Rahix

> That looks great already, but I'm wondering if there's a way to turn that into an iterator based approach, so instead of `commit()` you'd simply iterate on your iterator...

@RandomInsano: I think we should move the discussion about owning addresses to a separate issue and keep this one purely about transactions. Would you be ok with opening a new...

> I think the read function above would need to take a mutable iterator for ownership to make sense? `read` does not take input, so it does not need an...

> Why? You can still easily transfer the values into your sized buffer, like: I'd agree that we don't need size specifications if we were to only allow single reads...

Hmm, if you want to leave the freedom of implementation you can use a pattern similar to the one used with eg. the [`ToggleableOutputPin`](https://docs.rs/embedded-hal/0.2.2/embedded_hal/digital/trait.ToggleableOutputPin.html)

Oh sorry, the name should probably be just `Mutex`, `BusMutex` is what it is called in `shared-bus`, I forgot to change it ...

Well, first of all I would just suggest to add a `Mutex` trait to *embedded-hal* that is implemented by each bsp. This allows drivers to have a generic synchronization primitive...

I want to make a point for keeping the `try_` prefix. I came to like it quite a bit while building a HAL against the alpha-version because it allows a...

Your question is very unspecific, which is probably the reason no one could answer it. 1. Your title seemingly has no connection to the question you asked. Can you clarify...

> Do we `impl SpiDevice for T`? In SPI it was not possible due to CS, but here it is possible. It can make usage easier (no need for wrapping...