NRF_UICR->CUSTOMER read/write on 52832

RE: NRF_UICR->CUSTOMER read/write on 52832

Kenneth — Tue, 09 Mar 2021 17:07:16 GMT

[quote user="QuentinFarizon"]Well I guess that's the solution to my issue ?[/quote]

Yes.

Fyi, you can find an overview which peripherals are accessible here when softdevice is enabled/disabled:
https://infocenter.nordicsemi.com/topic/sds_s112/SDS/s1xx/sd_resource_reqs/hw_block_interrupt_vector.html

The same also apply to the nRF52811, however since the nRF52811 don't have a MWU that can enforce this limitation, then you may get away with it on the nRF52811, though it can cause various issues.

Best regards,
Kenneth

RE: NRF_UICR->CUSTOMER read/write on 52832

QuentinFarizon — Tue, 09 Mar 2021 14:18:38 GMT

Well I guess that's the solution to my issue ? I didn't know SD prevented writing UICR, and it wasn't the case on the 52811...

RE: NRF_UICR->CUSTOMER read/write on 52832

Kenneth — Tue, 09 Mar 2021 14:09:11 GMT

It's the first time I hear that the softdevice is enabled. If the softdevice is enabled, then the NVMC is a peripheral that is in use by the softdevice. If you want to write to the UICR, then you must do this when the softdevice is disabled.

Kenneth

RE: NRF_UICR->CUSTOMER read/write on 52832

QuentinFarizon — Tue, 09 Mar 2021 13:28:32 GMT

How is that different ? I get the invalid memory access when trying to write NRF_UICR->CUSTOMER

RE: NRF_UICR->CUSTOMER read/write on 52832

Kenneth — Tue, 09 Mar 2021 13:25:10 GMT

Your latest problem is different from your initial case, so I suggest you search for "SOFTDEVICE: INVALID MEMORY ACCESS" on devzone (for instance by google), you should find some suggestions to look for. I am more interested if you have been able to make a sample project for the UICR reset issue.

Kenneth

RE: NRF_UICR->CUSTOMER read/write on 52832

QuentinFarizon — Tue, 09 Mar 2021 11:58:47 GMT

I succeeded in making it work by calling sd_softdevice_disable first, and then enabling SD after.

Is this expected ?

RE: NRF_UICR->CUSTOMER read/write on 52832

QuentinFarizon — Tue, 09 Mar 2021 11:24:50 GMT

Hello,

The markings are : N52832 QFAAE1 1845D

I was (by mistake) shadowing app_error_fault_handler so I was missing this log :
SOFTDEVICE: INVALID MEMORY ACCESS

I see the same thing without a debugger, with only the logs

The power supply is steady yes, I have an Otii

RE: NRF_UICR->CUSTOMER read/write on 52832

Kenneth — Tue, 09 Mar 2021 08:53:06 GMT

Please start working on a simplified project I can unzip to a folder you specify in an nRF5 SDK you specify. Do you have access to an nRF52 DK you can use for comparison here?

If you have "crashes" I assume you mean asserts?

Please also answer my question: "Can you provide the chip laser markings that is printed on top of the nRF52832 IC?"

Do you see the same with debugger not connected?

Is the power supply steady here?

Kenneth

RE: NRF_UICR->CUSTOMER read/write on 52832

QuentinFarizon — Mon, 08 Mar 2021 18:04:59 GMT

Hello Kenneth,

I have the same result with both CONFIG_NFCT_PINS_AS_GPIOS and CONFIG_GPIO_AS_PINRESET undefined.

I have stabilized my port, and I'm now in a state where everything works fine when commenting out the erasing of the first two bytes of NRF_UICR->CUSTOMER

When uncommenting, project crashes every few seconds (even by itself, when I'm not using any i/o)

I have added some logs :

NRF_LOG_INFO("Erasing UICR start");
NRF_LOG_FLUSH();
NRF_NVMC->CONFIG = NVMC_CONFIG_WEN_Wen;
NRF_LOG_INFO("Erasing UICR 1");
NRF_LOG_FLUSH();
while (NRF_NVMC->READY == NVMC_READY_READY_Busy){}
NRF_UICR->CUSTOMER[0] = 0x0;
NRF_UICR->CUSTOMER[1] = 0x0;
NRF_NVMC->CONFIG = NVMC_CONFIG_WEN_Ren;
while (NRF_NVMC->READY == NVMC_READY_READY_Busy){}
NRF_LOG_INFO("Erasing UICR end");
NRF_LOG_FLUSH();

Output :

00> <info> app: APP START
00> <info> app: Erasing UICR start
00> <info> app: Reset_reason: 0x00000004.
00> <info> app: Reset_reason resetted: 0x00000000.
00> <info> app: APP START
00> <info> app: Erasing UICR start
00> <info> app: Reset_reason: 0x00000004.
00> <info> app: Reset_reason resetted: 0x00000000.

(...)

RE: NRF_UICR->CUSTOMER read/write on 52832

Kenneth — Mon, 08 Mar 2021 16:22:36 GMT

Do you see the same with CONFIG_NFCT_PINS_AS_GPIOS and/or CONFIG_GPIO_AS_PINRESET not defined?

