System freeze with white screen

Bug #46470 reported by Davor Cubranic
32
This bug affects 3 people
Affects Status Importance Assigned to Milestone
linux (Ubuntu)
Won't Fix
Undecided
Unassigned
linux-source-2.6.15 (Ubuntu)
Won't Fix
Medium
Unassigned
linux-source-2.6.20 (Ubuntu)
Won't Fix
Undecided
Unassigned

Bug Description

Dapper Flight 7 on a Shuttle ST20G5 (AMD64 and a Radeon X200 IGP)

I am experiencing occasional system freezes. It happens in X only -- the entire screen goes white and the computer does not respond any more to anything but power-cycle. I had the same issue in Breezy, then switched to Gentoo 2006.0 where everything worked fine, and now that I'm trying out Dapper I see the issue is still there.

I've had this happen with open-source Radeon driver on Dapper and ATI's fglrx drivers and the generic VESA on Breezy. As I already mentioned, this did not happen on Gentoo (both 2.6.15 and 2.6.16 kernels) nor on Windows XP sp2 (dual-boot).

Tags: cft-2.6.27
Revision history for this message
Davor Cubranic (cubranic) wrote :

The problem is still there with up-to-date 6.06 LTS and a 2.6.15-27 kernel. I have recently switched from the 64-bit kernel to 32-bit because some of the libraries I need come only in 32 bits, and the freezes seem to be getting more frequent. It usually happens within an hour, especially once I start up Firefox and/or Thunderbird.

This is a very serious issue for me, as it prevents me from using my Ubuntu box for any serious work. I would be happy to do any further investigation that would help in narrowing down the problem, but am at a loss at where to even start. I've never seen Linux crash so hard with a production kernel, there are no kernel panic messages on the screen that might indicate the source of the problem. I have suspected hardware fault, but I can run memtest for hours with no reported errors, and I drive the machine much harder in Windows XP than I do in Ubuntu and it never crashes. If anyone can advise what to do next, please let me know.

Revision history for this message
Davor Cubranic (cubranic) wrote :

I think I may have found the cause: enabling AMD Cool&Quiet (powernowd) when the internal graphics card is used. (Why Shuttle would design an AMD-based motherboard with an IGP that crashes when these two features are actually used? Good question.)

I said above that I never saw this crash in Windows, but that's because I had installed the AMD CPU drivers but not actually turned on C&Q (which on Windows is done through Power Options). Once I did turn it on, I saw the crash happen just as it did under Ubuntu and connected the dots with other reports on Shuttle's web forums. I then turned off powernowd init script and haven't had any crashes since. (Gentoo must have had it disabled by default.)

Therefore, I believe this bug can be closed. Once somebody sees this, that is. In six months that it's been open, there hasn't been a single comment by Ubuntu's team, other than changing the importance from 'unconfirmed' to 'medium'. I can appreciate the difficulty of keeping up with the increase in reports that comes with Ubuntu's growing popularity, but you should keep in mind that one of the surest ways for a project to look amateurish and turn off its users is to have slow and unresponsive bug triage.

I hope you can take a look at Mozilla's handling of incoming bugs, all of which get an initial response very quickly even if actually fixing them takes much longer.

Revision history for this message
Kees Cook (kees) wrote :

Just to clarify: if you disable powernowd, you no longer see the hangs, is that correct?

Revision history for this message
Davor Cubranic (cubranic) wrote :

> Just to clarify: if you disable powernowd, you no longer
> see the hangs, is that correct?

That's correct. I haven't had a hang since I turned it off, which is at least three weeks of almost daily use now.

Revision history for this message
Angus Kerr (angus-kerr) wrote :

Hi

I have the same issue with Feisty (7.04) wih kernel 2.6.20 and all updates installed.

System freezes to a white screen a few minutes after launch of Firefox (also up-to-date). The only recovery option is a hard reset. I have not looked at logs to see what happened in the kernel.

Disabling powernowd (sudo chmod -x /etc/init.d/powernowd) did nothing to help - the problem is still there. I am running an elitegroup ECS RS-482 Motherboard with built in ATI 200 Graphics card and AMD64 3500 proc.

System is currently unusable (but is rock-steady under Windows XP - runs for weeks not requiring a reboot). Would love to dump Windows, but need a stable Ubuntu system. I am considering other options - (Gentoo did not seem to have this problem)

Please keep this bug open. And anything else I can do on my side to get to the bottom of this and get the issue resolved.

