… | |
… | |
82 | |
82 | |
83 | =head1 WHAT TO READ WHEN IN A HURRY |
83 | =head1 WHAT TO READ WHEN IN A HURRY |
84 | |
84 | |
85 | This manual tries to be very detailed, but unfortunately, this also makes |
85 | This manual tries to be very detailed, but unfortunately, this also makes |
86 | it very long. If you just want to know the basics of libev, I suggest |
86 | it very long. If you just want to know the basics of libev, I suggest |
87 | reading L</ANATOMY OF A WATCHER>, then the L<E/XAMPLE PROGRAM> above and |
87 | reading L</ANATOMY OF A WATCHER>, then the L</EXAMPLE PROGRAM> above and |
88 | look up the missing functions in L</GLOBAL FUNCTIONS> and the C<ev_io> and |
88 | look up the missing functions in L</GLOBAL FUNCTIONS> and the C<ev_io> and |
89 | C<ev_timer> sections in L</WATCHER TYPES>. |
89 | C<ev_timer> sections in L</WATCHER TYPES>. |
90 | |
90 | |
91 | =head1 ABOUT LIBEV |
91 | =head1 ABOUT LIBEV |
92 | |
92 | |
… | |
… | |
569 | kernel is more efficient (which says nothing about its actual speed, of |
569 | kernel is more efficient (which says nothing about its actual speed, of |
570 | course). While stopping, setting and starting an I/O watcher does never |
570 | course). While stopping, setting and starting an I/O watcher does never |
571 | cause an extra system call as with C<EVBACKEND_EPOLL>, it still adds up to |
571 | cause an extra system call as with C<EVBACKEND_EPOLL>, it still adds up to |
572 | two event changes per incident. Support for C<fork ()> is very bad (you |
572 | two event changes per incident. Support for C<fork ()> is very bad (you |
573 | might have to leak fd's on fork, but it's more sane than epoll) and it |
573 | might have to leak fd's on fork, but it's more sane than epoll) and it |
574 | drops fds silently in similarly hard-to-detect cases |
574 | drops fds silently in similarly hard-to-detect cases. |
575 | |
575 | |
576 | This backend usually performs well under most conditions. |
576 | This backend usually performs well under most conditions. |
577 | |
577 | |
578 | While nominally embeddable in other event loops, this doesn't work |
578 | While nominally embeddable in other event loops, this doesn't work |
579 | everywhere, so you might need to test for this. And since it is broken |
579 | everywhere, so you might need to test for this. And since it is broken |
… | |
… | |
1393 | transition between them will be described in more detail - and while these |
1393 | transition between them will be described in more detail - and while these |
1394 | rules might look complicated, they usually do "the right thing". |
1394 | rules might look complicated, they usually do "the right thing". |
1395 | |
1395 | |
1396 | =over 4 |
1396 | =over 4 |
1397 | |
1397 | |
1398 | =item initialiased |
1398 | =item initialised |
1399 | |
1399 | |
1400 | Before a watcher can be registered with the event loop it has to be |
1400 | Before a watcher can be registered with the event loop it has to be |
1401 | initialised. This can be done with a call to C<ev_TYPE_init>, or calls to |
1401 | initialised. This can be done with a call to C<ev_TYPE_init>, or calls to |
1402 | C<ev_init> followed by the watcher-specific C<ev_TYPE_set> function. |
1402 | C<ev_init> followed by the watcher-specific C<ev_TYPE_set> function. |
1403 | |
1403 | |
… | |
… | |
2858 | |
2858 | |
2859 | This mode of operation can be useful together with an C<ev_check> watcher, |
2859 | This mode of operation can be useful together with an C<ev_check> watcher, |
2860 | to do something on each event loop iteration - for example to balance load |
2860 | to do something on each event loop iteration - for example to balance load |
2861 | between different connections. |
2861 | between different connections. |
2862 | |
2862 | |
2863 | See L<< /Abusing an C<ev_check> watcher for its side-effect >> for a longer |
2863 | See L</Abusing an ev_check watcher for its side-effect> for a longer |
2864 | example. |
2864 | example. |
2865 | |
2865 | |
2866 | =head3 Watcher-Specific Functions and Data Members |
2866 | =head3 Watcher-Specific Functions and Data Members |
2867 | |
2867 | |
2868 | =over 4 |
2868 | =over 4 |
… | |
… | |
2961 | connections have a chance of making progress. |
2961 | connections have a chance of making progress. |
2962 | |
2962 | |
2963 | Using an C<ev_check> watcher is almost enough: it will be called on the |
2963 | Using an C<ev_check> watcher is almost enough: it will be called on the |
2964 | next event loop iteration. However, that isn't as soon as possible - |
