEVBACKEND_KQUEUE\fR (value 8, most \s-1BSD\s0 clones)" 4 469.el .IP "\f(CWEVBACKEND_KQUEUE\fR (value 8, most \s-1BSD\s0 clones)" 4
309.IX Item "EVBACKEND_KQUEUE (value 8, most BSD clones)" 470.IX Item "EVBACKEND_KQUEUE (value 8, most BSD clones)"
310Kqueue deserves special mention, as at the time of this writing, it 471Kqueue deserves special mention, as at the time of this writing, it
311was broken on all BSDs except NetBSD (usually it doesn't work with 472was broken on all BSDs except NetBSD (usually it doesn't work with
312anything but sockets and pipes, except on Darwin, where of course its 473anything but sockets and pipes, except on Darwin, where of course its
313completely useless). For this reason its not being \*(L"autodetected\*(R" unless 474completely useless). For this reason its not being \*(L"autodetected\*(R"
314you explicitly specify the flags (i.e. you don't use \s-1EVFLAG_AUTO\s0). 475unless you explicitly specify it explicitly in the flags (i.e. using
476\&\f(CW\*(C`EVBACKEND_KQUEUE\*(C'\fR).
315.Sp 477.Sp
316It scales in the same way as the epoll backend, but the interface to the 478It scales in the same way as the epoll backend, but the interface to the
317kernel is more efficient (which says nothing about its actual speed, of 479kernel is more efficient (which says nothing about its actual speed, of
318course). While starting and stopping an I/O watcher does not cause an 480course). While starting and stopping an I/O watcher does not cause an
319extra syscall as with epoll, it still adds up to four event changes per 481extra syscall as with epoll, it still adds up to four event changes per
325.ie n .IP """EVBACKEND_PORT"" (value 32, Solaris 10)" 4 487.ie n .IP """EVBACKEND_PORT"" (value 32, Solaris 10)" 4
326.el .IP "\f(CWEVBACKEND_PORT\fR (value 32, Solaris 10)" 4 488.el .IP "\f(CWEVBACKEND_PORT\fR (value 32, Solaris 10)" 4
327.IX Item "EVBACKEND_PORT (value 32, Solaris 10)" 489.IX Item "EVBACKEND_PORT (value 32, Solaris 10)"
328This uses the Solaris 10 port mechanism. As with everything on Solaris, 490This uses the Solaris 10 port mechanism. As with everything on Solaris,
329it's really slow, but it still scales very well (O(active_fds)). 491it's really slow, but it still scales very well (O(active_fds)).
492.Sp
493Please note that solaris ports can result in a lot of spurious
494notifications, so you need to use non-blocking I/O or other means to avoid
495blocking when no data (or space) is available.
330.ie n .IP """EVBACKEND_ALL""" 4 496.ie n .IP """EVBACKEND_ALL""" 4
331.el .IP "\f(CWEVBACKEND_ALL\fR" 4 497.el .IP "\f(CWEVBACKEND_ALL\fR" 4
332.IX Item "EVBACKEND_ALL" 498.IX Item "EVBACKEND_ALL"
333Try all backends (even potentially broken ones that wouldn't be tried 499Try all backends (even potentially broken ones that wouldn't be tried
334with \f(CW\*(C`EVFLAG_AUTO\*(C'\fR). Since this is a mask, you can do stuff such as 500with \f(CW\*(C`EVFLAG_AUTO\*(C'\fR). Since this is a mask, you can do stuff such as
338.Sp 504.Sp
339If one or more of these are ored into the flags value, then only these 505If one or more of these are ored into the flags value, then only these
340backends will be tried (in the reverse order as given here). If none are 506backends will be tried (in the reverse order as given here). If none are
341specified, most compiled-in backend will be tried, usually in reverse 507specified, most compiled-in backend will be tried, usually in reverse
342order of their flag values :) 508order of their flag values :)
509.Sp
510The most typical usage is like this:
511.Sp
512.Vb 2
513\& if (!ev_default_loop (0))
514\& fatal ("could not initialise libev, bad $LIBEV_FLAGS in environment?");
515.Ve
516.Sp
517Restrict libev to the select and poll backends, and do not allow
518environment settings to be taken into account:
519.Sp
520.Vb 1
521\& ev_default_loop (EVBACKEND_POLL | EVBACKEND_SELECT | EVFLAG_NOENV);
522.Ve
523.Sp
524Use whatever libev has to offer, but make sure that kqueue is used if
525available (warning, breaks stuff, best use only with your own private
526event loop and only if you know the \s-1OS\s0 supports your types of fds):
527.Sp
528.Vb 1
529\& ev_default_loop (ev_recommended_backends () | EVBACKEND_KQUEUE);
530.Ve
343.RE 531.RE
344.IP "struct ev_loop *ev_loop_new (unsigned int flags)" 4 532.IP "struct ev_loop *ev_loop_new (unsigned int flags)" 4
345.IX Item "struct ev_loop *ev_loop_new (unsigned int flags)" 533.IX Item "struct ev_loop *ev_loop_new (unsigned int flags)"
346Similar to \f(CW\*(C`ev_default_loop\*(C'\fR, but always creates a new event loop that is 534Similar to \f(CW\*(C`ev_default_loop\*(C'\fR, but always creates a new event loop that is
347always distinct from the default loop. Unlike the default loop, it cannot 535always distinct from the default loop. Unlike the default loop, it cannot
348handle signal and child watchers, and attempts to do so will be greeted by 536handle signal and child watchers, and attempts to do so will be greeted by
349undefined behaviour (or a failed assertion if assertions are enabled). 537undefined behaviour (or a failed assertion if assertions are enabled).
538.Sp
539Example: Try to create a event loop that uses epoll and nothing else.
540.Sp
541.Vb 3
542\& struct ev_loop *epoller = ev_loop_new (EVBACKEND_EPOLL | EVFLAG_NOENV);
543\& if (!epoller)
544\& fatal ("no epoll found here, maybe it hides under your chair");
545.Ve
350.IP "ev_default_destroy ()" 4 546.IP "ev_default_destroy ()" 4
351.IX Item "ev_default_destroy ()" 547.IX Item "ev_default_destroy ()"
352Destroys the default loop again (frees all memory and kernel state 548Destroys the default loop again (frees all memory and kernel state
353etc.). This stops all registered event watchers (by not touching them in 549etc.). None of the active event watchers will be stopped in the normal
354any way whatsoever, although you cannot rely on this :). 550sense, so e.g. \f(CW\*(C`ev_is_active\*(C'\fR might still return true. It is your
551responsibility to either stop all watchers cleanly yoursef \fIbefore\fR
552calling this function, or cope with the fact afterwards (which is usually
553the easiest thing, youc na just ignore the watchers and/or \f(CW\*(C`free ()\*(C'\fR them
554for example).
355.IP "ev_loop_destroy (loop)" 4 555.IP "ev_loop_destroy (loop)" 4
356.IX Item "ev_loop_destroy (loop)" 556.IX Item "ev_loop_destroy (loop)"
357Like \f(CW\*(C`ev_default_destroy\*(C'\fR, but destroys an event loop created by an 557Like \f(CW\*(C`ev_default_destroy\*(C'\fR, but destroys an event loop created by an
358earlier call to \f(CW\*(C`ev_loop_new\*(C'\fR. 558earlier call to \f(CW\*(C`ev_loop_new\*(C'\fR.
359.IP "ev_default_fork ()" 4 559.IP "ev_default_fork ()" 4
388Returns one of the \f(CW\*(C`EVBACKEND_*\*(C'\fR flags indicating the event backend in 588Returns one of the \f(CW\*(C`EVBACKEND_*\*(C'\fR flags indicating the event backend in
389use. 589use.
390.IP "ev_tstamp ev_now (loop)" 4 590.IP "ev_tstamp ev_now (loop)" 4
391.IX Item "ev_tstamp ev_now (loop)" 591.IX Item "ev_tstamp ev_now (loop)"
392Returns the current \*(L"event loop time\*(R", which is the time the event loop 592Returns the current \*(L"event loop time\*(R", which is the time the event loop
393got events and started processing them. This timestamp does not change 593received events and started processing them. This timestamp does not
394as long as callbacks are being processed, and this is also the base time 594change as long as callbacks are being processed, and this is also the base
395used for relative timers. You can treat it as the timestamp of the event 595time used for relative timers. You can treat it as the timestamp of the
396occuring (or more correctly, the mainloop finding out about it). 596event occuring (or more correctly, libev finding out about it).
397.IP "ev_loop (loop, int flags)" 4 597.IP "ev_loop (loop, int flags)" 4
398.IX Item "ev_loop (loop, int flags)" 598.IX Item "ev_loop (loop, int flags)"
399Finally, this is it, the event handler. This function usually is called 599Finally, this is it, the event handler. This function usually is called
400after you initialised all your watchers and you want to start handling 600after you initialised all your watchers and you want to start handling
401events. 601events.
402.Sp 602.Sp
403If the flags argument is specified as 0, it will not return until either 603If the flags argument is specified as \f(CW0\fR, it will not return until
404no event watchers are active anymore or \f(CW\*(C`ev_unloop\*(C'\fR was called. 604either no event watchers are active anymore or \f(CW\*(C`ev_unloop\*(C'\fR was called.
605.Sp
606Please note that an explicit \f(CW\*(C`ev_unloop\*(C'\fR is usually better than
607relying on all watchers to be stopped when deciding when a program has
608finished (especially in interactive programs), but having a program that
609automatically loops as long as it has to and no longer by virtue of
610relying on its watchers stopping correctly is a thing of beauty.
405.Sp 611.Sp
406A flags value of \f(CW\*(C`EVLOOP_NONBLOCK\*(C'\fR will look for new events, will handle 612A flags value of \f(CW\*(C`EVLOOP_NONBLOCK\*(C'\fR will look for new events, will handle
407those events and any outstanding ones, but will not block your process in 613those events and any outstanding ones, but will not block your process in
408case there are no events and will return after one iteration of the loop. 614case there are no events and will return after one iteration of the loop.
409.Sp 615.Sp
410A flags value of \f(CW\*(C`EVLOOP_ONESHOT\*(C'\fR will look for new events (waiting if 616A flags value of \f(CW\*(C`EVLOOP_ONESHOT\*(C'\fR will look for new events (waiting if
411neccessary) and will handle those and any outstanding ones. It will block 617neccessary) and will handle those and any outstanding ones. It will block
412your process until at least one new event arrives, and will return after 618your process until at least one new event arrives, and will return after
413one iteration of the loop. 619one iteration of the loop. This is useful if you are waiting for some
620external event in conjunction with something not expressible using other
621libev watchers. However, a pair of \f(CW\*(C`ev_prepare\*(C'\fR/\f(CW\*(C`ev_check\*(C'\fR watchers is
622usually a better approach for this kind of thing.
414.Sp 623.Sp
415This flags value could be used to implement alternative looping
416constructs, but the \f(CW\*(C`prepare\*(C'\fR and \f(CW\*(C`check\*(C'\fR watchers provide a better and
417more generic mechanism.
418.Sp
419Here are the gory details of what ev_loop does: 624Here are the gory details of what \f(CW\*(C`ev_loop\*(C'\fR does:
420.Sp 625.Sp
421.Vb 15 626.Vb 18
422\& 1. If there are no active watchers (reference count is zero), return. 627\& * If there are no active watchers (reference count is zero), return.
423\& 2. Queue and immediately call all prepare watchers. 628\& - Queue prepare watchers and then call all outstanding watchers.
424\& 3. If we have been forked, recreate the kernel state. 629\& - If we have been forked, recreate the kernel state.
425\& 4. Update the kernel state with all outstanding changes. 630\& - Update the kernel state with all outstanding changes.
426\& 5. Update the "event loop time". 631\& - Update the "event loop time".
427\& 6. Calculate for how long to block. 632\& - Calculate for how long to block.
428\& 7. Block the process, waiting for events. 633\& - Block the process, waiting for any events.
634\& - Queue all outstanding I/O (fd) events.
429\& 8. Update the "event loop time" and do time jump handling. 635\& - Update the "event loop time" and do time jump handling.
430\& 9. Queue all outstanding timers. 636\& - Queue all outstanding timers.
431\& 10. Queue all outstanding periodics. 637\& - Queue all outstanding periodics.
432\& 11. If no events are pending now, queue all idle watchers. 638\& - If no events are pending now, queue all idle watchers.
433\& 12. Queue all check watchers. 639\& - Queue all check watchers.
434\& 13. Call all queued watchers in reverse order (i.e. check watchers first). 640\& - Call all queued watchers in reverse order (i.e. check watchers first).
641\& Signals and child watchers are implemented as I/O watchers, and will
642\& be handled here by queueing them when their watcher gets executed.
435\& 14. If ev_unloop has been called or EVLOOP_ONESHOT or EVLOOP_NONBLOCK 643\& - If ev_unloop has been called or EVLOOP_ONESHOT or EVLOOP_NONBLOCK
436\& was used, return, otherwise continue with step #1. 644\& were used, return, otherwise continue with step *.
645.Ve
646.Sp
647Example: Queue some jobs and then loop until no events are outsanding
648anymore.
649.Sp
650.Vb 4
651\& ... queue jobs here, make sure they register event watchers as long
652\& ... as they still have work to do (even an idle watcher will do..)