Thanks

Revision history for this message
Davor Cubranic (cubranic) wrote :

Are you positive "chmod" disabled it? I believe powermod may be sourced in /etc/init.d/powernowd.early, which does not require the file to be executable. This is what I did in my powernowd instead:

--- /etc/init.d/powernowd 2007/05/09 06:32:08 1.3
+++ /etc/init.d/powernowd 2007/05/09 06:37:07
@@ -25,6 +25,10 @@
 set -e

 load_modules() {
+ # disable powernowd because it's causing crashes on ST20G5 with
+ # the on-board video
+ return 0
+
         #stop the kernel printk'ing at all while we load.
         PRINTK=`cat /proc/sys/kernel/printk`
         [ "$VERBOSE" = no ] && echo "1 1 1 1" > /proc/sys/kernel/printk
@@ -104,6 +108,10 @@
 }

 start() {
+ # disable powernowd because it's causing crashes on ST20G5 with
+ # the on-board video
+ return 0
+
                log_begin_msg "Starting $DESC... "
        if use_ondemand
        then

Revision history for this message
Angus Kerr (angus-kerr) wrote :

Thanks Davor

I used sysv-rc-conf and disabled powernowd. This solved the problem, I haven't had a crash since then.

Revision history for this message
Davor Cubranic (cubranic) wrote : Re: [Bug 46470] Re: System freeze with white screen

On Wednesday 10 October 2007 14:23:38 Angus Kerr wrote:
> Thanks Davor
>
> I used sysv-rc-conf and disabled powernowd. This solved the problem,
> I haven't had a crash since then.

I hadn't known about that tool. Neat. Glad you're system is back to
being stable. I know how good it felt for me.

Davor

Revision history for this message
Gareth Fitzworthington (mapping-gp-deactivatedaccount) wrote :

This bug has had no activity for a considerable period. This is a check to see if there is still interest in investigating this bug report.
Is this still an issue against Gutsy and/or Hardy ?
If so, does above mentioned workaround still apply?

Changed in linux-source-2.6.15:
status: New → Incomplete
Changed in linux-source-2.6.20:
status: New → Incomplete
Revision history for this message
Angus Kerr (angus-kerr) wrote :

This is still an issue with Gutsy and Hardy.

The workaround still applies. I am concerned that a lot of new users will be put off because of this and other driver related issues......Like with Gutsy the boot CD - X works in 640x480 mode. In Hardy, X is broken and you need to boot up in failsafe mode and then enable the protected driver.

I believe that it will affect a lot of users with built-in ATI Graphics cards. I have seen a lot of activity regarding the issue.

I will try this with Gentoo and other distros and report back.

Changed in linux-source-2.6.20:
assignee: nobody → ubuntu-kernel-team
status: Incomplete → Confirmed
Changed in linux-source-2.6.15:
assignee: nobody → ubuntu-kernel-team
status: Incomplete → Confirmed
Changed in linux-source-2.6.24:
assignee: nobody → ubuntu-kernel-team
status: New → Confirmed
Revision history for this message
Chuck Leduc (celeduc) wrote :

Also confirming that this is still an issue with Hardy.

Revision history for this message
Leann Ogasawara (leannogasawara) wrote :

The Ubuntu Kernel Team is planning to move to the 2.6.27 kernel for the upcoming Intrepid Ibex 8.10 release. As a result, the kernel team would appreciate it if you could please test this newer 2.6.27 Ubuntu kernel. There are one of two ways you should be able to test:

1) If you are comfortable installing packages on your own, the linux-image-2.6.27-* package is currently available for you to install and test.

--or--

2) The upcoming Alpha5 for Intrepid Ibex 8.10 will contain this newer 2.6.27 Ubuntu kernel. Alpha5 is set to be released Thursday Sept 4. Please watch http://www.ubuntu.com/testing for Alpha5 to be announced. You should then be able to test via a LiveCD.

Please let us know immediately if this newer 2.6.27 kernel resolves the bug reported here or if the issue remains. More importantly, please open a new bug report for each new bug/regression introduced by the 2.6.27 kernel and tag the bug report with 'linux-2.6.27'. Also, please specifically note if the issue does or does not appear in the 2.6.26 kernel. Thanks again, we really appreicate your help and feedback.

Revision history for this message
Indigo42 (john-indigo42) wrote :

I am thrilled to find an answer to this issue. I'm on an AMD Athlon 64 3200+ with a built in ATI card. I have tried every ATI driver/config combo under the sun to no avail.

I am willing to try this out and help you test. I am on 7.10 right now. I reinstalled from 8.04 in hopes to fix this. It seemed that 7.1 wasn't doing it until about 20 min ago. I had been up for 4 days with no problems. (mostly rsh'ed in though, screen stayed off most of the time.

Do I need to have update to 8.04?

I have nothing against upgrading if it's easier, or will give you better results for testing.

 Is there a "tried and true" way of crashing it that you know of? Seems like I only blow up in Firefox after a random amount of time.

I am anxious to help out! Untill then, I am going to try the workaround.

Please advise.

Revision history for this message
Indigo42 (john-indigo42) wrote :

Sorry...need to clarify.

I will try the kernel upgrade, I can't afford to run an Alpha on this box.

Where would I find the Kernel package?

Thanks,

Revision history for this message
Davor Cubranic (cubranic) wrote :

Kernel upgrade is not going to help you, see above in comment #6 how I disabled powernowd by editing /etc/init.d/powernowd.

