[rs690m] [Gateway] Graphics corruption with ati x1200

First: I've upgraded from 9.10 to 10.04 beta. My system updates are current.
  I have a radeon rs690m (x1200)

This is the problem I ran into:
  * Small, horizontal white lines that appear a bit like static. Not very noticeable, and not present currently.
  * Medium and large horizontal lines / areas of corruption.
    * Stripes are often screen-wide
    * Stripes can be scrolled, e.g., in Firefox or Nautilus.
    * Refresh of graphical object clears up corruption -- E.g., a mouseover of a link or highlighting an icon
  * Mouse Pointer gets corrupted at some points, until refreshed (e.g., when the cursor changes from pointer to question mark, or to a hand, or from hand to arrow, etc.).
    * Corruption occurs regardless of whether or not Desktop Effects are enabled (important later)
      * Corruption might only occur after desktop effects *have* been enabled during this run.
  * Some fonts become corrupted. If they do, the letters corrupted stay corrupt. I.e., if "k" gets corrupted, and I type "k" elsewhere in the same font, that too will be corrupted.

It would be really nice to fix this before the release of Lucid -- I can't imagine what would happen with all the rs690 users out there. Not that there are a lot, but, there are enough.

Workaround (sort of):
In /etc/modprobe.d/radeon-kms.conf, I set:
 options radeon modeset=0

Why just "sort of"?
  * Logging into the (only) account created when the system was 9.10 fails, at best booting me back to the login screen.
  * Logging into an account created since the system has been upgraded to 10.04 works fine, desktop effects and all, no graphics glitches.
  * Logging into the 9.10-made account with desktop effects disabled works fine.
  * Enabling desktop effects once logged in makes the system visually unusable -- cannot switch to console, graphical screen black except for mouse, but the system is still running. I can ctrl-alt-delete from a 'text' console, although there is no text -- it just shows what was last in the graphics buffer (a black screen with the mouse pointer present).

Note that although there were glitches graphically, desktop effects or not, the system worked with desktop effects enabled when kms was enabled.

I think there are probably two problems here:
1) KMS causing or making visible some kind of horrible memory badness that I don't understand
2) Desktop Effects config in 9.10 can be set to a state that doesn't play nicely in 10.04 with KMS disabled.

Any further information I will be happy to provide. I have a workable system, I can just transfer everything over to a new account on my system, but I want to make sure others aren't affected problematically -- and it would be nice just to use my old account.

-Brian

Note: Included screenshots are *screenshots*, not photos. The actual images in memory are corrupt.

..please let me know if I can be of further assistance.

Thanks, this looks suitable to be forwarded upstream to bugs.freedesktop.org

If this is of any help, I believe, I am too suffering from this bug on a freshly installed beta-1 of Lucid.
The video card is reported as:
01:00.0 VGA compatible controller: ATI Technologies Inc Mobility Radeon X2300.

I noticed the static-like blinking horizontal ~6-12 pixels long artifacts during installation. The static is noticeable even during system bootup, even before the X is started. I cannot capture this static while taking snapshots.

The workaround suggested by the reporter relieves me from the ugly static.

Brian Visel - I've forwarded this bug upstream to http://bugs.freedesktop.org/show_bug.cgi?id=27529 - please subscribe yourself to this bug, in case they need further information or wish you to test something. Thanks ahead of time!

I have "static", it does not show up on screenshots.
Another interesting thing: "static" is dependent on color. There are absolutely no glitches on clean colors (red, green, blue). Corruption begins on tones closer to gray, and on the perfect gray there is the heaviest "static".
By gray's lightness: there are no glitches on black, they become visible on gray approximately lighter than 60 (of 256).

"static" itself appears white.

Posted a video, to check if its the same Bug i'm getting.

got this since somewhere in between Alpha3 and Beta1 of lucid.
No problems in Karmic at all.

with last package update amount of "static" increased heavily.

to zEn: yes this is the same thing
same shape, different color

@Psy: thanks for the confirmation.

My corruption is gone since 2 days:
I believe this was when kernel .20 arrived.

btw:
ATI Technologies Inc M52 [Mobility Radeon X1300] on a HP compaq nc6400

installed kernel upgrade 2.6.32-21 (metapackage 2.6.32-22 - is it normal?) nothing changed.

...radeon HD2400 on asus A8Sr

After upgrading kernel to 2.6.32-20 the static is gone.

Brian, two people have reported that with the -20 kernel the static is gone. Is this the case for you as well?

