xorg consumes 99.9% CPU and hangs on Kubuntu Dapper

update on this:

updated via apt-get to edgy knot 3, now using kernel 2.6.17-9-386. I figured if I had a broken dapper that could not be fixed then possibly having a bleeding edgy on a running laptop would be a step up.

Unfortunately this hasn't fixed the problem. xorg is still having shooting up and sitting at 99.9% CPU and needs to be restarted remotely. The symptoms are mouse cursor working and multicolored 1px lines through the screen or completely messed up screen.
Can sometimes log into KDE but opening any app will cause xorg to hit 99.9% CPU

op - 23:01:01 up 2 min, 3 users, load average: 1.08, 0.62, 0.25
Tasks: 85 total, 3 running, 82 sleeping, 0 stopped, 0 zombie
Cpu(s): 99.7%us, 0.0%sy, 0.0%ni, 0.0%id, 0.0%wa, 0.0%hi, 0.3%si, 0.0%st
Mem: 515780k total, 212032k used, 303748k free, 9472k buffers
Swap: 1510068k total, 0k used, 1510068k free, 124524k cached

  PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND
 3917 root 25 0 96364 11m 2928 R 99.9 2.3 1:36.08 Xorg
    1 root 16 0 1632 532 444 S 0.0 0.1 0:01.38 init

running an strace on the process id show that X is doing the following :

sigreturn() = ? (mask now [])
--- SIGALRM (Alarm clock) @ 0 (0) ---
sigreturn() = ? (mask now [])
--- SIGALRM (Alarm clock) @ 0 (0) ---
sigreturn() = ? (mask now [])
--- SIGALRM (Alarm clock) @ 0 (0) ---
sigreturn() = ? (mask now [])
--- SIGALRM (Alarm clock) @ 0 (0) ---
sigreturn() = ? (mask now [])
--- SIGALRM (Alarm clock) @ 0 (0) ---
sigreturn() = ? (mask now [])
--- SIGALRM (Alarm clock) @ 0 (0) ---
sigreturn() = ? (mask now [])
--- SIGALRM (Alarm clock) @ 0 (0) ---
sigreturn() = ? (mask now [])
--- SIGALRM (Alarm clock) @ 0 (0) ---
sigreturn() = ? (mask now [])

ad nauseam....

Doing a google shows that some other people are finding a similar problem and there doesn't seem to be a fix, some people say yes this fixes it, other people say that fix doesn't work on my box.
Notably this massive thread at
http://www.nvnews.net/vbulletin/showthread.php?t=31858&page=27&pp=15

I am running the nv drivers, not the nvidia binary drivers and do not have the nvidia-kernel-common package installed. I stripped the system back to just nv in the hope that would fix.

other items in xorg.conf:

not loading glx
renderaccel is false / off

So hopefully someone can have a look at this and see what is happening....

Could you test this on Feisty or newer and determine if the problem is still occurring? If it is, do you have any further insights into what may causing the bug, so that we may report it upstream?

Hi Bryce

I haven't seen this since upgrading to later releases (edgy/feisty) and for the life of me can't remember the solution for getting a usable laptop between Knot 3 and Edgy stable.
The only issue I've had with X was the well documented bug #96430, which is unrelated to this. I suggest that this be closed unless it is being reported in duplicates.

Thanks
James

I have a system with the same problem.
It's running Dapper right now, should I upgrade it to Feisty?

The problem occured with the i810 driver at first, used for the onboard video on the mainboard. The screen would just go black. I could ssh into the box fine & CPU usage would settle after a while.

The screen still stayed black though.

I thought it may be a videodriver problem & hoped to solve the problem by adding a GeForce FX 5500, but it seems the same problem still exists.

Sounds like this was fixed in Edgy or Feisty, so am closing as fixed. If anyone is on Dapper LTS and need the fix backported for that, please nominate it for Dapper LTS. If you're not tied to Dapper LTS, it sounds like trying out Edgy or Feisty would be worthwhile for fixing this issue.

I am seeing a problem that sounds identical to this with Feisty.

I am using XUbuntu with XFCE and an nVidia MX 440 videocard. The problem manifests quickly and frequently when using Firefox, but this morning when I went to my machine, the screen froze halfway through drawing the screensaver login window. I don't have another computer in that house for remote access, but am now at my Windows PC and can SSH in without problems. Doing so, "top" reports Xorg at 99.9% cpu usage.

I can reproduce the problem easily and once I get my work laptop into the same house as my Linux server I will be happy to do anything required to reproduce the problem.

Adding Xorg logfile.

Hi,

I'm having this issue on an up to date Gutsy with kernel 2.6.22-14-386.
I reopen an comment.

I have the same problem as described above.
Xorg hangs with a CPU 99.9%.
Ctrl+Alt+Backspace doesn't work.
Mouse can be moved on the screen but no reactions of any programs.
I can access the computer with ssh over the network and did a Xorg.0.log-copy which I attach.
The message mainly is "select return 0"

I also attach a lspci-list.

Pls. tell me if I can help with some other information from my system.

By the way, to receive this log, I started Xorg by

startx -- -logverbose 999 -verbose 999

Here's an lspci output of my system.

This bug is very similar to bug 56184.
As this issue is discussed since several years on the german forum

http://forum.ubuntuusers.de/topic/88548/?highlight=h%C3%A4nger

and in many other forums I encrease the priority, I hope you don't mind.

closing again as fixed. f.dingler, you might want to open a new bug against xserver-xorg-video-sis.