On 28/07/17 13:59, Peter Zijlstra wrote: > On Fri, Jul 28, 2017 at 01:16:24PM +0100, Dietmar Eggemann wrote: >>>> IIRC the topology you had in mind was MC + DIE level with n (n > 2) DIE >>>> level sched groups.
[...] >>> If I then start a 3rd loop, I see 100% 50%,50%. I then kill the 100%. >>> Then instantly they balance and I get 2x100% back. >> >> Yeah, could reproduce on IVB-EP (2x10x2). > > OK, I have one of those. What should I do, because I didn't actually see > anything odd. Me neither. Sorry, I was unclear. I meant I could reproduce your example, that one of the 50% task moves to the idle cpu on this machine. [...]
