MACE: High-Coverage and Robust Memory Analysis For Commodity Operating Systems

Memory analysis on on commodity OSes faces several challenges: partial knowledge of kernel data structures, difficulty in handling ambiguous pointers, and lack of robustness by relying on soft constraints. To address these challenges, we present MACE, a memory analysis system that achieves high coverage and robustness by evaluating pointer constraints globally.

Author(s):

Qian Feng    
Syracuse University
United States

Aravind Prakash    
Syracuse University
United States

Heng Yin    
Syracuse University
United States

Zhiqiang Lin    
University of Texas at Dallas
United States