Browse Prior Art Database

Extended Association of Error Codes

IP.com Disclosure Number: IPCOM000116778D
Original Publication Date: 1995-Nov-01
Included in the Prior Art Database: 2005-Mar-31
Document File: 4 page(s) / 94K

Publishing Venue

IBM

Related People

Basham, R: AUTHOR [+3]

Abstract

As a microcode product evolves and ages, it is often difficult to maintain control over loosely associated data which drives related functions. As an example, error codes are typically generated to identify a unique fault in a machine. Associated with this unique fault may be error recovery action and one or more failing components for service action. Once detected, errors must be incorporated into sense data and reported to the host operating system. When these instances of associated data are maintained independently, it is not always trivial to understand how an error definition or redefinition will affect these other algorithms. Inferences made against dynamic sequences of errors can also be a problem.

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

Extended Association of Error Codes

      As a microcode product evolves and ages, it is often difficult
to maintain control over loosely associated data which drives related
functions.  As an example, error codes are typically generated to
identify a unique fault in a machine.  Associated with this unique
fault may be error recovery action and one or more failing components
for service action.  Once detected, errors must be incorporated into
sense data and reported to the host operating system.  When these
instances of associated data are maintained independently, it is not
always trivial to understand how an error definition or redefinition
will affect these other algorithms.  Inferences made against dynamic
sequences of errors can also be a problem.

      This invention discloses a design approach that requires a
formal, close association of all data required to define the extended
relationships that an error code maintains within a tape susbystem.
Several conventions were defined for error codes:
  1.  Error codes are defined as a set of encoded hexidecimal
constants
       which are reported in host sense data following an exception
       condition.
  2.  Error codes point to very specific failures as opposed to a
wide
       spectrum of error conditions.  This is essential since error
       codes are used to drive error recovery, host error recovery
       indicators, and failing component (FRU) information.
  3.  Error codes are nominally considered to describe a unique
       failure.  In most cases, an error code description will be
posted
       by only one microcode object or module.  Exceptions to this
rule
       cover those conditions describing a general fault that could
       occur anywhere in the system (e.g., ADDRESS_OUT_OF_RANGE).
       Naming uniqueness is driven by appending the name of the
       detecting code module or name.  Numerical uniqueness is
achieved
       by varying the hex error code; as the number of objects which
       generate the same error description increase, so do the
       associated error codes.

          Error code nomenclature includes a unique prefix of the
    form FSCssnnnnnn, where ss is an abbreviation of the detecting
    microcode subsystem and nnnnnn is the name of the microcode
 ...