Browse Prior Art Database

Boundary Aligner

IP.com Disclosure Number: IPCOM000044539D
Original Publication Date: 1984-Dec-01
Included in the Prior Art Database: 2005-Feb-06
Document File: 2 page(s) / 41K

Publishing Venue

IBM

Related People

Awsienko, O: AUTHOR [+4]

Abstract

A technique is described whereby an algorithm is implemented, through the use of a shifter network, to align data pointed to by the starting addresses, called boundary alignment, in master and slave direct memory access (DMA) architecture. In multiple interconnected computer systems, before a master processor can read or write to a slave processor's memory directly, it must align the data to the slave's starting address so as to protect the integrity of the data. Before data can be transferred to the slave, data located at the starting address of the master's memory is moved to the starting address pointed to by the slave. This technique is called boundary alignment. There are two steps involved in the boundary alignment: 1.

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 55% of the total text.

Page 1 of 2

Boundary Aligner

A technique is described whereby an algorithm is implemented, through the use of a shifter network, to align data pointed to by the starting addresses, called boundary alignment, in master and slave direct memory access (DMA) architecture. In multiple interconnected computer systems, before a master processor can read or write to a slave processor's memory directly, it must align the data to the slave's starting address so as to protect the integrity of the data. Before data can be transferred to the slave, data located at the starting address of the master's memory is moved to the starting address pointed to by the slave. This technique is called boundary alignment. There are two steps involved in the boundary alignment: 1. When data is being transferred for the first time in the packet and there is a mismatch between the starting addresses of the master and slave. 2. Whenever data is being transferred to the slave. Fig. 1 shows the conditions requiring boundary alignment between the master and slave, when the data is being transferred for the first time.

This is determined by the last two bits of the master and slave addresses. If the last two bits of both the master and slave are the same, then no alignment is necessary. Since the last two bits of both the addresses determine the amount of mismatch, the type of shift required, either right or left, is also determined. If it is determined that a right shift is required, as in the case of master...