TreasureCache: Hiding Cache Evictions Against Side-Channel Attacks

被引:1
作者
Li, Mengming [1 ,2 ]
Bu, Kai [2 ,3 ]
Miao, Chenlu [3 ]
Ren, Kui [3 ,4 ]
机构
[1] Zhejiang Univ, Sch Software Technol, Hangzhou 310027, Peoples R China
[2] ZJU Hangzhou Global Sci & Technol Innovat Ctr, Hangzhou 311215, Peoples R China
[3] Zhejiang Univ, Coll Comp Sci & Technol, Hangzhou 310027, Peoples R China
[4] Zhejiang Prov Key Lab Blockchain & Cyberspace Gove, Hangzhou 310027, Peoples R China
基金
中国国家自然科学基金; 国家重点研发计划;
关键词
Side-channel attacks; Timing; Security; Hardware; Software; Computer science; Behavioral sciences; Cache side-channel attack; eviction-hidden buffer; secure replacement policy;
D O I
10.1109/TDSC.2024.3354991
中图分类号
TP3 [计算技术、计算机技术];
学科分类号
0812 ;
摘要
Cache side-channel attacks remain a stubborn source of cross-core secret leakage. Such attacks exploit the timing difference between cache hits and misses. Most defenses thus choose to prevent cache evictions. Given that two possible types of evictions-flush-based and conflict-based-use different architectural features, these defenses have to integrate hybrid defense strategies, incur OS modification, and sacrifice performance to completely throttle cache side-channel attacks. In this article, we present TreasureCache against cache side-channel attacks without modifying OS or sacrificing performance. Instead of preventing cache evictions with various costs, we advocate to allow cache evictions as is and hide exploitable evictions in our specialized small eviction-hidden buffer. The buffer guarantees a fast hit time comparative to LLC hits. This instantly closes the timing gap between accessing exploitable blocks when they are in and out of the LLC. Moreover, with the help of our buffer, we no longer have to disable flush instructions or shared memory. A lightweight constant-time flush instruction can help TreasureCache to prevent both flush-based and conflict-based side-channel attacks. We validate TreasureCache security and performance through extensive experiments. With a hardware overhead of less than 0.5%, TreasureCache reduces the secret-leakage resolution by about 1,000 times without introducing any performance slowdown.
引用
收藏
页码:4574 / 4588
页数:15
相关论文
共 65 条
[1]  
Aciiçmez O, 2007, CSAW'07: PROCEEDINGS OF THE 2007 ACM COMPUTER SECURITY ARCHITECTURE WORKSHOP, P11
[2]  
Aciiçqmez O, 2008, LECT NOTES COMPUT SC, V4964, P256, DOI 10.1007/978-3-540-79263-5_16
[3]  
Agarwal A, 2022, P IEEE S SECUR PRIV, P699, DOI [10.1109/SP46214.2022.00034, 10.1109/SP46214.2022.9833711]
[4]   MuonTrap: Preventing Cross-Domain Spectre-Like Attacks by Capturing Speculative State [J].
Ainsworth, Sam ;
Jones, Timothy M. .
2020 ACM/IEEE 47TH ANNUAL INTERNATIONAL SYMPOSIUM ON COMPUTER ARCHITECTURE (ISCA 2020), 2020, :132-144
[5]  
[Anonymous], 2024, Ruby home page
[6]  
Binkert Nathan, 2011, Computer Architecture News, V39, P1, DOI 10.1145/2024716.2024718
[7]  
Briongos S., 2020, P USENIX SEC S
[8]  
Brumley BB, 2009, LECT NOTES COMPUT SC, V5912, P667, DOI 10.1007/978-3-642-10366-7_39
[9]   SPEC CPU2017-Next-Generation Compute Benchmark [J].
Bucek, James ;
Lange, Klaus-Dieter ;
von Kistowski, Joakim .
COMPANION OF THE 2018 ACM/SPEC INTERNATIONAL CONFERENCE ON PERFORMANCE ENGINEERING (ICPE '18), 2018, :41-42
[10]   SGXPECTRE: Stealing Intel Secrets from SGX Enclaves via Speculative Execution [J].
Chen, Guoxing ;
Chen, Sanchuan ;
Xiao, Yuan ;
Zhang, Yinqian ;
Lin, Zhiqiang ;
Lai, Ten H. .
2019 4TH IEEE EUROPEAN SYMPOSIUM ON SECURITY AND PRIVACY (EUROS&P), 2019, :142-157