2964 | next event loop iteration. However, that isn't as soon as possible - |
2965 | without external events, your C<ev_check> watcher will not be invoked. |
2965 | without external events, your C<ev_check> watcher will not be invoked. |
2966 | |
|
|
2967 | |
2966 | |
2968 | This is where C<ev_idle> watchers come in handy - all you need is a |
2967 | This is where C<ev_idle> watchers come in handy - all you need is a |
2969 | single global idle watcher that is active as long as you have one active |
2968 | single global idle watcher that is active as long as you have one active |
2970 | C<ev_check> watcher. The C<ev_idle> watcher makes sure the event loop |
2969 | C<ev_check> watcher. The C<ev_idle> watcher makes sure the event loop |
2971 | will not sleep, and the C<ev_check> watcher makes sure a callback gets |
2970 | will not sleep, and the C<ev_check> watcher makes sure a callback gets |
… | |
… | |
3251 | |
3250 | |
3252 | =head2 C<ev_fork> - the audacity to resume the event loop after a fork |
3251 | =head2 C<ev_fork> - the audacity to resume the event loop after a fork |
3253 | |
3252 | |
3254 | Fork watchers are called when a C<fork ()> was detected (usually because |
3253 | Fork watchers are called when a C<fork ()> was detected (usually because |
3255 | whoever is a good citizen cared to tell libev about it by calling |
3254 | whoever is a good citizen cared to tell libev about it by calling |
3256 | C<ev_default_fork> or C<ev_loop_fork>). The invocation is done before the |
3255 | C<ev_loop_fork>). The invocation is done before the event loop blocks next |
3257 | event loop blocks next and before C<ev_check> watchers are being called, |
3256 | and before C<ev_check> watchers are being called, and only in the child |
3258 | and only in the child after the fork. If whoever good citizen calling |
3257 | after the fork. If whoever good citizen calling C<ev_default_fork> cheats |
3259 | C<ev_default_fork> cheats and calls it in the wrong process, the fork |
3258 | and calls it in the wrong process, the fork handlers will be invoked, too, |
3260 | handlers will be invoked, too, of course. |
3259 | of course. |
3261 | |
3260 | |
3262 | =head3 The special problem of life after fork - how is it possible? |
3261 | =head3 The special problem of life after fork - how is it possible? |
3263 | |
3262 | |
3264 | Most uses of C<fork()> consist of forking, then some simple calls to set |
3263 | Most uses of C<fork()> consist of forking, then some simple calls to set |
3265 | up/change the process environment, followed by a call to C<exec()>. This |
3264 | up/change the process environment, followed by a call to C<exec()>. This |
… | |
… | |
3659 | already been invoked. |
3658 | already been invoked. |
3660 | |
3659 | |
3661 | A common way around all these issues is to make sure that |
3660 | A common way around all these issues is to make sure that |
3662 | C<start_new_request> I<always> returns before the callback is invoked. If |
3661 | C<start_new_request> I<always> returns before the callback is invoked. If |
3663 | C<start_new_request> immediately knows the result, it can artificially |
3662 | C<start_new_request> immediately knows the result, it can artificially |
3664 | delay invoking the callback by e.g. using a C<prepare> or C<idle> watcher |
3663 | delay invoking the callback by using a C<prepare> or C<idle> watcher for |
3665 | for example, or more sneakily, by reusing an existing (stopped) watcher |
3664 | example, or more sneakily, by reusing an existing (stopped) watcher and |
3666 | and pushing it into the pending queue: |
3665 | pushing it into the pending queue: |
3667 | |
3666 | |
3668 | ev_set_cb (watcher, callback); |
3667 | ev_set_cb (watcher, callback); |
3669 | ev_feed_event (EV_A_ watcher, 0); |
3668 | ev_feed_event (EV_A_ watcher, 0); |
3670 | |
3669 | |
3671 | This way, C<start_new_request> can safely return before the callback is |
3670 | This way, C<start_new_request> can safely return before the callback is |
… | |
… | |
3679 | |
3678 | |
3680 | This brings the problem of exiting - a callback might want to finish the |
3679 | This brings the problem of exiting - a callback might want to finish the |
3681 | main C<ev_run> call, but not the nested one (e.g. user clicked "Quit", but |
3680 | main C<ev_run> call, but not the nested one (e.g. user clicked "Quit", but |
3682 | a modal "Are you sure?" dialog is still waiting), or just the nested one |
3681 | a modal "Are you sure?" dialog is still waiting), or just the nested one |
