Section: New Results

Parallel Code Generation of Synchronous Programs for a Many-core Architecture

Participant : Matthieu Moy.

Embedded systems tend to require more and more computational power. Many-core architectures are good candidates since they offer power and are considered more time predictable than classical multi-cores.

Data-flow Synchronous languages such as Lustre or Scade are widely used for avionic critical software. Programs are described by networks of computational nodes. Implementation of such programs on a many-core architecture must ensure a bounded response time and preserve the functional behavior by taking interference into account.

We consider the top-level node of a Lustre application as a software architecture description where each sub-node corresponds to a potential parallel task. Given a mapping (tasks to cores), we automatically generate code suitable for the targeted many-core architecture. This code uses hardware synchronization mechanisms and time-triggered execution. This minimizes memory interferences and allows usage of a framework to compute the Worst-Case Response Time.

This work was accepted for publication at the DATE 2018 conference  [82].