Bugzilla – Bug 593463
[i855GM] Flickering and rendering Corruption when Desktop Effects are enabled.
Last modified: 2012-08-02 15:57:26 UTC
User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.9.0.15) Gecko/2009102100 SUSE/3.0.15-0.1.2 Firefox/3.0.15 A Fujitsu-Siemens Amilo 7400M laptop can not boot to X with an of F3/F4 live CD boot options in 32-bit openSUSE-11.3 milestone4 live CD. It is possible to boot to run level 3. This bug report may be similar to a bug report raised on openSUSE-11.2 (passed upstream), except its symptoms are more serious in 11.3. 11.2 similar bug report is here: https://bugzilla.novell.com/show_bug.cgi?id=555619 In openSUSE-11.2 it was possible to boot with option acpi=off, or with patience boot in Safe Mode. This does not work in 11.3 milestone4. According to the bug raised on the kernel, there may be a patch, but custom compiling a kernel is beyond my expertise: https://bugs.freedesktop.org/show_bug.cgi?id=24789 Output of "lspci -nn" on this laptop, illustrating graphic hardware detail: 00:02.0 VGA compatible controller [0300]: Intel Corporation 82852/855GM Integrated Graphics Device [8086:3582] (rev 02) I will understand if this bug report is closed due to the problem being upstream, but I thought important I do document the bug still existing in openSUSE 11.3. Reproducible: Always Steps to Reproduce: 1. Simple boot from a liveCD, any boot option. 2. 3. Actual Results: PC always fails to boot X. . Expected Results: PC should be able to boot to X. It can not with any option. This laptop boots fine with openSUSE-11.1, but
I should clarify 1st line of previous post: A Fujitsu-Siemens Amilo 7400M laptop (with Intel 855GM graphics) can not boot to X with any of the KDE4 32-bit openSUSE-11.3 Milestone4 live CD boot options (ie none of the options obtained by pressing F3/F4 will help).
I tested this Fujitsu-Siemens Amilo 7400M laptop (with Intel 855GM graphics) again with 11.3 Milestone6 (build 0577) 32-bit openSUSE KDE4 liveCD, trying various configurations. There has been some improvement. The situation now is: a. nominal boot will go straight to X window (KDE) but upon first mouse click the mouse "click" function becomes unresponsive, although mouse will still move. Keyboard is partly functional. <ALT><F2> will not bring up run dialog. <ALT><F12>allows cursor movement of mouse, but <enter> will NOT simulate mouse click. The X window screen appears unresponsive to any applications. <CTRL><ALT><F2>brings to full screen text login. I noted from Xorg.0.log that "intel" driver was loaded by X. I have dmesg, xsession-errors and Xorg.0.log files available and can provide upon request. b. nominal boot with extra boot code acpi=OFF. Boot stops indefinitely at "Set System Time to Current Hardware clock", but pressing <CTRL><C> causes boot to continue. X comes. Fbdev driver in use again and this time KDE4 runs nominally with FBDEV driver. Wireless (Intel Pro 2100) functions. I have dmesg, xsession-errors and Xorg.0.log files available and can provide upon request. c. boot to run level 3 with extra boot code acpi=OFF. Installed various sax2 applications via USB stick and rpm command. Ran sax2 with: sax2 -r -m 0=vesa. Then ran startx. X comes up and runs nominally. Copied the xorg.conf created to a USB stick. I have dmesg, xsession-errors and Xorg.0.log files available and can provide upon request. d. reboot to run level 3 nominally (no extra boot codes). Copied from USB stick the vesa optimised xorg.conf file from step-c into /etc/X11.xorg.conf. Then ran "startX" and X comes up and runs nominally. This ok function under VESA was a surprise. I have dmesg, xsession-errors and Xorg.0.log files available and can provide upon request. e. logoff X back to run level 3 and installed (via USB stick) various sax2 rpms. Ran "sax2 -r -m 0=intel". Sax2 fails to configure. I have sax2 log and can post upon request. f. reboot to run level 3 nominally (no extra boot codes). copied a functional xorg.conf file for Intel driver from a running openSUSE-11.1 system on same hardware. ran startx. X comes up, but have same behaviour as reported in "a" above where X window display becomes unresponsive. I have dmesg, xsession-errors and Xorg.0.log files available and can provide upon request. g. reboot to run level 3 nominal (with acpi=off boot code). Copied a functional xorg.conf file for Intel driver from a running openSUSE-11.1 system on same hardware. Ran startx. X will not start. [speculation: the <CTRL><C> needed to restart boot process impacted some process that impacted X loading]. Based on the above, my guess is there are at least three hiccups here. (1) kernel will not smoothly boot with acpi=off as a <CTRL><C> is needed at a specific place in boot (2) sax2 can not create an Intel xorg.conf file. (3) Intel graphic driver does not function well with 2.6.3x.x kernels with this specific graphic chipset.
