Browse Prior Art Database

Method of Saving Error Data

IP.com Disclosure Number: IPCOM000038888D
Original Publication Date: 1987-Mar-01
Included in the Prior Art Database: 2005-Feb-01
Document File: 3 page(s) / 74K

Publishing Venue

IBM

Related People

Davis, GW: AUTHOR [+3]

Abstract

Extended error data can be saved for analysis in digital equipment incorporating a microprocessor and diskette by recording the data on the diskette when a high level interrupt occurs, such as interrupt level O. An Extended Errorlog can be achieved by a sequence of identifying error type and necessary recovery; recording a "snapshot" of the present state of hardware and microcode data in microprocessor memory; suspending awaiting jobs; and initiating commands to hardware and lower interrupt levels for an orderly recovery and diskette recording of the error data. An arrangement of microcode architecture for automatically saving extended error data is described with reference to the figure. The microcode units operate as follows: A.

This text was extracted from a PDF file.
At least one non-text object (such as an image or picture) has been suppressed.
This is the abbreviated version, containing approximately 60% of the total text.

Page 1 of 3

Method of Saving Error Data

Extended error data can be saved for analysis in digital equipment incorporating a microprocessor and diskette by recording the data on the diskette when a high level interrupt occurs, such as interrupt level O. An Extended Errorlog can be achieved by a sequence of identifying error type and necessary recovery; recording a "snapshot" of the present state of hardware and microcode data in microprocessor memory; suspending awaiting jobs; and initiating commands to hardware and lower interrupt levels for an orderly recovery and diskette recording of the error data. An arrangement of microcode architecture for automatically saving extended error data is described with reference to the figure. The microcode units operate as follows: A. Machine Check/Program Check (MC/PC) Code 10

performs the above functions, displays a

descriptive error message, and performs a "soft

reset" to clear the cause of the interrupt.

B. Slim File Channel Adapter Code (CAC) 11 does the

actual writing of the Errorlog Data on the

diskette on interrupt level 5.

C. Supervisor 12 runs at interrupt level 6 and

maintains communication among functional tasks

designated 13a - 13h, each having priority level

7, and all other interrupt levels. Supervisor 12

selects the task in control on interrupt 7, based

on job queue and relative priority that decreases

in descending order from Status Task 13a to

Initial Power Load Task 13h.

D. Status Task 13a is the only one that communicates

with MC/PC code and all decisions regarding error

severity, messages, recovery and Extended Errorlog

Data are determined here.

E. Slim File Control Task 13e communicates through

Supervisor 12 with Slim File CAC 11 for diskette

reading and writing. In operation, interrupt level O is assumed to represent a need for immediate attention and MC/PC Code performs some of the above functions and does others through Status Task 13a. An example of a sequence is as follows: 1. The cause of the interrupt is identified and

reset, if possible, to permit subsequent

interrupts to be detected.

2. Data representing the current state of the

hardware is duplicated in the temporary memory

buffer for the Extended Errorlog Data and includes

data traces preceding the interrupt.

3. Interrupt levels 1 - 4 are disabled through a...