A differential fuzz testing approach for CPU verification that uses register-coverage to find bugs in RTL designs.
DifuzzRTL is a differential fuzz testing framework for CPU verification that automatically instruments register-coverage into RTL designs, generates and mutates ISA instructions, and cross-checks results against an ISA simulator to detect bugs. It was presented at IEEE S&P 2021 and introduces a novel coverage metric to comprehensively capture RTL design states.
CPU designers, hardware verification engineers, and security researchers working on RTL implementations who need systematic bug detection in CPU designs.
DifuzzRTL offers a novel register-coverage metric that better captures RTL design states compared to traditional coverage approaches, combined with differential testing against ISA simulators for more effective CPU bug detection.
DifuzzRTL is a differential fuzz testing framework specifically designed for CPU verification. It introduces a novel coverage metric called register-coverage to comprehensively capture RTL design states and guide input generation, then cross-checks results against an ISA simulator to detect bugs.
DifuzzRTL takes a differential approach to CPU verification, combining coverage-guided fuzzing with cross-checking against reference models to systematically uncover bugs in RTL implementations.
Open-Awesome is built by the community, for the community. Submit a project, suggest an awesome list, or help improve the catalog on GitHub.