3683 | and not the main one (e.g. user clocked "Ok" in a modal dialog), or some |
3682 | and not the main one (e.g. user clocked "Ok" in a modal dialog), or some |
3684 | other combination: In these cases, C<ev_break> will not work alone. |
3683 | other combination: In these cases, a simple C<ev_break> will not work. |
3685 | |
3684 | |
3686 | The solution is to maintain "break this loop" variable for each C<ev_run> |
3685 | The solution is to maintain "break this loop" variable for each C<ev_run> |
3687 | invocation, and use a loop around C<ev_run> until the condition is |
3686 | invocation, and use a loop around C<ev_run> until the condition is |
3688 | triggered, using C<EVRUN_ONCE>: |
3687 | triggered, using C<EVRUN_ONCE>: |
3689 | |
3688 | |
… | |
… | |
3950 | libev sources can be compiled as C++. Therefore, code that uses the C API |
3949 | libev sources can be compiled as C++. Therefore, code that uses the C API |
3951 | will work fine. |
3950 | will work fine. |
3952 | |
3951 | |
3953 | Proper exception specifications might have to be added to callbacks passed |
3952 | Proper exception specifications might have to be added to callbacks passed |
3954 | to libev: exceptions may be thrown only from watcher callbacks, all |
3953 | to libev: exceptions may be thrown only from watcher callbacks, all |
3955 | other callbacks (allocator, syserr, loop acquire/release and periodioc |
3954 | other callbacks (allocator, syserr, loop acquire/release and periodic |
3956 | reschedule callbacks) must not throw exceptions, and might need a C<throw |
3955 | reschedule callbacks) must not throw exceptions, and might need a C<throw |
3957 | ()> specification. If you have code that needs to be compiled as both C |
3956 | ()> specification. If you have code that needs to be compiled as both C |
3958 | and C++ you can use the C<EV_THROW> macro for this: |
3957 | and C++ you can use the C<EV_THROW> macro for this: |
3959 | |
3958 | |
3960 | static void |
3959 | static void |
… | |
… | |
4121 | Associates a different C<struct ev_loop> with this watcher. You can only |
4120 | Associates a different C<struct ev_loop> with this watcher. You can only |
4122 | do this when the watcher is inactive (and not pending either). |
4121 | do this when the watcher is inactive (and not pending either). |
4123 | |
4122 | |
4124 | =item w->set ([arguments]) |
4123 | =item w->set ([arguments]) |
4125 | |
4124 | |
4126 | Basically the same as C<ev_TYPE_set>, with the same arguments. Either this |
4125 | Basically the same as C<ev_TYPE_set> (except for C<ev::embed> watchers>), |
4127 | method or a suitable start method must be called at least once. Unlike the |
4126 | with the same arguments. Either this method or a suitable start method |
4128 | C counterpart, an active watcher gets automatically stopped and restarted |
4127 | must be called at least once. Unlike the C counterpart, an active watcher |
4129 | when reconfiguring it with this method. |
4128 | gets automatically stopped and restarted when reconfiguring it with this |
|
|
4129 | method. |
|
|
4130 | |
|
|
4131 | For C<ev::embed> watchers this method is called C<set_embed>, to avoid |
|
|
4132 | clashing with the C<set (loop)> method. |
4130 | |
4133 | |
4131 | =item w->start () |
4134 | =item w->start () |
4132 | |
4135 | |
4133 | Starts the watcher. Note that there is no C<loop> argument, as the |
4136 | Starts the watcher. Note that there is no C<loop> argument, as the |
4134 | constructor already stores the event loop. |
4137 | constructor already stores the event loop. |
… | |
… | |
4237 | =item Lua |
4240 | =item Lua |
4238 | |
4241 | |
4239 | Brian Maher has written a partial interface to libev for lua (at the |
4242 | Brian Maher has written a partial interface to libev for lua (at the |
4240 | time of this writing, only C<ev_io> and C<ev_timer>), to be found at |
4243 | time of this writing, only C<ev_io> and C<ev_timer>), to be found at |
4241 | L<http://github.com/brimworks/lua-ev>. |
4244 | L<http://github.com/brimworks/lua-ev>. |
|
|
4245 | |
|
|
4246 | =item Javascript |
|
|
4247 | |
|
|
4248 | Node.js (L<http://nodejs.org>) uses libev as the underlying event library. |
|
|
4249 | |
|
|
4250 | =item Others |
|
|
4251 | |
|
|
4252 | There are others, and I stopped counting. |
4242 | |
4253 | |
4243 | =back |
4254 | =back |
4244 | |
4255 | |
4245 | |
4256 | |
4246 | =head1 MACRO MAGIC |
4257 | =head1 MACRO MAGIC |
… | |
… | |
4545 | |
4556 | |
4546 | If programs implement their own fd to handle mapping on win32, then this |
