Messages in this thread | | | Date | Fri, 14 Mar 2003 17:59:48 +0200 | From | Valentin Nechayev <> | Subject | Re: [patch, rfc] lt-epoll ( level triggered epoll ) ... |
| |
Tue, Mar 11, 2003 at 14:27:50, jamie wrote about "Re: [patch, rfc] lt-epoll ( level triggered epoll ) ...":
> Actually I think _this_ is cleanest: A three-way flag per registered > fd interest saying whether to: > > 1. Report 0->1 edges for this interest. (Initial 1 counts as an event). > 2. Continually report 1 levels for this interest. > 3. One-shot, report the first time 1 is noted and unregister. > > ET poll is equivalent to 1. LT poll is equivalent to 2. dnotify's > one-shot mode is equivalent to 3.
kqueue can do all three variants (1st with EV_CLEAR, 3rd with EV_ONESHOT).
So, result of this whole epoll work is trivially predictable - Linux will have analog of "overbloated" and "poorly designed" kqueue, but more poor and with incompatible interface, adding its own stone to hell of different APIs. Congratulations.
Linus was true: Linux is just for fun, not for work.
I say nothing bad for the real work to implement it. But the person who said "Do poor, do incompatible, but do our own" should be blamed. You know him.
-netch- - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
| |