Using SPEC CPU2006 to evaluate the sequential and parallel code generated by commercial and open-source compilers

Abstract

The role of the compiler is fundamental to exploit the hardware capabilities of a system running a particular application, minimizing the sequential execution time and, in some cases, offering the possibility of parallelizing part of the code automatically. This paper relies on the SPEC CPU2006 v1.1 benchmark suite to evaluate the performance of the code generated by three widely-used compilers (Intel C++/Fortran Compiler 11.0, Sun Studio 12 and GCC 4.3.2). Performance is measure in terms of base speed for reference problem sizes. Both sequential and automatic parallel performance obtained is analyzed, using different hardware architectures and configurations. The study includes a detailed description of the different problems that arise while compiling SPEC CPU2006 benchmarks with these tools, an information difficult to obtain elsewhere.

Having in mind that performance is a moving target in the field of compilers, our evaluation shows that the sequential code generated by both Sun and Intel compilers for the SPEC CPU2006 integer benchmarks present a similar performance, while the floating-point code generated by Intel compiler is faster than its competitors. With respect to the auto-parallelization options offered by Intel and Sun compilers, our study shows that their benefits only apply to some floating-point benchmarks, with an average speedup of 1.2× with four processors. Meanwhile, the GCC suite evaluated is not capable of compiling the SPEC CPU2006 benchmark with auto-parallelization options enabled.