Synchronize error handling for a plurality of partitions
View Patent ↗In accordance with at least some embodiments, a system comprises a plurality of partitions, each partition having its own error handler. The system further comprises a plurality of resources assignable to the plurality of partitions. The system further comprises management logic coupled to the plurality of partitions and the plurality of resources. The management logic comprises an error management tool that synchronizes operation of the error handlers in response to an error.
1. A system, comprising:
a plurality of partitions, each partition having its own error handler;
a plurality of resources assignable to the plurality of partitions; and
management logic accessible to the plurality of partitions and the plurality of resources,
wherein the management logic comprises an error management tool that synchronizes operation of the error handlers in response to an error, synchronization comprising re-enabling, based on a combination error log that logs errors for each partition, recognition of viral errors for a partition only after each partition is cleaned.
2. The system of claim 1 wherein the error occurs in a resource shared by a plurality of the partitions.
3. The system of claim 1 wherein each error handler comprises a machine-check abort (MCA) handler.
4. The system of claim 1 wherein the plurality of resources comprises an input/output hub (IOH) with a plurality of root complexes assignable to different partitions.
5. The system of claim 4 wherein the management logic synchronizes operation of the error handlers in response to an IOH error.
6. The system of claim 5 wherein the error management tool causes each error handler to perform error logging and cleanup and then to wait for IOH cleanup and recovery.
7. The system of claim 1 wherein the error management tool comprises an error logging services tool that coordinates generating an error log for all partitions when there is an error detected for any of the partitions.
8. The system of claim 7 wherein the error management tool provides the error log to all the partitions.
9. A computer-readable storage medium comprising partition error synchronization software that, when executed, causes a processor to:
detect an error in a resource shared by a plurality of partitions; and
synchronize operation of error handlers for each of the plurality of partitions in response to the error, synchronization comprising re-enabling, based on a combination error log that logs errors for each partition, recognition of viral errors for a partition only after each partition is cleaned.
10. The computer-readable storage medium of claim 9 wherein the partition error synchronization software further causes the processor to synchronize operation of machine-check abort (MCA) handlers for each of the plurality of partitions in response to the error.
11. The computer-readable storage medium of claim 9 wherein the partition error synchronization software further causes the processor to detect an error in an input/output hub (IOH) with a plurality of root complexes assignable to different partitions.
12. The computer-readable storage medium of claim 11 wherein the partition error synchronization software further causes the processor to direct each error handler to perform error logging and cleanup and then to wait for IOH cleanup and recovery.
13. The computer-readable storage medium of claim 9 wherein the partition error synchronization software further causes the processor to coordinate generating an error log for all of said partitions when there is an error detected for any of said partitions.
14. The computer-readable storage medium of claim 13 wherein the partition error synchronization software further causes the processor to provide the error log to all of said partitions.
15. A method, comprising:
detecting, by a processor, an error in a resource shared by a plurality of partitions; and
synchronizing error handling, by the processor, across the plurality of partitions in response to the error, synchronization comprising re-enabling, based on a combination error log that logs errors for each partition, recognition of viral errors for a partition only after each partition is cleaned.
16. The method of claim 15 wherein said synchronizing error handling across the plurality of partitions comprises synchronizing operation of a machine-check abort (MCA) handler for each of said partitions.
17. The method of claim 15 wherein said detecting an error in a resource shared by a plurality of partitions comprises detecting an error in an input/output hub (IOH) with a plurality of root complexes assignable to different partitions.
18. The method of claim 17 wherein said synchronizing error handling across the plurality of partitions comprises directing each partition to perform error logging and cleanup and then to wait for IOH cleanup and recovery.
19. The method of claim 15 further comprising coordinating generation of an error log for all of said partitions when there is an error detected for any of said partitions.
20. The method of claim 19 further comprising providing the error log to all of said partitions.