Oral/Lecture

Adaptive Test Generation for Fast Functional Coverage Closure

In this work, we propose a low-overhead framework that automatically extracts the coverage dependency graph associated with a design, and leverages the dependency information for adaptive test stimuli generation. Our experimental results, on open-source as well as large-scale real-world designs results, confirm that our approach significantly speeds up the coverage convergence with no human effort required.

Azade Nazi, Google Research
Qijing Huang, UC Berkeley
Hamid Shojaei, Google
Hodjat Asghari Esfeden, Google
Azalia Mirhosseini, Google Research, Brain
Richard Ho, Google