From: Ian Munsie <imunsie@au1.ibm.com>
To: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Arnaldo Carvalho de Melo <acme@ghostprotocols.net>,
linux-kernel <linux-kernel@vger.kernel.org>,
Paul Mackerras <paulus@samba.org>, Ingo Molnar <mingo@elte.hu>
Subject: Re: [PATCH 3/6] perf: Add timestamp to COMM and MMAP events
Date: Tue, 30 Nov 2010 10:45:40 +1100 [thread overview]
Message-ID: <1291073899-sup-7083@au1.ibm.com> (raw)
In-Reply-To: <1291031690.32004.20.camel@laptop>
Hi Peter,
Excerpts from Peter Zijlstra's message of Mon Nov 29 22:54:50 +1100 2010:
> On Mon, 2010-11-29 at 17:06 +1100, Ian Munsie wrote:
> > This goes away if PERF_SAMPLE_CPU is added to the mix so that the
> > timestamps are offset from the first event of *their* CPU, otherwise
> > time-epoch can go negative, as it has done here.
>
> How does that happen, I though the power7 sched_clock() was fully
> synchronized and monotonic across all cores?
Oh the timestamps coming from the kernel are fine, I should have been
clearer in my email. This bug is purely restricted to the userspace code
that prints them out - in the perf_session__print_tstamp function. It
tries to print the timestamps out so that they start at 0 and if it
doesn't know what CPU the events came from it uses the very first event
it sees as epoch, but at the moment that is not necessarily going to be
the earliest event.
Thinking about it overnight, I realise that this should also go away if
we sort them before printing things out at all, which is exactly what
I'm proposing to do in my patches.
Cheers,
-Ian
next prev parent reply other threads:[~2010-11-29 23:45 UTC|newest]
Thread overview: 16+ messages / expand[flat|nested] mbox.gz Atom feed top
2010-11-23 5:35 perf: Timestamp COMM and MMAP events for correct attribution Ian Munsie
2010-11-23 5:35 ` [PATCH 1/6] perf: Display unknown rather than IP for unidentified DSO Ian Munsie
2010-11-23 5:35 ` [PATCH 2/6] perf: Move all output for perf report -D into trace_event Ian Munsie
2010-11-23 5:35 ` [PATCH 3/6] perf: Add timestamp to COMM and MMAP events Ian Munsie
2010-11-23 11:08 ` Peter Zijlstra
2010-11-23 12:41 ` Arnaldo Carvalho de Melo
2010-11-25 15:47 ` Arnaldo Carvalho de Melo
2010-11-26 21:54 ` Arnaldo Carvalho de Melo
2010-11-29 6:06 ` Ian Munsie
2010-11-29 11:54 ` Peter Zijlstra
2010-11-29 23:45 ` Ian Munsie [this message]
2010-11-30 14:02 ` Arnaldo Carvalho de Melo
2010-11-23 5:35 ` [PATCH 4/6] perf: Add timestamp to READ and LOST events Ian Munsie
2010-11-23 5:35 ` [PATCH 5/6] perf tool: Process COMM and MMAP events in order Ian Munsie
2010-11-23 5:35 ` [PATCH 6/6] perf tool: Process READ and LOST " Ian Munsie
2010-11-23 6:37 ` perf: Timestamp COMM and MMAP events for correct attribution Ian Munsie
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=1291073899-sup-7083@au1.ibm.com \
--to=imunsie@au1.ibm.com \
--cc=a.p.zijlstra@chello.nl \
--cc=acme@ghostprotocols.net \
--cc=linux-kernel@vger.kernel.org \
--cc=mingo@elte.hu \
--cc=paulus@samba.org \
/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