All of lore.kernel.org
 help / color / mirror / Atom feed
From: Clark Williams <williams@redhat.com>
To: Bei Guan <gbtju85@gmail.com>
Cc: Steven Rostedt <rostedt@goodmis.org>,
	linux-rt-users <linux-rt-users@vger.kernel.org>
Subject: Re: Where can I get the latest cyclictest or the version 0.72
Date: Tue, 6 Dec 2011 10:08:13 -0600	[thread overview]
Message-ID: <20111206100813.5aa8b287@redhat.com> (raw)
In-Reply-To: <CAEQjb-TVaBZHbRBodLsBBPjJAMCd1xw9c_iLA-wX4ZmhRuacgQ@mail.gmail.com>

[-- Attachment #1: Type: text/plain, Size: 1853 bytes --]

On Mon, 5 Dec 2011 21:05:32 -0500
Bei Guan <gbtju85@gmail.com> wrote:

> 2011/12/5 Clark Williams <williams@redhat.com>
> 
> > Bei,
> >
> > That's a git tree. You'll need to compile the images from there.
> >
> > What distro are you using? Debian delivers rt-tests as a .deb, so you
> > could just apt-get it from there. I don't think Fedora delivers
> > rt-tests as an RPM.
> >
> I have two distros, Fedora and Ubuntu. Under the Fedora, I just use "make
> install" and it seems to work well. Now I can use the command cyclictest in
> Fedora.
> 
> But, under the Ubuntu, you mean I can use apt-get. Is the following right?
> apt-get install github.com/clrkwllms/rt-tests.git
>  (But it doesn't work. Please correct me.)

No you don't need to specify the git repository; there should be a
package already set up in the stable debian: apt-get install rt-tests

> 
> 
> Another question, how to get the histogram graphic using the cyclictest. I
> use the following command but I cannot get it.
> 
> [root@localhost ~]# cyclictest -h -s -p 80 -n -i 10000 -l 1000
> WARNING: High resolution timers not available
> policy: fifo: loadavg: 0.13 0.10 0.09 4/229 5283
> 
> T: 0 ( 5283) P:80 I:10000 C:   1000 Min:  14420 Act:14496 Avg:14712 Max:
> 24567
> WARNING: unable to open events/enable
> 

The histogram option (-h) takes an argument, which is the max value
to record in microseconds. So, -h100 would generate a 100 element
histogram (1 microsecond per line). Since it doesn't look like you're
running on an RT kernel (no hrtimers available), you'll have a wide
variance in response times, so you'll need a big histogram range to
catch everything. 

Try this:

# cyclictest -t -p80 -m -i 10000 -l 1000 -h 10000

Then be ready for 10000 lines of output when you control-C out of
cyclictest. 

Clark



[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 198 bytes --]

  reply	other threads:[~2011-12-06 16:08 UTC|newest]

Thread overview: 8+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
     [not found] <CAEQjb-TXX8bUULQVWOAQZjAxj7DVrA2ehs039qEH5zSe2BSfLA@mail.gmail.com>
2011-12-05 14:42 ` Where can I get the latest cyclictest or the version 0.72 Steven Rostedt
2011-12-05 14:59   ` Clark Williams
2011-12-05 15:30     ` Bei Guan
2011-12-05 16:36       ` Clark Williams
2011-12-06  2:05         ` Bei Guan
2011-12-06 16:08           ` Clark Williams [this message]
2011-12-07  1:24             ` Bei Guan
2011-12-06 16:13           ` Steven Rostedt

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=20111206100813.5aa8b287@redhat.com \
    --to=williams@redhat.com \
    --cc=gbtju85@gmail.com \
    --cc=linux-rt-users@vger.kernel.org \
    --cc=rostedt@goodmis.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.