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

Comparing libev/ev.pod (file contents):
Revision 1.208 by root, Wed Oct 29 10:24:23 2008 UTC vs.
Revision 1.213 by root, Wed Nov 5 02:48:45 2008 UTC

 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 n is the total number of fds (or the highest fd),
 epoll scales either O(1) or O(active_fds).
-The epoll syscalls are the most misdesigned of the more advanced event
+The epoll mechanism deserves honorable mention as the most misdesigned
-mechanisms: problems include silently dropping fds, requiring a system
+of the more advanced event mechanisms: mere annoyances include silently
-call per change per fd (and unnecessary guessing of parameters), problems
+dropping file descriptors, requiring a system call per change per file
+descriptor (and unnecessary guessing of parameters), problems with dup and
-with dup and so on. The biggest issue is fork races, however - if a
+so on. The biggest issue is fork races, however - if a program forks then
-program forks then I<both> parent and child process have to recreate the
+I<both> parent and child process have to recreate the epoll set, which can
-epoll set, which can take considerable time (one syscall per fd) and is of
+take considerable time (one syscall per file descriptor) and is of course
-course hard to detect.
+hard to detect.
-Epoll is also notoriously buggy - embedding epoll fds should work, but
+Epoll is also notoriously buggy - embedding epoll fds I<should> work, but
-of course doesn't, and epoll just loves to report events for totally
+of course I<doesn't>, and epoll just loves to report events for totally
 I<different> file descriptors (even already closed ones, so one cannot
 even remove them from the set) than registered in the set (especially
 on SMP systems). Libev tries to counter these spurious notifications by
 employing an additional generation counter and comparing that against the
-events to filter out spurious ones.
+events to filter out spurious ones, recreating the set when required.
 While stopping, setting and starting an I/O watcher in the same iteration
-will result in some caching, there is still a system call per such incident
+will result in some caching, there is still a system call per such
-(because the fd could point to a different file description now), so its
+incident (because the same I<file descriptor> could point to a different
-best to avoid that. Also, C<dup ()>'ed file descriptors might not work
+I<file description> now), so its best to avoid that. Also, C<dup ()>'ed
-very well if you register events for both fds.
+file descriptors might not work very well if you register events for both
+file descriptors.
 Best performance from this backend is achieved by not unregistering all
 watchers for a file descriptor until it has been closed, if possible,
 i.e. keep at least one watcher active per fd at all times. Stopping and
 starting a watcher (without re-setting it) also usually doesn't cause
 extra overhead. A fork can both result in spurious notifications as well
 as in libev having to destroy and recreate the epoll object, which can
 take considerable time and thus should be avoided.
+All this means that, in practise, C<EVBACKEND_SELECT> is as fast or faster
+then epoll for maybe up to a hundred file descriptors. So sad.
 While nominally embeddable in other event loops, this feature is broken in
 all kernel versions tested so far.
 This backend maps C<EV_READ> and C<EV_WRITE> in the same way as
 C<EVBACKEND_POLL>.
 =item C<EVBACKEND_KQUEUE>  (value 8, most BSD clones)
-Kqueue deserves special mention, as at the time of this writing, it was
+Kqueue deserves special mention, as at the time of this writing, it
-broken on all BSDs except NetBSD (usually it doesn't work reliably with
+was broken on all BSDs except NetBSD (usually it doesn't work reliably
-anything but sockets and pipes, except on Darwin, where of course it's
+with anything but sockets and pipes, except on Darwin, where of course
-completely useless). For this reason it's not being "auto-detected" unless
+it's completely useless). Unlike epoll, however, whose brokenness
-you explicitly specify it in the flags (i.e. using C<EVBACKEND_KQUEUE>) or
+is by design, these kqueue bugs can (and eventually will) be fixed
-libev was compiled on a known-to-be-good (-enough) system like NetBSD.
+without API changes to existing programs. For this reason it's not being
+"auto-detected" unless you explicitly specify it in the flags (i.e. using
+C<EVBACKEND_KQUEUE>) or libev was compiled on a known-to-be-good (-enough)
+system like NetBSD.
 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 C<ev_embed> watchers for more info.
 C<stat> on that path in regular intervals (or when the OS says it changed)
