arXiv subjects

Thomas Carle

Publications and source records attributed to Thomas Carle.

4 recordsLinked to original sources

arXiv2026

Convolutions Predictable Offloading to an Accelerator: Formalization and Optimization

Convolutional neural networks (CNNs) require a large number of multiply-accumulate (MAC) operations. To meet real-time constraints, they often need to be executed on specialized accelerators composed of an on-chip memory and a processing unit. However, the on-chip memory is often insufficient to store all the data required to compute a CNN layer. Thus, the computation must be performed in several offloading steps. We formalise such sequences of steps and apply our formalism to a state of the art decomposition of convolutions. In order to find optimal strategies in terms of duration, we encode the problem with a set of constraints. A Python-based simulator allows to analyse in-depth computed strategies.

cs.AR↗

arXiv2026

Extension of ACETONE C code generator for multi-core architectures

As the industry's interest in machine learning has grown in recent years, some solutions have emerged to safely embed them in safety-critical systems, such as the C code generator ACETONE. However, this framework is limited to generating sequential code, which cannot make most of the multi-core architectures. In this paper, we initiate an extension of ACETONE for the generation of parallel code by formally defining our processor assignment problem and surveying the state of the art on existing solutions. In the final paper, we will introduce the completed extension, including the implementation of the scheduling heuristic, the creation of templates implementing synchronization mechanisms, and an evaluation of the worst-case execution time of the framework's layers.

cs.DC↗

arXiv2022

Computing Execution Times with eXecution Decision Diagrams in the Presence of Out-Of-Order Resources

Worst-Case Execution Time (WCET) is a key component for the verification of critical real-time applications. Yet, even the simplest microprocessors implement pipelines with concurrently-accessed resources, such as the memory bus shared by fetch and memory stages. Although their in-order pipelines are, by nature, very deterministic, the bus can cause out-of-order accesses to the memory and, therefore, timing anomalies: local timing effects that can have global effects but that cannot be easily composed to estimate the global WCET. To cope with this situation, WCET analyses have to generate important over-estimations in order to preserve safety of the computed times or have to explicitly track all possible executions. In the latter case, the presence of out-of-order behavior leads to a combinatorial blowup of the number of pipeline states for which efficient state abstractions are difficult to design. This paper proposes instead a compact and exact representation of the timings in the pipeline, using eXecution Decision Diagram (XDD) [1]. We show how XDD can be used to model pipeline states all along the execution paths by leveraging the algebraic properties of XDD. This computational model allows to compute the exact temporal behavior at control flow graph level and is amenable to efficiently and precisely support WCET calculation in presence of out-of-order bus accesses. This model is finally experimented on the TACLe benchmark suite and we observe good performance making this approach appropriate for industrial applications.

eess.SY↗

arXiv2021

Static extraction of memory access profiles for multi-core interference analysis of real-time tasks

We present a static analysis framework for real-time task systems running on multi-core processors. Our method analyzes tasks in isolation at the binary level and generates worst-case timing and memory access profiles. These profiles can then be combined to perform an interference analysis at the task system level, as part of a multi-core Worst-Case Response Time (WCRT) analysis. In this paper we introduce a formal description of the models and algorithmic building blocks composing our framework. We also discuss how the memory access profiles generated by our method could be used to feed existing state-of-the-art WCRT frameworks. To the best of our knowledge, it is the first time that a method is documented on how to produce sound, safe and precise inputs for interference analysis methods.

eess.SY↗