GPU Memory Access Controller Secure Mode Segmentation

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Modern operating systems, particularly open platforms like Android, lack strong trust in protecting secure content due to vulnerabilities in kernel mode drivers, which can be altered by third parties, leading to insecure storage and copying of digital rights managed content.

Innovation Solution

Implementing a GPU with a memory access controller that operates in secure and unsecure modes, restricting access to secure memory regions to prevent unauthorized copying of protected content, even if the driver is altered to place the GPU in an unsecure mode.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Ease of operation

If kernel mode drivers are used to control GPU operation, then GPU functionality is achieved, but security is compromised because drivers can be altered by third parties

Engineering Contradiction:
ImproveGPU control capabilityVSAvoidcontent protection security
Core Design Contradiction:
Ease of operationVSReliability

Solution Approach 1:

The patent segments the GPU into multiple operational modes (secure mode and unsecure mode) with distinct functionality. In secure mode, the GPU restricts memory access to protected content regions, while in unsecure mode, full functionality is available. This segmentation allows the system to maintain both ease of operation through driver control and reliability through hardware-enforced security boundaries.

Inventive Principle:
Principle #1Segmentation

2Adaptability or versatility

If open source drivers are used for GPU control, then system accessibility is improved, but security against unauthorized content copying deteriorates

Engineering Contradiction:
Improvedriver compatibilityVSAvoidunauthorized content copying
Core Design Contradiction:
Adaptability or versatilityVSObject-affected harmful factors

Solution Approach 1:

The patent introduces secure memory regions as an intermediary layer between the GPU and unsecure system memory. This intermediary secure memory is protected from unauthorized access and copying, allowing open source drivers to maintain adaptability and compatibility while the hardware-enforced secure memory boundary prevents harmful content copying.

Inventive Principle:
Principle #24Intermediary (Mediator)

3Adaptability or versatility

If GPU access to all memory regions is allowed, then processing flexibility is improved, but content protection is compromised

Engineering Contradiction:
Improvememory access flexibilityVSAvoidsecure content integrity
Core Design Contradiction:
Adaptability or versatilityVSReliability

Solution Approach 1:

The patent applies local quality by creating different access permissions for different memory regions. Secure memory regions have restricted access permissions that prevent copying, while unsecure memory regions maintain full access flexibility. This allows the GPU to process content with appropriate security constraints applied locally to each memory region, maintaining both flexibility and protection.

Inventive Principle:
Principle #3Local quality

Data Source

PatentEP3332345B1Hardware enforced content protection for graphics processing units
Publication Date: 2020.03.11 QUALCOMM INC
  • EP3332345B1 patent drawingFigure 1
  • EP3332345B1 patent drawingFigure 2
  • EP3332345B1 patent drawingFigure 3

AI summary

This disclosure proposes techniques for graphics processing. In one example, a graphics processing unit (GPU) is configured to access a memory according to one of an unsecure mode and a secure mode. The GPU may include a memory access controller configured to direct memory transactions from at least one hardware unit of the GPU to a secure context bank in a memory controller when the GPU is operating in a secure mode, and configured to direct memory transactions from the at least one hardware unit of the GPU to an unsecure context bank in the memory controller when the GPU is operating in the unsecure mode.