1.4.2 CAN Bootloader system level execution flow

Basic Bootloader system level execution flow

• The Bootloader code starts executing on a device Reset

• If there are no conditions to enter the firmware upgrade mode, the Bootloader starts executing the user application

  • Refer to Bootloader Trigger Methods for different conditions to enter firmware upgrade mode

• The Bootloader performs Flash erase/program operations while in the firmware upgrade mode

Fail Safe Update Bootloader system level execution flow (MIPS Based MCUs)

• Bootloader always maps Bank 1 to lower region at boot up irrespective of cause for reset (Hard/Soft).

  • This is required because if Swap bit was set and a soft reset was triggered the value is retained but if Swap bit was set and Hard reset was triggered the value is cleared

• Once Swap bit is cleared and Bank 1 is mapped to lower region it performs below operation

  • If Bank 1 serial number is greater than Bank 2 serial number it just continues for trigger check or runs application from Bank 1, As Bank 1 is already mapped to lower region in above step (Bank 1 is Active Bank)

  • If Bank 2 serial number is greater than Bank 1 serial number it maps Bank 2 to lower region by setting the Swap bit and proceeds to trigger check or runs application from Bank 2 (Bank 2 is Active Bank)

• Whenever Bootlader programs new application in the Inactive bank and swap bank command is received it reads the serial number from Active bank, increments by 1 and then writes to Inactive Bank serial

  • Inactive Bank Serial number = Active Bank serial number + 1

  • Start address of inactive bank is equal start of mid of flash

• If the Host application requests to update the Active Bank and the address falls into the active bank serial sector it sends an error response and aborts the programming operation