I don't know that the static is the same issue. The thing I'm terming "static" is stuff that looks like analogue static. E.g., stuff that is active around window borders or between significant changes of color. That *particular* issue has gone away, or isn't manifesting.

What I'm terming "Corruption", which is a significantly worse problem, is corruption of the image(s) in memory which aren't fixed until the image is refreshed by the program; e.g., mouseover an icon, etc. I don't think I ever saw the -20 kernel upgrade, but I'm on 2.6.32-21. I just booted without the workaround, and have the same problem still. It's virtually unuseable (currently, my "t" is corrupt, parts of the firefox pane are totally corrupt, corruptions rampant across the screen).

Still don't know if this was the right bug description for me.
The things I had and are visible in the video i posted earlier are still present if i boot in kernel -19 gone with >-20.

so for my bug obiviously was a kernel issue and not radeon!

But this was the closest bug report i could find...

zEn: I think that you had the static issue, not the corruption issue. The static issue isn't bothering me, either.

The Corruption issue is still present, though.

Bryce, should I try to compile the kernel with the patch on fd.o bugzilla 27529? If so, do you offhand know a good link on patching and compiling the kernel? I don't mind searching that myself, but figure if you know one offhand and I need to compile, that will reduce my search time, and therefore give me more time to work on compiling and reporting back the results of the patch.

Hi Brian, I don't have an authoritative link offhand. I think you can basically apt-get the source, apply the patch inline, and do a regular rebuild. I don't typically do kernel builds so am not totally up on the procedure.

Given the upstream bug has a linux patch, it sounds likely this is a kernel drm bug, so am going to turn this bug report over to the kernel team for further tracking.

Once a viable patch comes to light, if the kernel team doesn't pick it up, please ping JFo (jeremyfoshee) to get the issue some SRU attention if appropriate.

I believe I'm having this exact same bug on a similar machine. LT3108h. I'm not too well adjusted to linux but if you'd like me to run any reports to see if I can help let me know. Screenshot matches similar issues I am getting, the corruption seems more random on my side and gets heavier the longer I leave it alone.

http://ubuntuone.com/p/1uI/

What happens after a while.

According to the changelog, this has been fixed upstream in kernel 2.6.35-rc4:

commit 0888e883ea5ff8fac27e813256d6c1eaede5a234
Author: Alex Deucher
Date: Sat Jun 12 11:50:13 2010 -0400

    drm/radeon/kms: fix bandwidth calculation when sideport is present

    Fixes fdo bug 27529:
    https://bugs.freedesktop.org/show_bug.cgi?id=27529

    Reported-by: steckdenis
    Signed-off-by: Alex Deucher
    Signed-off-by: Dave Airlie

A PPA of the abovementioned kernel can be found at http://kernel.ubuntu.com/~kernel-ppa/mainline/

Ubuntu 10.4 user with a pair of ATI FireMV 2250 cards (RV516 chip)
running Xinerama across three monitors (one monitor output is unused)
using radeon driver on an AMD-64 2.6.32-24-generic.

Over time, single characters in Firefox get mangled - in tab titles and
on the displayed web pages. Typing those characters into text boxes also
shows the mangling. Changing the Firefox font size up or down (CTRL-plus,
CTRL-minus) restores the characters to good at the new size; setting the
font back to the original size (CTRL-zero) returns the mangling again.
Sometimes the characters disappear entirely. Restarting Firefox didn't
help until I added "Option "AccelMethod" "XAA" to my xorg.conf file;
now, restarting Firefox clears the font problem. (It didn't before -
I had to restart the X server.)

Also, while paging through photos in Facebook, twice now (separate
instances of Firefox some time apart) a photo has drawn itself about 200
pixels up and to the left of where it should be in on Facebook web page,
even outside of the browser window. Dragging other windows across the
corrupt area cleared out the ghost photo.

Also, a few times going to a video in YouTube (in Firefox) would
display a non-functional static image of slewed black-on-white, similar to
https://bugs.launchpad.net/ubuntu/+source/linux/+bug/556782/+attachment/1283506/+files/Screenshot-3.png.
(Other people have called this "video static", but it is seems to be a wrongly
displayed image, not white noise.) Restarting Firefox cleared that up.

I didn't have any of these problems when I ran a single FireMV 2250 with
an Nvidia card for the third monitor, but that was on a previous kernel
(2.6.32-23-generic) as well. It's the dual 2250 (on the new kernel)
that is now misbehaving.

Downloading the xorg-edgers radeon driver 6.13.99 didn't help.