I just notice both are handled in system_nrf52.c, and if you have any reset, then they may trigger a soft reset. So you would then not see the actual first reset since the actual first reset is then "masked" by the soft reset caused by either of these defines.

Kenneth

RE: NRF_UICR->CUSTOMER read/write on 52832

QuentinFarizon — Mon, 08 Mar 2021 15:59:26 GMT

This is my own board. This is a board that have been produced with a 52811, and on which the 52811 has been later replaced by a 52832 (drop-in replacement)

RE: NRF_UICR->CUSTOMER read/write on 52832

Kenneth — Mon, 08 Mar 2021 15:49:45 GMT

Is there any way you can create an example I can run on an nRF52-DK here?

Can you provide the chip laser markings that is printed on top of the nRF52832 IC?

I am not sure what nRF52 hardware you are using here? Is it a nRF52-DK or your own board?

Kenneth

RE: NRF_UICR->CUSTOMER read/write on 52832

QuentinFarizon — Mon, 08 Mar 2021 14:00:04 GMT

I have the error with and without the bootloader (and my bootloader is unmodified from the examples).

Here is the RESETREAS code :

    uint32_t reset_reason = NRF_POWER->RESETREAS;
    NRF_LOG_INFO("Reset_reason: 0x%08x.\n", reset_reason);
    NRF_POWER->RESETREAS = 0xffffffff;
    reset_reason = NRF_POWER->RESETREAS;
    NRF_LOG_INFO("Reset_reason resetted: 0x%08x.\n", reset_reason);

RE: NRF_UICR->CUSTOMER read/write on 52832

Kenneth — Mon, 08 Mar 2021 10:18:53 GMT

I think there must be something we are overlooking here, can it be that you have a bootloader that is handling/doing something here? Can you share the implementation you have of the RESETREAS in the start of main()?

RE: NRF_UICR->CUSTOMER read/write on 52832

QuentinFarizon — Fri, 05 Mar 2021 14:39:34 GMT

Thanks for your help.

To be sure, I've commented out all calls to NVIC_SystemReset and sd_nvic_SystemReset (including in SDK files).
I have also checked that RESEATREAS is reset by reading after resetting.
I have tried removing CONFIG_GPIO_AS_PINRESET configuration.

The reboot still occurs in all cases (always with RESETREAS 0x4).

RE: NRF_UICR->CUSTOMER read/write on 52832

Kenneth — Thu, 04 Mar 2021 14:08:49 GMT

There is two reasons why the soft reset bit may be set in RESETREAS:

1. Somewhere in your project you are calling NVIC_SystemReset() or sd_nvic_SystemReset().

2. You did not clear the RESEATREAS by writing set bits to 1 on previous reset (this can be checked by reading the bits after clearing to make sure they are all 0):
https://infocenter.nordicsemi.com/topic/com.nordic.infocenter.nrf52832.ps.v1.1/power.html#register.RESETREAS

Can you just double check if any of the above is the case here?

[quote user="QuentinFarizon"]On 52832 I have set CONFIG_GPIO_AS_PINRESET to 1
Could this have an impact ? [/quote]

I don't see how that should affect the soft reset being set, but you can try to remove and see if that changes things.

RE: NRF_UICR->CUSTOMER read/write on 52832

QuentinFarizon — Wed, 03 Mar 2021 18:46:04 GMT

On 52832 I have set CONFIG_GPIO_AS_PINRESET to 1
Could this have an impact ?

I have added the hardfault handler but I don't see any logs from there, nor do any breakpoints break here.

RE: NRF_UICR->CUSTOMER read/write on 52832

Kenneth — Wed, 03 Mar 2021 18:07:04 GMT

Since this is not triggered from app_error_fault_handler(), then I suspect it is triggered by HardFault_process().

Try to add/change #define HARDFAULT_HANDLER_ENABLED 1 to sdk_config.h.

In addition:

Hardfault_handler_<toolchain>.c needs to be included in order for HardFault_c_handler() (by hardfault_implementation.c) to be invoked on a fault exception. Did you include one of the handler files from \components\libraries\hardfault\nrf52\handler (different versions depending on toolchain)?

Edit: The code snippet you shared seems to work on an nRF52832 I have here without any issues.

Best regards,
Kenneth

RE: NRF_UICR->CUSTOMER read/write on 52832

QuentinFarizon — Wed, 03 Mar 2021 16:15:55 GMT

RESEATREAS is 0x4 -> Reset from soft

RE: NRF_UICR->CUSTOMER read/write on 52832

Kenneth — Wed, 03 Mar 2021 15:27:42 GMT

I don't understand what the problem may be, but using that register will help find what the reset reason is, that will help us understand next step.

Kenneth

RE: NRF_UICR->CUSTOMER read/write on 52832

QuentinFarizon — Wed, 03 Mar 2021 14:56:54 GMT

I don't have anything on my code about that. Is this is a known difference between 52811/52832 ?

RE: NRF_UICR->CUSTOMER read/write on 52832

Kenneth — Wed, 03 Mar 2021 14:41:47 GMT

What is the NRF_POWER->RESETREAS on startup? Make sure to clear all bits in RESETREAS after reading by writing 1 to them, else they may be kept on next startup.

Kenneth