Browse Prior Art Database

High Performance Functionally Integrated Memory Address Register for a Small Processor

IP.com Disclosure Number: IPCOM000085900D
Original Publication Date: 1976-Jun-01
Included in the Prior Art Database: 2005-Mar-03
Document File: 2 page(s) / 42K

Publishing Venue

IBM

Related People

Eggebrecht, LC: AUTHOR

Abstract

The memory address circuit of the figure improves upon the basic technique of buffering an address register output, such that a new address can be incremented while the old address is made stable and applied to the memory for a memory access.

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 65% of the total text.

Page 1 of 2

High Performance Functionally Integrated Memory Address Register for a Small Processor

The memory address circuit of the figure improves upon the basic technique of buffering an address register output, such that a new address can be incremented while the old address is made stable and applied to the memory for a memory access.

In this circuit the function of a ripple counter is provided by a full look-ahead adder 11 and a memory array 12. The use of adder 11 results in a performance improvement, since the adder will perform a plus one add faster than a ripple carry counter can increment. The array 12 simply stores the new incremented address until it is clocked into the memory address register (MAR) output buffer
13.

By using an array 12 at this point an automatic interrupt address handling capability is available. The array 12 can be used to store interrupt addresses and save return addresses by simply controlling the array address lines. Since the array 12 is also used for next address generation, it is always current.

By adding data bus selectors 15 and 16 to the inputs of the adder 11, a wide range of effective memory address generation is available. This allows adder 11 to be used both for address incrementing and effective address generation.

Using the circuitry of the figure the following forms of effective address generation are available: MAR + MAR + 1 Increment to next address. MAR = Data Bus Out (DBO) Direct branch addressing. MAR = DBO + MAR Relative...