How to control test flow based on a sequence being finished some time later by rtl

UVM
,
1

My DUT is a processing block. I simply have to send it data, and then at the other end, process the expected data and keep track of the processed data so I know when it is “done”.

The idea would be to then send more data in, wait for “done” and then loop through.

I think I should be doing this in the test, so the test would start my sequence at agent A, while agent B could capture the processed data and know when it is done.

But I don’t know how agent B can signal to the test that it is done. What is the best way to solve this?

2

Sequences can not only be used to send data to the DUT, but they can also be used to capture data from the DUT. You can create a sequence for agent B that is designed to receive the required amount of data from the DUT, and when the sequence is complete, you know that you are finished.

3

In reply to cgales:

Interesting. I understand in principle, but not necessarily practice. Do you have any examples you could point me to?