An Evil Copy: How the Loader Betrays You
Xinyang Ge, Mathias Payer, Trent Jaeger
Abstract
Dynamic loading is a core feature used on current systems to (i) enable modularity and reuse, (ii) reduce memory footprint by sharing code pages of libraries and executables among processes, and (iii) simplify update procedures by eliminating the need to recompile executables when a library is updated. The Executable and Linkable Format (ELF) is a generic specification that describes how executable programs are stitched together from object files produced from source code to libraries and executables. Programming languages allow fine-grained control over variables, including access and memory protections, so programmers may write defense mechanisms assuming that the permissions specified at the source and/or compiler level will hold at runtime. Unfortunately, information about memory protection is lost during compilation. We identify one case that has significant security implications: when instantiating a process, constant external variables that are referenced in executables are forcefully relocated to a writable memory segment without warning. The loader trades security for compatibility due to the lack of memory protection information on the relocated external variables. We call this new attack vector COREV for Copy Relocation Violation. An adversary may use a memory corruption vulnerability to modify such "read-only" constant variables like vtables, function pointers, format strings, and file names to bypass defenses (like FORTIFY SOURCE or CFI) and to escalate privileges. We have studied all Ubuntu 16.04 LTS packages and found that out of 54,045 packages, 4,570 packages have unexpected copy relocations that change read-only permissions to read-write, presenting new avenues for attack. The attack surface is broad with 29,817 libraries exporting relocatable read-only variables. The set of 6,399 programs with actual copy relocation violations includes ftp servers, apt-get, and gettext. We discuss the cause, effects, and a set of possible mitigation strategies for the COREV attack vector. Permission to freely reproduce all or part of this paper for noncommercial purposes is granted provided that copies bear this notice and the full citation on the first page. Reproduction for commercial purposes is strictly prohibited without the prior written consent of the Internet Society, the first-named author (for reproduction of an entire paper only), and the author's employer if the paper was prepared within the scope of employment.
Ask about this paper
Your agent reads all of it.
Lune indexed this paper to the last equation, along with the top-tier papers that cite it. Ask a question and the answer quotes them.
Cited by top-tier papers2
- The Taming of the Stack: Isolating Stack Data from Memory ErrorsKaiming Huang, Yongzhe Huang, Mathias Payer, Zhiyun Qian et al.NDSS 2022
- Too Subtle to Notice: Investigating Executable Stack Issues in Linux SystemsHengkai Ye, Hong HuNDSS 2025
Builds on2
Related papers
- Debloating Software through Piece-Wise Compilation and LoadingAnh Quach, Aravind Prakash, Lok-Kwong YanUSENIX Security 2018 · 153 citations
- When Good Kernel Defenses Go Bad: Reliable and Stable Kernel Exploits via Defense-Amplified TLB Side-Channel LeaksLukas Maar, Lukas Giner, Daniel Gruss, Stefan MangardUSENIX Security 2025
- Crashing Through Defenses: Exploiting Segfaults and Chaining Around Intel CETMarcos Bajo, Ritvik Goyal, Apostolos Chatzianagnostou, Christian RossowS&P 2026
- Address Oblivious Code Reuse: On the Effectiveness of Leakage Resilient DiversityRobert Rudd, Richard Skowyra, David Bigelow, Veer Dedhia et al.NDSS 2017 · 78 citations
- What Cannot Be Read, Cannot Be Leveraged? Revisiting Assumptions of JIT-ROP DefensesGiorgi Maisuradze, Michael Backes, Christian RossowUSENIX Security 2016 · 41 citations
