Configuration Memory Error Logging for FPGA Soft Error Analysis

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

The miniaturization of semiconductor devices leads to issues with soft errors caused by terrestrial radiation, which are difficult to identify and manage, especially in programmable devices like FPGAs, as they result in malfunctions and data loss, and existing techniques struggle to store and reproduce error information effectively.

Innovation Solution

A programmable device with a configuration memory that includes means for direct reading and writing, error detection, and non-volatile storage of error data, allowing for the identification and retention of error locations and times, even during power off, to enhance reproducibility and reliability.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If configuration memory is used to store logic circuit information in programmable devices, then device functionality and adaptability are improved, but vulnerability to soft errors from terrestrial radiation increases

Engineering Contradiction:
Improveprogrammable device functionalityVSAvoidresistance to soft errors
Core Design Contradiction:
Adaptability or versatilityVSReliability

Solution Approach 1:

The configuration memory is divided into multiple storage units or blocks, each capable of independent error detection and correction. This segmentation allows localized error handling without affecting the entire configuration, thereby maintaining reliability while preserving the adaptability of the programmable device.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

Error detection and correction codes are pre-integrated into the configuration memory structure before the device operates. This preliminary action ensures that errors from terrestrial radiation are detected and corrected before they can cause functional failures, maintaining both reliability and adaptability.

Inventive Principle:
Principle #10Preliminary action

2Volume of moving object

If miniaturization is pursued to reduce device size, then device compactness is improved, but susceptibility to soft errors from radiation increases

Engineering Contradiction:
Improvedevice sizeVSAvoidradiation sensitivity
Core Design Contradiction:
Volume of moving objectVSObject-affected harmful factors

Solution Approach 1:

Error protection mechanisms are selectively applied to critical regions of the miniaturized device that are most susceptible to radiation effects. This local quality approach provides targeted protection without increasing overall device size, addressing both compactness and radiation sensitivity.

Inventive Principle:
Principle #3Local quality

Solution Approach 2:

The device employs composite structures combining radiation-hardened materials and standard miniaturized components. This composite approach allows the device to maintain small size while specific protective layers or structures mitigate radiation-induced soft errors.

Inventive Principle:
Principle #40Composite materials

3Reliability

If soft error restoration through restart is implemented, then device operation is recovered, but error cause identification becomes difficult

Engineering Contradiction:
Improveoperational recoveryVSAvoiderror diagnostic information
Core Design Contradiction:
ReliabilityVSLoss of information

Solution Approach 1:

The system implements feedback mechanisms that capture and store error information from configuration memory before restoration occurs. This feedback loop preserves diagnostic data about the soft error cause, enabling analysis while still allowing operational recovery through restart.

Inventive Principle:
Principle #23Feedback

Solution Approach 2:

Error information and configuration data are copied to persistent storage before restoration operations. This copying ensures that diagnostic information is preserved even when the original configuration is restored, allowing error cause identification while maintaining operational reliability.

Inventive Principle:
Principle #26Copying

4Loss of information

If conventional error storage techniques are used, then general error information is captured, but specific configuration memory error details are lost

Engineering Contradiction:
Improveerror information retentionVSAvoiderror location accuracy
Core Design Contradiction:
Loss of informationVSMeasurement precision

Solution Approach 1:

The error storage system is segmented to capture specific details about configuration memory errors, including precise location, type, and context information. This segmented approach preserves both general error information and specific diagnostic details, improving measurement precision while maintaining comprehensive error retention.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

Error information is stored in multiple dimensions or layers of detail, from general error classification to specific configuration memory address and bit information. This multi-dimensional storage approach ensures both comprehensive error retention and high measurement precision for error analysis.

Inventive Principle:
Principle #17Another dimension (Dimensionality change)

Data Source

PatentUS10095570B2Programmable device, error storage system, and electronic system device
Publication Date: 2018.10.09 HITACHI LTD
  • US10095570B2 patent drawing
  • US10095570B2 patent drawing
  • US10095570B2 patent drawing

AI summary

The present invention aims to provide a programmable device with a configuration memory that can hold the state of the occurrence abnormal situation that is difficult to assume such as a failure occurring in the programmable device due to the terrestrial radiation of the configuration memory, even during power off, in order to improve the reproducibility in device testing based on the held error information. The programmable device with the configuration memory includes: an error detection section for detecting an error in the configuration memory, and outputting the detected error as well as an address in which the error occurred, as error information; and an error information holding section provided with a non-volatile memory to store the output error information.