[CRIU] [PATCH] [RFC] perf: Here is an example how to trace syscalls

[Pavel Emelyanov]

On 10/03/2014 07:22 PM, Andrew Vagin wrote:
> On Fri, Oct 03, 2014 at 07:18:53PM +0400, Pavel Emelyanov wrote:
>> On 09/30/2014 04:05 PM, Andrey Vagin wrote:
>>> perf doesn't affect performance so significant as strace and
>>> it allows to trace parasite code.
>>>
>>> All processes are executed in a new perf_evets cgroup and
>>> "perf record" collects events only for processes from this group.
>>
>> Presumably I need to do some preparations:
>>
>> Execute zdtm/live/static/env00
>> ./env00 --pidfile=env00.pid --outfile=env00.out --envname=ENV_00_TEST
>> Dump 22781
>> no access to cgroup /sys/fs/cgroup/perf_event/test
> 
> Could you mount the perf_event cgroup?

Already there

# cat /proc/self/mountinfo | fgrep perf
33 23 0:29 / /sys/fs/cgroup/perf_event rw,nosuid,nodev,noexec,relatime shared:17 - cgroup cgroup rw,perf_event
# ls /sys/fs/cgroup/perf_event/
cgroup.clone_children  cgroup.procs  cgroup.sane_behavior  notify_on_release  release_agent  tasks  zdtm

> [avagin at localhost ~]$ cat /proc/self/mountinfo  | grep perf
> 31 21 0:28 / /sys/fs/cgroup/perf_event rw,nosuid,nodev,noexec,relatime
> shared:17 - cgroup cgroup rw,perf_event
> 
>>
>>  usage: perf record [<options>] [<command>]
>>     or: perf record [<options>] -- <command> [<options>]
>>
>>     -G, --cgroup <name>   monitor event in cgroup name only
>> Use the following command to analize perf data:
>> perf trace report -i /root/src/criu/test/dump/static/env00/22781/1/perf.dump
>> WARNING: env00 returned 129 and left running for debug needs
>> Test: zdtm/live/static/env00, Result: FAIL
>>
>>
>>> Signed-off-by: Andrey Vagin <avagin at openvz.org>
>>> ---
>>>  test/zdtm.sh | 11 ++++++++++-
>>>  1 file changed, 10 insertions(+), 1 deletion(-)
>>>
>>> diff --git a/test/zdtm.sh b/test/zdtm.sh
>>> index b5133c3..1eba6d7 100755
>>> --- a/test/zdtm.sh
>>> +++ b/test/zdtm.sh
>>> @@ -1,5 +1,8 @@
>>>  #!/bin/bash
>>>  
>>> +test -d /sys/fs/cgroup/perf_event/zdtm || mkdir /sys/fs/cgroup/perf_event/zdtm || exit 1
>>> +echo $$ > /sys/fs/cgroup/perf_event/zdtm/tasks
>>> +
>>>  ARCH=`uname -m | sed			\
>>>  		-e s/i.86/i386/		\
>>>  		-e s/sun4u/sparc64/	\
>>> @@ -582,8 +585,14 @@ EOF
>>>  
>>>  		save_fds $PID  $ddump/dump.fd
>>>  		save_maps $PID  $ddump/dump.maps
>>> -		setsid $CRIU_CPT $dump_cmd -D $ddump -o dump.log -v4 -t $PID $gen_args $cpt_args
>>> +		echo $$ > /sys/fs/cgroup/perf_event/tasks
>>> +		perf record -o $ddump/perf.dump -e raw_syscalls:sys_exit,raw_syscalls:sys_enter --cgroup test,test -a bash <<EOF
>>> +echo 0 > /sys/fs/cgroup/perf_event/zdtm/tasks &&
>>> +$CRIU_CPT $dump_cmd -D $ddump -o dump.log -v4 -t $PID $gen_args $cpt_args
>>> +EOF
>>>  		retcode=$?
>>> +		echo Use the following command to analize perf data:
>>> +		echo perf trace report -i $ddump/perf.dump
>>>  
>>>  		#
>>>  		# Here we may have two cases: either checkpoint is failed
>>>
>>
> .
>