Function transpose

src/backend/cuda/transpose.cpp:23–36

template<typename T>
Array<T> transpose(const Array<T> &in, const bool conjugate) {
    const dim4 &inDims = in.dims();

    dim4 outDims = dim4(inDims[1], inDims[0], inDims[2], inDims[3]);

    Array<T> out = createEmptyArray<T>(outDims);

    const bool is32multiple =
        inDims[0] % kernel::TILE_DIM == 0 && inDims[1] % kernel::TILE_DIM == 0;

    kernel::transpose<T>(out, in, conjugate, is32multiple);

    return out;
}

#define INSTANTIATE(T) \
    template Array<T> transpose(const Array<T> &in, const bool conjugate);