From my Xorg.0.log (full log below):

    X.Org X Server 1.7.6
    Release Date: 2010-03-17
    X Protocol Version 11, Revision 0
    Build Operating System: Linux 2.6.24-27-server x86_64 Ubuntu
    Current Operating System: Linux idallen-oak.home.idallen.ca
       2.6.32-24-generic #42-Ubuntu SMP Fri Aug 20 14:21:58 UTC 2010 x86_64
    Kernel command line: BOOT_IMAGE=/vmlinuz-2.6.32-24-generic
       root=UUID=dc727a13-5170-4697-aad8-849090d05426 ro text
    Build Date: 21 July 2010 01:03:39PM
    xorg-server 2:1.7.6-2ubuntu7.3 (For technical support please see
       http://www.ubuntu.com/support)
    Current version of pixman: 0.16.4
    (II) Loading /usr/lib/xorg/modules/drivers/radeon_drv.so
    (II) Module radeon: vendor="X.Org Foundation"
       compiled for 1.7.6, module version = 6.13.99
       Module class: X.Org Video Driver
       ABI class: X.Org Video Driver, version 6.0

xorg.conf:

The "AccelMethod" "XAA" were added to try to get rid of the font corruption.
It didn't eliminate it, but it might have made it go away with a Firefox
restart, whereas restarting Firefox previously didn't help.

Section "ServerLayout"
 Identifier "Ian"
 Screen "Screen Right" 0 0
 Screen "Screen Middle" LeftOf "Screen Right"
 Screen "Screen Left" LeftOf "Screen Middle"
 Option "Xinerama"
EndSection

Section "Screen"
 Identifier "Sc...

Ian - I'm not sure that you have the same bug, as I don't imagine your ATI FireMV 2250 cards (RV516 chip) have sideport memory, but does this issue still occur with the Ubuntu 10.10 "Maverick Meerkat" beta Live CD available at http://www.ubuntu.com/testing/maverick/beta ?

madbiologist - alas, I don't have a "non-production" machine on which to try the beta.

The corruption does disappear when I slide the window containing the corruption onto either of the other monitors,
which are on the second FireMV 2250 card. I suppose I should swap cards and see if the corruption swaps, in the
off chance that it might be hardware.

If the bug report doesn't belong grouped here, I wonder where it should go. I didn't see anything obvious other than this bug.

Swapping the cards sounds like a worthwhile test.

You can run the beta from the Live CD if you want. Just download the .iso image and then burn it to a blank CD. When you reboot Ubuntu 10.10 "Maverick Meerkat" beta will offer you the option to try without installing. If you choose that option it will run from the CD in conjunction with a RAMdisk in your system's memory. Your hard disks will not be touched, they will be mounted in read-only mode. When you shut down the CD will be ejected and you will be prompted to remove it and close the CD tray (if it exists) before power-off. The next time you boot your system it will be as you left it before using the Live CD.

madbiologist - yea, I know all that (Unix user since 1976, sysadmin for
30 years). Two things work against trying it:

- I'd need to bring over my custom xorg.conf file and restart X to enable
  the two cards to work together - I can't just run the CD as is

- the problem is intermittent and may or may not appear for an hour or
  two - I can't commit that time this week

I said:

> The corruption does disappear when I slide the window containing the
> corruption onto either of the other monitors,

Today, I have some corruption (on the zero glyph in a particular font at
a particular size) that stays corrupt when I slide the Firefox window
from the Right monitor to the Middle monitor (which crosses from the
first FireMV to the second). But this corruption vanishes when I keep
sliding from the Middle to the Left monitor, both of which are on the
same FireMV card. (And the corruption returns when I slide the window
back to Middle or to Right.) I don't think this is a hardware problem
peculiar to one of the cards.

Hi! I just upgraded to 10.10 via upgrade-manager -u and then updated the system again. And still, by just visiting http://www.wissenschaftsjournalismus.ch the screen reliably crashes to the look I attached... Well, I can very well steer around this site, which belongs to the swiss association of science journalists... but anyhow...

At last I found the bug affecting me!
I first run into the bug while testing Lucid Lynx from the very beginning, then found out some way to bypass it (don't remeber how, it was a real mess). I have just installed Maverick Meerkat RC on my netbook and the problem is here again.

