How can I find source of problem?

RE: How can I find source of problem?

Karl Ylvisaker — Fri, 26 Mar 2021 15:39:32 GMT

[quote user="cemalb"]the problem seems to be fixed. [/quote]

I am happy to hear that the problem seems to be resolved!
While we may not have added a lot of new features during this back-and-forth, we certainly have improved the overall quality of your project, and minimized a lot of potential for future frustration and failure.

[quote user="cemalb"]If I encounter the same problem again, I would bother you :) [/quote]

Sure, that sounds all right with me!

[quote user="cemalb"]Thank you for your help and patience.[/quote]

No problem at all, I am happy to help! :)

Please do not hesitate to open a new ticket if you should encounter any other issues or questions in the future.

Good luck with your development!

Best regards,
Karl

RE: How can I find source of problem?

cemalb — Fri, 26 Mar 2021 15:34:41 GMT

In fact, it cannot be said that I have made much changes in the codes. Although we cannot explain it exactly, the problem seems to be fixed. If I encounter the same problem again, I would bother you :)

Thank you for your help and patience.

RE: How can I find source of problem?

Karl Ylvisaker — Fri, 26 Mar 2021 15:31:23 GMT

It is unfortunately very hard to say exactly why the creation of those two timers failed before you started checking their returned error codes.. but it looks like their creation failed, which led to you call to app_timer_start also failing. I can't really say much more than that, because I do not have any information about whether the conditions of your testing has changed since then, or if you have made any other changes to the code, etc.

Best regards,
Karl

RE: How can I find source of problem?

cemalb — Fri, 26 Mar 2021 15:27:37 GMT

So what happened to the NRF_ERROR_INVALID_STATE error?Why does that error message no longer occur?

RE: How can I find source of problem?

Karl Ylvisaker — Fri, 26 Mar 2021 15:23:49 GMT

[quote user="cemalb"]I added the error code.the program does not fail anymore. I could not understand why. So adding the error code solved everything?[/quote]

No, not exactly. you were passing in a declared variable to your error check, but you never initialized it. When you declare a variable that only means that the compiler will allocate the required memory for that variable type, it does not provide it with a value. So the variable will hold whatever value existed in the allocated memory prior to its allocation - which is pretty much random.

When you then passed this random-valued error code to the error check it immediately failed, because it was != NRF_SUCCESS, causing your application to reset.

I am happy to hear that your program is not failing any more, either way!

Best regards,
Karl

RE: How can I find source of problem?

cemalb — Fri, 26 Mar 2021 15:17:00 GMT

I'm really angry with myself for my carelessness.

I added the error code.the program does not fail anymore. I could not understand why. So adding the error code solved everything?

RE: How can I find source of problem?

Karl Ylvisaker — Fri, 26 Mar 2021 15:08:09 GMT

You are not checking the returned error of your call to app_timer_stop on line 377, you are checking the randomly instantiated value of err_code.

Best regards,
Karl

RE: How can I find source of problem?

cemalb — Fri, 26 Mar 2021 15:02:02 GMT

😅

RE: How can I find source of problem?

Karl Ylvisaker — Fri, 26 Mar 2021 15:00:05 GMT

I see from your screenshot that this is in fact line 385 and 386.
Please share the snippet of the function containing line 377.

[quote user="cemalb"]I forgot to share the code on line 377 because I was tired. sorry. [/quote]

No need to apologize - I hope you get some rest soon! :)

Best regards,
Karl

RE: How can I find source of problem?

cemalb — Fri, 26 Mar 2021 14:56:15 GMT

RE: How can I find source of problem?

cemalb — Fri, 26 Mar 2021 14:54:44 GMT

I forgot to share the code on line 377 because I was tired. sorry. Since I do not reply to your last answer, I am writing here.

At line 377, a function is called to stop a timer:

void stopTimer_open_door_timer(){
ret_code_t err_code;
err_code = app_timer_stop(m_open_door_timer_id); /* Stop open door timer. */
APP_ERROR_CHECK(err_code);
}

RE: How can I find source of problem?

Karl Ylvisaker — Fri, 26 Mar 2021 14:40:14 GMT

[quote user="cemalb"]Ok, I will add immediately.[/quote]

Great! If you now see that the previous developer never checked any error codes you might have to review your whole project for this, but rest assured that it is essential to check error codes to know whether a function has completed successfully or not.

[quote user="cemalb"]

I guess the error has changed. the last error is:

<error> app: ERROR 536907338 [Unknown error code] at XXX\V1.5.9\examples\btc_ble_peripheral\xxxBluetooth\timer_init.c:377
PC at: 0x000336CB
<error> app: End of error report

[/quote]

What function is being called at line 377 of your timer_init.c file?
Keep in mind that I do not have you project, so there is no way for me to know what is happening here unless you tell me.
Please also share the snippet of code of the function containing line 377 of timer_init.c.

