A Survey of Binary Code Similarity

Binary code similarityapproaches compare two or more pieces of binary to identify their similarities and differences. The ability enables many real-world applications on scenarios where source may not be available such as patch analysis, bug search, malware detection analysis. Over the past 22 years numerous similarity approaches have been proposed, but research area has yet systematically analyzed. This article presents first survey similarity. It analyzes 70 approaches, which are systematized four aspects: (1) they enable, (2) approach characteristics, (3) how implemented, (4) benchmarks methodologies used evaluate them. In addition, discusses scope origins area, its evolution over decades, challenges that lie ahead.