MIRAGE: Mitigating Conflict-Based Cache Attacks with a Practical Fully-Associative Design
Gururaj Saileshwar, Moinuddin K. Qureshi
摘要
Shared caches in modern processors are vulnerable to conflict-based attacks, whereby an attacker monitors the access pattern of a victim by engineering cache-set conflicts. Recent mitigations propose a randomized mapping of addresses to cache locations to obfuscate addresses that can conflict with a target address. Unfortunately, such designs continue to select eviction candidates from a small subset of the resident cache lines, which makes such designs vulnerable to algorithms that can quickly identify the conflicting addresses. This paper presents Mirage, a practical design for a fully associative cache, wherein eviction candidates are selected randomly from among all the lines resident in the cache, to be immune to set-conflicts. A key challenge in naively adopting such designs for large shared caches (containing tens of thousands of lines) is the complexity of cache-lookup, as that can require searching through all the lines resident in the cache in such designs. Mirage practically enables a fully-associative design, while maintaining the access latency similar to a traditional set-associative cache using: (1) Pointer-based indirection from the tag-store to the data-store, which allows a newly installed address to evict data of any resident line, (2) Skewed-associative tag-store with extra invalid tags, wherein incoming addresses can be installed without set-conflicts, and (3) Load-aware placement that maximizes the availability of sets with invalid tags, to eliminate set-conflicts. Our analysis shows Mirage provides the global-eviction property of a fully-associative cache throughout the system lifetime (violations of full-associativity, i.e set-conflicts, occur less than once in 10^4 to 10^17 years), offering a principled defense against set-conflict based attacks. Mirage incurs negligible slowdown (0.3%) and 12-15% extra storage compared to the recently proposed Scatter-Cache.
问问这篇 Paper
智能体会读完全文。
Lune 把这篇 Paper 索引到了最后一个公式,引用它的顶会 Paper 也一样。你提问,回答直接引用原文。
引用它的顶会 Paper30
- AQUA: Scalable Rowhammer Mitigation by Quarantining Aggressor Rows at RuntimeAnish Saxena, Gururaj Saileshwar, Prashant J. Nair, Moinuddin K. QureshiMICRO 2022 · 被引用 57 次
- Prime+Scope: Overcoming the Observer Effect for High-Precision Cache Contention AttacksAntoon Purnal, Furkan Turan, Ingrid VerbauwhedeCCS 2021 · 被引用 55 次
- Scalable and Secure Row-Swap: Efficient and Safe Row Hammer Mitigation in Memory SystemsJeonghyun Woo, Gururaj Saileshwar, Prashant J. NairHPCA 2023 · 被引用 45 次
- MeshUp: Stateless Cache Side-channel Attack on CPU MeshJunpeng Wan, Yanxiang Bi, Zhe Zhou, Zhou LiS&P 2022 · 被引用 42 次
- MOESI-prime: preventing coherence-induced hammering in commodity workloadsKevin Loughlin, Stefan Saroiu, Alec Wolman, Yatin A. Manerkar 等ISCA 2022 · 被引用 34 次
它引用的顶会 Paper12
- Prefetch Side-Channel Attacks: Bypassing SMAP and Kernel ASLRDaniel Gruss, Clémentine Maurice, Anders Fogh, Moritz Lipp 等CCS 2016 · 被引用 278 次
- Strong and Efficient Cache Side-Channel Protection using Hardware Transactional MemoryDaniel Gruss, Julian Lettner, Felix Schuster, Olga Ohrimenko 等USENIX Security 2017 · 被引用 254 次
- ScatterCache: Thwarting Cache Attacks via Cache Set RandomizationMario Werner, Thomas Unterluggauer, Lukas Giner, Michael Schwarz 等USENIX Security 2019 · 被引用 221 次
- Prime+Abort: A Timer-Free High-Precision L3 Cache Attack using Intel TSXCraig Disselkoen, David Kohlbrenner, Leo Porter, Dean M. TullsenUSENIX Security 2017 · 被引用 186 次
- Robust Website Fingerprinting Through the Cache Occupancy ChannelAnatoly Shusterman, Lachlan Kang, Yarden Haskal, Yosef Meltser 等USENIX Security 2019 · 被引用 159 次
相关 Paper
- Are Randomized Caches Truly Random? Formal Analysis of Randomized-Partitioned CachesAnirban Chakraborty, Sarani Bhattacharya, Sayandeep Saha, Debdeep MukhopadhyayHPCA 2023 · 被引用 5 次
- The Maya Cache: A Storage-efficient and Secure Fully-associative Last-level CacheAnubhav Bhatla, Navneet, Biswabandan PandaISCA 2024 · 被引用 12 次
- SeqAss: Using SeqUential Associative Caches to Mitigate Conflict-Based Cache Attacks with Reduced Cache Misses and Performance OverheadWei Song, Zhidong Wang, Jinchi Han, Da Xie 等S&P 2026
- Randomized Last-Level Caches Are Still Vulnerable to Cache Side-Channel Attacks! But We Can Fix ItWei Song, Boya Li, Zihan Xue, Zhenzhen Li 等S&P 2021 · 被引用 67 次
- SoK: So, You Think You Know All About Secure Randomized Caches?Anubhav Bhatla, Hari Rohit Bhavsar, Sayandeep Saha, Biswabandan PandaUSENIX Security 2025
