2.1. Prerequisites

2.1.1. Build Host

A relatively recent Linux distribution is recommended. CI runs are done using Ubuntu 22.04 LTS (64-bit).

2.1.2. Toolchain

The clang-20 toolchain is recommended for building Hafnium and the test infrastructure. On Ubuntu, the toolchain can be installed from the LLVM apt repository (https://apt.llvm.org/).

Note

Using a toolchain version greater than, or significantly lesser than the one specified is not guaranteed to work.

Note

You may also use the Docker container if you are unable to install the toolchain on your host machine, see the Using Docker section.

2.1.3. Dependencies

2.1.3.1. Build

The following command install the dependencies for the Hafnium build:

sudo apt install make libssl-dev flex bison python3 python3-serial python3-pip device-tree-compiler

In addition, install the following python libraries using pip:

pip3 install fdt click

The file kokoro/static_checks.sh runs a series of static code checks into Hafnium’s codebase. Hafnium follows the linux kernel coding guidelines. As such, the static code checks using the ‘checkpatch.pl’ script from linux source tree. To setup and download ‘checkpatch.pl’:

./build/setup_checkpatch.sh

Then test it works with:

make checkpatch

2.1.3.2. System and Python Packages for test infrastructure

The Shrinkwrap tool is a core component of the hftest scripts and is included as a submodule. It relies on the following system packages to function correctly:

sudo apt-get install -y git netcat-openbsd python3 python3-pip telnet

Further, install the required Python packages:

pip3 install --user fdt pyyaml termcolor tuxmake

Note

These manual installation steps are not required if you’re using the Docker based test environment. All necessary dependencies—system packages, Python modules, and Shrinkwrap environment configuration—are already pre-installed and automated inside the Docker image provided by the Hafnium repository (see Using Docker).

2.1.3.3. Documentation

To create a rendered copy of this documentation locally you can use the Sphinx tool to build and package the plain-text documents into HTML-formatted pages.

For building a local copy of the documentation you will need:

Python 3 (3.8 or later)
PlantUML (1.2017.15 or later)
Poetry (Python dependency manager)

Below is an example set of instructions to get a working environment (tested on Ubuntu):

sudo apt install python3 python3-pip plantuml
curl -sSL https://install.python-poetry.org | python3 -

Run the command below to install using Poetry, Python dependencies to build the documentation:

poetry install --with docs

Poetry will create a new virtual environment and install all dependencies listed in pyproject.toml. You can get information about this environment, such as its location and the Python version, with the command:

poetry env info