This directory contains minimal, standalone proof-of-concept programs to showcase the different attack primitives described in Section 3 of the paper and summarized in Table 1.
Attack primitives are organized in hierarchical subdirectories as per Table 1 and MSP430 device number.
The subdirectory for the novel controlled call corruption primitive also includes relevant negative results discussed in the paper.
Refer to the top-level README on how to import and run the CCS projects with IPE enabled.
Every project contains a README with the expected output. For the architectural attack primitives, this should print the secret value 0x1337
(whereas reading directly yields the dummy value 0x3fff
):
Secret directly: 3fff, storage: 1337
For the side-channel attack primitives, the expected output shows that different behaviors can be distinguished.