Here's my graphic card and my hardware specs and some screenshots attached
lspci | grep VGA
01:05.0 VGA compatible controller: ATI Technologies Inc RS690M [Radeon X1200 Series]

I see corruption that changes depending on to which of the three LCD
screens I drag my Firefox window. The worst corruption is far right.
I see less corruption when I drag the window to the middle LCD, and
dragging the window to the leftmost LCD looks fine. The middle and
right LCDs are on the same ATI FireMV 2250 card. The leftmost LCD is on
a second FireMV 2250 card. The corruption isn't constant. Things look
fine when I first reboot and get worse as time goes on.

I'm hoping Ubuntu 10.10 fixes things.

see little white things at the top of the first screenshot?
Imagine them applied transparently all over the screen. Appearing most severely on gray parts of the screen. And they do not appear on screenshots. No other corruption, except this noise.
That's my case on Radeon HD2400.

I don't think that's the same issue, as the issue we have can be seen in screenshots, and while that might not seem like a significant difference, it is nevertheless a massive difference.

At least the pattern and shape of white stripes looks similar. Well, either both our problems end simultaneously... or not.

Also observed on Gateway LT3119u (aka Acer ZA8) laptop, with radeon rs690m, running Maverick.

Not a solution, but my Gateway LT3103u was essentially useless after this bug was introduced. My cheap/easy solution for now is to just rename the radeon.ko kernel module to radeon.ko.orig. After I did that, the corruption disappeared (as did any hope of accelerated 3d graphics, but hey, it's a netbook man).

sudo mv /lib/modules/`uname -r`/kernel/drivers/gpu/drm/radeon/radeon.ko /lib/modules/`uname -r`/kernel/drivers/gpu/drm/radeon/radeon.ko.orig

Then reboot (chances are, radeon.ko is loaded and it's a bear to unload)

I also have this problem, also on a Gateway LT3119u (aka Acer ZA8) laptop running 10.10 with that RS690 chipset. I followed John Gorkos's suggestion but it didn't solve it.

I tried a newer kernel (specifically this one below) but it didn't solve the graphics corruption.

http://kernel.ubuntu.com/~kernel-ppa/mainline/v2.6.36-maverick/linux-image-2.6.36-020636-generic_2.6.36-020636.201010210905_amd64.deb

A workaround for now is to disable KMS.

sudo echo options radeon modeset=0 > /etc/modprobe.d/radeon-kms.conf

For some users (particularly users with encrypted volumes) KMS is enabled very early in the boot process and in order to pick up these changes you also need to run sudo update-initramfs -u.

Then reboot.

I have the same issue on Ubuntu 10.10!

I have experienced this bug as well with 10.10 - I have included a screen shot, here are my specs;
Gateway ZA8

after using lspci -v | less

00:00.0 Host bridge: ATI Technologies Inc RS690 Host Bridge
        Subsystem: Acer Incorporated [ALI] Device 028c
        Flags: bus master, 66MHz, medium devsel, latency 64

00:01.0 PCI bridge: ATI Technologies Inc RS690 PCI to PCI Bridge (Internal gfx) (prog-if 00 [Normal decode])
        Flags: bus master, 66MHz, medium devsel, latency 64
        Bus: primary=00, secondary=01, subordinate=01, sec-latency=64
        I/O behind bridge: 00009000-00009fff
        Memory behind bridge: f0000000-f01fffff
        Prefetchable memory behind bridge: 00000000d0000000-00000000dfffffff
        Capabilities: [44] HyperTransport: MSI Mapping Enable+ Fixed+
        Capabilities: [b0] Subsystem: ATI Technologies Inc RS690 PCI to PCI Bridge (Internal gfx)
        Kernel modules: shpchp

00:05.0 PCI bridge: ATI Technologies Inc RS690 PCI to PCI Bridge (PCI Express Port 1) (prog-if 00 [Normal decode])
        Flags: bus master, fast devsel, latency 0
        Bus: primary=00, secondary=03, subordinate=03, sec-latency=0
        I/O behind bridge: 0000a000-0000afff
        Memory behind bridge: 80000000-800fffff
        Prefetchable memory behind bridge: 00000000cfd00000-00000000cfdfffff
:
Anyways, my days of device driver writing are several years old...can anyone please help with this?

A week ago, I was ready to declare this bug "fixed" in the Natty Alpha code. Then, two days ago I went through a dist-upgrade evolution and now the corruption is back worse than ever. Right now, not even the "radeon modeset=0" fix will make it work (which worked for the longest time for me). Whatever just came down the pipe in the Alpha for 11.04 is a giant step backward. Since over 500MB of packages changed when I did the dist-upgrade, I can't pinpoint precisely what package change it was that hosed me again, but I suspect it's the xserver-xorg-video-ati package.
None of my hardware has changed, specs the same as above.

