I’m analyzing the performance of the whole app and can’t make Vtune to pay attention to routines compiled by nvcc (host routines of course), Vtune simply ignores them and does not enumerate them. Is it possible to fix it ?
I’m analyzing the performance of the whole app and can’t make Vtune to pay attention to routines compiled by nvcc (host routines of course), Vtune simply ignores them and does not enumerate them. Is it possible to fix it ?