View previous topic :: View next topic |
Author |
Message |
pokkie n00b
Joined: 29 Jun 2004 Posts: 13 Location: Reading, UK
|
Posted: Tue Jun 29, 2004 3:29 pm Post subject: Warning : many lost ticks, time source seems to be instable |
|
|
Hello all,
I have recently begun the long process of installing gentoo on my new Athlon 64. When I try to run
I get the following error msg repeatedly :
Code: | Warning: many lost ticks.
Your time source seems to be instable or some driver is hogging resources.
|
I am using gcc-3.3.3 on the latest 2.6.7-r6 kernel.
Any help would be greatly appreciated.
Regards,
-- Pokkie |
|
Back to top |
|
|
Geoffrey n00b
Joined: 25 Jun 2004 Posts: 6
|
Posted: Tue Jun 29, 2004 3:34 pm Post subject: |
|
|
I personally don't have an AMD 64, but I think your computer is saying that your system clock is having problems. Try typing "date" on the shell, and then typing it again 1 hour later and seeing if your clock is consistent. |
|
Back to top |
|
|
pokkie n00b
Joined: 29 Jun 2004 Posts: 13 Location: Reading, UK
|
Posted: Tue Jun 29, 2004 3:36 pm Post subject: |
|
|
Righto, will try that.
Thanks for the suggestion.
-- Pokkie |
|
Back to top |
|
|
pokkie n00b
Joined: 29 Jun 2004 Posts: 13 Location: Reading, UK
|
Posted: Tue Jun 29, 2004 5:34 pm Post subject: |
|
|
Rigto, timed my clock for 1 hour and it didn't loose 1 second.
Any other suggestion would be most welcome.
Regards,
-- Pokkie |
|
Back to top |
|
|
asjapulk n00b
Joined: 15 Apr 2004 Posts: 71
|
Posted: Tue Jun 29, 2004 6:12 pm Post subject: DOT |
|
|
I have the same problem. It kind of went away when I disabled dynamic overclocking (MSI Neo FIS2R m/b), but now it's back again.
I also have another problem:
I get
Code: | * Failed to set hardware clock to system clock |
at boot time?
I set my time with rdate -s xxx.xx right but at next boot it's wrong again (+3 hours). |
|
Back to top |
|
|
Pink Veteran
Joined: 24 Jul 2003 Posts: 1062
|
Posted: Tue Jun 29, 2004 6:14 pm Post subject: |
|
|
I don't use a 64 but there is an option in the kernel which the notes say 'fixes' the lost ticks problem.
You may want to try that, (don't forget to mount /boot before copying over the new bzImage )
It's under Power Management -> acpi -> Power Management Timer Support.
This may help, but as I say I haven't got a 64... |
|
Back to top |
|
|
pokkie n00b
Joined: 29 Jun 2004 Posts: 13 Location: Reading, UK
|
Posted: Tue Jun 29, 2004 9:30 pm Post subject: Re: DOT |
|
|
asjapulk wrote: | I have the same problem. It kind of went away when I disabled dynamic overclocking (MSI Neo FIS2R m/b), but now it's back again. |
Heh, I just disabled dynamic overclocking on my MSI Neo Platinum and the errors seem to have disapeared. I also changed my memory timing, i think it is called, from T2 to T1.
PickledOnion wrote: | It's under Power Management -> acpi -> Power Management Timer Support. |
Should it be enabled or disabled?
Regards,
-- Pokkie |
|
Back to top |
|
|
fumtu99 n00b
Joined: 05 Sep 2002 Posts: 59
|
Posted: Wed Jun 30, 2004 5:05 am Post subject: |
|
|
Um, I don't think it makes any difference, if you're running in 64-bit mode: that option ("Power Management Timer Support") doesn't appear in the Power Management->ACPI menu, if you're using the x86_64 architecture, and the only place where the flag would be used is in the arch/i386/kernel part of the source tree. This is bad news for me: I'm getting tons of those messages in my log file. I suspect that the only way I can disable it is by disabling frequency scaling, something I'd sort of like to keep for my notebook, just to keep it from burning up batteries quite so quickly. The funny thing is that if I run in 32-bit mode, I don't seem to need it; it might be time to submit a bug report, but I think that the bug is not in Gentoo, per se, but rather in the current state of the 64-bit kernel, so I'm not sure where I'd submit it.
