Paper

y2k38-checker is a tool that identifies and reports code with potential Year 2038 problem issues in C language source code.

Requirements:

• Docker / Docker Compose
• OS: Ubuntu (TODO: macOS / Windows)

1. Download the releases.

2. Unzip the downloaded file.

   unzip y2k38-checker-<version>.zip

   Then, the following directory structure is created.

   y2k38-checker/
   ├── checker/
   │ ├── bin/y2k38-checker  # detection tool binary
   │ ├── lib/               # Clang plugin library
   │ └── scripts/           # scripts for running the detection tool
   ├── dataset/             # example for C source code
   ├── volumes/             # target source code
   └── .devcontainer/
     ├── Dockerfile
     ├── docker-compose.yml
     └── devcontainer.json
   

3. Add the path of the created the directory in .devcontainer/docker-compose.yml

   services:
      y2k38-checker-app:
         build:
            context: ..
            dockerfile: .devcontainer/Dockerfile
         tty: true
         volumes:
            - ..:/root/y2k38-checker/volumes/
            - type: bind
   -          source: /home/cysec/develop/.y2k38-checker/analysis-objects/
   +          source: <path/to/dir>
            target: /root/analysis-objects

4. Build & Run the docker container with CLI or DevContainer

   cd y2k38-checker
   docker-compose build # only first time
   docker-compose run y2k38-checker

   Alternatively, start it in the devcontainer of VSCode.

5. Run the detection tool with the following command.

Check the source code in the volumes/ directory with the detection tool. If the option --consent is specified, the analysis results will be used for research purposes.

python3 run.py --consent

python3 ./checker/script/run.py --help
# Usage: python3 run.py [option]

# Options:
#   -h, --help     Print this help message and exit
#   --consent <Yes/No>      Consent: Agree to the use of analysis results in our research

./checker/bin/check-y2k38 --help

# Usage: check-y2k38 [options] <source0>
#
# Options:
#   -h, --help                      - Print this help message and exit
#   -v, --version                   - Print the version number and exit
#   -p=<build-path>                 - Path to a compile_commands.json file
#
# Example:
#   ./check-y2k38 -p=path/to/compile_commands.json

clang -fplugin=./lib/libread-y2k38-checker-plugin.so -c <file>.c

1. Clone the repository

   git clone https://github.com/cysec-lab/y2k38-checker.git

2. Create the directory for the detecting target source code, and add files to be analyzed.

   mkdir <path/to/dir>
   cp -r <files/to/be/analyzed> <path/to/dir>

3. Download LLVM library

   cd ./checker/
   curl -L https://github.com/llvm/llvm-project/releases/download/llvmorg-11.0.0/clang+llvm-11.0.0-x86_64-linux-gnu-ubuntu-20.04.tar.xz | tar -Jxf -

   • https://github.com/llvm/llvm-project/releases/tag/llvmorg-11.0.0

4. Add the path of the created the directory in .devcontainer/docker-compose.yml

   services:
      y2k38-checker-app:
         build:
            context: ..
            dockerfile: .devcontainer/Dockerfile
         tty: true
         volumes:
            - ..:/root/y2k38-checker/volumes/
            - type: bind
   -          source: /home/cysec/develop/.y2k38-checker/analysis-objects/
   +          source: <path/to/dir>
            target: /root/analysis-objects

5. Build & Run the docker container with CLI or DevContainer

   cd y2k38-checker
   docker-compose build # only first time
   docker-compose run y2k38-checker

   Alternatively, start it in the devcontainer of VSCode.

1. Move to the checker/ directory

   cd ./checker

2. Build with CMake

   mkdir build
   cd ../y2k38-checker/build
   cmake -DCMAKE_EXPORT_COMPILE_COMMANDS=True \
      -DLLVM_DIR=../clang+llvm-11.0.0-x86_64-linux-gnu-ubuntu-20.04/lib/cmake/llvm/ \
      ../clang-analyzer
   make

   Then, the plugin library is created in the build/lib directory.

TODO