… | |
… | |
159 | When libev detects a usage error such as a negative timer interval, then |
159 | When libev detects a usage error such as a negative timer interval, then |
160 | it will print a diagnostic message and abort (via the C<assert> mechanism, |
160 | it will print a diagnostic message and abort (via the C<assert> mechanism, |
161 | so C<NDEBUG> will disable this checking): these are programming errors in |
161 | so C<NDEBUG> will disable this checking): these are programming errors in |
162 | the libev caller and need to be fixed there. |
162 | the libev caller and need to be fixed there. |
163 | |
163 | |
|
|
164 | Via the C<EV_FREQUENT> macro you can compile in and/or enable extensive |
|
|
165 | consistency checking code inside libev that can be used to check for |
|
|
166 | internal inconsistencies, suually caused by application bugs. |
|
|
167 | |
164 | Libev also has a few internal error-checking C<assert>ions, and also has |
168 | Libev also has a few internal error-checking C<assert>ions. These do not |
165 | extensive consistency checking code. These do not trigger under normal |
|
|
166 | circumstances, as they indicate either a bug in libev or worse. |
169 | trigger under normal circumstances, as they indicate either a bug in libev |
|
|
170 | or worse. |
167 | |
171 | |
168 | |
172 | |
169 | =head1 GLOBAL FUNCTIONS |
173 | =head1 GLOBAL FUNCTIONS |
170 | |
174 | |
171 | These functions can be called anytime, even before initialising the |
175 | These functions can be called anytime, even before initialising the |
… | |
… | |
511 | This backend maps C<EV_READ> to C<POLLIN | POLLERR | POLLHUP>, and |
515 | This backend maps C<EV_READ> to C<POLLIN | POLLERR | POLLHUP>, and |
512 | C<EV_WRITE> to C<POLLOUT | POLLERR | POLLHUP>. |
516 | C<EV_WRITE> to C<POLLOUT | POLLERR | POLLHUP>. |
513 | |
517 | |
514 | =item C<EVBACKEND_EPOLL> (value 4, Linux) |
518 | =item C<EVBACKEND_EPOLL> (value 4, Linux) |
515 | |
519 | |
516 | Use the linux-specific epoll(7) interface (for both pre- and post-2.6.9 |
520 | Use the Linux-specific epoll(7) interface (for both pre- and post-2.6.9 |
517 | kernels). |
521 | kernels). |
518 | |
522 | |
519 | For few fds, this backend is a bit little slower than poll and select, but |
523 | For few fds, this backend is a bit little slower than poll and select, but |
520 | it scales phenomenally better. While poll and select usually scale like |
524 | it scales phenomenally better. While poll and select usually scale like |
521 | O(total_fds) where total_fds is the total number of fds (or the highest |
525 | O(total_fds) where total_fds is the total number of fds (or the highest |
… | |
… | |
574 | This backend maps C<EV_READ> and C<EV_WRITE> in the same way as |
578 | This backend maps C<EV_READ> and C<EV_WRITE> in the same way as |
575 | C<EVBACKEND_POLL>. |
579 | C<EVBACKEND_POLL>. |
576 | |
580 | |
577 | =item C<EVBACKEND_LINUXAIO> (value 64, Linux) |
581 | =item C<EVBACKEND_LINUXAIO> (value 64, Linux) |
578 | |
582 | |
579 | Use the linux-specific linux aio (I<not> C<< aio(7) >> but C<< |
583 | Use the Linux-specific Linux AIO (I<not> C<< aio(7) >> but C<< |
580 | io_submit(2) >>) event interface available in post-4.18 kernels. |
584 | io_submit(2) >>) event interface available in post-4.18 kernels (but libev |
|
|
585 | only tries to use it in 4.19+). |
|
|
586 | |
|
|
587 | This is another Linux train wreck of an event interface. |
581 | |
588 | |
582 | If this backend works for you (as of this writing, it was very |
589 | If this backend works for you (as of this writing, it was very |
583 | experimental), it is the best event interface available on linux and might |
590 | experimental), it is the best event interface available on Linux and might |
584 | be well worth enabling it - if it isn't available in your kernel this will |
591 | be well worth enabling it - if it isn't available in your kernel this will |
585 | be detected and this backend will be skipped. |
592 | be detected and this backend will be skipped. |