I note the intel bug report https://bugs.freedesktop.org/show_bug.cgi?id=24789 on what could be the same problem suggests there may be a patch here: https://bugs.freedesktop.org/show_bug.cgi?id=27187 This is all above my understanding. I assume that "patch" (not sure if the fix in that referenced bug report is mature enough to be called a patch) is NOT in the 2.6.34 kernel in 11.3 M6 ?
There are so many posts on the bugs.freedesktop.org its difficult to see the proposed solution ... but it appears post #1 here may provide guidance to the proposed solution: https://bugs.freedesktop.org/show_bug.cgi?id=26345#c61
Apoglogies for the typo, that should read post#61. There are so many posts on the bugs.freedesktop.org its difficult to see the proposed solution ... but it appears post #61 here may provide guidance to the proposed solution: https://bugs.freedesktop.org/show_bug.cgi?id=26345#c61
I do not know if it will help, but I've learned that the Ubuntu "community" have come up with implementating a patch for the 855GM problem here: https://launchpad.net/~glasen/+archive/855gm-fix which portedly works with a 2.6.32 kernel. And a remastered Ubuntu liveCD here with the patch: http://glasen-hardt.de/?p=568 ... If thought useful, I could download the remastered Ubuntu CD and try it out, although I confess I am not a Ubuntu fan (just the opposite and very very far from it) and hence I am not particularly keen on doing this download of a Ubuntu liveCD, unless advised it would be helpful for me to try as a test.
This thread (mentioned above on 2010-05-02 14:09:53UTC) is also actively trying to solve the Intel i855GM bug: http://bugs.freedesktop.org/show_bug.cgi?id=27187 and this thread (also mentioned above) solve the Intel i845G bug: https://bugs.freedesktop.org/show_bug.cgi?id=26345 , where I can not tell if they similar problems on slightly different hardware or completely different.
Reference the "Ubuntu Community" fix for what I believe to be this problem, as noted here: > I do not know if it will help, but I've learned that the Ubuntu "community" > have come up with implementating a patch for the 855GM problem here: > https://launchpad.net/~glasen/+archive/855gm-fix which portedly works with a > 2.6.32 kernel. And a remastered Ubuntu liveCD here with the patch: > http://glasen-hardt.de/?p=568 ... > If thought useful, I could download the remastered Ubuntu CD and try it out, > although I confess I am not a Ubuntu fan (just the opposite and very very far > from it) and hence I am not particularly keen on doing this download of a > Ubuntu liveCD, unless advised it would be helpful for me to try as a test. I downloaded the iso file for the Ubuntu community CD, burned it, and then today booted my Fujitsu Siemens Amilo 7400M laptop with the i855GM graphics. The freeze that I experienced on openSUSE-11.2 and 11.3 M4 and M6 does NOT occur with this special Ubuntu community test version. I checked the Xorg.0.log file i that Ubuntu liveCD and the "Intel" driver (version 2.11) is being used with a Ubuntu 2.6.32-22-generic kernel (which I presume has been specially patched with the fix noted here https://launchpad.net/~glasen/+archive/855gm-fix. There are so many various fixes around, this is a bit confusing.
Comment 21 on but 555619 is likely relevant (as is that entire thread) https://bugzilla.novell.com/show_bug.cgi?id=555619 as I believe this to be the same problem. In essence, even though there are various "fixes" around, as near as I can determine, the patch in the https://bugs.freedesktop.org/show_bug.cgi?id=27187 thread addresses the problem better. The patch is here: https://launchpad.net/~glasen/+archive/855gm-fix I do NOT know its status relative to the Linux-2.6 git tree. (I don't know how to check that).
