Extended Split-Issue: Enabling Flexibility in the Hardware Implementation of NUAL VLIW DSPs

摘要

VLIW architecture based DSPs have become widespread due to the combined benefits of simple hardware and compiler-extractedinstruction-level parallelism. However, the VLIW instruction set architecture and its hardware implementation are tightly coupled,especially so for Non-Unit Assumed Latency (NUAL) VLIWs. The problem of object code compatibility across processors having different numbers of functional units or hardware latencies has beenthe Achilles' heel of this otherwise powerful architecture. In this paper, we propose eXtended Split-Issue (XSI), a novel mechanismthat breaks the instruction packet syntax of an NUAL VLIW compiler without violating the dataflow dependences. XSI provides a designerthe freedom of disassociating the hardware implementation of the NUAL VLIW processor from the instruction set architecture. Further, we investigate fairly radical (in the context of VLIW) changes to thehardware—like removing an adder, adding a multiplier, and incorporating simultaneous multithreading (SMT)—to show that ourtechnique works for a variety of hardware configurations without compromising on performance. The technique can be used in both single-threaded and multi-threaded architectures to achieve a levelof flexibility heretofore unavailable in the VLIW arena.