Recheck definition of multiprocessor schedule

I'm not sure if (cpu -> time -> option Job) is the most intuitive definition for global scheduling. When constructing any global schedule, we always compute the set of jobs that should be schedule at time t (horizontal, inductive construction). But we never join complete uniprocessor schedules (vertical construction).

This issue has been (implicitly) resolved by Maxime's way of representing multiprocessor schedules as part of the new generic schedules framework was introduced as part of f210238b.

closed