TM4C129ENCPDT MPU vs Hard Fault

Well, I guess my previous message was cut somehow (there was a problem with my browser?)

I was saying that I had (not so big) issue with the MPU. Instead of the MPU fault being invoked, a hard fault is invoked. I can work around this with the code:

void hardFaultHandler()
{
if(SCB->HFSR & 0x40000000 != 0 && SCB->CFSR & 0xff != 0)
{
/* MPU fault */
}
else
{
/* not an MPU fault */
}
}

So this is not a big issue, but was wondering why is the mpuHandler not called?

Thank you,
Manuel

Hello Bob,

Thank you for your answer. It makes sense that the fault is being escalated to a hard fault, like you said.

I'm having trouble figuring out why. I'm my test I purposely configure the MPU to certain memory regions and on the test, in thread mode without privileges, the test access a memory region outside the protected on the MPU (but still on RAM).

So I guess the scenario:
- is not the first one that you shown (because the fault happens in thread mode)
- is not the second one (because the fault happens in thread mode)
- is not the third one (because the fault happens in thread mode)
- must be the fourth

I've tried to find if I need to enable the MPU trap handler somehow, and could only find the MPUCTRL register, which is set to:
(1U << 2U) | (0U << 1U) | (1U << 0U).

Do you have any suggestion?

Thank you,
Manuel

Hello,

Finally I found the answer. The MPU exception can be enabled/disabled through the SHCSR register (and this was the thing that was missing).

Thanks again,

Manuel