public inbox for linux-kernel@vger.kernel.org
 help / color / mirror / Atom feed
From: Karim Yaghmour <karim@opersys.com>
To: Lee Revell <rlrevell@joe-job.com>
Cc: stefan.eletzhofer@eletztrick.de,
	linux-kernel <linux-kernel@vger.kernel.org>,
	Philippe Gerum <rpm@xenomai.org>
Subject: Re: [ANNOUNCE] Linux 2.6 Real Time Kernel
Date: Sat, 09 Oct 2004 16:53:26 -0400	[thread overview]
Message-ID: <41684FC6.1070803@opersys.com> (raw)
In-Reply-To: <1097352885.1428.60.camel@krustophenia.net>


Lee Revell wrote:
> In theory, I think yes, if all IRQs on the system run in threads except
> the saw interrupt, and the RT task that controls the saw runs at a
> higher priority than all the IRQ threads.  You can guarantee that other
> interrupts won't delay the saw, because the saw irq is the only thing on
> the system that runs in interrupt context.  With the current VP
> implementation you are still bounded by the longest non-preemptible code
> path in the kernel AKA the longest time that a spinlock is held. 
> Replacing most spinlocks with mutexes reduces this to less than 20 code
> paths according to Mvista, which then can be individually audited for
> RT-safeness. 
> 
> That being said, no way would I put my hand under the saw with the
> current implementation.  But, unless I am missing something, it seems
> like this kind of determinism is possible with the Mvista design.

It may be a question of taste, but even if that did work, which I am
not convinced of, it seems to me that it's awfully convoluted.
With the current interrupt pipeline mechanism part of Adeos, on
which RTAI and RTAI fusion are built, I can give you absolute hard-rt
deterministic guarantees while keeping the spinlocks intact, and not
having to check for the rt-safeness of any part of the kernel. You
just write the time-sensitive saw driver int handler in front of
Linux in the ipipe and you're done: 100% deterministic hard-rt,
regardless of the application load and the driver set.

> I will check that out, I have not looked at RTAI in over a year.

Here are some interesting links:

RTAI/fusion presentation by Philipppe Gerum last July (see slide 25
for some interesting numbers):
http://www.enseirb.fr/~kadionik/rmll2004/presentation/philippe_gerum.pdf
Here's a thread that explains the details about RTAI/fusion:
https://mail.rtai.org/pipermail/rtai/2004-June/thread.html#7909
Here's the ipipe core API:
http://home.gna.org/adeos/doc/api/interface_8h.html

Karim
-- 
Author, Speaker, Developer, Consultant
Pushing Embedded and Real-Time Linux Systems Beyond the Limits
http://www.opersys.com || karim@opersys.com || 1-866-677-4546



  reply	other threads:[~2004-10-09 20:51 UTC|newest]

