Processor Key Management for Data Scrambling Security
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Computer processors face security vulnerabilities due to unauthorized access through shared memory and cache coherence protocols, as seen in issues like Meltdown and Spectre, which allow malicious programs to access sensitive data.
Innovation Solution
Implementing data scrambling within the processor, where original data is rearranged using a random key to create a scrambled form that can only be unscrambled with a secret key, preventing unauthorized access by ensuring only authorized applications can recover the original data, and using multiplexers to efficiently reorder bits without significant performance impact.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If data is stored in shared memory and cache for processor access, then processor performance and data accessibility are improved, but security vulnerability increases due to unauthorized access
Solution Approach 1:
The memory space is segmented into different protection domains, each with its own access control mechanism. The system divides the memory into protected memory regions that can be accessed by specific processors or processors with specific permissions, preventing unauthorized access while maintaining performance for authorized operations.
Solution Approach 2:
Different memory regions have different security properties and access control mechanisms. Each memory region can be configured with specific protection levels, allowing the system to provide strong security where needed while maintaining high performance for authorized access in other regions.
2Speed
If speculative execution is used to execute instructions before determination, then processing speed is improved, but security vulnerability increases due to unauthorized data access
Solution Approach 1:
The system introduces an intermediary mechanism between the speculative execution unit and the memory access unit. This intermediary acts as a gatekeeper that validates access rights before allowing data to be read from memory during speculative execution, preventing unauthorized access while maintaining the performance benefits of speculative execution.
3Object-affected harmful factors
If data scrambling is implemented to prevent unauthorized access, then data security is improved, but processing complexity increases
Solution Approach 1:
The data scrambling function is merged with the existing memory access and processing units. Rather than adding a separate complex scrambling system, the scrambling operation is integrated into the data path, combining security functionality with existing hardware components to minimize additional complexity.
Solution Approach 2:
The system uses its own execution units to perform the scrambling and unscrambling operations. The processors use their existing arithmetic and logic capabilities to encrypt and decrypt data without requiring external specialized hardware, thereby reducing overall system complexity while maintaining security.
Data Source
AI summary
Systems, apparatuses, and methods related to a computer system having a processor and a main memory storing scrambled data are described. The processor may have a secure zone configured to store keys and an unscrambled zone configured to operate on unscrambled data. The processor can convert the scrambled data into the unscrambled data in the unscrambled zone using the keys retrieved from the secure zone in response to execution of instructions configured to operate on the unscrambled data. Another processor may also be coupled with the memory, but can be prevented from accessing the unscrambled data in the unscrambled zone.