I get the following output:

[codebox][ydd@localhost tmp]$ make

mkdir -p /usr/local/NVIDIA_CUDA_SDK/common//…/lib

mkdir -p obj/release

mkdir -p ./bin//release

/usr/local/cuda/bin/nvcc -o obj/release/test.cu_sm_13_o -c test.cu --compiler-options -fno-strict-aliasing -I. -I/usr/local/cuda/include -I/usr/local/NVIDIA_CUDA_SDK/common//…/common/inc -DUNIX -O3 -arch sm_13

g++ -fPIC -o ./bin//release/HelloCUDA obj/release/test.cu_sm_13_o -L/usr/local/cuda/lib -L/usr/local/NVIDIA_CUDA_SDK/common//…/lib -L/usr/local/NVIDIA_CUDA_SDK/common//…/common/lib/linux -lcudart -L/usr/local/cuda/lib -L/usr/local/NVIDIA_CUDA_SDK/common//…/lib -L/usr/local/NVIDIA_CUDA_SDK/common//…/common/lib/linux -lcutil

[ydd@localhost tmp]$ ./bin/release/HelloCUDA

vnew[0] is 0; vnew2[0] is 0;vnew[0] is 0

vnew[1] is 1; vnew2[1] is 1;vnew[1] is 1

vnew[2] is 2; vnew2[2] is 2;vnew[2] is 2

vnew[3] is 3; vnew2[3] is 3;vnew[3] is 3

vnew[4] is 4; vnew2[4] is 4;vnew[4] is 4

vnew[5] is 5; vnew2[5] is 5;vnew[5] is 5

vnew[6] is 6; vnew2[6] is 6;vnew[6] is 6

vnew[7] is 7; vnew2[7] is 7;vnew[7] is 7

vnew[8] is 8; vnew2[8] is 8;vnew[8] is 8

vnew[9] is 9; vnew2[9] is 9;vnew[9] is 9

vnew[10] is 10; vnew2[10] is 10;vnew[10] is 10

vnew[11] is 11; vnew2[11] is 11;vnew[11] is 11

vnew[12] is 12; vnew2[12] is 12;vnew[12] is 12

vnew[13] is 13; vnew2[13] is 13;vnew[13] is 13

vnew[14] is 14; vnew2[14] is 14;vnew[14] is 14

vnew[15] is 15; vnew2[15] is 15;vnew[15] is 15

vnew[16] is 16; vnew2[16] is 16;vnew[16] is 16

vnew[17] is 17; vnew2[17] is 17;vnew[17] is 17

vnew[18] is 18; vnew2[18] is 18;vnew[18] is 18

vnew[19] is 19; vnew2[19] is 19;vnew[19] is 19

[ydd@localhost tmp]$

[/codebox]

I think that’s what you wanted, isn’t it?

What card do you have in your machine, and what’s your compile command?