Rachit Nigam

One other possibility is allowing for some kind of `fork` and `join` primitive. It might be particularly interesting if you were allowed to `fork` an FSM's execution and then `join`...

I just realized that the above won't work because: ``` 1: { b; t { b.done ? 2 : 1 } ``` Would repeatedly attempt to `fork thread`. We need...

Huh, I _think_ that would love my problem! Wonderful catch!

Hm, I'm pretty sure this is a minimal version because any implementation will require a mechanism to trigger several FSMs in parallel (`fork`) and check their `done` signal (`join`).

Hey @KelvinChung2000! Thanks for checking in. Are you interested in the FSM implementation for a specific reason?

#2404 might help solve this

Yup, you need to find the right table using the parser’s API and then extract the information!

Yup, looks like the right place to me. It is surprising and perhaps a little concerning that no BRAMs are used. Maybe the internal memories used in FIFOs are not...

@ethanuppal what's the status of this PR and what would it take to get it merged?

There are two options: 1. We figure out a different approach for the testbench that does not rely on the macro magic but is still useful for users writing complex...