Re: [PATCH v3] perf/trace : Fix repetitious traces of perf on tracepoint

From: Arnaldo Carvalho de Melo
Date: Wed Jan 17 2018 - 09:26:01 EST


Em Wed, Jan 17, 2018 at 10:33:42AM -0300, Arnaldo Carvalho de Melo escreveu:
> Em Tue, Jan 16, 2018 at 04:06:22PM +0100, Jiri Olsa escreveu:
> > On Tue, Jan 16, 2018 at 08:40:38PM +0800, Cheng Jian wrote:
> > > after this patch, perf script(parent-1040, child-1041):

> > > test_fork 1040 [002] 36.536079: sched:sched_wakeup_new: comm=test_fork pid=1041 prio=120 target_cpu=003

> > > It will match it only once for tracing task(child-1041).

> > > Signed-off-by: Cheng Jian <cj.chengjian@xxxxxxxxxx>

> > the duplicated events are gone.. solution looks ok to me

> > Reviewed-and-Tested-by: Jiri Olsa <jolsa@xxxxxxxxxx>

> I noticed that in the past as well, will try to test this too.

Unpatched kernel:

[root@jouet ~]# perf trace --no-syscalls -e sched:sched_wakeup_new time sleep 1
0.000 sched:sched_wakeup_new:time:13571 [120] success=1 CPU:003)
18446744073709.547 sched:sched_wakeup_new:time:13571 [120] success=1 CPU:003)
18446744073709.551 sched:sched_wakeup_new:time:13571 [120] success=1 CPU:003)
18446744073709.551 sched:sched_wakeup_new:time:13571 [120] success=1 CPU:003)
18446744073709.551 sched:sched_wakeup_new:time:13571 [120] success=1 CPU:003)
0.00user 0.00system 0:01.00elapsed 0%CPU (0avgtext+0avgdata 1996maxresident)k
0inputs+0outputs (0major+71minor)pagefaults 0swaps
[root@jouet ~]#

Patched kernel:

[root@seventh ~]# perf trace --no-syscalls -e sched:sched_wakeup_new time sleep 1
0.000 sched:sched_wakeup_new:comm=time pid=339 prio=120 target_cpu=002)
0.001 sched:sched_wakeup_new:comm=time pid=339 prio=120 target_cpu=002)
0.00user 0.00system 0:01.00elapsed 0%CPU (0avgtext+0avgdata 2084maxresident)k
0inputs+0outputs (0major+75minor)pagefaults 0swaps
[root@seventh ~]#