A Systematic Look at Ciphertext Side Channels on AMD SEV-SNP
Mengyuan Li, Luca Wilke, Jan Wichelmann, Thomas Eisenbarth, Radu Teodorescu, Yinqian Zhang
摘要
Hardware-assisted memory encryption offers strong confidentiality guarantees for trusted execution environments like Intel SGX and AMD SEV. However, a recent study by Li et al. presented at USENIX Security 2021 has demonstrated the CipherLeaks attack, which monitors ciphertext changes in the special VMSA page. By leaking register values saved by the VM during context switches, they broke state-of-the-art constant-time cryptographic implementations, including RSA and ECDSA in the OpenSSL. In this paper, we perform a comprehensive study on the ciphertext side channels. Our work suggests that while the CipherLeaks attack targets only the VMSA page, a generic ciphertext side-channel attack may exploit the ciphertext leakage from any memory pages, including those for kernel data structures, stacks and heaps. As such, AMD’s existing countermeasures to the CipherLeaks attack, a firmware patch that introduces randomness into the ciphertext of the VMSA page, is clearly insufficient. The root cause of the leakage in AMD SEV’s memory encryption—the use of a stateless yet unauthenticated encryption mode and the unrestricted read accesses to the ciphertext of the encrypted memory—remains unfixed. Given the challenges faced by AMD to eradicate the vulnerability from the hardware design, we propose a set of software countermeasures to the ciphertext side channels, including patches to the OS kernel and cryptographic libraries. We are working closely with AMD to merge these changes into affected open-source projects.
问问这篇 Paper
智能体会读完全文。
Lune 把这篇 Paper 索引到了每一个公式,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 Paper45
- HECKLER: Breaking Confidential VMs with Malicious InterruptsBenedict Schlüter, Supraja Sridhara, Mark Kuhne, Andrin Bertschi 等USENIX Security 2024 · 被引用 48 次
- WeSee: Using Malicious #VC Interrupts to Break AMD SEV-SNPBenedict Schlüter, Supraja Sridhara, Andrin Bertschi, Shweta ShindeS&P 2024 · 被引用 42 次
- CacheWarp: Software-based Fault Injection using Selective State ResetRuiyi Zhang, Lukas Gerlach, Daniel Weber, Lorenz Hetterich 等USENIX Security 2024 · 被引用 36 次
- Bifrost: Analysis and Optimization of Network I/O Tax in Confidential Virtual MachinesDingji Li, Zeyu Mi, Chenhui Ji, Yifan Tan 等USENIX ATC 2023 · 被引用 31 次
- TEE.Fail: Breaking Trusted Execution Environments via DDR5 Memory Bus InterpositionJalen Chuang, Alexander Seto, Nicolás Berrios, Stephan van Schaik 等S&P 2026 · 被引用 29 次
它引用的顶会 Paper8
- DRAMA: Exploiting DRAM Addressing for Cross-CPU AttacksPeter Pessl, Daniel Gruss, Clémentine Maurice, Michael Schwarz 等USENIX Security 2016 · 被引用 500 次
- CIPHERLEAKS: Breaking Constant-time Cryptography on AMD SEV via the Ciphertext Side ChannelMengyuan Li, Yinqian Zhang, Huibo Wang, Kang Li 等USENIX Security 2021 · 被引用 130 次
- Exploiting Unprotected I/O Operations in AMD's Secure Encrypted VirtualizationMengyuan Li, Yinqian Zhang, Zhiqiang Lin, Yan SolihinUSENIX Security 2019 · 被引用 104 次
- SEVurity: No Security Without Integrity : Breaking Integrity-Free Memory Encryption with Minimal AssumptionsLuca Wilke, Jan Wichelmann, Mathias Morbitzer, Thomas EisenbarthS&P 2020 · 被引用 72 次
- Ginseng: Keeping Secrets in Registers When You Distrust the Operating SystemMin Hong Yun, Lin ZhongNDSS 2019 · 被引用 48 次
相关 Paper
- Cipherfix: Mitigating Ciphertext Side-Channel Attacks in SoftwareJan Wichelmann, Anna Pätschke, Luca Wilke, Thomas EisenbarthUSENIX Security 2023
- CipherH: Automated Detection of Ciphertext Side-channel Vulnerabilities in Cryptographic ImplementationsSen Deng, Mengyuan Li, Yining Tang, Shuai Wang 等USENIX Security 2023
- Shadows in Cipher Spaces: Exploiting Tweak Repetition in Hardware Memory EncryptionWei Peng, Yinshuai Li, Yinqian ZhangUSENIX Security 2025
- Heracles: Chosen Plaintext Attack on AMD SEV-SNPBenedict Schlüter, Christoph Wech, Shweta ShindeCCS 2025 · 被引用 1 次
- CounterSEVeillance: Performance-Counter Attacks on AMD SEV-SNPStefan Gast, Hannes Weissteiner, Robin Leander Schröder, Daniel GrussNDSS 2025
