RAM Parity Error Handling During Memory Test Without Restart

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Conventional methods for handling RAM parity errors in computer systems often result in system restarts or halts, causing disruptions in service, and there is a need for a more efficient way to manage these errors without such interruptions.

Innovation Solution

A method and system that utilize a lookup table with entries containing data field locations, sizes, and parity error handler types to identify and correct faulty bytes, employing various error handling strategies like overwriting with default values, tied variables, or refreshing bytes, without triggering system restarts or halts.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If conventional RAM parity error interrupt handler is used, then parity errors are detected, but system operation is disrupted due to restarts or halts

Engineering Contradiction:
Improveparity error detectionVSAvoidsystem operation continuity
Core Design Contradiction:
ReliabilityVSProductivity

Solution Approach 1:

The patent extracts the error handling logic from the critical system path by implementing an asynchronous error handler that processes parity errors separately from the main execution flow. The faulty byte address is captured and processed in the background using a byte swap mechanism, allowing the main system to continue operating without interruption.

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

The patent performs preliminary error capture by storing the faulty byte address in a dedicated register immediately when a parity error is detected. This preliminary action allows the error to be processed asynchronously later, preventing disruption to ongoing system operations while ensuring the error is not lost.

Inventive Principle:
Principle #10Preliminary action

2Ease of manufacture

If system restart or halt is implemented for parity errors, then error handling is simple, but service disruption occurs

Engineering Contradiction:
Improveerror handling implementationVSAvoidservice disruption
Core Design Contradiction:
Ease of manufactureVSObject-affected harmful factors

Solution Approach 1:

The patent introduces an intermediary byte swap mechanism that translates parity errors into correctable data states. Instead of directly restarting the system, the faulty byte address is used to swap bytes in a buffer, effectively correcting the error through an intermediate processing step that maintains system continuity.

Inventive Principle:
Principle #24Intermediary (Mediator)

3Measurement precision

If lookup table with error handler types is implemented, then error handling precision is improved, but device complexity increases

Engineering Contradiction:
Improveerror handling precisionVSAvoidlookup table structure
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent segments the error handling approach by creating a lookup table that divides memory into different data field types (word, double word, floating point, etc.). Each segment has predefined handling characteristics, allowing precise error recovery tailored to the specific data type while maintaining a structured and manageable implementation.

Inventive Principle:
Principle #1Segmentation

Data Source

PatentEP4425497B1Ram test with parity error detection and handling
Publication Date: 2025.12.31 HAMILTON SUNDSTRAND CORP
  • EP4425497B1 patent drawingFigure 1
  • EP4425497B1 patent drawingFigure 2
  • EP4425497B1 patent drawingFigure 3

AI summary

A computer system operates a method of detecting and handling a RAM parity error in the context of a RAM test. The computer system includes a memory storing therein a lookup table having a plurality of entries, each entry including a data location and a RAM parity error handler type. A processor obtains an address of a faulty byte detected during a RAM test on the parity RAM, searches the lookup table for an entry that contains the faulty byte based on the address of the faulty byte, obtains the RAM parity error handler type from the entry, and fixes a data field impacted by the faulty byte based on the RAM parity error handler type.