[ViewVC] Diff of: cvs/libev/ev.3

Comparing libev/ev.3 (file contents):
Revision 1.110 by root, Thu Jun 20 22:44:59 2019 UTC vs.
Revision 1.118 by root, Sat Dec 21 16:11:51 2019 UTC

 .\}
 .rm #[ #] #H #V #F C
 .\" ========================================================================
 .\"
 .IX Title "LIBEV 3"
-.TH LIBEV 3 "2019-06-20" "libev-4.25" "libev - high performance full featured event loop"
+.TH LIBEV 3 "2019-12-21" "libev-4.31" "libev - high performance full featured event loop"
 .\" For nroff, turn off justification.  Always turn off hyphenation; it makes
 .\" way too many mistakes in technical documents.
 .if n .ad l
 .nh
 .SH "NAME"
 watchers\fR, which are relatively small C structures you initialise with the
 details of the event, and then hand it over to libev by \fIstarting\fR the
 watcher.
 .SS "\s-1FEATURES\s0"
 .IX Subsection "FEATURES"
-Libev supports \f(CW\*(C`select\*(C'\fR, \f(CW\*(C`poll\*(C'\fR, the Linux-specific \f(CW\*(C`epoll\*(C'\fR, the
+Libev supports \f(CW\*(C`select\*(C'\fR, \f(CW\*(C`poll\*(C'\fR, the Linux-specific aio and \f(CW\*(C`epoll\*(C'\fR
-BSD-specific \f(CW\*(C`kqueue\*(C'\fR and the Solaris-specific event port mechanisms
+interfaces, the BSD-specific \f(CW\*(C`kqueue\*(C'\fR and the Solaris-specific event port
-for file descriptor events (\f(CW\*(C`ev_io\*(C'\fR), the Linux \f(CW\*(C`inotify\*(C'\fR interface
+mechanisms for file descriptor events (\f(CW\*(C`ev_io\*(C'\fR), the Linux \f(CW\*(C`inotify\*(C'\fR
-(for \f(CW\*(C`ev_stat\*(C'\fR), Linux eventfd/signalfd (for faster and cleaner
+interface (for \f(CW\*(C`ev_stat\*(C'\fR), Linux eventfd/signalfd (for faster and cleaner
 inter-thread wakeup (\f(CW\*(C`ev_async\*(C'\fR)/signal handling (\f(CW\*(C`ev_signal\*(C'\fR)) relative
 timers (\f(CW\*(C`ev_timer\*(C'\fR), absolute timers with customised rescheduling
 (\f(CW\*(C`ev_periodic\*(C'\fR), synchronous signals (\f(CW\*(C`ev_signal\*(C'\fR), process status
 change events (\f(CW\*(C`ev_child\*(C'\fR), and event watchers dealing with the event
 loop mechanism itself (\f(CW\*(C`ev_idle\*(C'\fR, \f(CW\*(C`ev_embed\*(C'\fR, \f(CW\*(C`ev_prepare\*(C'\fR and
 When libev detects a usage error such as a negative timer interval, then
 it will print a diagnostic message and abort (via the \f(CW\*(C`assert\*(C'\fR mechanism,
 so \f(CW\*(C`NDEBUG\*(C'\fR will disable this checking): these are programming errors in
 the libev caller and need to be fixed there.
 .PP
+Via the \f(CW\*(C`EV_FREQUENT\*(C'\fR macro you can compile in and/or enable extensive
+consistency checking code inside libev that can be used to check for
+internal inconsistencies, suually caused by application bugs.
+.PP
-Libev also has a few internal error-checking \f(CW\*(C`assert\*(C'\fRions, and also has
+Libev also has a few internal error-checking \f(CW\*(C`assert\*(C'\fRions. These do not
-extensive consistency checking code. These do not trigger under normal
-circumstances, as they indicate either a bug in libev or worse.
+trigger under normal circumstances, as they indicate either a bug in libev
+or worse.
 .SH "GLOBAL FUNCTIONS"
 .IX Header "GLOBAL FUNCTIONS"
 These functions can be called anytime, even before initialising the
 library in any way.
 .IP "ev_tstamp ev_time ()" 4
 want to handle signals only in specific threads and want to avoid libev
 unblocking the signals.
 .Sp
 It's also required by \s-1POSIX\s0 in a threaded program, as libev calls
 \&\f(CW\*(C`sigprocmask\*(C'\fR, whose behaviour is officially unspecified.
