Watchdog mechanism with fault recovery
View Patent ↗A method for handling watchdog events of an electronic device includes detecting a watchdog fault in a normal mode, which is a watchdog event in which a watchdog trigger is not correctly serviced; entering from the normal mode into a first escalation level of nx escalation levels upon detection of the watchdog fault, wherein nx is an integer equal to or greater than 1; detecting correct watchdog events, which are watchdog events in which a watchdog trigger is correctly serviced; and concurrently detecting watchdog faults, leaving the first escalation level if a first escalation condition is met, and recovering in a recovering step back from any of the nx escalation levels to a previous level or mode, if a de-escalation condition is met. An electronic device embodiment includes a CPU and program instructions for carrying out the method.
1. A method comprising:
detecting an incorrect watchdog trigger in a normal mode;
entering from the normal mode into a first state of a first escalation level of a plurality of escalation levels upon detection of the watchdog fault;
entering from the first state of the first escalation level to a second state of the first escalation level if a correct watchdog trigger is detected;
returning to the first state of the first escalation level from the second state of the first escalation level if the incorrect watchdog trigger is detected;
entering from the first escalation level into a second escalation level of the plurality of escalation levels if:
a number of detections of the watchdog fault while at the first escalation level exceeds a predetermined maximum; or
the watchdog fault is not corrected within a predetermined period; and
returning from the second state of the first escalation level to the normal mode if consecutive detections of the correct watchdog trigger exceeds a predetermined threshold.
2. The method of claim 1 , wherein the step of entering from the first escalation level into a second escalation level further comprises entering into a first state of the second escalation level.
3. The method of claim 2 , wherein the method further comprises entering from the first state of the second escalation level to a second state of the second escalation level if a correct watchdog trigger is detected.
4. The method of claim 3 , wherein the method further comprises returning to the first state of the second escalation level to the second state of the second escalation level if the incorrect watchdog trigger is detected.
5. The method of claim 4 , wherein the predetermined maximum and the predetermined period further comprise a first predetermined maximum and a first predetermined period, respectively, and wherein the method further comprises entering from the second escalation level into a safe mode if:
a number of detections of the watchdog fault while at the second escalation level exceeds a second predetermined maximum; or
the watchdog fault is not corrected within a second predetermined period.
6. The method of claim 5 , wherein the step of entering from the normal mode into the first state of the first escalation level of the plurality of escalation levels upon detection of the watchdog fault further comprises setting a first parameter and a second parameter to zero.
7. The method of claim 6 , wherein the method further comprises iterating the first parameter each time the incorrect watchdog trigger is detected while at the first escalation level so as to generate the number of detections of the watchdog fault while at the first escalation.
8. The method of claim 7 , wherein the method further comprises iterating the second parameter each time the incorrect watchdog trigger is detected while at the second escalation level so as to generate the number of detections of the watchdog fault while at the second escalation.
9. The method of claim 8 , wherein the threshold further comprises a first threshold, and wherein the method comprises returning from the second state of the second escalation level to the normal mode if consecutive detections of the correct watchdog trigger exceeds a second predetermined threshold.
10. An apparatus having a processor and a non-transitory computer readable storage medium with computer instructions embodied thereon, the computer instructions performing the steps:
detecting an incorrect watchdog trigger in a normal mode;
entering from the normal mode into a first state of a first escalation level of a plurality of escalation levels upon detection of the watchdog fault;
entering from the first state of the first escalation level to a second state of the first escalation level if a correct watchdog trigger is detected;
returning to the first state of the first escalation level from the second state of the first escalation level if the incorrect watchdog trigger is detected;
entering from the first escalation level into a second escalation level of the plurality of escalation levels if:
a number of detections of the watchdog fault while at the first escalation level exceeds a predetermined maximum; or
the watchdog fault is not corrected within a predetermined period; and
returning from the second state of the first escalation level to the normal mode if consecutive detections of the correct watchdog trigger exceeds a predetermined threshold.
11. The computer instructions of claim 10 , entering from the first escalation level into a second escalation level further comprises entering into a first state of the second escalation level.
12. The computer instructions of claim 11 , wherein the computer instructions further comprises entering from the first state of the second escalation level to a second state of the second escalation level if a correct watchdog trigger is detected.
13. The computer instructions of claim 12 , wherein the computer instructions further comprises returning to the first state of the second escalation level to the second state of the second escalation level if the incorrect watchdog trigger is detected.
14. The computer instructions of claim 13 , wherein the predetermined maximum and the predetermined period further comprise a first predetermined maximum and a first predetermined period, respectively, and wherein the computer instructions further comprises entering from the second escalation level into a safe mode if:
a number of detections of the watchdog fault while at the second escalation level exceeds a second predetermined maximum; or
the watchdog fault is not corrected within a second predetermined period.
15. The computer instructions of claim 14 , wherein entering from the normal mode into the first state of the first escalation level of the plurality of escalation levels upon detection of the watchdog fault further comprises setting a first parameter and a second parameter to zero.
16. The computer instructions of claim 15 , wherein the computer instructions further comprises iterating the first parameter each time the incorrect watchdog trigger is detected while at the first escalation level so as to generate the number of detections of the watchdog fault while at the first escalation.
17. The computer instructions of claim 16 , wherein the computer instructions further comprises iterating the second parameter each time the incorrect watchdog trigger is detected while at the second escalation level so as to generate the number of detections of the watchdog fault while at the second escalation.
18. The computer instructions of claim 17 , wherein the threshold further comprises a first threshold, and wherein the computer instructions comprises returning from the second state of the second escalation level to the normal mode if consecutive detections of the correct watchdog trigger exceeds a second predetermined threshold.