HardFault exception when jumping from application to system bootloader

[DAUSILI]

Hello,
when trying to jump from my application to system bootloader, I get an HardFault exception.

This is the jumping routine that I am using, based on the following ST tutorial:
https://community.st.com/t5/stm32-mcus/how-to-jump-to-system-bootloader-from-application-code-on-stm32/ta-p/49424

 

 

#define SYSTEM_MEMORY_ADDR 0x1FFF0000
#define BOOTLOADER_ENTRY_ADDR (SYSTEM_MEMORY_ADDR + 4)

void JumpToBootloader(void)
{
 ADC_HandleTypeDef hadc;
 UART_HandleTypeDef huart;
 SPI_HandleTypeDef hspi;
 TIM_HandleTypeDef htim;
 void (*f)(void);
 uint8_t i;

 /* Disable all interrupts */
 __disable_irq();

 /* Disable Systick timer */
 SysTick->CTRL = 0;

 /* Set the clock to the default state */
 HAL_RCC_DeInit();

 /* Disable all other peripherls */
 hadc.Instance = ADC1;
 HAL_ADC_DeInit(&hadc);

 huart.Instance = USART2;
 HAL_UART_DeInit(&huart);

 hspi.Instance = SPI1;
 HAL_SPI_DeInit(&hspi);

 htim.Instance = TIM2;
 HAL_TIM_Base_DeInit(&htim);

 /* Clear Interrupt Enable Register & Interrupt Pending Register */
 for (i = 0; i < 5; i++)
 {
 NVIC->ICER[i]=0xFFFFFFFF;
 NVIC->ICPR[i]=0xFFFFFFFF;
 }

 /* Re-enable all interrupts */
 __enable_irq();

 /* Set the main stack pointer to the boot loader stack */
 __set_MSP(*(uint32_t *) SYSTEM_MEMORY_ADDR);

 /* jump to bootloader */
 f = (void (*)(void)) BOOTLOADER_ENTRY_ADDR;
 f();
}

 

Thank you,
Daniele

[TDK]

What does information within the SCB registers tell you about the reason for the hard fault? Use a fault analyzer if needed.

The calls to HAL_*_DeInit aren't as intended. These should be called using the same global handles that were used in the HAL_*_Init calls. Unclear if this is the source.

[Jeff Tenney]

You probably need to map system memory to 0x00000000 via SYSCFG->CFGR1.  And you probably need to set the VTOR back to 0 as well (SCB->VTOR).

[Jeff Tenney]

---

@DAUSILI wrote:

 

 /* Clear Interrupt Enable Register & Interrupt Pending Register */
 for (i = 0; i < 5; i++)
 {
 NVIC->ICER[i]=0xFFFFFFFF;
 NVIC->ICPR[i]=0xFFFFFFFF;
 }

 

The M0+ has only 1 ICER and ICPR register, not 5.  So this code is writing to registers that don't exist.  That's probably the cause of the hardfault.

[DAUSILI]

Hello Jeff,
the cause of the malfunction was due to a wrong VTOR alignment.
The main memory was mapped at 0x00000000, so before jumping to system memory I correct the VTOR value in this way:

#define SYSTEM_MEMORY_ADDR 0x1FFF0000

__ISB();
SCB->VTOR = SYSTEM_MEMORY_ADDR;
__DSB();

Thank you,

Daniele

[gbm]

This is incorrect and quite different from the original code:

f = (void (*)(void)) BOOTLOADER_ENTRY_ADDR;

See my post in the bootloader thread for a correct and simpler version