Rupert Swarbrick

We've just run into another problem in OpenTitan where this would be really handy: There's a module `A` in our design that will instantiate a module, `B`, if some parameter,...

Following up on this, I fixed my installation, so this isn't urgent. But the code above is intrinsically wrong: there are two ways to get onto the error path, and...

Ahah. Now I understand what's going on. The problem is that the `split_var` hint works but there's also a big `always_comb` block that also needs splitting. I'm going to push...

@wallento: That would be perfect! @wsnyder: That's neat, but is a slightly awkward fit when you've got a long development cycle (like we're seeing with OpenTitan). There, it's nice if...

> Reading back on this I'm not quite clear how this was suggested to be used, @rswarbrick can you provide an example use case please? About the original "issue": it...

Looks like you're not using a supported version of Spike. See https://ibex-core.readthedocs.io/en/latest/03_reference/cosim.html for details of what you need.

Hmm, I just checked this locally and it seems to work for me, so I think it's probably a problem with your setup. I'd suggest taking a look at `build/lowrisc_ibex_ibex_simple_system_cosim_0/sim-verilator/config.mk`...

The easiest approach is probably to extend the stuff in `dv/uvm/core_ibex` to support the "please run this binary for me" flow. At the moment, that doesn't exist (it's on the...

For the `small` configuration, at least, the problem seems to be that we're running with no PMP. But the Spike configuration assumes otherwise! The way to get this to work...

@GregAC: What concrete steps do you think need taking before we can close this issue? At the moment, it'll probably remain open forever, which is not massively helpful.