… | |
… | |
483 | =item C<EVBACKEND_EPOLL> (value 4, Linux) |
483 | =item C<EVBACKEND_EPOLL> (value 4, Linux) |
484 | |
484 | |
485 | Use the linux-specific epoll(7) interface (for both pre- and post-2.6.9 |
485 | Use the linux-specific epoll(7) interface (for both pre- and post-2.6.9 |
486 | kernels). |
486 | kernels). |
487 | |
487 | |
488 | For few fds, this backend is a bit little slower than poll and select, |
488 | For few fds, this backend is a bit little slower than poll and select, but |
489 | but it scales phenomenally better. While poll and select usually scale |
489 | it scales phenomenally better. While poll and select usually scale like |
490 | like O(total_fds) where n is the total number of fds (or the highest fd), |
490 | O(total_fds) where total_fds is the total number of fds (or the highest |
491 | epoll scales either O(1) or O(active_fds). |
491 | fd), epoll scales either O(1) or O(active_fds). |
492 | |
492 | |
493 | The epoll mechanism deserves honorable mention as the most misdesigned |
493 | The epoll mechanism deserves honorable mention as the most misdesigned |
494 | of the more advanced event mechanisms: mere annoyances include silently |
494 | of the more advanced event mechanisms: mere annoyances include silently |
495 | dropping file descriptors, requiring a system call per change per file |
495 | dropping file descriptors, requiring a system call per change per file |
496 | descriptor (and unnecessary guessing of parameters), problems with dup, |
496 | descriptor (and unnecessary guessing of parameters), problems with dup, |
… | |
… | |
499 | 0.1ms) and so on. The biggest issue is fork races, however - if a program |
499 | 0.1ms) and so on. The biggest issue is fork races, however - if a program |
500 | forks then I<both> parent and child process have to recreate the epoll |
500 | forks then I<both> parent and child process have to recreate the epoll |
501 | set, which can take considerable time (one syscall per file descriptor) |
501 | set, which can take considerable time (one syscall per file descriptor) |
502 | and is of course hard to detect. |
502 | and is of course hard to detect. |
503 | |
503 | |
504 | Epoll is also notoriously buggy - embedding epoll fds I<should> work, but |
504 | Epoll is also notoriously buggy - embedding epoll fds I<should> work, |
505 | of course I<doesn't>, and epoll just loves to report events for totally |
505 | but of course I<doesn't>, and epoll just loves to report events for |
506 | I<different> file descriptors (even already closed ones, so one cannot |
506 | totally I<different> file descriptors (even already closed ones, so |
507 | even remove them from the set) than registered in the set (especially |
507 | one cannot even remove them from the set) than registered in the set |
508 | on SMP systems). Libev tries to counter these spurious notifications by |
508 | (especially on SMP systems). Libev tries to counter these spurious |
509 | employing an additional generation counter and comparing that against the |
509 | notifications by employing an additional generation counter and comparing |
510 | events to filter out spurious ones, recreating the set when required. Last |
510 | that against the events to filter out spurious ones, recreating the set |
|
|
511 | when required. Epoll also errornously rounds down timeouts, but gives you |
|
|
512 | no way to know when and by how much, so sometimes you have to busy-wait |
|
|
513 | because epoll returns immediately despite a nonzero timeout. And last |
511 | not least, it also refuses to work with some file descriptors which work |
514 | not least, it also refuses to work with some file descriptors which work |
512 | perfectly fine with C<select> (files, many character devices...). |
515 | perfectly fine with C<select> (files, many character devices...). |
513 | |
516 | |
514 | Epoll is truly the train wreck analog among event poll mechanisms, |
517 | Epoll is truly the train wreck among event poll mechanisms, a frankenpoll, |
515 | a frankenpoll, cobbled together in a hurry, no thought to design or |
518 | cobbled together in a hurry, no thought to design or interaction with |
516 | interaction with others. |
519 | others. Oh, the pain, will it ever stop... |
517 | |
520 | |
518 | While stopping, setting and starting an I/O watcher in the same iteration |
521 | While stopping, setting and starting an I/O watcher in the same iteration |
519 | will result in some caching, there is still a system call per such |
522 | will result in some caching, there is still a system call per such |
520 | incident (because the same I<file descriptor> could point to a different |
523 | incident (because the same I<file descriptor> could point to a different |
521 | I<file description> now), so its best to avoid that. Also, C<dup ()>'ed |
524 | I<file description> now), so its best to avoid that. Also, C<dup ()>'ed |
… | |
… | |
825 | This is useful if you are waiting for some external event in conjunction |
828 | This is useful if you are waiting for some external event in conjunction |
826 | with something not expressible using other libev watchers (i.e. "roll your |
829 | with something not expressible using other libev watchers (i.e. "roll your |
