Locking under Pfair scheduling

Abstract

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.

Analysis and implementation of the multiprocessor bandwidth inheritance protocol

“…As it comes to sharing resources in reservation-based systems, the first proposals were made by Caccamo and Sha [10], by Niz et al [40] and by Holman and Anderson [26]. Regarding hierarchical systems 2 , Behnam et al [3] and Fisher et al [21] pro-posed specific protocols to deal with shared resources.…”

A Flexible Real-Time Locking Protocol for Multiprocessors

“…In order to adapt the FMLP for use under PD 2 , two changes to the GSN-EDF variant of the FMLP are required. First, if a short resource request R cannot be complete by the end of the scheduling quantum, then R cannot be issued until the next scheduling quantum; this is a technique first suggested by Holman and Anderson in [16]. Second, if T j i is directly blocked by T b a , then T j i does not suspend, but rather, whenever T j i is scheduled, T b a can execute instead (thus allowing T b a to "inherit" T j i 's scheduling allocations).…”

“…Also, accesses of global critical sections are required to be non-nested. In work involving global scheduling, approaches for implementing non-nested locks have been presented by Holman and Anderson [16] for PD 2 and by Devi et al [12] for G-EDF.…”

Sirap

“…However this is off the point of this paper. 2 The idea of self-blocking has been also considered in different contexts, for example, in CBS-R [6] and zone based protocol (ZB) [12]. Our work is different from those in the sense that CBS-R used a similar idea for supporting soft real-time tasks, and ZB used it in a pfair-scheduling environment, while we use it for hard real-time tasks under hierarchical scheduling.…”