653\& ev_loop (my_loop, 0);
654\& ... jobs done. yeah!
437.Ve 655.Ve
438.IP "ev_unloop (loop, how)" 4 656.IP "ev_unloop (loop, how)" 4
439.IX Item "ev_unloop (loop, how)" 657.IX Item "ev_unloop (loop, how)"
440Can be used to make a call to \f(CW\*(C`ev_loop\*(C'\fR return early (but only after it 658Can be used to make a call to \f(CW\*(C`ev_loop\*(C'\fR return early (but only after it
441has processed all outstanding events). The \f(CW\*(C`how\*(C'\fR argument must be either 659has processed all outstanding events). The \f(CW\*(C`how\*(C'\fR argument must be either
455example, libev itself uses this for its internal signal pipe: It is not 673example, libev itself uses this for its internal signal pipe: It is not
456visible to the libev user and should not keep \f(CW\*(C`ev_loop\*(C'\fR from exiting if 674visible to the libev user and should not keep \f(CW\*(C`ev_loop\*(C'\fR from exiting if
457no event watchers registered by it are active. It is also an excellent 675no event watchers registered by it are active. It is also an excellent
458way to do this for generic recurring timers or from within third-party 676way to do this for generic recurring timers or from within third-party
459libraries. Just remember to \fIunref after start\fR and \fIref before stop\fR. 677libraries. Just remember to \fIunref after start\fR and \fIref before stop\fR.
678.Sp
679Example: Create a signal watcher, but keep it from keeping \f(CW\*(C`ev_loop\*(C'\fR
680running when nothing else is active.
681.Sp
682.Vb 4
683\& struct ev_signal exitsig;
684\& ev_signal_init (&exitsig, sig_cb, SIGINT);
685\& ev_signal_start (loop, &exitsig);
686\& evf_unref (loop);
687.Ve
688.Sp
689Example: For some weird reason, unregister the above signal handler again.
690.Sp
691.Vb 2
692\& ev_ref (loop);
693\& ev_signal_stop (loop, &exitsig);
694.Ve
460.SH "ANATOMY OF A WATCHER" 695.SH "ANATOMY OF A WATCHER"
461.IX Header "ANATOMY OF A WATCHER" 696.IX Header "ANATOMY OF A WATCHER"
462A watcher is a structure that you create and register to record your 697A watcher is a structure that you create and register to record your
463interest in some event. For instance, if you want to wait for \s-1STDIN\s0 to 698interest in some event. For instance, if you want to wait for \s-1STDIN\s0 to
464become readable, you would create an \f(CW\*(C`ev_io\*(C'\fR watcher for that: 699become readable, you would create an \f(CW\*(C`ev_io\*(C'\fR watcher for that:
500*)\*(C'\fR), and you can stop watching for events at any time by calling the 735*)\*(C'\fR), and you can stop watching for events at any time by calling the
501corresponding stop function (\f(CW\*(C`ev_<type>_stop (loop, watcher *)\*(C'\fR. 736corresponding stop function (\f(CW\*(C`ev_<type>_stop (loop, watcher *)\*(C'\fR.
502.PP 737.PP
503As long as your watcher is active (has been started but not stopped) you 738As long as your watcher is active (has been started but not stopped) you
504must not touch the values stored in it. Most specifically you must never 739must not touch the values stored in it. Most specifically you must never
505reinitialise it or call its set macro. 740reinitialise it or call its \f(CW\*(C`set\*(C'\fR macro.
506.PP
507You can check whether an event is active by calling the \f(CW\*(C`ev_is_active
508(watcher *)\*(C'\fR macro. To see whether an event is outstanding (but the
509callback for it has not been called yet) you can use the \f(CW\*(C`ev_is_pending
510(watcher *)\*(C'\fR macro.
511.PP 741.PP
512Each and every callback receives the event loop pointer as first, the 742Each and every callback receives the event loop pointer as first, the
513registered watcher structure as second, and a bitset of received events as 743registered watcher structure as second, and a bitset of received events as
514third argument. 744third argument.
515.PP 745.PP
540The signal specified in the \f(CW\*(C`ev_signal\*(C'\fR watcher has been received by a thread. 770The signal specified in the \f(CW\*(C`ev_signal\*(C'\fR watcher has been received by a thread.
541.ie n .IP """EV_CHILD""" 4 771.ie n .IP """EV_CHILD""" 4
542.el .IP "\f(CWEV_CHILD\fR" 4 772.el .IP "\f(CWEV_CHILD\fR" 4
543.IX Item "EV_CHILD" 773.IX Item "EV_CHILD"
544The pid specified in the \f(CW\*(C`ev_child\*(C'\fR watcher has received a status change. 774The pid specified in the \f(CW\*(C`ev_child\*(C'\fR watcher has received a status change.
775.ie n .IP """EV_STAT""" 4
776.el .IP "\f(CWEV_STAT\fR" 4
777.IX Item "EV_STAT"
778The path specified in the \f(CW\*(C`ev_stat\*(C'\fR watcher changed its attributes somehow.
545.ie n .IP """EV_IDLE""" 4 779.ie n .IP """EV_IDLE""" 4
546.el .IP "\f(CWEV_IDLE\fR" 4 780.el .IP "\f(CWEV_IDLE\fR" 4
547.IX Item "EV_IDLE" 781.IX Item "EV_IDLE"
548The \f(CW\*(C`ev_idle\*(C'\fR watcher has determined that you have nothing better to do. 782The \f(CW\*(C`ev_idle\*(C'\fR watcher has determined that you have nothing better to do.
549.ie n .IP """EV_PREPARE""" 4 783.ie n .IP """EV_PREPARE""" 4
559\&\f(CW\*(C`ev_loop\*(C'\fR has gathered them, but before it invokes any callbacks for any 793\&\f(CW\*(C`ev_loop\*(C'\fR has gathered them, but before it invokes any callbacks for any
560received events. Callbacks of both watcher types can start and stop as 794received events. Callbacks of both watcher types can start and stop as
561many watchers as they want, and all of them will be taken into account 795many watchers as they want, and all of them will be taken into account
562(for example, a \f(CW\*(C`ev_prepare\*(C'\fR watcher might start an idle watcher to keep 796(for example, a \f(CW\*(C`ev_prepare\*(C'\fR watcher might start an idle watcher to keep
563\&\f(CW\*(C`ev_loop\*(C'\fR from blocking). 797\&\f(CW\*(C`ev_loop\*(C'\fR from blocking).
798.ie n .IP """EV_EMBED""" 4
799.el .IP "\f(CWEV_EMBED\fR" 4
800.IX Item "EV_EMBED"
801The embedded event loop specified in the \f(CW\*(C`ev_embed\*(C'\fR watcher needs attention.
802.ie n .IP """EV_FORK""" 4
803.el .IP "\f(CWEV_FORK\fR" 4
804.IX Item "EV_FORK"
805The event loop has been resumed in the child process after fork (see
806\&\f(CW\*(C`ev_fork\*(C'\fR).
564.ie n .IP """EV_ERROR""" 4 807.ie n .IP """EV_ERROR""" 4
565.el .IP "\f(CWEV_ERROR\fR" 4 808.el .IP "\f(CWEV_ERROR\fR" 4
566.IX Item "EV_ERROR" 809.IX Item "EV_ERROR"
567An unspecified error has occured, the watcher has been stopped. This might 810An unspecified error has occured, the watcher has been stopped. This might
568happen because the watcher could not be properly started because libev 811happen because the watcher could not be properly started because libev
573Libev will usually signal a few \*(L"dummy\*(R" events together with an error, 816Libev will usually signal a few \*(L"dummy\*(R" events together with an error,
574for example it might indicate that a fd is readable or writable, and if 817for example it might indicate that a fd is readable or writable, and if
575your callbacks is well-written it can just attempt the operation and cope 818your callbacks is well-written it can just attempt the operation and cope
576with the error from \fIread()\fR or \fIwrite()\fR. This will not work in multithreaded 819with the error from \fIread()\fR or \fIwrite()\fR. This will not work in multithreaded
577programs, though, so beware. 820programs, though, so beware.
821.Sh "\s-1GENERIC\s0 \s-1WATCHER\s0 \s-1FUNCTIONS\s0"
822.IX Subsection "GENERIC WATCHER FUNCTIONS"
823In the following description, \f(CW\*(C`TYPE\*(C'\fR stands for the watcher type,
824e.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.
825.ie n .IP """ev_init"" (ev_TYPE *watcher, callback)" 4
826.el .IP "\f(CWev_init\fR (ev_TYPE *watcher, callback)" 4
827.IX Item "ev_init (ev_TYPE *watcher, callback)"
828This macro initialises the generic portion of a watcher. The contents
829of the watcher object can be arbitrary (so \f(CW\*(C`malloc\*(C'\fR will do). Only
830the generic parts of the watcher are initialised, you \fIneed\fR to call
831the type-specific \f(CW\*(C`ev_TYPE_set\*(C'\fR macro afterwards to initialise the
832type-specific parts. For each type there is also a \f(CW\*(C`ev_TYPE_init\*(C'\fR macro
833which rolls both calls into one.
834.Sp
835You can reinitialise a watcher at any time as long as it has been stopped
836(or never started) and there are no pending events outstanding.
837.Sp
838The callback is always of type \f(CW\*(C`void (*)(ev_loop *loop, ev_TYPE *watcher,
839int revents)\*(C'\fR.
840.ie n .IP """ev_TYPE_set"" (ev_TYPE *, [args])" 4
841.el .IP "\f(CWev_TYPE_set\fR (ev_TYPE *, [args])" 4
842.IX Item "ev_TYPE_set (ev_TYPE *, [args])"
843This macro initialises the type-specific parts of a watcher. You need to
844call \f(CW\*(C`ev_init\*(C'\fR at least once before you call this macro, but you can
845call \f(CW\*(C`ev_TYPE_set\*(C'\fR any number of times. You must not, however, call this
846macro on a watcher that is active (it can be pending, however, which is a
847difference to the \f(CW\*(C`ev_init\*(C'\fR macro).
848.Sp
849Although some watcher types do not have type-specific arguments
850(e.g. \f(CW\*(C`ev_prepare\*(C'\fR) you still need to call its \f(CW\*(C`set\*(C'\fR macro.
851.ie n .IP """ev_TYPE_init"" (ev_TYPE *watcher, callback, [args])" 4
852.el .IP "\f(CWev_TYPE_init\fR (ev_TYPE *watcher, callback, [args])" 4
853.IX Item "ev_TYPE_init (ev_TYPE *watcher, callback, [args])"
854This convinience macro rolls both \f(CW\*(C`ev_init\*(C'\fR and \f(CW\*(C`ev_TYPE_set\*(C'\fR macro
855calls into a single call. This is the most convinient method to initialise
856a watcher. The same limitations apply, of course.
857.ie n .IP """ev_TYPE_start"" (loop *, ev_TYPE *watcher)" 4
858.el .IP "\f(CWev_TYPE_start\fR (loop *, ev_TYPE *watcher)" 4
859.IX Item "ev_TYPE_start (loop *, ev_TYPE *watcher)"
860Starts (activates) the given watcher. Only active watchers will receive
861events. If the watcher is already active nothing will happen.
862.ie n .IP """ev_TYPE_stop"" (loop *, ev_TYPE *watcher)" 4
863.el .IP "\f(CWev_TYPE_stop\fR (loop *, ev_TYPE *watcher)" 4
864.IX Item "ev_TYPE_stop (loop *, ev_TYPE *watcher)"
865Stops the given watcher again (if active) and clears the pending
866status. It is possible that stopped watchers are pending (for example,
867non-repeating timers are being stopped when they become pending), but
868\&\f(CW\*(C`ev_TYPE_stop\*(C'\fR ensures that the watcher is neither active nor pending. If
869you want to free or reuse the memory used by the watcher it is therefore a
870good idea to always call its \f(CW\*(C`ev_TYPE_stop\*(C'\fR function.
871.IP "bool ev_is_active (ev_TYPE *watcher)" 4
872.IX Item "bool ev_is_active (ev_TYPE *watcher)"
873Returns a true value iff the watcher is active (i.e. it has been started
874and not yet been stopped). As long as a watcher is active you must not modify
875it.
876.IP "bool ev_is_pending (ev_TYPE *watcher)" 4
877.IX Item "bool ev_is_pending (ev_TYPE *watcher)"
878Returns a true value iff the watcher is pending, (i.e. it has outstanding
879events but its callback has not yet been invoked). As long as a watcher
880is pending (but not active) you must not call an init function on it (but
881\&\f(CW\*(C`ev_TYPE_set\*(C'\fR is safe) and you must make sure the watcher is available to
882libev (e.g. you cnanot \f(CW\*(C`free ()\*(C'\fR it).
883.IP "callback ev_cb (ev_TYPE *watcher)" 4
884.IX Item "callback ev_cb (ev_TYPE *watcher)"
885Returns the callback currently set on the watcher.
886.IP "ev_cb_set (ev_TYPE *watcher, callback)" 4
887.IX Item "ev_cb_set (ev_TYPE *watcher, callback)"
888Change the callback. You can change the callback at virtually any time
889(modulo threads).
