On 08/01/2015 03:53, Fam Zheng wrote: >>> > > 1) Remove 1ns PR_SET_TIMERSLACK in timerfd+epoll, this doesn't make >>> > > qemu_poll >>> > > faster than the old qemu_poll_ns, but may have other positive effects >>> > > that >>> > > compensate the cost. >> > >> > Sounds like a random hack. What is the reasoning for messing with timer >> > slack? > In a test this doesn't work. The reason is that timer slack affects poll > sys calls' timeout, therefore they are correlated. Anyway, I've left this.
1ns is definitely too much, it's just accuracy theater (and I am responsible for it). We don't care about 3 clock cycles when a context switch is at least 500. We probably could change it to 100 ns (0.1 us) with no practical effects; the default is 50 us. Paolo