We present several locking synchronization protocols for Pfair-scheduled multiprocessor systems. We focus on two classes of protocols. The first class is only applicable in systems in which all critical sections are short relative to the length of the scheduling quantum. In this case, efficient synchronization can be achieved by ensuring that all locks have been released before tasks are preempted. This is accomplished by exploiting the quantum-based nature of Pfair scheduling, which provides a priori knowledge of all possible preemption points. The second and more general protocol class is applicable to any system. For this class, we consider the use of a client-server model. We also discuss the viability of inheritance-based protocols in Pfair-scheduled systems.

We consider the implementation of a Pfair scheduler on a symmetric multiprocessor (SMP). Simulations presented herein suggest that bus contention resulting from simultaneous scheduling decisions can substantially degrade performance. To correct this problem, we propose a staggered model for Pfair scheduling under which scheduling points are uniformly distributed over time.

We revisit the problem of supertasking in Pfair-scheduled multiprocessor systems by presenting a generalized "reweighting" algorithm. The generalized algorithm we present breaks new ground by permitting tasks to have noninteger execution costs, by incorporating blocking terms into the analysis, and by assuming a more flexible global-scheduling model. To demonstrate the efficacy of the supertasking approach, we present an experimental evaluation of our algorithm that suggests that reweighting may often result in almost no schedulability loss in practice.

We consider the problem of object sharing in Pfair-scheduled multiprocessor systems. We primarily focus on systems that use lock-free shared objects, although some lock-based alternatives are briefly considered as well. Our work demonstrates that the tight synchrony that exists in Pfair-scheduled systems can be exploited to reduce object-sharing overheads when lock-free objects are used.

We propose two server implementations for multiplexing aperiodic and recurrent real-time tasks in fair-scheduled multiprocessor systems. We also provide admission-control tests for scheduling hard aperiodic tasks. Further, we consider several complexities, most of which arise because of the parallelism available in multiprocessor systems, and present techniques to handle them. Finally, we present experimental results that demonstrate the effectiveness of our implementations.

We consider two classes of locking synchronization protocols for pfair-scheduled multiprocessor systems: short critical-section protocols and long critical-section protocols. For the former class, we demonstrate that efficient synchronization can be achieved by ensuring that all locks have been released before tasks are preempted. For the latter class, we propose the use of statically-weighted resource servers. We also discuss several inheritance-based protocols as possible alternatives.

We reconsider the "supertask" approach, in which a set of Pfair tasks is scheduled as a single task. We define a "safe" weight threshold for a given supertask in a multiprocessor system with either hard or soft deadlines and either strict or relaxed rate constraints.