@ROMemories where are we with this?

where are we with this?

So that one is annoying, because the esp-hal crate selects a runtime crate (*-rt) based on the processor architecture (riscv32 or xtensa):

https://github.com/esp-rs/esp-hal/blob/0dc8dcf8e2f25bf0c26f41a1cb2bbcb98433bceb/esp-hal/Cargo.toml#L66-L70

If none is selected then we get the errors the CI is currently failing with ("unresolved import esp_riscv_rt"); however, if we set a processor architecture using --target=riscv32imac-unknown-none-elf for instance, then we get the following error:

error[E0463]: can't find crate for `test`

error: `#[panic_handler]` function required, but not found

For more information about this error, try `rustc --explain E0463`.
error: could not compile `riot-rs-esp` (lib test) due to 2 previous errors

I'm not sure what the right move would be here; should we try to patch esp-hal to select the runtime crate based on the already existing riscv and xtensa features (in your fork)?