Saved in:
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2509.04010 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866912570163593216 |
|---|---|
| author | Adjonyo, Olivier Bardin, Sebastien Bellini, Emanuele Dione, Gilbert Ndollane Ameen, Mahmudul Faisal Al Merget, Robert Recoules, Frederic Sellami, Yanis |
| author_facet | Adjonyo, Olivier Bardin, Sebastien Bellini, Emanuele Dione, Gilbert Ndollane Ameen, Mahmudul Faisal Al Merget, Robert Recoules, Frederic Sellami, Yanis |
| contents | The PQDSS standardization process requires cryptographic primitives to be free from vulnerabilities, including timing and cache side-channels. Resistance to timing leakage is therefore an essential property, and achieving this typically relies on software implementations that follow constant-time principles. Moreover, ensuring that all implementations are constant-time is crucial for fair performance comparisons, as secure implementations often incur additional overhead. Such analysis also helps identify scheme proposals that are inherently difficult to implement in constant time. Because constant-time properties can be broken during compilation, it is often necessary to analyze the compiled binary directly. Since manual binary analysis is extremely challenging, automated analysis becomes highly important. Although several tools exist to assist with such analysis, they often have usability limitations and are difficult to set up correctly. To support the developers besides the NIST committee in verifying candidates, we developed a toolchain that automates configuration, execution, and result analysis for several widely used constant-time analysis tools. We selected TIMECOP and Binsec/Rel2 to verify constant-time policy compliance at the binary level, and dudect and RTLF to detect side-channel vulnerabilities through statistical analysis of execution time behavior. We demonstrate its effectiveness and practicability by evaluating the NIST PQDSS round 1 and round 2 implementations. We reported 26 issues in total to the respective developers, and 5 of them have already been fixed. We also discuss our different findings, as well as the benefits of shortcomings of the different tools. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2509_04010 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Systematic Timing Leakage Analysis of NIST PQDSS Candidates: Tooling and Lessons Learned Adjonyo, Olivier Bardin, Sebastien Bellini, Emanuele Dione, Gilbert Ndollane Ameen, Mahmudul Faisal Al Merget, Robert Recoules, Frederic Sellami, Yanis Cryptography and Security The PQDSS standardization process requires cryptographic primitives to be free from vulnerabilities, including timing and cache side-channels. Resistance to timing leakage is therefore an essential property, and achieving this typically relies on software implementations that follow constant-time principles. Moreover, ensuring that all implementations are constant-time is crucial for fair performance comparisons, as secure implementations often incur additional overhead. Such analysis also helps identify scheme proposals that are inherently difficult to implement in constant time. Because constant-time properties can be broken during compilation, it is often necessary to analyze the compiled binary directly. Since manual binary analysis is extremely challenging, automated analysis becomes highly important. Although several tools exist to assist with such analysis, they often have usability limitations and are difficult to set up correctly. To support the developers besides the NIST committee in verifying candidates, we developed a toolchain that automates configuration, execution, and result analysis for several widely used constant-time analysis tools. We selected TIMECOP and Binsec/Rel2 to verify constant-time policy compliance at the binary level, and dudect and RTLF to detect side-channel vulnerabilities through statistical analysis of execution time behavior. We demonstrate its effectiveness and practicability by evaluating the NIST PQDSS round 1 and round 2 implementations. We reported 26 issues in total to the respective developers, and 5 of them have already been fixed. We also discuss our different findings, as well as the benefits of shortcomings of the different tools. |
| title | Systematic Timing Leakage Analysis of NIST PQDSS Candidates: Tooling and Lessons Learned |
| topic | Cryptography and Security |
| url | https://arxiv.org/abs/2509.04010 |