Revision history for this message
Angus Kerr (angus-kerr) wrote :

NB: This email and its contents are subject to the Eskom Holdings Limited EMAIL LEGAL NOTICE

which can be viewed at http://www.eskom.co.za/email_legalnotice

------------------------------------------------------------------------

Hi There

I have posted a workaround here: http://mybroadband.co.za/vb/showthread.php?t=116504

You should be able to get going after following this post. I agree that the kernel upgrade will not fix the problem - but it's worth a try!

>>> Indigo42 <email address hidden> 02/09/08 08:42 >>>

Sorry...need to clarify.

I will try the kernel upgrade, I can't afford to run an Alpha on this

box.

Where would I find the Kernel package?

Thanks,

--

System freeze with white screen

https://bugs.launchpad.net/bugs/46470

You received this bug notification because you are a direct subscriber

of the bug.

Revision history for this message
Leann Ogasawara (leannogasawara) wrote :

The 18 month support period for Feisty Fawn 7.04 has reached it's end of life - http://www.ubuntu.com/news/ubuntu-7.04-end-of-life . As a result we are closing the linux-source-2.6.20 task.

Intrepid Ibex 8.10 recently came out last month. It contains an updated version of powernowd as well as the kernel. Might not hurt to give it a test if you are able to. Thanks.

Changed in linux-source-2.6.20:
status: Confirmed → Won't Fix
Revision history for this message
Launchpad Janitor (janitor) wrote : Kernel team bugs

Per a decision made by the Ubuntu Kernel Team, bugs will longer be assigned to the ubuntu-kernel-team in Launchpad as part of the bug triage process. The ubuntu-kernel-team is being unassigned from this bug report. Refer to https://wiki.ubuntu.com/KernelTeamBugPolicies for more information. Thanks.

Revision history for this message
Davor Cubranic (cubranic) wrote :

Leann, how about also closing this bug for source-2.6.15, since you closed it for 2.6.20? Five years on, I would actually prefer to see this bug closed to having it clog my bug list. It's not a kernel bug anyways, but a faulty hardware design by Shuttle computers.

Revision history for this message
Zsombor (gzsombor) wrote :

Unfortunately for me this bug is happens pretty frequently - with a graphics card: ATI Technologies Inc Radeon XPRESS 200M 5955 (PCIE), and with kernel @2.6.38-8-generic #42-Ubuntu SMP Mon Apr 11 03:31:50 UTC 2011 i686 athlon i386 GNU/Linux".
As I use my laptop as an emergency terminal, I can live with the frequent freezes, but it's a pain nevertheless.

Revision history for this message
Davor Cubranic (cubranic) wrote :

@Zsombor: it really is a hardware problem. You have to turn off AMD's "Cool and quiet" feature that slows down the CPU when it's idle to reduce power consumption. It looks like there is a real problem with the onboard graphics card when the CPU speeds back up again. I was getting white screens under both Linux and Windows until CnQ was turned off.

So either disable powernowd (run 'sudo update-rc.d ondemand disable') or turn off Cool-and-quiet in the BIOS if your laptop has that option.

Also, you don't mention your Ubuntu release, but judging by the kernel version, I assume it is Natty.

Revision history for this message
Zsombor (gzsombor) wrote :

Yes, it's Natty. I'm not totally sure that this is only hardware problem - I had this problem for a couple of releases, since the fglrx driver stopped supporting my pretty old graphic card, so I though this is a deficiency of the open source driver. As there is nothing in the log files, after it happens, it can be anything, it's a very fortunate accident, that I found this bug report / solution. It would be nice, if this cool and quiet feature can be disabled automaticly on the old machines, or some kernel log, that list's the last operation which tried to perform, something like that you can get after a reboot :
$ cat /proc/lastmsg
x.y.driver : if you see this, there can be a hardware problem, during executing Z hardware functionality

I know, it can potentially slow down the system, if these messages needs to be written to the disk every time, but it would be nice, wouldn't it ? :)

Revision history for this message
Zsombor (gzsombor) wrote :

Unfortunately, it seems that this happens with the performance CPU governor (after disabling the ondemand with that update-rc.d script). The only difference, that the laptop become hot, and the ventilator never stops - so the performance cpu frequence governor is working, I'm sure). And firefox (or the terminal, or the chromium) freeze pretty frequently, when I want to scroll up or down. Not always, but nearly after 20-30 tries.

$ uname -a
Linux arrakis 2.6.38-8-generic #42-Ubuntu SMP Mon Apr 11 03:31:50 UTC 2011 i686 athlon i386 GNU/Linux

Revision history for this message
Brad Figg (brad-figg) wrote :

This bug was opened against a version of the kernel that is no longer supported. If there is a kernel issue w.r.t. this problem on a supported kernel version, please open a new bug.

Changed in linux (Ubuntu):
status: Confirmed → Won't Fix
Changed in linux-source-2.6.15 (Ubuntu):
status: Confirmed → Won't Fix
To post a comment you must log in.
This report contains Public information  
Everyone can see this information.

Duplicates of this bug

Other bug subscribers

Remote bug watches

Bug watches keep track of this bug in other bug trackers.