Sonata Pinmux driver

[HU90m]

A driver for Sonata v1.0's Pinmux

[marnovandermaas]

Thanks Hugo, this looks correct to me from a code review.

[HU90m]

I think in the case of the Pinmux it's best to make sure sink names match the canonical name used documentation, RTL and schematic. The sinks are enum classes, so breaking the code style shouldn't cause much confusion here.

[nwf]

It looks like platform-pinmux.hh is generated from sonata-system's platform-pinmux.hh.tpl via its top_gen. It'd be good to...

• have a comment to that effect in the file
• update the .hh.tpl in light of the changes made here (NOLINT, names of some function arguments and locals, &c).

FWIW, I ran git diff --word-diff having clobbered sonata-system v0.3-403-g658a741's sw/cheri/common/platform-pinmux.hh with the one from this PR; that might be the easiest way to find the requisite changes to the template?

I have not attempted to validate that top_gen is consistent between the RTL and C++ it generates, much preferring to merely assume so. :)

[HU90m]

↑ rebased onto main, ↓ added comment pointing to what generates the file.

See lowRISC/sonata-system#376

[nwf]

Thanks for doing that. One nit aside, this LGTM.

[marnovandermaas]

lowRISC/sonata-system#376 is now merged.

[nwf]

Want to update the platform-pinmux.hh herein with the result of that PR? I think it's good to go after that.

[HU90m]

Want to update the platform-pinmux.hh herein with the result of that PR? I think it's good to go after that.

The sonata-system repo uses the lowRISC C++ formatting, unlike sonata-software, so the output of the topgen script has to be lowRISC formatted.

This PR has already been updated with the updated with the latest output, only run through clang-format with the CHERIoT settings.

[davidchisnall]

I missed in review, but this seems to have an Apache license. Is it possible to relicense to match the rest of the repository?