Towards Automated Vulnerability Analysis in ARM-based Virtualization 


Vol. 14,  No. 12, pp. 1051-1057, Dec.  2025
10.3745/TKIPS.2025.14.12.1051


PDF
  Abstract

This study systematically analyzes the attack surface of ARM-based virtualization in comparison with x86 and proposes a methodology for identifying ARM-specific vulnerabilities. The methodology comprises three stages—extraction of address-translation, coverage-guided fuzzing, and multi-layered detection. In particular, provides reproducible instrumentation procedures and a cross-validation framework for low-level mechanisms such as NV and the TLB, enabling practitioners to rapidly detect and confirm issues in ARM environments. Applied to KVM/arm64, the methodology revealed and reproduced two concrete vulnerabilities: an ASID matching error and a TLB invalidation-range calculation error. We addressed both with small patches and validated the fixes using the proposed observation procedure, demonstrating a practical approach to vulnerability analysis in ARM virtualization.

  Statistics


  Cite this article

[IEEE Style]

D. Lee, G. Jin, G. Lee, D. Ko, J. Yang, H. Oh, "Towards Automated Vulnerability Analysis in ARM-based Virtualization," The Transactions of the Korea Information Processing Society, vol. 14, no. 12, pp. 1051-1057, 2025. DOI: 10.3745/TKIPS.2025.14.12.1051.

[ACM Style]

Dongha Lee, Gyujeong Jin, Geonha Lee, Daehyeon Ko, Jaewon Yang, and Hyungyu Oh. 2025. Towards Automated Vulnerability Analysis in ARM-based Virtualization. The Transactions of the Korea Information Processing Society, 14, 12, (2025), 1051-1057. DOI: 10.3745/TKIPS.2025.14.12.1051.