Structure of Array How to allocate and use it?

Accelerated Computing CUDA CUDA Programming and Performance
#1

Hi,I’m a newbie in cuda programming and I want to allocate and process a structure of array; in the structure there are two 2D array representing an image in its real and imaginary part. I’ve tried this but it doesn’t work (segmentation fault :ph34r: ):

  size_t size1, size2;
  int BLOCK_SIZE;
  int GRID_SIZE;

  BLOCK_SIZE=512;
  GRID_SIZE=8;
   
  fcomplex *rcDev;
  fftw_complex *tfftwDev;
  size1 = nvp*nazft* sizeof(fcomplex); 
  size2 = nazft* sizeof(fftw_complex);
  cudaMalloc((void**)&rcDev,size1);
  cudaMalloc((void**)&tfftwDev,size2);
  cudaMemcpy(rcDev, rc, size1, cudaMemcpyHostToDevice);
  cudaMemcpy( tfftwDev, tfftw, size2, cudaMemcpyHostToDevice );
  kernelForwardAzFFT_2 <<< dimGrid, dimBlock >>>( rcDev, tfftwDev , i, nazft);
  cudaMemcpy(rc,rcDev,size1,cudaMemcpyDeviceToHost);
  cudaFree(rcDev);	
  cudaFree(tfftwDev);

where “i” nvp and nazft are fixed, the kernel i used is:

global void kernelForwardAzFFT_2(fcomplex* rcDev, fftw_complex* tfftwDev, int i, int nazft)
{
int j=blockIdx.x * blockDim.x+threadIdx.x;

    rcDev[i*nazft+j].re = tfftwDev[j].re;
    rcDev[i*nazft+j].im = tfftwDev[j].im;

}

I think it may be a trivial question, please help me.