Multi-Core Scheduling via Relative Priority Coordination
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current multi-core systems in RTOS lack an effective method for relative priority-based scheduling across cores, leading to inefficiencies in task management and resource allocation, particularly in asymmetric multiprocessing configurations where tasks on different cores do not share priority relations.
Innovation Solution
A method and electronic device for handling relative priority-based scheduling by determining and scheduling tasks across multiple cores based on relative priority, allowing tasks to follow priority relations with other cores, and using signaling messages for force suspension or preemption to ensure optimal task execution.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If tasks are permanently allocated to separate cores in AMP configurations, then each core has dedicated resource control, but scheduling flexibility and inter-core priority coordination are lost
Solution Approach 1:
The patent implements asymmetric multiprocessing by allowing each core to have its own operating system instance while introducing a master-slave relationship for scheduling coordination. The master core maintains global priority information and coordinates with slave cores, creating an asymmetric architecture that combines dedicated resource control with centralized scheduling flexibility.
Solution Approach 2:
The patent introduces a scheduling coordinator acting as an intermediary between cores. This coordinator receives scheduling decisions from the master OS instance and distributes them to slave cores, enabling inter-core priority coordination without requiring permanent task allocation or full SMP complexity.
2Adaptability or versatility
If a single OS controls all cores in SMP configurations, then scheduling flexibility is improved, but system complexity and coordination overhead increase
Solution Approach 1:
The patent segments the operating system into multiple instances, with one master OS instance on a designated core and slave OS instances on other cores. This segmentation allows each core to have local OS control while the master instance maintains global scheduling authority, reducing the complexity of full SMP while preserving scheduling flexibility.
Solution Approach 2:
The patent adds a hierarchical dimension to the scheduling architecture by introducing master-slave relationships among OS instances. Instead of flat SMP coordination, the system uses a two-level hierarchy where the master OS instance makes scheduling decisions and slave instances execute them, reducing coordination overhead.
3Productivity
If tasks are bounded to specific cores, then resource allocation efficiency improves, but inter-core priority relations and preemption capabilities are lost
Solution Approach 1:
The patent performs preliminary actions by having the master OS instance pre-calculate and distribute scheduling decisions to slave cores before task execution. This allows bounded tasks to maintain their core assignments while the system pre-establishes priority relationships and preemption rules across cores, enabling both efficiency and coordination.
Data Source
AI summary
A method for handling a relative priority based scheduling procedure in an electronic device. The method includes receiving at least one task, from an application, to access a first core from a plurality of cores in the electronic device. Further, the method includes determining a relative priority with at least one second core from the plurality of cores in the electronic device based on the at least one task. Further, the method includes scheduling the at least one task in the first core from the plurality of cores in the electronic device based on the relative priority.