linux-kernel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Jim Cromie <jim.cromie@gmail.com>
To: Arnaldo Carvalho de Melo <acme@ghostprotocols.net>
Cc: linux-kernel@vger.kernel.org, Jim Cromie <jim.cromie@gmail.com>
Subject: [PATCH 2/5] perf-stat: fix +- nan% in --no-aggr runs
Date: Wed,  7 Sep 2011 17:14:01 -0600	[thread overview]
Message-ID: <1315437244-3788-3-git-send-email-jim.cromie@gmail.com> (raw)
In-Reply-To: <1315437244-3788-1-git-send-email-jim.cromie@gmail.com>

without this patch, running:
$ sudo ./perf stat -r20 --no-aggr -a perl -e '$i++ for 1..100000'

I get computations like this:

CPU0             12.488247 task-clock                #    1.224 CPUs utilized            ( +-  -nan% )
CPU1             12.488909 task-clock                #    1.225 CPUs utilized            ( +-  -nan% )
CPU2             12.500221 task-clock                #    1.226 CPUs utilized            ( +-  -nan% )
CPU3             12.481713 task-clock                #    1.224 CPUs utilized            ( +-  -nan% )

but with patch, I get:

CPU0              8.233682 task-clock                #    0.754 CPUs utilized            ( +-  0.00% )
CPU1              8.226318 task-clock                #    0.754 CPUs utilized            ( +-  0.00% )
CPU2              8.210737 task-clock                #    0.752 CPUs utilized            ( +-  0.00% )
CPU3              8.201691 task-clock                #    0.751 CPUs utilized            ( +-  0.00% )

Note that without --no-aggr, I get non-0 statistics both before and after patch:

        231.986022 task-clock                #    4.030 CPUs utilized            ( +-  0.97% )
               212 context-switches          #    0.001 M/sec                    ( +- 12.07% )
                 9 CPU-migrations            #    0.000 M/sec                    ( +- 25.80% )
               466 page-faults               #    0.002 M/sec                    ( +-  3.23% )
       174,318,593 cycles                    #    0.751 GHz                      ( +-  1.06% )

I couldnt see anything wrong in the caller, so fixed it in
stddev_stats().  ISTM that 0.00 is better than nan, since perf stat
was passed -A (--no-aggr) so no standard deviation should be expected,
and nan is suggestive of a deeper error.

When running with --no-aggr, perhaps we should suppress the statistics
printing entirely, or do so when they are 0.00.

Signed-off-by: Jim Cromie <jim.cromie@gmail.com>
---
 tools/perf/builtin-stat.c |    9 +++++++--
 1 files changed, 7 insertions(+), 2 deletions(-)

diff --git a/tools/perf/builtin-stat.c b/tools/perf/builtin-stat.c
index 5e3206e..c6e47d2 100644
--- a/tools/perf/builtin-stat.c
+++ b/tools/perf/builtin-stat.c
@@ -254,8 +254,13 @@ static double avg_stats(struct stats *stats)
  */
 static double stddev_stats(struct stats *stats)
 {
-	double variance = stats->M2 / (stats->n - 1);
-	double variance_mean = variance / stats->n;
+	double variance, variance_mean;
+
+	if (!stats->n)
+		return 0.0;
+
+	variance = stats->M2 / (stats->n - 1);
+	variance_mean = variance / stats->n;
 
 	return sqrt(variance_mean);
 }
-- 
1.7.4.4


  parent reply	other threads:[~2011-09-07 23:14 UTC|newest]

Thread overview: 25+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2011-05-21  8:56 [PATCH 0/2] perf stat: add -l <N> option to redirect stderr elsewhere Jim Cromie
2011-05-21  8:56 ` [PATCH 1/2] " Jim Cromie
2011-05-21 10:34   ` Ingo Molnar
2011-05-21 20:41     ` Arnaldo Carvalho de Melo
2011-09-01 21:58       ` Jim Cromie
2011-09-02 14:35         ` Arnaldo Carvalho de Melo
2011-09-02 18:02           ` Arnaldo Carvalho de Melo
2011-09-02 18:04           ` Jim Cromie
2011-09-02 18:58             ` Arnaldo Carvalho de Melo
2011-09-07 23:13               ` [patch 0/5] perf stat --log-fd=N Jim Cromie
2011-09-07 23:14                 ` [PATCH 1/5] perf stat: add --log-fd <N> option to redirect stderr elsewhere Jim Cromie
2011-09-07 23:14                 ` Jim Cromie [this message]
2011-09-07 23:14                 ` [PATCH 3/5] perf stat: suppress printing std-dev when its 0 Jim Cromie
2011-09-07 23:14                 ` [PATCH 4/5] perf stat: allow tab as cvs delimiter Jim Cromie
2011-09-07 23:14                 ` [PATCH 5/5] perf stat: fix spelling in comment Jim Cromie
2011-05-21  8:56 ` [PATCH 2/2] dont commify big numbers by default, let -B do it Jim Cromie
2011-05-21 10:33   ` Ingo Molnar
2011-05-24 22:05     ` Jim Cromie
2011-05-24 22:11       ` [PATCH] add --simple output mode to perf-stat, based upon csv-output Jim Cromie
2011-05-25 12:44       ` [PATCH 2/2] dont commify big numbers by default, let -B do it Ingo Molnar
2011-05-26 19:41         ` Jim Cromie
2011-05-26 19:50           ` [PATCH 0/3] perf-stat: refactor print/formatting into print-ops Jim Cromie
2011-05-26 19:50             ` [PATCH 1/3] perf-stat: refactor print/formatting into print-ops for pretty, csv Jim Cromie
2011-05-26 19:50             ` [PATCH 2/3] perf-stat: fix +- nan% in -Aa runs Jim Cromie
2011-05-26 19:50             ` [PATCH 3/3] perf-stat: clean up <no count> handling, CPUx prefixing Jim Cromie

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=1315437244-3788-3-git-send-email-jim.cromie@gmail.com \
    --to=jim.cromie@gmail.com \
    --cc=acme@ghostprotocols.net \
    --cc=linux-kernel@vger.kernel.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;
as well as URLs for NNTP newsgroup(s).