Xorg uses 100% CPU

Bug #87351 reported by Martin Alias
10
Affects Status Importance Assigned to Milestone
linux-restricted-modules-2.6.20 (Ubuntu)
Won't Fix
Low
Unassigned

Bug Description

I'm testing Kubuntu Feisty Fawn and it's works ok in general, but when I leave the PC for a minutes may be an hour and I return to use it I find that the CPU is at 100% and making a "top" I could see that the problem is with xorg process, and after that the system doesnt respond, it just gets locked and I have to reset the PC

ProblemType: Bug
Date: Fri Feb 23 13:10:04 2007
DistroRelease: Ubuntu 7.04
Uname: Linux Athlon 2.6.18.2 #1 SMP Wed Nov 29 00:07:49 ART 2006 i686 GNU/Linux

Revision history for this message
Brian Murray (brian-murray) wrote :

Thanks for your bug report. Feisty Fawn, Ubuntu 7.04, actually uses the 2.6.20 kernel. Is your system up to date? Additionally, when you leave the system does a screen saver kick in? Thanks in advance.

Changed in xorg:
assignee: nobody → brian-murray
status: Unconfirmed → Needs Info
Revision history for this message
Martin Alias (tinchio) wrote :

Hi, yes what you say about the kernel its true, I already have the 2.6.20 kernel, but I'm using this because it's the one I have configured to use with vmware, but if you want I could try this 2.6.20 a few days to see whats happens, I believe it'ill be the same result. About the second question I don't understand it at all (english it's not my natural language); but I have a screensaver configured at 10 minutes and it's configured so that I don't have to type the password each time. If you could please ask this question with differents words i wouldl answer it.

Revision history for this message
Brian Murray (brian-murray) wrote :

Could you either try using the 2.6.20 kernel or turning off your screensaver? This will help us narrow down where the problem lies. Thanks again.

Revision history for this message
Martin Alias (tinchio) wrote :

Ok here i am again, I've tried both, not just turning off the screensaver but also using the 2.6.20 kernel, and still the same negative results, any other idea?

Revision history for this message
Timo Aaltonen (tjaalton) wrote :

could you attach the output of /usr/share/bug/xserver-xorg-core/script

Revision history for this message
Martin Alias (tinchio) wrote :

the /usr/share/bug/xserver-xorg/script is empty (there's no "xserver-xorg-core"). Any suggestion? Thanks for your time

Revision history for this message
Timo Aaltonen (tjaalton) wrote :

heh, indeed.. wait for the xorg-server-1.2 upload which will be in feisty shortly, and then try again :)

Revision history for this message
Timo Aaltonen (tjaalton) wrote :

ok, the new xorg-server is now in. The correct command to run is '/usr/share/bug/xserver-xorg-core/script 3>&1 file' and attach the file.

Revision history for this message
Martin Alias (tinchio) wrote :

Ok i did that step, here I attach the flie

Revision history for this message
Timo Aaltonen (tjaalton) wrote :

that's the script I want you to run with the parameters I gave you ;) That directs the output to a file, and then attach that file to this bug.

Revision history for this message
Brian Murray (brian-murray) wrote :

I have had a similar issue and executed the script creating the attached log file.

Revision history for this message
Timo Aaltonen (tjaalton) wrote :

Brian, from Xorg.0.log:

(II) fglrx(0): driver needs X.org 7.1.x.y with x.y >= 0.0
(II) fglrx(0): detected X.org 7.1.0.0
(EE) fglrx(0): GART is not initialized, disabling DRI
(WW) fglrx(0): ***********************************************
(WW) fglrx(0): * DRI initialization failed! *
(WW) fglrx(0): * (maybe driver kernel module missing or bad) *
(WW) fglrx(0): * 2D acceleraton available (MMIO) *
(WW) fglrx(0): * no 3D acceleration available *
(WW) fglrx(0): ********************************************* *

maybe that's what's your problem.

Revision history for this message
Martin Alias (tinchio) wrote :

Ok, i've just done the same, here i attach de log

Revision history for this message
Bryce Harrington (bryce) wrote :

Perhaps try reinstalling the fglrx module in synaptic?

Revision history for this message
Timo Aaltonen (tjaalton) wrote :

Tinchio: dri is disabled for you as well.

