You can't, but you could create a test bench and verify that the behavior is cycle acurate the same as a simulated openc910. That's probably better than doing n

You can't, but you could create a test bench and verify that the behavior is cycle acurate the same as a simulated openc910. That's probably better than doing nothing. We know that openC910 isn't the exact same chip as the C910, as it doesn't include the draft vector extension.