Hannes, [Sorry I'm late to the discussion here, had a few fires going in addition to the end of the kernel cycle] > And we shouldn't forget that the main issue which killed all previous > implementations was a missing QoS guarantee. That and the fact that our arbitrary stacking situation was hard to resolve. The QoS guarantee was somewhat addressed by Fred in T10. But I agree we need some sort of toggle. -- Martin K. Petersen Oracle Linux Engineering
