Awesome

Platypus Attack - Educational Samples

What is the Platypus Attack

The Platypus attack is a collection of software-based power side-channel attacks, published in November 2020. This repository provides samples for the attacks discussed in the actual paper.

For more information please visit the official site. Or you can find the paper at here.

Note

All these examples were tested on Intel CPUs, AMD CPUs also expose the RAPL interface, and therefore, these examples should work on AMD as well. Except the KASLR break which is using TSX.

Setup

These samples use the exposed RAPL interface, a Linux-based OS with the PowerCap framework installed. The presence of the interface can be tested by executing:

sudo cat /sys/class/powercap/intel-rapl:0/energy_uj 

If the above results in error, please execute the following command:

sudo apt update
sudo apt install powercap-utils 

All of the provided samples contain Makefiles to build the example. Therefore a basic compiler toolchain (gcc, g++, ld, make) must be installed.

If no toolchain is installed, execute:

sudo apt update
sudo apt install build-essential

Some of the examples provide a visual representation of the data by executing a python script. Therefore, make sure python3 is installed:

sudo apt update
sudo apt install python3
sudo apt install pip3

The plot scripts use numpy and matplotlib, install these with:

python3 -m pip install numpy
python3 -m pip install matplotlib
 

Example Summary

Ex1

Get in touch with the measurement interface.

Ex2

Compare the energy consumption of different instructions.

Ex3

Compare the energy consumption of different data operands.

Ex4

A basic covert channel which measures the energy periodically in the background.

Ex5

Kernel Address Space Layout randomization break using TSX. (This example only works when the CPU supports TSX)

Ex6

Using a simulation of SGX-Step to extract secret information of a square and multiply algorithm, which is non constant time.

Ex7

Using a simulation of SGX-Step to extract secret information of a square and multiply algorithm, which is constant time.