Thread overview: 96+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2004-10-09  5:59 [ANNOUNCE] Linux 2.6 Real Time Kernel Sven-Thorsten Dietrich
2004-10-09  6:40 ` Lee Revell
2004-10-09  7:33   ` Daniel Walker
2004-10-09  7:42     ` Lee Revell
2004-10-09 23:40       ` Matthias Urlichs
2004-10-09  8:52     ` Lee Revell
2004-10-09 23:20     ` Dave Hansen
2004-10-09 23:24       ` Lee Revell
2004-10-09 10:51 ` Måns Rullgård
2004-10-09 13:15 ` Måns Rullgård
2004-10-09 21:20   ` Lee Revell
2004-10-09 21:35     ` Måns Rullgård
2004-10-09 21:37       ` Lee Revell
2004-10-09 21:45         ` Måns Rullgård
2004-10-09 21:55       ` Lee Revell
2004-10-09 22:21         ` Måns Rullgård
2004-10-09 23:52       ` Lee Revell
2004-10-10  0:05         ` Måns Rullgård
2004-10-10  0:45           ` Lee Revell
2004-10-10  1:05             ` Måns Rullgård
2004-10-10  1:09               ` Lee Revell
2004-10-10  0:43       ` Micha Feigin
2004-10-10  1:08         ` Måns Rullgård
2004-10-09 17:41 ` Karim Yaghmour
2004-10-09 18:30   ` Lee Revell
2004-10-09 21:26     ` stefan.eletzhofer
2004-10-09 19:30       ` Lee Revell
2004-10-09 19:38         ` Måns Rullgård
2004-10-09 21:38         ` stefan.eletzhofer
2004-10-09 19:47           ` Lee Revell
2004-10-09 20:11             ` Karim Yaghmour
2004-10-09 20:14               ` Lee Revell
2004-10-09 20:53                 ` Karim Yaghmour [this message]
2004-10-09 20:59                   ` Lee Revell
2004-10-09 20:20             ` Robert Love
2004-10-09 20:25               ` Lee Revell
2004-10-10  1:15 ` Lee Revell
2004-10-10  8:46 ` Ingo Molnar
2004-10-10 19:41   ` Daniel Walker
2004-10-10 19:46     ` Ingo Molnar
2004-10-10 21:20     ` Andrew Morton
2004-10-10 21:59       ` Ingo Molnar
2004-10-11 17:53         ` Daniel Walker
2004-10-11 20:49           ` Ingo Molnar
2004-10-11 21:44             ` Sven Dietrich
2004-10-11 21:54               ` Ingo Molnar
2004-10-11 23:05                 ` Sven Dietrich
2004-10-12  5:50                   ` Ingo Molnar
2004-10-14  5:09                     ` Dipankar Sarma
2004-10-14  7:18                       ` Ingo Molnar
2004-10-15 14:59                         ` Paul E. McKenney
2004-10-15 15:45                           ` Ingo Molnar
2004-10-15 16:40                             ` Paul E. McKenney
2004-10-15 16:45                               ` Paul E. McKenney
2004-10-17 17:12                       ` Ingo Molnar
2004-10-12 18:50             ` Daniel Walker
2004-10-12 19:46               ` [Ext-rt-dev] " Thomas Gleixner
2004-10-12 20:31                 ` Sven Dietrich
2004-10-12 20:37                   ` Thomas Gleixner
2004-10-13  0:30                   ` George Anzinger
2004-10-12 21:12                 ` Bill Huey
2004-10-12 21:24                   ` Bill Huey
2004-10-12 21:32                     ` Thomas Gleixner
2004-10-12 23:13                       ` Bill Huey
2004-10-12 21:41                   ` Sven Dietrich
2004-10-12 22:57                     ` Bill Huey
2004-10-12 23:17                       ` Adam Heath
2004-10-12 23:36                         ` Lee Revell
2004-10-12 23:25                       ` Thomas Gleixner
2004-10-13  2:02                       ` K.R. Foley
2004-10-13 13:39                       ` Martijn Sipkema
2004-10-13 13:26                         ` La Monte H.P. Yarroll
2004-10-13 15:04                           ` Martijn Sipkema
2004-10-13 14:55                             ` Kurt Wall
2004-10-13 14:52                               ` Christoph Hellwig
2004-10-13 15:56                         ` Lee Revell
2004-10-13 16:13                           ` Robert Love
2004-10-13 17:14                           ` Martijn Sipkema
2004-10-13  3:55                     ` Bill Huey
2004-10-12 22:00                   ` Thomas Gleixner
2004-10-12 22:36                     ` Bill Huey
2004-10-12 23:10                       ` Thomas Gleixner
2004-10-12 23:33                         ` Bill Huey
2004-10-12 23:37                           ` Thomas Gleixner
2004-10-12 23:52                             ` Bill Huey
2004-10-13  0:59                         ` Valdis.Kletnieks
2004-10-10 12:21 ` John Richard Moser
2004-10-10 17:26   ` Lee Revell
2004-10-10 18:45     ` John Richard Moser
2004-10-10 20:20       ` Ingo Molnar
2004-10-10 20:44         ` John Richard Moser
2004-10-10 17:29   ` Daniel Walker
  -- strict thread matches above, loose matches on Subject: below --
2004-10-09 11:38 John Hedditch
2004-10-09 22:54 Sven Dietrich
2004-10-11 15:28 Vadim Lebedev
2004-10-11 15:50 ` Eugeny S. Mints

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=41684FC6.1070803@opersys.com \
    --to=karim@opersys.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=rlrevell@joe-job.com \
    --cc=rpm@xenomai.org \
    --cc=stefan.eletzhofer@eletztrick.de \
    /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