Yet Another Mirage of Breaking MIRAGE: Debunking Occupancy-based Side-Channel Attacks on Fully Associative Randomized Caches

TL;DR

This paper debunks recent claims that occupancy-based side-channel attacks can break MIRAGE's randomized cache, showing that previous results were due to flawed simulation conditions that do not reflect real-world operation.

Contribution

It identifies and corrects a modeling flaw in prior work, demonstrating that MIRAGE's security holds under realistic randomized eviction scenarios.

Findings

Attack success depends on flawed simulation conditions.

Realistic eviction randomness prevents the attack.

MIRAGE remains secure against occupancy-based side channels.

Abstract

Recent work presented at USENIX Security 2025 (SEC'25) claims that occupancy-based attacks can recover AES keys from the MIRAGE randomized cache. In this paper, we examine these claims and find that they arise from a modeling flaw in the SEC'25 paper. Most critically, the SEC'25 paper's simulation of MIRAGE uses a constant seed to initialize the random number generator used for global evictions in MIRAGE, causing every AES encryption they trace to evict the same deterministic sequence of cache lines. This artificially creates a highly repeatable timing pattern that is not representative of a realistic implementation of MIRAGE, where eviction sequences vary randomly between encryptions. When we instead randomize the eviction seed for each run, reflecting realistic operation, the correlation between AES T-table accesses and attacker runtimes disappears, and the attack fails. These findings show that the reported leakage is an artifact of incorrect modeling, and not an actual vulnerability in MIRAGE.