Statistical Timing Analysis for Cycle Time Independent Tests
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for statistical static timing analysis (SSTA) of cycle time independent tests in high-performance digital circuits are inefficient, leading to excessive calculations and increased runtime due to full propagation of timing data throughout the design, even for tests that are frequency independent.
Innovation Solution
A method and system that identify and filter timing quantities feeding cycle time dependent tests through fast forward and backward propagation marking steps, propagating statistical timing only where values are needed for downstream cycle time independent tests, thereby reducing wasteful calculations and runtime.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If full propagation of timing data is performed throughout the design, then complete timing analysis coverage is achieved, but calculation complexity and runtime increase excessively
Solution Approach 1:
The patent segments the timing analysis process into two distinct phases: a first phase that performs full propagation to identify all cycle time independent tests, and a second phase that performs propagation only to those specific tests. This segmentation allows the system to maintain complete coverage while avoiding redundant calculations in the second phase.
Solution Approach 2:
The patent performs a preliminary full propagation in the first phase to identify which timing tests are cycle time independent. This preliminary action enables the second phase to target only the relevant tests, avoiding unnecessary calculations while ensuring complete coverage of all independent tests.
2Reliability
If full propagation of timing data is performed throughout the design, then all timing tests are analyzed, but runtime increases excessively
Solution Approach 1:
The timing analysis is divided into two sequential phases: Phase 1 performs complete propagation to identify cycle time independent tests, and Phase 2 performs propagation only to those identified tests. This segmentation reduces runtime in Phase 2 while maintaining complete coverage of independent tests across both phases.
Solution Approach 2:
The first phase serves as a preliminary action that identifies the subset of tests requiring detailed analysis. By performing this identification first, the second phase can focus computational resources only on cycle time independent tests, significantly reducing overall runtime while preserving complete coverage.
3Measurement precision
If statistical timing propagation is performed for all signal labels, then complete timing data is obtained, but wasteful calculations increase
Solution Approach 1:
The patent extracts and identifies only the cycle time independent tests from the complete set of timing tests. By taking out this specific subset, the system performs statistical timing propagation only where necessary, eliminating wasteful calculations on cycle time dependent tests while maintaining complete and precise timing data for the independent tests.
Data Source
AI summary
A computer program product for performing selected timing comparisons in a digital electronic design includes propagating from signal sources to timing comparisons of one or multiple signal labels. The signal label includes signal source identifiers and signal path cycle adjust information. Timing comparisons are determined in which signal label values at each input of the timing comparison are required to compute the selected timing comparisons. The propagation back from the timing comparisons are needed signal labels, followed by the propagation and computing timing data from the signal source applied to the propagated signal labels corresponding to the required signal labels.