Compiler optimizations may result in improvements to code size and/or performance. Consider one or more of the benchmark programs from the SPEC CPU2017 or the EEMBC benchmark suites. Use the RISC-V processor or a processor available to you along with the GNU C compiler to optimize the benchmark program(s) using no optimization, –O1, –O2, and –O3. Compare the performance and size of the resulting programs. Also compare your results to Figure A.21.

Figure A.21.

Program, compiler optimization level lucas, level 3 lucas, level 2 lucas, level 1 lucas, level 0 mcf, level 3 mcf, level 2 mcf, level 1 mcf, level 0 0% 11% 12% 21% Branches/calls I Floating-point ALU ops Loads-stores Integer ALU ops 76% 76% 84% 20% 40% 60% 80% Percentage of unoptimized instructions executed 100% 100% 100%