HALF: Process Hollowing Analysis Framework for Binary Programs with the Assistance of Kernel Modules
Abstract
Binary program analysis is still very important in system security. There are many practical achievements in binary code analysis, but fine-grained analysis such as dynamic taint analysis, is constantly studied due to the problem of deployability, high memory usage, and performance overhead, so as to better adapt to the new analysis scenario, such as memory corruption exploits and sandbox evasion malware. This paper presents a new binary program analysis framework, in order to improve the usability and performance of fine-grained analysis. The framework mainly uses the kernel module to further expand the analysis capability of the traditional dynamic binary instrumentation. Then, based on the idea of decoupling analysis, the analysis environment is constructed in the container process through process hollowing techniques in a new way. It can reuse the functions of the existing dynamic binary instrumentation platforms and also reduce the impact on the execution of the target program. The prototype is implemented on the Windows platform. The validity and performance of the framework are verified by a large number of experiments with benchmark and actual programs. The effectiveness of the framework is also verified by the analysis of actual exploit programs and malicious code, demonstrating the value of the practical application.