James |
|
Back to top |
|
|
RAPUL l33t
Joined: 29 Dec 2002 Posts: 664 Location: Valencia (SPAIN)
|
Posted: Wed Jun 30, 2004 1:46 pm Post subject: Bios time. |
|
|
Change time in BIOS.
Sometimes a corruption of CMOS gives unexpected errors. _________________ Entropy rulz world.
Redundancy sux.
World is full of redundancy.
World sux. |
|
Back to top |
|
|
thaldyron Apprentice
Joined: 25 Sep 2002 Posts: 227 Location: On Earth
|
Posted: Wed Jun 30, 2004 4:07 pm Post subject: Re: Bios time. |
|
|
RAPUL wrote: | Change time in BIOS.
Sometimes a corruption of CMOS gives unexpected errors. |
I'm having exactly the same problem so I think it's rather a kernel issue than everyones' CMOS clock went nuts.
When I boot 2.6.5-r1 (64bit) the warning doesn't occur so it seems to be a bug that was introduced in gentoo-dev-2.6.7. |
|
Back to top |
|
|
asjapulk n00b
Joined: 15 Apr 2004 Posts: 71
|
Posted: Wed Jun 30, 2004 4:28 pm Post subject: still |
|
|
My "Lost ticks" disappeared when I installed right driver in kernel for my PATA disk.
Thou still got the problem with * Failed to set hardware clock to system clock" at boot and shutdown:(
So if you get
Code: | Warning: many lost ticks.
Your time source seems to be instable or some driver is hogging resources. |
You can:
1. Disable Dynamic overclocking @BIOS
2. Make sure that you have right device drivers installed in kernel. (IDE etc) |
|
Back to top |
|
|
thaldyron Apprentice
Joined: 25 Sep 2002 Posts: 227 Location: On Earth
|
Posted: Thu Jul 01, 2004 1:54 am Post subject: Re: still |
|
|
asjapulk wrote: |
You can:
1. Disable Dynamic overclocking @BIOS
2. Make sure that you have right device drivers installed in kernel. (IDE etc) |
I didn't change any IDE drivers etc, and my notebook isn't overclockable anyways. I really think this is a kernel issue since so _many_ fellows are having the same problems.
Edit: Just tried gentoo-dev-sources-2.6.7-r7 but the problem remains. |
|
Back to top |
|
|
jman n00b
Joined: 02 Jul 2004 Posts: 2
|
Posted: Fri Jul 02, 2004 6:48 am Post subject: Asus K8V SE Deluxe, or any VIA chipset board. |
|
|
I have a hunch that most of you are using a VIA chipset board. I have an Asus K8V SE Deluxe, and as soon as I recompiled my kernel with VIA VP IDE support, and disabled all of the other IDE drivers the problem is gone. I was having a lot of stability issues as well, but they seem to have cleared up. For those with notebooks, it's the speed-step type feature that is causing problems. There is a setting somewhere in the kernel options to fix it. For those with desktops, I guess the "Cool-n-Quiet" feature could cause the same kind of issues, but it's enabled on my system and it seems to be fine. In closing, I'm not so sure this is as much a kernel issue as it is a configuration issue. With new technology we have to worry about bugs in the code, and in our configuration. Good luck everyone. There is no question that Gentoo is the best way to flex your AMD64's muscles! |
|
Back to top |
|
|
Flappie n00b
Joined: 01 Jul 2004 Posts: 8 Location: Netherlands
|
Posted: Fri Jul 02, 2004 11:29 am Post subject: |
|
|
Nopes i have the same problem and i have a sis chipset |
|
Back to top |
|
|
torklingberg Tux's lil' helper
Joined: 30 May 2004 Posts: 86
|
Posted: Fri Jul 02, 2004 3:23 pm Post subject: |
|
|
Similar problem here, but the warning in dmesg is:
Code: |
warning: many lost ticks.
Your time source seems to be instable or some driver is hogging [b]interupts[/b].
Losing some ticks... checking if CPU frequency changed.
|
I have an Asus K8V motherboard, and gentoo-dev-sources-2.6.7-r7. I have frequency scaling enabled in kernel config, but I would like to keep it. I do not get this warning from a 32-bit x86 kernel.
