Question

This code compiled properly on Visual Studio 2010 on Windows, but I’m getting this error on Linux, g++. Could anyone please explain how to fix this?

int bits;
T scale;
std::vector<U> TwoPowers;
U cropper;

template <typename T, typename U>
ConvertToBits<T,U>::ConvertToBits(int bits, T scale)
{
    this->bits = bits;
    this->scale = scale;
    for(long i = 0; i < 64; i++)
    {
        TwoPowers.push_back(static_cast<U>(pow(2.,i))); //error appears here
    }
    cropper = 0;
    for(int i = 0; i < bits; i++)
    {
        cropper += TwoPowers[i];
    }
}

Error message:

error: there are no arguments to ‘pow’ that depend on a template parameter, so a declaration of ‘pow’ must be available [-fpermissive]

Thank you.

Answer 1

You need to #include <cmath> and also link in the math library with -lm (on some systems).