Well, we can run TLA+ in a headless mode without GUI. It is required to
and install Java package. Running TLA+ in command line:
java -Xmx30G -cp ../tla2tools.jar tlc2.TLC MC -deadlock -workers 12. The
-Xmx30G denotes the amount of memory to allocate to the model checker
-workers 12 the number of worker threads (should be equal to the number
of cores on machine). The setting
-deadlock ensures that TLC explores the
full reachable state space, not searching for deadlocks.
Using TLA+ in console is inconvenient a bit. We will use it with tlaplus_jupyter. It is a plugin for Jupyter, popular system for interactive computations. By default it supports only Python kernel, but it is possible to extend it with kernels that add support for other programming languages.
Create Python virtual environment and install
tlaplus-jupyter package in it:
$ python -m venv tlaplus $ . tlaplus/bin/activate (tlaplus_env) $ pip install tlaplus_jupyter (tlaplus_env) $ python -m tlaplus_jupyter.install Installing Jupyter kernel spec Downloading tla2tools.jar to /home/sergeyb/Downloads/tlaplus_env/lib/python3.7/site-packages/tlaplus_jupyter/vendor/tla2tools.jar (tlaplus_env) $ jupyter notebook
Now we are ready to start Jupyter notebook:
(tlaplus_env) $ jupyter notebook
To create a new TLA+ notebook click on the New button and select TLA+ in a dropdown menu. It is also handy to enable line numbering inside cells (View -> Toggle Line Numbers) since syntax checker refers to problems by their line numbers.
Learn more about TLA+