I am interested in gathering some application properties by reading GPU hardware counters for some of the workloads (not graphic workloads but general purpose workloads written using opencl to run on GPU) on Intel on-chip GPU in my i7-6700K (skylake) processor. To be specific I would interested in knowing metrics like EU utilization, stall cycles etc. The intel_perf_counter (https://github.com/tiagovignatti/intel-gpu-tools/blob/master/tools/intel_perf_counters.c intel-gpu-tools/intel_perf_counters.c at master · tiagovignatti/intel-gpu-tools · GitHub) tool can do that but it does not work for the new generation processor. Can anyone point me to any documentation on how to do this?
