SREQ and DOG set in RESETREASON register

We have fairly regularly observed that the reset reason register gives two reasons for reset: software request and watchdog.

Is it plausible that these bits are genuinely set at the same time? If so, under which conditions?

Alternatively, is it possible that this register accumulates reasons and we are supposed to clear them?

Our software consists of a bootloader (the one shipped with the thread support SDK) and an application image.

Parents
  • Hi,

    Is it plausible that these bits are genuinely set at the same time? If so, under which conditions?

    No. A reset can only have one cause.

    Alternatively, is it possible that this register accumulates reasons and we are supposed to clear them

    Yes. You are right, the REASEREAS register is cumulative, and cleared by writnig to the field you want to clear. 

Reply
  • Hi,

    Is it plausible that these bits are genuinely set at the same time? If so, under which conditions?

    No. A reset can only have one cause.

    Alternatively, is it possible that this register accumulates reasons and we are supposed to clear them

    Yes. You are right, the REASEREAS register is cumulative, and cleared by writnig to the field you want to clear. 

Children