586 | |
593 | |
587 | This backend can batch oneshot requests and supports a user-space ring |
594 | This backend can batch oneshot requests and supports a user-space ring |
588 | buffer to receive events. It also doesn't suffer from most of the design |
595 | buffer to receive events. It also doesn't suffer from most of the design |
589 | problems of epoll (such as not being able to remove event sources from |
596 | problems of epoll (such as not being able to remove event sources from |
590 | the epoll set), and generally sounds too good to be true. Because, this |
597 | the epoll set), and generally sounds too good to be true. Because, this |
591 | being the linux kernel, of course it suffers from a whole new set of |
598 | being the Linux kernel, of course it suffers from a whole new set of |
592 | limitations. |
599 | limitations, forcing you to fall back to epoll, inheriting all its design |
|
|
600 | issues. |
593 | |
601 | |
594 | For one, it is not easily embeddable (but probably could be done using |
602 | For one, it is not easily embeddable (but probably could be done using |
595 | an event fd at some extra overhead). It also is subject to a system wide |
603 | an event fd at some extra overhead). It also is subject to a system wide |
596 | limit that can be configured in F</proc/sys/fs/aio-max-nr> - each loop |
604 | limit that can be configured in F</proc/sys/fs/aio-max-nr>. If no AIO |
597 | currently requires C<61> of this number. If no aio requests are left, this |
|
|
598 | backend will be skipped during initialisation. |
605 | requests are left, this backend will be skipped during initialisation, and |
|
|
606 | will switch to epoll when the loop is active. |
599 | |
607 | |
600 | Most problematic in practise, however, is that not all file descriptors |
608 | Most problematic in practice, however, is that not all file descriptors |
601 | work with it. For example, in linux 5.1, tcp sockets, pipes, event fds, |
609 | work with it. For example, in Linux 5.1, TCP sockets, pipes, event fds, |
602 | files, F</dev/null> and a few others are supported, but ttys do not work |
610 | files, F</dev/null> and many others are supported, but ttys do not work |
603 | properly (a known bug that the kernel developers don't care about, see |
611 | properly (a known bug that the kernel developers don't care about, see |
604 | L<https://lore.kernel.org/patchwork/patch/1047453/>), so this is not |
612 | L<https://lore.kernel.org/patchwork/patch/1047453/>), so this is not |
605 | (yet?) a generic event polling interface. |
613 | (yet?) a generic event polling interface. |
606 | |
614 | |
|
|
615 | Overall, it seems the Linux developers just don't want it to have a |
|
|
616 | generic event handling mechanism other than C<select> or C<poll>. |
|
|
617 | |
607 | To work around this latter problem, the current version of libev uses |
618 | To work around all these problem, the current version of libev uses its |
608 | epoll as a fallback for file deescriptor types that do not work. Epoll |
619 | epoll backend as a fallback for file descriptor types that do not work. Or |
609 | is used in, kind of, slow mode that hopefully avoids most of its design |
620 | falls back completely to epoll if the kernel acts up. |
610 | problems and requires 1-3 extra syscalls per active fd every iteration. |
|
|
611 | |
621 | |
612 | This backend maps C<EV_READ> and C<EV_WRITE> in the same way as |
622 | This backend maps C<EV_READ> and C<EV_WRITE> in the same way as |
613 | C<EVBACKEND_POLL>. |
623 | C<EVBACKEND_POLL>. |
614 | |
624 | |
615 | =item C<EVBACKEND_KQUEUE> (value 8, most BSD clones) |
625 | =item C<EVBACKEND_KQUEUE> (value 8, most BSD clones) |
616 | |
626 | |
617 | Kqueue deserves special mention, as at the time of this writing, it |
627 | Kqueue deserves special mention, as at the time this backend was |
618 | was broken on all BSDs except NetBSD (usually it doesn't work reliably |
628 | implemented, it was broken on all BSDs except NetBSD (usually it doesn't |
