I think it would be a good idea to employ an error detection and correction (EDAC) scheme for everything stored on the FRAM. That is the persistent state and the telemetry memory. Since the FRAM is not very big, triple modular redundancy with its 200 % memory overhead is not feasible for large portions of the data (like the telemetry memory) but maybe for the persistent variables. On the other hand, it's probably a good idea to use a single EDAC scheme for "everything".
We don't have the resources to do this properly now, so I'll move this to the lowest priority and unblock other issues previously depending on this one.
Description
I think it would be a good idea to employ an error detection and correction (EDAC) scheme for everything stored on the FRAM. That is the persistent state and the telemetry memory. Since the FRAM is not very big, triple modular redundancy with its 200 % memory overhead is not feasible for large portions of the data (like the telemetry memory) but maybe for the persistent variables. On the other hand, it's probably a good idea to use a single EDAC scheme for "everything".