/**== Copy the contents of the F table inside the pF table ==**/
    for (i=0;i<mF;i++) {
		for(j=0;j<nF;j++){
			pF[(i+mT-1)*(nF + 2*nT -2)+nT-1+j]=F[i*nF+j];
		}
    }

	CUDA_SAFE_CALL(cudaMallocPitch((void **)&dev_pF,&pitchF,(nF + 2*nT -2)*sizeof(int),mF + 2*mT -2));

	
	CUDA_SAFE_CALL(cudaMemcpy2D((void *)(dev_pF),pitchF,(void *)pF,(nF + 2*nT -2)*sizeof(int),(nF + 2*nT -2)*sizeof(int),(mF + 2*mT -2),cudaMemcpyHostToDevice));

pF[i][j]

                        int *rF=(int *)((char *)pF + (pitchF)*(i));
                        rF[j];