HolBA SCAM V tutorial

We provide an introductory VM for SCAM-V (Validation of Abstract Side-Channel Models for Computer Architectures).

Getting started

1. Download and import the VM (see below).
2. Boot and login with user scamv (pw: scamv).
3. The file README.html as well as a terminal will open by itself.
4. In the terminal, the current working directory is ~/scamv, where you find HolBA/SCAM-V and the data directories, together with the directory introduction.
5. The file README.md (source for README.html) resides in introduction together with basic scripts to easily run the tools and reproduce the results.
6. Read the import information below before continuing!

IMPORTANT INFORMATION

• Generating and running the experiments in parallel is possible by importing the VM multiple times under different names.
• Generating all our experiments may take several days, even if run in parallel.
• Running the generated experiments is the real bottleneck and requires a somewhat special hardware setup, which can be laboriously replicated from the documentation. Alternatively, one can ask us for remote access to our existing setup.

Directories and documentation

• HolBA

  • Binary analysis tool HolBA with SCAM-V embedded.
  • https://github.com/kth-step/HolBA/blob/master/README.md

• HolBA/src/tools/scamv

  • SCAM-V libraries and entry point.
  • https://github.com/kth-step/HolBA/tree/master/src/tools/scamv/README.md

• HolBA/src/tools/scamv/examples

  • SCAM-V configurations and run scripts.
  • https://github.com/kth-step/HolBA/blob/master/src/tools/scamv/examples/README.md

• HolBA_opt

  • HolBA/SCAM-V dependencies.

• HolBA_opt/embexp/EmbExp-Box

  • Infratsructure for remotely/locally connecting to a board (e.g., Raspberry Pi 3).
  • https://github.com/kth-step/EmbExp-Box/blob/master/README.md

• HolBA_opt/embexp/EmbExp-ProgPlatform

  • Software platform for executing low-level hardware experiments (e.g., on Raspberry Pi 3).
  • https://github.com/kth-step/EmbExp-ProgPlatform/blob/master/README.md

• HolBA_logs/EmbExp-Logs

  • Experiment repository for the paper with one git branch per experiment set.
  • https://github.com/kth-step/EmbExp-Logs/blob/master/README.md

VM image

• SHA256: c3325cacf70b90b8f8b91ff7dbab76103c35f9ed2f11220b709939ffc34f11a7
• mirror1
• mirror2

Image parameters

• VM type: VirtualBox (.ova)
• Size: about 10GB
• OS: Debian 10.3.0 x64
• RAM: 2GB
• CPU cores: 2

Tested host platform parameters

• OS: Windows 10 Pro 1909
• VirtualBox: 6.0.18
• RAM: 16GB
• CPU: Intel Core i7-6650U
• CPU cores: 2 (4 threads)
• CPU frequency: 2.2GHz (3.4Ghz turbo)