* perf: can I record in one devce, and analyze in another device
@ 2017-05-28 13:21 cee1
0 siblings, 0 replies; only message in thread
From: cee1 @ 2017-05-28 13:21 UTC (permalink / raw)
To: linux-perf-users
Hi all,
I have a perf.data recored in an ARM device with a kernel of old
version. And 'perf script' in my PC, and I noticed following lines:
"""
perf 7952 [001] 5118.977020: sched:sched_stat_runtime:
comm=perf pid=7952 runtime=3659688 [ns] vruntime=1394034874931 [ns]
perf 7952 [001] 5118.977037: sched:sched_switch:
perf:7952 [120] R ==> nanohub:414 [50]
...
nanohub 414 [001] 5118.977116: sched:sched_switch:
nanohub:414 [50] S ==> perf:7952 [120]
...
.mobileqq:video 7908 [002] 5118.977172: sched:sched_stat_runtime:
comm=.mobileqq:video pid=7908 runtime=81563 [ns] vruntime=842128189897
[ns]
.mobileqq:video 7908 [002] 5118.977194: sched:sched_stat_runtime:
comm=perf pid=7952 runtime=97031 [ns] vruntime=1394034971962 [ns]
"""
The last line, pid=7908 on CPU 002 "sched_stat_runtime" of pid=7952 on CPU 001?
IIUC sched_stat_runtime is invoked in update_curr, which will stat the
"current task" of the rq. How can task 7952 on CPU 001 be "stated" by
the current task on CPU 002?
Or it is wrong to analyze perf.data in a different device?
--
Regards,
- cee1
^ permalink raw reply [flat|nested] only message in thread
only message in thread, other threads:[~2017-05-28 13:21 UTC | newest]
Thread overview: (only message) (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2017-05-28 13:21 perf: can I record in one devce, and analyze in another device cee1
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).