error: a host function call cannot be configured

Hello,

I write a cu file for solving least squares with the following structure:

global
void d_getLS(int cL, int cT, int cI, float *d_sData, float *d_uData, float *d_Mat)
{

}

/**

  • Runs the algorithm.
    */
    void OPS::run() {
    cudacall(cudaMemcpy(d_sData, h_sData, pow(cL,2) * nI * sizeof(float), cudaMemcpyHostToDevice));
    cudacall(cudaMemcpy(d_uData, h_uData, pow(cL,2) * nI * sizeof(float), cudaMemcpyHostToDevice));
    cudacall(cudaMalloc(&d_Mat, nIt * cT * sizeof(float *)));

     ...
    

    d_getLS <<<grid, blockDim>>> (cL, cT, cI, d_sData, d_uData, d_Mat);


    }

and obtain the following error (at the position d_getLS) when compiling:
error: a host function call cannot be configured

1 error detected in the compilation of “/tmp/tmpxft_000024e2_00000000-7_bBBb.cpp1.ii”.

I hope I have been sufficiently clear to make the problem understandable.