In Comment#9 that should read "bug 55519" and not "but 555619"
I note the 32-bit openSUSE-11.3 milestone 7 (build 0625) does not boot at all to X on this Fujtisu-Siemens Amilo 7400M laptop (with the 855GM graphics), in either 1024x768, 800x600, VESA, nor Safe Settings with or without the "acpi=off" option. Adding the "nomodeset" boot parameter makes no difference. In all cases when X tries to load one ends up with a black screen and a small while dash in the upper left corner. A characteristic of all these failed boots is the boot initially failing with the last message: [ ... 55.013924] rtc0: alarms up to one day, 114 bytes nvram and a <CTRL><C> is then needed for boot to continue. Then after continuing it fails again with "dash" in upper left screen. I do not know if this is associated with 855GM error noted earlier, but I suspect it may be. It is no longer possible to boot X at all with this laptop in 11.3. It is possible to boot to run level 3 by selecting the F3 text mode. One does not get the pause noted above if there are no additional boot parameters added (such as acpi=off). However from run level 3 it is not possible to boot to X without getting the above mentioned freeze. I believe the solution to this 855GM problem is as noted in comment#8 above with this patch: https://launchpad.net/~glasen/+archive/855gm-fix
A further note to the above, on openSUSE-11.3 Milestone7 (32-bit KDE liveCD) I was able to get the laptop to boot X with the fbdev driver by booting to run level 3 (by selecting text mode) and then when in run level 3 adding the line: Device "fbdev" #to the file /etc/X11/xorg.conf.d/50-device.conf , followed by startx. I still believe the solution to this 855GM problem is as noted in comment#8 above with this patch: https://launchpad.net/~glasen/+archive/855gm-fix
wrt the freedesktop 27187 bug fix (which I believe is the one used for the Ubuntu community patch), I note that post #158 http://bugs.freedesktop.org/show_bug.cgi?id=27187#c158 in turn points to this bug fix: http://bugs.freedesktop.org/attachment.cgi?id=35548 ... Its not clear to me how final/mature that fix may be.
Why the priority P5=none? I think the first-generation Centrino platform (codename "Carmel") with an Intel Mobile 855 Express series chipset (codenamed Odem or Montara) and Intel Extreme Graphics 2 was quite popular and will be still much in use. Is the kernel 2.6.34 harder to patch than the 2.6.32 used in Ubuntu 10.4?
I've found this effects Ubuntu 10.04 (Gnome) and heard it effects Fedora 13. The system locks when it starts X. My system is a Dell Latitude D400 (I don't have more specs at this time). I also ran across this on a Toshiba (cannot recall the model at this time) I managed to get Ubuntu to boot up into graphical mode by appending "i915.modeset=1" to the bootup script. Unfortunately this wasn't enough for video playback and webcam. I applied the patch for the system (steps described below) which has allowed the system to boot in without any issues and be able to play video files (flash, etc.). It does not, however, provide for Desktop effects or webcam video (Cheese). Running into these freezes up X, yet I was able to get into a terminal with Cntl+F2 (ttyl2?) to restart the system. +-~-~-~-~-~-~-~-~-~-~-~-~-~-~-~-~-~-~-~-~-~-~-~-~-~-~-+ ## Add repositories sudo add-apt-repository ppa:glasen/855gm-fix sudo add-apt-repository ppa:glasen/intel-driver #(optional) ## Update sudo apt-get update sudo apt-get upgrade ## Install sudo apt-get install dkms linux-headers-generic 855gm-fix-dkms ## use only if using KMS (i915.modeset=1) and is optional sudo add-apt-repository ppa:glasen/libdrm #(Needed for intel-driver) sudo apt-get install intel-drivers +-~-~-~-~-~-~-~-~-~-~-~-~-~-~-~-~-~-~-~-~-~-~-~-~-~-~-+ I hope to be able to set up my system with openSUSE 11.3M* for further testing.
My research indicated this impacts all distributions with a kernel after the 2.6.27 kernel. Some examples: (1) Fedora: http://forums.fedoraforum.org/showthread.php?t=245296&highlight=855GM and http://forums.fedoraforum.org/showthread.php?t=245898&highlight=855gm (2) Ubuntu: http://ubuntuforums.org/showthread.php?t=1472282&highlight=855GM and http://ubuntuforums.org/showthread.php?t=1475128&highlight=855GM and http://ubuntuforums.org/showthread.php?t=1464239&highlight=855GM and https://wiki.ubuntu.com/X/Bugs/Lucidi8xxFreezes (4) LinuxMint: http://forums.linuxmint.com/viewtopic.php?f=59&t=47825 (5) Archlinux: http://bugs.archlinux.org/task/16974 ... plus others. Fedora-13 tried the drm-intel-big-hammer.patch which did not work for most users. Currently the fix-i855-cache-coherency-v9.patch in post #158 of http://bugs.freedesktop.org/show_bug.cgi?id=27187#c158 appears to be the most hopeful fix. But as near as I can determine, it still has not been submitted upstream to linux git tree. If you wish to test the fix-i855-cache-coherency-v9.patch, to my knowledge there is no openSUSE build with this patch. Hence one needs to test it on an unofficial community Ubuntu liveCD which one can find the URL for the torrent here: http://glasen-hardt.de/?p=568 A problem with testing this 855GM patch, is all Linux distros have been broken for the 855GM for so long, its possible other problems have crept in, that have not been tested and discovered, and its difficult to sort those others, from this one.