Revision history for this message
kurazu (kurazu) wrote :

Same 100% CPU utilisation problem here, after upgrading from Edgy to Feisty. Problem exists if I leave system for some (long) time, especially when I lock the screen. After coming back I'm able to unlock the screen, the desktop is running, superkaramba widgets are running smoothly, showing 100% CPU util, ktorrent seems to be running smoothly as well (I can see network traffic on one of the widgets). The icons on the desktop are displaying hover effect, but panels are not responding (hidden panels are not showing). Sometimes when I'm leaving for not so long, xorg CPU util. goes back to normal after few seconds of interaction with system. I switched screensaver to "blank screen" from "solar winds", but it didn't help.

Revision history for this message
animae (animae13) wrote :

Same problem here too...I think that it wappens when i am using azureus( at least this is the reason why i leave the computer open when i am not using it) and a couple of times that i left it whithout usng azureus it hasen't occured.. Next time it happened again i will attach the script

Revision history for this message
animae (animae13) wrote :

Here it is (this is what you want right???)

Revision history for this message
Conn O Griofa (psyke83) wrote :

Hi,

I'm not suffering from the issue reported in this bug, but I wanted to suggest that this bug gets unmarked as a duplicate of bug #88815. If people are still experiencing problems with Feisty, then it's not related to bug #88815, due to slowdown caused by xcb which has since been fixed (by not building with xcb support).

Revision history for this message
kurazu (kurazu) wrote :

I'm posting also output of the "ps -eouser,fname,pid,pcpu,pmem" command taken at the moment of 100% CPU utilisation happening (switched to console to do that). I agree with Conn, that this is not the same bug as #88815.

Changed in xorg:
assignee: brian-murray → nobody
Revision history for this message
sclose (scot-close) wrote :

I don't know if this is the same problem, but ever since I upgraded from Edgy to Feisty, every time I try to record in Audacity, my CPU usage goes to 100%, with Xorg being the reason. I never had this problem with Edgy. Otherwise, everything is normal. It is only when I press the record button in Audacity that the problem occurs.

Revision history for this message
animae (animae13) wrote :

I came and i saw that my cpu was 100% for once more BUT i didn't reboot it.I went out and when i came again the cpu usage was NORMAL....I think that this is pretty important information and i wanted to share it.Also for a little bit more safe restart keep opened and not minimized a terminal window so if it happend again just to write "sudo reboot"(Also the fact that you can still write on an opened terminal is also important no??)

Revision history for this message
kurazu (kurazu) wrote :

The bug with 100% cpu usage is happening also when working for long time in one window and not switching to another or to the desktop (fullscreen Battle for Wesnoth or fullscreen vmplayer). This one window is active, you can continue to work within it (performance is bad though). Sometimes, when one act fast and switch windows not long after 100% cpu usage starts, everything goes back to normal after few seconds.
This bug is really troublesome.

Revision history for this message
Jamey Sharp (sharpone) wrote :

This doesn't sound like an XCB-related bug, especially since Ubuntu currently isn't using XCB, so I've marked it no longer a duplicate of #88815.

Bryce Harrington (bryce)
Changed in xorg:
importance: Undecided → Low
status: Incomplete → Triaged
Revision history for this message
Patrick Salami (pat-entitycom) wrote :

it looks like this is a duplicate of bug 109507 and bug 120347, as well as bug 104481 and bug 51991.
I have exactly the same problem and I haven't found a solution yet, but I will try disabling superkaramba to see if the problem persists. If there is a path for superkaramba that addresses this issue (because according to the bug reports, a lot of people have reported that superkaramba is indeed the problem), that would be ideal.

Revision history for this message
kurazu (kurazu) wrote :

Disabling superkaramba helped on my machine. sk seems to be the problem.

Revision history for this message
Timo Aaltonen (tjaalton) wrote :

There are a lot of reasons why the GPU would soft lock (ie. mouse moves, Xorg eating 100% CPU and can be killed remotely), and this bug seemed to be about the binary driver fglrx (and apparently nvidia too). There's not much we can do about those, so I'm just filing this against linux-restricted-modules package.

Changed in xorg:
status: Triaged → Confirmed
Revision history for this message
Bryce Harrington (bryce) wrote : linux-restricted-modules-2.6.20 is obsolete

This package has become obsolete so we're closing out the bug report as WONTFIX.
Thanks for reporting it though!

Changed in linux-restricted-modules-2.6.20:
status: Confirmed → Won't Fix
To post a comment you must log in.
This report contains Public information  
Everyone can see this information.

Other bug subscribers

Remote bug watches

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