From: David Ahern <daahern@cisco.com>
To: Ingo Molnar <mingo@elte.hu>
Cc: Thomas Gleixner <tglx@linutronix.de>,
linux-perf-users@vger.kernel.org,
LKML <linux-kernel@vger.kernel.org>,
acme@ghostprotocols.net, Peter Zijlstra <peterz@infradead.org>,
Frederic Weisbecker <fweisbec@gmail.com>,
Paul Mackerras <paulus@samba.org>,
John Stultz <johnstul@us.ibm.com>, Borislav Petkov <bp@alien8.de>
Subject: Re: [PATCH 3/6] perf record: add time-of-day option
Date: Thu, 03 Mar 2011 07:33:24 -0700 [thread overview]
Message-ID: <4D6FA6B4.7060407@cisco.com> (raw)
In-Reply-To: <20110303085148.GA16832@elte.hu>
On 03/03/2011 01:51 AM, Ingo Molnar wrote:
>
> * David Ahern <daahern@cisco.com> wrote:
>
>>> How does all that deal with CLOCK_REALTIME being affected by NTP and
>>> settimeofday? Not really, as far as I can tell. It somehow works, but
>>> that depends on the frequency of your event injection.
>>
>> It is sampled at some periodic rate to get NTP changes. Right now it is
>> hardcoded at once an hour. The frequency option can be added to the
>> --tod parameter.
>
> As Thomas mentioned, we probably need something more complete than that.
Just responded to Thomas' email with a simpler proposal than munging on
tracepoints and trying to track time-of-day implicitly.
> Still your approach is obviously useful and i'd like to stress that explicitly. Have
Thanks for stating that.
> you considered another related feature, feeding printk lines as special 'string
> events' into the perf ringbuffer?
>
> It would round up your scheme very nicely: that way you'd have a single, global,
> GTOD-correlated event store/flow for basically every system event you might be
> interested in. You could switch events (and tracepoints) on/off based on need,
> controlling the type and rate of information in a very finegrained way.
>
> If the printk approach works out i'd even suggest that such a facility would need a
> separate tool within perf: 'perf syslog' or 'perf log'. It would heavily reuse
> perf-script and other perf internal facilities, obviously - but you would not be
> tied to any particular sub-tool implementation.
>
> EDAC type events could feed into this as well, giving the tool a broader 'system
> health' aspect as well, beyond the 'system performance analysis' aspect.
I can look into it (as time allows) once I get the dumping of software
events done.
David
>
> Thanks,
>
> Ingo
next prev parent reply other threads:[~2011-03-03 14:33 UTC|newest]
Thread overview: 32+ messages / expand[flat|nested] mbox.gz Atom feed top
2011-02-28 3:52 [PATCH 0/6 v3] perf events: Add realtime clock event, time-of-day strings to script output David Ahern
2011-02-28 3:52 ` [PATCH 1/6] perf events: Introduce realtime clock event David Ahern
2011-02-28 3:52 ` [PATCH 2/6] perf events: plumbing for PERF_SAMPLE_READ and read_format David Ahern
2011-02-28 3:52 ` [PATCH 3/6] perf record: add time-of-day option David Ahern
2011-03-01 14:29 ` Peter Zijlstra
2011-03-01 14:35 ` David Ahern
2011-03-01 15:35 ` Peter Zijlstra
2011-03-01 15:41 ` David Ahern
2011-03-01 16:00 ` Peter Zijlstra
2011-03-01 16:09 ` David Ahern
2011-03-01 16:37 ` Peter Zijlstra
2011-03-01 16:45 ` David Ahern
2011-03-01 17:07 ` Arnaldo Carvalho de Melo
2011-03-01 17:09 ` Arnaldo Carvalho de Melo
2011-03-01 22:28 ` Peter Zijlstra
2011-03-01 22:35 ` David Ahern
2011-03-02 14:16 ` Thomas Gleixner
2011-03-02 14:28 ` David Ahern
2011-03-02 17:28 ` Thomas Gleixner
2011-03-03 14:29 ` David Ahern
2011-03-03 8:51 ` Ingo Molnar
2011-03-03 14:33 ` David Ahern [this message]
2011-02-28 3:52 ` [PATCH 4/6] perf script: dump software events too David Ahern
2011-03-01 14:09 ` Frederic Weisbecker
2011-03-01 14:18 ` David Ahern
2011-03-01 15:11 ` Frederic Weisbecker
2011-03-01 16:11 ` David Ahern
2011-03-01 16:24 ` Frederic Weisbecker
2011-03-01 16:49 ` Arnaldo Carvalho de Melo
2011-02-28 3:52 ` [PATCH 5/6] perf script: Prepend lines with time-of-day string David Ahern
2011-02-28 3:52 ` [PATCH 6/6] perf stat: treat realtime-clock as nsec counter David Ahern
2011-02-28 3:55 ` [PATCH 0/6 v3] perf events: Add realtime clock event, time-of-day strings to script output David Ahern
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=4D6FA6B4.7060407@cisco.com \
--to=daahern@cisco.com \
--cc=acme@ghostprotocols.net \
--cc=bp@alien8.de \
--cc=fweisbec@gmail.com \
--cc=johnstul@us.ibm.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-perf-users@vger.kernel.org \
--cc=mingo@elte.hu \
--cc=paulus@samba.org \
--cc=peterz@infradead.org \
--cc=tglx@linutronix.de \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
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).