Publications

While the number of IoT devices grows at an exhilarating pace their security remains stagnant. Imposing secure coding standards across all vendors is infeasible. Testing individual devices allows an analyst to evaluate their security post deployment. Any discovered vulnerabilities can then be disclosed to the vendors in order to assist them in securing their products. The search for vulnerabilities should ideally be automated for efficiency and furthermore be device-independent for scalability. We present FirmFuzz, an automated device-independent emulation and dynamic analysis framework for Linux-based firmware images. It employs a greybox-based generational fuzzing approach coupled with static analysis and system introspection to provide targeted and deterministic bug discovery within a firmware image. We evaluate FirmFuzz by emulating and dynamically analyzing 32 images (from 27 unique devices) with a network accessible from the host performing the emulation. During testing, FirmFuzz discovered seven previously undisclosed vulnerabilities across six different devices: two IP cameras and four routers. So far, 4 CVE's have been assigned.

Cyber-physical systems (CPSs) are increasingly targeted in high-profile cyber attacks. Examples of such attacks include Stuxnet, which targeted nuclear centrifuges; Crashoverride, and Triton, which targeted power grids; and the Mirai botnet, which targeted internet-of-things (IoT) devices such as cameras to carry out a large-scale distributed denial-of-service (DDoS) attack. Such attacks demonstrate the importance of securing current and future cyber-physical systems. Therefore, next-generation operating systems (OSes) for CPS need to be designed to provide security features necessary, as well as be secure in and of themselves. CPSs are designed with one of three broad classes of OSes: (a) bare-metal applications with effectively no operating system, (b) embedded systems executing on impoverished platforms running an embedded or real-time operating system (RTOS) such as FreeRTOS, or (c) more performant platforms running general purpose OSes such as Linux, sometimes tuned for real-time performance such as through the PREEMPT_RT patch. In cases (a) and (b), the OS, if any, is very minimal to facilitate improved resource utilization in real-time or latency-sensitive applications, especially running on impoverished hardware platforms. In such OSes, security is often overlooked, and many important security features (e.g. process/kernel memory isolation) are notably absent. In case (c), the general-purpose OS inherits many of the security-related features that are critical in enterprise and general-purpose applications, such as virtual memory and address-space layout randomization (ASLR). However, the highly complex nature of general-purpose OSes can be problematic in the development of CPSs, as they are highly non-deterministic and difficult to formally reason about for cyber-physical applications, which often have real-time constraints. These issues motivate the need for a next generation OS that is highly capable, predictable and deterministic for real-time performance, but also secure in the face of many of the next generation of cyber threats. In order to design such a next-generation OS, it is necessary to first reflect on the types of threats that CPSs face, including the attacker intentions and types of effects that can be achieved, as well as the type of access that attackers have. While threat models are not the same for all CPSs, it is important to understand how the threat models for CPSs compare to general-purpose or enterprise computing environments. We discuss these issues next (Sec. 2), before providing insights and recommendations for approaches to incorporate in next-generation OSes for CPS in Sec. 3.

Control flow integrity (CFI) has been proposed as an approach to defend against control-hijacking memory corruption attacks. CFI works by assigning tags to indirect branch targets statically and checking them at runtime. Coarse-grained enforcements of CFI that use a small number of tags to improve the performance overhead have been shown to be ineffective. As a result, a number of recent efforts have focused on fine-grained enforcement of CFI as it was originally proposed. In this work, we show that even a finegrained form of CFI with unlimited number of tags and a shadow stack (to check calls and returns) is ineffective in protecting against malicious attacks. We show that many popular code bases such as Apache and Nginx use coding practices that create flexibility in their intended control flow graph (CFG) even when a strong static analyzer is used to construct the CFG. These flexibilities allow an attacker to gain control of the execution while strictly adhering to a fine-grained CFI. We then construct two proof-of-concept exploits that attack an unlimited tag CFI system with a shadow stack. We also evaluate the difficulties of generating a precise CFG using scalable static analysis for real-world applications. Finally, we perform an analysis on a number of popular applications that highlights the availability of such attacks.

Memory corruption attacks continue to be a major vector of attack for compromising modern systems. Numerous defenses have been proposed against memory corruption attacks, but they all have their limitations and weaknesses. Stronger defenses such as complete memory safety for legacy languages (C/C++) incur a large overhead, while weaker ones such as practical control flow integrity have been shown to be ineffective. A recent technique called code pointer integrity (CPI) promises to balance security and performance by focusing memory safety on code pointers thus preventing most control-hijacking attacks while maintaining low overhead. CPI protects access to code pointers by storing them in a safe region that is protected by instruction level isolation. On x86-32, this isolation is enforced by hardware; on x86-64 and ARM, isolation is enforced by information hiding. We show that, for architectures that do not support segmentation in which CPI relies on information hiding, CPI's safe region can be leaked and then maliciously modified by using data pointer overwrites. We implement a proof-of-concept exploit against Nginx and successfully bypass CPI implementations that rely on information hiding in 6 seconds with 13 observed crashes. We also present an attack that generates no crashes and is able to bypass CPI in 98 hours. Our attack demonstrates the importance of adequately protecting secrets in security mechanisms and the dangers of relying on difficulty of guessing without guaranteeing the absence of memory leaks.