From: Jiri Olsa <jolsa@redhat.com>
To: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: acme@redhat.com, mingo@elte.hu, paulus@samba.org,
cjashfor@linux.vnet.ibm.com, fweisbec@gmail.com,
linux-kernel@vger.kernel.org,
Thomas Gleixner <tglx@linutronix.de>
Subject: Re: [PATCH 3/3] perf, tool: Add new event group management
Date: Tue, 20 Mar 2012 23:36:36 +0100 [thread overview]
Message-ID: <20120320223636.GB1549@m> (raw)
In-Reply-To: <1332269081.18960.444.camel@twins>
On Tue, Mar 20, 2012 at 07:44:41PM +0100, Peter Zijlstra wrote:
> On Tue, 2012-03-20 at 19:15 +0100, Jiri Olsa wrote:
> > Examples (first event in brackets is group leader):
> >
> > # 1 group (cpu-clock,task-clock)
> > perf record --group -e cpu-clock,task-clock ls
> > perf record --group parsed -e cpu-clock,task-clock ls
> >
> > # 2 groups (cpu-clock,task-clock) (minor-faults,major-faults)
> > perf record --group parsed -e cpu-clock,task-clock \
> > -e minor-faults,major-faults ls
> >
> > # 1 group (cpu-clock,task-clock,minor-faults,major-faults)
> > perf record --group -e cpu-clock,task-clock \
> > -e minor-faults,major-faults ls
> >
> > # 2 groups (cpu-clock,task-clock) (minor-faults,major-faults)
> > perf record --group parsed -e cpu-clock,task-clock \
> > -e minor-faults,major-faults -e instructions ls
> >
> > # 1 group (cpu-clock,task-clock,minor-faults,major-faults,instructions)
> > perf record --group -e cpu-clock,task-clock \
> > -e minor-faults,major-faults -e instructions ls
>
> I can't help but dislike the --group/--group parsed thing...
>
> How about something like:
>
> event_group = "{", events, "}" , [ ":", event_group_mod ]
>
> Such that you can write things like:
>
> perf record -e "{cpu-clock,cache-misses,cache-references}"
>
> perf stat -e "{cpu-clock,cycles},{cpu-clock,cache-misses,cache-references}"
> perf stat -e "{cpu-clock,cycles}" -e "{cpu-clock,cache-misses,cache-references}"
there should be no problem with that
>
> I'm not sure what the current behaviour of --group is, if you create a
> group like this, do they all sample?
I think so.. the --group switch just set the group_fd for the event syscall, nothing else
>
> If so, we need some option like:
>
> perf record -e "{cpu-clock,cache-misses,cache-references}:1"
>
> to mean, only sample on cpu-clock but use PERF_SAMPLE_READ and
> PERF_FORMAT_GROUP to read all siblings on every cpu-clock sample.
never happy, are you.. ;) seems like nice feature, will check
>
> Now the disadvantage is that {} needs quotes on bash, the advantage is
> that its completely natural on how to construct groups, without weird
> --group/--group parsed. Also it provides a place for group modifiers.
looks like Arnaldo could use [] but I get different results:
[jolsa@dhcp-26-214 perf]$ ./perf stat -e [cycles,instructions] ls
invalid or unsupported event: 't'
need to check.. maybe escaping is small price for good readability,
but I'll think about some other way as well.
thanks for comments,
jirka
next prev parent reply other threads:[~2012-03-20 22:36 UTC|newest]
Thread overview: 28+ messages / expand[flat|nested] mbox.gz Atom feed top
2012-03-20 18:15 [RFC 0/3] perf tool: Add new event group management Jiri Olsa
2012-03-20 18:15 ` [PATCH 1/3] perf, tool: Fix various casting issues for 32 bits Jiri Olsa
2012-03-23 8:29 ` [tip:perf/urgent] perf tools: " tip-bot for Jiri Olsa
2012-03-20 18:15 ` [PATCH 2/3] perf, tool: Fix modifier to be applied on correct events Jiri Olsa
2012-03-23 8:30 ` [tip:perf/urgent] perf tools: " tip-bot for Jiri Olsa
2012-03-20 18:15 ` [PATCH 3/3] perf, tool: Add new event group management Jiri Olsa
2012-03-20 18:44 ` Peter Zijlstra
2012-03-20 20:46 ` Ingo Molnar
2012-03-20 22:18 ` Arnaldo Carvalho de Melo
2012-03-21 9:15 ` Peter Zijlstra
2012-03-21 9:52 ` Ingo Molnar
2012-03-21 11:54 ` Peter Zijlstra
2012-03-21 12:15 ` Ingo Molnar
2012-03-21 15:13 ` Arnaldo Carvalho de Melo
2012-03-22 7:56 ` Ingo Molnar
2012-03-22 12:41 ` Arnaldo Carvalho de Melo
2012-03-22 13:54 ` Ingo Molnar
2012-03-22 14:07 ` Jiri Olsa
2012-03-20 22:36 ` Jiri Olsa [this message]
2012-03-21 11:54 ` Peter Zijlstra
2012-03-21 10:42 ` Jiri Olsa
2012-03-21 15:11 ` Arnaldo Carvalho de Melo
2012-03-21 22:27 ` Jiri Olsa
2012-03-22 12:40 ` Arnaldo Carvalho de Melo
2012-04-10 13:19 ` Jiri Olsa
2012-03-20 20:08 ` Arnaldo Carvalho de Melo
2012-03-20 22:16 ` Jiri Olsa
2012-03-21 2:12 ` Namhyung Kim
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=20120320223636.GB1549@m \
--to=jolsa@redhat.com \
--cc=a.p.zijlstra@chello.nl \
--cc=acme@redhat.com \
--cc=cjashfor@linux.vnet.ibm.com \
--cc=fweisbec@gmail.com \
--cc=linux-kernel@vger.kernel.org \
--cc=mingo@elte.hu \
--cc=paulus@samba.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).