I still have this problem in Natty. Badly.

The "radeon modeset=0" (disabling KMS) workaround still works, sortof. It reduces the frequency/amount of corruption. However, the underlying problem is still there. ..and with Natty as it is, disabling KMS also disables compositing, unlike with Maverick.

At this point, the default install of Ubuntu is effectively useless for many on Radeon rs690m (a fairly common card), and has limited functionality if you apply the workaround.

Even though it's not currently reflected on Launchpad (tracking of the fd.o bugzilla seems to be borked currently), freedesktop.org has closed that bug as 'fixed'. However, this didn't seem certain (you can look at the bug yourself to see what I mean). I don't know if it is actually fixed, and I can't test it as I don't know the process of getting code from freedesktop.org into a compiled binary.

Jerome Glisse, who closed the bug on freedesktop.org said that we shouldn't open another bug report with them on this issue until I (or someone else with the problem) can get that done.

Any ideas for resources/howtos on that (fd.o git -> ubuntu binaries)?

Oh. Something else I noticed, although I don't know if it's important:

When an application (let's say the terminal), is updating regularly, it will affect areas outside of itself on the screen, which we've all discussed -- bleedover from one element on the screen to another.

But even when that program is hidden by the screensaver, if the application has been affecting other areas of the screen, it will continue to do so even when not visible on the screen. I was in the process of something long on the command line, and the screensaver came on, blanking the screen. ..and the gnome-terminal visuals kept overwriting the blackness of the screen.

This bug also affects the terminal, at least when the x session is running. (graphics corruptions show up in the terminal as streaks or distorted images from the x session)

If pulling the xf86-video-ati drivers from git means anything, this has not yet been fixed. There is another bug open with xorg, (35457).

I have had this issue for over 2 years, been plaguing me ever since I bought my packard bell dotma with the RS690M x1200, I thought I was suffering alone ! I got round it with the radeon modeset=0 as a kernel boot param.

However now doing that in 11.4 results in Unity not launching which is a step too far for me now.

This bug certainly is NOT resolved in the radeon driver.

Jerome Glisse closing the bug was unfortunate!

I agree, and it skews the bug report statistics. :-( But, all we can do is re-post it on freedesktop.org, and hope it gets attention (which has been done -- at least the posting part, and it had attention for a bit, and then activity ceased). It's a shame, because I thought when I started seeing activity on the first bug-post on freedesktop.org, it would finally be resolved after a year and a half of having the difficulty and waiting -- but then it got confused with another bug, and marked 'fixed' when it wasn't. Sometimes I'm tempted to just shout "Help, help!", but I know it's just more noise. ..if *anyone* knows who to mail that:
1) can move this bug along, and
2) whom it wouldn't be rude to email about this, and
3) preferably actually *wants* to see this fixed,
please let me know!

I don't know if the full gravity of the bug is realized. In it's current form, in Natty, it means:
* No hardware accelleration
* Thus no compositing
* No Unity
* This is a regression. The older driver (without KMS) worked. The new one (with KMS) does not.

I'm thinking of just downgrading to Lucid. ..but there are a lot of packages in Lucid that are outdated for my needs, so that's a lot of manual installation. :-/ I've also been keeping Natty on my system so I can help debug, but it seems no-one's debugging it. If someone is, do you mind just posting a quick update?

Sorry I forgot to subscribe this bug.
I will look up 35457 and see if I can assist.

This is not good for the radeon driver guys as the RS690M is a common chip-set in a lot of netbooks ... the target audience of Unity in Ubuntu 11.4.

A lot of netbook users will be unable to use Unity next week on go live!

FYI I have just pulled the latest unstable branch of the radeon xorg driver and recompiled .... still no fix.

I wonder if Wayland would be a solution !

Thank you! :-) IANAE, but I think Wayland uses the same driver.. ..but if Wayland works, I'll use it. :-)

Hi all, this workaround doesn't help me
[code]
sudo su
echo options radeon modeset=0 > /etc/modprobe.d/radeon-kms.conf
exit
[/code]

I am a complete newbie so I cannot provide much data, i can tell my pc is a packard bell gateway zA8.
chipset rs690m and vga X1200.

I tried to install, Ubuntu 10.10 and 11.04 (New Natty).... no way of geting it working with the workaround here described. screen goes well untill ghraphics gets initialized and everything gets screwed up. ctrl + alt + F1 still works and I can access console right. (except some ocasions it completelly hangs).

help please ....

