… | |
… | |
511 | This backend maps C<EV_READ> to C<POLLIN | POLLERR | POLLHUP>, and |
511 | This backend maps C<EV_READ> to C<POLLIN | POLLERR | POLLHUP>, and |
512 | C<EV_WRITE> to C<POLLOUT | POLLERR | POLLHUP>. |
512 | C<EV_WRITE> to C<POLLOUT | POLLERR | POLLHUP>. |
513 | |
513 | |
514 | =item C<EVBACKEND_EPOLL> (value 4, Linux) |
514 | =item C<EVBACKEND_EPOLL> (value 4, Linux) |
515 | |
515 | |
516 | Use the linux-specific epoll(7) interface (for both pre- and post-2.6.9 |
516 | Use the Linux-specific epoll(7) interface (for both pre- and post-2.6.9 |
517 | kernels). |
517 | kernels). |
518 | |
518 | |
519 | For few fds, this backend is a bit little slower than poll and select, but |
519 | 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 |
520 | 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 |
521 | 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 |
574 | This backend maps C<EV_READ> and C<EV_WRITE> in the same way as |
575 | C<EVBACKEND_POLL>. |
575 | C<EVBACKEND_POLL>. |
576 | |
576 | |
577 | =item C<EVBACKEND_LINUXAIO> (value 64, Linux) |
577 | =item C<EVBACKEND_LINUXAIO> (value 64, Linux) |
578 | |
578 | |
579 | Use the linux-specific linux aio (I<not> C<< aio(7) >> but C<< |
579 | 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. |
580 | io_submit(2) >>) event interface available in post-4.18 kernels (but libev |
|
|
581 | only tries to use it in 4.19+). |
|
|
582 | |
|
|
583 | This is another Linux train wreck of an event interface. |
581 | |
584 | |
582 | If this backend works for you (as of this writing, it was very |
585 | 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 |
586 | 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 |
587 | be well worth enabling it - if it isn't available in your kernel this will |
585 | be detected and this backend will be skipped. |
588 | be detected and this backend will be skipped. |
586 | |
589 | |
587 | This backend can batch oneshot requests and supports a user-space ring |
590 | 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 |
591 | 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 the |
592 | problems of epoll (such as not being able to remove event sources from |
590 | epoll set), and generally sounds too good to be true. Because, this being |
593 | the epoll set), and generally sounds too good to be true. Because, this |
591 | the linux kernel, of course it suffers from a whole new set of limitations. |
594 | being the Linux kernel, of course it suffers from a whole new set of |
|
|
595 | limitations, forcing you to fall back to epoll, inheriting all its design |
|
|
596 | issues. |
592 | |
597 | |
593 | For one, it is not easily embeddable (but probably could be done using |
598 | For one, it is not easily embeddable (but probably could be done using |
594 | an event fd at some extra overhead). It also is subject to a system wide |
599 | an event fd at some extra overhead). It also is subject to a system wide |
595 | limit that can be configured in F</proc/sys/fs/aio-max-nr> - each loop |
600 | limit that can be configured in F</proc/sys/fs/aio-max-nr>. If no AIO |
596 | currently requires C<61> of this number. If no aio requests are left, this |
|
|
597 | backend will be skipped during initialisation. |
601 | requests are left, this backend will be skipped during initialisation, and |
|
|
602 | will switch to epoll when the loop is active. |
598 | |
603 | |
599 | Most problematic in practise, however, is that not all file descriptors |
604 | Most problematic in practice, however, is that not all file descriptors |
600 | work with it. For example, in linux 5.1, tcp sockets, pipes, event fds, |
605 | work with it. For example, in Linux 5.1, TCP sockets, pipes, event fds, |
601 | files, F</dev/null> and a few others are supported, but ttys do not work |
606 | files, F</dev/null> and many others are supported, but ttys do not work |
602 | properly (a known bug that the kernel developers don't care about, see |
607 | properly (a known bug that the kernel developers don't care about, see |
603 | L<https://lore.kernel.org/patchwork/patch/1047453/>), so this is not |
608 | L<https://lore.kernel.org/patchwork/patch/1047453/>), so this is not |
604 | (yet?) a generic event polling interface. |
609 | (yet?) a generic event polling interface. |
605 | |
610 | |
606 | Overall, it seems the linux developers just don't want it to have a |
