FAQs
English (US) 日本語 | Japanese 中文 | Chinese (Simplified)

upgrade the APP and jump to Boot directly from the APP (without restarting), but HardFault error

cancel
Turn on suggestions
Auto-suggest helps you quickly narrow down your search results by suggesting possible matches as you type.
Showing results for 
Show  only  | Search instead for 
Did you mean: 

upgrade the APP and jump to Boot directly from the APP (without restarting), but HardFault error

‎09-06-2020 06:48 PM
1,002 Views
ljq
Contributor II

  • Chip: S32K146    OS:FreeRtos

  • App Jump Boot
    • #define portNVIC_SYSTICK_CTRL_REG ( * ( ( volatile uint32_t * ) 0xe000e010 ) )
      #define portNVIC_SYSTICK_CURRENT_VALUE_REG ( * ( ( volatile uint32_t * ) 0xe000e018 ) )
      void EnterUpgradeModeForJump(void)
      {
          T8U i;
          EnterCriticalCode();
          DisableAllIRQs();
          DISABLE_INTERRUPTS();
      for(i=0;i<SYSTEM_TASK_ID_SYSMONITOR;i++)
      {
      System_Task_Deactivate(i);
      }
      Sys_Board_Hw_DeInit();
      Hal_FlashInit_Api();
      Update_Program_Mcu_SysInfo();
      Hal_FlashDeInit_Api();
      /* Stop and clear the SysTick. */
      portNVIC_SYSTICK_CTRL_REG = 0UL;
      portNVIC_SYSTICK_CURRENT_VALUE_REG = 0UL;
      ExitCriticalCode();
      App_JumpTo_Boot_API();
      }
    • App Deinit the hardware resources used before the jump (except for Clock and IO, there is no Deinit that can be called by Clock,IO port cannot be powered off), and reinitialize the Clock and serial ports in Boot
    • can jump to Boot normally and upgrade, but HardFault occurs during the upgrade process
    • If the code of Boot is rebooted and upgraded, it is OK, only if the App is directly jumped is an error
0 Kudos
Reply
3 Replies

‎09-08-2020 02:51 AM
988 Views
lukaszadrapa
NXP TechSupport
NXP TechSupport

Hi,

first, I would try to find the source of hard fault. This application note could be useful:

https://www.nxp.com/docs/en/application-note/AN12201.pdf

There's obviously something what is not de-initialized before jump. This is always a risk when jumping from one application to another without reset. Especially, I would check all interrupts - if interrupt flags, local enable bits in peripherals, interrupt priorities are reverted back to default state.

Regards,

Lukas

 

0 Kudos
Reply

‎09-10-2020 11:49 PM
978 Views
ljq
Contributor II

void BusFault_Handler(void)
{

BFSR = (S32_SCB->CFSR & 0x0000FF00);
/* BusFault Status Register (BFSR) CFSR[15:8] indicates the cause of BusFault
* [15] BFARVALID = 1 BFAR holds a valid fault address
* [13] LSPERR = 1 Bus fault occurred during floating-point lazy state preservation
* [12] STKERR = 1 Stacking for an exception entry has caused BusFault, BFAR address is not valid
* [11] UNSTKERR = 1 Unstack for an exception return has caused BusFault, BFAR address is not valid
* [10] IMPRECISERR = 1 Imprecise data bus error, BFAR address is not valid
* [9] PRECISERR = 1 Precise data bus error, BFAR address is valid
* [8] IBUSERR = 1 error on prefetching an instruction, BFAR address is invalid
* */
if(BFSR & S32_SCB_CFSR_BFARVALID_MASK){ // If the address in BFAR register is valid
BFAR = S32_SCB->BFAR; // BFAR holds the address of the location that generated the BusFault
}
if(BFSR & S32_SCB_CFSR_PRECISERR_MASK) asm("BKPT");
if(BFSR & S32_SCB_CFSR_IMPRECISERR_MASK) asm("BKPT");
if(BFSR & S32_SCB_CFSR_LSPERR_MASK) asm("BKPT");
if(BFSR & S32_SCB_CFSR_STKERR_MASK) asm("BKPT");
if(BFSR & S32_SCB_CFSR_UNSTKERR_MASK) asm("BKPT");
if(BFSR & S32_SCB_CFSR_IBUSERR_MASK) asm("BKPT");

asm("MOVS R0, #4");
asm("MOV R1, LR");
asm("TST R0, R1");
asm("BEQ __MSP");
asm("MRS R0, PSP");
asm("B getStackFrame");
asm("__MSP:");
asm("MRS R0, MSP");
asm("B getStackFrame");
}

 

  • Each time it is stuck in the following error :BFSR = 0x8200

BusFault_Handler(): if(BFSR & S32_SCB_CFSR_PRECISERR_MASK) asm("BKPT");

BFAR point to __isr_vector

0 Kudos
Reply

‎08-10-2021 11:13 PM
819 Views
pedritorres21
Contributor I

void BusFault_Handler(void)
{

BFSR = (S32_SCB->CFSR & 0x0000FF00);
/* BusFault Status Register (BFSR) CFSR[15:8] indicates the cause of BusFault
* [15] BFARVALID = 1 BFAR holds a valid fault address
* [13] LSPERR = 1 Bus fault occurred during floating-point lazy state preservation
* [12] STKERR = 1 Stacking for an exception entry has caused BusFault, BFAR address is not valid
* [11] UNSTKERR = 1 Unstack for an exception return has caused BusFault, BFAR address is not valid
* [10] IMPRECISERR = 1 Imprecise data bus error, BFAR address is not valid
* [9] PRECISERR = 1 Precise data bus error, BFAR address is valid
* [8] IBUSERR = 1 error on prefetching an instruction, BFAR address is invalid
* */
if(BFSR & S32_SCB_CFSR_BFARVALID_MASK){ // If the address in BFAR register is valid
BFAR = S32_SCB->BFAR; // BFAR holds applinked apk address of the location that generated the BusFault
}
if(BFSR & S32_SCB_CFSR_PRECISERR_MASK) asm("BKPT");
if(BFSR & S32_SCB_CFSR_IMPRECISERR_MASK) asm("BKPT");
if(BFSR & S32_SCB_CFSR_LSPERR_MASK) asm("BKPT");
if(BFSR & S32_SCB_CFSR_STKERR_MASK) asm("BKPT");
if(BFSR & S32_SCB_CFSR_UNSTKERR_MASK) asm("BKPT");
if(BFSR & S32_SCB_CFSR_IBUSERR_MASK) asm("BKPT");

asm("MOVS R0, #4");
asm("MOV R1, LR");
asm("TST R0, R1");
asm("BEQ __MSP");
asm("MRS R0, PSP");
asm("B getStackFrame");
asm("__MSP:");
asm("MRS R0, MSP");
asm("B getStackFrame");
}

 

  • Each time it is stuck in the following error :BFSR = 0x8200

BusFault_Handler(): if(BFSR & S32_SCB_CFSR_PRECISERR_MASK) asm("BKPT");

BFAR point to __isr_vector

I am facing the same issue.

0 Kudos
Reply