Our current project is to integrate EB's osek protocol stack (communication, network management, diagnosis) based on mcal. After the integration, the program will enter the abnormal interrupt during the running, and the internal watchdog cannot reset the microcontroller.
It has now been found that there are two situations in which an exception will be entered:
- The first run after each use of the debugger flash program, enter machine_check_int_ha when running to fccu configuration. How should I go to troubleshoot the problem? After resetting the program with the debugger, the program can run normally temporarily, but the second case will be encountered.
- The current controller is configured with Rx messages of about 10 frames. If the controller only sends out, the program can run normally for a long time; use CANoe to simulate all Rx messages sent to the controller, and the program will enter the exception after running for several tens of seconds.
After the program enters the exception, the greenhills probe debugger we are using cannot get the value of the exception register, so there is no way to check the cause of the exception.
A separate mcal program (without network management and diagnostics) has been tested and has not experienced an exception when accepting messages. EB's osek protocol stack itself is not based on the mcal program. They can only test their own programs to see if they can see the same phenomenon.
Best and Regards