Attach nvprof to process

I want to profile a program which has two phases. Assume after 10 seconds it enters the main phase. If I use nvprof for the whole run, it will include the first phase data in the result. The document at seems to be a windows and visual studio method.

Is there any way to attach nvprof to a running process?


Instead, you should use profiling controls such as cudaProfilerStart() and cudaProfilerStop() within your program to determine the phase to profile.

Then profile it normally with nvprof, while selecting the nvprof option to start with profiling disabled.