578.Sh "\s-1ASSOCIATING\s0 \s-1CUSTOM\s0 \s-1DATA\s0 \s-1WITH\s0 A \s-1WATCHER\s0" 890.Sh "\s-1ASSOCIATING\s0 \s-1CUSTOM\s0 \s-1DATA\s0 \s-1WITH\s0 A \s-1WATCHER\s0"
579.IX Subsection "ASSOCIATING CUSTOM DATA WITH A WATCHER" 891.IX Subsection "ASSOCIATING CUSTOM DATA WITH A WATCHER"
580Each watcher has, by default, a member \f(CW\*(C`void *data\*(C'\fR that you can change 892Each watcher has, by default, a member \f(CW\*(C`void *data\*(C'\fR that you can change
581and read at any time, libev will completely ignore it. This can be used 893and read at any time, libev will completely ignore it. This can be used
582to associate arbitrary data with your watcher. If you need more data and 894to associate arbitrary data with your watcher. If you need more data and
603\& struct my_io *w = (struct my_io *)w_; 915\& struct my_io *w = (struct my_io *)w_;
604\& ... 916\& ...
605\& } 917\& }
606.Ve 918.Ve
607.PP 919.PP
608More interesting and less C\-conformant ways of catsing your callback type 920More interesting and less C\-conformant ways of casting your callback type
609have been omitted.... 921instead have been omitted.
922.PP
923Another common scenario is having some data structure with multiple
924watchers:
925.PP
926.Vb 6
927\& struct my_biggy
928\& {
929\& int some_data;
930\& ev_timer t1;
931\& ev_timer t2;
932\& }
933.Ve
934.PP
935In this case getting the pointer to \f(CW\*(C`my_biggy\*(C'\fR is a bit more complicated,
936you need to use \f(CW\*(C`offsetof\*(C'\fR:
937.PP
938.Vb 1
939\& #include <stddef.h>
940.Ve
941.PP
942.Vb 6
943\& static void
944\& t1_cb (EV_P_ struct ev_timer *w, int revents)
945\& {
946\& struct my_biggy big = (struct my_biggy *
947\& (((char *)w) - offsetof (struct my_biggy, t1));
948\& }
949.Ve
950.PP
951.Vb 6
952\& static void
953\& t2_cb (EV_P_ struct ev_timer *w, int revents)
954\& {
955\& struct my_biggy big = (struct my_biggy *
956\& (((char *)w) - offsetof (struct my_biggy, t2));
957\& }
958.Ve
610.SH "WATCHER TYPES" 959.SH "WATCHER TYPES"
611.IX Header "WATCHER TYPES" 960.IX Header "WATCHER TYPES"
612This section describes each watcher in detail, but will not repeat 961This section describes each watcher in detail, but will not repeat
613information given in the last section. 962information given in the last section. Any initialisation/set macros,
963functions and members specific to the watcher type are explained.
964.PP
965Members are additionally marked with either \fI[read\-only]\fR, meaning that,
966while the watcher is active, you can look at the member and expect some
967sensible content, but you must not modify it (you can modify it while the
968watcher is stopped to your hearts content), or \fI[read\-write]\fR, which
969means you can expect it to have some sensible content while the watcher
970is active, but you can also modify it. Modifying it may not do something
971sensible or take immediate effect (or do anything at all), but libev will
972not crash or malfunction in any way.
614.ie n .Sh """ev_io"" \- is this file descriptor readable or writable" 973.ie n .Sh """ev_io"" \- is this file descriptor readable or writable?"
615.el .Sh "\f(CWev_io\fP \- is this file descriptor readable or writable" 974.el .Sh "\f(CWev_io\fP \- is this file descriptor readable or writable?"
616.IX Subsection "ev_io - is this file descriptor readable or writable" 975.IX Subsection "ev_io - is this file descriptor readable or writable?"
617I/O watchers check whether a file descriptor is readable or writable 976I/O watchers check whether a file descriptor is readable or writable
618in each iteration of the event loop (This behaviour is called 977in each iteration of the event loop, or, more precisely, when reading
619level-triggering because you keep receiving events as long as the 978would not block the process and writing would at least be able to write
620condition persists. Remember you can stop the watcher if you don't want to 979some data. This behaviour is called level-triggering because you keep
621act on the event and neither want to receive future events). 980receiving events as long as the condition persists. Remember you can stop
981the watcher if you don't want to act on the event and neither want to
982receive future events.
622.PP 983.PP
623In general you can register as many read and/or write event watchers per 984In general you can register as many read and/or write event watchers per
624fd as you want (as long as you don't confuse yourself). Setting all file 985fd as you want (as long as you don't confuse yourself). Setting all file
625descriptors to non-blocking mode is also usually a good idea (but not 986descriptors to non-blocking mode is also usually a good idea (but not
626required if you know what you are doing). 987required if you know what you are doing).
627.PP 988.PP
628You have to be careful with dup'ed file descriptors, though. Some backends 989You have to be careful with dup'ed file descriptors, though. Some backends
629(the linux epoll backend is a notable example) cannot handle dup'ed file 990(the linux epoll backend is a notable example) cannot handle dup'ed file
630descriptors correctly if you register interest in two or more fds pointing 991descriptors correctly if you register interest in two or more fds pointing
631to the same underlying file/socket etc. description (that is, they share 992to the same underlying file/socket/etc. description (that is, they share
632the same underlying \*(L"file open\*(R"). 993the same underlying \*(L"file open\*(R").
633.PP 994.PP
634If you must do this, then force the use of a known-to-be-good backend 995If you must do this, then force the use of a known-to-be-good backend
635(at the time of this writing, this includes only \f(CW\*(C`EVBACKEND_SELECT\*(C'\fR and 996(at the time of this writing, this includes only \f(CW\*(C`EVBACKEND_SELECT\*(C'\fR and
636\&\f(CW\*(C`EVBACKEND_POLL\*(C'\fR). 997\&\f(CW\*(C`EVBACKEND_POLL\*(C'\fR).
998.PP
999Another thing you have to watch out for is that it is quite easy to
1000receive \*(L"spurious\*(R" readyness notifications, that is your callback might
1001be called with \f(CW\*(C`EV_READ\*(C'\fR but a subsequent \f(CW\*(C`read\*(C'\fR(2) will actually block
1002because there is no data. Not only are some backends known to create a
1003lot of those (for example solaris ports), it is very easy to get into
1004this situation even with a relatively standard program structure. Thus
1005it is best to always use non-blocking I/O: An extra \f(CW\*(C`read\*(C'\fR(2) returning
1006\&\f(CW\*(C`EAGAIN\*(C'\fR is far preferable to a program hanging until some data arrives.
1007.PP
1008If you cannot run the fd in non-blocking mode (for example you should not
1009play around with an Xlib connection), then you have to seperately re-test
1010wether a file descriptor is really ready with a known-to-be good interface
1011such as poll (fortunately in our Xlib example, Xlib already does this on
1012its own, so its quite safe to use).
637.IP "ev_io_init (ev_io *, callback, int fd, int events)" 4 1013.IP "ev_io_init (ev_io *, callback, int fd, int events)" 4
638.IX Item "ev_io_init (ev_io *, callback, int fd, int events)" 1014.IX Item "ev_io_init (ev_io *, callback, int fd, int events)"
639.PD 0 1015.PD 0
640.IP "ev_io_set (ev_io *, int fd, int events)" 4 1016.IP "ev_io_set (ev_io *, int fd, int events)" 4
641.IX Item "ev_io_set (ev_io *, int fd, int events)" 1017.IX Item "ev_io_set (ev_io *, int fd, int events)"
642.PD 1018.PD
643Configures an \f(CW\*(C`ev_io\*(C'\fR watcher. The fd is the file descriptor to rceeive 1019Configures an \f(CW\*(C`ev_io\*(C'\fR watcher. The \f(CW\*(C`fd\*(C'\fR is the file descriptor to
644events 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 | 1020rceeive events for and events is either \f(CW\*(C`EV_READ\*(C'\fR, \f(CW\*(C`EV_WRITE\*(C'\fR or
645EV_WRITE\*(C'\fR to receive the given events. 1021\&\f(CW\*(C`EV_READ | EV_WRITE\*(C'\fR to receive the given events.
1022.IP "int fd [read\-only]" 4
1023.IX Item "int fd [read-only]"
1024The file descriptor being watched.
1025.IP "int events [read\-only]" 4
1026.IX Item "int events [read-only]"
1027The events being watched.
1028.PP
1029Example: Call \f(CW\*(C`stdin_readable_cb\*(C'\fR when \s-1STDIN_FILENO\s0 has become, well
1030readable, but only once. Since it is likely line\-buffered, you could
1031attempt to read a whole line in the callback.
1032.PP
1033.Vb 6
1034\& static void
1035\& stdin_readable_cb (struct ev_loop *loop, struct ev_io *w, int revents)
1036\& {
1037\& ev_io_stop (loop, w);
1038\& .. read from stdin here (or from w->fd) and haqndle any I/O errors
1039\& }
1040.Ve
1041.PP
1042.Vb 6
1043\& ...
1044\& struct ev_loop *loop = ev_default_init (0);
1045\& struct ev_io stdin_readable;
1046\& ev_io_init (&stdin_readable, stdin_readable_cb, STDIN_FILENO, EV_READ);
1047\& ev_io_start (loop, &stdin_readable);
1048\& ev_loop (loop, 0);
1049.Ve
646.ie n .Sh """ev_timer"" \- relative and optionally recurring timeouts" 1050.ie n .Sh """ev_timer"" \- relative and optionally repeating timeouts"
647.el .Sh "\f(CWev_timer\fP \- relative and optionally recurring timeouts" 1051.el .Sh "\f(CWev_timer\fP \- relative and optionally repeating timeouts"
648.IX Subsection "ev_timer - relative and optionally recurring timeouts" 1052.IX Subsection "ev_timer - relative and optionally repeating timeouts"
649Timer watchers are simple relative timers that generate an event after a 1053Timer watchers are simple relative timers that generate an event after a
650given time, and optionally repeating in regular intervals after that. 1054given time, and optionally repeating in regular intervals after that.
651.PP 1055.PP
652The timers are based on real time, that is, if you register an event that 1056The timers are based on real time, that is, if you register an event that
653times out after an hour and you reset your system clock to last years 1057times out after an hour and you reset your system clock to last years
687.IP "ev_timer_again (loop)" 4 1091.IP "ev_timer_again (loop)" 4
688.IX Item "ev_timer_again (loop)" 1092.IX Item "ev_timer_again (loop)"
689This will act as if the timer timed out and restart it again if it is 1093This will act as if the timer timed out and restart it again if it is
690repeating. The exact semantics are: 1094repeating. The exact semantics are:
691.Sp 1095.Sp
1096If the timer is pending, its pending status is cleared.
1097.Sp
692If the timer is started but nonrepeating, stop it. 1098If the timer is started but nonrepeating, stop it (as if it timed out).
693.Sp 1099.Sp
694If the timer is repeating, either start it if necessary (with the repeat 1100If the timer is repeating, either start it if necessary (with the
695value), or reset the running timer to the repeat value. 1101\&\f(CW\*(C`repeat\*(C'\fR value), or reset the running timer to the \f(CW\*(C`repeat\*(C'\fR value.
696.Sp 1102.Sp
697This sounds a bit complicated, but here is a useful and typical 1103This sounds a bit complicated, but here is a useful and typical
698example: Imagine you have a tcp connection and you want a so-called idle 1104example: Imagine you have a tcp connection and you want a so-called idle
699timeout, that is, you want to be called when there have been, say, 60 1105timeout, that is, you want to be called when there have been, say, 60
700seconds of inactivity on the socket. The easiest way to do this is to 1106seconds of inactivity on the socket. The easiest way to do this is to
701configure an \f(CW\*(C`ev_timer\*(C'\fR with after=repeat=60 and calling ev_timer_again each 1107configure an \f(CW\*(C`ev_timer\*(C'\fR with a \f(CW\*(C`repeat\*(C'\fR value of \f(CW60\fR and then call
702time you successfully read or write some data. If you go into an idle 1108\&\f(CW\*(C`ev_timer_again\*(C'\fR each time you successfully read or write some data. If
703state where you do not expect data to travel on the socket, you can stop 1109you go into an idle state where you do not expect data to travel on the
704the timer, and again will automatically restart it if need be. 1110socket, you can \f(CW\*(C`ev_timer_stop\*(C'\fR the timer, and \f(CW\*(C`ev_timer_again\*(C'\fR will
1111automatically restart it if need be.
1112.Sp
1113That means you can ignore the \f(CW\*(C`after\*(C'\fR value and \f(CW\*(C`ev_timer_start\*(C'\fR
1114altogether and only ever use the \f(CW\*(C`repeat\*(C'\fR value and \f(CW\*(C`ev_timer_again\*(C'\fR:
1115.Sp
1116.Vb 8
1117\& ev_timer_init (timer, callback, 0., 5.);
1118\& ev_timer_again (loop, timer);
1119\& ...
1120\& timer->again = 17.;
1121\& ev_timer_again (loop, timer);
1122\& ...
1123\& timer->again = 10.;
1124\& ev_timer_again (loop, timer);
1125.Ve
1126.Sp
1127This is more slightly efficient then stopping/starting the timer each time
1128you want to modify its timeout value.
