A Comparison of VLIW and Traditional DSP Architectures for Compiled Code

Although programmable digital signal processors comprise a significant fraction of the processors sold in the world, their basic architectures have changed little since they were originally developed. The evolu-tion and implementation of these processors has been based more on commonly held beliefs than quantitative data. In this paper, we show that by changing to a VLIW model with more registers, orthogonal instructions, and better flexibility for instruction-level parallelism, it is possible to achieve at least a factor of 1.3–2 in performance gain over the traditional DSP architectures on a suite of DSP benchmarks. When accounting for the effect of restrictive register use in traditional DSP architectures, we argue that the actual performance gain is at least a factor of 1.8–2.8. To counter an argument about extra chip area, we show that the cost of adding more registers is minimal when the overall area of the processor and the performance benefits are considered. Although a VLIW architecture has a much lower instruction density, we also show that the average number of instructions is actually reduced because there are fewer memory operations. A significant contribution to the better performance of the VLIW architecture is the ability to express more instances of parallelism than the restricted parallelism of the more traditional architectures. However, efficient techniques for encoding long instructions are required to make the higher flexibility and better perfor-mance of VLIW architectures feasible