Adaptive Test Generation for IC Functional Coverage
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing coverage-driven test generation methods for integrated circuit design verification rely heavily on human expertise and costly data-driven techniques, leading to significant overhead and scalability issues.
Innovation Solution
A system that automatically generates test stimuli by extracting a coverage dependency graph and adaptively updating distribution constraints for random variables, thereby improving coverage and reducing verification overhead.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If coverage-driven test generation uses manual verification expert updates to increase coverage, then test coverage is improved, but verification overhead and time consumption increase significantly
Solution Approach 1:
The system enables self-service by automatically extracting coverage dependency graphs and updating distribution constraints without requiring manual verification expert intervention. The automated mechanism analyzes coverage statistics and independently determines optimal test stimulus distributions, allowing the verification system to improve coverage without proportional increases in human expertise time.
Solution Approach 2:
The system implements feedback by continuously monitoring coverage statistics from simulation cycles and using this information to update distribution constraints. The coverage dependency graph captures relationships between test stimuli and coverage metrics, enabling the system to adaptively adjust test generation strategies based on real-time coverage performance to achieve optimal coverage closure.
2Manufacturing precision
If coverage-driven test generation uses data driven techniques to learn coverage-to-input relationships, then test coverage is improved, but computational cost and overhead increase
Solution Approach 1:
The system extracts the coverage dependency graph from the verification data, separating the essential coverage-to-input relationships from the complex simulation data. By extracting only the critical dependency information needed for distribution constraint updates, the system achieves improved coverage without requiring costly comprehensive data-driven learning models.
Solution Approach 2:
The system changes parameters by updating distribution constraints based on extracted coverage dependencies rather than using complex data-driven models. This parameter-based approach modifies the test stimulus generation directly through coverage-guided distribution adjustments, achieving improved coverage with lower computational overhead compared to full data-driven learning techniques.
3Loss of time
If the system automatically updates distribution constraints using coverage dependency graphs, then verification overhead is reduced, but the complexity of the test generation mechanism increases
Solution Approach 1:
The system segments the test generation process into distinct modular components: coverage dependency graph extraction, coverage statistics collection, distribution constraint updating, and test stimulus generation. This segmentation allows each component to be independently optimized and maintained, reducing overall system complexity despite the automated nature of the process.
Solution Approach 2:
The coverage dependency graph serves as an intermediary structure that mediates between coverage statistics and distribution constraint updates. This intermediate representation simplifies the relationship between raw coverage data and test stimulus generation, making the automated mechanism more manageable and less complex while still achieving significant overhead reduction.
Data Source
AI summary
Methods, systems, and apparatus for adaptively generating test stimuli for testing a hardware design for an integrated circuit. In one aspect, a system comprises one or more computers configured to obtain graph data representing a coverage dependency graph associated with a hardware design for an integrated circuit. At each of a plurality of simulation cycles, the one or more computers obtain a set of coverage statistics as of the simulation cycle and update respective distribution constraints for one or more random variables in a set of random variables using the coverage dependency graph and the coverage statistics. After the updating, the one or more computers generate one or more test stimuli by, for each test stimulus, sampling a respective value for each of the random variables based on the respective distribution constraints. The one or more computers simulate a performance of the integrated circuit for each of the test stimuli.