Synopsys Cloud
Cloud native EDA tools & pre-optimized hardware platforms
Request a Free Trial
As the complexity of SoCs is increasing, so is the complexity and verification of IPs and development of VIPs used to verify it. Evolving protocol standards require a dynamic test suite for the IPs and VIPs and consistent requirements to meet mandated functional and code coverage metrics. This typically requires:
- lots of simulation cycles
- large compute resources due to running redundant tests
- and significant manual effort in debug to close the coverage holes
How does VSO.ai help?
Synopsys VSO.ai addresses these challenges in functional coverage closure with its advanced architecture, including coverage connectivity engines, a coverage-directed constraint solver, and comprehensive coverage root cause analysis and reporting.
VSO.ai works at the coarse-grained test level which integrates with the test environment and provides automated, adaptive test optimization that learns as the results change. Running the tests with the highest ROI first while eliminating redundant tests accelerates coverage closure and saves compute resources.
The tool also works at the fine-grained level within the simulator to improve the test quality of results (QoR) by adapting the constrained-random stimulus to better target unexercised coverage points. This not only accelerates coverage closure but also drives convergence to a higher percentage value.
The last mile closure challenge is addressed by automated, AI-driven analysis of coverage results. VSO.ai performs root cause analysis (RCA) to determine why specific coverage points are not being reached. If the tool can resolve the situation itself, it will. Otherwise, it presents the team with actionable results, such as identifying conflicting constraints. It integrates with tools such as URG and Synopsys Verdi® for better visualization and analysis of constraints.
How Synopsys Optimized IP Regression Flows
The Synopsys IP team was looking to optimize the flows for IPs by bringing the regression sign-off to under 3 days and Synopsys’ own VSO.ai was instrumental in achieving that goal.
During their exploration, VSO.ai was tried on the USB3x digital IP and mixed signal PHY layer IPs on the medium-large sized regressions of up to 4000 tests. The Synopsys IP MPHY team deployed VSO.ai through the plug-in to Synopsys VC Execution Manager, a regression management tool, making for a seamless experience in execution and continuous tuning of the ML model for the best QoR and TTR. The VSO.ai plug-in in VC Execution Manager offered a single cockpit solution for regression and coverage results, optimization outcomes and health of the ML model.
As seen in results in the table below, VSO.ai has been integrated as part of the release regressions for these IPs which significantly reduced the need for manual intervention.
| Default Run | Ask-tell Runs | Gain | ||||
|---|---|---|---|---|---|---|
| Regression | No. of Total Runs | Regression Run Time in Hours | No. of Total Runs | Regression Runtime in Hours | Reduction in No. of Runs | Reduction in Runtime |
| USB3x Feature 1 | 1225 | 55 | 770 | 35 | 1.59X | 1.57X |
| USB3x Feature 2 | 666 | 25 | 334 | 11.5 | 1.99X | 2.17X |
When tested on PCIe GEN6 PHY the same functional coverage of 61% was achieved in 3X fewer runs (226 vs. 710).
Results with PCIe GEN6 PHY
VSO.ai was also able to demonstrate similar results with code coverage where the same Line, FSM and Toggle coverages could be achieved in 2x fewer tests.
Results with AMBA CHI VIP
Concluding Results
The implementation of VSO.ai demonstrated remarkable improvements in the regression testing process for IP and VIP environments. The ability to achieve a 10x reduction in the number of seeds required to reach 77% coverage, and subsequently improving to 97% coverage by addressing specific testbench constructs, underscores the efficacy of VSO.ai in optimizing test environments. Even for a mature VIP like AMBA, VSO.ai has proven its value by enhancing the quality of the testbench environment. By leveraging this unique and automated solution, teams can significantly boost their confidence in delivering high-quality IPs and VIPs, while effectively addressing common challenges related to quality, resource management, and efficiency.
Continue Reading
Blog
6 min read
/
Mar 31, 2025
A Novel Approach to Speeding Up PCIe 6.0+ Designs - Compliance Load Board Verification
Blog
4 min read
/
Jan 27, 2025
Validation of AUTOSAR Classic ECUs Running Rust SWCs: A Safer Path to Automotive Software
Blog
2 min read
/
Sep 12, 2024
Industry First Verification IP for Arm AMBA DTI-G