1129.IP "ev_tstamp repeat [read\-write]" 4
1130.IX Item "ev_tstamp repeat [read-write]"
1131The current \f(CW\*(C`repeat\*(C'\fR value. Will be used each time the watcher times out
1132or \f(CW\*(C`ev_timer_again\*(C'\fR is called and determines the next timeout (if any),
1133which is also when any modifications are taken into account.
1134.PP
1135Example: Create a timer that fires after 60 seconds.
1136.PP
1137.Vb 5
1138\& static void
1139\& one_minute_cb (struct ev_loop *loop, struct ev_timer *w, int revents)
1140\& {
1141\& .. one minute over, w is actually stopped right here
1142\& }
1143.Ve
1144.PP
1145.Vb 3
1146\& struct ev_timer mytimer;
1147\& ev_timer_init (&mytimer, one_minute_cb, 60., 0.);
1148\& ev_timer_start (loop, &mytimer);
1149.Ve
1150.PP
1151Example: Create a timeout timer that times out after 10 seconds of
1152inactivity.
1153.PP
1154.Vb 5
1155\& static void
1156\& timeout_cb (struct ev_loop *loop, struct ev_timer *w, int revents)
1157\& {
1158\& .. ten seconds without any activity
1159\& }
1160.Ve
1161.PP
1162.Vb 4
1163\& struct ev_timer mytimer;
1164\& ev_timer_init (&mytimer, timeout_cb, 0., 10.); /* note, only repeat used */
1165\& ev_timer_again (&mytimer); /* start timer */
1166\& ev_loop (loop, 0);
1167.Ve
1168.PP
1169.Vb 3
1170\& // and in some piece of code that gets executed on any "activity":
1171\& // reset the timeout to start ticking again at 10 seconds
1172\& ev_timer_again (&mytimer);
1173.Ve
705.ie n .Sh """ev_periodic"" \- to cron or not to cron" 1174.ie n .Sh """ev_periodic"" \- to cron or not to cron?"
706.el .Sh "\f(CWev_periodic\fP \- to cron or not to cron" 1175.el .Sh "\f(CWev_periodic\fP \- to cron or not to cron?"
707.IX Subsection "ev_periodic - to cron or not to cron" 1176.IX Subsection "ev_periodic - to cron or not to cron?"
708Periodic watchers are also timers of a kind, but they are very versatile 1177Periodic watchers are also timers of a kind, but they are very versatile
709(and unfortunately a bit complex). 1178(and unfortunately a bit complex).
710.PP 1179.PP
711Unlike \f(CW\*(C`ev_timer\*(C'\fR's, they are not based on real time (or relative time) 1180Unlike \f(CW\*(C`ev_timer\*(C'\fR's, they are not based on real time (or relative time)
712but on wallclock time (absolute time). You can tell a periodic watcher 1181but on wallclock time (absolute time). You can tell a periodic watcher
713to trigger \*(L"at\*(R" some specific point in time. For example, if you tell a 1182to trigger \*(L"at\*(R" some specific point in time. For example, if you tell a
714periodic watcher to trigger in 10 seconds (by specifiying e.g. c<ev_now () 1183periodic watcher to trigger in 10 seconds (by specifiying e.g. \f(CW\*(C`ev_now ()
715+ 10.>) and then reset your system clock to the last year, then it will 1184+ 10.\*(C'\fR) and then reset your system clock to the last year, then it will
716take a year to trigger the event (unlike an \f(CW\*(C`ev_timer\*(C'\fR, which would trigger 1185take a year to trigger the event (unlike an \f(CW\*(C`ev_timer\*(C'\fR, which would trigger
717roughly 10 seconds later and of course not if you reset your system time 1186roughly 10 seconds later and of course not if you reset your system time
718again). 1187again).
719.PP 1188.PP
720They can also be used to implement vastly more complex timers, such as 1189They can also be used to implement vastly more complex timers, such as
801.IX Item "ev_periodic_again (loop, ev_periodic *)" 1270.IX Item "ev_periodic_again (loop, ev_periodic *)"
802Simply stops and restarts the periodic watcher again. This is only useful 1271Simply stops and restarts the periodic watcher again. This is only useful
803when you changed some parameters or the reschedule callback would return 1272when you changed some parameters or the reschedule callback would return
804a different time than the last time it was called (e.g. in a crond like 1273a different time than the last time it was called (e.g. in a crond like
805program when the crontabs have changed). 1274program when the crontabs have changed).
1275.IP "ev_tstamp interval [read\-write]" 4
1276.IX Item "ev_tstamp interval [read-write]"
1277The current interval value. Can be modified any time, but changes only
1278take effect when the periodic timer fires or \f(CW\*(C`ev_periodic_again\*(C'\fR is being
1279called.
1280.IP "ev_tstamp (*reschedule_cb)(struct ev_periodic *w, ev_tstamp now) [read\-write]" 4
1281.IX Item "ev_tstamp (*reschedule_cb)(struct ev_periodic *w, ev_tstamp now) [read-write]"
1282The current reschedule callback, or \f(CW0\fR, if this functionality is
1283switched off. Can be changed any time, but changes only take effect when
1284the periodic timer fires or \f(CW\*(C`ev_periodic_again\*(C'\fR is being called.
1285.PP
1286Example: Call a callback every hour, or, more precisely, whenever the
1287system clock is divisible by 3600. The callback invocation times have
1288potentially a lot of jittering, but good long-term stability.
1289.PP
1290.Vb 5
1291\& static void
1292\& clock_cb (struct ev_loop *loop, struct ev_io *w, int revents)
1293\& {
1294\& ... its now a full hour (UTC, or TAI or whatever your clock follows)
1295\& }
1296.Ve
1297.PP
1298.Vb 3
1299\& struct ev_periodic hourly_tick;
1300\& ev_periodic_init (&hourly_tick, clock_cb, 0., 3600., 0);
1301\& ev_periodic_start (loop, &hourly_tick);
1302.Ve
1303.PP
1304Example: The same as above, but use a reschedule callback to do it:
1305.PP
1306.Vb 1
1307\& #include <math.h>
1308.Ve
1309.PP
1310.Vb 5
1311\& static ev_tstamp
1312\& my_scheduler_cb (struct ev_periodic *w, ev_tstamp now)
1313\& {
1314\& return fmod (now, 3600.) + 3600.;
1315\& }
1316.Ve
1317.PP
1318.Vb 1
1319\& ev_periodic_init (&hourly_tick, clock_cb, 0., 0., my_scheduler_cb);
1320.Ve
1321.PP
1322Example: Call a callback every hour, starting now:
1323.PP
1324.Vb 4
1325\& struct ev_periodic hourly_tick;
1326\& ev_periodic_init (&hourly_tick, clock_cb,
1327\& fmod (ev_now (loop), 3600.), 3600., 0);
1328\& ev_periodic_start (loop, &hourly_tick);
1329.Ve
806.ie n .Sh """ev_signal"" \- signal me when a signal gets signalled" 1330.ie n .Sh """ev_signal"" \- signal me when a signal gets signalled!"
807.el .Sh "\f(CWev_signal\fP \- signal me when a signal gets signalled" 1331.el .Sh "\f(CWev_signal\fP \- signal me when a signal gets signalled!"
808.IX Subsection "ev_signal - signal me when a signal gets signalled" 1332.IX Subsection "ev_signal - signal me when a signal gets signalled!"
809Signal watchers will trigger an event when the process receives a specific 1333Signal watchers will trigger an event when the process receives a specific
810signal one or more times. Even though signals are very asynchronous, libev 1334signal one or more times. Even though signals are very asynchronous, libev
811will try it's best to deliver signals synchronously, i.e. as part of the 1335will try it's best to deliver signals synchronously, i.e. as part of the
812normal event processing, like any other event. 1336normal event processing, like any other event.
813.PP 1337.PP
823.IP "ev_signal_set (ev_signal *, int signum)" 4 1347.IP "ev_signal_set (ev_signal *, int signum)" 4
824.IX Item "ev_signal_set (ev_signal *, int signum)" 1348.IX Item "ev_signal_set (ev_signal *, int signum)"
825.PD 1349.PD
826Configures the watcher to trigger on the given signal number (usually one 1350Configures the watcher to trigger on the given signal number (usually one
827of the \f(CW\*(C`SIGxxx\*(C'\fR constants). 1351of the \f(CW\*(C`SIGxxx\*(C'\fR constants).
1352.IP "int signum [read\-only]" 4
1353.IX Item "int signum [read-only]"
1354The signal the watcher watches out for.
828.ie n .Sh """ev_child"" \- wait for pid status changes" 1355.ie n .Sh """ev_child"" \- watch out for process status changes"
829.el .Sh "\f(CWev_child\fP \- wait for pid status changes" 1356.el .Sh "\f(CWev_child\fP \- watch out for process status changes"
830.IX Subsection "ev_child - wait for pid status changes" 1357.IX Subsection "ev_child - watch out for process status changes"
831Child watchers trigger when your process receives a \s-1SIGCHLD\s0 in response to 1358Child watchers trigger when your process receives a \s-1SIGCHLD\s0 in response to
832some child status changes (most typically when a child of yours dies). 1359some child status changes (most typically when a child of yours dies).
833.IP "ev_child_init (ev_child *, callback, int pid)" 4 1360.IP "ev_child_init (ev_child *, callback, int pid)" 4
834.IX Item "ev_child_init (ev_child *, callback, int pid)" 1361.IX Item "ev_child_init (ev_child *, callback, int pid)"
835.PD 0 1362.PD 0
840\&\fIany\fR process if \f(CW\*(C`pid\*(C'\fR is specified as \f(CW0\fR). The callback can look 1367\&\fIany\fR process if \f(CW\*(C`pid\*(C'\fR is specified as \f(CW0\fR). The callback can look
841at the \f(CW\*(C`rstatus\*(C'\fR member of the \f(CW\*(C`ev_child\*(C'\fR watcher structure to see 1368at the \f(CW\*(C`rstatus\*(C'\fR member of the \f(CW\*(C`ev_child\*(C'\fR watcher structure to see
842the status word (use the macros from \f(CW\*(C`sys/wait.h\*(C'\fR and see your systems 1369the status word (use the macros from \f(CW\*(C`sys/wait.h\*(C'\fR and see your systems
843\&\f(CW\*(C`waitpid\*(C'\fR documentation). The \f(CW\*(C`rpid\*(C'\fR member contains the pid of the 1370\&\f(CW\*(C`waitpid\*(C'\fR documentation). The \f(CW\*(C`rpid\*(C'\fR member contains the pid of the
844process causing the status change. 1371process causing the status change.
1372.IP "int pid [read\-only]" 4
1373.IX Item "int pid [read-only]"
1374The process id this watcher watches out for, or \f(CW0\fR, meaning any process id.
1375.IP "int rpid [read\-write]" 4
1376.IX Item "int rpid [read-write]"
1377The process id that detected a status change.
1378.IP "int rstatus [read\-write]" 4
1379.IX Item "int rstatus [read-write]"
1380The process exit/trace status caused by \f(CW\*(C`rpid\*(C'\fR (see your systems
1381\&\f(CW\*(C`waitpid\*(C'\fR and \f(CW\*(C`sys/wait.h\*(C'\fR documentation for details).
1382.PP
1383Example: Try to exit cleanly on \s-1SIGINT\s0 and \s-1SIGTERM\s0.
1384.PP
1385.Vb 5
1386\& static void
1387\& sigint_cb (struct ev_loop *loop, struct ev_signal *w, int revents)
1388\& {
1389\& ev_unloop (loop, EVUNLOOP_ALL);
1390\& }
1391.Ve
1392.PP
1393.Vb 3
1394\& struct ev_signal signal_watcher;
1395\& ev_signal_init (&signal_watcher, sigint_cb, SIGINT);
1396\& ev_signal_start (loop, &sigint_cb);
1397.Ve
1398.ie n .Sh """ev_stat"" \- did the file attributes just change?"
1399.el .Sh "\f(CWev_stat\fP \- did the file attributes just change?"
1400.IX Subsection "ev_stat - did the file attributes just change?"
1401This watches a filesystem path for attribute changes. That is, it calls
1402\&\f(CW\*(C`stat\*(C'\fR regularly (or when the \s-1OS\s0 says it changed) and sees if it changed
1403compared to the last time, invoking the callback if it did.
1404.PP
1405The path does not need to exist: changing from \*(L"path exists\*(R" to \*(L"path does
1406not exist\*(R" is a status change like any other. The condition \*(L"path does
1407not exist\*(R" is signified by the \f(CW\*(C`st_nlink\*(C'\fR field being zero (which is
1408otherwise always forced to be at least one) and all the other fields of
1409the stat buffer having unspecified contents.
1410.PP
1411The path \fIshould\fR be absolute and \fImust not\fR end in a slash. If it is
1412relative and your working directory changes, the behaviour is undefined.
1413.PP
1414Since there is no standard to do this, the portable implementation simply
1415calls \f(CW\*(C`stat (2)\*(C'\fR regularly on the path to see if it changed somehow. You
1416can specify a recommended polling interval for this case. If you specify
1417a polling interval of \f(CW0\fR (highly recommended!) then a \fIsuitable,
1418unspecified default\fR value will be used (which you can expect to be around
1419five seconds, although this might change dynamically). Libev will also
1420impose a minimum interval which is currently around \f(CW0.1\fR, but thats
1421usually overkill.
