Summary
Many security attacks on computer memory rely on measuring how long things take to happen, but some protections can make those timing clues harder to spot. This paper shows that regular software running without special permissions can use electric power measurements inside processors to figure out how memory is organized and what data is being accessed, even without timing information. The researchers created a method called HOMARD that uses power data to reverse-engineer memory addresses and recover secret information from memory. Their approach works on popular devices where traditional timing attacks fail, proving power measurements alone can provide a new way to spy on data.
What this means in practice
- •For security engineers: Detect memory vulnerabilities on ARM devices by monitoring unprivileged power sensor data to identify exploitable side channels when timing-based methods fail.
- •For hardware designers: Design memory systems and SoCs that reduce information leakage by considering power-based side channels revealed by millisecond-resolution sensors.
Authors
Eliott Quéré, Li-Chung Chiang, Carina Fiedler, Thomas Rokicki, Maria Méndez Real, Alessandro Palumbo, Lilian Bossuet, Rubén Salvador, Daniel Gruss
Abstract
Software-based microarchitectural attacks have mainly exploited timing differences to infer information from DRAM. However, defenses and optimizations can prevent timing-based approaches from resolving precise microarchitectural behaviors. We show that unprivileged millisecond- and milliampere-resolution power sensors on modern Arm SoCs can recover DRAM mapping functions and reveal the data dependence of memory accesses without timers. We present HOMARD, a methodology that reads DRAM activity from this telemetry, and two attack primitives. HOMARD-RE is a hierarchy-aware linear-algebraic method that reverse-engineers DRAM addressing functions from power alone. HOMARD-DATA recovers the Hamming weight of a fetched cache line through joint-power rail templating and secret byte values through correlation power analysis on the same rails. Across an AMD ZCU102, a Raspberry Pi 5, and a multi-chip-based LPDDR4 NVIDIA Jetson TX2, HOMARD-RE succeeds where row-buffer timing methods fail. On the Pi 5, HOMARD-DATA classifies 9 Hamming-weight levels at 81.5% accuracy and recovers secret bytes in minutes by correlating rail current with data-dependent bit toggling between neighbors in a row. Both primitives work with an architectural load or a mispredicted Spectre-RSB path on which no value is architecturally read. These results establish unprivileged power telemetry as an independent side channel for profiled microarchitectural attacks, particularly when timing falls silent.