LibSVF -- A Library for computing SVF

Library to help in computing Side-channel Vulnerability Factor, a metric defined in "Side-channel Vulnerability Factor: A Metric for Measuring Information Leakage", published in ISCA 2012.

http://www.cs.columbia.edu/~simha/preprint_isca12_svf.pdf

Compilation

Compilation is not necessary. LibSVF is nearly all templates, so it is contained entirely in header files. You can compile and run the unit tests with scons.

Usage

Usage examples for libsvf can be found in tests/*.cpp. As a small example, here is a very basic usage using scalars as the data collection points:

#include <svf/svf.hpp>

struct ScalarDist {
    template <typename T>
    double operator()(T a, T b) {
        assert(!isnan(a));
        assert(!isnan(b));
        return fabs(a - b);
    }
};

// The first two specify the oracle data type and distance function.
// The last two are the same but for the side-channel trace.
SVF::SVF<int, ScalarDist, int, ScalarDist> svf;

for (size_t i=0; i<1000; i++) {
    svf.pushTimestep(i, i);
}

double svfVal = svf.computeSVF();
// This will be 1.0 since points are the same

Here's an example of sparse vectors and doubles as data points. In this example, the "attacker" can see only the length of the oracle vector, but there is still some correlation:

#include <svf/svf.hpp>
#include <svf/sparse_vector.hpp>

struct ScalarDist {
    template <typename T>
    double operator()(T a, T b) {
        assert(!isnan(a));
        assert(!isnan(b));
        return fbs(a - b);
    }
};

// ...

SVF::SVF<SVF::SparseVector<double>, SVF::SparseVector<double>::EuclideanDistance<>,
         double, ScalarDist> svf; 

for (size_t i=0; i<100; i++) {
    // Just the length seems a decent approximation for 6
    // dimensions
    SVF::SparseVector<double> a = randVec(6);
    svf.pushTimestep(a, a.length());
}

double svfVal = svf.computeSVF();
// This will be between 0 and 1

Documentation

Yeah, right! Read the code and the unit tests.

Credits

SVF research is conducted at the Computer Architecture and Security Technologies Lab in the Computer Science department at Columbia University in the city of New York. The original inventors of SVF are John Demme, Robert Martin, Adam Waksman and Simha Sethumadhavan. Their work is/was supported by grants FA 99500910389 (AFOSR), FA 865011C7190 (DARPA), FA 87501020253 (DARPA), CCF/TC 1054844 (NSF) and gifts from Microsoft Research, WindRiver Corp, Xilinx and Synopsys Inc. Opinions, findings, conclusions and recommendations expressed in this material are those of the authors and do not necessarily reflect the views of the US Government or commercial entities.

LibSVF code written and occasionally maintained by John Demme jdd@cs.columbia.edu.