SPRUJ28E November 2021 – September 2024 AM68 , AM68A , TDA4AL-Q1 , TDA4VE-Q1 , TDA4VL-Q1
ECC is a mechanism for providing increased protection against soft errors by allowing single bit errors to be detected and corrected and double bit errors to be detected. The ECC Control Register can be used to inject single bit or double bit errors. For more information on this register, refer to Section 12.10.4.2.2. These errors are then reported through the Error Signal Module (ESM).
The ECC aggregator plays a crucial role in maintaining data accuracy by correcting single bit errors and detecting double bit errors, although it cannot correct the latter. This process helps in enhancing the overall system reliability. Each ECC aggregator connects to two kinds of ESM events – Correctable error events (for SEC events) and Un-correctable error events (for DED events and parity and redundancy checker types). ESM is used for notification of events from endpoints which are not Inject-Only. Inject-Only endpoints notify errors through IP-specific mechanisms. Sometimes a separate ESM event (not via ECC aggregator) is available. Please refer to the individual IP sections for more details on the ESM events generated.
The ECC aggregator provides support for injecting single and double bit errors. In the case of RAM wrapper type endpoints, errors can be injected to a specific row or can be requested to be injected on the next row access (n_row). Errors can be requested to be injected once or on repeated reads. For Interconnect endpoint types, single and double bit errors (depending on the checker type) can be injected.