This discussion has been locked.
You can no longer post new replies to this discussion. If you have a question you can start a new discussion

RTX

Hello,

How would RTX fail apart from stack overflow?

Currently, my software runs for approx. 10 hours and fail using RTX. I have increase the task overflow and also the interrupt and user mode.

Regards
Paul

  • The nice thing about stack overflows is that you do not know how the application will fail. Just that it can be unpredictable and in some situations very spectacular.

    You use an ARM - have you looked at the analysis that the compiler can produce, with max stack depth and amount of info needed for auto variables etc?

    Or are you using recursive calls?

    Memory overwrites anywhere? All buffers of correct size and no uninitialized pointers?

  • A memory leak?

    (eg repeated calls to malloc() without matching calls to free() )

    Christopher
    ==

  • some tips :

    - for each dynamic allocation buffer array, use a semaphore .

    - Create a high priority task with a dedicated mailbox that trap errors. Personnaly , I use a task that get the last 100 errors with a timestamp, the faulty task ID and a specific error code.

    - And as Christopher said : verify the malloc/free _alloc_box / _free_box couples ... :-)