Best regards,
Karl

RE: How can I find source of problem?

cemalb — Fri, 26 Mar 2021 14:34:54 GMT

[quote userid="87869" url="~/f/nordic-q-a/72111/how-can-i-find-source-of-problem/302165#302165"]Please make sure to add APP_ERROR_CHECK to all places where it is appropriate in your code,
not just the timer_init function.[/quote]

Ok, I will add immediately.

[quote userid="87869" url="~/f/nordic-q-a/72111/how-can-i-find-source-of-problem/302165#302165"]Are you still getting the NRF_ERROR_INVALID_STATE returned from your two app_timer_start calls, or has that too disappeared? If so, what error code is being returned currently?[/quote]

I guess the error has changed. the last error is:

<error> app: ERROR 536907338 [Unknown error code] at XXX\V1.5.9\examples\btc_ble_peripheral\xxxBluetooth\timer_init.c:377
PC at: 0x000336CB
<error> app: End of error report

RE: How can I find source of problem?

Karl Ylvisaker — Fri, 26 Mar 2021 14:25:03 GMT

[quote user="cemalb"]I replied late because I had to deal with other projects.[/quote]

No problem at all - we continue whenever you have the chance! :)

[quote user="cemalb"]While creating timers, I added error codes for all of them, but they didn't return any error codes. they all return 0.[/quote]

I am happy to see that you have added this!
Please make sure to add APP_ERROR_CHECK to all places where it is appropriate in your code,
not just the timer_init function.

So they all return successfully, that is good news!
Are you still getting the NRF_ERROR_INVALID_STATE returned from your two app_timer_start calls, or has that too disappeared? If so, what error code is being returned currently?

Best regards,
Karl

RE: How can I find source of problem?

cemalb — Fri, 26 Mar 2021 13:11:32 GMT

Hi, I replied late because I had to deal with other projects.

[quote userid="87869" url="~/f/nordic-q-a/72111/how-can-i-find-source-of-problem/300962#300962"]Please add error code checks to all functions currently missing them, see what error code is being returned by the functions, and then check the API Reference to see why the function call failed.[/quote]

While creating timers, I added error codes for all of them, but they didn't return any error codes. they all return 0.

RE: How can I find source of problem?

Karl Ylvisaker — Fri, 19 Mar 2021 13:53:18 GMT

Hello again,

[quote user="cemalb"]I already start all timers with the timers_init() function inside the main() function.[/quote]

No, you currently attempt to start all the timers. You never actually check whether you were successful in doing so or not.
You are not checking any of the error codes returned from all your calls to app_timer_create.
Please always check the API reference / documentation of functions you are using that you have not written yourself - such as SDK functions - before using them, and please always check the returned error codes of functions you end up using in your code.
If you do not check the returned error codes, you will have no way of knowing whether a function call has failed for whatever reason, or if you application to could proceed as usual.
Checking the returned error codes is crucial - it is also why we have taken the time to implemented them in the first place.

Please add error code checks to all functions currently missing them, see what error code is being returned by the functions, and then check the API Reference to see why the function call failed.
Do this, and let me know if it resolves your issue.

Looking forward to resolve these issues together.

Best regards,
Karl

RE: How can I find source of problem?

cemalb — Fri, 19 Mar 2021 11:13:19 GMT

Hi,

[quote userid="87869" url="~/f/nordic-q-a/72111/how-can-i-find-source-of-problem/300870#300870"]Please check that you have created the two timers in question, prior to attempting to start them.[/quote]

I already start all timers with the timers_init() function inside the main() function.

RE: How can I find source of problem?

Karl Ylvisaker — Fri, 19 Mar 2021 10:14:20 GMT

Hello,

[quote user="cemalb"]I did not change it. it is 6 by default.[/quote]

Thank you for confirming.

[quote user="cemalb"]I forgot to return the error code value for the function that caused the error. I fixed it. the error code is[/quote][quote user="cemalb"]The codes in the relevant line are as follows:[/quote][quote user="cemalb"]but how can i solve this? [/quote]

When an error code is returned from a function, you should always check the functions API reference to determine the cause of the error. In this case, the app_timer_start function documentation about the returned error reads:

NRF_ERROR_INVALID_STATE

If the application timer module has not been initialized or the timer has not been created.

So, you have either not initialized the timer module, or created this particular timer you are trying to use. Since your other timers are working as intended, I would think the latter is the case here.
Please check that you have created the two timers in question, prior to attempting to start them.

Best regards,
Karl

RE: How can I find source of problem?

cemalb — Fri, 19 Mar 2021 10:01:25 GMT

Hi,

[quote userid="87869" url="~/f/nordic-q-a/72111/how-can-i-find-source-of-problem/300624#300624"]If you were using interrupt priority 0 for your UART, then that would definitely be a problem, yes.
Now you are using the NRFX_UART_DEFAULT_CONFIG_IRQ_PRIORITY, is this still the default value, or have you modified this? [/quote]

I did not change it. it is 6 by default.

