No source available for "0x6d2" ???

RE: No source available for "0x6d2" ???

bigturtle — Tue, 27 Mar 2018 14:08:16 GMT

I changed the RAM_BASE start address as you suggest and it works ^^.

Thank You

RE: No source available for "0x6d2" ???

Håkon Alseth — Thu, 08 Mar 2018 08:54:51 GMT

I am not able to compile your project successfully, so I have to debug it based on your .out file, and it looks like you have optimization enabled. How does your .ld file look like? The .ld script was missing from your .zip file. From your .map file, it looks like the RAM_BASE start address is 0x20001870, which is too low. You should be at 0x20002080 for s130 with 1 peripheral, and 0 central roles.

Could you try adjusting your RAM start address to match other SDK examples and see if that works?

Cheers,

Håkon

RE: No source available for "0x6d2" ???

bigturtle — Wed, 07 Mar 2018 11:02:58 GMT

I commented out the printf() call and I disabled the flow control but it did not work... My device is still invisible on my phone.

RE: No source available for "0x6d2" ???

Håkon Alseth — Tue, 06 Mar 2018 07:57:26 GMT

Thanks for uploading the project.

You're calling printf in your project, but you do not have retarget.c included in your build. This will cause issues, as the printf call will rely on finding _write() and _read(), and when it does not; it will go into a undefined state.

In addition, you have using UART with flow control enabled:

    const app_uart_comm_params_t comm_params =
    {
        RX_PIN_NUMBER,
        TX_PIN_NUMBER,
        RTS_PIN_NUMBER,
        CTS_PIN_NUMBER,
        APP_UART_FLOW_CONTROL_ENABLED,
        false,
        UART_BAUDRATE_BAUDRATE_Baud115200
    };

This requires that you open the terminal before the application will continue.

Could you try adding retarget.c to your build (or comment out the printf() call) ?

For debugging purposes, you can also disable the flow control, as this would require you to open the terminal in order for the application to run properly.

Best regards,

Håkon

RE: No source available for "0x6d2" ???

bigturtle — Mon, 05 Mar 2018 15:46:33 GMT

devzone.nordicsemi.com/.../ble_5F00_uart_5F00_test_5F00_sdk11.zip

like this ?

RE: No source available for "0x6d2" ???

Håkon Alseth — Mon, 05 Mar 2018 15:11:40 GMT

You can .zip the whole example and attach it to your post when replying (Insert -> Insert image/video/file)

RE: No source available for "0x6d2" ???

bigturtle — Mon, 05 Mar 2018 13:41:37 GMT

can you explain me how to send it please ?

RE: No source available for "0x6d2" ???

Håkon Alseth — Thu, 08 Feb 2018 20:14:23 GMT

Could you share one example that does not work at your end (just .zip the whole example)? one of the example applications from any SDK that you've altered should do.

RE: No source available for "0x6d2" ???

bigturtle — Thu, 08 Feb 2018 16:54:03 GMT

I do a "mass_erase" before flashing softdevice thanks to openOCD. Judging from this openocd document

at page 87 we can read :

nrf5 mass_erase
Erases the contents of the code memory and user information configuration
registers as well. It must be noted that this command works only for chips that
do not have factory pre-programmed region 0 code.

So I think it erases all.

RE: No source available for "0x6d2" ???

Håkon Alseth — Thu, 08 Feb 2018 15:15:17 GMT

Revision 3 includes all variants (128k/16k, 256k/16k, and 256k/32k), it is just that SDK 12 has projects for the "nRF51x22_xxAC" variant, which is the one with 32k RAM.

You are stuck in the softdevice MBR, which is seldom a good sign. Could it be that you have an older bootloader present in flash and UICR? I see that you're using openOCD, do you have the ability to do a erase_all from that debugger?

Here's the sequence for issuing an erase_all to a nRF51 device: https://devzone.nordicsemi.com/b/blog/posts/recover-the-nrf51-when-nrjprog-and-nrfgo-studio-co

Cheers,

Håkon

RE: No source available for "0x6d2" ???

bigturtle — Wed, 07 Feb 2018 20:16:12 GMT

Thanks for your reply

Q1: Yes

Q2: Yes

Q3: No

This turned out to be a big headache so I tried to change SDK and SD

And... I made this work 2 days ago by using SDK 10.0.0 and S130 v1.0.0... but these are old versions.

Does it mean all of the SDK and softdevice which are compatible with rev. 3 judging from this

are not compatible with every version of rev 3 ?

RE: No source available for "0x6d2" ???

Håkon Alseth — Mon, 05 Feb 2018 13:54:12 GMT

Hi,

Q1: Could you ensure that you have set the clock_lf_cfg equal to what is stated in this thread?

Q2: Are you always stuck in the same address space (looping in 0x750)?

Q3: Are you able to run any other ble example successfully (remember to change LFCLK source first!)?

Cheers,

Håkon

RE: No source available for "0x6d2" ???

bigturtle — Mon, 05 Feb 2018 10:46:41 GMT

Ok now I'm stuck in SVCALL(SD_SOFTDEVICE_ENABLE, uint32_t, sd_softdevice_enable(nrf_clock_lf_cfg_t const * p_clock_lf_cfg, nrf_fault_handler_t fault_handler)) in ble_stack_init().

When I step into this function it stops and shows me this:

No source available for "0x750"

I tried to use internal RC clock but it didn't work...

RE: No source available for "0x6d2" ???

bigturtle — Fri, 02 Feb 2018 16:20:09 GMT

this is why it stops... in main.c in the function buttons_leds_init the function bsp_init returns 8 and that error code means :

Invalid state, operation disallowed in this state (judging from nrf_error.h)

Can I have some explanation on this error please?