Rtl simulators tested with grammar guided large model fuzzing
GI-RTL-Fuzz: Grammar-Informed RTL Generation with LLMs for End-to-End Differential Simulator Fuzzing
Hardware ArchitectureCryptography and Security
Summary
RTL simulators help design computer chips but can have bugs that cause errors. The authors introduce a tool that uses AI to create complex test programs for these simulators, guided by the rules of the programming language. This method finds hidden bugs by comparing outputs from different simulators. Their approach uncovered many new errors in how these simulators check and run designs.
What this means in practice
- •For integrated circuit design teams: Improve chip design reliability by finding and fixing RTL simulator bugs using AI-generated test cases guided by language grammar.
- •For hardware verification engineers: Help identify inconsistencies across different RTL simulators with automated tests to increase confidence in hardware simulation results.
Authors
Ziyue Zheng, Zhi Qu, Yangdi Lyu
Abstract
Register-Transfer Level (RTL) simulators are fundamental to modern IC design workflows, yet like any complex software, they frequently contain critical bugs that compromise simulation accuracy. To systematically stress-test these tools, we introduce GI-RTL-Fuzz, a grammar-informed fuzzing framework that leverages Large Language Models (LLMs) to generate high-quality RTL test cases. Our key innovation is guiding the LLM's generation process using Backus-Naur Form (BNF) grammar coverage information, enabling the exploration of complex and rarely-tested language features. By performing differential testing against different RTL simulators, our framework uncovered a significant number of previously unknown inconsistencies in both the linting process and simulation results of major Verilog simulators.