A Systematic Evaluation of Transient Execution Attacks and Defenses

Claudio Canella, Jo Van Bulck, Michael Schwarz, Moritz Lipp, Benjamin von Berg, Philipp Ortner, Frank Piessens, Dmitry Evtyushkin, Daniel Gruss

Research output: Chapter in Book/Report/Conference proceedingConference contributionResearchpeer-review

Abstract

Research on transient execution attacks including Spectre and Meltdown showed that exception or branch misprediction events might leave secret-dependent traces in the CPU’s microarchitectural state. This observation led to a proliferation of new Spectre and Meltdown attack variants and even more ad-hoc defenses (e.g., microcode and software patches). Both the industry and academia are now focusing on finding effective defenses for known issues. However, we only have limited insight on residual attack surface and the completeness of the proposed defenses.

In this paper, we present a systematization of transient execution attacks. Our systematization uncovers 6 (new) transient execution attacks that have been overlooked and not been investigated so far: 2 new exploitable Meltdown effects: Meltdown-PK (Protection Key Bypass) on Intel, and Meltdown-BND (Bounds Check Bypass) on Intel and AMD; and 4 new Spectre mistraining strategies. We evaluate the attacks in our classification tree through proof-of-concept implementations on 3 major CPU vendors (Intel, AMD, ARM). Our Systematization yields a more complete picture of the attack surface and allows for a more systematic evaluation of defenses. Through this systematic evaluation, we discover that most defenses, including deployed ones, cannot fully mitigate all attack variants.
Original languageEnglish
Title of host publicationProceedings of the 28th USENIX Security Symposium
PublisherUSENIX Association
Pages249-266
Number of pages17
Publication statusPublished - 14 Aug 2019

Publication series

NamearXiv.org e-Print archive
PublisherCornell University Library

Fingerprint

Program processors
Core meltdown
Firmware
Industry

Keywords

  • cs.CR

Cite this

Canella, C., Bulck, J. V., Schwarz, M., Lipp, M., Berg, B. V., Ortner, P., ... Gruss, D. (2019). A Systematic Evaluation of Transient Execution Attacks and Defenses. In Proceedings of the 28th USENIX Security Symposium (pp. 249-266). (arXiv.org e-Print archive). USENIX Association.

A Systematic Evaluation of Transient Execution Attacks and Defenses. / Canella, Claudio; Bulck, Jo Van; Schwarz, Michael; Lipp, Moritz; Berg, Benjamin von; Ortner, Philipp; Piessens, Frank; Evtyushkin, Dmitry; Gruss, Daniel.

Proceedings of the 28th USENIX Security Symposium. USENIX Association, 2019. p. 249-266 (arXiv.org e-Print archive).

Research output: Chapter in Book/Report/Conference proceedingConference contributionResearchpeer-review

Canella, C, Bulck, JV, Schwarz, M, Lipp, M, Berg, BV, Ortner, P, Piessens, F, Evtyushkin, D & Gruss, D 2019, A Systematic Evaluation of Transient Execution Attacks and Defenses. in Proceedings of the 28th USENIX Security Symposium. arXiv.org e-Print archive, USENIX Association, pp. 249-266.
Canella C, Bulck JV, Schwarz M, Lipp M, Berg BV, Ortner P et al. A Systematic Evaluation of Transient Execution Attacks and Defenses. In Proceedings of the 28th USENIX Security Symposium. USENIX Association. 2019. p. 249-266. (arXiv.org e-Print archive).
Canella, Claudio ; Bulck, Jo Van ; Schwarz, Michael ; Lipp, Moritz ; Berg, Benjamin von ; Ortner, Philipp ; Piessens, Frank ; Evtyushkin, Dmitry ; Gruss, Daniel. / A Systematic Evaluation of Transient Execution Attacks and Defenses. Proceedings of the 28th USENIX Security Symposium. USENIX Association, 2019. pp. 249-266 (arXiv.org e-Print archive).
@inproceedings{407a60b30d1f46cf8b82002c24ebbca2,
title = "A Systematic Evaluation of Transient Execution Attacks and Defenses",
abstract = "Research on transient execution attacks including Spectre and Meltdown showed that exception or branch misprediction events might leave secret-dependent traces in the CPU’s microarchitectural state. This observation led to a proliferation of new Spectre and Meltdown attack variants and even more ad-hoc defenses (e.g., microcode and software patches). Both the industry and academia are now focusing on finding effective defenses for known issues. However, we only have limited insight on residual attack surface and the completeness of the proposed defenses.In this paper, we present a systematization of transient execution attacks. Our systematization uncovers 6 (new) transient execution attacks that have been overlooked and not been investigated so far: 2 new exploitable Meltdown effects: Meltdown-PK (Protection Key Bypass) on Intel, and Meltdown-BND (Bounds Check Bypass) on Intel and AMD; and 4 new Spectre mistraining strategies. We evaluate the attacks in our classification tree through proof-of-concept implementations on 3 major CPU vendors (Intel, AMD, ARM). Our Systematization yields a more complete picture of the attack surface and allows for a more systematic evaluation of defenses. Through this systematic evaluation, we discover that most defenses, including deployed ones, cannot fully mitigate all attack variants.",
keywords = "cs.CR",
author = "Claudio Canella and Bulck, {Jo Van} and Michael Schwarz and Moritz Lipp and Berg, {Benjamin von} and Philipp Ortner and Frank Piessens and Dmitry Evtyushkin and Daniel Gruss",
year = "2019",
month = "8",
day = "14",
language = "English",
series = "arXiv.org e-Print archive",
publisher = "USENIX Association",
pages = "249--266",
booktitle = "Proceedings of the 28th USENIX Security Symposium",
address = "United States",

