Instruction Level Parallelism (ILP) extraction for multi-cluster VLIW processors is a very hard task. In this paper, we propose a retargetable architecture that can exploit ILP and thread level parallelism jointly, thus allowing an easier parallelism extraction and improving the performance with respect to traditional multicluster VLIW processors.
Citation:
Domenico Barretta, William Fornaciari, Mariagiovanna Sami, Daniele Bagni, "Multithreaded Extension to Multicluster VLIW Processors for Embedded Applications," date, vol. 2, pp.748-749, Design, Automation and Test in Europe (DATE'05) Volume 2, 2005