827 | own C<ev_run>"). However, a pair of C<ev_prepare>/C<ev_check> watchers is |
830 | own C<ev_run>"). However, a pair of C<ev_prepare>/C<ev_check> watchers is |
828 | usually a better approach for this kind of thing. |
831 | usually a better approach for this kind of thing. |
829 | |
832 | |
830 | Here are the gory details of what C<ev_run> does: |
833 | Here are the gory details of what C<ev_run> does (this is for your |
|
|
834 | understanding, not a guarantee that things will work exactly like this in |
|
|
835 | future versions): |
831 | |
836 | |
832 | - Increment loop depth. |
837 | - Increment loop depth. |
833 | - Reset the ev_break status. |
838 | - Reset the ev_break status. |
834 | - Before the first iteration, call any pending watchers. |
839 | - Before the first iteration, call any pending watchers. |
835 | LOOP: |
840 | LOOP: |
… | |
… | |
2151 | |
2156 | |
2152 | Another way to think about it (for the mathematically inclined) is that |
2157 | Another way to think about it (for the mathematically inclined) is that |
2153 | C<ev_periodic> will try to run the callback in this mode at the next possible |
2158 | C<ev_periodic> will try to run the callback in this mode at the next possible |
2154 | time where C<time = offset (mod interval)>, regardless of any time jumps. |
2159 | time where C<time = offset (mod interval)>, regardless of any time jumps. |
2155 | |
2160 | |
2156 | For numerical stability it is preferable that the C<offset> value is near |
2161 | The C<interval> I<MUST> be positive, and for numerical stability, the |
2157 | C<ev_now ()> (the current time), but there is no range requirement for |
2162 | interval value should be higher than C<1/8192> (which is around 100 |
2158 | this value, and in fact is often specified as zero. |
2163 | microseconds) and C<offset> should be higher than C<0> and should have |
|
|
2164 | at most a similar magnitude as the current time (say, within a factor of |
|
|
2165 | ten). Typical values for offset are, in fact, C<0> or something between |
|
|
2166 | C<0> and C<interval>, which is also the recommended range. |
2159 | |
2167 | |
2160 | Note also that there is an upper limit to how often a timer can fire (CPU |
2168 | Note also that there is an upper limit to how often a timer can fire (CPU |
2161 | speed for example), so if C<interval> is very small then timing stability |
2169 | speed for example), so if C<interval> is very small then timing stability |
2162 | will of course deteriorate. Libev itself tries to be exact to be about one |
2170 | will of course deteriorate. Libev itself tries to be exact to be about one |
2163 | millisecond (if the OS supports it and the machine is fast enough). |
2171 | millisecond (if the OS supports it and the machine is fast enough). |
… | |
… | |
4204 | F<event.h> that are not directly supported by the libev core alone. |
4212 | F<event.h> that are not directly supported by the libev core alone. |
4205 | |
4213 | |
4206 | In standalone mode, libev will still try to automatically deduce the |
4214 | In standalone mode, libev will still try to automatically deduce the |
4207 | configuration, but has to be more conservative. |
4215 | configuration, but has to be more conservative. |
4208 | |
4216 | |
|
|
4217 | =item EV_USE_FLOOR |
|
|
4218 | |
|
|
4219 | If defined to be C<1>, libev will use the C<floor ()> function for its |
|
|
4220 | periodic reschedule calculations, otherwise libev will fall back on a |
|
|
4221 | portable (slower) implementation. If you enable this, you usually have to |
|
|
4222 | link against libm or something equivalent. Enabling this when the C<floor> |
|
|
4223 | function is not available will fail, so the safe default is to not enable |
|
|
4224 | this. |
|
|
4225 | |
4209 | =item EV_USE_MONOTONIC |
4226 | =item EV_USE_MONOTONIC |
4210 | |
4227 | |
4211 | If defined to be C<1>, libev will try to detect the availability of the |
4228 | If defined to be C<1>, libev will try to detect the availability of the |
4212 | monotonic clock option at both compile time and runtime. Otherwise no |
4229 | monotonic clock option at both compile time and runtime. Otherwise no |
4213 | use of the monotonic clock option will be attempted. If you enable this, |
4230 | use of the monotonic clock option will be attempted. If you enable this, |
… | |
… | |
5224 | The physical time that is observed. It is apparently strictly monotonic :) |
5241 | The physical time that is observed. It is apparently strictly monotonic :) |
5225 | |
5242 | |
5226 | =item wall-clock time |
5243 | =item wall-clock time |
5227 | |
5244 | |
5228 | The time and date as shown on clocks. Unlike real time, it can actually |
5245 | The time and date as shown on clocks. Unlike real time, it can actually |
5229 | be wrong and jump forwards and backwards, e.g. when the you adjust your |
5246 | be wrong and jump forwards and backwards, e.g. when you adjust your |
5230 | clock. |
5247 | clock. |
5231 | |
5248 | |
5232 | =item watcher |
5249 | =item watcher |
5233 | |
5250 | |
5234 | A data structure that describes interest in certain events. Watchers need |
5251 | A data structure that describes interest in certain events. Watchers need |