Has anyone filed it in Bugzilla? Or found out what config error causes this? |
|
Back to top |
|
|
jman n00b
Joined: 02 Jul 2004 Posts: 2
|
Posted: Fri Jul 02, 2004 3:37 pm Post subject: |
|
|
For those who are sure they have the right IDE drivers, then it is a frequency scaling issue. "Cool-n-Quiet" on desktops. It does work fine on mine though. |
|
Back to top |
|
|
crontab n00b
Joined: 13 Aug 2003 Posts: 6 Location: Netherlands
|
Posted: Fri Jul 02, 2004 4:41 pm Post subject: |
|
|
Hi,
I've had the same problem.
I solved it by compiling the via82xx into the kernel not as a module. _________________ #!/usr/bin/perl -w
use Gentoo;
require AMD64 qw(3200+); |
|
Back to top |
|
|
thaldyron Apprentice
Joined: 25 Sep 2002 Posts: 227 Location: On Earth
|
Posted: Fri Jul 02, 2004 5:44 pm Post subject: |
|
|
crontab wrote: | Hi,
I've had the same problem.
I solved it by compiling the via82xx into the kernel not as a module. |
This didn't help me either. I'm pretty sure i've got the right IDE drivers because I didn't have that problem with 2.6.5...
@torklingberg: I'm getting exactly the same error msg. |
|
Back to top |
|
|
TiccTech n00b
Joined: 13 Jun 2004 Posts: 42 Location: Oregon, US
|
Posted: Sat Jul 03, 2004 6:22 pm Post subject: |
|
|
The same problem is biting me -- 2.6.5 kernel works fine; 2.6.7-r6 or 2.6.7-r7 both lose ticks. Kernels compiled with the same (as close as possible) options. SiS760/964 chipset.
I also have a problem on 2.6.7 of the HID module failing to load. I didn't see anyone else mention that so it's not necessarily related.
Anyway, I guess it's time to try building some different kernels to see if that helps...
-TiccTech |
|
Back to top |
|
|
thaldyron Apprentice
Joined: 25 Sep 2002 Posts: 227 Location: On Earth
|
Posted: Sat Jul 03, 2004 9:20 pm Post subject: |
|
|
TiccTech wrote: |
I also have a problem on 2.6.7 of the HID module failing to load. I didn't see anyone else mention that so it's not necessarily related.
|
The hid module is called "usbhid" in 2.6.7! |
|
Back to top |
|
|
TiccTech n00b
Joined: 13 Jun 2004 Posts: 42 Location: Oregon, US
|
Posted: Sat Jul 03, 2004 11:32 pm Post subject: |
|
|
thaldyron wrote: | The hid module is called "usbhid" in 2.6.7! |
I'd missed that change -- thanks for the tip!
Where is that kind of change usually documented?
And, getting back on topic for this thread,
I seem to have gotten rid of my "many lost ticks" problem. I'd gone through my kernel config and attempted to clean up a bunch of stuff. So I don't know exactly which change(s) made the difference, and I probably won't bother to isolate the problem.
Here's a list of what I changed, in case it's of help to anyone still trying to track down this problem on their system. Left-hand .config reports zillions of "many lost ticks" messages, right-hand .config is clean. 2.6.7-r7 kernel, SiS760/964 chipset.
