Trait ChipImpl

Source

pub trait ChipImpl:
    HlComms
    + Send
    + Sync
    + 'static {
    // Required methods
    fn update_init_state(
        &mut self,
        status: &mut InitStatus,
    ) -> Result<ChipInitResult, PlatformError>;
    fn get_arch(&self) -> Arch;
    fn get_telemetry(&self) -> Result<Telemetry, PlatformError>;
    fn arc_msg(&self, msg: ArcMsgOptions) -> Result<ArcMsgOk, PlatformError>;
    fn get_neighbouring_chips(
        &self,
    ) -> Result<Vec<NeighbouringChip>, PlatformError>;
    fn as_any(&self) -> &dyn Any;
    fn get_device_info(&self) -> Result<Option<DeviceInfo>, PlatformError>;
}

Expand description

Defines common functionality for all chips. This is a convinence interface that allows chip type agnostic code to be written.

As a general rule the chip should not be accessed without an explicit request from the user. This means that chip initialization must be explicity called and for example if the user has not explicity stated that they want to enumerate remote chips, then we won’t even start looking at remote readiness. This is to avoid situations where a problematic state is reached and causes an abort even if that capability is not needed.

Required Methods§

Source

fn update_init_state( &mut self, status: &mut InitStatus, ) -> Result<ChipInitResult, PlatformError>

Update the initialization state of the chip. The primary purpose of this function is to tell the caller when it is safe to starting interacting with the chip.

However the secondary purpose is to provide information about what chip functions are currently available for use. For example if the arc is not ready, then we should not try to send an arc message. Or in a more complex example, if the arc is ready, but the ethernet is not (for example the ethernet fw is hung) then we will be able to access the local arc, but won’t be able to access any remote chips.

Source