Task placement and selection of data consistency mechanisms for real-time multicore applications

Abstract

Multicores are today used in several automotive, controls and avionics systems supporting real-time functionality. When real-time tasks allocated on different cores cooperate through the use of shared communication resources, they need to be protected by mechanisms that guarantee access in a mutual exclusive way with bounded worst-case blocking time. Lock-based mechanisms such as MPCP and MSRP have been developed to fulfill this demand, and research papers are today tackling the problem of finding the optimal task placement in multicores while trying to meet the deadlines against blocking times. In this work, we provide a method that improves on existing task placement algorithms for systems that use MSRP to protect shared resources. Furthermore, we leverage an additional opportunity provided by wait-free methods as an alternative data consistency mechanism in those cases where the shared resource is communication or state memory. The selective use of waitfree communication can further extend the range of schedulable systems and consequently the design space, at the cost of memory.

Weakly Hard Schedulability Analysis for Fixed Priority Scheduling of Periodic Real-Time Tasks

“…A possible extension of the approach proposed in this paper could target multiprocessor systems with partitioned scheduling and shared resources as applicable to several automotive applications [1,32].…”

A model-based synthesis flow for automotive CPS

“…The constraints on block-to-runnable mapping are (5)(6)(7)(8). Constraints (9-11) define the causality relations among tasks, and (12) requires that the block-to-runnable mapping does not introduce any false input-output dependencies.…”

“…The analysis of real-time communication mechanisms based on spin locks for AUTOSAR runnables in multicores is presented in [14], and mechanisms for a flowpreserving AUTOSAR implementation of synchronous signals are presented and compared in [13]. The optimal placement of tasks with deadline constraints communicating with locking primitives in multicores is discussed in [24,5], with an MILP solution and e↵ective heuristics presented.…”

“…The first task synthesis algorithm is a search heuristic extended from the solution presented for single-core platforms in [7] and the allocation heuristics in [24], [5]. The algorithm conducts a greedy allocation with limited backtrack, and has a post-processing step that tries to further improve the solution.…”

Multiprocessor Real-Time Locking Protocols: A Systematic Review

“…Over the years, many results pertaining to the task and resource mapping problems as well as related optimization problems have appeared [13,34,70,76,87,88,99,100,101,111,112,115,134,149,152,154,168,180,189,195,198,208]. Particularly well-known is Lakshmanan et al's task-set partitioning heuristic for use with the MPCP [123].…”