Detection of abnormal operation caused by interrupt processing

Patent metadata

Field	Value
Publication number	US-9436627-B2
Application number	US-201214239832-A
Country	US
Kind code	B2
Filing date	Jul 25, 2012
Priority date	Aug 25, 2011
Publication date	Sep 6, 2016
Grant date	Sep 6, 2016

Abstract

A controller for controlling interrupt processing in a multiple-interrupt system is provided. The controller includes multiple watchdog timers (WDTs), each provided for each of interrupt priorities. The controller includes interrupt priority selectors, each of which receives each interrupt request signal and outputs an activation signal to a corresponding WDT according to the priority of the interrupt request signal. The controller includes an interrupt processing circuit, which when a WDT has timed out, outputs, to a processor, an interrupt request signal having a priority one or more levels higher than the priority corresponding to the WDT. When multiple causes of interrupt are assigned to one of the interrupt priorities, the interrupt processing circuit gives priority to an interrupt request signal caused by the timeout of a WDT lower in priority level than the interrupt priority to detect that an abnormal operation has occurred in interrupt processing having the lower level priority.

First claim

The invention claimed is:

1. A method, in a data processing system, of detecting an abnormal operation caused by interrupt processing in a multiple-interrupt system, comprising:
preparing a plurality of watchdog timers (WDTs), wherein each WDT has a predetermined time-out value for a respective interrupt priority;
starting a given WDT responsive to an interrupt request having a corresponding priority;
responsive to detecting the given WDT has timed out, accepting an interrupt request having a priority one or more levels higher than a given interrupt priority corresponding to the given WDT; and
responsive to determining a cause of interrupt is assigned to the given interrupt priority, giving priority to a lower level interrupt request caused by a timeout of a WDT having an interrupt priority one level lower than the given interrupt priority to detect that an abnormal operation has occurred in processing the lower level interrupt request.

2. The method according to claim 1 , further comprising:
responsive to a first interrupt request of the cause of interrupt, starting a WDT corresponding to the first interrupt request.

3. The method according to claim 1 , further comprising resetting each of the plurality of WDTs responsive to a corresponding interrupt request being accepted and interrupt processing being started by a processor.

4. The method according to claim 1 , wherein accepting the interrupt request comprises accepting a plurality of interrupt requests having priorities one or more levels higher than the given interrupt priority.

5. The method according to claim 1 , further comprising resetting the data processing system responsive to a WDT corresponding to an interrupt whose priority is highest has timed out.

6. The method according to claim 1 , further comprising outputting, to a processor, an interrupt request signal having a priority one or more levels higher than the given interrupt priority corresponding to the given WDT.

7. The method according to claim 6 , further comprising suppressing output of any other interrupt request signal having a priority level less than or equal to a corresponding priority upon outputting the interrupt request signal.

8. The method according to claim 1 , wherein the predetermined timeout value of the given WDT is greater than the predetermined timeout value of a WDT having an interrupt priority lower than the interrupt priority of the given WDT.

9. The method according to claim 1 , wherein the predetermined timeout value of the given WDT is less than the predetermined timeout value of a WDT having an interrupt priority higher than the interrupt priority of the given WDT.

10. The method according to claim 1 , further comprising:
responsive to detecting the given WDT has timed out, asserting an interrupt request signal corresponding to a higher level interrupt priority.

11. The method of claim 10 , further comprising:
responsive to asserting the interrupt request signal corresponding to the higher level interrupt priority, starting a WDT corresponding to the higher level interrupt priority.

12. A controller, in a data processing system, for controlling interrupt processing in a multiple-interrupt system, comprising:
a plurality of watchdog timers (WDTs) each having a predetermined time-out value provided for a respective interrupt priority;
an interrupt priority selector for receiving an interrupt request signal from a device and outputting an activation signal to a corresponding one of the plurality of WDTs according to the priority of each interrupt request signal; and
an interrupt processing circuit which, responsive to detecting a given WDT has timed out, outputs, to a processor, an interrupt request signal having a priority one or more levels higher than a given interrupt priority corresponding to the given WDT,
wherein responsive to determining a cause of interrupt is assigned to the given interrupt priority, the interrupt processing circuit gives priority to a lower level interrupt request signal caused by a timeout of a WDT having an interrupt priority one level lower than the given interrupt priority to detect that an abnormal operation has occurred in processing the lower level interrupt request.

13. The controller according to claim 12 , wherein the interrupt processing circuit suppresses output of any other interrupt request signal having a priority level less than or equal to a corresponding priority upon outputting the interrupt request signal.

14. The controller according to claim 12 , wherein each of the plurality of WDTs is reset responsive to a corresponding interrupt request being accepted and interrupt processing being started by the processor.

15. The controller according to claim 12 , wherein responsive to detecting a WDT having a highest interrupt priority has timed out, the WDT having the highest priority outputs a signal for resetting the system.

16. The controller according to claim 12 , wherein responsive to a first interrupt request of the cause of interrupt, the interrupt processing circuit starts a WDT corresponding to the first interrupt request.

17. The controller according to claim 12 , wherein accepting the interrupt request comprises accepting a plurality of interrupt requests having priorities one or more levels higher than the given interrupt priority.

18. The controller according to claim 12 , wherein the predetermined timeout value of the given WDT is greater than the predetermined timeout value of a WDT having an interrupt priority lower than the interrupt priority of the given WDT.

19. The controller according to claim 12 , wherein the predetermined timeout value of the given WDT is less than the predetermined timeout value of a WDT having an interrupt priority higher than the interrupt priority of the given WDT.

20. The controller according to claim 12 , wherein responsive to the given WDT timing out, the given WDT asserts an interrupt request signal corresponding to a higher level interrupt priority;
wherein responsive to assertion of the higher level interrupt priority, a WDT corresponding to the higher level interrupt priority starts.

Assignees

• Shiratori Toshiyuki
• IBM

Inventors

• Shiratori Toshiyuki

Classifications

• G06F11/0751

  Error or fault detection not based on redundancy (power supply failures G06F1/30; network fault management H04L41/06) · CPC title

• G06F11/0757

  by exceeding a time limit, i.e. time-out, e.g. watchdogs · CPC title

• G06F13/26 Primary

  with priority control · CPC title

• G06F11/0706

  the processing taking place on a specific hardware platform or in a specific software environment · CPC title