+.ie n .IP """EVFLAG_NOTIMERFD""" 4
+.el .IP "\f(CWEVFLAG_NOTIMERFD\fR" 4
+.IX Item "EVFLAG_NOTIMERFD"
+When this flag is specified, the libev will avoid using a \f(CW\*(C`timerfd\*(C'\fR to
+detect time jumps. It will still be able to detect time jumps, but takes
+longer and has a lower accuracy in doing so, but saves a file descriptor
+per loop.
 .Sp
-This flag's behaviour will become the default in future versions of libev.
+The current implementation only tries to use a \f(CW\*(C`timerfd\*(C'\fR when the first
+\&\f(CW\*(C`ev_periodic\*(C'\fR watcher is started and falls back on other methods if it
+cannot be created, but this behaviour might change in the future.
 .ie n .IP """EVBACKEND_SELECT""  (value 1, portable select backend)" 4
 .el .IP "\f(CWEVBACKEND_SELECT\fR  (value 1, portable select backend)" 4
 .IX Item "EVBACKEND_SELECT (value 1, portable select backend)"
 This is your standard \fBselect\fR\|(2) backend. Not \fIcompletely\fR standard, as
 libev tries to roll its own fd_set with no limits on the number of fds,
 This backend maps \f(CW\*(C`EV_READ\*(C'\fR to \f(CW\*(C`POLLIN | POLLERR | POLLHUP\*(C'\fR, and
 \&\f(CW\*(C`EV_WRITE\*(C'\fR to \f(CW\*(C`POLLOUT | POLLERR | POLLHUP\*(C'\fR.
 .ie n .IP """EVBACKEND_EPOLL""   (value 4, Linux)" 4
 .el .IP "\f(CWEVBACKEND_EPOLL\fR   (value 4, Linux)" 4
 .IX Item "EVBACKEND_EPOLL (value 4, Linux)"
-Use the linux-specific \fBepoll\fR\|(7) interface (for both pre\- and post\-2.6.9
+Use the Linux-specific \fBepoll\fR\|(7) interface (for both pre\- and post\-2.6.9
 kernels).
 .Sp
 For few fds, this backend is a bit little slower than poll and select, but
 it scales phenomenally better. While poll and select usually scale like
 O(total_fds) where total_fds is the total number of fds (or the highest
 All this means that, in practice, \f(CW\*(C`EVBACKEND_SELECT\*(C'\fR can be as fast or
 faster than epoll for maybe up to a hundred file descriptors, depending on
 the usage. So sad.
 .Sp
 While nominally embeddable in other event loops, this feature is broken in
-all kernel versions tested so far.
+a lot of kernel revisions, but probably(!) works in current versions.
+.Sp
+This backend maps \f(CW\*(C`EV_READ\*(C'\fR and \f(CW\*(C`EV_WRITE\*(C'\fR in the same way as
+\&\f(CW\*(C`EVBACKEND_POLL\*(C'\fR.
+.ie n .IP """EVBACKEND_LINUXAIO""   (value 64, Linux)" 4
+.el .IP "\f(CWEVBACKEND_LINUXAIO\fR   (value 64, Linux)" 4
+.IX Item "EVBACKEND_LINUXAIO (value 64, Linux)"
+Use the Linux-specific Linux \s-1AIO\s0 (\fInot\fR \f(CWaio(7)\fR but \f(CWio_submit(2)\fR) event interface available in post\-4.18 kernels (but libev
+only tries to use it in 4.19+).
+.Sp
+This is another Linux train wreck of an event interface.
+.Sp
+If this backend works for you (as of this writing, it was very
+experimental), it is the best event interface available on Linux and might
+be well worth enabling it \- if it isn't available in your kernel this will
+be detected and this backend will be skipped.
+.Sp
+This backend can batch oneshot requests and supports a user-space ring
+buffer to receive events. It also doesn't suffer from most of the design
+problems of epoll (such as not being able to remove event sources from
+the epoll set), and generally sounds too good to be true. Because, this
+being the Linux kernel, of course it suffers from a whole new set of
+limitations, forcing you to fall back to epoll, inheriting all its design
+issues.
+.Sp
+For one, it is not easily embeddable (but probably could be done using
+an event fd at some extra overhead). It also is subject to a system wide
+limit that can be configured in \fI/proc/sys/fs/aio\-max\-nr\fR. If no \s-1AIO\s0
+requests are left, this backend will be skipped during initialisation, and
+will switch to epoll when the loop is active.
