Browse Prior Art Database

Dynamic Reconfiguring of Personal Computer Systems After Accidental Exchange of Adapter Position in the Micro Channel

IP.com Disclosure Number: IPCOM000105536D
Original Publication Date: 1993-Aug-01
Included in the Prior Art Database: 2005-Mar-20
Document File: 2 page(s) / 103K

Publishing Venue

IBM

Related People

Dayan, RA: AUTHOR [+2]

Abstract

Described is a technique to avoid unnecessary running of set configuration utility.

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

Dynamic Reconfiguring of Personal Computer Systems After Accidental Exchange of Adapter Position in the Micro Channel

      Described is a technique to avoid unnecessary running of set
configuration utility.

Once the set configuration utility has the final configuration data,
it is stored in a non-volatile storage, referred to as NVRAM, for use
by Power On Self Test (POST).  Programmable Option Select (POS)
information is saved in NVRAM, with one entry provided per Micro
Channel* (MC) slot.  Set configuration utility will then force the
user to run through POST either by exiting the set configuration menu
or by the keystroke sequence Alt-Ctrl-Del.  POST will reset the MC
through the I/O space of the microprocessor.  POST will then retrieve
the encoded POS data from NVRAM, and deliver this data to both the
planar and adapters.  The adapters and planar are now considered
configured.

      The POS data and the adapter ID for each MC slot in system is
stored in the NVRAM of the system.  Each MC slot is associated with a
unique entry in NVRAM.  When the system is powered on, POST compares
the adapter ID read from the MC slot, slot_adapter_ID, to the adapter
ID stored in the NVRAM entry corresponding to the particular slot,
NVRAM_adapter_ID.  If a comparison fails, currently POST will display
an error code and require the user to use the reference diskette to
reconfigure the system.

      In the described technique, it will be necessary to first
define some terms.  There are 'n' MC slots in a system.  The current
MC slot being configured will be defined as the 'i'th slot.  The
'j'th slot is any other slot where 'i' is not equal to 'j' and 'j' is
greater than 'i'.

      Accordingly, POST will use the following algorithm when
detecting a configuration error, instead of immediately flagging an
error and requiring reconfiguration.  Upon encountering a mismatch of
adapter IDs (during NVRAM/slot comparison), POST will set the
indicator, Config_changed_flag, and will continue its search for the
current slot_adapter_ID in the NVRAM entries, NVRAM_adapter_ID,
corresponding to the remaining slots.

      For example, in slot 'i', the slot_adapter_ID read is 'XXXX'.
The NVRAM_adapter_ID in the slot 'i'th entry is 'AAAA'.  Starting at
the 'i'+1th entry in NVRAM, a NVRAM_adapter_ID of 'XXXX' is checked
for.  If the adapter ID 'XXXX' is found in NVRAM, then POST will read
the current resident slot_adapter_ID to the corresponding MC slot.
This will be referred to as the 'j'th...