Control Flow Integrity in IoT Devices with Performance Counters and DWT
Authors:
- 1 December 2020
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 171-176
Abstract
IoT devices are open to traditional control flow integrity (CFI) attacks resulting from buffer overflow and return-oriented programming like techniques. They often have limited computational capacity ruling out many of the traditional heavy-duty software countermeasures. In this work, we deploy hardware/software solutions to detect CFI attacks. Some of the medium capability IoT devices, for example based on Raspberry Pi, contain ARM Cortex A-53 (Pi 3) or Cortex A-73 (Pi 4) processors. These processors include hardware counters to count microarchitecture level events affecting performance. Lighter weight IoT devices, say based on ARM Cortex M4 or M7, include DWT (Debug, Watch & Trace) module. When control flow anomalies caused by attacks such as buffer overflow or return oriented programming (ROP) occur, they leave a microarchitectural footprint. Hardware counters reflect such footprints to flag control flow anomalies. This paper is geared towards buffer overflow and ROP control flow anomaly detection in embedded programs. The targeted program entities are main event loops and task/event handlers. The proposed anomaly detection mechanism is evaluated on ArduPilot [1] - a popular autopilot software on a Raspberry Pi 3 with PMU and DWT. A self-navigation program is evaluated on an iCreate Roomba platform with an ARM Cortex M4 processor with DWT only. We are able to achieve 97-99%+ accuracy with 1-10 micro-second time overhead per control flow anomaly check.Keywords
This publication has 8 references indexed in Scilit:
- On the Detection of Kernel-Level Rootkits Using Hardware Performance Counters2017
- ASLR on the Line: Practical Cache Attacks on the MMU2017
- Jump over ASLR: Attacking branch predictors to bypass ASLR2016
- ROPecker: A Generic and Practical Approach For Defending Against ROP Attacks2014
- LIBSVMACM Transactions on Intelligent Systems and Technology, 2011
- Exploiting Hardware Performance Counters2008
- Control-flow integrity2005
- On the effectiveness of address-space randomization2004