Home
TritonParse is a comprehensive visualization and analysis tool for Triton IR files, designed to help developers analyze, debug, and understand Triton kernel compilation processes.
- Installation - Complete setup instructions
- Quick Start Tutorial - Your first TritonParse experience
- System Requirements - Prerequisites and compatibility
- Usage Guide - Generate traces and analyze kernels
- Web Interface Guide - Master the visualization interface
- File Formats - Understanding input/output formats
- Troubleshooting - Common issues and solutions
- Architecture Overview - System design and components
- API Reference - Python API documentation
- Contributing - Development setup and guidelines
- Code Formatting - Formatting standards and tools
- Source Mapping - IR stage mapping explained
- Environment Variables - Configuration options
- Performance Optimization - Tips for large traces
- Custom Deployments - Self-hosting and customization
- Basic Examples - Simple usage scenarios
- Advanced Examples - Complex use cases
- FAQ - Frequently asked questions
- Glossary - Technical terms and definitions
- Interactive Kernel Explorer - Browse kernel information and stack traces
- Multi-format IR Support - View TTGIR, TTIR, LLIR, PTX, and AMDGCN
- Side-by-side Comparison - Compare IR stages with synchronized highlighting
- Interactive Code Views - Click-to-highlight corresponding lines
- Compilation Tracing - Capture detailed Triton compilation events
- Stack Trace Integration - Full Python stack traces for debugging
- Metadata Extraction - Comprehensive kernel metadata and statistics
- NDJSON Output - Structured logging format for easy processing
- GitHub Pages - Ready-to-use online interface
- Local Development - Full development environment
- Standalone HTML - Self-contained deployments
# Clone the repository
git clone https://github.com/pytorch-labs/tritonparse.git
cd tritonparse
# Install dependencies
pip install -e .import tritonparse.structured_logging
# Initialize logging
tritonparse.structured_logging.init("./logs/")
# Your Triton/PyTorch code here
...
# Parse logs
import tritonparse.utils
tritonparse.utils.unified_parse(source="./logs/", out="./parsed_output")Visit https://pytorch-labs.github.io/tritonparse/ and load your trace files!
Frontend: React 19, TypeScript, Vite, Tailwind CSS, Monaco Editor Backend: Python, Triton integration, structured logging Deployment: GitHub Pages, local development server
- Live Tool: https://pytorch-labs.github.io/tritonparse/
- GitHub Repository: https://github.com/pytorch-labs/tritonparse
- Issues: GitHub Issues
- Discussions: GitHub Discussions
We welcome contributions! Please see our Contributing Guide for details on:
- Development setup
- Code formatting standards
- Pull request process
- Issue reporting
This project is licensed under the BSD-3 License. See the LICENSE file for details.
Note: This tool is designed for developers working with Triton kernels and GPU computing. Basic familiarity with CUDA, GPU programming concepts, and the Triton language is recommended for effective use.