[quote userid="87869" url="~/f/nordic-q-a/72111/how-can-i-find-source-of-problem/300624#300624"]It seems the error is triggered at line 299 of your 'timer_init.c' file. Which function is being called here? Could you also share the entire function being executed in line 299 of timer_init.c, so I may take a look?[/quote]

I forgot to return the error code value for the function that caused the error. I fixed it. the error code is:

<error> app: ERROR 8 [NRF_ERROR_INVALID_STATE] at D:\1_1_xxx_xxx_xxx\V.1.5.9_CML\V1.5.9\V1.5.9\examples\xxx_ble_peripheral\XXX_Project\timer_init.c:299
PC at: 0x000335C7
<error> app: End of error report

The codes in the relevant line are as follows:

also when I cancel this line, it gives the same warning for line 285:

I guess there was a problem while creating the timers.I use 5 more timers other than these two timers. however, there was no such problem for them.

but how can i solve this?

RE: How can I find source of problem?

Karl Ylvisaker — Thu, 18 Mar 2021 10:18:19 GMT

Hello,

[quote user="cemalb"]

its initial value is 0. this conflicts with the softdevice priority value.

Could this be what caused the problem?

[/quote]

If you were using interrupt priority 0 for your UART, then that would definitely be a problem, yes.
Now you are using the NRFX_UART_DEFAULT_CONFIG_IRQ_PRIORITY, is this still the default value, or have you modified this?

[quote user="cemalb"]<error> app: ERROR 211843 [Unknown error code] at D:\1_1_kale_kilit_yazilim\V.1.5.9_CML\V1.5.9\V1.5.9\examples\btc_ble_peripheral\BTC_Project\timer_init.c:299 PC at: 0x000335C5
<error> app: End of error report[/quote]

It seems the error is triggered at line 299 of your 'timer_init.c' file. Which function is being called here? Could you also share the entire function being executed in line 299 of timer_init.c, so I may take a look?

Best regards,
Karl

RE: How can I find source of problem?

cemalb — Wed, 17 Mar 2021 09:28:45 GMT

When I insert this line and debug it I got another error message:

<error> app: ERROR 211843 [Unknown error code] at D:\1_1_kale_kilit_yazilim\V.1.5.9_CML\V1.5.9\V1.5.9\examples\btc_ble_peripheral\BTC_Project\timer_init.c:299 PC at: 0x000335C5
<error> app: End of error report

RE: How can I find source of problem?

cemalb — Wed, 17 Mar 2021 09:20:15 GMT

Hi,

I have just realized this situation.

The comm_params variable is used when making the initial settings for the uart. however, the initial value of the interrupt_priority element belonging to the nrf_drv_uart_config_t struct is not given while doing this setting. therefore, its initial value is 0. this conflicts with the softdevice priority value.

Could this be what caused the problem?

RE: How can I find source of problem?

cemalb — Wed, 17 Mar 2021 08:45:09 GMT

[quote userid="87869" url="~/f/nordic-q-a/72111/how-can-i-find-source-of-problem/299856#299856"]Did you ever check whether any of the peripherals you are using in your project attempt to set a priority that conflicts with the priority levels reserved for the SoftDevice that I mentioned earlier?[/quote]

wdt, timer and uart are used in the project. interrupt priority level setted 6 for all of them. I think there is no conflict since softdevice 132 uses interrupt priority level 0-1-4.

RE: How can I find source of problem?

cemalb — Wed, 17 Mar 2021 06:30:11 GMT

I forgot to share this. I am so sorry.

devzone.nordicsemi.com/.../6443.report_5F00_only_5F00_difference.htm

I will check it out.

RE: How can I find source of problem?

Karl Ylvisaker — Mon, 15 Mar 2021 14:47:13 GMT

Hello again,

[quote user="cemalb"]Yes, unfortunately. it gives the same error again. I also set up for the comparator. i just created a report showing the only differences. Do you think there is a difference that causes our problem? [/quote]

The error points towards the 107th line of nrfx_glue.h which is an assert that triggers if an invalid priority level is attempted used.
Did you ever check whether any of the peripherals you are using in your project attempt to set a priority that conflicts with the priority levels reserved for the SoftDevice that I mentioned earlier?
Those top priority levels are only available when the SoftDevice is not present.

I do not suspect that a change has been made to the SDK, but it is still good for you and the future development to be aware of any changes the previous developer might have made, if any. So, it is still good that you have taken the time to verify the integrity of the SDK you have been handed from the previous developer.

[quote user="cemalb"]By the way, I fixed the spaces in the folder names as you said. Thank you for your warning.[/quote]

Great, I am happy to hear that you found it helpful!

Best regards,
Karl