py-pixi 0.0.1

A Python package for program analysis and custom instruction generation

Pixi (Pass for Instruction eXtraction and Inference)

Pixi is a tool for analyzing LLVM IR and generating custom hardware instructions to optimize performance-critical code sections in the von-Neumann pipeline. The project follows the idea of everything is computation both hardware and software but they come with different computational cost models.

Requirements

• Python 3.10+
• LLVM and Clang (automatically set up by the installation process)
• Ninja build system
• CMake 3.13.4+

Installation

# Clone the repository
git clone https://github.com/JoyenBenitto/Pixi.git
cd Pixi

# Install the package in development mode
pip install -e .

# Set up LLVM dependencies
pixi setup -c <jobs>

[!Note] The setup process clones and configures the LLVM project, which may take some time during the initial setup. Subsequent setups will be much faster.

Usage

Pixi provides a comprehensive command-line interface for analyzing code, generating LLVM IR, and creating custom hardware instructions.

Command Reference

Command	Description
setup	Sets up the LLVM environment for Pixi

License

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