Dynamic GPU Memory Bandwidth Limiting for Audio Video Playback
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In bandwidth-limited systems, such as SoCs used in electronic devices, the varying memory bandwidth requirements for audio/video decoding and GPU processing lead to decreased performance when both are high, affecting smooth playback and user experience.
Innovation Solution
A method is introduced to dynamically limit the memory bandwidth of a GPU by detecting decoding times of frames and adjusting the maximum granted bandwidth based on target times, ensuring sufficient bandwidth allocation for audio/video decoding.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If the memory bandwidth is allocated to both audio/video decoder and GPU simultaneously at high levels, then the performance of both components can be maintained, but the total memory bandwidth requirement exceeds the system's limited capacity causing performance degradation
Solution Approach 1:
The patent implements dynamic memory bandwidth allocation by continuously monitoring the decoding performance of the audio/video decoder and adjusting the GPU's memory bandwidth grant amount in real-time. The CPU acts as a dynamic controller that modifies the max grant amount based on current system conditions, transforming the static bandwidth allocation into a dynamic adaptive system that responds to changing workload demands.
Solution Approach 2:
The patent changes the parameter of memory bandwidth allocation by adjusting the max grant amount for the GPU based on the audio/video decoder's performance metrics. When the decoder performs well (below threshold), the GPU receives higher bandwidth; when the decoder struggles (above threshold), the GPU's bandwidth is reduced. This parameter adjustment resolves the contradiction by adapting bandwidth distribution to actual system needs.
2Reliability
If the memory bandwidth is prioritized to the audio/video decoder to maintain playback smoothness, then the audio/video playback quality is improved, but the GPU performance deteriorates due to reduced memory bandwidth
Solution Approach 1:
The patent dynamically changes the memory bandwidth parameter allocated to the GPU based on the audio/video decoder's performance. By monitoring decoding times and comparing them against a threshold, the system adjusts the max grant amount to ensure the decoder receives sufficient bandwidth for smooth playback while allowing the GPU to utilize remaining bandwidth capacity, thus resolving the trade-off between playback reliability and GPU performance.
Solution Approach 2:
The patent implements a feedback mechanism where the CPU monitors the audio/video decoder's performance metrics (decoding times) and uses this feedback to adjust the GPU's memory bandwidth allocation. This closed-loop control ensures that when playback smoothness is compromised, the system responds by allocating more bandwidth to the decoder, while maintaining optimal GPU performance when conditions allow.
3Productivity
If the memory bandwidth is allocated dynamically based on decoding performance, then the overall system efficiency is improved, but the system complexity increases due to additional monitoring and control mechanisms
Solution Approach 1:
The patent implements a self-service mechanism where the CPU automatically monitors decoding performance and adjusts GPU bandwidth allocation without external intervention. The system serves itself by using its own processing capacity to manage resource allocation, eliminating the need for additional complex control hardware or external management systems, thus improving efficiency while limiting the increase in complexity.
Data Source
AI summary
A method for dynamically limiting a memory bandwidth of a graphics processing unit (GPU) is applicable to a bandwidth-limited system. The bandwidth-limited system includes an audio/video decoder and a GPU. The method includes detecting a plurality of decoding times of a plurality of frames of an audio/video decoded by an audio/video decoder, and adjusting a max grant amount of a memory bandwidth of a GPU according to the plurality of decoding times of the plurality of frames and a target time. Therefore, in the case where the total memory bandwidth is limited, the memory bandwidth of the GPU is limited by the performance of the corresponding audio/video decoder such that, during audio/video playback, the effect of audio/video playback can be prevented from being affected and at the same time a better graphical user interface can be provided.