HardFault_Handler occurs when SPI function is added to the Bluetooth connection software

[Blance]

Hi all:

I have a software version that allows Bluetooth to be connected to the phone normally,

When I tried to use the SPI interface, it happened as long as I used

"nrf_drv_spi_init(&spi, p_spi_cfg, spi_event_handler, NULL)" function,

test the BLE connection with the mobile phone again, it will appear about 3 seconds after the connection is successful, and the HardFault_Handler will be triggered

Troubleshoot the problem in depth and found that when the initial is executed to the following figure "m_contexts[inst_idx]=p_context", even if I leave directly later and do not perform any SPI transmission, HardFault_Handler will be triggered after the BLE connection,Any suggestions?

// <e> SPI_ENABLED - nrf_drv_spi - SPI/SPIM peripheral driver - legacy layer
//==========================================================
#ifndef SPI_ENABLED
#define SPI_ENABLED 1
#endif
// <o> SPI_DEFAULT_CONFIG_IRQ_PRIORITY  - Interrupt priority
 

// <i> Priorities 0,2 (nRF51) and 0,1,4,5 (nRF52) are reserved for SoftDevice
// <0=> 0 (highest) 
// <1=> 1 
// <2=> 2 
// <3=> 3 
// <4=> 4 
// <5=> 5 
// <6=> 6 
// <7=> 7 

#ifndef SPI_DEFAULT_CONFIG_IRQ_PRIORITY
#define SPI_DEFAULT_CONFIG_IRQ_PRIORITY 6
#endif

// <o> NRF_SPI_DRV_MISO_PULLUP_CFG  - MISO PIN pull-up configuration.
 
// <0=> NRF_GPIO_PIN_NOPULL 
// <1=> NRF_GPIO_PIN_PULLDOWN 
// <3=> NRF_GPIO_PIN_PULLUP 

#ifndef NRF_SPI_DRV_MISO_PULLUP_CFG
#define NRF_SPI_DRV_MISO_PULLUP_CFG 0
#endif

// <e> SPI0_ENABLED - Enable SPI0 instance
//==========================================================
#ifndef SPI0_ENABLED
#define SPI0_ENABLED 1
#endif
// <q> SPI0_USE_EASY_DMA  - Use EasyDMA
 

#ifndef SPI0_USE_EASY_DMA
#define SPI0_USE_EASY_DMA 0
#endif

// </e>

// <e> SPI1_ENABLED - Enable SPI1 instance
//==========================================================
#ifndef SPI1_ENABLED
#define SPI1_ENABLED 0
#endif
// <q> SPI1_USE_EASY_DMA  - Use EasyDMA
 

#ifndef SPI1_USE_EASY_DMA
#define SPI1_USE_EASY_DMA 1
#endif

// </e>

// <e> SPI2_ENABLED - Enable SPI2 instance
//==========================================================
#ifndef SPI2_ENABLED
#define SPI2_ENABLED 0
#endif
// <q> SPI2_USE_EASY_DMA  - Use EasyDMA
 

#ifndef SPI2_USE_EASY_DMA
#define SPI2_USE_EASY_DMA 0
#endif

// </e>

// <q> SPIM_NRF52_ANOMALY_109_WORKAROUND_ENABLED  - Enables nRF52 anomaly 109 workaround for SPIM.
 

// <i> The workaround uses interrupts to wake up the CPU by catching
// <i> a start event of zero-length transmission to start the clock. This 
// <i> ensures that the DMA transfer will be executed without issues and
// <i> that the proper transfer will be started. See more in the Errata 
// <i> document or Anomaly 109 Addendum located at 
// <i> https://infocenter.nordicsemi.com/

#ifndef SPIM_NRF52_ANOMALY_109_WORKAROUND_ENABLED
#define SPIM_NRF52_ANOMALY_109_WORKAROUND_ENABLED 0
#endif

[Edvin]

Hello,

From where do you call nrf_drv_spi_init()? Is it inside some sort of interrupt? Does it also happen if you try to call it directly from your main() function before you enable the softdevice?

Best regards,

Edvin

[Blance]

Hi:

My main purpose is to make another "nrf_block_dev_spi" to control Flash based on "nrf_block_dev_qspi",

so the trigger will be in :

main->fatfs_init->disk_initialize(0)->nrf_blk_dev_init.

I also tried to use GPIO to simulate SPI, the operation is normal, but The transmission rate is slow,

I still hope to use hw spi to speed up the process

[Edvin]

The nRF52832 doesn't have QSPI, so if you drag in something from that, you may write to registers that doesn't have the same functionality in the nRF52832 as it does in the nRF52840.