Finding Suitable Variability Abstractions for Family-Based Analysis

For program families (Software Product Lines), specially designed variability-aware static (dataflow) analyses allow analyzing all variants (products) of the family, simultaneously, in a single run without generating any of the variants explicitly. They are also known as lifted or family-based analyses. The variability-aware analyses may be too costly or even infeasible for families with a large number of variants. In order to make them computationally cheaper, we can apply variability abstractions which aim to tame the combinatorial explosion of the number of variants (configurations) and reduce it to something more tractable. % by manipulating the configuration space of the family. However, the number of possible abstractions is still intractably large to search naively, with most abstractions being too imprecise or too costly.

In this work, we propose a technique to efficiently find suitable variability abstractions from a large family of abstractions for a variability-aware static analysis. The idea is to use a pre-analysis to estimate the impact of variability-specific parts of the program family on the analysis's precision. Then we use the pre-analysis results to find out when and where the analysis should turn off or on its variability-awareness. We demonstrate the practicality of this approach on several Java benchmarks.