Marek Materzok

Also there should be a howto for modifying Transactron and Coreblocks at the same time.

> Any clue why that test might be failing? It seems unrelated to me, but I'm not sure. Unfortunately, I don't know. There seems to be some nondeterminism in the...

Also, I'm not sure I like this change. Maybe it would be cleaner, and ultimately more useful, if models remembered their parent model. This way, you could reconstruct the "navigation...

> maybe that'll also fix the strange testing issue. Wouldn't count on it. The issue unfortunately exists on the master branch, and it is nondeterministic - re-running the test sometimes...

Possibly moving to WebAssembly compiled Yosys would allow more flexibility and reduce server dependency. Unfortunately, for my purposes I also need a linter (I currently use Verilator for that).

I'm sorry, I fail to understand your problem. Could you create a minimal testcase exhibiting the wrong behavior?

As far as I can see, the simulation works correctly. Number input doesn't change while typing, so that potentially wrong intermediate values don't show up on the wire. Input is...

This was a design decision. I didn't want incorrect signals to propagate so easily. I suggest two ways of mitigating this problem. One, use a color or some symbol (do...

> Whats more I am not sure if we want to have tests based on statistics, because this can cause instability. We're not supposed to implement some dreamed up hash...

The boilerplate reduction is nice, but this _is_ a little bit hacky. Maybe, instead of a custom `amaranth.Value` subclass, `typing.Annotated` could be used to store the same information?