I'm trying to profile my OpenCL application on PyOpenCL via Intel VTune profiler. I have Intel HD Graphics 630 installed in my machine. From my understanding, that can serve as a GPU to run my OpenCL computations on.
I start the profiler via the launch application selection, pointing the profiler to the python script that I am executing. However, in the summary report generated, I noticed that the profiler picks up Nvidia GeForce as the GPU to profile. I have an Nvidia GeForce 1060 graphics card mounted into the system, just that the driver has not yet been installed in the OS.
I would like to know if it is possible to select the GPU used in the profiling. Attaching a screen shot of the summary page here for debugging.
VTune can collect GPU HW metrics for OpenCL kernels and profile them only if kernels are executed on Intel GPU. So it is not necessary to select the GPU for profiling in VTune. Your code just need to choose the Intel GPU as the target device for OpenCL kernels and it is enough for VTune. Does your application run kernels on Intel GPU?
Yes the application runs its kernel on Intel HD graphics. So what I'm getting from your post is that the application simply needs to run on HD graphics and VTune will automatically pick it up in its GPU hotspot analysis?
In this case, 2 follow up questions here:
... the application simply needs to run on HD graphics and VTune will automatically pick it up in its GPU hotspot analysis?
Can VTune profile kernels on NVIDIA CUDA?
No, VTune cannot profile CUDA kernels.
My code has been showing 0% GPU utilization after repeated execution. The python script that I am profiling is configured to select the runtime environment as Intel HD Graphics. Can I get some suggestions as to how I can troubleshoot this issue?
Could you please share the result for further investigation?