I just read this article (Accellera DAC 2014 Breakfast—What Engineers Really Think About UVM). They hit a few interesting points.
I'd like to add another few points that I think were missed in this discussion.
So after I described my troubles with the UVM register model's coverage in my previous post, I revisited the issue in a slightly different context and discovered the current solution was rather lacking.
In the previous example, coverage was taken for a value written via the register …
A couple of weeks ago, I created a status register callback class.
The purpose of it was to create a built-in scoreboard allowing me to predict the value of a status register based on the known status signal's input.
I'll maybe post the code in a sometime in the future …
Read moreBack when I started using UVM, I had no use for a register model for the specific unit I was testing. As such, sequences were written which needed a sequencer which would hand the items in the sequence to the driver.
Over time, we moved to use the register model …
Read more