APP_TIMER_V2 does not appear to generate timer interrupts

RE: APP_TIMER_V2 does not appear to generate timer interrupts

Karl Ylvisaker — Mon, 13 Jun 2022 12:47:52 GMT

Hello RVM,

I am glad to hear that this seems to resolve your issue - please do not hesitate to let us know if testing proves otherwise.
I will mark the ticket as 'closed' as requested, but you can re-open it at any time if it should be necessary.

Best regards,
Karl

RE: APP_TIMER_V2 does not appear to generate timer interrupts

RVM — Wed, 08 Jun 2022 22:52:07 GMT

[quote userid="87869" url="~/f/nordic-q-a/87642/app_timer_v2-does-not-appear-to-generate-timer-interrupts/370791#370791"]

I have discussed this with some colleagues and I think it comes down to that you will have to make sure that the LFCLK start task is triggered regardless of the WDT status.
You could for example do this through the HAL layer directly:

Fullscreen

[/quote]

I have tried this briefly and it seems to work -- time will tell how robust this is going to be.

I will mark this as 'answered'. Please feel free to close this ticket.

Cheers

RVM

RE: APP_TIMER_V2 does not appear to generate timer interrupts

Karl Ylvisaker — Fri, 03 Jun 2022 06:24:10 GMT

Hello again, RMV

Thank you for your patience with this.

[quote user="RMV"]FYI, I have noticed that when I reset the MCU (via the 5-pin SWD interface) and stop at the reset vector, the RUNSTATUS register in the WDT has the value '1' i.e. WDT is enabled. This does not match the product specification which states that this register should be '0' at reset.[/quote]

I have reached out to the nRF5 SDK developers about this, and checked up on the internal issue that my colleague Susheel submitted in the other ticket, but there unfortunately does not seem to exist an estimate for when this fix might make it into an SDK release.

[quote user="RMV"]I really need to know ASAP whether or not this issue is going to be resolved, or there is some workaround outside of the SDK that I can use to get the LFCLK being recognized by the Applicatiion Timer?[/quote]

...
        nrf_clock_task_trigger(NRF_CLOCK_TASK_LFCLKSTART);
...

You can see the detailed explanation in this ticket.
Try this, and let me know if it resolves the issue you are seeing.

By the way, are you aware of the Errata 132 for the nRF52832? My colleague Håkon mentions its significance in the referenced ticket as well.

Best regards,
Karl

RE: APP_TIMER_V2 does not appear to generate timer interrupts

RVM — Thu, 02 Jun 2022 22:15:07 GMT

Hi Karl

I really need to know ASAP whether or not this issue is going to be resolved, or there is some workaround outside of the SDK that I can use to get the LFCLK being recognized by the Applicatiion Timer?

Also, any clues on the hardware anamoly I noted below, in my update from a few days ago?

RE: APP_TIMER_V2 does not appear to generate timer interrupts

RVM — Sun, 29 May 2022 17:07:32 GMT

FYI, I have noticed that when I reset the MCU (via the 5-pin SWD interface) and stop at the reset vector, the RUNSTATUS register in the WDT has the value '1' i.e. WDT is enabled. This does not match the product specification which states that this register should be '0' at reset.

Later on this bit is tested (in the source files associated with NRF clock driver) to determine the status of the WDT component, and since it returns a TRUE the run status of the LFCLOCK is set/assumed to be ON.

I suspect this is what is causing my issues but I need to dig a bit further into this.

I also vaguely recall another ticket refer to this same issue and simply comment out this fragment of code as a local workaround for that user's use case.

RMV

RE: APP_TIMER_V2 does not appear to generate timer interrupts

Karl Ylvisaker — Sun, 29 May 2022 13:40:17 GMT

Hello again, RVM

[quote user="RMV"]No, I am not using the FreeRTOS -- just provided the link to describe an almost identical problem situation that I am encountering.[/quote]

Thank you for clarifying - this is very helpful for me to know.

I will ask my colleague Susheel if there has been any progress in the issue, as you also have asked in the other ticket.

Best regards,
Karl

RE: APP_TIMER_V2 does not appear to generate timer interrupts

RVM — Fri, 20 May 2022 15:26:31 GMT

[quote userid="87869" url="~/f/nordic-q-a/87642/app_timer_v2-does-not-appear-to-generate-timer-interrupts/368799#368799"]Aha - I see. I was not aware that you were using FreeRTOS + the watchdog module. I see that an update has been requested in the ticket, and that my colleague Susheel will be back in office next week. You can subscribe to the ticket in order to receive a notification whenever there is an update, if you would like.[/quote]

Hi Karl,

No, I am not using the FreeRTOS -- just provided the link to describe an almost identical problem situation that I am encountering.

