Nowadays the advanced applications required in our lives have led to a significant increase in the complexity of circuits, which enhances the possibility of occurring design errors. Hence an automated, powerful, and scalable debugging approach is needed. Therefore, this paper proposes a scalable approach for localizing multiple bugs in Register-Transfer level (RTL) designs by using neural networks. The main idea is that hit-statements which are covered by failed test-vectors are more suspicious than those covered by passed test-vectors. We use coverage data as samples of our data set, label these samples, and tune the neural network model. Then we encode hit-statements and give them to the tuned model as new samples. The model classifies hit-statements. Hit-statements that take the failed labels, labels related to the failed test-vectors, are more suspicious of containing bugs. The results demonstrate that the proposed methodology outperforms recent approaches Tarsel and CirFix by localizing 80% of bugs at Top-1. The results also imply that our methodology increases the F1-score metric by 1.13× in comparison with existing RTL debugging techniques, which are prediction-based.

Existing debugging and correction approaches suffer from weaknesses such as scalability, reproducing new bugs, and lacking a strategy to deal with multiple bugs. Hence, this article proposes FixRTL, a fully automated scalable methodology for localizing and correcting multiple bugs in Register-Transfer level (RTL) designs. FixRTL consists of three phases: (1) Constructing Samples , (2) Debugging , and (3) Correction . First, we simulate the design under verification (DUV), extract coverage data, and construct our samples. Since we are looking for buggy hit-statements, we use the proposed feature burst (FB) clustering algorithm in the Debugging Phase . The algorithm applies samples as train data, categorizes the encoded hit-statements into bursts, and uses them as test data to predict their cluster. Then we rank hit-statements based on their probability of containing bugs per cluster. In the Correction Phase , we apply a proposed mutation-based framework to correct high-ranked hit-statements. The results show that FixRTL reduces the percentage of hit-statements that must be examined to localize bugs on average by 44.3%. The results also demonstrate that FixRTL corrects 67% of injected bugs while recent existing works correct up to 25%. Moreover, unlike recent works, FixRTL offers corrections that match the grand truth.