RE: Problem with emulated System OFF mode with NRF52

Remo — Mon, 17 Oct 2016 07:59:43 GMT

Thank you very much for this detailed explanation!

RE: Problem with emulated System OFF mode with NRF52

Ole Bauck — Fri, 14 Oct 2016 12:05:27 GMT

According to nRF51 and nRF52 specification when entering emulated system off it is recommended to add an infinite loop directly after entering System OFF. Problem here is that sd_power_system_off() will execute instructions after entering emulated system off (including returning to the application) before the application code is run. On nRF51 this leads to a hardfault.

You can check this by stopping code execution in the debugger after calling sd_power_system_off(). The ISR number (xPSR->ISR in the registers panel) will be 3, which is the HardFault_IRQn (see nrf51.h, you have to subtract 16 from the ISR number as this starts at 1 and the enum starts at -15).

On the nRF52 it does not hardfault and the function returns, which is why it seems that the two chips are acting different to emulated system off.

I will report this internally and see if we can change the sd_power_system_off() function.

For your other question this is a bug with the nRF52 and is found in the errata:

GPIO: PIN_CNF is not retained when in debug interface mode

...

Symptoms

GPIO pin configuration is reset on wakeup from System OFF.