Björn Quentin

> The obscure stacks are a little unhelpful. Getting useful backtraces would be good

Maybe we should disable the peripheral on drop (a `System::disable` would be needed) I am also thinking if we should actually reset the peripheral in `enable` ... not just clear...

That it's only happening in release mode doesn't make much sense to me but from a brief look something like this ```rust i8080.send(CMD_CSCON, 0, &[0xC3]).unwrap(); // Enable extension command 2...

I thought about adding the "valid_ram_address" check in `prepare_transfer_without_start` but then I was wondering why using DROM doesn't trigger a DMA error and found this  Seems there are a...

> Short term though, should the example be patched to ensure the buffers are in DRAM? Or should it wait for the driver to be changed? I don't have a...

> ... and just generally improve examples as able I think some of the examples are really tests (especially the crypto-accelerators but also others). These are better turned into HIL...

Maybe it would be good to have a way to specify which core the interrupt should be handled on. IIRC we are considering removing the interrupt handler from the constructors...

this will be taken over by @JurajSadel

Thanks! Making `read_byte` public is fine but there are a few problems with this PR - public functions should include API docs - this adds a changelog in the root...

@georgik during the pre-release test for the current esp-hal release I updated https://github.com/bjoernQ/rust-esp32s3-ili9341 - that might help