Ryan Summers

I don't see any current support for interrupts in this crate, so any PR would need to be added to configure the interrupt. If you'd like to make a PR...

No worries, would be happy to guid. I imagine we want a new API added to `UninitializedDevice: ``` pub fn enable_interrupt(&mut self, interrupt: InterruptType) ``` or something similar. Also we...

`accept` would need to return a _different_, _newly-allocated_ socket. That's why it hasn't been implemented - TCP servers are way more complex and aren't generally implemented on embedded targets because...

Looks good to me! I'll add CHANGELOG entry. Thanks!

Hey there, I'm reading through your post and addressing specific points as I see fit. As some background, I've been a firmware developer for the last ~8 years, most of...

Have you checked out [`probe-rs`](https://github.com/probe-rs/probe-rs) yet? I know they target mainly ARM cores, but I think it would be well-worth checking out the work there and potentially contributing with them....

I'm happy to accept a PR for a new opt-in feature, but I do not want to adopt 1.0.0 of the e-h as the default until it is formally released.

Actually, with the advent of `embedded-hal` v1.0, this crate is no longer needed. `embedded-hal-bus` handles the synchronization mechanisms required. I even added in support for the RTIC mutexes directly: *...

The summary that @jordens has above for the examples makes the most sense with what already exists in `miniconf` for sure. While the extra `/0` on newtypes is annoying, I...

> Such that (using your snippet) `/foo` would have serialized value `Bar` This seems reasonable to me, since there's no other value represented. So for the simple unit enum case,...