GNN-RE: Graph Neural Networks for Reverse Engineering of Gate-Level Netlists

Abstract

This work introduces a generic, machine learning (ML)-based platform for functional reverse engineering (RE) of circuits. Our proposed platform <monospace xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">GNN-RE</monospace> leverages the notion of graph neural networks (GNNs) to: 1) represent and analyze flattened/unstructured gate-level netlists; 2) automatically identify the boundaries between the modules or subcircuits implemented in such netlists; and 3) classify the subcircuits based on their functionalities. For GNNs in general, each graph node is tailored to learn about its own features and its neighboring nodes, which is a powerful approach for the detection of any kind of subgraphs of interest. For <monospace xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">GNN-RE</monospace> , in particular, each node represents a gate and is initialized with a feature vector that reflects on the functional and structural properties of its neighboring gates. <monospace xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">GNN-RE</monospace> also learns the global structure of the circuit, which facilitates identifying the boundaries between subcircuits in a flattened netlist. Initially, to provide high-quality data for training of <monospace xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">GNN-RE</monospace> , we deploy a comprehensive dataset of foundational designs/components with differing functionalities, implementation styles, bit widths, and interconnections. <monospace xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">GNN-RE</monospace> is then tested on the unseen shares of this custom dataset, as well as the EPFL benchmarks, the ISCAS-85 benchmarks, and the 74X series benchmarks. <monospace xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">GNN-RE</monospace> achieves an average accuracy of 98.82% in terms of mapping individual gates to modules, all without any manual intervention or postprocessing. We also release our code and source data.