Configurable IC Error-Correction Circuitry for Soft-Error Recovery
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Configurable integrated circuits (ICs) face issues with random bit errors in their configuration memory due to high-energy particles, leading to unpredictable system failures, and existing solutions like scrubbing and triple modular redundancy are either time-consuming or require significant additional IC real estate.
Innovation Solution
A configurable IC with integrated error detection and correction circuitry that continuously checks and corrects configuration data, using error correction codes and a two-tiered multiplexer structure to provide error-corrected data to reconfigurable circuits at a faster rate, with the ability to write corrected data back into memory.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If scrubbing is used to periodically check configuration memory, then error detection capability is improved, but system productivity decreases due to time taken away from external processes
Solution Approach 1:
The configurable IC performs self-diagnosis and self-correction of configuration memory errors through integrated error correction circuitry that operates autonomously during normal operation, eliminating the need for external scrubbing operations and maintaining continuous productivity while improving reliability
2Reliability
If triple modular redundancy is applied to protect against soft errors, then system reliability is improved, but IC area increases significantly requiring more than three times as much real estate
Solution Approach 1:
The error correction circuitry is integrated within the existing configurable IC structure, merging error detection and correction functions with the logic circuits and configuration memory rather than using separate redundant circuits, thereby achieving reliability improvement without proportional increase in IC area
Solution Approach 2:
The patent uses error correction codes that add relatively few check bits to the configuration data (e.g., 7 check bits for 63 data bits in Hamming code), changing the data representation parameters to provide comprehensive error protection while occupying minimal additional space compared to full triple modular redundancy
3Reliability
If configuration memory is checked periodically by external system, then error detection is improved, but the likelihood of system fault between checks increases
Solution Approach 1:
The integrated error correction circuitry operates continuously during normal IC operation, constantly monitoring and correcting configuration memory errors rather than performing periodic checks, thereby eliminating gaps where faults could occur undetected while maintaining system productivity
Data Source
AI summary
An integrated circuit (IC) performs error detection and correction on configuration data. The IC includes a configuration memory for storing configuration data and error correction data, and error correction circuitry for receiving the configuration data, correcting a particular error in the received configuration data when the particular error exists in the configuration data, and outputting the configuration data without the particular error. The IC further includes a configurable circuit (e.g., a configurable logic circuit or a configurable interconnect circuit) that receives the error-corrected configuration data from the error correction circuitry, and circuitry to write the corrected configuration data and error data back into the configuration memory.