Cheers

RVM

RE: APP_TIMER_V2 does not appear to generate timer interrupts

Karl Ylvisaker — Fri, 20 May 2022 10:42:45 GMT

Hello again, RMV

Thank you for your extreme patience with this.

[quote user="RMV"]On a different, but related, note I want to understand what it means when the documentation says:
"The RTC library does not directly start or stop the RTC but rather schedules a request to do so in the future".[/quote]

Could you pinpoint exactly where in the documentation you read this, so that I may take a look?
The app timer definitively starts the RTC (if it is not already running), but perhaps this is referring to the case when the SoftDevice is present and enabled, in which the application looses all access to direct clock control, and may only 'request' clock operations through the SoftDevice API?

[quote user="RMV"]There was, however, one more thing that I had to take care of before I got this fully working and that had to do with a bug in the RTC clock driver in SDK v17.x.y, as documented here: here[/quote]

Aha - I see. I was not aware that you were using FreeRTOS + the watchdog module. I see that an update has been requested in the ticket, and that my colleague Susheel will be back in office next week. You can subscribe to the ticket in order to receive a notification whenever there is an update, if you would like.

Best regards,
Karl

RE: APP_TIMER_V2 does not appear to generate timer interrupts

RVM — Fri, 13 May 2022 16:10:05 GMT

[quote userid="87869" url="~/f/nordic-q-a/87642/app_timer_v2-does-not-appear-to-generate-timer-interrupts/367798#367798"]I am happy to hear that you were able to pinpoint the root of your issue, and that it is now working as intended, great! :)[/quote]

Hi Karl,

There was, however, one more thing that I had to take care of before I got this fully working and that had to do with a bug in the RTC clock driver in SDK v17.x.y, as documented here: here

RE: APP_TIMER_V2 does not appear to generate timer interrupts

Karl Ylvisaker — Fri, 13 May 2022 10:19:53 GMT

Hello again, RVM

Thank you for your patience with this.
I am happy to hear that you were able to pinpoint the root of your issue, and that it is now working as intended, great! :)

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

Good luck with your development!

Best regards,
Karl

RE: APP_TIMER_V2 does not appear to generate timer interrupts

RVM — Thu, 12 May 2022 15:21:50 GMT

On a different, but related, note I want to understand what it means when the documentation says:
"The RTC library does not directly start or stop the RTC but rather schedules a request to do so in the future".

What am I missing here? If the call to start the timer does not actually start the RTC then how would the RTC even be prepared to generate an IRQ that a scheduler can work with and use 'in the future'?

RE: APP_TIMER_V2 does not appear to generate timer interrupts

RVM — Wed, 11 May 2022 19:36:09 GMT

Hi Karl,

Thanks for all your insights and help.

I finally figured out the issue with my new design -- I was initializing the scheduler twice.

RVM

RE: APP_TIMER_V2 does not appear to generate timer interrupts

RVM — Tue, 10 May 2022 17:35:40 GMT

[quote userid="87869" url="~/f/nordic-q-a/87642/app_timer_v2-does-not-appear-to-generate-timer-interrupts/366615#366615"]If the timer is set up correctly to trigger a compare event every 1 second it will wake up the device from the SYSTEM_ON sleep to process the event. Once the event has been processed it will go back to sleep, unless there are other events or functions that need processing before it does.[/quote]

I did not go through any steps to set up the core RTC module itself -- I assumed that the app timer module would take care of it when it is initialized.

If that is not the case, what exactly are the steps to set up the RTC? I did not realize that simply enabling the RTC would not be enough and that I have to set up extra 'comparators'?

