RM46L852: Mysterious Data Abort outside of the RAM

Hi, 

I would like to have your opinion on an exception that I'm encountering using the RM46 with FreeRTOS, HalCoGen and SEGGER RTT.

As you can see, the exception is a data abort.

As you can see it's a Precise External Abort.
To my understanding, the fault location is located before the stack. I don't understand why.

I don't understand why there would be a data abort on a RESERVED portion of memory.

If I unroll the call stack, I end-up in the TI memcpy implementation

Any idea why this is happening, or what I could do to better understand the problem?

Thanks for your help,
Gabriel