On Fri, 2010-12-03 at 19:00 +0530, Srivatsa Vaddagiri wrote: > On Fri, Dec 03, 2010 at 02:23:39PM +0100, Peter Zijlstra wrote: > > Right, so another approach might be to simply swap the vruntime between > > curr and p. > > Can't that cause others to stave? For ex: consider a cpu p0 having these tasks: > > p0 -> A0 B0 A1 > > A0/A1 have entered some sort of AB<->BA spin-deadlock, as a result A0 wants to > direct yield to A1 which wants to direct yield to A0. If we keep swapping their > runtimes, can't it starve B0? No, because they do receive service (they spend some time spinning before being interrupted), so the respective vruntimes will increase, at some point they'll pass B0 and it'll get scheduled. -- 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
