On 14 July 2017 at 13:53, Etemadi, Mohammad mohammad.etemadi@intel.com wrote:
Hello Mathieu
We also like to trace minor page fault in the kernel when a user thread hits a page fault.
I am using the following perf command to start the trace for a given thread.
perf record -e cs_etm/@8008046000.etf/ --per-thread --pid 2231&
I never tested this scenario but it should work... There shouldn't be any difference between a PID and a utility name. Do you have namespace or containers enabled? That could explain things.
Trace captured has only user space activity and I do not see any kernel trace.
Is this the correct perf command?
Regards, Reza