The high level of heterogeneity of modern embedded systems forces designers to use different computational models and formalisms, thus making reuse and integration very diffcult tasks. Reducing such an heterogeneity to a homogeneous implementation is a key solution to allow both simulation and validation of the system. Furthermore, the implementation may be executed on highly optimized architectures or used as a starting point for re-design flows. This paper proposes two novel flows to gain a homogeneous implementation of a starting heterogeneous system, thus showing how heterogeneity can be reconciled to a single language, still preserving correctness. The target languages are SystemC-AMS, that enhances support for continuous behaviors and allows complete validation, and C++, a executable implementation that can be the starting point of re-design fows. The approaches are compared with respect to state-of-the-art techniques in terms of performance and accuracy, also through the application to a complex case study.
Code Generation Alternatives to Reduce Heterogeneous Embedded Systems to Homogeneity
FUMMI, Franco;LORA, MICHELE;STEFANNI, Francesco;
2014-01-01
Abstract
The high level of heterogeneity of modern embedded systems forces designers to use different computational models and formalisms, thus making reuse and integration very diffcult tasks. Reducing such an heterogeneity to a homogeneous implementation is a key solution to allow both simulation and validation of the system. Furthermore, the implementation may be executed on highly optimized architectures or used as a starting point for re-design flows. This paper proposes two novel flows to gain a homogeneous implementation of a starting heterogeneous system, thus showing how heterogeneity can be reconciled to a single language, still preserving correctness. The target languages are SystemC-AMS, that enhances support for continuous behaviors and allows complete validation, and C++, a executable implementation that can be the starting point of re-design fows. The approaches are compared with respect to state-of-the-art techniques in terms of performance and accuracy, also through the application to a complex case study.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.