Abstract
Leakage simulators offer the tantalising promise of easy and quick testing of software with respect to the presence of side channel leakage. The quality of their build in leakage models is therefore crucial, this includes the faithful inclusion of micro-architectural leakage. Micro-architectural leakage is a reality even on low- to mid-range commercial processors, such as the ARM Cortex M series. Dealing with it seems initially infeasible in a “grey box” setting: how should we describe it if micro-architectural elements are not publicly known? We demonstrate, for the first time, that it is feasible, using a recent leakage modelling technique, to reverse engineer significant elements of the micro-architectural leakage of a commercial processor. Our approach first recovers the micro-architectural leakage of each stage in the pipeline, and the leakage of elements that are known to produce glitches. Using the reverse engineered leakage features we build an enhanced version of the popular leakage simulator ELMO.
Original language | English |
---|---|
Title of host publication | Advances in Cryptology – EUROCRYPT 2022 - 41st Annual International Conference on the Theory and Applications of Cryptographic Techniques, 2022, Proceedings |
Editors | Orr Dunkelman, Stefan Dziembowski |
Publisher | Springer |
Pages | 284-311 |
Number of pages | 28 |
ISBN (Electronic) | 9783031070822 |
ISBN (Print) | 9783031070815 |
DOIs | |
Publication status | Published - 25 May 2022 |
Event | 41st Annual International Conference on the Theory and Applications of Cryptographic Techniques, EUROCRYPT 2022 - Trondheim, Norway Duration: 30 May 2022 → 3 Jun 2022 |
Publication series
Name | Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) |
---|---|
Volume | 13277 |
ISSN (Print) | 0302-9743 |
ISSN (Electronic) | 1611-3349 |
Conference
Conference | 41st Annual International Conference on the Theory and Applications of Cryptographic Techniques, EUROCRYPT 2022 |
---|---|
Country/Territory | Norway |
City | Trondheim |
Period | 30/05/22 → 3/06/22 |
Bibliographical note
Funding Information:Acknowledments. We would like to thank Ben Marshall for his invaluable insights, which guided us through various mazes in our leakage modelling efforts. Si Gao and Elisabeth Oswald were funded in part by the ERC via the grant SEAL (Project Reference 725042). This work has been supported in part by EPSRC via grant EP/R012288/1, under the RISE (http://www.ukrise.org) programme.
Publisher Copyright:
© 2022, International Association for Cryptologic Research.
ASJC Scopus subject areas
- Theoretical Computer Science
- General Computer Science