}

TY - GEN

T1 - A Systematic Evaluation of Transient Execution Attacks and Defenses

AU - Canella, Claudio

AU - Bulck, Jo Van

AU - Schwarz, Michael

AU - Lipp, Moritz

AU - Berg, Benjamin von

AU - Ortner, Philipp

AU - Piessens, Frank

AU - Evtyushkin, Dmitry

AU - Gruss, Daniel

PY - 2019/8/14

Y1 - 2019/8/14

N2 - Research on transient execution attacks including Spectre and Meltdown showed that exception or branch misprediction events might leave secret-dependent traces in the CPU’s microarchitectural state. This observation led to a proliferation of new Spectre and Meltdown attack variants and even more ad-hoc defenses (e.g., microcode and software patches). Both the industry and academia are now focusing on finding effective defenses for known issues. However, we only have limited insight on residual attack surface and the completeness of the proposed defenses.In this paper, we present a systematization of transient execution attacks. Our systematization uncovers 6 (new) transient execution attacks that have been overlooked and not been investigated so far: 2 new exploitable Meltdown effects: Meltdown-PK (Protection Key Bypass) on Intel, and Meltdown-BND (Bounds Check Bypass) on Intel and AMD; and 4 new Spectre mistraining strategies. We evaluate the attacks in our classification tree through proof-of-concept implementations on 3 major CPU vendors (Intel, AMD, ARM). Our Systematization yields a more complete picture of the attack surface and allows for a more systematic evaluation of defenses. Through this systematic evaluation, we discover that most defenses, including deployed ones, cannot fully mitigate all attack variants.

AB - Research on transient execution attacks including Spectre and Meltdown showed that exception or branch misprediction events might leave secret-dependent traces in the CPU’s microarchitectural state. This observation led to a proliferation of new Spectre and Meltdown attack variants and even more ad-hoc defenses (e.g., microcode and software patches). Both the industry and academia are now focusing on finding effective defenses for known issues. However, we only have limited insight on residual attack surface and the completeness of the proposed defenses.In this paper, we present a systematization of transient execution attacks. Our systematization uncovers 6 (new) transient execution attacks that have been overlooked and not been investigated so far: 2 new exploitable Meltdown effects: Meltdown-PK (Protection Key Bypass) on Intel, and Meltdown-BND (Bounds Check Bypass) on Intel and AMD; and 4 new Spectre mistraining strategies. We evaluate the attacks in our classification tree through proof-of-concept implementations on 3 major CPU vendors (Intel, AMD, ARM). Our Systematization yields a more complete picture of the attack surface and allows for a more systematic evaluation of defenses. Through this systematic evaluation, we discover that most defenses, including deployed ones, cannot fully mitigate all attack variants.

KW - cs.CR

M3 - Conference contribution

T3 - arXiv.org e-Print archive

SP - 249

EP - 266

BT - Proceedings of the 28th USENIX Security Symposium

PB - USENIX Association

ER -