GlueZilla: Efficient and Scalable Software to Hardware Binding using Rowhammer

GlueZilla: Efficient and Scalable Software to Hardware Binding using Rowhammer

연구 분야: Analysis

학회: International Conference on Detection of Intrusions and Malware, and Vulnerability Assessment

Industrial-scale reverse engineering affects the majority of companies in the mechanical and plant engineering sector and imposes significant economic damages. Although reverse engineering mitigations exist, economic damage has not been impacted, indicating that they have failed to address the problem. A closer investigation shows that industrial-scale reverse engineering typically only expends efforts on replicating hardware, since software can often be copied verbatim—no reverse engineering effort required. We present GLUEZILLA, a system that binds software to hardware through user-space rowhammer PUFs. GLUEZILLA transforms programs such that they only exhibit their intended behavior on the single machine they are bound to at compile time. When run on any other machine, the programs will exhibit a different functionality. GLUEZILLA relies on unclonable machine features and thereby forces counterfeiters to not clone just the hardware but also the software. Cloning both hard- and software drives up reverse engineering costs, thereby also decreasing the economic viability of industrial-scale reverse engineering. GLUEZILLA works on commodity hardware and does not rely on expensive hardware components. Our evaluation shows that GLUEZILLA is effective and incurs 16% run-time performance overhead in a practical case.