619 | with anything but sockets and pipes, except on Darwin, where of course |
629 | work reliably with anything but sockets and pipes, except on Darwin, |
620 | it's completely useless). Unlike epoll, however, whose brokenness |
630 | where of course it's completely useless). Unlike epoll, however, whose |
621 | is by design, these kqueue bugs can (and eventually will) be fixed |
631 | brokenness is by design, these kqueue bugs can be (and mostly have been) |
622 | without API changes to existing programs. For this reason it's not being |
632 | fixed without API changes to existing programs. For this reason it's not |
623 | "auto-detected" unless you explicitly specify it in the flags (i.e. using |
633 | being "auto-detected" on all platforms unless you explicitly specify it |
624 | C<EVBACKEND_KQUEUE>) or libev was compiled on a known-to-be-good (-enough) |
634 | in the flags (i.e. using C<EVBACKEND_KQUEUE>) or libev was compiled on a |
625 | system like NetBSD. |
635 | known-to-be-good (-enough) system like NetBSD. |
626 | |
636 | |
627 | You still can embed kqueue into a normal poll or select backend and use it |
637 | You still can embed kqueue into a normal poll or select backend and use it |
628 | only for sockets (after having made sure that sockets work with kqueue on |
638 | only for sockets (after having made sure that sockets work with kqueue on |
629 | the target platform). See C<ev_embed> watchers for more info. |
639 | the target platform). See C<ev_embed> watchers for more info. |
630 | |
640 | |
631 | It scales in the same way as the epoll backend, but the interface to the |
641 | It scales in the same way as the epoll backend, but the interface to the |
632 | kernel is more efficient (which says nothing about its actual speed, of |
642 | kernel is more efficient (which says nothing about its actual speed, of |
633 | course). While stopping, setting and starting an I/O watcher does never |
643 | course). While stopping, setting and starting an I/O watcher does never |
634 | cause an extra system call as with C<EVBACKEND_EPOLL>, it still adds up to |
644 | cause an extra system call as with C<EVBACKEND_EPOLL>, it still adds up to |
635 | two event changes per incident. Support for C<fork ()> is very bad (you |
645 | two event changes per incident. Support for C<fork ()> is very bad (you |
636 | might have to leak fd's on fork, but it's more sane than epoll) and it |
646 | might have to leak fds on fork, but it's more sane than epoll) and it |
637 | drops fds silently in similarly hard-to-detect cases. |
647 | drops fds silently in similarly hard-to-detect cases. |
638 | |
648 | |
639 | This backend usually performs well under most conditions. |
649 | This backend usually performs well under most conditions. |
640 | |
650 | |
641 | While nominally embeddable in other event loops, this doesn't work |
651 | While nominally embeddable in other event loops, this doesn't work |
… | |
… | |
4967 | called. If set to C<2>, then the internal verification code will be |
4977 | called. If set to C<2>, then the internal verification code will be |
4968 | called once per loop, which can slow down libev. If set to C<3>, then the |
4978 | called once per loop, which can slow down libev. If set to C<3>, then the |
4969 | verification code will be called very frequently, which will slow down |
4979 | verification code will be called very frequently, which will slow down |
4970 | libev considerably. |
4980 | libev considerably. |
4971 | |
4981 | |
|
|
4982 | Verification errors are reported via C's C<assert> mechanism, so if you |
|
|
4983 | disable that (e.g. by defining C<NDEBUG>) then no errors will be reported. |
|
|
4984 | |
4972 | The default is C<1>, unless C<EV_FEATURES> overrides it, in which case it |
4985 | The default is C<1>, unless C<EV_FEATURES> overrides it, in which case it |
4973 | will be C<0>. |
4986 | will be C<0>. |
4974 | |
4987 | |
4975 | =item EV_COMMON |
4988 | =item EV_COMMON |
4976 | |
4989 | |