Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

OpenLS-DGF: An Adaptive Open-Source Dataset Generation Framework for Machine Learning Tasks in Logic Synthesis

Liwei Ni, Rui Wang, Miao Liu, Xingyu Meng, Xiaoze Lin, Junfeng Liu, Guojie Luo, Zhufei Chu, Weikang Qian, Xiaoyan Yang, Biwei Xie, Xingquan Li, Huawei Li

TL;DR

The versatility of OpenLS-D-v1 is demonstrated through four distinct downstream tasks: circuit classification, circuit ranking, quality of results (QoR) prediction, and probability prediction, and the experimental results show the generated dataset from OpenLS-DGF achieves prominent diversity and applicability.

Abstract

This paper introduces OpenLS-DGF, an adaptive logic synthesis dataset generation framework, to enhance machine learning~(ML) applications within the logic synthesis process. Previous dataset generation flows were tailored for specific tasks or lacked integrated machine learning capabilities. While OpenLS-DGF supports various machine learning tasks by encapsulating the three fundamental steps of logic synthesis: Boolean representation, logic optimization, and technology mapping. It preserves the original information in both Verilog and machine-learning-friendly GraphML formats. The verilog files offer semi-customizable capabilities, enabling researchers to insert additional steps and incrementally refine the generated dataset. Furthermore, OpenLS-DGF includes an adaptive circuit engine that facilitates the final dataset management and downstream tasks. The generated OpenLS-D-v1 dataset comprises 46 combinational designs from established benchmarks, totaling over 966,000 Boolean circuits. OpenLS-D-v1 supports integrating new data features, making it more versatile for new challenges. This paper demonstrates the versatility of OpenLS-D-v1 through four distinct downstream tasks: circuit classification, circuit ranking, quality of results (QoR) prediction, and probability prediction. Each task is chosen to represent essential steps of logic synthesis, and the experimental results show the generated dataset from OpenLS-DGF achieves prominent diversity and applicability. The source code and datasets are available at https://github.com/Logic-Factory/ACE/blob/master/OpenLS-DGF/readme.md.

OpenLS-DGF: An Adaptive Open-Source Dataset Generation Framework for Machine Learning Tasks in Logic Synthesis

TL;DR

The versatility of OpenLS-D-v1 is demonstrated through four distinct downstream tasks: circuit classification, circuit ranking, quality of results (QoR) prediction, and probability prediction, and the experimental results show the generated dataset from OpenLS-DGF achieves prominent diversity and applicability.

Abstract

This paper introduces OpenLS-DGF, an adaptive logic synthesis dataset generation framework, to enhance machine learning~(ML) applications within the logic synthesis process. Previous dataset generation flows were tailored for specific tasks or lacked integrated machine learning capabilities. While OpenLS-DGF supports various machine learning tasks by encapsulating the three fundamental steps of logic synthesis: Boolean representation, logic optimization, and technology mapping. It preserves the original information in both Verilog and machine-learning-friendly GraphML formats. The verilog files offer semi-customizable capabilities, enabling researchers to insert additional steps and incrementally refine the generated dataset. Furthermore, OpenLS-DGF includes an adaptive circuit engine that facilitates the final dataset management and downstream tasks. The generated OpenLS-D-v1 dataset comprises 46 combinational designs from established benchmarks, totaling over 966,000 Boolean circuits. OpenLS-D-v1 supports integrating new data features, making it more versatile for new challenges. This paper demonstrates the versatility of OpenLS-D-v1 through four distinct downstream tasks: circuit classification, circuit ranking, quality of results (QoR) prediction, and probability prediction. Each task is chosen to represent essential steps of logic synthesis, and the experimental results show the generated dataset from OpenLS-DGF achieves prominent diversity and applicability. The source code and datasets are available at https://github.com/Logic-Factory/ACE/blob/master/OpenLS-DGF/readme.md.

Paper Structure

This paper contains 53 sections, 2 theorems, 5 equations, 17 figures, 8 tables.

Table of Contents

  1. Introduction
  2. Background and Related Work
  3. Background
  4. Boolean circuit and Functional completeness
  5. Logic Optimization and Technology Mapping
  6. Static Timing Analysis
  7. Graph Neural Network (GNN)
  8. Related Work
  9. OpenLS-DGF
  10. Overview
  11. Dataset Generation Steps
  12. Step 1: Generic Technology Circuit Synthesis.
  13. Step 2: And-Inverter Graph Generation.
  14. Step 3: Logic Optimization Recipes.
  15. Step 4: Logic Blasting.
  16. ...and 38 more sections

Key Result

Theorem 1

The logic blasting method preserves the dependency relationships of the original circuit.

Figures (17)

  • Figure 1: The Logic Synthesis flow.
  • Figure 2: The adaptive logic synthesis dataset generation framework of OpenLS-DGF.
  • Figure 3: Components of an item in OpenLS-D-v1. For convenience, the logic types {OIG, XAG, MIG, PRIMARY} are abbreviated with ellipses.
  • Figure 4: The UML class diagram of the generic circuit class.
  • Figure 5: Node correspondence between the three types of graph: Boolean circuit, Circuit, and torch_geometry.
  • ...and 12 more figures

Theorems & Definitions (4)

  • Definition 1: Functional completeness
  • Theorem 1
  • Proof 1
  • Lemma 1