🚀 ttperf - TT-Metal Performance Profiler

A streamlined CLI tool for profiling Tenstorrent's TT-Metal tests and extracting device kernel performance metrics

✨ Features

• 🔍 Automated Profiling: Seamlessly runs Tenstorrent's TT-Metal profiler with pytest
• 📊 CSV Analysis: Automatically extracts and parses performance CSV files
• ⚡ Real-time Output: Shows profiling progress in real-time
• 📈 Performance Metrics: Calculates total DEVICE KERNEL DURATION
• 🎯 Simple CLI: Easy-to-use command-line interface
• 🛠️ Flexible: Supports named profiles and various test paths
• 🚀 Operation-based Profiling: Profile specific operations by name (e.g., ttperf add)
• ⚙️ Dynamic Configuration: Customize tensor shape, dtype, and layout for operations

🚀 Quick Start

Installation

# Install from PyPI (recommended)
pip install ttperf

Or install from source:

# Clone the repository
git clone https://github.com/Aswintechie/ttperf.git
cd ttperf

# Install the package
pip install -e .

Configuration

ttperf automatically searches for your TT-Metal installation using a simple two-step process:

# Option 1: Set PYTHONPATH to your tt-metal location
export PYTHONPATH=/path/to/your/tt-metal
ttperf add

# Option 2: Run from within tt-metal directory (or any subdirectory)
cd /path/to/your/tt-metal
ttperf relu
# Or from a subdirectory
cd /path/to/your/tt-metal/tests
ttperf matmul

tt-metal Path Search Order:

1. PYTHONPATH environment variable (if specified)
2. Current working directory (walks up directory tree to find tt-metal root)

Basic Usage

# Run profiling on a specific test
ttperf test_performance.py

# Run with a custom profile name
ttperf my_profile pytest test_performance.py

# Run on a specific test method
ttperf tests/test_ops.py::test_matmul

# Profile specific operations by name
ttperf add
ttperf relu
ttperf matmul

# Profile operations with custom profile names
ttperf my_add_profile add
ttperf my_relu_profile relu

# Profile operations with custom configuration
ttperf add --shape 1,1,32,32 --dtype bfloat16 --layout tile
ttperf relu --shape 1,1,64,64 --dtype float32 --layout row_major

# Profile operations with memory configuration
ttperf add --dram                                # Use DRAM memory (default)
ttperf relu --l1                                 # Use L1 memory
ttperf add --shape 1,1,64,64 --l1                # Combined options

📋 Usage Examples

Test File Profiling

ttperf test_conv.py

Named Profile

ttperf conv_benchmark pytest test_conv.py

Specific Test Method

ttperf tests/ops/test_matmul.py::test_basic_matmul

Operation-based Profiling

# Basic operations
ttperf add
ttperf subtract
ttperf multiply
ttperf divide

# Activation functions
ttperf relu
ttperf sigmoid
ttperf tanh
ttperf gelu

# Mathematical operations
ttperf sqrt
ttperf exp
ttperf log
ttperf sin
ttperf cos

# Comparison operations
ttperf gt
ttperf lt
ttperf eq
ttperf ne

# Reduction operations
ttperf max
ttperf min
ttperf mean
ttperf sum

# Backward operations
ttperf add_bw
ttperf relu_bw
ttperf sigmoid_bw

Dynamic Configuration

# Custom tensor shape
ttperf add --shape 1,1,32,32
ttperf relu --shape 2,3,64,128

# Custom data type
ttperf add --dtype float32
ttperf multiply --dtype int32

# Custom memory layout
ttperf add --layout row_major
ttperf relu --layout tile

# Combined configuration
ttperf add --shape 1,1,64,64 --dtype float32 --layout row_major
ttperf gelu --shape 2,1,32,32 --dtype bfloat16 --layout tile

# Memory configuration options
ttperf add --memory-config dram                  # Explicit DRAM
ttperf relu --memory-config l1                   # Explicit L1  
ttperf add --dram --shape 1,1,128,128            # DRAM with custom shape
ttperf relu --l1 --dtype float32                 # L1 with custom dtype

List All Supported Operations

ttperf --list-ops
# or
ttperf -l

Output Example

🔧 Using custom configuration:
   Shape: (1, 1, 32, 32)
   Dtype: bfloat16
   Layout: tile
🏷️ Auto-generated profile name: temp_test_add
▶️ Running: ./tools/tracy/profile_this.py -n temp_test_add -c "pytest temp_test_add.py"

... (profiling output) ...

📁 Found CSV path: /path/to/profile_results.csv
⏱️ DEVICE KERNEL DURATION [ns] total: 1234567.89 ns

🛠️ How It Works

1. Command Parsing: Analyzes input arguments to determine profile name and test path/operation
2. Operation Detection: If an operation name is provided, maps it to the corresponding test method
3. Dynamic Configuration: If custom configuration is provided, generates a temporary test file with the specified parameters
4. Profile Execution: Runs the Tenstorrent's TT-Metal profiler with the specified test
5. Output Monitoring: Streams profiling output in real-time
6. CSV Extraction: Parses the output to find the generated CSV file path
7. Performance Analysis: Reads the CSV and calculates total device kernel duration

📊 Performance Metrics

The tool extracts the following key metrics:

• DEVICE KERNEL DURATION [ns]: Total time spent in device kernels
• CSV Path: Location of the detailed profiling results
• Real-time Progress: Live output during profiling

⚙️ Configuration Options

Shape Configuration

• Format: Comma-separated integers (e.g., 1,1,32,32)
• Default: (1, 1, 1024, 1024)
• Example: --shape 2,3,64,128

Data Type Configuration

• Valid Options: bfloat16, float32, int32
• Default: bfloat16
• Example: --dtype float32

Layout Configuration

• Valid Options: tile, row_major
• Default: tile
• Example: --layout row_major

🔧 Requirements

• Python 3.8+
• pandas
• Tenstorrent's TT-Metal development environment
• pytest

📁 Project Structure

ttperf/
├── ttperf.py          # Main CLI implementation
├── pyproject.toml     # Project configuration
├── README.md          # This file
└── .gitignore         # Git ignore rules

🤝 Contributing

Contributions are welcome! Please feel free to submit a Pull Request.

1. Fork the repository
2. Create your feature branch (git checkout -b feature/amazing-feature)
3. Commit your changes (git commit -m 'Add some amazing feature')
4. Push to the branch (git push origin feature/amazing-feature)
5. Open a Pull Request

📝 License

This project is licensed under the MIT License - see the LICENSE file for details.

⚠️ Disclaimer

This tool is an independent utility that interfaces with Tenstorrent's TT-Metal profiling tools. It is not affiliated with or endorsed by Tenstorrent Inc. The tool serves as a convenience wrapper around existing TT-Metal profiling infrastructure.

🐛 Issues

If you encounter any issues, please create an issue on GitHub.

👨‍💻 Author

Aswin Z

• GitHub: @Aswintechie
• Portfolio: aswinlocal.in

🌟 Acknowledgments

• Tenstorrent's TT-Metal development team for the profiling tools
• Python community for excellent libraries like pandas

---

Made with ❤️ for the Tenstorrent TT-Metal community