RTL Redundant Code Removal Using Equivalence Checking
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
RTL optimization in digital circuits is challenging due to complexity, requiring substantial time and expertise to determine optimal signal widths, multiplexor branches, and operand relevance, which existing tools inadequately address.
Innovation Solution
Automatic generation of optimized RTL via redundant code removal using equivalence checking to confirm functionality, allowing for the identification and removal of unreachable or constant-valued code without human intervention.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If manual RTL optimization is performed, then optimization quality can be controlled, but time consumption and expertise requirements increase substantially
Solution Approach 1:
The system enables automatic RTL optimization by having the optimization tool perform the optimization process itself without requiring manual intervention. The equivalence checking tool automatically analyzes RTL code, identifies redundant operations, and generates optimized code while maintaining functional equivalence, thereby eliminating the time-consuming manual optimization process.
Solution Approach 2:
The patent replaces manual mechanical optimization processes with automated computational methods. The equivalence checking tool uses formal verification methods and computational algorithms to automatically detect and remove redundant code, substituting the manual expert analysis with an automated system that achieves similar optimization quality without requiring human expertise.
2Extent of automation
If existing optimization tools are used, then automation is improved, but they inadequately address complex RTL optimization challenges
Solution Approach 1:
The system incorporates feedback mechanisms where the equivalence checking tool continuously verifies the optimized RTL code against the original to ensure functional equivalence. This feedback loop ensures that automation does not compromise optimization effectiveness, as the tool validates each optimization step and only accepts changes that maintain correct functionality.
Solution Approach 2:
The patent changes the approach by introducing formal verification parameters and equivalence checking criteria that enable reliable automated optimization. By establishing rigorous verification parameters, the system achieves both high automation levels and reliable optimization effectiveness, overcoming the limitations of existing tools that lack such comprehensive verification capabilities.
3Productivity
If redundant code removal is performed, then optimization efficiency improves, but determination of code relevance becomes more difficult
Solution Approach 1:
The system replaces manual detection of code relevance with automated equivalence checking mechanisms. The tool uses formal verification methods to automatically determine which code portions are redundant and can be safely removed, eliminating the difficulty of manual relevance detection while maintaining high optimization efficiency.
Solution Approach 2:
The equivalence checking tool serves as an intermediary between the optimization process and the RTL code. It mediates by automatically analyzing code relevance, identifying redundant portions, and validating optimization changes, thereby simplifying the detection process and enabling efficient automated optimization without requiring direct manual assessment of code relevance.
Data Source
AI summary
Described herein is a technique for automatic generation of optimized RTL via redundant code removal. By automatically introducing local mutations into the original RTL and using equivalence checking tools to confirm that the functionality it is not affected, optimized RTL can be produced automatically without requiring human intervention.