mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
From: Ingo Molnar <mingo@elte.hu>
To: James R Bruce <bruce@andrew.cmu.edu>
Cc: Karim Yaghmour <karim@opersys.com>,
	Kristian Benoit <kbenoit@opersys.com>,
	linux-kernel@vger.kernel.org, paulmck@us.ibm.com, bhuey@lnxw.com,
	andrea@suse.de, tglx@linutronix.de, pmarques@grupopie.com,
	nickpiggin@yahoo.com.au, ak@muc.de, sdietrich@mvista.com,
	dwalker@mvista.com, hch@infradead.org, akpm@osdl.org,
	rpm@xenomai.org
Subject: Re: PREEMPT_RT vs ADEOS: the numbers, part 1
Date: Sun, 12 Jun 2005 16:54:01 +0200	[thread overview]
Message-ID: <20050612145401.GB15864@elte.hu> (raw)
In-Reply-To: <38010.210.137.194.5.1118573221.squirrel@210.137.194.5>


* James R Bruce <bruce@andrew.cmu.edu> wrote:

> Ingo, if you could document the right options required for decent 
> performace somewhere it would be quite helpful (maybe in 
> Documentation/rt-preempt?).  My first test of Preempt-RT showed 
> unexpectedly high overhead for a fairly benign network load (120 UDP 
> packets/sec), but that was likely the result of leaving some debugging 
> options on.

agreed, this needs to be addressed.

in the latest patch (-48-17 or later) i have changed the debugging 
options to default to off. (this wont turn them off if your .config has 
them turned on already, but will turn them off for new testers' 
.configs)

I also added a prominent boot-time message that, if certain 
high-overhead debugging options are enabled, says:

*****************************************************************************
*                                                                           *
* WARNING, the following debugging options are turned on in your .config:   *
*                                                                           *
*        CONFIG_DEBUG_RT_LOCKING_MODE                                       *
*        CONFIG_RT_DEADLOCK_DETECT                                          *
*        CONFIG_DEBUG_PREEMPT                                               *
*        CONFIG_CRITICAL_PREEMPT_TIMING                                     *
*        CONFIG_CRITICAL_IRQSOFF_TIMING                                     *
*        CONFIG_LATENCY_TRACE                                               *
*        CONFIG_DEBUG_SLAB                                                  *
*                                                                           *
* they may increase runtime overhead and latencies considerably!            *
*                                                                           *
*****************************************************************************

wrt. documentation - i'm not a big doc writer, but i'm taking patches
:-)

	Ingo

  reply	other threads:[~2005-06-12 14:55 UTC|newest]

Thread overview: 51+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2005-06-11  4:36 Kristian Benoit
2005-06-11  6:14 ` Nick Piggin
2005-06-11  9:15   ` Sven-Thorsten Dietrich
2005-06-11 14:15     ` Kristian Benoit
2005-06-12 15:48       ` Philippe Gerum
2005-06-11 14:39     ` Karim Yaghmour
2005-06-11 22:14       ` Philippe Gerum
2005-06-11 13:57   ` Kristian Benoit
2005-06-11 14:28   ` Zwane Mwaikambo
2005-06-11  7:08 ` Ingo Molnar
2005-06-11  7:44   ` Nick Piggin
2005-06-11  9:27     ` Sven-Thorsten Dietrich
2005-06-12 15:31       ` Philippe Gerum
2005-06-11 14:28     ` Kristian Benoit
2005-06-11 10:37   ` Ingo Molnar
2005-06-11 14:23     ` Kristian Benoit
2005-06-11 14:56       ` Ingo Molnar
2005-06-11 14:31   ` Karim Yaghmour
2005-06-11 14:52     ` Ingo Molnar
2005-06-11 14:52       ` Karim Yaghmour
2005-06-11 17:40       ` Karim Yaghmour
2005-06-11 18:15         ` Ingo Molnar
2005-06-11 18:34           ` Ingo Molnar
2005-06-11 22:27             ` Karim Yaghmour
2005-06-12 10:47           ` James R Bruce
2005-06-12 14:54             ` Ingo Molnar [this message]
2005-06-13  2:39               ` Kristian Benoit
2005-06-11 19:14 ` Ingo Molnar
2005-06-11 22:31   ` Karim Yaghmour
2005-06-12  6:11     ` Ingo Molnar
2005-06-12 15:26       ` Daniel Walker
2005-06-12 19:29       ` Karim Yaghmour
2005-06-12 20:59         ` Ingo Molnar
2005-06-13  0:45           ` Andrea Arcangeli
2005-06-13  1:20             ` Karim Yaghmour
2005-06-13  5:47               ` Ingo Molnar
2005-06-12 22:20     ` Andrea Arcangeli
2005-06-12 23:03       ` Sven-Thorsten Dietrich
2005-06-13  0:53         ` randy_dunlap
2005-06-13  1:12           ` Karim Yaghmour
2005-06-13  6:51           ` Sven-Thorsten Dietrich
2005-06-13 15:00     ` Ingo Molnar
2005-06-13 15:12       ` Kristian Benoit
2005-06-11 19:27 ` Ingo Molnar
2005-06-12  4:21   ` Karim Yaghmour
2005-06-11 20:14 ` Paul E. McKenney
2005-06-11 22:33   ` Karim Yaghmour
2005-06-12 20:36     ` Paul E. McKenney
2005-06-12 11:11 ` James R Bruce
2005-06-12 19:49   ` Karim Yaghmour
2005-06-13  2:32     ` Kristian Benoit

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=20050612145401.GB15864@elte.hu \
    --to=mingo@elte.hu \
    --cc=ak@muc.de \
    --cc=akpm@osdl.org \
    --cc=andrea@suse.de \
    --cc=bhuey@lnxw.com \
    --cc=bruce@andrew.cmu.edu \
    --cc=dwalker@mvista.com \
    --cc=hch@infradead.org \
    --cc=karim@opersys.com \
    --cc=kbenoit@opersys.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=nickpiggin@yahoo.com.au \
    --cc=paulmck@us.ibm.com \
    --cc=pmarques@grupopie.com \
    --cc=rpm@xenomai.org \
    --cc=sdietrich@mvista.com \
    --cc=tglx@linutronix.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

all inboxes | Powered by JetHome®