Multi-core Code Generation from Polychronous Programs with Time-Predictable Properties (ACVI 2014) - Archive ouverte HAL Accéder directement au contenu
Communication Dans Un Congrès Année : 2014

Multi-core Code Generation from Polychronous Programs with Time-Predictable Properties (ACVI 2014)

Résumé

Synchronous programming models capture concurrency in computation quite naturally, especially in its dataflow multi-clock (polychronous) flavor. With the rising importance of multi-core processors in safety-critical embedded systems or cyber-physical systems (CPS), there is a growing need for model-driven generation of multi-threaded code for multi-core systems. This paper proposes a build method of timepredictable system on multi-core, based on synchronous-model development. At the modeling level, the synchronous abstraction allows deterministic time semantics. Thus synchronous programming is a good choice for time-predictable system design. At the compiler level, the verified compiler from the synchronous language SIGNAL to our intermediate representation (S-CGA, a variant of guarded actions) and to multi-threaded code, preserves the time predictability. At the platform level, we propose a time-predictable multi-core architecture model in AADL (Architecture Analysis and Design Language), and then we map the multi-threaded code to this model. Therefore, our method integrates time predictability across several design layers.
Fichier principal
Vignette du fichier
Yang_13210.pdf (1.39 Mo) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)
Loading...

Dates et versions

hal-01150341 , version 1 (11-05-2015)

Licence

Paternité

Identifiants

  • HAL Id : hal-01150341 , version 1
  • OATAO : 13210

Citer

Zhibin Yang, Jean-Paul Bodeveix, M Filali. Multi-core Code Generation from Polychronous Programs with Time-Predictable Properties (ACVI 2014). 1st International Workshop on Architecture Centric Virtual Integration (ACVI 2014), Sep 2014, Valencia, Spain. pp.1-10. ⟨hal-01150341⟩
143 Consultations
46 Téléchargements

Partager

Gmail Facebook X LinkedIn More