Real-time scheduling algorithms for multiprocessor systems have been the subject of considerable recent interest. For such an algorithm to be truly useful in practice, support for semaphore-based locking must be provided. However, for many global scheduling algorithms, no such mechanisms have been proposed. Furthermore, in the partitioned case, most prior semaphore schemes are either inefficient or restrict critical sections considerably. In this paper, a new flexible multiprocessor locking scheme is presented that can be applied under both partitioning and global scheduling. This scheme allows unrestricted critical-section nesting, but has been designed to deal with the common case of short non-nested accesses efficiently.

This paper describes a real-time extension to Linux calledLITMUS RT, whichisbeingdesignedtosupport real-timeworkloadsonmultiprocessorandmulticore platforms. Therecentshiftbychipmakerstomulticoredesigns, combinedwithbuildinginterestwithinthe open-sourcecommunityinsupportingreal-timefeatures inLinux,makesthisresearchquitetimely. ThedevelopmentofLITMUS RT wasdrivenbyadesiretobridge thegapbetweenthoseworkingonalgorithmicissues pertainingtomultiprocessorreal-timeresourceallocationand operating-systems researchers working to improve real-time support within operating systems such as Linux.

CONTENTS DEDICATION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . iii ACKNOWLEDGMENTS . . . . . . . . . . . . . . . . . . . . . . . . . . . iv LIST OF TABLES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . viii LIST OF FIGURES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ix SUMMARY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xii I INTRODUCTION . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 1.1 Problem Statement . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 1.2 Thesis Contributions . . . . . . . . . . . . . . . . . . . . . . . . . . 3 1.3 Thesis Organization and Roadmap . . . . . . . . . . . . . . . . . . . 5 II BACKGROUND AND PREVIOUS WORK . . . . . . . . . . . . 6 2.1 Locking Schemes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6 2.1.1 Hardware Instructions for Locking . . . . . . . . . . . . . . . 8 2.1.2 Traditional Spin-Lock . . . . . . . . . . . . . . . . . . . . . . 1

In the domain of multiprocessor real-time systems, there hasbeen a wealth of recent work on scheduling, but relatively little work on the equally-important topic of synchronization.When synchronizing accesses to shared resources, four basic options exist: lock-free execution, wait-free execution, spin-based locking, and suspension-based locking. To our knowledge, no empirical multiprocessor-based evaluation of thesebasic techniques that focuses on real-time systems has ever been conducted before. In this paper, we present such anevaluation and report on our efforts to incorporate synchronization support in the testbed used in this effort.

When locking protocols are used in real-time systems, bounds on blocking times are required when ensuring timing constraints. While the term “blocking ” is well-understood in the context of uniprocessor real-time systems, the same is not true in the multiprocessor case. In this paper, two definitions of blocking are presented that are applicable to suspensionbased multiprocessor locking protocols. The need for two definitions arises because of differences in how suspensions are handled in existing schedulability analysis. For each definition, locking protocols are presented that have asymptotically optimal blocking behavior. In particular, protocols are presented for any job-level static-priority global or partitioned scheduling algorithm.

This paper presents the first suspension-based real-time locking protocols for clustered schedulers. Such schedulers pose challenges from a locking perspective because they exhibit aspects of both partitioned and global scheduling, which seem to necessitate fundamentally different means for bounding priority inversions. A new mechanism to bound such inversions, termed priority donation, is presented and used to derive protocols for mutual exclusion, reader-writer exclusion, and k-exclusion. Each protocol has asymptotically optimal blocking bounds under certain analysis assumptions. The latter two protocols are also the first of their kind for the special cases of global and partitioned scheduling.

We extend the FMLP to partitioned static-priority scheduling and derive corresponding worst-case blocking bounds. Further, we present the first implementation of the PCP, SRP, D-PCP, M-PCP, and FMLP synchronization protocols in a unified framework in a general-purpose OS and discuss design issues that were beyond the scope of prior algorithmic-oriented work on real-time synchronization.

This paper presents the first real-time multiprocessor locking protocol that supports fine-grained nested resource requests. This locking protocol relies on a novel technique for ordering the satisfaction of resource requests to ensure a bounded duration of priority inversions for nested requests. This technique can be applied on partitioned, clustered, and globally scheduled systems in which waiting is realized by either spinning or suspending. Furthermore, this technique can be used to construct fine-grained nested locking protocols that are efficient under spin-based, suspension-oblivious or suspension-aware analysis of priority inversions. Locking protocols built upon this technique perform no worse than coarse-grained locking mechanisms, while allowing for increased parallelism in the average case (and, depending upon the task set, better worst-case performance). 1

System Model and Assumptions Our system model is a real-time system consisting of a set of processors and a set of periodic tasks. The system uses off-line scheduling to guarantee to meet the timing constraints of all tasks and adopts rollback or replication technique as fault-tolerance strategy. Let the elemental unit (EU) be the smallest non-preemptable execution unit [MA91]. Each EU is characterized by its release time, execution time, and deadline. Each task can be described by an elemental unit graph (EUG), where an EUG is a directed acyclic graph; each node is an EU and each directed edge represents the precedence relation and communication pattern. Since tasks are periodic, the window size for examining the execution behavior of the cyclic tasks can be the least common multiple (LCM) of the periods of all the tasks. We define such time interval as a frame. Let minframe of an EU be the time interval between its release time and its deadline. We adopt a fault model which permits ...

The Concurrency Control in Real Time Database Systems has to ensure not only the consistency of the data in the multi-user environments like in traditional databases, but also the temporal consistency of the data, and that all transactions meet their deadline. Therefore, the Concurrency Control in Real Time Database Systems (RTDBS) is much more complicated than the Concurrency Control in traditional database systems (DBS). A comprehensive model for the concurrency control and formal technique for reasoning about its behaviour will support the design, analysis and veri cation of concurrency control protocols. In this paper, we propose such a model based on Duration Calculus. With our formal model, we give a formal veri cation of some conditions for maintaining the temporal consistency of the data, and a formal veri cation of the correctness of the Read/Write Priority Ceiling Protocol (R/WPCP) for the concurrency control in real-time database systems.