-TiccTech
Code: | # diff .config.old .config
28,29c28,29
< # CONFIG_POSIX_MQUEUE is not set
< # CONFIG_BSD_PROCESS_ACCT is not set
---
> CONFIG_POSIX_MQUEUE=y
> CONFIG_BSD_PROCESS_ACCT=y
55c55
< # CONFIG_KMOD is not set
---
> CONFIG_KMOD=y
167,170c167,168
< CONFIG_BLK_DEV_RAM=y
< CONFIG_BLK_DEV_RAM_SIZE=4096
< CONFIG_BLK_DEV_INITRD=y
< CONFIG_LBD=y
---
> # CONFIG_BLK_DEV_RAM is not set
> # CONFIG_LBD is not set
183c181
< CONFIG_IDEDISK_MULTI_MODE=y
---
> # CONFIG_IDEDISK_MULTI_MODE is not set
315,318c313
< CONFIG_FUSION=y
< CONFIG_FUSION_MAX_SGE=40
< # CONFIG_FUSION_ISENSE is not set
< # CONFIG_FUSION_CTL is not set
---
> # CONFIG_FUSION is not set
449,450c444
< CONFIG_AMD8111_ETH=y
< # CONFIG_AMD8111E_NAPI is not set
---
> # CONFIG_AMD8111_ETH is not set
580c574
< CONFIG_SERIAL_8250_CONSOLE=y
---
> # CONFIG_SERIAL_8250_CONSOLE is not set
589d582
< CONFIG_SERIAL_CORE_CONSOLE=y
620,621c613
< CONFIG_RAW_DRIVER=y
< CONFIG_MAX_RAW_DEVS=256
---
> # CONFIG_RAW_DRIVER is not set
919c911
< CONFIG_USB_HID=y
---
> CONFIG_USB_HID=m
922a915,920
>
> #
> # USB HID Boot Protocol drivers
> #
> # CONFIG_USB_KBD is not set
> # CONFIG_USB_MOUSE is not set
1017,1018c1015,1016
< CONFIG_AUTOFS_FS=y
< # CONFIG_AUTOFS4_FS is not set
---
> # CONFIG_AUTOFS_FS is not set
> CONFIG_AUTOFS4_FS=m
1171c1169
< CONFIG_CRC32=y
---
> CONFIG_CRC32=m
|
|
|
Back to top |
|
|
herring Apprentice
Joined: 22 Jul 2002 Posts: 187 Location: Norway
|
Posted: Mon Jul 12, 2004 10:40 am Post subject: |
|
|
MSI K8N Neo platinum, AMD64 3200+ and FX5700ULTRA. stage1 installation from gcc34-amd64-2004.2-test3. Nothing fancy; CFLAGS="-march=k8 -O2 -pipe", CHOST="x86_64-pc-linux-gnu", USE="", ACCEPT_KEYWORDS="amd64".
Im only getting "lost ticks" after starting xorg-x11. I am using nvidia driver (ver 1.0.5332-r1) .
Could it be this driver, or xorg-x11 itself being task-intensive ? |
|
Back to top |
|
|
herring Apprentice
Joined: 22 Jul 2002 Posts: 187 Location: Norway
|
Posted: Mon Jul 12, 2004 11:02 am Post subject: |
|
|
Starting xorg-x11 without nvidia (using a vga driver 600x480) produced lost ticks as well...
EDIT> Btw I am using gentoo-dev-sources (linux-2.6.7-gentoo-r7) and fluxbox
Not sure if this info is useful:
From within xorg-x11 (X) running tail -f /var/log/messages in a xterm to monitor new tics
Iam not able to produce lost tics by running intensive tasks as emerging or starting programs like firefox
However, if I run glxinfo or glxgear, I am able to reproduce 2-3 error messages every time.
Last edited by herring on Mon Jul 12, 2004 3:02 pm; edited 1 time in total |
|
Back to top |
|
|
vyzivus Apprentice
Joined: 05 Jul 2004 Posts: 173 Location: Slovakia
|
Posted: Mon Jul 12, 2004 12:41 pm Post subject: |
|
|
Yo!
I have exactly the same problem with my acer aspire 1501LMi. I tried all options from .config suggested by TiccTech (thanks for tips for useless devices however the messages didn't disappear. Well I haven't checked them all: I don't see the point in compiling USB_HID as a module.
I tried to turn the CPU frequency scaling off but it was a disaster: avg temp was 5C higher than normal and fan never stopped. Only thing that disappeared was the line:
Losing some ticks... checking if CPU frequency changed.
I already posted the bug to amd64 gentoo devs, just check out
https://bugs.gentoo.org/show_bug.cgi?id=56073
I just compiled with AUTOFS4_FS=m, let's see what it does. _________________ I thought what I'd do was, I'd pretend I was one of those deaf-mutes or should I? |
|
Back to top |
|
|
vyzivus Apprentice
Joined: 05 Jul 2004 Posts: 173 Location: Slovakia
|
Posted: Mon Jul 12, 2004 1:30 pm Post subject: |
|
|
Ok, the avg fan was 15C higher above normal (50C), and compiling with AUTOFS4_FS=m didn't help either. I even tried compiling via8233 sound (i hope that was meant by via82xx , but ticks are still merciless. 2.6.7-r9 kernel, "VT82C586A/B/VT82C686/A/B/VT823x/A/C PIPC Bus Master IDE", using "VIA82CXXX chipset support" kernel driver. _________________ I thought what I'd do was, I'd pretend I was one of those deaf-mutes or should I? |
|
Back to top |
|
|
|