STMH7 ETH_DMAMR_SWR fails

[BengofEtech]

Hey,

i got a specific Problem with the STM32H7 Ethernet, sometimes when performing

  SET_BIT(heth->Instance->DMAMR, ETH_DMAMR_SWR);
 
  /* Get tick */
  tickstart = HAL_GetTick();
 
  /* Wait for software reset */
  uint8_t force_reset = 0;
  while (READ_BIT(heth->Instance->DMAMR, ETH_DMAMR_SWR) > 0U)
  {
    if(((HAL_GetTick() - tickstart ) > ETH_SWRESET_TIMEOUT))
    {
      /* Set Error Code */
      heth->ErrorCode = HAL_ETH_ERROR_TIMEOUT;
      /* Set State as Error */
      heth->gState = HAL_ETH_STATE_ERROR;
      /* Return Error */
      return HAL_ERROR;
      break;
    }
  }

The error gets detectet, so ETH_DMAMR_SWR doesnt get activated within this timeout period. Is there aother way to reset the ETH Controller except from restarting the whole mcu?

I tried __HAL_RCC_ETH1MAC_FORCE_RESET but it didnt work out.

[BengofEtech]

okay much more interesting:

  /* Get tick */
  tickstart = HAL_GetTick();
 
  /* Wait for software reset */
  uint8_t force_reset = 0;
  while (READ_BIT(heth->Instance->DMAMR, ETH_DMAMR_SWR) > 0U)
  {
    if(((HAL_GetTick() - tickstart ) > ETH_SWRESET_TIMEOUT))
    {
      /* Set Error Code */
      heth->ErrorCode = HAL_ETH_ERROR_TIMEOUT;
      /* Set State as Error */
      heth->gState = HAL_ETH_STATE_ERROR;
      /* Return Error */
      force_reset = 1;
      break;
    }
  }
 
 
  tickstart = HAL_GetTick();
  if(force_reset)
  {
//	  tickstart = HAL_GetTick();
	  __HAL_RCC_ETH1MAC_FORCE_RESET();
	  HAL_Delay(100);
	  __HAL_RCC_ETH1MAC_RELEASE_RESET();
	  HAL_Delay(100);
	 SET_BIT(heth->Instance->DMAMR, ETH_DMAMR_SWR);
	  while (READ_BIT(heth->Instance->DMAMR, ETH_DMAMR_SWR) > 0U)
	  {
	    if(((HAL_GetTick() - tickstart ) > ETH_SWRESET_TIMEOUT))
	    {
	      /* Set Error Code */
	      heth->ErrorCode = HAL_ETH_ERROR_TIMEOUT;
	      /* Set State as Error */
	      heth->gState = HAL_ETH_STATE_ERROR;
	      /* Return Error */
 
	      return HAL_ERROR;
	    }
	  }
  }

when optimising AND uncomment the tickstart, the whole programm wont run correctly. Anyone expirienced things like that?

[BengofEtech]

there seems something wrong with data alligmnent. cant imagine, it could be something written on the flash. any experiences?

as far as i can tell, usually adding non related code in flash should not interfere with the programs ability to function, which is clearly the case here.

[JackSparrow]

I have run into the same issue with my STM32H753 MCU(https://community.st.com/t5/stm32-mcus-products/stm32h753-ethernet-initialization-failure/td-p/602033)

Were you able to fix this issue?

[MBeau]

Same issue here on a H747 DISC0. Same c-code as yesterday, starting my program today and constantly running into this behaviour.

This line of code triggers the Hardfault_Handler()

SET_BIT(heth->Instance->DMAMR, ETH_DMAMR_SWR);

But when debugging I noticed that this function is called already by MX_ETH_Init (I guess so). So the SET_BIT function was called already and is running into the HardFault_Handler, when called the second time.
After commenting out "HAL_ETH_Init(&heth)" in my main.c, I didn't run into the error any more :)

Still wondering why I didn't run into the HardFault_Handler yesterday, when HAL_ETH_Init" was called twice (1x MX Init + 1x main.c)

Hope this will help anyone :)
Cheers