RE: APP_TIMER_V2 does not appear to generate timer interrupts

RVM — Tue, 10 May 2022 17:32:30 GMT

[quote userid="87869" url="~/f/nordic-q-a/87642/app_timer_v2-does-not-appear-to-generate-timer-interrupts/367140#367140"]On a sidenote, I notice that the naming in the supplied code does not match exactly - is this not the exact code you are running / is it pseudocode just to highlight the steps you are taking?
[/quote]

Hi,

Yeah, it looks like this is not exactly what I use to test -- the timer handler names are identical in my build 🙂

RE: APP_TIMER_V2 does not appear to generate timer interrupts

Karl Ylvisaker — Tue, 10 May 2022 10:38:20 GMT

Hello again,

[quote user="RMV"]The two things (and there may be more) that come to mind are that without the above code, either the RTC module is not fully functional or the Timer IRQ does not have a high enough priority to be noticed by the NVIC.[/quote]

The call to nrf_sdh_enable_request() attempts to start the SoftDevice.
The RTC0 instance will be blocked when the SoftDevice is enabled, but this should not matter to you since you are using the RTC1 instance.
Unfortunately I do not notice anything off about the steps you take to initialize and start the RTC.
Taking a closer look at the app_timer2.c implementation it seems to me that the instance used - RTC1 - is hardcoded, and not based on the APP_TIMER_V2_RTC1_ENABLED define.
I will create an internal ticket to have this examined more closely.

Could you also share with me the APP_TIMER define's from your sdk_config.h?

On a sidenote, I notice that the naming in the supplied code does not match exactly - is this not the exact code you are running / is it pseudocode just to highlight the steps you are taking?

Best regards,
Karl

RE: APP_TIMER_V2 does not appear to generate timer interrupts

RVM — Mon, 09 May 2022 18:41:36 GMT

My experimentation indicates that all I need to do to get the timer to trigger the IRQ as desired is to 'enable the SD Handler" as below just before starting the timer....

if (!nrf_sdh_is_enabled()) {
  lcl_err_code = nrf_sdh_enable_request();
}

The two things (and there may be more) that come to mind are that without the above code, either the RTC module is not fully functional or the Timer IRQ does not have a high enough priority to be noticed by the NVIC.

RE: APP_TIMER_V2 does not appear to generate timer interrupts

RVM — Fri, 06 May 2022 17:11:37 GMT

Hi Karl,

When I do not define the symbol 'APP_TIMER_V2' then this is the design pattern
that has always worked for me with SDK 15.3.0, 16.0.0 and 17.0.1.

/*   *** NEVER GETS CALLED WHEN USING SDK 17.2.0 AND 
     'APP_TIMER_V2' AND 'APP_TIMER_V2_RTC1_ENABLED' ARE DEFINED *** */
void PCT_CORE_TimerHandler_OneSecond(void *timer_data)
{
  // Basic tick counter.
  static uint32_t ticks = 0;
  ticks++;
  return;
}


APP_TIMER_DEF(OneSecond_TimerId);

/* Initialize timer module */
lcl_err_code = app_timer_init();
APP_ERROR_CHECK(lcl_err_code);  // No errors observed
  
/* Create the Reed Switch activation timer */
lcl_err_code = app_timer_create(&OneSecond_TimerId, APP_TIMER_MODE_REPEATED, TimerHandler_OneSecond);
APP_ERROR_CHECK(lcl_err_code); // No errors observed.



#define TIMER_START(T,D)        { ret_code_t lcl_err_code; lcl_err_code = app_timer_start(T, APP_TIMER_TICKS(D), NULL); APP_ERROR_CHECK(lcl_err_code); }

TIMER_START(PCT_CORE_OneSecond_TimerId, 1000);

I am not sure what other steps are required for 'configuration and initialization' of this design.

I also see no difference whether or not I use the scheduler to manage the timer's functionality

RE: APP_TIMER_V2 does not appear to generate timer interrupts

Karl Ylvisaker — Fri, 06 May 2022 08:02:33 GMT

Hello RVM,

[quote user=""]My first question is: I do not quite understand "why do I need to enable the RTC1"? -- and when I try to build without using this symbol then I get a build error that there is no RTC instance defined in the build.[/quote]

The App timer library is based on the RTC1, which is why you will need to enable this RTC instance.

[quote user=""]My next comment (based on observations) is that when I enable RTC1 then I notice that the processor enters the nrf_power_mgmt phase but never fires the timer handler that I expect to be triggered every 1 second. I have validated that[/quote]

Could you show me the code for the configuration and initialization of your app timer instance?
If the timer is set up correctly to trigger a compare event every 1 second it will wake up the device from the SYSTEM_ON sleep to process the event. Once the event has been processed it will go back to sleep, unless there are other events or functions that need processing before it does.

Best regards,
Karl