STM32N6 Hard Fault from FreeRTOS

[Neuraleanus]

I am attempting to setup a new Hello World project on the STM32N6 using the following:

1.  FreeRTOS running two tasks, a timer, and a semaphore
2. A couple of hardware timers
3. A USART
4. The PSSI doing 16 bit bi-directional transfers
5. The BSP package

Running the debugger I get

followed by a Hard Fault:

The code is running on a Nucleo-N657XO-Q development board.  My development tools are:

1.  STMCube 6.18.1
2. STM32Cube_FW_N6_V1.4.0
3. IAR 9.60.4

Something not correctly setup or a bad board?

[Andrew Neil]

IAR 9.60.4

You’d be better asking IAR for support with their IDE, and what to do about that particular IAR message.

Did you try Tools > Options, as it suggests ?

 

General tips on debugging Cortex-M Hard Faults.

[Neuraleanus]

I have been doing the code construction in stages, prior to adding the timers, both soft and hard, the following code worked:

Perhaps this is indication of a memory access issue?

I see nothing obvious in tools-->options.

 

[Andrew Neil]

I see nothing obvious in tools-->options.

 

Again, that’s an IAR thing: best to ask them about that - it is their product, and you’ve paid for support as part of your licence.

[FBL]

Hi ​@Neuraleanus 

I assume ​@Andrew Neil is pointing to this interesting article How to debug a HardFault on an Arm® Cortex®-M STM32 | Community

From the information provided, this does not look like a toolchain issue. I would recommend to comment out all printf() calls first to narrow down the issue stack / lib c / UART problem vs RTOS / peripheral / memory issue

BFAR 0xC usually means NULL pointer + offset.

The value of PC just before the exception has occurred shown in the debug window (can be found in the stack (Exception frame) points to the faulty instruction.