611 | Overall, it seems the Linux developers just don't want it to have a |
607 | generic event handling mechanism other than C<select> or C<poll>. |
612 | generic event handling mechanism other than C<select> or C<poll>. |
608 | |
613 | |
609 | To work around the fd type problem, the current version of libev uses |
614 | To work around all these problem, the current version of libev uses its |
610 | epoll as a fallback for file deescriptor types that do not work. Epoll |
615 | epoll backend as a fallback for file descriptor types that do not work. Or |
611 | is used in, kind of, slow mode that hopefully avoids most of its design |
616 | falls back completely to epoll if the kernel acts up. |
612 | problems and requires 1-3 extra syscalls per active fd every iteration. |
|
|
613 | |
617 | |
614 | This backend maps C<EV_READ> and C<EV_WRITE> in the same way as |
618 | This backend maps C<EV_READ> and C<EV_WRITE> in the same way as |
615 | C<EVBACKEND_POLL>. |
619 | C<EVBACKEND_POLL>. |
616 | |
620 | |
617 | =item C<EVBACKEND_KQUEUE> (value 8, most BSD clones) |
621 | =item C<EVBACKEND_KQUEUE> (value 8, most BSD clones) |
618 | |
622 | |
619 | Kqueue deserves special mention, as at the time of this writing, it |
623 | Kqueue deserves special mention, as at the time this backend was |
620 | was broken on all BSDs except NetBSD (usually it doesn't work reliably |
624 | implemented, it was broken on all BSDs except NetBSD (usually it doesn't |
621 | with anything but sockets and pipes, except on Darwin, where of course |
625 | work reliably with anything but sockets and pipes, except on Darwin, |
622 | it's completely useless). Unlike epoll, however, whose brokenness |
626 | where of course it's completely useless). Unlike epoll, however, whose |
623 | is by design, these kqueue bugs can (and eventually will) be fixed |
627 | brokenness is by design, these kqueue bugs can be (and mostly have been) |
624 | without API changes to existing programs. For this reason it's not being |
628 | fixed without API changes to existing programs. For this reason it's not |
625 | "auto-detected" unless you explicitly specify it in the flags (i.e. using |
629 | being "auto-detected" on all platforms unless you explicitly specify it |
626 | C<EVBACKEND_KQUEUE>) or libev was compiled on a known-to-be-good (-enough) |
630 | in the flags (i.e. using C<EVBACKEND_KQUEUE>) or libev was compiled on a |
627 | system like NetBSD. |
631 | known-to-be-good (-enough) system like NetBSD. |
628 | |
632 | |
629 | You still can embed kqueue into a normal poll or select backend and use it |
633 | You still can embed kqueue into a normal poll or select backend and use it |
630 | only for sockets (after having made sure that sockets work with kqueue on |
634 | only for sockets (after having made sure that sockets work with kqueue on |
631 | the target platform). See C<ev_embed> watchers for more info. |
635 | the target platform). See C<ev_embed> watchers for more info. |
632 | |
636 | |
633 | It scales in the same way as the epoll backend, but the interface to the |
637 | It scales in the same way as the epoll backend, but the interface to the |
634 | kernel is more efficient (which says nothing about its actual speed, of |
638 | kernel is more efficient (which says nothing about its actual speed, of |
635 | course). While stopping, setting and starting an I/O watcher does never |
639 | course). While stopping, setting and starting an I/O watcher does never |
636 | cause an extra system call as with C<EVBACKEND_EPOLL>, it still adds up to |
640 | cause an extra system call as with C<EVBACKEND_EPOLL>, it still adds up to |
637 | two event changes per incident. Support for C<fork ()> is very bad (you |
641 | two event changes per incident. Support for C<fork ()> is very bad (you |
638 | might have to leak fd's on fork, but it's more sane than epoll) and it |
642 | might have to leak fds on fork, but it's more sane than epoll) and it |
639 | drops fds silently in similarly hard-to-detect cases. |
643 | drops fds silently in similarly hard-to-detect cases. |
640 | |
644 | |
641 | This backend usually performs well under most conditions. |
645 | This backend usually performs well under most conditions. |
642 | |
646 | |
643 | While nominally embeddable in other event loops, this doesn't work |
647 | While nominally embeddable in other event loops, this doesn't work |