Julien Forget

Scheduling Dependent Periodic Tasks without Synchronization Mechanisms

This article studies the scheduling of critical embedded systems, which consist of a set of communicating periodic tasks with constrained deadlines. Currently, tasks are usually sequenced manually, partly because available scheduling policies do not ensure the determinism of task communications. Ensuring this determinism requires scheduling policies supporting task precedence constraints (which we call dependent tasks), which are used to force the order in which communicating tasks execute. We propose fixed priority scheduling policies for different classes of dependent tasks: with simultaneous or arbitrary release times, with simple precedences (between tasks of the same period) or extended precedences (between tasks of different periods). We only consider policies that do not require synchronization mechanisms (like semaphores). This completely prevents deadlocks or scheduling anomalies without requiring further proofs.

Multi-task Implementation of Multi-periodic Synchronous Programs

This article presents a complete scheme for the integration and the development of multi-periodic critical embedded systems. A system is formally specified as a modular and hierarchical assembly of several locally mono-periodic synchronous functions into a globally multi-periodic synchronous system. To support this, we introduce a real-time software architecture description language, named Prelude, which is built upon the synchronous languages and which provides a high level of abstraction for describing the functional and the real-time architecture of a multi-periodic control system. A program is translated into a set of real-time tasks that can be executed on a monoprocessor real-time platform with an on-line priority-based scheduler such as Deadline-Monotonic or Earliest-Deadline-First. The compilation is formally proved correct, meaning that the generated code respects the real-time semantics of the original program (respect of periods, deadlines, release dates and precedences) as well as its functional semantics (respect of variable consumption).

A Multi-Periodic Synchronous Data-Flow Language

Implementing real-time critical systems is an increasingly complex process that calls for high-level formal programming languages. Existing languages mainly focus on mono-periodic systems, implementing multi-periodic systems with these languages is possible but inefficient. As a result, current practice usually consists in writing one program for each different rate and then letting a real-time operating system handle the multi-rate aspects. This can be a source of non-determinism as communications between processes of different rates are not precisely defined. We propose a new language, built upon synchronous data-flow languages, to handle multi-rate systems properly. It has strong formal semantics, which prevents non-deterministic communications, and relies on real-time primitives that enable efficient use of existing multi-periodic schedulers.

A real-time architecture design language for multi-rate embedded control systems

This paper presents a language dedicated to the description of the software architecture of complex embedded control systems. The language relies on the synchronous approach but extends it to support efficiently systems with multiple real-time constraints, such as deadline constraints or periodicity constraints. It provides a high-level of abstraction and benefits from the formal properties of synchronous languages. The language defines a small set of rate transition operators, which enable the description of user-defined deterministic multi-rate communication patterns between components of different rates. The compiler of the language automatically translates a program into a set of communicating real-time tasks implemented as concurrent C threads that can be executed on a standard real-time operating system.

Minimizing a real-time task set through Task Clustering

In the industry, real-time systems are specified as a set of hundreds of functionalities with timing constraints. Implementing those functionalities as threads in a one-to-one relation is not realistic due to the overhead caused by the large number of threads. In this paper, we present task clustering, which aims at minimizing the number of threads while preserving the schedulability. We prove that our clustering problem is NP-Hard and describe a heuristic to tackle it. Our approach has been applied to fixed-task or fixed-job priority based scheduling policies as Deadline Monotonic (DM) or Earliest Deadline First (EDF).

A heuristic to minimize the cardinality of a real-time task set by automated task clustering

We propose in this paper a method to automatically map functionalities (blocks of code corresponding to high-level features) with real-time constraints to tasks (or threads). We aim at reducing the number of tasks functions are mapped to, while preserving the schedulability of the initial system. We consider independent tasks running on a single processor. Our approach has been applied with fixed-task or fixed-job priorities assigned in a Deadline Monotonic (DM) or a Earliest Deadline First (EDF) manner.

End-to-end latency computation in a multi-periodic design

The specification of a critical real-time application often includes quantitative temporal properties, imposed by the designer, that need to be respected by the implementation. In this paper we focus on end-to-end latency constraints, i.e. the amount of time required before an input is taken into account by the corresponding output. Such applications usually consist of a set of periodic communicating tasks. In this paper, we describe an application using the language Prelude, dedicated to the specification of multi-periodic systems, and show that end-to-end latencies can be computed automatically based on the formal semantics of the language.

Implementing Multi-Periodic Critical Systems: from Design to Code Generation

This article presents a complete scheme for the development of Critical Embedded Systems with Multiple Real-Time Constraints. The system is programmed with a language that extends the synchronous approach with high-level real-time primitives. It enables to assemble in a modular and hierarchical manner several locally mono-periodic synchronous systems into a globally multi-periodic synchronous system. It also allows to specify flow latency constraints. A program is translated into a set of real-time tasks. The generated code (\C\ code) can be executed on a simple real-time platform with a dynamic-priority scheduler (EDF). The compilation process (each algorithm of the process, not the compiler itself) is formally proved correct, meaning that the generated code respects the real-time semantics of the original program (respect of periods, deadlines, release dates and precedences) as well as its functional semantics (respect of variable consumption).

A Synchronous Language with Partial Delay Specification for Real-Time Systems Programming

High-level formal programming languages require system designers to provide a very precise description of the system during early development phases, which may in some cases lead to arbitrary choices (i.e. the designer “overspecifies” the system). In this paper, we propose an extension of synchronous dataflow languages where the designer can specify that he does not care whether some communication is immediate or delayed. It is then up to the compiler to choose where to introduce delays, in a way that breaks causality cycles and satisfies latency requirements imposed on the system.

Symbolic WCET Computation

Parametric Worst-case execution time (WCET) analysis of a sequential program produces a formula that represents the worst-case execution time of the program, where parameters of the formula are user-defined parameters of the program (as loop bounds, values of inputs, or internal variables, etc). In this article we propose a novel methodology to compute the parametric WCET of a program. Unlike other algorithms in the literature, our method is not based on Integer Linear Programming (ILP). Instead, we follow an approach based on the notion of symbolic computation of WCET formulae. After explaining our methodology and proving its correctness, we present a set of experiments to compare our method against the state of the art. We show that our approach dominates other parametric analyses and produces results that are very close to those produced by non-parametric ILP-based approaches, while keeping very good computing time.