SNIP_CLASSIC_STD_SECTION_5_2_6_8_Error_Handling

5.2.6.8 Error Handling In RTE, error and status information is defined with the data type Std_ReturnType, see Section 5.5.1. It is possible to distinguish between infrastructure errors and application errors. Infrastructure errors are caused by a resource failure or an invalid input parameter. Infrastructure errors usually occur in the basic software or hardware along the communication path of a data element. Application errors are reported by a SW-C or by AUTOSAR services. RTE has the capability to treat application errors that are forwarded • by return value in client server communication or • by signal invalidation in sender receiver communication with data semantics. Errors that are detected during an RTE API call are notified to the caller using the API’s return value. [SWS_Rte_01034] Upstream requirements: SRS_Rte_00094 ⌈Error states (including ’no error’) shall only be passed as return value of the RTE API to the AUTOSAR SW-C.⌋ Requirement [SWS_Rte_01034] ensures that, irrespective of whether the API is blocking or non-blocking, the error is collected at the same time the data is made available to the caller thus ensuring that both items are accessed consistently. Certain RTE API calls operate asynchronously from the underlying communication mechanism. In this case, the return value from the API indicates only errors detected during that API call. Errors detected after the API has terminated are returned using a different mechanism [SWS_Rte_01111]. RTE also provides an ’implicit’ API for direct access to virtually shared memory. This API does not return any errors. The underlying communication is decoupled. Instead, an API is provided to pick up the current status of the corresponding data element.