Browse Prior Art Database

CHAINED EVENTS VECTOR

IP.com Disclosure Number: IPCOM000006757D
Original Publication Date: 1993-Mar-01
Included in the Prior Art Database: 2002-Jan-30
Document File: 2 page(s) / 82K

Publishing Venue

Motorola

Related People

Richard Soja: AUTHOR

Abstract

Current microprocessor based systems use inter- rupt vectors which contain only one address-that of the start address of an interrupt handler to process the inter- rupt event. An application which uses different vectors to cause execution of the same group of software rou- tines in a variety of orders requires a software overhead, usually in the form of a jump-k-subroutine table, to organ- ize the order of execution.

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 50% of the total text.

Page 1 of 2

MO-LA INC. Technical Developments Volume 18 March 1993

CHAINED EVENTS VECTOR

by Richard Soja

   Current microprocessor based systems use inter- rupt vectors which contain only one address-that of the start address of an interrupt handler to process the inter- rupt event. An application which uses different vectors to cause execution of the same group of software rou- tines in a variety of orders requires a software overhead, usually in the form of a jump-k-subroutine table, to organ- ize the order of execution.

  The proposed 'Chained Events Vector' is a hard- ware implementation which overcomes the so&ware over- head required to order the execution of different subroutines.

  The single start address contained in the vector is replaced by a number of addresses which are used in sequence under hardware control. This eliminates the time and space overhead associated with the storage, decoding and execution of microprocessor instructions.

Each address may be stored in a number of ways:
1. As an absolute address for a software routine.
2. As an offset which is added to a base address, to give the absolute start address ofthe software routine.
3. As an offset into a table of addresses which contains the absolute start address ofthe software routine.

  The advantage of using an absolute address is that the time taken to update the program counter value is reduced.

  The advantage of using offsets is that the number of bits allocated to each offset value may be reduced in size without compromising the flexibility. In item 2 the

base address, and in item 3 the base address of the table of addresses, may be determined in a variety of ways. It maybe-
a. fixed in ha...