Method and apparatus for disabling defective components in a computer system

• US 7,137,020 B2
• Filed: 04/14/2003
• Issued: 11/14/2006
• Est. Priority Date: 05/17/2002
• Status: Active Grant

• Current Assignee: Oracle America, Inc. (Oracle Corporation)
• Original Assignee: Sun Microsystems Incorporated (Oracle Corporation)
• Inventors: Gilstrap, Raymond J., Gang, Fongyan, Gordon, David S., Hain, Daniel J.
• Primary Examiner(s): Trujillo, James K.
• Application Number:
  US10/412,998
• Publication Number:
  US 20030217256A1
• Time in Patent Office: 1,310 Days
• Field of Search: None
• US Class Current: 713/324
• CPC Class Codes: G06F 11/2035 without idle spare hardware
• CPC Class Codes: G06F 11/2043 where the redundant components share a common memory address space

Abstract

A method for configuring a computer system includes identifying a defective component in the computer system; disabling at least the defective component; and dynamically reconfiguring the computer system to allow continued operation with the defective component disabled. A computer system includes a plurality of components and a system controller. The system controller is adapted to identify a defective component from the plurality of components, disable at least the defective component, and dynamically reconfigure the computer system to allow continued operation with the defective component disabled.

Claims

1. A method for configuring a computer system, wherein the computer system includes a field replaceable unit comprising a plurality of components including a defective component, and wherein the field replaceable unit comprises a first field replaceable unit, and the defective component comprises a second field replaceable unit having an associated field replaceable unit identification memory configured to store error information associated with the second field replaceable unit, the method comprising:

identifying the defective component in the computer system;

disabling at least the defective component;

dynamically reconfiguring the computer system to allow continued operation with the defective component disabled;

allowing continued operation of at least a portion of the plurality of components of the field replaceable unit other than the defective component;

generating a component map of the components in the first field replaceable unit, the component map including enable information regarding the components;

accessing the component map to disable a subset of components associated with the defective component; and

accessing the error information in the associated field replaceable unit identification memory to set a defective error bit to disable the second field replaceable unit.

2. The method of claim 1, wherein the plurality of components comprises at least a processor and an associated memory device, and the method further comprises:

disabling the processor responsive to the memory device being the defective component; and

disabling the memory device responsive to the processor being the defective component.

3. The method of claim 1, wherein the plurality of components comprises a processor and a plurality of associated memory devices, the defective component comprising a defective memory device from the plurality of associated memory devices, and the method further comprises:

disabling the defective memory device; and

enabling the processor and the memory devices in the plurality of associated memory devices other than the defective memory device.

4. The method of claim 3, wherein the memory devices are arranged in banks, and the method further comprises disabling one of the plurality of associated memory devices in a common bank with the defective memory device.

5. A method for configuring a computer system that includes a field replaceable unit comprising a plurality of components including a defective component, and wherein the plurality of components comprises a processor and a plurality of associated memory devices, and the defective component comprises a defective memory device from the plurality of associated memory devices, and wherein the field replaceable unit comprises a first field replaceable unit, and the defective memory device comprises a second field replaceable unit having an associated field replaceable unit identification memory configured to store error information associated with the second field replaceable unit, the method comprising:

identifying the defective component in the computer system;

disabling the defective memory device;

dynamically reconfiguring the computer system to allow continued operation with the defective component disabled;

allowing continued operation of at least a portion of the plurality of components of the field replaceable unit other than the defective component;

enabling the processor and the memory devices in the plurality of associated memory devices other than the defective memory device;

generating a component map of the components in the first field replaceable unit, the component map including enable information regarding the components;

accessing the component map to disable one of the plurality of associated memory devices in a common bank with the defective memory device; and

accessing the error information in the associated field replaceable unit identification memory to set a defective error bit to disable the defective memory device.

6. A computer system, comprising:

a field replaceable unit comprising a plurality of components; and

a system controller adapted to identify a defective component from the plurality of components and to dynamically reconfigure the computer system to allow continued operation with the defective component disabled, and wherein the system controller is further configured to allow continued operation of at least a portion of the plurality of components of the field replaceable unit other than the defective component, wherein the field replaceable unit comprises a first field replaceable unit, and the defective component comprises a second field replaceable unit having an associated field replaceable unit identification memory configured to store error information associated with the second field replaceable unit, and the system controller is further configured to generate a component map including enable information regarding the components in the first field replaceable unit, to access the component map to disable a subset of the plurality of components associated with the defective component, to and access the error information in the associated field replaceable unit identification memory to set a defective error bit to disable the second field replaceable unit.

7. The system of claim 6, wherein the plurality of components comprises at least a processor and an associated memory device, and the system controller is further configured to disable the processor responsive to the memory device being the defective component and disable the memory device responsive to the processor being the defective component.

8. The system of claim 6, wherein the plurality of components comprises a processor and a plurality of associated memory devices, the defective component comprising a defective memory device from the plurality of associated memory devices, and the system controller is further configured to disable the defective memory device and enable the processor and the memory devices in the plurality of associated memory devices other than the defective memory device.

9. The system of claim 8, wherein the memory devices are arranged in banks, and the system controller is further configured to disable one of the plurality of associated memory devices in a common bank with the defective memory device.

10. The system of claim 8, wherein the field replaceable unit comprises a third field replaceable unit and the defective memory device comprises a fourth field replaceable unit having an associated field replaceable unit identification memory configured to store error information associated with the fourth field replaceable unit, and the system controller is further configured to generate a component map of the components in the third field replaceable unit, the component map including enable information regarding the plurality of components, access the component map to disable the one of the plurality of associated memory devices in a common bank with the defective memory device, and access the error information in the associated field replaceable unit identification memory to set a defective error bit to disable the defective memory device.