HAL: Upstream from OVL works

[jamesmunns]

Hey, I've been working on ATSAME70 support specifically in a repo as part of some contracting work (OVL => OneVariable UG Labs).

Since I've been focused on just this one chip, and a relatively limited set of features, it's probably not ready to merge 1:1, but just in case someone is interested, the repo is here, and I've published the crate under the name ovl-atsamx7x-hal for now, until this work gets upstreamed.

So far, I have the following components implemented:

• EFC - basically just the ability to set wait states
• GMAC - some corners rounded, really only tested to work with the KSZ8061 PHY, but capable of "doing ethernet" and has a smoltcp implementation capable of maintaining a TCP connection. Only supports one set of pins ATM
• PIOx - generally all 5 ports are supported, and some pin functionality is working (mostly setting ABCD peripheral modes, and output pins)
• PMC - Pretty rudimentary support, but enough to set some prescalers. I mostly use it to reach a single clock config (300MHz CPU clock, 150MHz MCK)
• RTT - A fixed 8192kHz frequency, usable for rtic-monotonic::Monotonic and groundhog::RollingTimer trait implementations
• Serial - I haven't really touched the upstream code here, I'm not using serial
• SPI - Basic "transfer" API support for the SPI0 peripheral. Only supports one set of pins ATM
• WDT - Basically just enough to shut it off

Feel free to take any of this work if you want/need it before I get a chance to upstream! I've kept the 0BSD license, so you can do what you'd like.

Edit (2022-04-22): Added RTT impl

[jamesmunns]

Oh, there are also quite a few working examples in the same70-bringup project, available here.

[tmplt]

Thanks for the notice. I'll take a more thorough look at it when able.

[michalfita]

@tmplt is there anything from that @jamesmunns work here that wasn't in #25?

[tmplt]

Only the GMAC and groundhog::RollingTimer implementations. I'll push a WIP PR for the former tomorrow. I do not consider the latter in-scope for the HAL currently.

[tmplt]

Superseded by #33. This issue can be closed.

[michalfita]

The #33 is a PR, as well as #32 and #31 which are continuation of work from #24.