Installation

Method 1: Docker Build (Recommended)

Docker provides a containerized environment with all dependencies pre-configured.

Prerequisites

• Docker installed on your system
• At least 8GB of available disk space

Build Steps

1. Clone the repository with submodules:

   git clone <repository-url>
   cd aieda_fork
   git submodule update --init --recursive
   

2. Build the Docker image:

   # Use the provided build script
   ./build_docker.sh
   
   # Or build manually
   docker build -f Dockerfile.base -t aieda:base .
   docker build -f Dockerfile.final -t aieda:latest .
   

3. Run the container:

   docker run --rm -it aieda:latest
   

4. Test the installation:

   docker run --rm aieda:latest python3 test/test_sky130_gcd.py
   

For detailed Docker instructions, refer to the build script and configuration files in the repository.

Method 2: UV Build (Development)

UV provides a fast Python package manager for local development.

Prerequisites

• Python 3.10 or higher
• UV package manager
• iEDA dependencies (C++ compiler, CMake, etc.)

Build Steps

1. Install UV:

   pip install uv
   

2. Verify UV installation:

   uv --version
   

3. Create and activate virtual environment:

   uv venv
   source .venv/bin/activate
   

4. Install dependencies:

   uv run python -m build
   

5. Build aieda package:

   uv build
   

6. Install aieda:

   uv pip install dist/aieda-0.1.0-py3-none-any.whl
   

7. Test the installation:

   uv run python test/test_sky130_gcd.py
   

iEDA Compilation (Required for UV build)

Before using AiEDA with UV, you need to compile the iEDA backend:

Compile iEDA:

mkdir build
cd build
cmake ..
make -j32 ieda_py