GPU Task Scheduling with Priority Ring Buffers for Smooth Display

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

As refresh rates increase in electronic device displays, the load on the graphics processing unit (GPU) also increases, leading to GPU tasks being queued in a first-created-first-executed order, which can cause freezing and frame dropping of the currently displayed picture due to long-executing tasks blocking later tasks, especially those related to the foreground display.

Innovation Solution

The GPU tasks are sorted in descending order of priorities and allocated resources in ring buffers such that higher-priority tasks are positioned closer to the head, allowing them to be executed first, thereby reducing freezing and frame dropping without increasing GPU frequency or power consumption.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Ease of operation

If GPU tasks are processed in first-created-first-executed order, then task execution follows simple FIFO scheduling, but high-priority display tasks are blocked by long-running low-priority tasks causing freezing and frame dropping

Engineering Contradiction:
Improvetask scheduling simplicityVSAvoiddisplay smoothness
Core Design Contradiction:
Ease of operationVSReliability

Solution Approach 1:

The patent implements dynamic task scheduling by introducing multiple priority levels and allowing tasks to be promoted or demoted based on their importance. High-priority display tasks can jump ahead of lower-priority tasks in the execution queue, making the scheduling system adaptive rather than static. This resolves the contradiction by maintaining simplicity through automated priority-based reordering while ensuring display tasks are not blocked.

Inventive Principle:
Principle #15Dynamics

Solution Approach 2:

The system incorporates feedback mechanisms where the scheduler continuously monitors task characteristics and system state, then adjusts execution order accordingly. Display tasks are identified and given higher priority based on feedback about their importance to user experience, allowing the system to dynamically respond to changing conditions and prevent freezing without complex manual intervention.

Inventive Principle:
Principle #23Feedback

2Productivity

If GPU frequency is increased to reduce task execution time, then task completion speed improves, but power consumption increases

Engineering Contradiction:
Improvetask execution speedVSAvoidGPU power consumption
Core Design Contradiction:
ProductivityVSUse of energy by moving object

Solution Approach 1:

Instead of changing the GPU frequency parameter, the patent changes the scheduling parameter - the order in which tasks are executed. By reordering tasks to prioritize display-related work, the system achieves faster effective completion of critical tasks without increasing power consumption. This resolves the contradiction by improving productivity through parameter optimization rather than resource escalation.

Inventive Principle:
Principle #35Parameter changes

3Speed

If more ring buffers are allocated to high-priority tasks, then high-priority tasks execute faster with more resources, but resource allocation complexity increases

Engineering Contradiction:
Improvehigh-priority task execution speedVSAvoidresource allocation mechanism
Core Design Contradiction:
SpeedVSDevice complexity

Solution Approach 1:

The patent segments the single resource pool into multiple priority-based queues or virtual rings, where high-priority tasks are allocated resources from dedicated segments. This segmentation allows fast execution of critical tasks while keeping the allocation mechanism relatively simple through clear separation of priorities, resolving the contradiction between speed and complexity.

Inventive Principle:
Principle #1Segmentation

Data Source

PatentEP4682715A1Task scheduling method, electronic device, storage medium, and chip
Publication Date: 2026.01.21 HONOR DEVICE CO LTD
  • EP4682715A1 patent drawingFigure 1
  • EP4682715A1 patent drawingFigure 2A
  • EP4682715A1 patent drawingFigure 2B

AI summary

This application provides a task scheduling method, an electronic device, a storage medium, and a chip, and relates to the field of terminal technologies. The method includes: acquiring a plurality of graphics processing unit (GPU) tasks, and sorting the plurality of GPU tasks in descending order of priorities of GPU task types; allocating resources to the plurality of GPU tasks in a plurality of ring buffers, so that resource proportions of GPU tasks sorted higher are higher at positions closer to heads of the plurality of ring buffers; and sequentially executing the plurality of GPU tasks. By using the method, in a scenario with heavy GPU load, freezing of a currently displayed picture of an electronic device can be reduced, thereby improving the user experience. This solution requires no GPU frequency increase, so that no GPU power consumption increase is caused.