I have come across a company guidelines which require me to do separate code coverage closure for the register bank and the design(with register bank excluded). Is it a normal industry wide practice to do the code coverage closure that way? In any case, I would love to hear the thoughts on why we may want to do the coverage closure separately?
Cheers
It would be great if you would encourage the people who write these guidelines to put in the justifications behind them. I’ve seen coding guidelines put in because of tool bugs that were fixed a decade ago.
I have not come across this before, but a reason I can think for this rule is the register blocks are generated by some IP and tested separately, maybe even with formal tools instead of simulation.