4557 | If programs implement their own fd to handle mapping on win32, then this |
4547 | macro can be used to override the C<close> function, useful to unregister |
4558 | macro can be used to override the C<close> function, useful to unregister |
4548 | file descriptors again. Note that the replacement function has to close |
4559 | file descriptors again. Note that the replacement function has to close |
4549 | the underlying OS handle. |
4560 | the underlying OS handle. |
|
|
4561 | |
|
|
4562 | =item EV_USE_WSASOCKET |
|
|
4563 | |
|
|
4564 | If defined to be C<1>, libev will use C<WSASocket> to create its internal |
|
|
4565 | communication socket, which works better in some environments. Otherwise, |
|
|
4566 | the normal C<socket> function will be used, which works better in other |
|
|
4567 | environments. |
4550 | |
4568 | |
4551 | =item EV_USE_POLL |
4569 | =item EV_USE_POLL |
4552 | |
4570 | |
4553 | If defined to be C<1>, libev will compile in support for the C<poll>(2) |
4571 | If defined to be C<1>, libev will compile in support for the C<poll>(2) |
4554 | backend. Otherwise it will be enabled on non-win32 platforms. It |
4572 | backend. Otherwise it will be enabled on non-win32 platforms. It |
… | |
… | |
4606 | above. This reduces dependencies and makes libev faster. |
4624 | above. This reduces dependencies and makes libev faster. |
4607 | |
4625 | |
4608 | =item EV_ATOMIC_T |
4626 | =item EV_ATOMIC_T |
4609 | |
4627 | |
4610 | Libev requires an integer type (suitable for storing C<0> or C<1>) whose |
4628 | Libev requires an integer type (suitable for storing C<0> or C<1>) whose |
4611 | access is atomic and serialised with respect to other threads or signal |
4629 | access is atomic with respect to other threads or signal contexts. No |
4612 | contexts. No such type is easily found in the C language, so you can |
4630 | such type is easily found in the C language, so you can provide your own |
4613 | provide your own type that you know is safe for your purposes. It is used |
4631 | type that you know is safe for your purposes. It is used both for signal |
4614 | both for signal handler "locking" as well as for signal and thread safety |
4632 | handler "locking" as well as for signal and thread safety in C<ev_async> |
4615 | in C<ev_async> watchers. |
4633 | watchers. |
4616 | |
4634 | |
4617 | In the absence of this define, libev will use C<sig_atomic_t volatile> |
4635 | In the absence of this define, libev will use C<sig_atomic_t volatile> |
4618 | (from F<signal.h>), which is usually good enough on most platforms, |
4636 | (from F<signal.h>), which is usually good enough on most platforms. |
4619 | although strictly speaking using a type that also implies a memory fence |
|
|
4620 | is required. |
|
|
4621 | |
4637 | |
4622 | =item EV_H (h) |
4638 | =item EV_H (h) |
4623 | |
4639 | |
4624 | The name of the F<ev.h> header file used to include it. The default if |
4640 | The name of the F<ev.h> header file used to include it. The default if |
4625 | undefined is C<"ev.h"> in F<event.h>, F<ev.c> and F<ev++.h>. This can be |
4641 | undefined is C<"ev.h"> in F<event.h>, F<ev.c> and F<ev++.h>. This can be |
… | |
… | |
5294 | thread" or will block signals process-wide, both behaviours would |
5310 | thread" or will block signals process-wide, both behaviours would |
5295 | be compatible with libev. Interaction between C<sigprocmask> and |
5311 | be compatible with libev. Interaction between C<sigprocmask> and |
5296 | C<pthread_sigmask> could complicate things, however. |
5312 | C<pthread_sigmask> could complicate things, however. |
5297 | |
5313 | |
5298 | The most portable way to handle signals is to block signals in all threads |
5314 | The most portable way to handle signals is to block signals in all threads |
5299 | except the initial one, and run the default loop in the initial thread as |
5315 | except the initial one, and run the signal handling loop in the initial |
5300 | well. |
5316 | thread as well. |
5301 | |
5317 | |
5302 | =item C<long> must be large enough for common memory allocation sizes |
5318 | =item C<long> must be large enough for common memory allocation sizes |
5303 | |
5319 | |
5304 | To improve portability and simplify its API, libev uses C<long> internally |
5320 | To improve portability and simplify its API, libev uses C<long> internally |
5305 | instead of C<size_t> when allocating its data structures. On non-POSIX |
5321 | instead of C<size_t> when allocating its data structures. On non-POSIX |