STM32H753I HardFault_Handler() __iar_packbits_init_single3

[KGane.1]

I have simple application( LED toggling) built for STM32H753I-EVAl board using IAR EWARM Toolchain. I am using IAR I-Jet debugger to download and debug application.

When I download application, i am getting Fault exception

"the processor has escalated a configurable-priority exception to hardfault

am imprecise data access error has occurred"

Call Stack:

HardFault_Handler()

<Exception frame>

[__iar_packbits_init_single3 + 0x19]

<Unknown handler>

Seems the Reset_Handler in startup code calls SystemInit() and then __iar_program_start. The control never comes out the function to main().

What could cause fault exception toolchain specific functions.? Am I missing something in initialization? 

Observation: Sometimes the application is running successfully. The LED toggling is perfectly fine. Tried running application on two boards. Seeing fault exception on both boards but not always.

[TDK]

Stay objective, stop guessing. Examine SCB and determine cause of hard fault and trace it to the source.

Probably a memory mismanagement issue.

edit: for imprecise faults, verify solid power scheme, decoupling caps, clock setup and wait states.

[KGane.1]

@TDK​ Thanks. Started looking into debugger log. Seems unhandled BusFault caused HardFault exception.

Sun Jan 16, 2022 19:20:21: Stopped by a vector catch: 

Sun Jan 16, 2022 19:20:21: BusFault fault escalated into HardFault 

Sun Jan 16, 2022 19:20:21: The BusFault handler is disabled 

Sun Jan 16, 2022 19:20:21: HardFault exception. 

Sun Jan 16, 2022 19:20:21: The processor has escalated a configurable-priority exception to HardFault. 

Sun Jan 16, 2022 19:20:21:  An imprecise data access error has occurred. 

Sun Jan 16, 2022 19:20:21:  

Sun Jan 16, 2022 19:20:21: Exception occured at: 0x800310e 

Sun Jan 16, 2022 19:20:21:  

Sun Jan 16, 2022 19:20:21: See the call stack for more information. 

What is confusing is why the failure is not consistent? Does it relate to timing issues in read/write of internal memory?

[TDK]

Not sure if you saw my edit. For imprecise error, verify clock and power settings, power rail, wait states.

[KGane.1]

@TDK​ Thanks. It seems data corruption during binary download is causing the problem. Download verification is failing. Trying to run failed binary is causing Hardfault.