Display Processor Dynamic Blend-Core Allocation for Multi-Layer Rendering
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The increasing demand for high dynamic range, frame rate, and resolution in image display requires a larger number of blend cores in the DPU, leading to an increased area of the blender, posing a challenge for implementation.
Innovation Solution
A method for processing display data that filters out to-be-blended source layers and assigns them to blend cores one-to-one, reducing the need for pre-allocation and minimizing the number of blend cores required, while maintaining the original hierarchy of the source layers.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If the number of blend cores is increased to handle more layers, then the blending capability and display quality are improved, but the area of the blender increases significantly
Solution Approach 1:
The patent implements dynamic allocation of blend cores based on the actual number of layers being blended. Instead of having a fixed number of blend cores, the system dynamically determines how many blend cores are needed based on the input layers, allowing the blender to adapt its configuration to different display requirements without wasting resources on unused blend cores.
Solution Approach 2:
The patent makes each blend core multi-functional by enabling them to handle multiple layers through dynamic assignment. A single blend core can process different layers depending on the current blending task, allowing the same hardware resource to serve multiple purposes and reducing the total number of blend cores needed.
2Productivity
If more blend cores are allocated for each source layer, then the blending processing is improved, but the hardware area and complexity increase
Solution Approach 1:
The system dynamically assigns blend cores to source layers based on the actual blending requirements. Instead of pre-allocating a fixed number of blend cores for each layer, the system determines the exact number of blend cores needed based on the input layers and their blending relationships, optimizing both processing efficiency and hardware utilization.
Solution Approach 2:
The patent changes the parameter of blend core allocation from a fixed static configuration to a dynamic variable configuration. The number of active blend cores is adjusted based on the number of source layers and their blending hierarchy, allowing the system to optimize hardware usage without compromising processing performance.
Data Source
AI summary
The present disclosure provides a method for processing display data, electronic device, and display processor unit. The method comprises obtaining a pixel coordinate of a target pixel point; searching for a to-be-blended source layer among multiple source layers, wherein the to-be-blended source layer comprises pixels to be displayed on the display axis, and the display axis is the coordinate axis of the pixel coordinate in the pixel coordinate system; assigning the to-be-blended source layer to different blend cores, wherein the to-be-blended source layer corresponds one-to-one with the blend core; and each blend core blending the assigned to-be-blended source layer to generate display data. The method allows the blending of multiple layers with a small number of blend cores, reducing the demand for the number of blend cores and minimizing the area of the blender.