$ cat t263.cu
#include <stdio.h>

#define cudaCheckErrors(msg) \
    do { \
        cudaError_t __err = cudaGetLastError(); \
        if (__err != cudaSuccess) { \
            fprintf(stderr, "Fatal error: %s (%s at %s:%d)\n", \
                msg, cudaGetErrorString(__err), \
                __FILE__, __LINE__); \
            fprintf(stderr, "*** FAILED - ABORTING\n"); \
            exit(1); \
        } \
    } while (0)

int main(){
  int capacity = 1048576;
  float* data_t = 0;
  float* data = 0;

  size_t msize = sizeof(float) * capacity;
  cudaHostAlloc((void**)&data_t, msize, cudaHostAllocPortable);
  cudaCheckErrors("cudaHostAlloc 1 fail");
  cudaHostAlloc((void**)&data, msize, cudaHostAllocPortable);
  cudaCheckErrors("cudaHostAlloc 2 fail");
  cudaMemcpy(data_t, data, msize, cudaMemcpyDefault);
  cudaCheckErrors("cudaMemcpy fail");
  printf("success\n");
  return 0;
}
$ nvcc -arch=sm_20 -o t263 t263.cu
$ ./t263
success
$ uname -a
Linux cluster1 2.6.18-194.el5 #1 SMP Fri Apr 2 14:58:14 EDT 2010 x86_64 x86_64 x86_64 GNU/Linux
$ nvcc --version
nvcc: NVIDIA (R) Cuda compiler driver
Copyright (c) 2005-2012 NVIDIA Corporation
Built on Fri_Sep_21_17:28:58_PDT_2012
Cuda compilation tools, release 5.0, V0.2.1221
$