Dismiss
InnovationQ will be updated on Sunday, Oct. 22, from 10am ET - noon. You may experience brief service interruptions during that time.
Browse Prior Art Database

Storage Path Check Stop Recovery

IP.com Disclosure Number: IPCOM000105505D
Original Publication Date: 1993-Aug-01
Included in the Prior Art Database: 2005-Mar-20
Document File: 2 page(s) / 28K

Publishing Venue

IBM

Related People

Beardsley, BC: AUTHOR [+6]

Abstract

A method for recovering from catastrophic faults in one of many tightly coupled microprocessors that share some common memory is disclosed. The microprocessors are able to indicate to each other that one of them has suffered a catastrophic fault and has not yet recovered. The other (non-faulted) microprocessors use this information to recover the integrity of the system with no adverse affects visible to the user of the system.

This text was extracted from an ASCII text file.
This is the abbreviated version, containing approximately 100% of the total text.

Storage Path Check Stop Recovery

      A method for recovering from catastrophic faults in one of many
tightly coupled microprocessors that share some common memory is
disclosed.  The microprocessors are able to indicate to each other
that one of them has suffered a catastrophic fault and has not yet
recovered.  The other (non-faulted) microprocessors use this
information to recover the integrity of the system with no adverse
affects visible to the user of the system.

      A non-faulted microprocessor is able to force another
microprocessor into its process for recovering from any catastrophic
faults.

      A hardware Status register has a bit for each microprocessor.
This bit indicates that the corresponding microprocessor has
encountered and not yet recovered from a catastrophic fault.  There
is another Control register that:

1.  Allows recovering microprocessors to reset the Status register
    (described above) so that the other (non-faulted) microprocessors
    know that its recovery process is successful.

2.  Allows a microprocessor to force another microprocessor into a
    catastrophic fault recovery process.