Memory ECC Poisoning for On-Die and System-Level Error Testing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
As memory devices scale to increase memory densities and storage capacities, ensuring reliability becomes increasingly challenging, particularly in detecting and correcting errors within the data stored, which existing error correcting code (ECC) systems may not adequately address, especially when testing on-die ECC circuits and system-level ECC functionalities in operational environments.
Innovation Solution
Implementing a method to 'poison' data bits within memory devices, allowing the host device to control which bits to invert during write or read operations, enabling the testing of on-die and system-level ECC functionalities by generating and correcting error codes, thereby validating the ECC circuit's functionality.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If memory devices scale to increase memory densities, then storage capacity is improved, but reliability deteriorates due to increased error rates
Solution Approach 1:
The patent applies preliminary action by pre-computing and storing expected ECC syndrome values (poison values) in a lookup table before data is written to memory. When data is read back, the actual syndrome is compared against these pre-computed values to detect errors, enabling error detection before the data is fully processed or used.
Solution Approach 2:
The patent introduces an intermediary mechanism (ECC circuit with poison value comparison) between the memory array and the data processing logic. This intermediary computes syndromes and compares them against pre-stored poison values to identify errors, acting as a mediator that filters out corrupted data before it reaches the main system.
2Reliability
If existing ECC systems are used for error detection, then error correction capability is provided, but testing of on-die ECC circuits becomes difficult when test modes are not accessible
Solution Approach 1:
The patent enables self-service testing by allowing the system to generate its own test patterns and expected results using the normal ECC computation logic. The host system can write data, have the ECC circuit compute syndromes, and then compare against pre-computed poison values, enabling the ECC circuit to test itself through normal operational paths without requiring external test modes.
Solution Approach 2:
The patent makes the ECC circuit multi-functional by enabling it to perform both its primary function (error detection and correction during normal operation) and a secondary function (self-testing) using the same hardware resources and operational paths. The poison value mechanism serves dual purposes: detecting errors in production and validating ECC functionality during testing.
3Reliability
If data bits are inverted to test ECC functionality, then ECC circuit validation is enabled, but data integrity is temporarily compromised
Solution Approach 1:
The patent applies preliminary action by pre-computing the expected syndrome values (poison values) that would result from specific bit inversions. This allows the system to intentionally create known error patterns, store the expected outcomes, and then validate that the ECC circuit correctly detects and corrects these errors, while maintaining data integrity through the validation process.
Data Source
AI summary
Memory devices, systems including memory devices, and methods of operating memory devices are described, in which memory devices are configured to poison data based on an indication provided by a host device coupled with the memory devices. The indication may include which one or more bits to poison (invert) at which stages of performing write or read operations. In some embodiments, the memory device may invert one or more bits according to the indication and then correct one or more errors associated with inverting the one or more bit to verify its on-die ECC functionality. In some embodiments, the memory device may provide the host device with poisoned data including one or more bits inverted according to the indication such that the host device may test system-level ECC functionality using the poisoned data.