1422.PP
1423This watcher type is not meant for massive numbers of stat watchers,
1424as even with OS-supported change notifications, this can be
1425resource\-intensive.
1426.PP
1427At the time of this writing, only the Linux inotify interface is
1428implemented (implementing kqueue support is left as an exercise for the
1429reader). Inotify will be used to give hints only and should not change the
1430semantics of \f(CW\*(C`ev_stat\*(C'\fR watchers, which means that libev sometimes needs
1431to fall back to regular polling again even with inotify, but changes are
1432usually detected immediately, and if the file exists there will be no
1433polling.
1434.IP "ev_stat_init (ev_stat *, callback, const char *path, ev_tstamp interval)" 4
1435.IX Item "ev_stat_init (ev_stat *, callback, const char *path, ev_tstamp interval)"
1436.PD 0
1437.IP "ev_stat_set (ev_stat *, const char *path, ev_tstamp interval)" 4
1438.IX Item "ev_stat_set (ev_stat *, const char *path, ev_tstamp interval)"
1439.PD
1440Configures the watcher to wait for status changes of the given
1441\&\f(CW\*(C`path\*(C'\fR. The \f(CW\*(C`interval\*(C'\fR is a hint on how quickly a change is expected to
1442be detected and should normally be specified as \f(CW0\fR to let libev choose
1443a suitable value. The memory pointed to by \f(CW\*(C`path\*(C'\fR must point to the same
1444path for as long as the watcher is active.
1445.Sp
1446The callback will be receive \f(CW\*(C`EV_STAT\*(C'\fR when a change was detected,
1447relative to the attributes at the time the watcher was started (or the
1448last change was detected).
1449.IP "ev_stat_stat (ev_stat *)" 4
1450.IX Item "ev_stat_stat (ev_stat *)"
1451Updates the stat buffer immediately with new values. If you change the
1452watched path in your callback, you could call this fucntion to avoid
1453detecting this change (while introducing a race condition). Can also be
1454useful simply to find out the new values.
1455.IP "ev_statdata attr [read\-only]" 4
1456.IX Item "ev_statdata attr [read-only]"
1457The most-recently detected attributes of the file. Although the type is of
1458\&\f(CW\*(C`ev_statdata\*(C'\fR, this is usually the (or one of the) \f(CW\*(C`struct stat\*(C'\fR types
1459suitable for your system. If the \f(CW\*(C`st_nlink\*(C'\fR member is \f(CW0\fR, then there
1460was some error while \f(CW\*(C`stat\*(C'\fRing the file.
1461.IP "ev_statdata prev [read\-only]" 4
1462.IX Item "ev_statdata prev [read-only]"
1463The previous attributes of the file. The callback gets invoked whenever
1464\&\f(CW\*(C`prev\*(C'\fR != \f(CW\*(C`attr\*(C'\fR.
1465.IP "ev_tstamp interval [read\-only]" 4
1466.IX Item "ev_tstamp interval [read-only]"
1467The specified interval.
1468.IP "const char *path [read\-only]" 4
1469.IX Item "const char *path [read-only]"
1470The filesystem path that is being watched.
1471.PP
1472Example: Watch \f(CW\*(C`/etc/passwd\*(C'\fR for attribute changes.
1473.PP
1474.Vb 15
1475\& static void
1476\& passwd_cb (struct ev_loop *loop, ev_stat *w, int revents)
1477\& {
1478\& /* /etc/passwd changed in some way */
1479\& if (w->attr.st_nlink)
1480\& {
1481\& printf ("passwd current size %ld\en", (long)w->attr.st_size);
1482\& printf ("passwd current atime %ld\en", (long)w->attr.st_mtime);
1483\& printf ("passwd current mtime %ld\en", (long)w->attr.st_mtime);
1484\& }
1485\& else
1486\& /* you shalt not abuse printf for puts */
1487\& puts ("wow, /etc/passwd is not there, expect problems. "
1488\& "if this is windows, they already arrived\en");
1489\& }
1490.Ve
1491.PP
1492.Vb 2
1493\& ...
1494\& ev_stat passwd;
1495.Ve
1496.PP
1497.Vb 2
1498\& ev_stat_init (&passwd, passwd_cb, "/etc/passwd");
1499\& ev_stat_start (loop, &passwd);
1500.Ve
845.ie n .Sh """ev_idle"" \- when you've got nothing better to do" 1501.ie n .Sh """ev_idle"" \- when you've got nothing better to do..."
846.el .Sh "\f(CWev_idle\fP \- when you've got nothing better to do" 1502.el .Sh "\f(CWev_idle\fP \- when you've got nothing better to do..."
847.IX Subsection "ev_idle - when you've got nothing better to do" 1503.IX Subsection "ev_idle - when you've got nothing better to do..."
848Idle watchers trigger events when there are no other events are pending 1504Idle watchers trigger events when there are no other events are pending
849(prepare, check and other idle watchers do not count). That is, as long 1505(prepare, check and other idle watchers do not count). That is, as long
850as your process is busy handling sockets or timeouts (or even signals, 1506as your process is busy handling sockets or timeouts (or even signals,
851imagine) it will not be triggered. But when your process is idle all idle 1507imagine) it will not be triggered. But when your process is idle all idle
852watchers are being called again and again, once per event loop iteration \- 1508watchers are being called again and again, once per event loop iteration \-
863.IP "ev_idle_init (ev_signal *, callback)" 4 1519.IP "ev_idle_init (ev_signal *, callback)" 4
864.IX Item "ev_idle_init (ev_signal *, callback)" 1520.IX Item "ev_idle_init (ev_signal *, callback)"
865Initialises and configures the idle watcher \- it has no parameters of any 1521Initialises and configures the idle watcher \- it has no parameters of any
866kind. There is a \f(CW\*(C`ev_idle_set\*(C'\fR macro, but using it is utterly pointless, 1522kind. There is a \f(CW\*(C`ev_idle_set\*(C'\fR macro, but using it is utterly pointless,
867believe me. 1523believe me.
1524.PP
1525Example: Dynamically allocate an \f(CW\*(C`ev_idle\*(C'\fR watcher, start it, and in the
1526callback, free it. Also, use no error checking, as usual.
1527.PP
1528.Vb 7
1529\& static void
1530\& idle_cb (struct ev_loop *loop, struct ev_idle *w, int revents)
1531\& {
1532\& free (w);
1533\& // now do something you wanted to do when the program has
1534\& // no longer asnything immediate to do.
1535\& }
1536.Ve
1537.PP
1538.Vb 3
1539\& struct ev_idle *idle_watcher = malloc (sizeof (struct ev_idle));
1540\& ev_idle_init (idle_watcher, idle_cb);
1541\& ev_idle_start (loop, idle_cb);
1542.Ve
868.ie n .Sh """ev_prepare""\fP and \f(CW""ev_check"" \- customise your event loop" 1543.ie n .Sh """ev_prepare""\fP and \f(CW""ev_check"" \- customise your event loop!"
869.el .Sh "\f(CWev_prepare\fP and \f(CWev_check\fP \- customise your event loop" 1544.el .Sh "\f(CWev_prepare\fP and \f(CWev_check\fP \- customise your event loop!"
870.IX Subsection "ev_prepare and ev_check - customise your event loop" 1545.IX Subsection "ev_prepare and ev_check - customise your event loop!"
871Prepare and check watchers are usually (but not always) used in tandem: 1546Prepare and check watchers are usually (but not always) used in tandem:
872prepare watchers get invoked before the process blocks and check watchers 1547prepare watchers get invoked before the process blocks and check watchers
873afterwards. 1548afterwards.
874.PP 1549.PP
1550You \fImust not\fR call \f(CW\*(C`ev_loop\*(C'\fR or similar functions that enter
1551the current event loop from either \f(CW\*(C`ev_prepare\*(C'\fR or \f(CW\*(C`ev_check\*(C'\fR
1552watchers. Other loops than the current one are fine, however. The
1553rationale behind this is that you do not need to check for recursion in
1554those watchers, i.e. the sequence will always be \f(CW\*(C`ev_prepare\*(C'\fR, blocking,
1555\&\f(CW\*(C`ev_check\*(C'\fR so if you have one watcher of each kind they will always be
1556called in pairs bracketing the blocking call.
1557.PP
875Their main purpose is to integrate other event mechanisms into libev. This 1558Their main purpose is to integrate other event mechanisms into libev and
876could be used, for example, to track variable changes, implement your own 1559their use is somewhat advanced. This could be used, for example, to track
877watchers, integrate net-snmp or a coroutine library and lots more. 1560variable changes, implement your own watchers, integrate net-snmp or a
1561coroutine library and lots more. They are also occasionally useful if
1562you cache some data and want to flush it before blocking (for example,
1563in X programs you might want to do an \f(CW\*(C`XFlush ()\*(C'\fR in an \f(CW\*(C`ev_prepare\*(C'\fR
1564watcher).
878.PP 1565.PP
879This is done by examining in each prepare call which file descriptors need 1566This is done by examining in each prepare call which file descriptors need
880to be watched by the other library, registering \f(CW\*(C`ev_io\*(C'\fR watchers for 1567to be watched by the other library, registering \f(CW\*(C`ev_io\*(C'\fR watchers for
881them and starting an \f(CW\*(C`ev_timer\*(C'\fR watcher for any timeouts (many libraries 1568them and starting an \f(CW\*(C`ev_timer\*(C'\fR watcher for any timeouts (many libraries
882provide just this functionality). Then, in the check watcher you check for 1569provide just this functionality). Then, in the check watcher you check for
900.IX Item "ev_check_init (ev_check *, callback)" 1587.IX Item "ev_check_init (ev_check *, callback)"
901.PD 1588.PD
902Initialises and configures the prepare or check watcher \- they have no 1589Initialises and configures the prepare or check watcher \- they have no
903parameters of any kind. There are \f(CW\*(C`ev_prepare_set\*(C'\fR and \f(CW\*(C`ev_check_set\*(C'\fR 1590parameters of any kind. There are \f(CW\*(C`ev_prepare_set\*(C'\fR and \f(CW\*(C`ev_check_set\*(C'\fR
904macros, but using them is utterly, utterly and completely pointless. 1591macros, but using them is utterly, utterly and completely pointless.
1592.PP
1593Example: To include a library such as adns, you would add \s-1IO\s0 watchers
1594and a timeout watcher in a prepare handler, as required by libadns, and
1595in a check watcher, destroy them and call into libadns. What follows is
1596pseudo-code only of course:
1597.PP
1598.Vb 2
1599\& static ev_io iow [nfd];
1600\& static ev_timer tw;
1601.Ve
1602.PP
1603.Vb 9
1604\& static void
1605\& io_cb (ev_loop *loop, ev_io *w, int revents)
1606\& {
1607\& // set the relevant poll flags
1608\& // could also call adns_processreadable etc. here
1609\& struct pollfd *fd = (struct pollfd *)w->data;
1610\& if (revents & EV_READ ) fd->revents |= fd->events & POLLIN;
1611\& if (revents & EV_WRITE) fd->revents |= fd->events & POLLOUT;
1612\& }
1613.Ve
1614.PP
1615.Vb 7
1616\& // create io watchers for each fd and a timer before blocking
1617\& static void
1618\& adns_prepare_cb (ev_loop *loop, ev_prepare *w, int revents)
1619\& {
1620\& int timeout = 3600000;truct pollfd fds [nfd];
1621\& // actual code will need to loop here and realloc etc.
1622\& adns_beforepoll (ads, fds, &nfd, &timeout, timeval_from (ev_time ()));
1623.Ve
1624.PP
1625.Vb 3
1626\& /* the callback is illegal, but won't be called as we stop during check */
1627\& ev_timer_init (&tw, 0, timeout * 1e-3);
1628\& ev_timer_start (loop, &tw);
1629.Ve
1630.PP
1631.Vb 6
1632\& // create on ev_io per pollfd
1633\& for (int i = 0; i < nfd; ++i)
1634\& {
1635\& ev_io_init (iow + i, io_cb, fds [i].fd,
1636\& ((fds [i].events & POLLIN ? EV_READ : 0)
1637\& | (fds [i].events & POLLOUT ? EV_WRITE : 0)));
1638.Ve
1639.PP
1640.Vb 5
1641\& fds [i].revents = 0;
1642\& iow [i].data = fds + i;
1643\& ev_io_start (loop, iow + i);
1644\& }
1645\& }
1646.Ve
1647.PP
1648.Vb 5
1649\& // stop all watchers after blocking
1650\& static void
1651\& adns_check_cb (ev_loop *loop, ev_check *w, int revents)
1652\& {
1653\& ev_timer_stop (loop, &tw);
1654.Ve
1655.PP
1656.Vb 2
1657\& for (int i = 0; i < nfd; ++i)
1658\& ev_io_stop (loop, iow + i);
1659.Ve
1660.PP
1661.Vb 2
1662\& adns_afterpoll (adns, fds, nfd, timeval_from (ev_now (loop));
1663\& }
1664.Ve
1665.ie n .Sh """ev_embed"" \- when one backend isn't enough..."
