Browse Prior Art Database

Data Structures for Storing Subaddress Information

IP.com Disclosure Number: IPCOM000114886D
Original Publication Date: 1995-Feb-01
Included in the Prior Art Database: 2005-Mar-30
Document File: 4 page(s) / 209K

Publishing Venue

IBM

Related People

Blackledge, JW: AUTHOR [+4]

Abstract

Disclosed is an architecture providing a method for storing Subaddress information in a conventional Programmable Option Select (POS) data table without requiring additional system resources. The information is arranged in a structure promoting flexibility and ease of use, enhancing the conventional definition of POS data while maintaining compatibility with implementations not using subaddress information.

This text was extracted from an ASCII text file.
This is the abbreviated version, containing approximately 52% of the total text.

Data Structures for Storing Subaddress Information

      Disclosed is an architecture providing a method for storing
Subaddress information in a conventional Programmable Option Select
(POS) data table without requiring additional system resources.  The
information is arranged in a structure promoting flexibility and ease
of use, enhancing the conventional definition of POS data while
maintaining compatibility with implementations not using subaddress
information.

      Table 1 shows the conventional use of the thirty-five byte
blocks in the POS table.  Each block is assigned to an adapter in a
certain slot.  Information pertinent to each adapter is stored in
Non-Volatile Random Access Memory (NVRAM), being loaded by a
configuration program commonly called "Setup," to be written to the
appropriate adapters during Power On Self Test (POST).  The
conventional POS definition includes specifying the adapter with an
ID before specifying information for loading ports 102h through 105h.
The remaining 28 bytes are reserved for expansion.

      The new architecture uses a data byte counter, with a non-zero
value in bit 7 of the POS Data Byte Count indicating the presence of
subaddress data.  A zero value in this bit indicates the presence of
non-subaddress POS data.  Since fewer than 32 bytes are possible, bit
7 provides a flag while maintaining compatibility with earlier
systems.

      When subaddress data is present, a flag byte is used to help
POST interpret the data for adapter initialization, allowing simple
and complex subaddress support.  A "one-to-one" format allows the
initialization of non-contiguous addresses, while a "range
initialization" format provides support for large arrays typically
needed for memory adapters.  Data can be incremented with the
address, or it can remain constant across the entire range.  The data
counter includes all bytes used for subaddressing.  If the count does
not match the data, POST does not initialize the subaddress registers
o...