RE: Running function stack trace

outspace — Fri, 29 Jan 2021 10:52:58 GMT

Sorry to reply you so late, thank you for your guidance

Because sometimes, for example, some dynamic memory overflows cause occasional program exceptions, I want to use this idea to determine and narrow the scope of the problem, I want to see the same context as the call stack window in the IDE, and further provide me with some ideas

Thank you very much.

RE: Running function stack trace

Kenneth — Wed, 11 Nov 2020 19:10:38 GMT

I think this can be done simpler. When the processor takes an exception, the processor pushes information onto the current stack. See here:

https://developer.arm.com/documentation/dui0552/a/the-cortex-m3-processor/exception-model/exception-entry-and-return

Thereby I believe you can get the information you want from the call stack. Typically in the nRF5 SDK you can for instance catch exceptions (hardfaults) by including (one of the files are compiler specific):

$(SDK_ROOT)/components/libraries/hardfault/hardfault_implementation.c
$(SDK_ROOT)/components/libraries/hardfault/nrf52/handler/hardfault_handler_gcc.c

After adding the above files and setting HARDFAULT_HANDLER_ENABLED 1 in sdk_config.h, you should be able to catch the hardfaults in HardFault_c_handler() and can process these as you like.