PCIe Link Equalization Training Parameter Reuse
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The existing PCIe link equalization training process is lengthy, causing bus unavailability and potential system crashes or data corruption, especially when retraining is required outside the initial boot sequence, which can compromise system robustness and lead to errors or crashes.
Innovation Solution
The solution involves reusing optimal parameter coefficients from a previous successful equalization training cycle to reduce the time required for equalization training, allowing the system to bypass the time-intensive search for new optimal parameters and complete the training faster.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If equalization training is performed to ensure favorable bus bit error rate at higher speeds, then link reliability is improved, but link availability deteriorates due to lengthy training time
Solution Approach 1:
The system performs equalization training during the initial system boot sequence before the operating system loads, so that the link is already trained and available when needed. This preliminary action ensures that subsequent retraining events are minimized or eliminated, resolving the contradiction by preparing the link in advance rather than training it when data transmission is needed.
Solution Approach 2:
The system changes the timing parameter of equalization training from on-demand (when data needs to be transmitted) to upfront (during boot sequence). This parameter change transforms the training process from a reactive operation that causes downtime to a proactive operation that occurs when system resources are abundant, thus improving link availability while maintaining reliability.
2Adaptability or versatility
If equalization retraining is performed outside initial boot sequence, then link adaptability is improved, but system robustness deteriorates due to operating system delays
Solution Approach 1:
The system performs equalization training during the initial boot sequence before the operating system loads, so that the link is already trained and available when needed. This preliminary action ensures that subsequent retraining events are minimized or eliminated, resolving the contradiction by preparing the link in advance rather than training it when data transmission is needed.
Solution Approach 2:
The system prevents the harmful effect of operating system delays by performing equalization training before the operating system is loaded. This preliminary anti-action counteracts the potential robustness issues that would arise from retraining during OS operation, allowing the system to maintain both adaptability and robustness.
3Speed
If PCIe link equalization training is performed at higher speeds, then communication speed is improved, but training time increases causing bus unavailability
Solution Approach 1:
The system performs equalization training during the initial system boot sequence before the operating system loads, so that the link is already trained and available when needed. This preliminary action ensures that subsequent retraining events are minimized or eliminated, resolving the contradiction by preparing the link in advance rather than training it when data transmission is needed.
Solution Approach 2:
The system changes the timing parameter of equalization training from on-demand (when data needs to be transmitted) to upfront (during boot sequence). This parameter change transforms the training process from a reactive operation that causes downtime to a proactive operation that occurs when system resources are abundant, thus improving link availability while maintaining reliability.
Data Source
AI summary
In a system according to one embodiment of the present disclosure, the system comprises a first device, a second device, a communications link, and a memory. The memory stores instructions that when executed by the system perform a method of communications link training. This method comprises requesting a speed change to a second speed for the first device communicating with the second device at a first speed via the communications link. A saved set of parameters are accessed for at least one of the first device and the second device. A first training cycle is performed for the first device and the second device at the second speed using the saved set of parameters for the at least one of the first device and second device. The reuse of parameters from a previous successful equalization training cycle reduces the time required to perform equalization training.


