RTC1 interrupt does't trigger, and what's event counter?

[akita]

I'm try to use RTC1 to generate overflow interrupt, but it doesn't trigger interrupt.

void RTC1_IRQHandler(void)
{
  // do something
}

main()
{
  NRF_RTC1->TASKS_STOP = 1;
  NRF_RTC1->INTENSET = 0x2; // enable OVRFLW interrupt
//  NVIC_SetVector(RTC1_IRQn, (uint32_t) &RTC1_IRQHandler); // needed?
  NVIC_EnableIRQ(RTC1_IRQn );
  NRF_RTC1->TASKS_START = 1;
  while(1)
}

In my understanding, the vector for RTC1_IRQn() is already assigned in start-up code, so I skipped NVIC_SetVector(), however, the interrupt doesn't trigger even if setting NVIC_SetVector().

By watching the value of RTC1->COUNTER, it increments automatically, and OVRFLW event occurs by enabling overflow event by setting "NRF_RTC1->EVTENSET = 0x2;"

I also confusing the function of event counter. Does the event counter increments at every event? For example, in the following code, the event counter, NRF_RTC1->EVENTS_TICK, becomes 1 after the first tick, but it keeps 1. Doesn't event counter increase more than 1? (In other words, can I check it simply for checking event occurred or not?)

main()
{
  NRF_RTC1->TASKS_STOP = 1;
  NRF_RTC1->EVTENSET = 0x1; // enable Tick event
  NRF_RTC1->TASKS_START = 1;
  while(1)
}

[FormerMember]

Did you remember to start the LFCLK before starting the RTC?

[Turbo J]

EVENTS_TICK register will never contain any other value than 1 or 0.

[akita]

Yes, I'm sure LFCLK has started in startup routine in mbed (I'm using mbed-compiler for programming). Does RTC1's counter increments even if LFCLK doesn't start? (In my situation ,I've confirmed that RTC1's counter increments at specified speed by prescaler.)

[FormerMember]

Could you double check if the LFCLK is running by reading the LFCLKSTAT register? That will make the debugging easier.

[akita]

I've checked LFCLKSTAT indicates that LFCLK is not running. I've tried to start it (writing 1 to LFCLKSTART), but LFCLKSTAT still indicates that LFCLK is not running. I'll continue to debug it, but I wonder that RTC1 actually increments, even if LFCLK seems to not running; this is right?