[novice question] OpenCL device code is not run

Accelerated Computing CUDA CUDA Programming and Performance
#1

(Probably lame question yet …)

Having installed NVIDIA gpucomputingsdk_4.2.9 and whatever else is needed (on Ubuntu 64 (10.04 ?) with NVIDIA GTX 285)

– CUDA stuff (compiled with nvcc) works properly

– OpenCL stuff however does not work properly

the openCL sample was compiled with this command line:

g++ -L/usr/local/cuda/lib64 -lOpenCL -I/usr/local/cuda/include opencl_hello_world.c -o hwOpenCL

But when hwOpenCL is started, from the output it is evident that the device code has not run at all.

How come?

#2

Can I get a yes/no/may-be answer to that question:

Is this the way to compile an OpenCL program:

g++ -L/usr/local/cuda/lib64 -lOpenCL -I/usr/local/cuda/include <source-files-list> -o <executable-nama>