A comparison of quantum compilers using a DAG-based or phase polynomial-based intermediate representation

Floating-point programs are challenging for symbolic execution due to the constraint solving problem. This paper empirically studies five existing symbolic execution methods for floating-point programs to evaluate their effectiveness and limitations. We have implemented the existing methods based on the state-of-the-art symbolic execution tool KLEE and constructed a real-world floating-point program benchmark for evaluation. We evaluate the existing methods with respect to statement coverage and the ability to detect floating-point exceptions. The results demonstrate that the existing methods complement each other. Based on the evaluation results, we propose a synergistic approach to improving the efficiency of the symbolic execution for floating-point programs. The experimental results indicate our synergistic method’s effectiveness in finding floating-point exceptions.