Hypervisor Abnormality Detection via Peripheral Access Recording
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for monitoring a hypervisor, especially third-party hypervisors with closed internal logic, fail to accurately detect abnormalities without accessing the hypervisor's internal state, and may miss abnormal states caused by internal issues or require prioritization that can prevent deadlock detection.
Innovation Solution
An electronic controller design that includes a virtual machine, a hypervisor, an access recording unit, a state recording unit, and a monitoring unit, which records peripheral access requests and states to determine hypervisor abnormalities based on input and output, allowing detection without accessing the hypervisor's internal state.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a monitoring virtual machine is installed to periodically execute test patterns and acquire internal state of the hypervisor, then abnormality detection capability is improved, but device complexity increases and the method cannot be applied to third-party hypervisors with closed internal logic
Solution Approach 1:
The patent extracts the monitoring function from the hypervisor's internal state and relocates it to the virtual machine layer. Instead of monitoring internal hypervisor state directly, the system observes peripheral access requests and responses between virtual machines and the hypervisor, thereby detecting abnormalities without increasing core system complexity or requiring internal hypervisor access.
Solution Approach 2:
The patent introduces peripheral access requests and responses as intermediary observables. These peripheral interactions serve as mediators that indirectly reveal hypervisor operational status without requiring direct access to internal hypervisor state, enabling monitoring of third-party hypervisors with closed internal logic.
2Speed
If a watchdog timer is used to monitor hypervisor operation with high priority counting, then abnormality detection speed is improved, but deadlock detection capability deteriorates due to priority inversion
Solution Approach 1:
The patent dynamically adjusts monitoring priorities based on the type of abnormality being detected. Instead of using a fixed high-priority watchdog timer that can cause priority inversion, the system adapts its monitoring approach to observe peripheral access patterns, enabling simultaneous detection of both fast abnormalities and deadlocks without priority conflicts.
Solution Approach 2:
The patent implements feedback mechanisms that monitor the completion status of peripheral access requests. By tracking whether requests are properly completed or stuck in pending state, the system can detect deadlocks through the absence of expected feedback, rather than relying solely on high-priority timing mechanisms.
Data Source
AI summary
An electronic controller includes: a virtual machine that accesses a first virtual driver to execute a process; a hypervisor that calls a first real driver of a first peripheral, based on a peripheral access request received from the first virtual driver; an access recording unit that calls the first virtual driver to record a peripheral access request transmitted to the hypervisor; a state recording unit that calls the first real driver to record a state of a register of the first peripheral; and a monitoring unit that monitors an operation of the hypervisor. The monitoring unit determines an abnormality of the hypervisor, based on a record by the access recording unit and on a record by the state recording unit.