Synthèse comportementale basée sur l'ordonnancement

Abstract : The objective of this thesis is the elaboration of a new design methodology for the behavioral synthesis. Until now, the behavioral synthesis has been viewed as the process of synthetizing a datapath/controller from an algorithmic specification. This synthesis flows generally includes several steps: the scheduling, the allocation, the datapath generation and the controller synthesis. While all these phases have been extensively studied in the EDA domain, the integration of the behavioral synthesis tools in the existing synthesis environments has not yet got the adaquate attention. And despite a strong interest and demand, the current behavioral synthesis has never been widely adopted by the designers. Solving this problem requires a redefinition of the behavioral synthesis. This thesis proposes a powerful behavioral synthesis flow only based on the scheduling phase. The scheduling is first extended by a datapath analysis for mixed applications, and then a generation efficiently ties the scheduled model to the RTL synthesis. This methodology has been integrated to a system level synthesis flow and its effectiveness has been evaluated on two industrial- scaled examples.
Complete list of metadatas

Cited literature [3 references]  Display  Hide  Download

https://tel.archives-ouvertes.fr/tel-00163733
Contributor : Lucie Torella <>
Submitted on : Wednesday, July 18, 2007 - 12:00:06 PM
Last modification on : Wednesday, May 16, 2018 - 6:30:04 PM
Long-term archiving on : Thursday, April 8, 2010 - 11:32:17 PM

Identifiers

  • HAL Id : tel-00163733, version 1

Collections

TIMA | UGA

Citation

Z. Sugar. Synthèse comportementale basée sur l'ordonnancement. Micro et nanotechnologies/Microélectronique. Institut National Polytechnique de Grenoble - INPG, 2000. Français. ⟨tel-00163733⟩

Share

Metrics

Record views

221

Files downloads

275