linux-perf-users.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
* perf stat -e cycles -p <pid> sleep 10
@ 2012-11-26 22:50 Chris Freehill
  2012-11-27  6:47 ` Chulmin Kim
  0 siblings, 1 reply; 4+ messages in thread
From: Chris Freehill @ 2012-11-26 22:50 UTC (permalink / raw)
  To: linux-perf-users

Hi,

Apologies if the answer to this is someplace I have not looked. I've searched 
this list and other sites/FAQ's. 

I've noticed that command listed in the subject line always yields

<not counted> cycles

unless the target process is suspended. I have noticed this with a few different
distributions. Is the fact that the target process has to be suspended at least
once a known design limitation or would it be considered a bug?

Thank you for your help.

Chris

^ permalink raw reply	[flat|nested] 4+ messages in thread

* Re: perf stat -e cycles -p <pid> sleep 10
  2012-11-26 22:50 perf stat -e cycles -p <pid> sleep 10 Chris Freehill
@ 2012-11-27  6:47 ` Chulmin Kim
  2012-11-27 12:25   ` Chris Freehill
  0 siblings, 1 reply; 4+ messages in thread
From: Chulmin Kim @ 2012-11-27  6:47 UTC (permalink / raw)
  To: Chris Freehill; +Cc: linux-perf-users

What kind of distributions you've tested?

In my case(ubuntu 11.04), it works correctly.

I suspect your machine's PMU status, wrong installation, or missing perf 
kernel module.




2012-11-27 오전 7:50, Chris Freehill 쓴 글:
> Hi,
>
> Apologies if the answer to this is someplace I have not looked. I've searched
> this list and other sites/FAQ's.
>
> I've noticed that command listed in the subject line always yields
>
> <not counted> cycles
>
> unless the target process is suspended. I have noticed this with a few different
> distributions. Is the fact that the target process has to be suspended at least
> once a known design limitation or would it be considered a bug?
>
> Thank you for your help.
>
> Chris
>
> --
> To unsubscribe from this list: send the line "unsubscribe linux-perf-users" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
>

^ permalink raw reply	[flat|nested] 4+ messages in thread

* Re: perf stat -e cycles -p <pid> sleep 10
  2012-11-27  6:47 ` Chulmin Kim
@ 2012-11-27 12:25   ` Chris Freehill
  2012-11-27 13:13     ` Chris Freehill
  0 siblings, 1 reply; 4+ messages in thread
From: Chris Freehill @ 2012-11-27 12:25 UTC (permalink / raw)
  To: linux-perf-users


Chulmin Kim <cmkim <at> core.kaist.ac.kr> writes:
> What kind of distributions you've tested?
> 
> In my case(ubuntu 11.04), it works correctly.
> 
> I suspect your machine's PMU status, wrong installation, or missing perf 
> kernel module.

Thank you for the reply.

I had tested on Ubuntu 12.04 and the Freescale Linux version (power arch--
P5020). Upon your response, I retested, and you are correct. It does work as 
expected on Ubuntu, so there must be an issue on the other version only. 

I will investigate some of the leads you suggested. I am a newbie with perf 
and perf development, so the suggestions help.

Thanks again.

^ permalink raw reply	[flat|nested] 4+ messages in thread

* Re: perf stat -e cycles -p <pid> sleep 10
  2012-11-27 12:25   ` Chris Freehill
@ 2012-11-27 13:13     ` Chris Freehill
  0 siblings, 0 replies; 4+ messages in thread
From: Chris Freehill @ 2012-11-27 13:13 UTC (permalink / raw)
  To: linux-perf-users


An update in case another newbie runs into this...this was user error.

I believe the reason I was seeing no events is that the processes (in the 
"failure" cases) I chose to monitor were not being allocated any time, so of 
course they generated no events. When I select the right processes, I get events.

^ permalink raw reply	[flat|nested] 4+ messages in thread

end of thread, other threads:[~2012-11-27 13:13 UTC | newest]

Thread overview: 4+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2012-11-26 22:50 perf stat -e cycles -p <pid> sleep 10 Chris Freehill
2012-11-27  6:47 ` Chulmin Kim
2012-11-27 12:25   ` Chris Freehill
2012-11-27 13:13     ` Chris Freehill

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).