Understanding of CUDA's cudaGetSymbolAddress - memory

Question

I am new to CUDA, now I'm trying to understand how cudaGetSymbolAddress works. I get an unexpected Segmentation Fault in a really simple code. What I do is the following:
I declare a global device variable(device_int)
In main() I ensure that the definition was correct by setting its value in a kernel
I create a pointer (host_pointer_to_device_int) in host memory and make it point to device_int via cudaGetSymbolAddress
I create one more pointer (host_pointer_to_host_int) and try to cudaMemcpy the value from host_pointer_to_device_int to host_pointer_to_host_int
All these operations finish with no errors, but I get Segmentation Fault when trying to print the host_pointer_to_host_int's value. Here is the code:
#include <iostream>
#include <cassert>
using namespace std;
__device__ int device_int;
__global__ void kernel()
{
device_int = 1000;
}
int main()
{
kernel<<<1, 1>>>();
assert(cudaGetLastError() == cudaSuccess); // The above operation executed successfully
int *host_pointer_to_device_int;
cudaGetSymbolAddress((void **)&host_pointer_to_device_int, device_int);
assert(cudaGetLastError() == cudaSuccess); // The above operation executed successfully
int *host_pointer_to_host_int;
// Copy the device_int's value
cudaMemcpy((void **)&host_pointer_to_host_int, host_pointer_to_device_int,
sizeof(int), cudaMemcpyDeviceToHost);
assert(cudaGetLastError() == cudaSuccess); // The above operation executed successfully
cout << *host_pointer_to_host_int << endl; // Segmentation fault
}

Answer 1

My mistake was not in misunderstanding how cudaGetSymbolAddress works, but in using cudaMemcpy with wrong parameters types: I expected cudaMemcpy to allocate memory for me, so I've cast my variables to the wrong types.
The corrected code is:
#include <iostream>
#include <cassert>
using namespace std;
__device__ int device_int;
__global__ void kernel()
{
device_int = 1000;
}
int main()
{
kernel<<<1, 1>>>();
assert(cudaGetLastError() == cudaSuccess);
int *host_pointer_to_device_int;
/* Get a pointer to device_int. After this, I won't be able to access it,
* but I'm going to copy its value with cudaMemcpy */
cudaGetSymbolAddress((void **)&host_pointer_to_device_int, device_int);
assert(cudaGetLastError() == cudaSuccess); // The above operation executed successfully
int host_int;
// Copy the device_int's value
cudaMemcpy(&host_int, host_pointer_to_device_int,
sizeof(int), cudaMemcpyDeviceToHost);
assert(cudaGetLastError() == cudaSuccess); // The above operation executed successfully
cout << host_int << endl; // Everything's fine!
}
Thanks to #talonmies for helping me in figuring it out.