On Wed, 17 Jun 2009, Gregory Haskins wrote: > Davide Libenzi wrote: > > > How much the (possible, but not certain) kernel thread context switch time > > weighs in the overall KVM IRQ service time? > > > > Generally each one is costing me about 7us on average. For something > like high-speed networking, we have a path that has about 30us of > base-line overhead. So one additional ctx-switch puts me at base+7 ( = > ~37us), two puts me in base+2*7 (= ~44us). So in that context (no pun > intended ;), it hurts quite a bit. I'll be the first to admit that not > everyone (most?) will care about latency, though. But FWIW, I do. And how a frame reception is handled in Linux nowadays? > True, but thats the notifiee's burden, not eventfd's. And its always > going to be opt-in. Even today, someone is free to either try to sleep > (which will oops on the might_sleep()), ... No, today you just can't sleep. As you can't sleep in any callback-registered wakeups, like epoll, for example. - Davide -- To unsubscribe from this list: send the line "unsubscribe kvm" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
