System and method for switching the role of boot processor to a spare processor responsive to detection of loss of lockstep in a boot processor

1. A method comprising:

detecting loss of lockstep (LOL) for a processor in a multi-processor system;

determining that the processor for which said LOL is detected is assigned a role of boot processor;

changing a system identifier, used by the system's operating system to recognize the boot processor, of said boot processor for which said LOL is detected, to a different value, which causes the system's operating system to not recognize the processor for which said LOL is detected as the boot processor; and

changing a system identifier of the spare processor to that of the boot processor, thereby switching the role of boot processor to a spare processor without shutting down the system's operating system.

2. The method of claim 1 wherein the spare processor is a hot spare that is held in idle until switched the role of boot processor.

3. The method of claim 1 wherein the spare processor is an active spare.

4. The method of claim 3 wherein said switching further comprises:

causing the system's operating system to idle the active spare.

5. The method of claim 1 wherein the spare processor is a processor module that includes a master and slave processors that operate in lockstep.

6. The method of claim 1 wherein said determining that the processor for which said LOL is detected is assigned the role of boot processor comprises:

accessing information stored to non-volatile data storage that identifies one of the processors in said multi-processor system that is assigned the role of boot processor.

7. The method of claim 1 further comprising:

determining the processor that is assigned the role of spare processor.

8. The method of claim 7 wherein said determining the processor that is assigned the role of spare processor comprises:

accessing information stored to non-volatile data storage that identifies one of the processors in said multi-processor system that is assigned the role of spare processor.

9. The method of claim 1 wherein said determining step is performed by system firmware.

10. The method of claim 1 wherein said switching step is performed by system firmware.

11. The method of claim 1 wherein the processor for which said LOL is detected includes a lockstep pair of processors, and wherein said switching comprises:

determining one of the lockstep pair of processors that is not the cause of said LOL; and

copying the state of the determined one of said lockstep pair of processors that is not the cause of said LOL to said spare processor.

12. The method of claim 11 wherein said spare processor includes a lockstep pair of processors, and wherein said copying comprises:

copying the state of the determined one of said lockstep pair of processors that is not the cause of said LOL to each of said lockstep pair of processors of said spare processor.

13. A method comprising:

system firmware determining that loss of lockstep (LOL) is detected for a lockstep pair of processors that are assigned the role of boot processor in a system;

determining one of said lockstep pair of processors that is not the cause of said LOL;

copying the state of the determined one of said lockstep pair of processors that is not the cause of said LOL to a spare processor;

changing a system identifier, used by the system's operating system to recognize the boot processor, of said lockstep pair of processors that are assigned the role of boot processor, to a different value, which causes the s system's operating system to not recognize the lock step pair of processors for which said LOL is detected as the processor; and

changing a system identifier of the spare processor to that of the boot processor, thereby switching the role of boot processor to said spare processor.

14. The method of claim 13 wherein the spare processor is a hot spare that is held in idle until switched the role of boot processor.

15. The method of claim 13 wherein the spare processor is an active spare.

16. The method of claim 15 further comprising:

causing the system's operating system to idle the active spare.

17. The method of claim 13 wherein said switching causes the system's operating system to recognize said spare processor as said boot processor without shutting down said operating system.

18. The method of claim 13 wherein said determining step is performed by system firmware.

19. The method of claim 13 wherein said copying step is performed by system firmware.

20. The method of claim 13 wherein said switching step is performed by system firmware.

21. A system comprising:

a plurality of processor modules that each include a master processor and a slave processor that operate in lockstep;

an operating system;

error detection logic operable to detect loss of lockstep (LOL) for at least one of said processor modules; and

system firmware operable, responsive to the detection of LOL for a first of said processor modules, to determine whether said first processor module is assigned a role of system boot processor;

wherein if determined that said first processor module is assigned the role of system boot processor, said system firmware is further operable to cause said operating system to recognize another processor module as system boot processor without shutting down said operating system.

22. The system of claim 21 wherein the another processor module is a processor module assigned a role of spare processor.

23. The system of claim 21 wherein the another processor module is held in idle until said system firmware causes said operating system to recognize said another processor module as said system boot processor.

24. The system of claim 21 wherein the system firmware causes the operating system to idle the another processor module before causing said operating system to recognize said another processor module as said system boot processor.

25. The system of claim 21 further comprising:

non-volatile data storage storing information that assigns the role of system boot processor to one of said plurality of processor modules.

26. A system comprising:

system firmware means for detecting loss of lockstep (LOL) for a processor in a multi-processor system;

means for determining whether the processor for which said LOL is detected is assigned the role of boot processor; and

system firmware means for switching the role of boot processor to a spare processor without shutting down the system's operating system if determined that the processor for which said LOL is detected is assigned the role of boot processor.

27. The system of claim 26 wherein said processor in said multi-processor system is an Itanium Processor Family (IPF) processor.