Memory Controller Data Mask Training for RAM Timing Alignment
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional RAM designs lack suitable measures for aligning data (DQ) and data mask (DM) signals with respect to data strobe (DQS) signals, leading to timing skew issues that worsen with increasing clock rates and varying operational conditions, resulting in unreliable high-speed data transfer.
Innovation Solution
A method and system for expeditiously training RAM to align DQ and DM signals with DQS signals by performing write training procedures that establish optimal delay values through incremental variations and comparisons, using a memory controller with delay circuits to selectively delay signals and ensure precise timing alignment.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If conventional RAM designs are used without built-in training measures, then device complexity is reduced, but timing alignment precision between DQ, DM, and DQS signals deteriorates at high speeds
Solution Approach 1:
The patent performs preliminary training operations before normal data transfer to establish optimal delay values. The memory controller executes write training procedures that incrementally adjust DQ and DM signal delays relative to DQS signals, storing the determined optimal delay values for subsequent use. This preliminary action ensures timing alignment is established in advance, compensating for the lack of built-in training measures in conventional RAM designs.
Solution Approach 2:
The memory controller performs self-training by automatically adjusting its own signal delays. The controller incrementally varies DQ and DM delay values, writes test patterns to the RAM, reads back the data, and compares results to determine optimal alignment. This self-service approach allows the system to autonomously establish timing margins without requiring additional hardware training circuits in the RAM device itself.
2Productivity
If clock rate is increased to improve data transfer speed, then productivity increases, but timing margin and reliability deteriorate
Solution Approach 1:
The patent implements dynamic delay adjustment where the memory controller adaptively sets DQ and DM signal delays based on actual timing measurements. The training procedure incrementally adjusts delay values and uses read-back verification to determine the optimal dynamic delay settings that maximize timing margins at the specific operating speed. This dynamic adjustment allows the system to maintain reliability even at increased clock rates by optimizing timing parameters for each operating condition.
Solution Approach 2:
The patent changes the delay parameters of DQ and DM signals to optimize timing alignment. By incrementally varying delay values during training and selecting the combination that produces the most accurate read-back data, the system establishes optimal timing parameters that maintain reliability at high transfer speeds. The memory controller stores these optimized parameter settings for use during normal high-speed operation.
3Measurement precision
If incremental delay variation and comparison procedures are implemented, then timing alignment precision improves, but training time and operational complexity increase
Solution Approach 1:
The patent performs the time-consuming incremental delay variation and comparison procedures as a preliminary training operation that is executed once during initialization or when timing conditions change. The memory controller systematically adjusts DQ and DM delays, performs write-read-compare cycles, and stores the determined optimal delay values. This preliminary action consolidates the time loss to an initial setup phase, allowing subsequent data transfers to benefit from the established timing alignment without repeating the lengthy training procedure.
Data Source
AI summary
A system and method providing timing alignment of a data mask (DM) signal with respect to a data strobe (DQS) signal for memory devices not designed for adjusting such alignment is provided. Alignment between data signals (DQ) and a DQS signal is first achieved during a first write training procedure where a data delay value is optimized for one of the DQS or DQ signals. Subsequently, using the optimum delay value from the first write training procedure, a second write training procedure is initiated. In the second write training procedure, timing alignment between the DM signal and the DQ signals is achieved by determining an optimal delay value of the DM signal relative to the DQS signal.