Getting Started

Archspec is a minimal-size python package that makes it easier to work with various aspects of a system architecture like CPU, network fabrics, etc. by providing:

• A set of human-understandable names for each entity
• Well defined APIs to query properties from objects or compare them

Note

This project grew out of Spack and is currently under active development. At present it only models CPU microarchitectures but extensions to other aspects are expected in the future.

System requirements

Archspec needs the following software:

Python	2.7 or 3.5+
six	^1.13.0
click	>=7.1.2,<8.0

It is a multiplatform project and currently works on linux and MacOS. Porting to Windows is expected in the future.

Installation from PyPI

Archspec is published on PyPI so to install any release you can simply use pip:

$ pip install archspec[==<required-release>]
$ python -c "import archspec; print(archspec.__version__)"
0.1.3

This is the simplest way to install the package and getting started using it.

Installing from GitHub Repository

Installing Archspec from a clone of its GitHub Repository requires poetry. The preferred method to install this tool is via its custom installer outside of any virtual environment:

$ curl -sSL https://raw.githubusercontent.com/python-poetry/poetry/master/get-poetry.py | python

You can refer to Poetry’s documentation for further details or for other methods to install it. Once poetry is available we can proceed with cloning the repository and installing Archspec:

$ which poetry
/home/user/.poetry/bin/poetry

$ git clone --recursive https://github.com/archspec/archspec.git
$ cd archspec

$ poetry install --no-dev
Creating virtualenv archspec-0fr1r4aA-py2.7 in /home/culpo/.cache/pypoetry/virtualenvs
Updating dependencies
Resolving dependencies... (17.7s)

Writing lock file


Package operations: 2 installs, 0 updates, 0 removals

  - Installing click (7.1.2)
  - Installing six (1.15.0)
  - Installing archspec (0.1.3)

$ poetry run python -c "import archspec; print(archspec.__version__)"
0.1.3

Poetry manages virtual environments for the user. Using poetry run is just one of the possibility offered by this tool, for further options you can refer to its documentation.

Install Development Dependencies

To install Archspec with its development dependencies you just need to omit the --no-dev option from the commands shown above:

$ git clone --recursive https://github.com/archspec/archspec.git
$ cd archspec

$ poetry install
Installing dependencies from lock file


Package operations: 21 installs, 0 updates, 0 removals

  [...]
  - Installing pytest (4.6.9)
  - Installing jsonschema (3.2.0)
  - Installing pytest-cov (2.8.1)
  - Installing archspec (0.1.3)

At this point you can run unit-tests, linters or other checks. When developing we recommend to use Python ^3.6 so that the latest versions of each development tool can be used:

$ poetry run pytest
============================================================== test session starts ===============================================================
platform linux -- Python 3.7.6, pytest-5.3.4, py-1.8.1, pluggy-0.13.1
rootdir: /home/culpo/tmp/archspec/docs-scratch/archspec
plugins: cov-2.8.1
collected 255 items

tests/test_archspec.py .                                                                                                                   [  0%]
tests/test_cpu.py ........................................................................................................................ [ 47%]
......................................................................................................................................     [100%]

============================================================== 255 passed in 0.73s ===============================================================

$ poetry run black --check archspec tests
All done! ✨ 🍰 ✨
9 files would be left unchanged.

$ poetry run pylint archspec
--------------------------------------------------------------------
Your code has been rated at 10.00/10 (previous run: 10.00/10, +0.00)