+.Sp
+Most problematic in practice, however, is that not all file descriptors
+work with it. For example, in Linux 5.1, \s-1TCP\s0 sockets, pipes, event fds,
+files, \fI/dev/null\fR and many others are supported, but ttys do not work
+properly (a known bug that the kernel developers don't care about, see
+<https://lore.kernel.org/patchwork/patch/1047453/>), so this is not
+(yet?) a generic event polling interface.
+.Sp
+Overall, it seems the Linux developers just don't want it to have a
+generic event handling mechanism other than \f(CW\*(C`select\*(C'\fR or \f(CW\*(C`poll\*(C'\fR.
+.Sp
+To work around all these problem, the current version of libev uses its
+epoll backend as a fallback for file descriptor types that do not work. Or
+falls back completely to epoll if the kernel acts up.
 .Sp
 This backend maps \f(CW\*(C`EV_READ\*(C'\fR and \f(CW\*(C`EV_WRITE\*(C'\fR in the same way as
 \&\f(CW\*(C`EVBACKEND_POLL\*(C'\fR.
 .ie n .IP """EVBACKEND_KQUEUE""  (value 8, most \s-1BSD\s0 clones)" 4
 .el .IP "\f(CWEVBACKEND_KQUEUE\fR  (value 8, most \s-1BSD\s0 clones)" 4
 .IX Item "EVBACKEND_KQUEUE (value 8, most BSD clones)"
-Kqueue deserves special mention, as at the time of this writing, it
+Kqueue deserves special mention, as at the time this backend was
-was broken on all BSDs except NetBSD (usually it doesn't work reliably
+implemented, it was broken on all BSDs except NetBSD (usually it doesn't
-with anything but sockets and pipes, except on Darwin, where of course
+work reliably with anything but sockets and pipes, except on Darwin,
-it's completely useless). Unlike epoll, however, whose brokenness
+where of course it's completely useless). Unlike epoll, however, whose
-is by design, these kqueue bugs can (and eventually will) be fixed
+brokenness is by design, these kqueue bugs can be (and mostly have been)
-without \s-1API\s0 changes to existing programs. For this reason it's not being
+fixed without \s-1API\s0 changes to existing programs. For this reason it's not
-\&\*(L"auto-detected\*(R" unless you explicitly specify it in the flags (i.e. using
+being \*(L"auto-detected\*(R" on all platforms unless you explicitly specify it
-\&\f(CW\*(C`EVBACKEND_KQUEUE\*(C'\fR) or libev was compiled on a known-to-be-good (\-enough)
+in the flags (i.e. using \f(CW\*(C`EVBACKEND_KQUEUE\*(C'\fR) or libev was compiled on a
-system like NetBSD.
+known-to-be-good (\-enough) system like NetBSD.
 .Sp
 You still can embed kqueue into a normal poll or select backend and use it
 only for sockets (after having made sure that sockets work with kqueue on
 the target platform). See \f(CW\*(C`ev_embed\*(C'\fR watchers for more info.
 .Sp
 It scales in the same way as the epoll backend, but the interface to the
 kernel is more efficient (which says nothing about its actual speed, of
 course). While stopping, setting and starting an I/O watcher does never
 cause an extra system call as with \f(CW\*(C`EVBACKEND_EPOLL\*(C'\fR, it still adds up to
 two event changes per incident. Support for \f(CW\*(C`fork ()\*(C'\fR is very bad (you
-might have to leak fd's on fork, but it's more sane than epoll) and it
+might have to leak fds on fork, but it's more sane than epoll) and it
 drops fds silently in similarly hard-to-detect cases.
 .Sp
 This backend usually performs well under most conditions.
 .Sp
 While nominally embeddable in other event loops, this doesn't work
 Example: Use whatever libev has to offer, but make sure that kqueue is
 used if available.
 .Sp
 .Vb 1
 \&   struct ev_loop *loop = ev_loop_new (ev_recommended_backends () | EVBACKEND_KQUEUE);
+.Ve
+.Sp
+Example: Similarly, on linux, you mgiht want to take advantage of the
+linux aio backend if possible, but fall back to something else if that
+isn't available.
+.Sp
+.Vb 1
+\&   struct ev_loop *loop = ev_loop_new (ev_recommended_backends () | EVBACKEND_LINUXAIO);
 .Ve
 .RE
 .IP "ev_loop_destroy (loop)" 4
 .IX Item "ev_loop_destroy (loop)"
 Destroys an event loop object (frees all memory and kernel state
 .IX Subsection "WATCHER PRIORITY MODELS"
 Many event loops support \fIwatcher priorities\fR, which are usually small
 integers that influence the ordering of event callback invocation
 between watchers in some way, all else being equal.
