Static call graph generation is a basic building block of static program analysis. Many other analyses are based on the resulting call graph and directly inherit their imprecisions and flaws. Therefore, a most precise and complete call graph should be the aim of this step. However, in practice it’s not that easy to evaluate call graph generation results, especially under realistic conditions.
In our new publication “Total Recall? How Good are Static Call Graphs Really?” just accepted at the ACM SIGSOFT International Symposium on Software Testing and Analysis (ISSTA), we construct dynamic baselines that show a realistic call graph for a certain entry point. This allows us to study the quality of static call graph construction. The paper is a collaborative effort with the Software Technology Group from Prof. Mira Mezini at TU Darmstadt.
The ACM SIGSOFT International Symposium on Software Testing and Analysis (ISSTA) is the leading research symposium on software testing and analysis, bringing together academics, industrial researchers, and practitioners to exchange new ideas, problems, and experience on how to analyze and test software systems. It has a CORE ranking of A.