1666.el .Sh "\f(CWev_embed\fP \- when one backend isn't enough..."
1667.IX Subsection "ev_embed - when one backend isn't enough..."
1668This is a rather advanced watcher type that lets you embed one event loop
1669into another (currently only \f(CW\*(C`ev_io\*(C'\fR events are supported in the embedded
1670loop, other types of watchers might be handled in a delayed or incorrect
1671fashion and must not be used).
1672.PP
1673There are primarily two reasons you would want that: work around bugs and
1674prioritise I/O.
1675.PP
1676As an example for a bug workaround, the kqueue backend might only support
1677sockets on some platform, so it is unusable as generic backend, but you
1678still want to make use of it because you have many sockets and it scales
1679so nicely. In this case, you would create a kqueue-based loop and embed it
1680into your default loop (which might use e.g. poll). Overall operation will
1681be a bit slower because first libev has to poll and then call kevent, but
1682at least you can use both at what they are best.
1683.PP
1684As for prioritising I/O: rarely you have the case where some fds have
1685to be watched and handled very quickly (with low latency), and even
1686priorities and idle watchers might have too much overhead. In this case
1687you would put all the high priority stuff in one loop and all the rest in
1688a second one, and embed the second one in the first.
1689.PP
1690As long as the watcher is active, the callback will be invoked every time
1691there might be events pending in the embedded loop. The callback must then
1692call \f(CW\*(C`ev_embed_sweep (mainloop, watcher)\*(C'\fR to make a single sweep and invoke
1693their callbacks (you could also start an idle watcher to give the embedded
1694loop strictly lower priority for example). You can also set the callback
1695to \f(CW0\fR, in which case the embed watcher will automatically execute the
1696embedded loop sweep.
1697.PP
1698As long as the watcher is started it will automatically handle events. The
1699callback will be invoked whenever some events have been handled. You can
1700set the callback to \f(CW0\fR to avoid having to specify one if you are not
1701interested in that.
1702.PP
1703Also, there have not currently been made special provisions for forking:
1704when you fork, you not only have to call \f(CW\*(C`ev_loop_fork\*(C'\fR on both loops,
1705but you will also have to stop and restart any \f(CW\*(C`ev_embed\*(C'\fR watchers
1706yourself.
1707.PP
1708Unfortunately, not all backends are embeddable, only the ones returned by
1709\&\f(CW\*(C`ev_embeddable_backends\*(C'\fR are, which, unfortunately, does not include any
1710portable one.
1711.PP
1712So when you want to use this feature you will always have to be prepared
1713that you cannot get an embeddable loop. The recommended way to get around
1714this is to have a separate variables for your embeddable loop, try to
1715create it, and if that fails, use the normal loop for everything:
1716.PP
1717.Vb 3
1718\& struct ev_loop *loop_hi = ev_default_init (0);
1719\& struct ev_loop *loop_lo = 0;
1720\& struct ev_embed embed;
1721.Ve
1722.PP
1723.Vb 5
1724\& // see if there is a chance of getting one that works
1725\& // (remember that a flags value of 0 means autodetection)
1726\& loop_lo = ev_embeddable_backends () & ev_recommended_backends ()
1727\& ? ev_loop_new (ev_embeddable_backends () & ev_recommended_backends ())
1728\& : 0;
1729.Ve
1730.PP
1731.Vb 8
1732\& // if we got one, then embed it, otherwise default to loop_hi
1733\& if (loop_lo)
1734\& {
1735\& ev_embed_init (&embed, 0, loop_lo);
1736\& ev_embed_start (loop_hi, &embed);
1737\& }
1738\& else
1739\& loop_lo = loop_hi;
1740.Ve
1741.IP "ev_embed_init (ev_embed *, callback, struct ev_loop *embedded_loop)" 4
1742.IX Item "ev_embed_init (ev_embed *, callback, struct ev_loop *embedded_loop)"
1743.PD 0
1744.IP "ev_embed_set (ev_embed *, callback, struct ev_loop *embedded_loop)" 4
1745.IX Item "ev_embed_set (ev_embed *, callback, struct ev_loop *embedded_loop)"
1746.PD
1747Configures the watcher to embed the given loop, which must be
1748embeddable. If the callback is \f(CW0\fR, then \f(CW\*(C`ev_embed_sweep\*(C'\fR will be
1749invoked automatically, otherwise it is the responsibility of the callback
1750to invoke it (it will continue to be called until the sweep has been done,
1751if you do not want thta, you need to temporarily stop the embed watcher).
1752.IP "ev_embed_sweep (loop, ev_embed *)" 4
1753.IX Item "ev_embed_sweep (loop, ev_embed *)"
1754Make a single, non-blocking sweep over the embedded loop. This works
1755similarly to \f(CW\*(C`ev_loop (embedded_loop, EVLOOP_NONBLOCK)\*(C'\fR, but in the most
1756apropriate way for embedded loops.
1757.IP "struct ev_loop *loop [read\-only]" 4
1758.IX Item "struct ev_loop *loop [read-only]"
1759The embedded event loop.
1760.ie n .Sh """ev_fork"" \- the audacity to resume the event loop after a fork"
1761.el .Sh "\f(CWev_fork\fP \- the audacity to resume the event loop after a fork"
1762.IX Subsection "ev_fork - the audacity to resume the event loop after a fork"
1763Fork watchers are called when a \f(CW\*(C`fork ()\*(C'\fR was detected (usually because
1764whoever is a good citizen cared to tell libev about it by calling
1765\&\f(CW\*(C`ev_default_fork\*(C'\fR or \f(CW\*(C`ev_loop_fork\*(C'\fR). The invocation is done before the
1766event loop blocks next and before \f(CW\*(C`ev_check\*(C'\fR watchers are being called,
1767and only in the child after the fork. If whoever good citizen calling
1768\&\f(CW\*(C`ev_default_fork\*(C'\fR cheats and calls it in the wrong process, the fork
1769handlers will be invoked, too, of course.
1770.IP "ev_fork_init (ev_signal *, callback)" 4
1771.IX Item "ev_fork_init (ev_signal *, callback)"
1772Initialises and configures the fork watcher \- it has no parameters of any
1773kind. There is a \f(CW\*(C`ev_fork_set\*(C'\fR macro, but using it is utterly pointless,
1774believe me.
905.SH "OTHER FUNCTIONS" 1775.SH "OTHER FUNCTIONS"
906.IX Header "OTHER FUNCTIONS" 1776.IX Header "OTHER FUNCTIONS"
907There are some other functions of possible interest. Described. Here. Now. 1777There are some other functions of possible interest. Described. Here. Now.
908.IP "ev_once (loop, int fd, int events, ev_tstamp timeout, callback)" 4 1778.IP "ev_once (loop, int fd, int events, ev_tstamp timeout, callback)" 4
909.IX Item "ev_once (loop, int fd, int events, ev_tstamp timeout, callback)" 1779.IX Item "ev_once (loop, int fd, int events, ev_tstamp timeout, callback)"
938.Ve 1808.Ve
939.Sp 1809.Sp
940.Vb 1 1810.Vb 1
941\& ev_once (STDIN_FILENO, EV_READ, 10., stdin_ready, 0); 1811\& ev_once (STDIN_FILENO, EV_READ, 10., stdin_ready, 0);
942.Ve 1812.Ve
943.IP "ev_feed_event (loop, watcher, int events)" 4 1813.IP "ev_feed_event (ev_loop *, watcher *, int revents)" 4
944.IX Item "ev_feed_event (loop, watcher, int events)" 1814.IX Item "ev_feed_event (ev_loop *, watcher *, int revents)"
945Feeds the given event set into the event loop, as if the specified event 1815Feeds the given event set into the event loop, as if the specified event
946had happened for the specified watcher (which must be a pointer to an 1816had happened for the specified watcher (which must be a pointer to an
947initialised but not necessarily started event watcher). 1817initialised but not necessarily started event watcher).
948.IP "ev_feed_fd_event (loop, int fd, int revents)" 4 1818.IP "ev_feed_fd_event (ev_loop *, int fd, int revents)" 4
949.IX Item "ev_feed_fd_event (loop, int fd, int revents)" 1819.IX Item "ev_feed_fd_event (ev_loop *, int fd, int revents)"
950Feed an event on the given fd, as if a file descriptor backend detected 1820Feed an event on the given fd, as if a file descriptor backend detected
951the given events it. 1821the given events it.
952.IP "ev_feed_signal_event (loop, int signum)" 4 1822.IP "ev_feed_signal_event (ev_loop *loop, int signum)" 4
953.IX Item "ev_feed_signal_event (loop, int signum)" 1823.IX Item "ev_feed_signal_event (ev_loop *loop, int signum)"
954Feed an event as if the given signal occured (loop must be the default loop!). 1824Feed an event as if the given signal occured (\f(CW\*(C`loop\*(C'\fR must be the default
1825loop!).
955.SH "LIBEVENT EMULATION" 1826.SH "LIBEVENT EMULATION"
956.IX Header "LIBEVENT EMULATION" 1827.IX Header "LIBEVENT EMULATION"
957Libev offers a compatibility emulation layer for libevent. It cannot 1828Libev offers a compatibility emulation layer for libevent. It cannot
958emulate the internals of libevent, so here are some usage hints: 1829emulate the internals of libevent, so here are some usage hints:
959.IP "* Use it by including <event.h>, as usual." 4 1830.IP "* Use it by including <event.h>, as usual." 4
970.IP "* The libev emulation is \fInot\fR \s-1ABI\s0 compatible to libevent, you need to use the libev header file and library." 4 1841.IP "* The libev emulation is \fInot\fR \s-1ABI\s0 compatible to libevent, you need to use the libev header file and library." 4
971.IX Item "The libev emulation is not ABI compatible to libevent, you need to use the libev header file and library." 1842.IX Item "The libev emulation is not ABI compatible to libevent, you need to use the libev header file and library."
972.PD 1843.PD
973.SH "\*(C+ SUPPORT" 1844.SH "\*(C+ SUPPORT"
974.IX Header " SUPPORT" 1845.IX Header " SUPPORT"
975\&\s-1TBD\s0. 1846Libev comes with some simplistic wrapper classes for \*(C+ that mainly allow
1847you to use some convinience methods to start/stop watchers and also change
1848the callback model to a model using method callbacks on objects.
1849.PP
1850To use it,
1851.PP
1852.Vb 1
1853\& #include <ev++.h>
1854.Ve
1855.PP
1856(it is not installed by default). This automatically includes \fIev.h\fR
1857and puts all of its definitions (many of them macros) into the global
1858namespace. All \*(C+ specific things are put into the \f(CW\*(C`ev\*(C'\fR namespace.
1859.PP
1860It should support all the same embedding options as \fIev.h\fR, most notably
1861\&\f(CW\*(C`EV_MULTIPLICITY\*(C'\fR.
1862.PP
1863Here is a list of things available in the \f(CW\*(C`ev\*(C'\fR namespace:
1864.ie n .IP """ev::READ""\fR, \f(CW""ev::WRITE"" etc." 4
1865.el .IP "\f(CWev::READ\fR, \f(CWev::WRITE\fR etc." 4
1866.IX Item "ev::READ, ev::WRITE etc."
1867These are just enum values with the same values as the \f(CW\*(C`EV_READ\*(C'\fR etc.
1868macros from \fIev.h\fR.
1869.ie n .IP """ev::tstamp""\fR, \f(CW""ev::now""" 4
1870.el .IP "\f(CWev::tstamp\fR, \f(CWev::now\fR" 4
1871.IX Item "ev::tstamp, ev::now"
1872Aliases to the same types/functions as with the \f(CW\*(C`ev_\*(C'\fR prefix.
1873.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
1874.el .IP "\f(CWev::io\fR, \f(CWev::timer\fR, \f(CWev::periodic\fR, \f(CWev::idle\fR, \f(CWev::sig\fR etc." 4
1875.IX Item "ev::io, ev::timer, ev::periodic, ev::idle, ev::sig etc."
1876For each \f(CW\*(C`ev_TYPE\*(C'\fR watcher in \fIev.h\fR there is a corresponding class of
1877the same name in the \f(CW\*(C`ev\*(C'\fR namespace, with the exception of \f(CW\*(C`ev_signal\*(C'\fR
1878which is called \f(CW\*(C`ev::sig\*(C'\fR to avoid clashes with the \f(CW\*(C`signal\*(C'\fR macro
1879defines by many implementations.
1880.Sp
1881All of those classes have these methods:
1882.RS 4
1883.IP "ev::TYPE::TYPE (object *, object::method *)" 4
1884.IX Item "ev::TYPE::TYPE (object *, object::method *)"
1885.PD 0
1886.IP "ev::TYPE::TYPE (object *, object::method *, struct ev_loop *)" 4
1887.IX Item "ev::TYPE::TYPE (object *, object::method *, struct ev_loop *)"
1888.IP "ev::TYPE::~TYPE" 4
1889.IX Item "ev::TYPE::~TYPE"
1890.PD
1891The constructor takes a pointer to an object and a method pointer to
1892the event handler callback to call in this class. The constructor calls
1893\&\f(CW\*(C`ev_init\*(C'\fR for you, which means you have to call the \f(CW\*(C`set\*(C'\fR method
1894before starting it. If you do not specify a loop then the constructor
1895automatically associates the default loop with this watcher.