 .PP
-In libev, Watcher priorities can be set using \f(CW\*(C`ev_set_priority\*(C'\fR. See its
+In libev, watcher priorities can be set using \f(CW\*(C`ev_set_priority\*(C'\fR. See its
 description for the more technical details such as the actual priority
 range.
 .PP
 There are two common ways how these these priorities are being interpreted
 by event loops:
 But really, best use non-blocking mode.
 .PP
 \fIThe special problem of disappearing file descriptors\fR
 .IX Subsection "The special problem of disappearing file descriptors"
 .PP
-Some backends (e.g. kqueue, epoll) need to be told about closing a file
+Some backends (e.g. kqueue, epoll, linuxaio) need to be told about closing
-descriptor (either due to calling \f(CW\*(C`close\*(C'\fR explicitly or any other means,
+a file descriptor (either due to calling \f(CW\*(C`close\*(C'\fR explicitly or any other
-such as \f(CW\*(C`dup2\*(C'\fR). The reason is that you register interest in some file
+means, such as \f(CW\*(C`dup2\*(C'\fR). The reason is that you register interest in some
-descriptor, but when it goes away, the operating system will silently drop
+file descriptor, but when it goes away, the operating system will silently
-this interest. If another file descriptor with the same number then is
+drop this interest. If another file descriptor with the same number then
-registered with libev, there is no efficient way to see that this is, in
+is registered with libev, there is no efficient way to see that this is,
-fact, a different file descriptor.
+in fact, a different file descriptor.
 .PP
 To avoid having to explicitly tell libev about such cases, libev follows
 the following policy:  Each time \f(CW\*(C`ev_io_set\*(C'\fR is being called, libev
 will assume that this is potentially a new file descriptor, otherwise
 it is assumed that the file descriptor stays the same. That means that
 reuse the same code path.
 .PP
 \fIThe special problem of fork\fR
 .IX Subsection "The special problem of fork"
 .PP
-Some backends (epoll, kqueue) do not support \f(CW\*(C`fork ()\*(C'\fR at all or exhibit
+Some backends (epoll, kqueue, linuxaio, iouring) do not support \f(CW\*(C`fork ()\*(C'\fR
-useless behaviour. Libev fully supports fork, but needs to be told about
+at all or exhibit useless behaviour. Libev fully supports fork, but needs
-it in the child if you want to continue to use it in the child.
+to be told about it in the child if you want to continue to use it in the
+child.
 .PP
 To support fork in your child processes, you have to call \f(CW\*(C`ev_loop_fork
 ()\*(C'\fR after a fork in the child, enable \f(CW\*(C`EVFLAG_FORKCHECK\*(C'\fR, or resort to
 \&\f(CW\*(C`EVBACKEND_SELECT\*(C'\fR or \f(CW\*(C`EVBACKEND_POLL\*(C'\fR.
 .PP
 \&   ev_win32.c      required on win32 platforms only
 \&
 \&   ev_select.c     only when select backend is enabled
 \&   ev_poll.c       only when poll backend is enabled
 \&   ev_epoll.c      only when the epoll backend is enabled
+\&   ev_linuxaio.c   only when the linux aio backend is enabled
+\&   ev_iouring.c    only when the linux io_uring backend is enabled
 \&   ev_kqueue.c     only when the kqueue backend is enabled
 \&   ev_port.c       only when the solaris port backend is enabled
 .Ve
 .PP
 \&\fIev.c\fR includes the backend files directly when enabled, so you only need
 higher, as it simplifies linking (no need for \f(CW\*(C`\-lrt\*(C'\fR).
 .IP "\s-1EV_USE_NANOSLEEP\s0" 4
 .IX Item "EV_USE_NANOSLEEP"
 If defined to be \f(CW1\fR, libev will assume that \f(CW\*(C`nanosleep ()\*(C'\fR is available
 and will use it for delays. Otherwise it will use \f(CW\*(C`select ()\*(C'\fR.
+.IP "\s-1EV_USE_EVENTFD\s0" 4
+.IX Item "EV_USE_EVENTFD"
+If defined to be \f(CW1\fR, then libev will assume that \f(CW\*(C`eventfd ()\*(C'\fR is
+available and will probe for kernel support at runtime. This will improve
+\&\f(CW\*(C`ev_signal\*(C'\fR and \f(CW\*(C`ev_async\*(C'\fR performance and reduce resource consumption.
