Compiler Optimization on Instruction Scheduling for a Specialized Real-Time Floating Point Soft-Core Processor

Abstract

This paper presents the authors' research in the field of specialized optimizing assembly language compilers for embedded real-time soft-core processor systems on FPGAs. With this soft-core processor, we are targeting a highly specialized field of applications that require large floating-point precision and other unique characteristics. Therefore, a specialized optimizing assembly language compiler is necessary in order to provide the needed machine code and optimize it in a way that efficient usage of the internal parallelism mechanisms is possible, resulting in major performance benefits on single-core, multi-core and vector processors. One important key feature is the design-time analyzability to meet the hard real-time constraints of any given problem.

Keywords

• dynamic compiler
• optimization methods
• processor scheduling
• scheduling algorithms
• vector processor