Compiler-directed ILP extraction for clustered VLIW/EPIC machines: predication, speculation and modulo scheduling

Compiler-directed ILP extraction techniques are critical to effectively exploiting the significant processing capacity of contemporaneous VLIW/EPIC machines. In this paper we propose a novel algorithm for ILP extraction targeting clustered EPIC machines that integrates three powerful techniques: predication, speculation and modulo scheduling. In addition, our framework schedules and binds operations, generating actual VLIW code. To the best of our knowledge, there is no other algorithm in the literature on predicated code optimizations that jointly considers speculation and modulo scheduling in the context of clustered EPIC machines. Our experimental results show that by jointly considering different extraction techniques in a resource aware context, the proposed algorithm can take maximum advantage of the resources available on the clustered machine, aggressively improving performance.