All of lore.kernel.org
 help / color / mirror / Atom feed
From: Arnaldo Carvalho de Melo <acme@kernel.org>
To: Jiri Olsa <jolsa@redhat.com>
Cc: Jiri Olsa <jolsa@kernel.org>,
	linux-kernel@vger.kernel.org,
	Adrian Hunter <adrian.hunter@intel.com>,
	Corey Ashford <cjashfor@linux.vnet.ibm.com>,
	David Ahern <dsahern@gmail.com>,
	Frederic Weisbecker <fweisbec@gmail.com>,
	Ingo Molnar <mingo@kernel.org>,
	Jean Pihet <jean.pihet@linaro.org>,
	Namhyung Kim <namhyung@kernel.org>,
	Paul Mackerras <paulus@samba.org>,
	Peter Zijlstra <a.p.zijlstra@chello.nl>
Subject: Re: [PATCH 17/19] perf tools: Always force PERF_RECORD_FINISHED_ROUND event
Date: Fri, 25 Jul 2014 11:14:13 -0300	[thread overview]
Message-ID: <20140725141413.GK7831@kernel.org> (raw)
In-Reply-To: <20140725113426.GC1214@krava.brq.redhat.com>

Em Fri, Jul 25, 2014 at 01:34:26PM +0200, Jiri Olsa escreveu:
> On Thu, Jul 24, 2014 at 06:34:51PM -0300, Arnaldo Carvalho de Melo wrote:
> > I think both changes are OK, but should be split in different patches,
 
> right, I'll split it

Thanks!
 
> > [root@zoo /]# perf stat -r 5 perf report --no-ordered-samples > /dev/null
> >    101,171,572,553      instructions              #    1.10  insns per cycle        
> >       30.249514999 seconds time elapsed                                          ( +-  0.48% )

> > [root@zoo /]# perf stat -r 5 perf report --ordered-samples > /dev/null
> >    105,982,144,263      instructions              #    1.04  insns per cycle        
> >       32.636483981 seconds time elapsed                                          ( +-  0.41% )

> so those 2 extra seconds is the ordering time, right? sounds ok

Yeah, but I think its worth investigating if using it is a strict
requirement in all cases, i.e. is it possible to receive out of order
events when sampling on a single CPU? Or a single CPU socket with a
coherent time source? etc.

Providing a way to disable this ordering to be used in corner cases
where it is not a strict requirement and the volume of samples is so
high that reducing processing time like shown above seems to be a
sensible thing to do.

We're in the business of optimizing stuff, huh? :-)
 
- Arnaldo

  reply	other threads:[~2014-07-25 14:14 UTC|newest]

Thread overview: 45+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2014-07-20 21:55 [PATCHv3 00/19] perf tools: Factor ordered samples queue Jiri Olsa
2014-07-20 21:55 ` [PATCH 01/19] perf tools: Fix accounting of " Jiri Olsa
2014-07-28  8:27   ` [tip:perf/core] perf session: " tip-bot for Jiri Olsa
2014-07-20 21:55 ` [PATCH 02/19] perf tools: Rename ordered_samples bool to ordered_events Jiri Olsa
2014-07-20 21:55 ` [PATCH 03/19] perf tools: Rename ordered_samples struct " Jiri Olsa
2014-07-20 21:55 ` [PATCH 04/19] perf tools: Rename ordered_events members Jiri Olsa
2014-07-20 21:55 ` [PATCH 05/19] perf tools: Add ordered_events_(new|delete) interface Jiri Olsa
2014-07-20 21:55 ` [PATCH 06/19] perf tools: Factor ordered_events_flush to be more generic Jiri Olsa
2014-07-20 21:55 ` [PATCH 07/19] perf tools: Limit ordered events queue size Jiri Olsa
2014-07-20 21:55 ` [PATCH 08/19] perf tools: Flush ordered events in case of allocation failure Jiri Olsa
2014-07-20 21:55 ` [PATCH 09/19] perf tools: Make perf_session_deliver_event global Jiri Olsa
2014-07-20 21:55 ` [PATCH 10/19] perf tools: Create ordered-events object Jiri Olsa
2014-07-20 21:55 ` [PATCH 11/19] perf tools: Use list_move in ordered_events_delete function Jiri Olsa
2014-07-20 21:55 ` [PATCH 12/19] perf tools: Add ordered_events_init function Jiri Olsa
2014-07-20 21:55 ` [PATCH 13/19] perf tools: Add ordered_events_free function Jiri Olsa
2014-07-20 21:55 ` [PATCH 14/19] perf tools: Add perf_config_u64 function Jiri Olsa
2014-07-20 21:55 ` [PATCH 15/19] perf tools: Add report.queue-size config file option Jiri Olsa
2014-07-20 21:56 ` [PATCH 16/19] perf tools: Add debug prints for ordered events queue Jiri Olsa
2014-07-20 21:56 ` [PATCH 17/19] perf tools: Always force PERF_RECORD_FINISHED_ROUND event Jiri Olsa
2014-07-24 21:34   ` Arnaldo Carvalho de Melo
2014-07-25 11:34     ` Jiri Olsa
2014-07-25 14:14       ` Arnaldo Carvalho de Melo [this message]
2014-07-25 15:45         ` Frederic Weisbecker
2014-07-25 16:12           ` Peter Zijlstra
2014-07-20 21:56 ` [PATCH 18/19] perf tools: Limit the ordered events queue by default to 100MB Jiri Olsa
2014-07-20 21:56 ` [PATCH 19/19] perf tools: Allow out of order messages in forced flush Jiri Olsa
2014-07-21  6:43 ` [PATCHv3 00/19] perf tools: Factor ordered samples queue Adrian Hunter
2014-07-21  8:02   ` Jiri Olsa
2014-07-21  8:47     ` Adrian Hunter
2014-07-21  9:54       ` Jiri Olsa
2014-07-21 12:09         ` Adrian Hunter
2014-07-21 12:35           ` Jiri Olsa
2014-07-21 12:58             ` Adrian Hunter
2014-07-21 16:31         ` Andi Kleen
2014-07-21 18:23           ` Adrian Hunter
2014-07-21 18:36             ` David Ahern
2014-07-21 18:44               ` Adrian Hunter
2014-07-24 14:19               ` Arnaldo Carvalho de Melo
2014-07-24 14:56                 ` Jiri Olsa
2014-07-24 15:10                   ` Arnaldo Carvalho de Melo
2014-07-24 15:20                     ` Jiri Olsa
2014-07-24 15:51                     ` David Ahern
2014-07-24 18:01                 ` Adrian Hunter
2014-07-21 19:39             ` Andi Kleen
2014-07-25 14:55 [PATCHv4 " Jiri Olsa
2014-07-25 14:56 ` [PATCH 17/19] perf tools: Always force PERF_RECORD_FINISHED_ROUND event Jiri Olsa

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=20140725141413.GK7831@kernel.org \
    --to=acme@kernel.org \
    --cc=a.p.zijlstra@chello.nl \
    --cc=adrian.hunter@intel.com \
    --cc=cjashfor@linux.vnet.ibm.com \
    --cc=dsahern@gmail.com \
    --cc=fweisbec@gmail.com \
    --cc=jean.pihet@linaro.org \
    --cc=jolsa@kernel.org \
    --cc=jolsa@redhat.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=mingo@kernel.org \
    --cc=namhyung@kernel.org \
    --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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.