Method, apparatus, and computer program product for deconfiguring a processor

#1. Field of the Invention

#The invention concerns diagnosing a processor in a computer system having a number of processors with a processing load distributed among the processors, and more particularly concerns deconfiguring the processor responsive to the diagnosis.

#2. Related Art

#FIG. 1 illustrates a computer system, according to an embodiment of the present invention.

#FIG. 2 illustrates a more conventional system, in which firmware does not pass information to an operating system error log.

#FIG. 3 illustrates a more conventional system, in which error checking is performed by special error checking hardware dedicated to error checking in each system processor.

#The claims at the end of this application set out novel features which applicants believe are characteristic of the invention. The invention, a preferred mode of use, further objectives and advantages, will best be understood by reference to the following detailed description of an illustrative embodiment read in conjunction with the accompanying drawings.

#Referring now to FIG. 1, a computer system 100 is illustrated, according to an embodiment of the invention. In the system 100, a processing load is distributed among a number of system processors 110. Test computations are performed at intervals during run time by floating point logic 112 of the processors 110, responsive to stored test instructions. The floating point logic 112 is the same hardware used in normal computing operations, and not specially designed logic dedicated to error checking.

#Instructions for execution by the processors 110 include firmware instructions 120, operating system instructions 130 that run with the support of the firmware 120, and application instructions 140 that run with the support of the firmware 120 and operating system 130. It should be appreciated that the firmware 120 runs on the processors 110 substantially independently of the operating system 130, and provides a lower level of interface and therefore greater access to hardware components such as processors 110 than does the operating system 130. The system 100 also has a service processor 115 that is separate from the system processors 110 used to run the operating system 130. The service processor 115 runs specialized firmware code 120 that includes code for portions of an initial program load process, including portions that control component testing. (The term “initial program load” refers to the process of taking the system 100 from a powered off, or non-running state, to the point of operating system 130 specific code. Initial program load may include running various tests. In a multiprocessor system all functioning processors 110 ordinarily go through the initial program load process. This is sometimes referred to as the “boot” process.) The service processor 115 has controlling access to the hardware of the system 100, including the ability to start and stop system processors and read fault isolation registers in the various components. The service processor 115 is also be available to help diagnose system problems that occur during run time.