I/O Error Recovery via Platform Communication Channel
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current error-containment logic, such as Live Error Recovery (LER) technology, fails to coordinate effectively with operating systems (OS) and virtual machine monitors (VMM) during input/output (I/O) error events, leading to potential system failures despite contained errors.
Innovation Solution
Implementing a platform communication channel (PCC) to notify the OS/VMM of I/O error-containment events, allowing for coordinated recovery by clearing error registers and transitioning I/O ports from a LinkDown to a LinkUp state, thereby preventing OS/VMM-level failures.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Object-affected harmful factors
If error-containment logic (LER) is used to contain I/O errors, then error containment capability is improved, but system reliability deteriorates due to lack of coordination with OS/VMM
Solution Approach 1:
The patent implements a feedback mechanism where the error-containment logic notifies the OS/VMM of error events through platform communication channels. This allows the OS/VMM to respond appropriately to contained errors, preventing system failures while maintaining the error containment capability. The notification system creates a closed-loop feedback path between the error containment hardware and the software layer.
Solution Approach 2:
The patent introduces platform communication channels as an intermediary mechanism between the error-containment logic and the OS/VMM. This intermediary enables coordinated error handling by allowing the error containment logic to notify higher-level software without direct coupling, resolving the coordination problem while maintaining system reliability.
2Ease of repair
If LER technology is implemented for error recovery, then error recovery capability is improved, but coordination with OS/VMM deteriorates due to insufficient communication mechanisms
Solution Approach 1:
The patent utilizes existing platform communication channels, which are multi-functional communication paths already present in the system architecture. These channels serve both normal system communication and error notification purposes, avoiding the need for dedicated error communication infrastructure and reducing overall system complexity.
Solution Approach 2:
The error-containment logic autonomously notifies the OS/VMM through the platform communication channels without requiring complex external coordination mechanisms. The system components serve themselves by utilizing existing communication infrastructure, simplifying the coordination process while maintaining effective error recovery capability.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
Embodiments of apparatus, computer-implemented methods, systems, devices, and computer-readable media are described herein for a computing device with a platform entity such as an interrupt handler configured to notify an operating system or virtual machine monitor executing on the computing device of an input/output error-containment event. In various embodiments, the interrupt handler may be configured to facilitate recovery of a link to an input/output device that caused the input/output error-containment event, responsive to a directive from the operating system or virtual machine monitor.