[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20101024100356.7af1f4dd@infradead.org>
Date: Sun, 24 Oct 2010 10:03:56 -0700
From: Arjan van de Ven <arjan@...radead.org>
To: Sitsofe Wheeler <sitsofe@...oo.com>
Cc: Peter Zijlstra <a.p.zijlstra@...llo.nl>,
Paul Mackerras <paulus@...ba.org>, Ingo Molnar <mingo@...e.hu>,
Arnaldo Carvalho de Melo <acme@...hat.com>,
linux-kernel@...r.kernel.org
Subject: Re: perf timechart segfault
On Sun, 24 Oct 2010 17:23:02 +0100
Sitsofe Wheeler <sitsofe@...oo.com> wrote:
> Hi,
>
> I've run into a couple of issues using perf timechart:
>
> 1. (On 2.6.36) If I issue
> echo 0 > /proc/sys/kernel/perf_event_paranoid
> as root before running
> perf timechart record echo
> as a normal user the following is printed:
>
> Fatal: Permission error - are you root?
> Consider tweaking /proc/sys/kernel/perf_event_paranoid.
>
> Shouldn't it just say it needs root permissions for a timechart
> record?
timechart needs root perms since it gets very global data.
>
> 2. (On -tip) Running
> perf timechart record echo
> outputs
> invalid or unsupported event: 'power:power_start'
> Run 'perf list' for a list of valid events
tip broke ABI :-(
That's one for Ingo to deal with for now.
>
> 3. (On 2.6.36) I can't seem to convince timechart to ever capture more
> than 34000 samples. When trying to record using
> perf timechart record -fga sleep 10
> and then creating a graph using perf timechart, it only ever
> outputs a few seconds worth.
does this happen even without the "fga" ?
I've used timechart to capture a minute of data before
(but... you need a rather quiet system for that to be useful.... you
otherwise get an impossibly big SVG file)
--
Arjan van de Ven Intel Open Source Technology Centre
For development, discussion and tips for power savings,
visit http://www.lesswatts.org
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
Powered by blists - more mailing lists