Browse Prior Art Database

Extending Video Memory Addressing

IP.com Disclosure Number: IPCOM000036249D
Original Publication Date: 1989-Sep-01
Included in the Prior Art Database: 2005-Jan-28
Document File: 6 page(s) / 154K

Publishing Venue

IBM

Related People

Bealkowski, R: AUTHOR [+2]

Abstract

A technique is described whereby the addressing architecture of video memories, as used in personal systems with standard video graphics array (VGA) capabilities, is extended so as to allow programs, or devices, to be able to address memory as a single block, with each byte having a unique address. By enabling the addressing to occur with each byte at unique addresses, block transfer and implicit access method support are provided for map-oriented memories.

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 42% of the total text.

Page 1 of 6

Extending Video Memory Addressing

A technique is described whereby the addressing architecture of video memories, as used in personal systems with standard video graphics array (VGA) capabilities, is extended so as to allow programs, or devices, to be able to address memory as a single block, with each byte having a unique address. By enabling the addressing to occur with each byte at unique addresses, block transfer and implicit access method support are provided for map-oriented memories.

Typically, VGA memories are organized as distinct memory maps to correspond to certain video modes. This organization can preclude the ability for the completion of block transfers of video memory in one operation. It has been determined that alternate bus masters, such as direct memory access (DMA) or coprocessors, should be able to access the video memory as a single block. Also, the software should be able to have the ability to vary the mechanism by which it accesses the video. As a result, the concept described herein provides the ability to extend the existing map structure so as to enable bus masters to perform a block transfer to all video memory devices in a single operation.

(Image Omitted)

Video modes provided by the VGA generally control the access to the video memory and support a section of memory, typically 256 KB, which is used to store the data necessary to produce an image on the display. The organization of the memory presented to a master, such as a host central processing unit, a DMA subsystem, or a coprocessor, will vary depending on the specific video mode employed. Memory access for various video modes for a 256 KB memory is generally organized as four maps of 64 KB each, as shown in Fig. 1. This type of architecture limits the addressing capabilities, since there is not a unique address provided for each byte of video memory. Individual maps must be selected through control registers in the VGA.

For software manipulation of the video memory, accessing each map individually through control ports in the VGA is cumbersome and impossible for bus masters to perform block transfers to all of the video memory units in a single operation. Therefore, the concept provides a means of enabling programs, or devices, to address the video memory as

(Image Omitted)

a single block, with each byte having a unique address. This extension is significant for applications, such as graphics, and where master devices require block transfer capabilities to and from the video memory.

The concept provides three types of addressing mode extensions for access to the video memory, as follows:

1) CONSECUTIVE - This addressing mode allows the memory to be accessed with each byte of video memory residing at a unique address when viewed from the system bus. In the CONSECUTIVE address mode, each

1

Page 2 of 6

successive map is assigned addresses after the preceding map. The maps will appear to be aligned one after the other. An example of the CONSECUTIVE...