0

is it possible to see full backtrace after getting segmentaion fault or unexpected end of program in TRACE 32. Like in DDD after bt command ?

I am merging a lot of source into a project (SW Update) and keep getting target reset, would like to know what was the last funcion called.

There is an option Stackframe or Stackframe with locals but it works only if the program is running.

Thanks in advance.

Jeremy
  • 45
  • 11
  • 1
    It might be helpful to know the processor architecture of the target you're debugging. In general, the goal would be to stop the CPU before the reset occurs, because afterwards, much of the required information will be gone. How to do this depends on the architecture and why the reset occurred in the first place. – wrtlprnft Sep 12 '18 at 09:54
  • Thanks for the reply. The tricky part is that i do not know where the error occurs. It is somewhere in the framework and with stepping in with the debugger a do not see the error. So the Idea was like with the DDD to see full backtrace. I am using RH850 F1L from Rensesas. – Jeremy Sep 17 '18 at 06:13

1 Answers1

1

View the call stack with command FRAME.view.

Keep in mind that a debugger can only show you the "backtrace" in case of an exception, if the call stack is not corrupt or gets hardly altered by the exception handler.

To find out exactly what was going on before interrupt or exception you would have to record the program flow. However only some CPUs are "traceable" and support the recording of the program flow. In case e.g. of an ARM CPU it would require the ETM component. If you CPU is traceable, but supports only offchip-trace you also require a suitable trace tool e.g. a "PowerTrace" if you want to use TRACE32.

Holger
  • 3,920
  • 1
  • 13
  • 35