Messages in this thread | | | Date | Tue, 24 Mar 2015 16:10:20 -0400 | From | Don Zickus <> | Subject | Re: [PATCH] perf tool: Fix ppid for synthesized fork events |
| |
On Thu, Mar 19, 2015 at 03:06:46PM -0600, David Ahern wrote: > On 3/19/15 2:56 PM, Don Zickus wrote: > >On Thu, Mar 19, 2015 at 11:41:15AM -0600, David Ahern wrote: > >>>363b785f38 added synthesized fork events and set a thread's parent id > >>>to itself. Since we are already processing/proc/<pid>/status the ppid > >>>can be determined properly. Make it so. > >Thanks David. My tester, Joe, is currently running other tests and then is > >out until Tuesday. I will try to provide test feedback by Tuesday or > >Wednesday if it can wait. > > ok. thanks for the heads up.
Hmm, preliminary tests, show a significant slow down in perf record and perf report. I would be against this patch for now. We will dig into what the problem is.
Joe is running a specjbb bench with lots of threads in the background and running:
perf mem record -a -e 'cpu/mem-loads,ldlat=50/pp' -e 'cpu/mem-stores/pp' sleep 10
multiple times to get an average. And also
perf mem report --stdio
to get that timing average too..
He does this with and without the patch. The difference is usually over 50% extra time with the patch for both the record timings and report timings. :-(
Cheers, Don
| |