All of lore.kernel.org
 help / color / mirror / Atom feed
From: Shashank Bhatia <shashank.bhatia@domain.hid>
To: Xenomai Help <xenomai@xenomai.org>
Subject: [Xenomai-core] Regarding latency benchmarks
Date: Tue, 23 Jun 2009 16:25:10 +0530	[thread overview]
Message-ID: <1245754510.25830.5.camel@domain.hid> (raw)

Dear All,

I was trying all latency benchmark programs, in one such session, i had
loaded my linux with some program which will read the hard drive to
create an iso image. This program had taken my Dual core processor to
100%. And both core showed up 100% in gnome-system-monitor.

While this was running, i tried the klatency benchmarks, i got some
strange results :

root@domain.hid# ./run 
*
*
* Type ^C to stop this application.
*
*
== Sampling period: 100 us
== Test mode: in-kernel periodic task
== All results in microseconds
warming up...
RTT|  00:00:00  (in-kernel periodic task, 100 us period, priority 99)
RTH|-lat min|-----lat avg|-----lat max|-overrun|----lat best|---lat 
RTD| -2.593|      -2.410|       2.229|       0|      -2.593|       2.229
RTD|100.000|       0.001|       0.000|   -2590|      -2.411|      -0.001
RTD| -4.821|    2419.968|      -0.001|   -2590|      -2.411|      -0.001
RTD|100.000|       0.001|       0.000|   -2585|      -2.381|      -0.001
RTD| -7.202|    2419.968|      -0.001|   -2585|      -2.381|      -0.001
RTD|100.000|       0.001|       0.000|   -2585|      -2.405|      -0.001
RTD| -9.607|    2419.968|      -0.001|   -2585|      -2.405|      -0.001
RTD|100.000|       0.001|       0.000|   -2587|      -2.380|      -0.001
RTD|-11.987|    2419.968|      -0.001|   -2587|      -2.380|      -0.001
RTD|100.000|       0.001|       0.000|   -2579|      -2.360|      -0.001
RTD|-14.347|    2419.968|      -0.001|   -2579|      -2.360|      -0.001
RTD|100.000|       0.001|       0.000|   -2598|      -2.359|      -0.001
RTD|-16.706|    2419.968|      -0.001|   -2598|      -2.359|      -0.001
RTD|100.000|       0.001|       0.000|   -2588|      -2.373|      -0.001
RTD|-19.079|    2419.968|      -0.001|   -2588|      -2.373|      -0.001
RTD|100.000|       0.001|       0.000|   -2668|      -2.397|      -0.001
RTD|-21.476|    2419.968|      -0.001|   -2668|      -2.397|      -0.001
RTD|100.000|       0.001|       0.000|   -2589|      -2.405|      -0.001
RTD|-23.881|    2419.968|      -0.001|   -2589|      -2.405|      -0.001
RTD|100.000|       0.001|       0.000|   -2588|      -2.406|      -0.001
RTD|-26.287|    2419.968|      -0.001|   -2588|      -2.406|      -0.001


Then i tried the same with user space latency benchmarks, and these are
the results i got:
root@domain.hid# ./run 
*
*
* Type ^C to stop this application.
*
*
== Sampling period: 100 us
== Test mode: periodic user-mode task
== All results in microseconds
warming up...
RTT|  00:00:01  (periodic user-mode task, 100 us period, priority 99)
RTH|--lat min|-----lat avg|-----lat max|-overrun|----lat best|---worst
RTD| -0.851|      -0.165|       8.607|       0|      -0.851|       8.607
RTD| -0.861|      -0.297|       8.141|       0|      -0.861|       8.607
RTD| -0.876|      -0.357|       6.562|       0|      -0.876|       8.607
RTD| -0.881|      -0.367|      10.051|       0|      -0.881|      10.051
RTD| -0.871|      -0.214|       6.487|       0|      -0.881|      10.051
RTD| -0.866|      -0.246|       9.038|       0|      -0.881|      10.051
RTD| -0.871|      -0.247|      10.677|       0|      -0.881|      10.677
RTD| -0.886|      -0.307|      10.597|       0|      -0.886|      10.677
RTD| -0.840|      -0.251|      11.880|       0|      -0.886|      11.880
RTD| -0.861|      -0.301|       8.873|       0|      -0.886|      11.880
RTD| -1.081|      -0.415|       6.582|       0|      -1.081|      11.880
RTD| -0.891|      -0.395|       7.925|       0|      -1.081|      11.880
RTD| -0.896|      -0.352|       8.622|       0|      -1.081|      11.880
RTD| -0.896|      -0.398|       9.259|       0|      -1.081|      11.880
RTD| -0.881|      -0.287|       7.459|       0|      -1.081|      11.880
RTD| -0.891|      -0.155|      10.081|       0|      -1.081|      11.880
RTD| -0.871|      -0.147|       8.572|       0|      -1.081|      11.880
RTD| -0.861|      -0.232|       6.432|       0|      -1.081|      11.880
RTD| -0.876|      -0.397|      12.181|       0|      -1.081|      12.181
RTD| -0.881|      -0.397|       6.858|       0|      -1.081|      12.181
RTD| -0.886|      -0.349|       6.667|       0|      -1.081|      12.181



My CPU is a dual core intel processor 1.8 GHz, and i am using
xenomai-2.4.8 with 2.6.28.9 kernel.


-Shashank


-------------------------------------
Hi-Tech Gears Limited, Gurgaon, India






             reply	other threads:[~2009-06-23 10:55 UTC|newest]

Thread overview: 2+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2009-06-23 10:55 Shashank Bhatia [this message]
2009-06-23 11:39 ` [Xenomai-core] Regarding latency benchmarks Gilles Chanteperdrix

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=1245754510.25830.5.camel@domain.hid \
    --to=shashank.bhatia@domain.hid \
    --cc=xenomai@xenomai.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.