We are encountering a system hang issue during AGX boot-up. To address this type of deadlock problem, we want to implement a watchdog “walk run” function. Please help us with the following questions:
1. Can we use WDT_RESET_OUT_N as a trigger signal and use an external MCU to monitor WDT_RESET_OUT_N to determine whether the AGX has hung? If this approach is not feasible, are there any alternative solutions to achieve the same functionality?
2. What is the functional purpose and operational logic of WDT_RESET_OUT_N?
3. Has the NVIDIA team encountered similar issues with other customers? If so, how was this handled in those cases?