I note problem still exists in 11.3 RC1 as the fix-i855-cache-coherency-v9.patch in post #158 of http://bugs.freedesktop.org/show_bug.cgi?id=27187#c158 appears not to have been submitted to linus git tree as more testing is ongoing. I changed version of this 11.3 bug report from M4 to RC1. This is also same bug as this openSUSE-11.2 bug https://bugzilla.novell.com/show_bug.cgi?id=555619 and it may be useful to close one of these two bug reports reference the other.
I note the behaviour on 11.3 RC2 is different on this same Fujitsu-Siemens Amilo 7400M with the i855GM graphics. X window with 11.3 RC2 lasts a bit longer (ie it does not freeze at the first mouse click with 11.3 RC2). However this 'non-freeze' behaviour is short lived. On my Fujitsu-Siemens Amilo 7400M, within a few minutes of KDE loading (on the liveCD) by inserting a USB stick, copying some files, removing the USB stick, I can crash the KDE destkop totally, such that one is left with a black screen with a single dash in the upper left corner. No key strokes will bring up another terminal. Typing (as a boot code) "vga=normal" or "vga=normal i915 modeset=1" made no difference to the problem (ie the crash still quickly occurs).
Problem with use of Intel driver still exits in 11.3 GM (final) when tested on Fujitsu-Siemens Amilo 7400M with i855GM. Similar to RC2, crash of X will happen typically within 10 to 30 minutes of booting X.
Please try the 2.6.36 kotd http://ftp.suse.com/pub/projects/kernel/kotd/master -or add the kernel-head repo: http://download.opensuse.org/repositories/Kernel:/HEAD/openSUSE_11.3/Kernel:HEAD.repo
> Please try the 2.6.36 kotd > http://ftp.suse.com/pub/projects/kernel/kotd/master > -or add the kernel-head repo: > http://download.opensuse.org/repositories/Kernel:/HEAD/openSUSE_11.3/Kernel:HEAD.repo Thankyou for the suggestion. Unfortunately I'm not able at this time to try the patch. This laptop currently has openSUSE-11.1 (with KDE-4.4.4) with the 2.6.27 kernel which works well, and this laptop running openSUSE is used extensively by my wife who for once in her life is happy with openSUSE. The hard drive is too small for a dual boot between different openSUSE versions, and I can not remove openSUSE-11.1 without risking the wrath of my wife. Unfortunately I can only test liveCDs with the kernels noted, and creating a liveCD with such a kernel is beyond my expertise. I note this bug is still being worked actively upstream on this thread http://bugs.freedesktop.org/show_bug.cgi?id=27187 . The current most promising patch is discussed in comment#233 with reasons why it has not (yet) been submitted to Linux GIT tree. I will continue to track this, and if there is any liveCD with the noted kernels created, I will test on any such liveCDs.
*** Bug 555619 has been marked as a duplicate of this bug. ***
(In reply to comment #21) > Thankyou for the suggestion. Unfortunately I'm not able at this time to try the > ... > I note this bug is still being worked actively upstream on this thread > http://bugs.freedesktop.org/show_bug.cgi?id=27187 . Then I'm going to close as RESOLVED/UPSTREAM.
What I can do, re: testing with the 2.6.36 RC? kernel on this laptop, is download and burn and then boot to a liveCD openSUSE-11.4 milestone2 (when it is made available later this week). If it has the fixes from the same 2.6.36 kernel, then I can test that way ... If there is a 2.6.36 RC ? kernel for openSUSE-11.1, I could also install that kernel with "rpm -i" (keeping the old kernel) and test that way. Those are the only ways that I know how to proceed with out getting into a difficult family situation. My apologies for my personal affairs limiting the testing.
