I'm working with an MCF5475 running Linux v2.6.10 from the Freescale BSP. I recently ran across information in the Coldfire Family Programmer's Reference Manual, Rev 3 in section 11.1.3 that states:
"Note that if a Coldfire processor encounters any fault while processing another fault, it immediately halts execution with a catastrophic fault-on-fault condition. A reset is required to force the processor to exit this halted state."
I'm suspicious that we maybe encountering this catastrophic fault-on-fault condition but I am having trouble proving this. I suspect this because on a very intermmitent basis, while I have my BDI2000 debugging tool plugged into the BDM port and when the system locks up, I halt the CPU using the debugger to investigate the registers and find that the registers have invalid values and that the BDI2000 loses communication with the CPU. Under normal circumstances, I halt and resume the CPU without issue using the BDI2000.
Is there a way to determine that the CPU experienced a catastrophic fault-on-fault condition by sticking a probe on one of the CPU pins or something along that line?