I have posted previously on fault injection methods to check group 1 fault behaviour. My ESM handler works as expected for other faults, nError goes low and remains low until I reset the fault records next power cycle. My ESM channel IDs are saved to Flash Emulated EEPROM Bank 7 (FEE) in my ESM handler.
I identified methods to inject faults with reference to the selftest routines and safe TI library. But I don't fully understand how to successfully inject faults for channels 35 and 36 such that I can complete writes to FEE after the ESM interrupt is triggered. I am satisfied that the system goes to a safe state - nError remains low for both Channel 35 and 36 and we enter our application fault state. But I would like to understand why I can't write the channel # to FEE.
Here is a screen shot of my Channel 35 Fault Injection code, am I missing some housekeeping here to permit writes to FEE after generating the fault? I know the save to FEE is not successful as on next powerup we trip nError again, normally my test logic would not run the fault injection if the last power cycle triggered and saved an ESM event to FEE.
here is Channel 36 fault injection. This is strange in that the first power up after download I do capture Channel 36 and get nError activity similar to Channel 35. Next power up I get nError activity but my ESM channel ID comes through as 255 which indicates the ESM handler is not called as that is my initialization value. Then I suddenly get a Channel 35 error? I only run the fault injection prior to entering the main loop so the sudden single bit error must be a leftover from previous runs?
Appreciate any insight in to how to successfully inject channel 35 and 36 faults, such that I capture the channel ID with ability to save in EEPROM bank 7.
thanks
Jamie
