Synchronous Deterministic Parallel Programming for Multi-Cores with ForeC

Embedded real-time systems are tightly integrated with their physical environment. Their correctness depends both on the outputs and timeliness of computations. The increasing use multi-core processors in such is pushing embedded programmers to be parallel programming experts. However, challenging because skills, experiences, knowledge needed avoid common traps pitfalls. This article proposes ForeC synchronous multi-threaded language for deterministic, parallel, reactive multi-cores. semantics designed greatly simplify understanding debugging programs. ensures that programs can compiled efficiently execution amenable static timing analysis. ForeC’s main innovation its shared variable provides thread isolation deterministic communication. All correct by construction deadlock free no non-deterministic constructs needed. We have benchmarked our compiler several medium-sized (e.g., a 2.274-line program up 26 threads distributed 10 cores, which was based 2.155-line non-multi-threaded C program). These benchmark show achieve better performance than Esterel, widely used imperative concurrent safety-critical systems, competitive OpenMP, popular desktop solution (which implements classical multi-threading, hence intrinsically non-deterministic). also demonstrate worst-case time estimated high degree precision.