This repository contains the scripts that can be used to reproduce the 3 tasks contained in the paper: Differential cryptanalysis with SAT, SMT, MILP, and CP: a detailed comparison for bit-oriented primitives.

Tests are completely automatized and csv files will be produced as output.

• CLAASP: the tests contained in this repository are intended to be run using the Cryptographic Library for the Automated Analysis of Symmetric Primitives. In order to be updated with the last fixes, clone the repository:

  $ git clone https://github.com/Crypto-TII/claasp

• Get this repository: once you have downloaded the CLAASP repository, you must clone this repository:

  $ cd claasp
  $ git clone https://github.com/Crypto-TII/constrain-solvers-comparison_of_bit-oriented_primitives.git scripts

  Note that the second command will clone the repository into a directory called scripts.
• Installing the library: CLAASP is built on top of SageMath, therefore you must install it first. You can install all the dependencies in your machine or use a Docker image. For a quick start, read docs/USER_GUIDE.md. The latter is the preferable method.
• You are ready to go. Inside the container created for CLAASP, type sage scripts/<test-name>. You should be able to see logs and output files in the scripts directory.

The tasks are the following ones:

1. the task 1 is the search for an optimal differential trail; note that there are easy and difficult instances;
2. the task 2 is the enumeration of all optimal trails;
3. the task 3 is the estimation of the probability of a differential.

Every task has its own single model version and multiple model one. If you have enough computational resources, you can go straight for the multiple one.

• Single model - in order to run tests for a single model just type:

  # sage scripts/task_STRING.sage -m MODEL

  • (mandatory) replace STRING with 1_easy, 1_difficult, 2 or 3;
  • (optional) replace MODEL with sat, smt, milp or cp. The default is sat.
• Multiple models - if you have enough computational resources, you can go straight for parallelization:

  # sage scripts/task_STRING_parallelization.sage -m MODELS

  • (mandatory) replace STRING with 1_easy, 1_difficult, 2 or 3;
  • (optional) replace MODELS with models you want to test. For instance, say that you want just tests for SAT and CP, you need to run:

    # sage scripts/task_STRING_parallelization.sage -m sat cp

    The defaults are all four models.

In order to not create too many containers, instead of running the CLAASP Makefile recipes, create an image:

$ docker build -f docker/Dockerfile -t claasp .

create a container using the image just created:

$ docker create -i --name tester -v `pwd`:/home/sage/tii-claasp -t claasp /bin/bash

and start the container:

$ docker start -ia tester

You have now to install the library. Since at this point you are inside the Docker container, you can run:

# make install

If you exit the container and need to go back to it, you just need to type the Docker start command. In the case you have pulled the CLAASP repository for updating purposes and the Dockerfile has not been changed, you can safely start the tester container and re-run a make install command.