mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
From: Robert Schwebel <r.schwebel@pengutronix.de>
To: Ingo Molnar <mingo@elte.hu>
Cc: linux-kernel@vger.kernel.org
Subject: Re: 2.6.19-rc6-rt4, changed yum repository
Date: Sun, 19 Nov 2006 15:42:42 +0100	[thread overview]
Message-ID: <20061119144242.GN18636@pengutronix.de> (raw)
In-Reply-To: <20061118163032.GA14625@elte.hu>

Ingo,

On Sat, Nov 18, 2006 at 05:30:32PM +0100, Ingo Molnar wrote:
> i've released the 2.6.18-rc6-rt4 tree, which can be downloaded from
> the usual place:

I have problems runinng cyclictest on later -rt kernels, up to -rt4.

This is a Dell Latitude D610, with Pentium M & ICH6 chipset.

Symptoms: 

- "cyclictest -n -p 90 -t 4" is just "running upwards"

  With 2.6.18-rt5 it looked like this, much better:

  isonoe:/home/rsc/tmp/rtpreempt/cyclictest-0.11# ./cyclictest -n -t 4 -p 90 -l 1000
  1.51 0.99 0.41 1/117 9885          
  
  T: 0 ( 9802) P:90 I:    1000 C:    1000 Min:       7 Act:      11 Avg:      16 Max:     561
  T: 1 ( 9803) P:89 I:    1500 C:     671 Min:       7 Act:      11 Avg:      15 Max:     498
  T: 2 ( 9804) P:88 I:    2000 C:     504 Min:       7 Act:      11 Avg:      16 Max:     533
  T: 3 ( 9805) P:87 I:    2500 C:     403 Min:       6 Act:      11 Avg:      13 Max:     394

- Using a relative timer (-r), I get huge latencies:

  isonoe:/home/rsc/tmp/rtpreempt/cyclictest-0.11# ./cyclictest -n -p 90 -t 4 -r -l 1000
  0.47 0.76 0.61 3/128 16112           
  
  T: 0 (15901) P:90 I:    1000 C:    1000 Min:    2963 Act:    2990 Avg:    4334 Max:    7039
  T: 1 (15902) P:89 I:    1500 C:     667 Min:    4003 Act:    6495 Avg:    6492 Max:    6541
  T: 2 (15903) P:88 I:    2000 C:     667 Min:    3498 Act:    5995 Avg:    5993 Max:    6041
  T: 3 (15904) P:87 I:    2500 C:     667 Min:    5433 Act:    5488 Avg:    5498 Max:    6898

This is with "lapic lapictimer" on the kernel commandline. I've manually
switched the clocksource from pit to tsc:

isonoe:/sys/devices/system/clocksource/clocksource0# cat available_clocksource 
jiffies tsc pit 
isonoe:/sys/devices/system/clocksource/clocksource0# cat current_clocksource 
tsc 

Some related dmesg output:

Kernel command line: root=/dev/sda7 ro lapic lapictimer
Found and enabled local APIC!
mapped APIC to ffffd000 (fee00000)
...
WARNING: experimental RCU implementation.
Clock event device pit configured with caps set: 07
...
Clock event device pit new caps set: 03
Clock event device lapic configured with caps set: 04
...
Time: pit clocksource has been installed.
...
*****************************************************************************
*                                                                           *
*  REMINDER, the following debugging options are turned on in your .config: *
*                                                                           *
*        CONFIG_CRITICAL_PREEMPT_TIMING                                     *
*        CONFIG_CRITICAL_IRQSOFF_TIMING                                     *
*        CONFIG_FUNCTION_TRACE                                              *
*                                                                           *
*  they may increase runtime overhead and latencies.                        *
*                                                                           *
*****************************************************************************

Last time this machine worked was with 2.6.18-rt5, 2.6.18-rt7 didn't work any
more. If it is helpful, I can do some further boot tests to find out which
patch broke it. 

Robert
-- 
 Dipl.-Ing. Robert Schwebel | http://www.pengutronix.de
 Pengutronix - Linux Solutions for Science and Industry
   Handelsregister:  Amtsgericht Hildesheim, HRA 2686
     Hannoversche Str. 2, 31134 Hildesheim, Germany
   Phone: +49-5121-206917-0 |  Fax: +49-5121-206917-9


  parent reply	other threads:[~2006-11-19 14:43 UTC|newest]

Thread overview: 8+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2006-11-18 16:30 Ingo Molnar
2006-11-19 14:39 ` Karsten Wiese
2006-11-19 13:43   ` Ingo Molnar
2006-11-19 20:56     ` Karsten Wiese
2006-11-19 21:14       ` Ingo Molnar
2006-11-21  0:04         ` Karsten Wiese
2006-11-19 14:42 ` Robert Schwebel [this message]
2006-11-20 14:14 ` Sergio Monteiro Basto

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=20061119144242.GN18636@pengutronix.de \
    --to=r.schwebel@pengutronix.de \
    --cc=linux-kernel@vger.kernel.org \
    --cc=mingo@elte.hu \
    /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®