Browse Prior Art Database

Circumvention Technique for Correcting Errors in Read Only Storage

IP.com Disclosure Number: IPCOM000074516D
Original Publication Date: 1971-May-01
Included in the Prior Art Database: 2005-Feb-23
Document File: 2 page(s) / 36K

Publishing Venue

IBM

Related People

Kushner, ML: AUTHOR [+2]

Abstract

This system provides a solution which allows some instructions to be executed from the R/W storage 10 to correct for errors in Read-Only Storage (ROS) 11, where ROS 11 is a storage device where instructions and constants are located therein at the time of manufacture.

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 54% of the total text.

Page 1 of 2

Circumvention Technique for Correcting Errors in Read Only Storage

This system provides a solution which allows some instructions to be executed from the R/W storage 10 to correct for errors in Read-Only Storage (ROS) 11, where ROS 11 is a storage device where instructions and constants are located therein at the time of manufacture.

For example, consider a storage system with 2/16/ words each of 16 bits, where the first half of storage is ROS and the remaining storage is read/write (R/W). A single storage address register (SAR) 12 is used to address the two storage devices 10 and 11 and a single storage data register (SDR) 13 receives the output thereof.

A central processor unit (CPU) 14 includes Address Generators 15 and 16 which generate instruction addresses (IA) and operand addresses (OA), respectively, to the storage system. Also included in CPU 14 are instruction buffer (IB) 17 and operand buffer (OB) 18 which receive storage system instruction and data, respectively, from the storage system.

The storage is considered to be partitioned into 256 word sectors. For the n/th/ sector S(n)(there are 256 sectors) two words, A(n) & B(n), are stored in unique locations in the R/W storage. The address of A(n)is:

(Image Omitted)

A(n) is a 16-bit number which contains the sector identifier in bits 0 - 7 nand the address within the sector where an error has occurred, in bits 8 - 15. If no error has occurred in sector n, A(n) is equal to zero. B(n)is the address of the first word of a correction program in R/W storage 10 during initial program load.

The sequence of ope...