Is it a good practice to have the agent config class handle in the agent’s interface. In the interface, I would like to use some knobs from the agent’s config class to write some assertions. What are the best practices for such kind of use case.
In reply to yasaswi93:
It depends. Some methodologies rely on interfaces to remain synthesizable for for use in the design, or integration with hardware accelerators. But you can easily create “drivers” to set your knobs from your class based testbench in that case.