 and sees if it changed compared to the last time, invoking the callback if
 it did.
 The path does not need to exist: changing from "path exists" to "path does
-not exist" is a status change like any other. The condition "path does
+not exist" is a status change like any other. The condition "path does not
-not exist" is signified by the C<st_nlink> field being zero (which is
+exist" (or more correctly "path cannot be stat'ed") is signified by the
-otherwise always forced to be at least one) and all the other fields of
+C<st_nlink> field being zero (which is otherwise always forced to be at
-the stat buffer having unspecified contents.
+least one) and all the other fields of the stat buffer having unspecified
+contents.
 The path I<must not> end in a slash or contain special components such as
 C<.> or C<..>. The path I<should> be absolute: If it is relative and
 your working directory changes, then the behaviour is undefined.
 This watcher type is not meant for massive numbers of stat watchers,
 as even with OS-supported change notifications, this can be
 resource-intensive.
 At the time of this writing, the only OS-specific interface implemented
-is the Linux inotify interface (implementing kqueue support is left as
+is the Linux inotify interface (implementing kqueue support is left as an
-an exercise for the reader. Note, however, that the author sees no way
+exercise for the reader. Note, however, that the author sees no way of
-of implementing C<ev_stat> semantics with kqueue).
+implementing C<ev_stat> semantics with kqueue, except as a hint).
 =head3 ABI Issues (Largefile Support)
 Libev by default (unless the user overrides this) uses the default
 compilation environment, which means that on systems with large file
 to exchange stat structures with application programs compiled using the
 default compilation environment.
 =head3 Inotify and Kqueue
-When C<inotify (7)> support has been compiled into libev (generally
+When C<inotify (7)> support has been compiled into libev and present at
-only available with Linux 2.6.25 or above due to bugs in earlier
+runtime, it will be used to speed up change detection where possible. The
-implementations) and present at runtime, it will be used to speed up
+inotify descriptor will be created lazily when the first C<ev_stat>
-change detection where possible. The inotify descriptor will be created
+watcher is being started.
-lazily when the first C<ev_stat> watcher is being started.
 Inotify presence does not change the semantics of C<ev_stat> watchers
 except that changes might be detected earlier, and in some cases, to avoid
 making regular C<stat> calls. Even in the presence of inotify support
 there are many cases where libev has to resort to regular C<stat> polling,
-but as long as the path exists, libev usually gets away without polling.
+but as long as kernel 2.6.25 or newer is used (2.6.24 and older have too
+many bugs), the path exists (i.e. stat succeeds), and the path resides on
+a local filesystem (libev currently assumes only ext2/3, jfs, reiserfs and
+xfs are fully working) libev usually gets away without polling.
 There is no support for kqueue, as apparently it cannot be used to
 implement this functionality, due to the requirement of having a file
 descriptor open on the object at all times, and detecting renames, unlinks
 etc. is difficult.
+=head3 C<stat ()> is a synchronous operation
+Libev doesn't normally do any kind of I/O itself, and so is not blocking
+the process. The exception are C<ev_stat> watchers - those call C<stat
+()>, which is a synchronous operation.
+For local paths, this usually doesn't matter: unless the system is very
+busy or the intervals between stat's are large, a stat call will be fast,
+as the path data is suually in memory already (except when starting the
+watcher).
+For networked file systems, calling C<stat ()> can block an indefinite
+time due to network issues, and even under good conditions, a stat call
+often takes multiple milliseconds.
+Therefore, it is best to avoid using C<ev_stat> watchers on networked
+paths, although this is fully supported by libev.
 =head3 The special problem of stat time resolution
 The C<stat ()> system call only supports full-second resolution portably,
 and even on systems where the resolution is higher, most file systems
 =back
 =head1 AUTHOR
-Marc Lehmann <libev@schmorp.de>.
+Marc Lehmann <libev@schmorp.de>, with repeated corrections by Mikael Magnusson.

Diff Legend

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

Comparing libev/ev.pod (file contents): Revision 1.208 by root, Wed Oct 29 10:24:23 2008 UTC vs. Revision 1.213 by root, Wed Nov 5 02:48:45 2008 UTC

Diff Legend

Comparing libev/ev.pod (file contents):
Revision 1.208 by root, Wed Oct 29 10:24:23 2008 UTC vs.
Revision 1.213 by root, Wed Nov 5 02:48:45 2008 UTC