The TMS320F2838x MCU product architecture provides different levels of fault indication from internal safety mechanisms using CPU Interrupt, Non Maskable Interrupt (NMI), assertion of ERRORSTS pin, assertion of CPU input reset and assertion of warm reset (XRSn). The fault response is the action that is taken by the TMS320F2838x MCU or system when a fault is indicated. Multiple potential fault responses are possible during a fault indication. The system integrator is responsible to determine which fault response should be taken to ensure consistency with the system safety concept. The fault indication ordered in terms of severity (device power down being the most severe) is shown in Figure 3-12.
- Device Power Down: This is the highest
priority fault response where the external component (see Section 3.4.5) detects malfunctioning
of the device or other system components and powers down the TMS320F2838xD/S MCU. From
this state, it is possible to re-enter cold boot to attempt recovery.
- Assertion of XRSn: The XRSn reset could be generated from an internal or external monitor that detects a critical fault having potential to violate safety goal. Internal sources generate this fault response when the TMS320F2838xD/S MCU is not able to handle the internal fault condition by itself (for example, CPU1 (master CPU) is not able to handle NMI by itself). From this state, it is possible to re-enter cold boot and attempt recovery.
- Assertion of CPU Reset: CPU Reset changes the state of the CPU from pre-operational or operational state to warm boot phase. The CPU Reset is generated from an internal monitor that detects any security violations. On a properly working system, the security violations may be the secondary effect due to a fault condition. In addition, CPU2 subsystem generates this fault response when it is not able to handle the internal fault condition by itself (for example, CPU2 is not able to handle NMI by itself). From this state, it is possible to re-enter warm boot phase and attempt recovery.
- Non Maskable Interrupt (NMI) and assertion of ERRORSTS pin: C28x CPU supports a Non Maskable Interrupt (NMI), which has a higher priority than all other interrupts. Each CPU subsystem is equipped with a NMIWD module responsible for generating NMI to the C28x CPU. ERRORSTS pin will also be asserted along with NMI. Depending on the system level requirements, the fault can be handled either internal to the TMS320F2838xD/S MCU using software or at the system level using the ERRORSTS pin information.
- CPU Interrupt: CPU interrupt allows events external to the CPU to generate a program sequence context transfer to an interrupt handler where software has an opportunity to manage the fault. The peripheral interrupt expansion (PIE) block multiplexes multiple interrupt sources into a smaller set of CPU interrupt inputs.