SCST alarm handling

[Aravind1]

hi,

we are working on s32k342 device .

1)we have few confusion's regarding SCST alarm handling.in SCST we are performing total 50 check's if any one of SCST check fail's means how we can
handle that check is there any prevention or it will be like a core issue(permanent fault).

2)is there any way to decide which are critical fault's and which are the non-critical fault's for handling alarm's .

3)if one power cycle SCST test fail mean whether it will be recover in next power cycle, if it recover means how we can ensure.

[danielmartynek]

Hi @Aravind1,

1.

Note that SCST is software responsible for permanent HW faults detecteion. In case a test periodically fails, this means that there is permanent HW fault in the core. Then, it is the application responsibility to react on such a fault.

2.

In general, all permanent faults in the core are critical faults.

3.

In case a fault is permanent, then it will not be recovered.
In case the fault is transient then it can be detected in one cycle and it will not be detected in next cycle. In such case it is again application responsibility to react to such a fault. Application may e.g. gather some statistics how often the transient faults occur and react accordingly.

Best regards,

Daniel