[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index] Re: [Xen-devel] cpuidle causing Dom0 soft lockups
>>> Keir Fraser <keir.fraser@xxxxxxxxxxxxx> 21.01.10 11:26 >>> >Hm, okay. Another thing to consider is not using ticket locks when running >on Xen. Regardless of slow wakeup times (and you can't really depend on them >being fast, in general, in a virtualised environment) experiments have shown >that, as you might expect, contended ticket locks perform *abysmally* in >virtualised environments. But that's exactly why we use the polling method: That way we don't need to wake up all potential waiters, but just the one that's going to get the lock next. We now even go further and detect whether the lock owner is not currently running, and go polling right away (instead of spinning on the lock a few hundred times first). >Turning off our major source of power-management wins by default in >Xen 4.0 is not really a goer. I can see your point. But how can you consider shipping with something apparently severely broken. As said before - the fact that this manifests itself by hanging many-vCPU Dom0 has the very likely implication that there are (so far unnoticed) problems with smaller Dom0-s. If I had a machine at hand that supports C3, I'd try to do some measurements with smaller domains... Jan _______________________________________________ Xen-devel mailing list Xen-devel@xxxxxxxxxxxxxxxxxxx http://lists.xensource.com/xen-devel
|
Lists.xenproject.org is hosted with RackSpace, monitoring our |