All of lore.kernel.org
 help / color / mirror / Atom feed
From: Patrick Roberts <patrick.m.roberts@domain.hid>
To: xenomai@xenomai.org
Subject: Re: [Xenomai-help] Xenomai on intel core i7
Date: Wed, 25 Aug 2010 15:01:45 +0000 (UTC)	[thread overview]
Message-ID: <loom.20100825T165507-176@domain.hid> (raw)
In-Reply-To: AANLkTinGZRptSbQPPF-NBiipx2jyt=oz_QHxfftWxMvx@domain.hid

Gajamohan Mohanarajah <gajamohan.m <at> gmail.com> writes:

> 
> I still have not tried. Just checking any known issues since the
> processor is relatively new.
> 
> Best Regards,
> Gajan
> 
> On 24 August 2010 23:45, Gilles Chanteperdrix
> <gilles.chanteperdrix <at> xenomai.org> wrote:
> > Gajamohan Mohanarajah wrote:
> >> Hi,
> >>
> >> Is there any problems when using Xenomai on an i7 (-880 / Dell
> >> Precision T1500).
> >> Any tips/suggestions/experiences are most welcome.
> >
> > Why do you ask? Did you have any issue when trying it?
> >
> > --
> >                                            Gilles.
> >
>

 I tried Xenomai on several platforms with the following results:
 
1) Dell laptop with i7 processors, multicore
    Typical latency was around 5 usec,  but at least once per second had a 
long pass of approximately 700 usec.   Tried the xenomai SMI work around with 
no noticable change in results.   Many iterations of kernel settings, periodic 
high latency passes in all.  For a very good paper on i7 xenomai see:
http://download.intel.com/design/intarch/papers/322386.pdf  they had more luck 
than I did.
 
 
2) Atom processor laptop
    Typical latency around 5 usec,  this machine would run 10 minutes or so 
with low latency but would eventually give a single sporadic high latency pass.
 
3)   PC/104 atom processor
    This was our desired target from the start the other two were just 
experiments.
    Very good success.  Ran four test for two hours each.  1) Non-reatime side 
unloaded, 2) flood ping to generate interrupts, 3) cpuBurn to load processor, 
and 4) combination of both cpuBurn and flood ping.
 
Our conclusion is the results are very hardware dependent.       Can't really 
tell you which processor to use.  I can only give you our results.

Patrick Roberts






  parent reply	other threads:[~2010-08-25 15:01 UTC|newest]

Thread overview: 7+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2010-08-24 21:38 [Xenomai-help] Xenomai on intel core i7 Gajamohan Mohanarajah
2010-08-24 21:45 ` Gilles Chanteperdrix
2010-08-25  7:09   ` Gajamohan Mohanarajah
2010-08-25  7:19     ` Gilles Chanteperdrix
2010-08-25 15:01     ` Patrick Roberts [this message]
2010-08-25 15:49       ` Gilles Chanteperdrix
2010-08-25  4:16 ` Prakash A S

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=loom.20100825T165507-176@domain.hid \
    --to=patrick.m.roberts@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.