+If undefined, it will be enabled if the headers indicate GNU/Linux + Glibc
+2.7 or newer, otherwise disabled.
+.IP "\s-1EV_USE_SIGNALFD\s0" 4
+.IX Item "EV_USE_SIGNALFD"
+If defined to be \f(CW1\fR, then libev will assume that \f(CW\*(C`signalfd ()\*(C'\fR is
+available and will probe for kernel support at runtime. This enables
+the use of \s-1EVFLAG_SIGNALFD\s0 for faster and simpler signal handling. If
+undefined, it will be enabled if the headers indicate GNU/Linux + Glibc
+2.7 or newer, otherwise disabled.
+.IP "\s-1EV_USE_TIMERFD\s0" 4
+.IX Item "EV_USE_TIMERFD"
+If defined to be \f(CW1\fR, then libev will assume that \f(CW\*(C`timerfd ()\*(C'\fR is
+available and will probe for kernel support at runtime. This allows
+libev to detect time jumps accurately. If undefined, it will be enabled
+if the headers indicate GNU/Linux + Glibc 2.8 or newer and define
+\&\f(CW\*(C`TFD_TIMER_CANCEL_ON_SET\*(C'\fR, otherwise disabled.
 .IP "\s-1EV_USE_EVENTFD\s0" 4
 .IX Item "EV_USE_EVENTFD"
 If defined to be \f(CW1\fR, then libev will assume that \f(CW\*(C`eventfd ()\*(C'\fR is
 available and will probe for kernel support at runtime. This will improve
 \&\f(CW\*(C`ev_signal\*(C'\fR and \f(CW\*(C`ev_async\*(C'\fR performance and reduce resource consumption.
 If defined to be \f(CW1\fR, libev will compile in support for the Linux
 \&\f(CW\*(C`epoll\*(C'\fR(7) backend. Its availability will be detected at runtime,
 otherwise another method will be used as fallback. This is the preferred
 backend for GNU/Linux systems. If undefined, it will be enabled if the
 headers indicate GNU/Linux + Glibc 2.4 or newer, otherwise disabled.
+.IP "\s-1EV_USE_LINUXAIO\s0" 4
+.IX Item "EV_USE_LINUXAIO"
+If defined to be \f(CW1\fR, libev will compile in support for the Linux aio
+backend (\f(CW\*(C`EV_USE_EPOLL\*(C'\fR must also be enabled). If undefined, it will be
+enabled on linux, otherwise disabled.
+.IP "\s-1EV_USE_IOURING\s0" 4
+.IX Item "EV_USE_IOURING"
+If defined to be \f(CW1\fR, libev will compile in support for the Linux
+io_uring backend (\f(CW\*(C`EV_USE_EPOLL\*(C'\fR must also be enabled). Due to it's
+current limitations it has to be requested explicitly. If undefined, it
+will be enabled on linux, otherwise disabled.
 .IP "\s-1EV_USE_KQUEUE\s0" 4
 .IX Item "EV_USE_KQUEUE"
 If defined to be \f(CW1\fR, libev will compile in support for the \s-1BSD\s0 style
 \&\f(CW\*(C`kqueue\*(C'\fR(2) backend. Its actual availability will be detected at runtime,
 otherwise another method will be used as fallback. This is the preferred
 in. If set to \f(CW1\fR, then verification code will be compiled in, but not
 called. If set to \f(CW2\fR, then the internal verification code will be
 called once per loop, which can slow down libev. If set to \f(CW3\fR, then the
 verification code will be called very frequently, which will slow down
 libev considerably.
+.Sp
+Verification errors are reported via C's \f(CW\*(C`assert\*(C'\fR mechanism, so if you
+disable that (e.g. by defining \f(CW\*(C`NDEBUG\*(C'\fR) then no errors will be reported.
 .Sp
 The default is \f(CW1\fR, unless \f(CW\*(C`EV_FEATURES\*(C'\fR overrides it, in which case it
 will be \f(CW0\fR.
 .IP "\s-1EV_COMMON\s0" 4
 .IX Item "EV_COMMON"

Diff Legend

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

Comparing libev/ev.3 (file contents): Revision 1.110 by root, Thu Jun 20 22:44:59 2019 UTC vs. Revision 1.118 by root, Sat Dec 21 16:11:51 2019 UTC

Diff Legend

Comparing libev/ev.3 (file contents):
Revision 1.110 by root, Thu Jun 20 22:44:59 2019 UTC vs.
Revision 1.118 by root, Sat Dec 21 16:11:51 2019 UTC