1896.Sp
1897The destructor automatically stops the watcher if it is active.
1898.IP "w\->set (struct ev_loop *)" 4
1899.IX Item "w->set (struct ev_loop *)"
1900Associates a different \f(CW\*(C`struct ev_loop\*(C'\fR with this watcher. You can only
1901do this when the watcher is inactive (and not pending either).
1902.IP "w\->set ([args])" 4
1903.IX Item "w->set ([args])"
1904Basically the same as \f(CW\*(C`ev_TYPE_set\*(C'\fR, with the same args. Must be
1905called at least once. Unlike the C counterpart, an active watcher gets
1906automatically stopped and restarted.
1907.IP "w\->start ()" 4
1908.IX Item "w->start ()"
1909Starts the watcher. Note that there is no \f(CW\*(C`loop\*(C'\fR argument as the
1910constructor already takes the loop.
1911.IP "w\->stop ()" 4
1912.IX Item "w->stop ()"
1913Stops the watcher if it is active. Again, no \f(CW\*(C`loop\*(C'\fR argument.
1914.ie n .IP "w\->again () ""ev::timer""\fR, \f(CW""ev::periodic"" only" 4
1915.el .IP "w\->again () \f(CWev::timer\fR, \f(CWev::periodic\fR only" 4
1916.IX Item "w->again () ev::timer, ev::periodic only"
1917For \f(CW\*(C`ev::timer\*(C'\fR and \f(CW\*(C`ev::periodic\*(C'\fR, this invokes the corresponding
1918\&\f(CW\*(C`ev_TYPE_again\*(C'\fR function.
1919.ie n .IP "w\->sweep () ""ev::embed"" only" 4
1920.el .IP "w\->sweep () \f(CWev::embed\fR only" 4
1921.IX Item "w->sweep () ev::embed only"
1922Invokes \f(CW\*(C`ev_embed_sweep\*(C'\fR.
1923.ie n .IP "w\->update () ""ev::stat"" only" 4
1924.el .IP "w\->update () \f(CWev::stat\fR only" 4
1925.IX Item "w->update () ev::stat only"
1926Invokes \f(CW\*(C`ev_stat_stat\*(C'\fR.
1927.RE
1928.RS 4
1929.RE
1930.PP
1931Example: Define a class with an \s-1IO\s0 and idle watcher, start one of them in
1932the constructor.
1933.PP
1934.Vb 4
1935\& class myclass
1936\& {
1937\& ev_io io; void io_cb (ev::io &w, int revents);
1938\& ev_idle idle void idle_cb (ev::idle &w, int revents);
1939.Ve
1940.PP
1941.Vb 2
1942\& myclass ();
1943\& }
1944.Ve
1945.PP
1946.Vb 6
1947\& myclass::myclass (int fd)
1948\& : io (this, &myclass::io_cb),
1949\& idle (this, &myclass::idle_cb)
1950\& {
1951\& io.start (fd, ev::READ);
1952\& }
1953.Ve
1954.SH "MACRO MAGIC"
1955.IX Header "MACRO MAGIC"
1956Libev can be compiled with a variety of options, the most fundemantal is
1957\&\f(CW\*(C`EV_MULTIPLICITY\*(C'\fR. This option determines wether (most) functions and
1958callbacks have an initial \f(CW\*(C`struct ev_loop *\*(C'\fR argument.
1959.PP
1960To make it easier to write programs that cope with either variant, the
1961following macros are defined:
1962.ie n .IP """EV_A""\fR, \f(CW""EV_A_""" 4
1963.el .IP "\f(CWEV_A\fR, \f(CWEV_A_\fR" 4
1964.IX Item "EV_A, EV_A_"
1965This provides the loop \fIargument\fR for functions, if one is required (\*(L"ev
1966loop argument\*(R"). The \f(CW\*(C`EV_A\*(C'\fR form is used when this is the sole argument,
1967\&\f(CW\*(C`EV_A_\*(C'\fR is used when other arguments are following. Example:
1968.Sp
1969.Vb 3
1970\& ev_unref (EV_A);
1971\& ev_timer_add (EV_A_ watcher);
1972\& ev_loop (EV_A_ 0);
1973.Ve
1974.Sp
1975It assumes the variable \f(CW\*(C`loop\*(C'\fR of type \f(CW\*(C`struct ev_loop *\*(C'\fR is in scope,
1976which is often provided by the following macro.
1977.ie n .IP """EV_P""\fR, \f(CW""EV_P_""" 4
1978.el .IP "\f(CWEV_P\fR, \f(CWEV_P_\fR" 4
1979.IX Item "EV_P, EV_P_"
1980This provides the loop \fIparameter\fR for functions, if one is required (\*(L"ev
1981loop parameter\*(R"). The \f(CW\*(C`EV_P\*(C'\fR form is used when this is the sole parameter,
1982\&\f(CW\*(C`EV_P_\*(C'\fR is used when other parameters are following. Example:
1983.Sp
1984.Vb 2
1985\& // this is how ev_unref is being declared
1986\& static void ev_unref (EV_P);
1987.Ve
1988.Sp
1989.Vb 2
1990\& // this is how you can declare your typical callback
1991\& static void cb (EV_P_ ev_timer *w, int revents)
1992.Ve
1993.Sp
1994It declares a parameter \f(CW\*(C`loop\*(C'\fR of type \f(CW\*(C`struct ev_loop *\*(C'\fR, quite
1995suitable for use with \f(CW\*(C`EV_A\*(C'\fR.
1996.ie n .IP """EV_DEFAULT""\fR, \f(CW""EV_DEFAULT_""" 4
1997.el .IP "\f(CWEV_DEFAULT\fR, \f(CWEV_DEFAULT_\fR" 4
1998.IX Item "EV_DEFAULT, EV_DEFAULT_"
1999Similar to the other two macros, this gives you the value of the default
2000loop, if multiple loops are supported (\*(L"ev loop default\*(R").
2001.PP
2002Example: Declare and initialise a check watcher, working regardless of
2003wether multiple loops are supported or not.
2004.PP
2005.Vb 5
2006\& static void
2007\& check_cb (EV_P_ ev_timer *w, int revents)
2008\& {
2009\& ev_check_stop (EV_A_ w);
2010\& }
2011.Ve
2012.PP
2013.Vb 4
2014\& ev_check check;
2015\& ev_check_init (&check, check_cb);
2016\& ev_check_start (EV_DEFAULT_ &check);
2017\& ev_loop (EV_DEFAULT_ 0);
2018.Ve
2019.SH "EMBEDDING"
2020.IX Header "EMBEDDING"
2021Libev can (and often is) directly embedded into host
2022applications. Examples of applications that embed it include the Deliantra
2023Game Server, the \s-1EV\s0 perl module, the \s-1GNU\s0 Virtual Private Ethernet (gvpe)
2024and rxvt\-unicode.
2025.PP
2026The goal is to enable you to just copy the neecssary files into your
2027source directory without having to change even a single line in them, so
2028you can easily upgrade by simply copying (or having a checked-out copy of
2029libev somewhere in your source tree).
2030.Sh "\s-1FILESETS\s0"
2031.IX Subsection "FILESETS"
2032Depending on what features you need you need to include one or more sets of files
2033in your app.
2034.PP
2035\fI\s-1CORE\s0 \s-1EVENT\s0 \s-1LOOP\s0\fR
2036.IX Subsection "CORE EVENT LOOP"
2037.PP
2038To include only the libev core (all the \f(CW\*(C`ev_*\*(C'\fR functions), with manual
2039configuration (no autoconf):
2040.PP
2041.Vb 2
2042\& #define EV_STANDALONE 1
2043\& #include "ev.c"
2044.Ve
2045.PP
2046This will automatically include \fIev.h\fR, too, and should be done in a
2047single C source file only to provide the function implementations. To use
2048it, do the same for \fIev.h\fR in all files wishing to use this \s-1API\s0 (best
2049done by writing a wrapper around \fIev.h\fR that you can include instead and
2050where you can put other configuration options):
2051.PP
2052.Vb 2
2053\& #define EV_STANDALONE 1
2054\& #include "ev.h"
2055.Ve
2056.PP
2057Both header files and implementation files can be compiled with a \*(C+
2058compiler (at least, thats a stated goal, and breakage will be treated
2059as a bug).
2060.PP
2061You need the following files in your source tree, or in a directory
2062in your include path (e.g. in libev/ when using \-Ilibev):
2063.PP
2064.Vb 4
2065\& ev.h
2066\& ev.c
2067\& ev_vars.h
2068\& ev_wrap.h
2069.Ve
2070.PP
2071.Vb 1
2072\& ev_win32.c required on win32 platforms only
2073.Ve
2074.PP
2075.Vb 5
2076\& ev_select.c only when select backend is enabled (which is by default)
2077\& ev_poll.c only when poll backend is enabled (disabled by default)
2078\& ev_epoll.c only when the epoll backend is enabled (disabled by default)
2079\& ev_kqueue.c only when the kqueue backend is enabled (disabled by default)
2080\& ev_port.c only when the solaris port backend is enabled (disabled by default)
2081.Ve
2082.PP
2083\&\fIev.c\fR includes the backend files directly when enabled, so you only need
2084to compile this single file.
2085.PP
2086\fI\s-1LIBEVENT\s0 \s-1COMPATIBILITY\s0 \s-1API\s0\fR
2087.IX Subsection "LIBEVENT COMPATIBILITY API"
2088.PP
2089To include the libevent compatibility \s-1API\s0, also include:
2090.PP
2091.Vb 1
2092\& #include "event.c"
2093.Ve
2094.PP
2095in the file including \fIev.c\fR, and:
2096.PP
2097.Vb 1
2098\& #include "event.h"
2099.Ve
2100.PP
2101in the files that want to use the libevent \s-1API\s0. This also includes \fIev.h\fR.
2102.PP
2103You need the following additional files for this:
2104.PP
2105.Vb 2
2106\& event.h
2107\& event.c
2108.Ve
2109.PP
2110\fI\s-1AUTOCONF\s0 \s-1SUPPORT\s0\fR
2111.IX Subsection "AUTOCONF SUPPORT"
2112.PP
2113Instead of using \f(CW\*(C`EV_STANDALONE=1\*(C'\fR and providing your config in
2114whatever way you want, you can also \f(CW\*(C`m4_include([libev.m4])\*(C'\fR in your
2115\&\fIconfigure.ac\fR and leave \f(CW\*(C`EV_STANDALONE\*(C'\fR undefined. \fIev.c\fR will then
2116include \fIconfig.h\fR and configure itself accordingly.
2117.PP
2118For this of course you need the m4 file:
2119.PP
2120.Vb 1
2121\& libev.m4
2122.Ve
2123.Sh "\s-1PREPROCESSOR\s0 \s-1SYMBOLS/MACROS\s0"
2124.IX Subsection "PREPROCESSOR SYMBOLS/MACROS"
2125Libev can be configured via a variety of preprocessor symbols you have to define
2126before including any of its files. The default is not to build for multiplicity
2127and only include the select backend.
2128.IP "\s-1EV_STANDALONE\s0" 4
2129.IX Item "EV_STANDALONE"
2130Must always be \f(CW1\fR if you do not use autoconf configuration, which
2131keeps libev from including \fIconfig.h\fR, and it also defines dummy
2132implementations for some libevent functions (such as logging, which is not
2133supported). It will also not define any of the structs usually found in
2134\&\fIevent.h\fR that are not directly supported by the libev core alone.
2135.IP "\s-1EV_USE_MONOTONIC\s0" 4
2136.IX Item "EV_USE_MONOTONIC"
2137If defined to be \f(CW1\fR, libev will try to detect the availability of the
2138monotonic clock option at both compiletime and runtime. Otherwise no use
2139of the monotonic clock option will be attempted. If you enable this, you
2140usually have to link against librt or something similar. Enabling it when
2141the functionality isn't available is safe, though, althoguh you have
2142to make sure you link against any libraries where the \f(CW\*(C`clock_gettime\*(C'\fR
2143function is hiding in (often \fI\-lrt\fR).
2144.IP "\s-1EV_USE_REALTIME\s0" 4
2145.IX Item "EV_USE_REALTIME"
2146If defined to be \f(CW1\fR, libev will try to detect the availability of the
2147realtime clock option at compiletime (and assume its availability at
2148runtime if successful). Otherwise no use of the realtime clock option will
2149be attempted. This effectively replaces \f(CW\*(C`gettimeofday\*(C'\fR by \f(CW\*(C`clock_get
2150(CLOCK_REALTIME, ...)\*(C'\fR and will not normally affect correctness. See tzhe note about libraries
2151in the description of \f(CW\*(C`EV_USE_MONOTONIC\*(C'\fR, though.
2152.IP "\s-1EV_USE_SELECT\s0" 4
2153.IX Item "EV_USE_SELECT"
2154If undefined or defined to be \f(CW1\fR, libev will compile in support for the
2155\&\f(CW\*(C`select\*(C'\fR(2) backend. No attempt at autodetection will be done: if no
2156other method takes over, select will be it. Otherwise the select backend
2157will not be compiled in.
