Operating System Error Handling via BMC Shared Memory
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional error handling models in information handling systems face security concerns and performance degradation due to their reliance on firmware-first approaches, while operating-system-first models fail to report errors to computing device management systems effectively.
Innovation Solution
An Information Handling System (IHS) that includes a primary processing system and a primary memory system, which provides an operating system engine to identify errors, generate error information, access error handling information stored in a reserved memory subsystem, and store this information in a shared memory subsystem within a Baseboard Management Controller (BMC) device, thereby notifying the BMC of the error information.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a firmware-first error handling model is used, then error detection and logging can be performed by BIOS, but security concerns arise due to SMI providing a common target for exploits and performance degradation occurs due to latency scaling with number of cores
Solution Approach 1:
The patent extracts the SMI/SMM mechanism from the error handling process and replaces it with direct operating system-level error handling. The operating system subsystem directly detects errors from hardware components and logs them without invoking firmware SMI handlers, thereby eliminating the security target and performance bottleneck while preserving error detection capability
Solution Approach 2:
The patent introduces a shared memory subsystem as an intermediary between the operating system subsystem and the BMC device. Error information is written to this shared memory by the OS subsystem and then read by the BMC, enabling communication without requiring firmware involvement or SMI mechanisms
2Object-affected harmful factors
If a conventional operating-system-first error handling model is used, then security concerns and performance degradation are eliminated, but error reporting to computing device management systems fails because error handling is not managed by BIOS firmware
Solution Approach 1:
The patent introduces a shared memory subsystem as an intermediary between the operating system subsystem and the BMC device. Error information is written to this shared memory by the OS subsystem and then read by the BMC, enabling communication without requiring firmware involvement or SMI mechanisms
Solution Approach 2:
The patent establishes a feedback mechanism where the BMC device monitors the shared memory subsystem for error information written by the operating system subsystem. This enables the management system to receive error reports and provide appropriate responses or actions based on the error conditions detected
Data Source
AI summary
An operating-system-first error handling system includes an operating system subsystem coupled to a BMC device including a shared memory subsystem, and to a reserved memory subsystem storing error handling information that is configured to direct the storage of error information in the shared memory subsystem in the BMC device. The operating system subsystem identifies a first error and generates first error information for the first error. The operating system subsystem then accesses the error handling information and, based on the error handling information, stores the first error information generated for the first error in the shared memory subsystem included in the BMC device. The operating system subsystem then notifies the BMC device of the first error information stored in the shared memory subsystem. The BMC device may then access the first error information in the shared memory subsystem and transmit the first error information to a computing device management system.