hardware_watchdog

Functions
void	watchdog_reboot (uint32_t pc, uint32_t sp, uint32_t delay_ms)
	Define actions to perform at watchdog timeout. More...
void	watchdog_start_tick (uint cycles)
	Start the watchdog tick. More...
void	watchdog_update (void)
	Reload the watchdog counter with the amount of time set in watchdog_enable.
void	watchdog_enable (uint32_t delay_ms, bool pause_on_debug)
	Enable the watchdog. More...
bool	watchdog_caused_reboot (void)
	Did the watchdog cause the last reboot? More...
bool	watchdog_enable_caused_reboot (void)
	Did watchdog_enable cause the last reboot? More...
uint32_t	watchdog_get_count (void)
	Returns the number of microseconds before the watchdog will reboot the chip. More...

Detailed Description

Hardware Watchdog Timer API

Supporting functions for the Pico hardware watchdog timer.

The RP2040 has a built in HW watchdog Timer. This is a countdown timer that can restart parts of the chip if it reaches zero. For example, this can be used to restart the processor if the software running on it gets stuck in an infinite loop or similar. The programmer has to periodically write a value to the watchdog to stop it reaching zero.

Example

 
#include <stdio.h>
#include "pico/stdlib.h"
#include "hardware/watchdog.h"
 
int main() {
    stdio_init_all();
 
    if (watchdog_caused_reboot()) {
        printf("Rebooted by Watchdog!\n");
        return 0;
    } else {
        printf("Clean boot\n");
    }
 
    // Enable the watchdog, requiring the watchdog to be updated every 100ms or the chip will reboot
    // second arg is pause on debug which means the watchdog will pause when stepping through code
    watchdog_enable(100, 1);
 
    for (uint i = 0; i < 5; i++) {
        printf("Updating watchdog %d\n", i);
        watchdog_update();
    }
 
    // Wait in an infinite loop and don't update the watchdog so it reboots us
    printf("Waiting to be rebooted by watchdog\n");
    while(1);
}

Function Documentation

watchdog_caused_reboot()

bool watchdog_caused_reboot

(

void

)

Did the watchdog cause the last reboot?

Returns

true If the watchdog timer or a watchdog force caused the last reboot

false If there has been no watchdog reboot since the last power on reset. A power on reset is typically caused by a power cycle or the run pin (reset button) being toggled.

watchdog_enable()

void watchdog_enable	(	uint32_t	delay_ms,
		bool	pause_on_debug
	)

Enable the watchdog.

Note

If watchdog_start_tick value does not give a 1MHz clock to the watchdog system, then the delay_ms parameter will not be in microseconds. See the datasheet for more details.

By default the SDK assumes a 12MHz XOSC and sets the watchdog_start_tick appropriately.

This method sets a marker in the watchdog scratch register 4 that is checked by watchdog_enable_caused_reboot. If the device is subsequently reset via a call to watchdog_reboot (including for example by dragging a UF2 onto the RPI-RP2), then this value will be cleared, and so watchdog_enable_caused_reboot will return false.

Parameters

delay_ms	Number of milliseconds before watchdog will reboot without watchdog_update being called. Maximum of 0x7fffff, which is approximately 8.3 seconds
pause_on_debug	If the watchdog should be paused when the debugger is stepping through code

watchdog_enable_caused_reboot()

bool watchdog_enable_caused_reboot

(

void

)

Did watchdog_enable cause the last reboot?

Perform additional checking along with watchdog_caused_reboot to determine if a watchdog timeout initiated by watchdog_enable caused the last reboot.

This method checks for a special value in watchdog scratch register 4 placed there by watchdog_enable. This would not be present if a watchdog reset is initiated by watchdog_reboot or by the RP2040 bootrom (e.g. dragging a UF2 onto the RPI-RP2 drive).

Returns

true If the watchdog timer or a watchdog force caused (see watchdog_caused_reboot) the last reboot and the watchdog reboot happened after watchdog_enable was called

false If there has been no watchdog reboot since the last power on reset, or the watchdog reboot was not caused by a watchdog timeout after watchdog_enable was called. A power on reset is typically caused by a power cycle or the run pin (reset button) being toggled.

watchdog_get_count()

uint32_t watchdog_get_count

(

void

)

Returns the number of microseconds before the watchdog will reboot the chip.

Returns

The number of microseconds before the watchdog will reboot the chip.

watchdog_reboot()

void watchdog_reboot	(	uint32_t	pc,
		uint32_t	sp,
		uint32_t	delay_ms
	)

Define actions to perform at watchdog timeout.

Note

If watchdog_start_tick value does not give a 1MHz clock to the watchdog system, then the delay_ms parameter will not be in microseconds. See the datasheet for more details.

By default the SDK assumes a 12MHz XOSC and sets the watchdog_start_tick appropriately.

Parameters

pc	If Zero, a standard boot will be performed, if non-zero this is the program counter to jump to on reset.
sp	If pc is non-zero, this will be the stack pointer used.
delay_ms	Initial load value. Maximum value 0x7fffff, approximately 8.3s.

watchdog_start_tick()

void watchdog_start_tick

(

uint

cycles

)

Start the watchdog tick.

Parameters

cycles

This needs to be a divider that when applied to the XOSC input, produces a 1MHz clock. So if the XOSC is 12MHz, this will need to be 12.