2158.IP "\s-1EV_SELECT_USE_FD_SET\s0" 4
2159.IX Item "EV_SELECT_USE_FD_SET"
2160If defined to \f(CW1\fR, then the select backend will use the system \f(CW\*(C`fd_set\*(C'\fR
2161structure. This is useful if libev doesn't compile due to a missing
2162\&\f(CW\*(C`NFDBITS\*(C'\fR or \f(CW\*(C`fd_mask\*(C'\fR definition or it misguesses the bitset layout on
2163exotic systems. This usually limits the range of file descriptors to some
2164low limit such as 1024 or might have other limitations (winsocket only
2165allows 64 sockets). The \f(CW\*(C`FD_SETSIZE\*(C'\fR macro, set before compilation, might
2166influence the size of the \f(CW\*(C`fd_set\*(C'\fR used.
2167.IP "\s-1EV_SELECT_IS_WINSOCKET\s0" 4
2168.IX Item "EV_SELECT_IS_WINSOCKET"
2169When defined to \f(CW1\fR, the select backend will assume that
2170select/socket/connect etc. don't understand file descriptors but
2171wants osf handles on win32 (this is the case when the select to
2172be used is the winsock select). This means that it will call
2173\&\f(CW\*(C`_get_osfhandle\*(C'\fR on the fd to convert it to an \s-1OS\s0 handle. Otherwise,
2174it is assumed that all these functions actually work on fds, even
2175on win32. Should not be defined on non\-win32 platforms.
2176.IP "\s-1EV_USE_POLL\s0" 4
2177.IX Item "EV_USE_POLL"
2178If defined to be \f(CW1\fR, libev will compile in support for the \f(CW\*(C`poll\*(C'\fR(2)
2179backend. Otherwise it will be enabled on non\-win32 platforms. It
2180takes precedence over select.
2181.IP "\s-1EV_USE_EPOLL\s0" 4
2182.IX Item "EV_USE_EPOLL"
2183If defined to be \f(CW1\fR, libev will compile in support for the Linux
2184\&\f(CW\*(C`epoll\*(C'\fR(7) backend. Its availability will be detected at runtime,
2185otherwise another method will be used as fallback. This is the
2186preferred backend for GNU/Linux systems.
2187.IP "\s-1EV_USE_KQUEUE\s0" 4
2188.IX Item "EV_USE_KQUEUE"
2189If defined to be \f(CW1\fR, libev will compile in support for the \s-1BSD\s0 style
2190\&\f(CW\*(C`kqueue\*(C'\fR(2) backend. Its actual availability will be detected at runtime,
2191otherwise another method will be used as fallback. This is the preferred
2192backend for \s-1BSD\s0 and BSD-like systems, although on most BSDs kqueue only
2193supports some types of fds correctly (the only platform we found that
2194supports ptys for example was NetBSD), so kqueue might be compiled in, but
2195not be used unless explicitly requested. The best way to use it is to find
2196out whether kqueue supports your type of fd properly and use an embedded
2197kqueue loop.
2198.IP "\s-1EV_USE_PORT\s0" 4
2199.IX Item "EV_USE_PORT"
2200If defined to be \f(CW1\fR, libev will compile in support for the Solaris
220110 port style backend. Its availability will be detected at runtime,
2202otherwise another method will be used as fallback. This is the preferred
2203backend for Solaris 10 systems.
2204.IP "\s-1EV_USE_DEVPOLL\s0" 4
2205.IX Item "EV_USE_DEVPOLL"
2206reserved for future expansion, works like the \s-1USE\s0 symbols above.
2207.IP "\s-1EV_USE_INOTIFY\s0" 4
2208.IX Item "EV_USE_INOTIFY"
2209If defined to be \f(CW1\fR, libev will compile in support for the Linux inotify
2210interface to speed up \f(CW\*(C`ev_stat\*(C'\fR watchers. Its actual availability will
2211be detected at runtime.
2212.IP "\s-1EV_H\s0" 4
2213.IX Item "EV_H"
2214The name of the \fIev.h\fR header file used to include it. The default if
2215undefined is \f(CW\*(C`<ev.h>\*(C'\fR in \fIevent.h\fR and \f(CW"ev.h"\fR in \fIev.c\fR. This
2216can be used to virtually rename the \fIev.h\fR header file in case of conflicts.
2217.IP "\s-1EV_CONFIG_H\s0" 4
2218.IX Item "EV_CONFIG_H"
2219If \f(CW\*(C`EV_STANDALONE\*(C'\fR isn't \f(CW1\fR, this variable can be used to override
2220\&\fIev.c\fR's idea of where to find the \fIconfig.h\fR file, similarly to
2221\&\f(CW\*(C`EV_H\*(C'\fR, above.
2222.IP "\s-1EV_EVENT_H\s0" 4
2223.IX Item "EV_EVENT_H"
2224Similarly to \f(CW\*(C`EV_H\*(C'\fR, this macro can be used to override \fIevent.c\fR's idea
2225of how the \fIevent.h\fR header can be found.
2226.IP "\s-1EV_PROTOTYPES\s0" 4
2227.IX Item "EV_PROTOTYPES"
2228If defined to be \f(CW0\fR, then \fIev.h\fR will not define any function
2229prototypes, but still define all the structs and other symbols. This is
2230occasionally useful if you want to provide your own wrapper functions
2231around libev functions.
2232.IP "\s-1EV_MULTIPLICITY\s0" 4
2233.IX Item "EV_MULTIPLICITY"
2234If undefined or defined to \f(CW1\fR, then all event-loop-specific functions
2235will have the \f(CW\*(C`struct ev_loop *\*(C'\fR as first argument, and you can create
2236additional independent event loops. Otherwise there will be no support
2237for multiple event loops and there is no first event loop pointer
2238argument. Instead, all functions act on the single default loop.
2239.IP "\s-1EV_PERIODIC_ENABLE\s0" 4
2240.IX Item "EV_PERIODIC_ENABLE"
2241If undefined or defined to be \f(CW1\fR, then periodic timers are supported. If
2242defined to be \f(CW0\fR, then they are not. Disabling them saves a few kB of
2243code.
2244.IP "\s-1EV_EMBED_ENABLE\s0" 4
2245.IX Item "EV_EMBED_ENABLE"
2246If undefined or defined to be \f(CW1\fR, then embed watchers are supported. If
2247defined to be \f(CW0\fR, then they are not.
2248.IP "\s-1EV_STAT_ENABLE\s0" 4
2249.IX Item "EV_STAT_ENABLE"
2250If undefined or defined to be \f(CW1\fR, then stat watchers are supported. If
2251defined to be \f(CW0\fR, then they are not.
2252.IP "\s-1EV_FORK_ENABLE\s0" 4
2253.IX Item "EV_FORK_ENABLE"
2254If undefined or defined to be \f(CW1\fR, then fork watchers are supported. If
2255defined to be \f(CW0\fR, then they are not.
2256.IP "\s-1EV_MINIMAL\s0" 4
2257.IX Item "EV_MINIMAL"
2258If you need to shave off some kilobytes of code at the expense of some
2259speed, define this symbol to \f(CW1\fR. Currently only used for gcc to override
2260some inlining decisions, saves roughly 30% codesize of amd64.
2261.IP "\s-1EV_PID_HASHSIZE\s0" 4
2262.IX Item "EV_PID_HASHSIZE"
2263\&\f(CW\*(C`ev_child\*(C'\fR watchers use a small hash table to distribute workload by
2264pid. The default size is \f(CW16\fR (or \f(CW1\fR with \f(CW\*(C`EV_MINIMAL\*(C'\fR), usually more
2265than enough. If you need to manage thousands of children you might want to
2266increase this value (\fImust\fR be a power of two).
2267.IP "\s-1EV_INOTIFY_HASHSIZE\s0" 4
2268.IX Item "EV_INOTIFY_HASHSIZE"
2269\&\f(CW\*(C`ev_staz\*(C'\fR watchers use a small hash table to distribute workload by
2270inotify watch id. The default size is \f(CW16\fR (or \f(CW1\fR with \f(CW\*(C`EV_MINIMAL\*(C'\fR),
2271usually more than enough. If you need to manage thousands of \f(CW\*(C`ev_stat\*(C'\fR
2272watchers you might want to increase this value (\fImust\fR be a power of
2273two).
2274.IP "\s-1EV_COMMON\s0" 4
2275.IX Item "EV_COMMON"
2276By default, all watchers have a \f(CW\*(C`void *data\*(C'\fR member. By redefining
2277this macro to a something else you can include more and other types of
2278members. You have to define it each time you include one of the files,
2279though, and it must be identical each time.
2280.Sp
2281For example, the perl \s-1EV\s0 module uses something like this:
2282.Sp
2283.Vb 3
2284\& #define EV_COMMON \e
2285\& SV *self; /* contains this struct */ \e
2286\& SV *cb_sv, *fh /* note no trailing ";" */
2287.Ve
2288.IP "\s-1EV_CB_DECLARE\s0 (type)" 4
2289.IX Item "EV_CB_DECLARE (type)"
2290.PD 0
2291.IP "\s-1EV_CB_INVOKE\s0 (watcher, revents)" 4
2292.IX Item "EV_CB_INVOKE (watcher, revents)"
2293.IP "ev_set_cb (ev, cb)" 4
2294.IX Item "ev_set_cb (ev, cb)"
2295.PD
2296Can be used to change the callback member declaration in each watcher,
2297and the way callbacks are invoked and set. Must expand to a struct member
2298definition and a statement, respectively. See the \fIev.v\fR header file for
2299their default definitions. One possible use for overriding these is to
2300avoid the \f(CW\*(C`struct ev_loop *\*(C'\fR as first argument in all cases, or to use
2301method calls instead of plain function calls in \*(C+.
2302.Sh "\s-1EXAMPLES\s0"
2303.IX Subsection "EXAMPLES"
2304For a real-world example of a program the includes libev
2305verbatim, you can have a look at the \s-1EV\s0 perl module
2306(<http://software.schmorp.de/pkg/EV.html>). It has the libev files in
2307the \fIlibev/\fR subdirectory and includes them in the \fI\s-1EV/EVAPI\s0.h\fR (public
2308interface) and \fI\s-1EV\s0.xs\fR (implementation) files. Only the \fI\s-1EV\s0.xs\fR file
2309will be compiled. It is pretty complex because it provides its own header
2310file.
2311.Sp
2312The usage in rxvt-unicode is simpler. It has a \fIev_cpp.h\fR header file
2313that everybody includes and which overrides some autoconf choices:
2314.Sp
2315.Vb 4
2316\& #define EV_USE_POLL 0
2317\& #define EV_MULTIPLICITY 0
2318\& #define EV_PERIODICS 0
2319\& #define EV_CONFIG_H <config.h>
2320.Ve
2321.Sp
2322.Vb 1
2323\& #include "ev++.h"
2324.Ve
2325.Sp
2326And a \fIev_cpp.C\fR implementation file that contains libev proper and is compiled:
2327.Sp
2328.Vb 2
2329\& #include "ev_cpp.h"
2330\& #include "ev.c"
2331.Ve
2332.SH "COMPLEXITIES"
2333.IX Header "COMPLEXITIES"
2334In this section the complexities of (many of) the algorithms used inside
2335libev will be explained. For complexity discussions about backends see the
2336documentation for \f(CW\*(C`ev_default_init\*(C'\fR.
2337.RS 4
2338.IP "Starting and stopping timer/periodic watchers: O(log skipped_other_timers)" 4
2339.IX Item "Starting and stopping timer/periodic watchers: O(log skipped_other_timers)"
2340.PD 0
2341.IP "Changing timer/periodic watchers (by autorepeat, again): O(log skipped_other_timers)" 4
2342.IX Item "Changing timer/periodic watchers (by autorepeat, again): O(log skipped_other_timers)"
2343.IP "Starting io/check/prepare/idle/signal/child watchers: O(1)" 4
2344.IX Item "Starting io/check/prepare/idle/signal/child watchers: O(1)"
2345.IP "Stopping check/prepare/idle watchers: O(1)" 4
2346.IX Item "Stopping check/prepare/idle watchers: O(1)"
2347.IP "Stopping an io/signal/child watcher: O(number_of_watchers_for_this_(fd/signal/pid % \s-1EV_PID_HASHSIZE\s0))" 4
2348.IX Item "Stopping an io/signal/child watcher: O(number_of_watchers_for_this_(fd/signal/pid % EV_PID_HASHSIZE))"
2349.IP "Finding the next timer per loop iteration: O(1)" 4
2350.IX Item "Finding the next timer per loop iteration: O(1)"
2351.IP "Each change on a file descriptor per loop iteration: O(number_of_watchers_for_this_fd)" 4
2352.IX Item "Each change on a file descriptor per loop iteration: O(number_of_watchers_for_this_fd)"
2353.IP "Activating one watcher: O(1)" 4
2354.IX Item "Activating one watcher: O(1)"
2355.RE
2356.RS 4
2357.PD
976.SH "AUTHOR" 2358.SH "AUTHOR"
977.IX Header "AUTHOR" 2359.IX Header "AUTHOR"
978Marc Lehmann <libev@schmorp.de>. 2360Marc Lehmann <libev@schmorp.de>.

Diff Legend

Removed lines
+ Added lines
< Changed lines
> Changed lines