For Tom Jones, I note in the kernel-head repo for openSUSE-11.1 there is a : (1) kernel-vanilla-2.6.36-rc6.25.1.i586.rpm for 11.1 (2) kernel-vanilla-base-2.6.36-rc6.25.1.i586.rpm for 11.1 I could install that "in parallel" on this openSUSE-11.1 install, and test that kernel. I believe it would not destroy the 2.6.27 kernel that is functioning, and hence I could test and still keep my wife happy. ... Would that be of help ? I also note that I did test (boot only) the openSUSE-11.4 Milestone-2 KDE and Gnome liveCDs on this laptop and both had major issues when booting. 11.4 M2 has an 2.6.36 rc? kernel(I think rc4 but I am not sure as I am not at my PCs right now). The symptom for Gnome was the mouse cursor disappeared. The keyboard partially functioned (enough for very limited tests). I did confirm the "intel" driver loaded via the /var/log/Xorg.0.log file. The symptom for KDE with 11.4 M2 was the mouse cursor disappeared AND the desktop plasma was very flakey, with application windows constantly disappearing behind the plasma such that they could not be seen. Its possible there were other issues that would appear after a period of time, but I could not functionally use the desktop with the above issues for more than 10 minutes. I will try the openSUSE-11.1 kernel-vanilla-2.6.36-rc6.25.1.i586.rpm this weekend if believed helpful.
Created attachment 393631 [details] Xorg.0.log when booting 2.6.36-rc6 kernel on openSUSE-11.1 i855GM h/w To Tony Jones. First, my MOST sincere and humble apologies for typing your name wrong in the above post. I wanted to say a lot and I rushed my post. My apologies. I downloaded and installed on 32-bit openSUSE-11.1 (on the Fujtisu-Siemens Amilo 7400M laptop with the Intel i855GM graphics) the kernel-default-2.6.36-rc6.25.1.i586.rpm and kernel-default-base-2.6.36-rc6.25.1.i586.rpm and then attempted to boot to those kernels. I can boot to run level 3 but not X window. KDE4-4 hangs each time I attempt this. Attached is /var/log/Xorg.0.log file from the failed boot.
It appears openSUSE-11.4 Milestone-6 with its 2.14.0 Intel driver (and Novell/SuSE-GmbH patches) and the 2.6.37-20 kernel (also with the Novell/SuSE-GmbH applied patches) is very close to solving this problem. I just tested the 32-bit KDE4 liveCD of build 1034 openSUSE-11.4 miletone6. I reported on this in post#30 of this forum thread: http://forums.opensuse.org/english/get-technical-help-here/pre-release-beta/453319-opensuse-11-4-milestone6-unofficial-comments-thread.html#post2284465 and I made and posted a video of the behaviour in this youtube video : http://www.youtube.com/watch?v=QRRyQn_h03Y& I think the changes (all upstream) need to both kernel and Intel driver are extensive to back port to openSUSE-11.3. So my recommendation is to close this bug report, with a note that the bug is to be addressed in openSUSE-11.4 or subsequent openSUSE version. Is that acceptable ?
I mean to say: I think the changes (all upstream) needed to both kernel and Intel driver are TOO extensive to back port to openSUSE-11.3. Hence I recommend this bug be closed, with a note that openSUSE-11.4 or later should address the problem.
Here is what this looks like on 32-bit openSUSE-11.4 M6 KDE 2.6.37-20 kernel (Intel 2.14.0 driver): http://www.youtube.com/watch?v=QRRyQn_h03Y& Here is what this looks like on 32-bit openSUSE-11.4 M6 Gnome 2.6.37-20 kernel (Intel 2.14.0 driver): http://www.youtube.com/watch?v=9-X3ZiYUbcc&f Here is what this looks like on 32-bit openSUSE-11.1 Gnome 2.6.27 kernel where it works nicely (stock package): http://www.youtube.com/watch?v=lfnAPDt_bn0&feature=related I note the 32-bit openSUSE-11.4 M6 Gnome desktop went to a 'black-unresponsive-screen' after about 4.5 hours but because of the timing of this occurrence at 11:30pm at night (me a volunteer going to bed) I was not able to get logs. I'll see about reproducing it (likely won't be difficult - but rather is just a matter of patience). I've seen similar 'black-unresponsive-screen' on the latest Fedora.
