Value Slice: A New Slicing Concept for Scalable Property Checking

Abstract

A backward slice is a commonly used preprocessing step for scaling property checking. For large programs though, the reduced size of the slice may still be too large for verifiers to handle. We propose an aggressive slicing method that, apart from slicing out the same statements as backward slice, also eliminates computations that only decide whether the point of property assertion is reachable. However, for precision, we also carefully identify and retain all computations that influence the values of the variables in the property. The resulting slice, called value slice, is smaller and scales better for property checking than backward slice. We carry experiments on property checking of industry strength programs using three comparable slicing techniques: backward slice, value slice and an even more aggressive slicing technique called thin slice that retains only those statements on which the variables in the property are data dependent. While backward slicing enables highest precision and thin slice scales best, value slice based property checking comes close to the best in both scalability and precision. This makes value slice a good compromise between backward and thin slice for property checking.