Error collection coordination for software-readable and non-software readable fault isolation registers in a computer system

• US 6,119,246 A
• Filed: 03/31/1997
• Issued: 09/12/2000
• Est. Priority Date: 03/31/1997
• Status: Expired due to Fees

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Kitamorn, Alongkorn, McLaughlin, Charles Andrew, Nulu, Sayileela
• Primary Examiner(s): Sheikh, Ayaz R.
• Assistant Examiner(s): JEAN, FRANTZ B
• Application Number:
  US08/829,018
• Time in Patent Office: 1,261 Days
• Field of Search: 395/182.18, 395/182.13, 395/183.1, 395/183.03, 364/265, 364/267, 364/268, 364/268.9, 364/280, 714/25, 714/27, 714/30, 714/34, 714/38, 714/42, 714/54, 714/15, 714/20
• US Class Current: 714/27
• CPC Class Codes: G06F 11/0775 Content or structure details of the error report, e.g. specific table structure, specific error fields
• CPC Class Codes: G06F 11/2736 using a dedicated service processor for test

Abstract

The present invention provides method and system aspects for performing error data gathering from fault isolation registers of a computer system following a machine check occurrence. A method aspect includes utilizing firmware to perform failure information retrieval in software accessible registers and initiating a service processor (SP) for failure data retrieval in non-software accessible registers. The method further includes coordinating the combination of the failure information retrieved and the failure data retrieved in an error log for use in isolation of a fault source in the computer system.

Claims

1. A method for coordinating failure information retrieval from non-software accessible registers and software accessible registers of a computer system, the method comprising:

performing failure information retrieval with the software accessible registers when a non-maskable interrupt occurs;

determining whether failure information retrieval utilizing a service processor (SP) is necessary;

performing failure information retrieval from non-software accessible registers by utilizing the SP by initiating the SP for register scan through receiving an attention signal and determining if a semaphore indicator has been set; and

forming an error log with error data retrieved during the failure information retrieval steps.

2. The method of claim 1 further comprising gathering failure information from the non-software accessible registers and storing the failure information in a predefined storage area.

3. The method of claim 2 further comprising indicating completion of the gathering.

4. The method of claim 3 wherein indicating completion further comprises releasing a semaphore indicator.

5. The method of claim 3 wherein indicating completion further comprises issuing a predefined return code.

6. The method of claim 3 wherein the predefined storage area comprises a partition of a nonvolatile random access memory (NVRAM).

7. A method for performing error data gathering from fault isolation registers of a computer system following a machine check occurrence, the method comprising:

utilizing firmware to perform failure information retrieval in software accessible registers;

initiating a service processor (SP) for failure data retrieval in non-software accessible registers by receiving an attention signal and determining if a semaphore indicator has been set; and

coordinating the combination of the failure information retrieved and the failure data retrieved in an error log for use in isolation of a fault source in the computer system.

8. The method of claim 7 wherein coordinating the combination further comprising utilizing a storage area commonly accessible to the firmware and the SP for the error log.

9. The method of claim 8 wherein utilizing further comprises utilizing a predefined partition of a memory device for forming the error log with a desired data structure.

10. The method of claim 9 wherein the memory device comprises a nonvolatile random access memory (NVRAM).

11. A computer system capable of identifying a source of a machine check condition from data in failure isolation registers, the computer system comprising:

processing means;

firmware means supported by the processing means for performing failure information retrieval from software-readable registers;

service processor (SP) means coupled to the processing means, the SP means performing failure information retrieval for non-software readable registers and being initiated for register scan by receiving an attention signal and determining if a semaphore indicator has been set; and

memory means coupled to the SP means and the processing means for storing failure information retrieved by the firmware means and the SP means in a predefined error log data structure.

12. The computer system of claim 11 wherein the processing means further supports an operating system, the operating system isolating a fault condition from the predefined error log data structure.

13. The computer system of claim 11 wherein the memory means comprises a nonvolatile random access memory (NVRAM).

14. The computer system of claim 11 wherein the firmware means initiates the SP means for failure information retrieval.