Browse Prior Art Database

Processing of Overlapping Operands

IP.com Disclosure Number: IPCOM000092556D
Original Publication Date: 1966-Dec-01
Included in the Prior Art Database: 2005-Mar-05
Document File: 2 page(s) / 43K

Publishing Venue

IBM

Related People

Adler, JG: AUTHOR [+4]

Abstract

Variable field length operands are usually processed one byte, i. e., eight bits, at a time. To save storage accessing time and for convenience, operand data is fetched and stored in eight byte words. This can cause error, if some part of a word being fetched comprises byte positions for which the data has been processed but which data is buffered in a word that has not been put into storage at this time.

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 53% of the total text.

Page 1 of 2

Processing of Overlapping Operands

Variable field length operands are usually processed one byte, i. e., eight bits, at a time. To save storage accessing time and for convenience, operand data is fetched and stored in eight byte words. This can cause error, if some part of a word being fetched comprises byte positions for which the data has been processed but which data is buffered in a word that has not been put into storage at this time.

In the drawing, an eight-byte source register 1 receives the data comprising a first source operand over a bus 2 from main storage. Sink register SK 3 receives the data comprising a second sink operand over bus 2. Byte address counter BC 4 for register 1 controls the outgating of the byte of register 1 which is identified by the setting of its BC 4 from register 1 to bus 5. A similar byte address counter BC 6 controls the outgating of a byte from register SK3 to bus 7.

The two bytes on busses 5 and 7 are combined in VFL execution unit 8 with the result set onto bus 9. BC 6 ingates the result on bus 9 back into the same byte of SK3 from which the processed byte was outgated. When the source word in register 1 has been utilized, a new source word is supplied over bus 2, but when the last byte of the sink operand has been processed, the resultant sink word is sent over bus 10 to main store before the new sink word is gated into SK3.

Due to the delay between processing of a sink byte and its storage in main store, if the byte address of a source byte is less than eig...