Reconvergence Integration - Quick Start

Get started with reconvergence-aware testability analysis in minutes.

What is Reconvergence Integration?

Version 1.0.1 automatically detects reconvergent fanout in circuits and adjusts COP/SCOAP metrics for improved accuracy. No manual setup required!

5-Minute Quick Start

1. One-Line COP Analysis

opentest auto-cop -i circuit.txt

That’s it! The tool automatically:

Creates DAG
Detects reconvergence
Runs COP with correlation adjustments
Enables parallel mode if needed

2. One-Line SCOAP Analysis

opentest auto-scoap -i circuit.txt

Automatically:

Creates DAG
Extracts fanout points
Runs SCOAP with reconvergence awareness
Processes cones in parallel

3. Python API

from opentestability.core.cop import run_cop

# Everything automatic!
run_cop('circuit.txt', 'results.txt',
        reconvergence_algorithm='auto',
        auto_parallel=True)

Key Benefits

Improved Accuracy

98%+ accurate reconvergence detection
Bayesian correlation adjustments
Better testability metrics at fanout points

Better Performance

3-4x speedup on large circuits (>100k gates)
Automatic parallel activation
Concurrent processing of reconvergent cones

Zero Configuration

Works out of the box
Intelligent algorithm selection
No manual DAG creation needed

Algorithm Selection

Auto (Recommended)

run_cop('circuit.txt', 'results.txt',
        reconvergence_algorithm='auto')  # Default

Let the tool choose the best algorithm.

Simple (Fast & Accurate)

run_cop('circuit.txt', 'results.txt',
        reconvergence_algorithm='simple')

98%+ accuracy
Fast execution
Best for most circuits

Advanced (Research-Grade)

run_cop('circuit.txt', 'results.txt',
        reconvergence_algorithm='advanced')

100% accuracy
Complex pattern detection
Best for pipelined designs

Parallel Computation

Automatic (Recommended)

run_cop('circuit.txt', 'results.txt',
        auto_parallel=True)  # Default

Parallel mode activates automatically for circuits >100,000 gates.

Manual Control

# Custom worker count
run_cop('circuit.txt', 'results.txt',
        max_workers=8)

# Force sequential
run_cop('circuit.txt', 'results.txt',
        auto_parallel=False)

Common Workflows

Workflow 1: Quick Analysis

# Simplest - let tool handle everything
opentest auto-cop -i circuit.txt
opentest auto-scoap -i circuit.txt

Workflow 2: Large Circuit

# Specify worker count for better performance
opentest auto-cop -i large_design.txt -w 8
opentest auto-scoap -i large_design.txt -w 8

Workflow 3: Research Analysis

# Use advanced algorithm for complete detection
opentest cop -i circuit.txt -r -a advanced

Workflow 4: Python Batch Processing

from opentestability.core.reconvergence.integration import integrate_with_cop
import glob

for circuit in glob.glob('data/parsed/*.txt'):
    integrate_with_cop(
        input_file=circuit,
        output_file=circuit.replace('.txt', '_cop.txt'),
        algorithm='auto',
        auto_parallel=True
    )

Performance Expectations

Small Circuits (< 100k gates)

Sequential processing
Sub-second analysis
No parallel overhead

Large Circuits (> 100k gates)

Automatic parallel activation
3-4x speedup
Efficient multi-core utilization

Example Benchmarks

Circuit	Gates	Algorithm	Time (Sequential)	Time (Parallel)	Speedup
ALU	1,200	Simple	0.12s	N/A	1.0x
Medium	45,000	Simple	2.4s	N/A	1.0x
SoC Block	180,000	Simple	18.5s	5.2s	3.6x
ASIC	520,000	Simple	95.2s	24.1s	3.9x

Direct API Examples

Example 1: ReconvergenceIntegrator

from opentestability.core.reconvergence.integration import ReconvergenceIntegrator

integrator = ReconvergenceIntegrator('circuit_dag.json')
reconv_data = integrator.detect_reconvergence(algorithm='simple')

sites = integrator.extract_reconvergent_sites()
print(f"Found {len(sites)} reconvergent sites")

Example 2: Integration Functions

from opentestability.core.reconvergence.integration import (
    integrate_with_cop,
    integrate_with_scoap
)

# COP with reconvergence
integrate_with_cop('circuit.txt', 'cop_results.txt', 
                   algorithm='auto', auto_parallel=True)

# SCOAP with reconvergence
integrate_with_scoap('circuit.txt', 'scoap_results.json',
                     algorithm='auto', auto_parallel=True)

Example 3: Parallel COP

from opentestability.core.cop.parallel import build_cop_with_parallel

C1, Obs = build_cop_with_parallel(
    nets, inputs, outputs, gates,
    reconvergent_sites=['node5', 'node12'],
    max_workers=4
)

Comparison with Traditional Approach

Traditional (Multi-Step)

# Step 1: Parse Verilog
opentest parse -i circuit.v

# Step 2: Create DAG
opentest dag -i circuit.json

# Step 3: Detect reconvergence
opentest simple -i circuit_dag.json

# Step 4: Run analysis (manual reconvergence data)
opentest cop -i circuit.txt

New (Integrated)

# One command - everything automatic!
opentest auto-cop -i circuit.txt

Benefits:

4 commands → 1 command
No manual file management
Automatic optimization
Better accuracy

Migration from v2.0.0

Old Code (Still Works!)

from opentestability.core.cop import run_cop

run_cop('circuit.txt', 'results.txt')  # Works unchanged

New Code (Enhanced)

from opentestability.core.cop import run_cop

run_cop('circuit.txt', 'results.txt',
        reconvergence_algorithm='auto',  # Add this line
        auto_parallel=True)              # Add this line

No breaking changes - just add two parameters for enhanced features!

Command Line Cheat Sheet

# Quick analysis
auto-cop -i circuit.txt
auto-scoap -i circuit.txt

# With custom workers
auto-cop -i circuit.txt -w 8
auto-scoap -i circuit.txt -w 4

# Specific algorithm
cop -i circuit.txt -r -a simple
scoap -i circuit.txt -r -a advanced

# Verbose output
auto-cop -i circuit.txt -v
auto-scoap -i circuit.txt -v

# Run tests
test-reconv -v

Troubleshooting

Q: “DAG file not found”

A: Auto commands create DAG automatically. If using manual workflow, run:

opentest dag -i circuit.json

Q: “Reconvergence detection failed”

A: Try simple algorithm (most reliable):

opentest auto-cop -i circuit.txt -r -a simple

Q: “Parallel not improving performance”

A: Your circuit may be too small. Parallel activates at >100k gates. Check:

opentest auto-cop -i circuit.txt -v  # See circuit size

Q: “How to disable parallel?”

A: Use Python API:

run_cop('circuit.txt', 'results.txt', auto_parallel=False)

Next Steps

Learn More:

Get Help:

Stay Updated: