Cache ECC Wordline Failure Detection Across Multiple ECC Words
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current error detection schemes in memory systems, particularly in cache designs, often fail to detect wordline errors effectively, leading to mis-correction of data and unacceptably high probabilities of incorrect error identification, especially when multiple bits fail simultaneously.
Innovation Solution
A controller is implemented to read multiple error correcting code (ECC) words from a wordline, detecting a first error in one ECC word and a second error in another ECC word, thereby identifying a failing wordline by treating combinations of errors as uncorrectable errors, reducing mis-correct probabilities through restricted 'CE' or 'UE' information from multiple ECC groups.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional error detection schemes are used to detect errors in memory, then single bit errors can be detected and corrected, but multiple bit errors (wordline errors) cannot be effectively detected, leading to mis-correction of data
Solution Approach 1:
The patent segments the error detection process into multiple independent syndrome checks, one for each ECC word within a wordline. Instead of treating the entire wordline as a single unit, each ECC word is checked separately using its own syndrome calculation, allowing identification of which specific ECC words contain errors and thereby detecting wordline-level failures.
Solution Approach 2:
The patent adds a new dimension to error detection by checking errors across multiple ECC words within the same wordline simultaneously. This multi-dimensional approach examines error patterns across different ECC words (ECC word 0, ECC word 1, etc.) to identify wordline errors that single-wordline checks would miss.
2Measurement precision
If comprehensive error detection is implemented to detect all error types, then detection accuracy improves, but the complexity of detection circuitry increases significantly
Solution Approach 1:
The detection circuitry is segmented into multiple independent syndrome check units, each handling one ECC word. This modular approach allows comprehensive error detection across all ECC words while keeping each individual check unit relatively simple and manageable.
Solution Approach 2:
Each ECC word within the wordline performs self-service error detection through its own syndrome calculation and check. The system leverages the existing ECC structure of each word to detect errors independently, eliminating the need for a single complex centralized detection circuit.
3Reliability
If multiple ECC words are checked for errors, then wordline failure detection capability improves, but the amount of detection circuitry and processing required increases
Solution Approach 1:
The detection function is segmented across multiple ECC words, with each word having its own syndrome check. This segmentation allows the system to detect wordline failures by examining multiple ECC words independently, distributing the detection workload rather than requiring a single complex circuit.
Solution Approach 2:
The syndrome check circuitry is designed to be universal and reusable across multiple ECC words. The same basic syndrome calculation and comparison logic is applied to each ECC word, allowing the system to detect errors in multiple words using the same detection mechanism rather than requiring different circuits for each word.
Data Source
AI summary
A technique for error detection is provided. A controller is configured to detect errors by using error correcting code (ECC), and a cache includes independent ECC words for storing data. The controller detects the errors in the ECC words for a wordline that is read. The controller detects a first error in a first ECC word on the wordline and a second error in a second ECC word on the wordline. The controller determines that the wordline is a failing wordline based on detecting the first error in the first ECC word and the second error in the second ECC word.