mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
From: john stultz <johnstul@us.ibm.com>
To: hauan@cmu.edu
Cc: Linux Kernel <linux-kernel@vger.kernel.org>
Subject: Re: 2.6.1 "clock preempt"?
Date: Tue, 20 Jan 2004 13:26:18 -0800	[thread overview]
Message-ID: <1074633977.16374.67.camel@cog.beaverton.ibm.com> (raw)
In-Reply-To: <1074630968.19174.49.camel@steinar.cheme.cmu.edu>

On Tue, 2004-01-20 at 12:36, Steinar Hauan wrote:
>   i've started to test the 2.6 series of kernels and observed a
>   strange thing: with moderate background load, the system clock
>   (i.e. time) seems to slow down to about 60% of normal speed
>   and the normally reliable ntp process (v4.2.0)
> 
>   details: working interactively with a couple of background dummy
>   processes (*1), my system clock slowed down approx 90 mins over
>   a period of approx 4 hrs real time.
>  
> (*1) infinite loop: 1 rip, 1 encode -- both run at nice 10
> 
>   the kernel logs show messages on the form:
> 
> localhost kernel: Losing too many ticks!
> localhost kernel: TSC cannot be used as a timesource.
>                        (Are you running with SpeedStep?)

How quickly do you see this message? Does it happen right at boot time,
or during load?


> localhost kernel: Falling back to a sane timesource.
> localhost kernel: set_rtc_mmss: can't update from 5 to 58
> 
>   without the background load, the system keeps perfect time.
> 
> ==> any ideas of what could be going on would be appreciated.

You seem to be losing lots of timer ticks. If this happens enough, the
system will fall back to the PIT thinking that your cpu frequency must
have changed w/o notification. 

You might want to try the attached patch to see if we're overreacting
when we fall back to the PIT or if your system really is having trouble.
Basically I'm just disabling the lost tick compensation code, which
means if your system is really having issues, you'll still lose time.

>   hardware details
>     Intel P4 2.53gz on Supermicro P4SAA mobo (Intel 7205 chipset)
>     1gb memory; multiple drives; Promise Ultra/133 raid controller
>   software
>     kernel 2.6.1 w/APIC and PREEMPT options turned on.

Also, do you see the problem when preempt is disabled (without using the
attached patch?)

thanks
-john

===== arch/i386/kernel/timers/timer_tsc.c 1.35 vs edited =====
--- 1.35/arch/i386/kernel/timers/timer_tsc.c	Wed Jan  7 00:31:11 2004
+++ edited/arch/i386/kernel/timers/timer_tsc.c	Tue Jan 20 13:22:54 2004
@@ -226,7 +226,7 @@
 	delta += delay_at_last_interrupt;
 	lost = delta/(1000000/HZ);
 	delay = delta%(1000000/HZ);
-	if (lost >= 2) {
+	if (0 && (lost >= 2)) {
 		jiffies_64 += lost-1;
 
 		/* sanity check to ensure we're not always losing ticks */



  reply	other threads:[~2004-01-20 21:27 UTC|newest]

Thread overview: 19+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2004-01-20 20:36 Steinar Hauan
2004-01-20 21:26 ` john stultz [this message]
2004-01-21 15:06   ` Steinar Hauan
2004-01-21 18:19     ` john stultz
2004-01-22 19:37       ` timothy parkinson
2004-01-22 19:42         ` john stultz
2004-01-22 19:50           ` timothy parkinson
2004-01-22 19:54             ` john stultz
2004-01-22 20:00               ` timothy parkinson
2004-01-22 21:21                 ` john stultz
2004-01-23 19:02                   ` timothy parkinson
2004-01-23 19:17                     ` john stultz
2004-01-23 19:36                       ` timothy parkinson
2004-01-23 20:06                         ` john stultz
2004-01-23 20:38                           ` timothy parkinson
2004-01-23 21:04                             ` john stultz
2004-01-27 21:30                             ` markus reichelt
2004-01-30 23:32                               ` Bartlomiej Zolnierkiewicz
2004-01-31  0:24                                 ` timothy parkinson

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=1074633977.16374.67.camel@cog.beaverton.ibm.com \
    --to=johnstul@us.ibm.com \
    --cc=hauan@cmu.edu \
    --cc=linux-kernel@vger.kernel.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 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®