Viewing profile — pphaneuf

As in "only the 3rd or 4th" (so I'm possibly clever), or as in "a bunch of people"? I don't know your sample size. :-)

Why would there be extra overhead when using edge triggered? There's definitely extra complexity on the client side, but it's close to what you're trying to do with super-poll (the…

Your results seem correct to me, even though I'm a bit disappointed at the actual number (I was hoping for 0.8 or 0.9, not 0.6), and the code looks good. I'm meaning to run it for …

Right on. Myself, I always figured epoll was a bit slower than poll when most/all of the fds are ready all the time, from reading the code. Now, there's an experiment to figure out…

I've never perceived epoll as being O(1), always O(N), the big thing being the N being how many events are dispatched, rather than the number of watched fds. I'm fine with the amou…

Hey, some idea I remembered (from some old paper): don't go from epoll to poll when the ATR gets too high, go to... NOTHING! Just read() or write(), as if something told you they w…

Adding/removing from the poll() array is more or less free, but with epoll, you have to use epoll_ctl(), so you'd definitely want an hysteresis to avoid bouncing things around too …

Question: as the ATR is going higher, so would the proportional time spent in poll or epoll, no? So if you have a thousand fds, and they're all active, you have to deal with a thou…

The best would be if it would be possible to code up superpoll to be adaptive, and in effect, benchmark itself to come to the same conclusion, dynamically. So if one day the kernel…

I think what he's saying is that it should be pretty easy to change between poll, epoll, and "superpoll", as you can easily abstract a common interface. Then, you could worry about…