iBinHunt: Binary Hunting with Inter-Procedural Control Flow

• Main Authors: MING, Jiang, PAN, Meng, GAO, Debin
• Format: text
• Language: English
• Published: Institutional Knowledge at Singapore Management University 2012
• Subjects: binary diffing; semantic difference; taint analysis; Information Security; Software Engineering
• Online Access: https://ink.library.smu.edu.sg/sis_research/1700; https://ink.library.smu.edu.sg/context/sis_research/article/2699/viewcontent/icisc12.pdf
• Institution: Singapore Management University

Techniques have been proposed to find the semantic differences between two binary programs when the source code is not available. Analyzing control flow, and in particular, intra-procedural control flow, has become an attractive technique in the latest binary diffing tools since it is more resistant to syntactic, but non-semantic, differences. However, this makes such techniques vulnerable to simple function obfuscation techniques (e.g., function inlining) attackers any malware writers could use. In this paper, we first show function obfuscation as an attack to such binary diffing techniques, and then propose iBinHunt which uses deep taint and automatic input generation to find semantic differences in inter-procedural control flows. Evaluation on comparing various versions of a http server and gzip shows that iBinHunt not only is capable of comparing inter-procedural control flows of two programs, but offers substantially better accuracy and efficiency in binary diffing.