In this work, we hope to expand the universe of security practitioners of open-source hardware by creating a bridge from hardware design languages (HDLs) to data science languages like Python and R through novel libraries that convert VCD (value change dump) files into data frames, the expected input type of the modern data science tools. We show how insights can be derived in high-level languages from register transfer level (RTL) trace data. Additionally, we show a promising future direction in hardware security research leveraging the parallelism of Spark to study transient execution CPU vulnerabilities, and provide reproducibility researchers via GitHub and Colab.
@article{arxiv.2505.06470,
title = {"vcd2df" -- Leveraging Data Science Insights for Hardware Security Research},
author = {Calvin Deutschbein and Jimmy Ostler and Hriday Raj},
journal= {arXiv preprint arXiv:2505.06470},
year = {2025}
}
Comments
6 pages, no figures, camera ready submission after acceptance at ACDSA 2025. Added co-author Hriday Raj during v2 as Hriday joined us for some Spark characterization as a domain specialist between initial drafts and reaching camera-readiness