Egbert- Can you look at this? openSUSE 11.4 M6 is flickering with Desktop effects enabled w/ Intel driver. See Comment #29 for videos of the problem. Lee- Can you include the files and information requested on this wiki page: http://en.opensuse.org/openSUSE:Bugreport_X Thanks, Brandon
I booted to the 11.4 RC1 liveCD to provide the information noted on the http://en.opensuse.org/openSUSE:Bugreport_X webpage. After obtaining the files/information (with desktop window/flickering) ... I also managed to reproduce the 'classic' 855GM X freeze (where graphics are frozen although mouse cursor moves and one can still access the PC via <ctrl><alt><f2>. And again this is on a PC with the graphics: 00:02.0 VGA compatible controller [0300]: Intel Corporation 82852/855GM Integrated Graphics Device [8086:3582] (rev 02) IMHO the 'freeze' is an upstream problem w/kernel/intel-graphics and not something that SuSE can solve (only something we can track for a solution if/when it appears). So I'll provide the attachments in 2 groups: (1) Before the freeze (2) after the Freeze. Special Desktop effects were left in their default configuration throughout. Attachments to follow in subsequent posts.
Created attachment 414929 [details] Xorg.0.log w/special desktop effects window flickering when moving w/11.4 RC1 32-bit KDE4
Created attachment 414930 [details] hwinfo-gfx w/special desktop effects enabled 11.4 RC1 32-bit KDE4 liveCD
Created attachment 414931 [details] head-n3 changelog files - this is a stock 32-bit 11.4 RC1 KDE4 liveCD so I put all in one file
Created attachment 414936 [details] xrandr - booting to 32-bit 11.4 RC1 KDE4 liveCD
Created attachment 414941 [details] xorg.0.log After FREEZE info-only 32-bit 11.4 RC1 KDE4 - classic 855GM freeze This is the Xorg.0.log obtained via <ctrl><alt><f2> and logging in to full screen terminal session after desktop freeze. Its the classic 855GM freeze (since after regression after 2.6.27 kernel) and I think this freeze an upstream problem that needs to be solved upstream.
Created attachment 414944 [details] /var/log/messages After FREEZE info-only 32-bit 11.4 RC1 KDE4 - classic 855GM freeze This is the /var/log/messages obtained via <ctrl><alt><f2> and logging in to full screen terminal session after desktop freeze. Its the classic 855GM freeze (since after regression after 2.6.27 kernel) and I think this freeze an upstream problem that needs to be solved upstream.
Created attachment 414945 [details] ~/.xsession-errors After FREEZE info-only 32-bit 11.4 RC1 KDE4 - classic 855GM freeze This is the ~/.xsession-errors obtained via <ctrl><alt><f2> and logging in to full screen terminal session after desktop freeze. Its the classic 855GM freeze (since after regression after 2.6.27 kernel) and I think this freeze an upstream problem that needs to be solved upstream.
Created attachment 414946 [details] Xorg.0.log (repost) After FREEZE info-only 32-bit 11.4 RC1 KDE4 - classic 855GM freeze This is a repost of the Xorg.0.log obtained via <ctrl><alt><f2> and logging in to full screen terminal session after desktop freeze. Its the classic 855GM freeze (since after regression after 2.6.27 kernel) and I think this freeze an upstream problem that needs to be solved upstream. Apologies for repost. Original post appears to have been misidentified as a binary file.
I will try to reproduce this on my 855.
Was it intended that this bug is still on the status "NEEDINFO"?
Yes, I can reproduce the issues here on my test system.
I note this commit (possibly upstream): agp/intel: Experiment with a 855GM GWB bit The reference for that commit is here: https://bugs.freedesktop.org/show_bug.cgi?id=25325 and I think this may be related to the problem.
I note the same problem still exists in openSUSE-12.1 Milestone2. Shall I raise a bug report on openSUSE-12.1 M2 on the same problem. I do not believe the "agp/intel: Experiment with a 855GM GWB bit" commit has made it it downtream yet.
We are aware of this - you don't need to issue a new ticket.
This bug is still present in a 64-bit openSUSE-12.1 Milestone-3 KDE4 liveCD.
*** Bug 673873 has been marked as a duplicate of this bug. ***
With the coming release of openSUSE 12.2, openSUSE kernel developers are focusing their efforts there. Reports against openSUSE 11.4 and prior will not get the attention needed to resolve them before openSUSE 12.2 is release and openSUSE 11.4 becomes unmaintained. Please re-test with openSUSE 12.1 or openSUSE RC2+ and re-open with an updated Product if you still encounter your issue. We apologize for this issue not getting the attention it deserves but we are focusing our resources in the area where they will have the most impact for our users. We're working hard to make openSUSE 12.2 the best openSUSE release yet!