Retargetable compilation for low power

Most research to date on energy minimization in DSP processors has focused on hardware solution. This paper examines the software-based factors affecting performance and energy consumption for architecture-aware compilation. In this paper, we focus on providing support for one architectural feature of DSPs that makes code generation difficult, namely the use of multiple data memory banks. This feature increases memory bandwidth by permitting multiple data memory accesses to occur in parallel when the referenced variables belong to different data memory banks and the registers involved conform to a strict set of conditions. We present novel instruction scheduling algorithms that attempt to maximize the performance, minimize the energy, and therefore, maximize the benefit of this architectural feature. Experimental results demonstrate that our algorithms generate high performance, low energy codes for the DPS architectural features with multiple data memory banks. Our algorithm led to improvements in performance and energy consumption of 48.3% and 66.6% respectively in our benchmark examples