Thanks!

Static is gone on my radeon hd2400 with upgrade to natty

Personally I apply the modeset=0 argment on bootup.
I do this by editing the /boot/grub/grub.conf file but that can be a bit risky if your a nubie and you may find it hard to correct any mistakes that stop your system booting.

However the fact that the radeon-kms.conf does not seem to be working for you makes me think that you are not using the radeon driver in X.org. You can check this by typing this command;

sudo grep LoadModule /var/log/Xorg.0.log.

If you do not see LoadModule "radeon" you could not be using the driver that setting applies to, explaining why you still see the corruption ... e.g. you could be using the radeonhd driver.

sorry typo ..

sudo grep LoadModule /var/log/Xorg.0.log (no dot on the end)

I can't use also 3d acceleration (but I have [almost] no screen corruption) on ati X1200, Dell Inspiron 1721. In fact, I filed Bug #755791 ; I didn't know about this bug report. At the moment, I see 2 bug reports in freedesktop.org . Is the problem already solved or not? If yes, when the solution can be tested?

Same behaviour on a Gateway LT3116h with custom bios which allows for 64bit OS to be installed. Under Xubuntu when loading flash video content or extending desktop to additional screen, all sorts of artifact occour. I posted bug#778860 which is dup of this. Back to winxp unfortunately :(

@Pablo: Yes, the workaround doesn't work for natty. Or, more correctly, it kindof works. It provides working 2d graphics (or for some, it seems, working, slow software rendering of 3d graphics). If 3d is not working, Unity (the new desktop environment) will not work. You can try installing the unity-2d package, which should run fine. You should not (or should only rarely) get corruption with that. You could also change your session (done at the login screen) to 'Ubuntu Classic'. For a graphical example of both of those, see:
http://www.liberiangeek.net/2011/04/install-unity-2d-in-ubuntu-11-04-natty-narwhal/

@Axel ohmygoodness you may have just given me an idea for a workaround for Natty users.

@Andre: I get weird behaviour on Natty, too, but it all goes back to the same things. I have the same log errors as you for this. As to the freedesktop.org bugs: There were actually multiple issues affecting graphics for the ATI rs690/x1200. One issue (handling of sideport ram, I believe) has been fixed. That was not the main issue for me, though it made some other peoples' problems go away.

@Tim I *might* have a workaround for you in about an hour..

@Anyone working on fixing the code for the bug:
The info here may be outdated, since Ubuntu has gone through an version change since the logs, etc were posted. Should logs etc. be re-posted? I know this is being worked on on the freedesktop.org site, but I don't know if there's cross-communication to where sharing logs here would be a good idea.

Well, no, the idea I had was totally not viable as a workaround. I was thinking of installing or creating a package of the older radeonhd drivers (which do work) for Natty, but but they have an older ABI and haven't been updated because they're being phased out. Looks like the version of X that is in Natty really is the nail in the driver we all know and love called radeonhd, which has gotten us through a year of difficulty with as-yet-incomplete radeon driver. :'-(

I'll later post a summary of the current state of knowledge regarding this bug.

If you are trying to disable KMS as a workaround (it worked for Maverick), in the Comment #4 of Bug #755791 , the user Bryce Harrington explained me that since in Natty the driver is not Radeon any more (it has been replaced by Gallium 3d drivers), that workaround will not work.

https://bugs.launchpad.net/ubuntu/+source/xserver-xorg-video-ati/+bug/755791/comments/4

Unfortunately the only options are: fixing the issue in this driver or using another driver (not a viable option since radeonhd and old radeon are deprecated).

Disabling KMS disables 3d support altogether. It's not a workaround in the sense of "This provides equivalent functionality", it's a workaround in the sense of "this provides a functional system." Disabling KMS actually ends up disabling the gallium driver altogether, and allows software rendering to take over. This could also be done with whatever option there is to disable the gallium driver, but I don't know that option, and can't find it right now, and don't have the time to search it out.

..although I do mean 'functional system' in the sense that it's functional, but limited. The default interface (Unity) won't work with software rendering, and therefore the "Ubuntu Classic" interface must be used. But this is by far not ideal, and a fix is definitely required.

Not to mention that any software which dependant on OpenGL is completely
unusable.

Yes, but in Maverick, disabling the KMS really solved the problem (3d accleration working).

I use as a "workaround" the following: 1) I do use the KMS (even if that doesn't solve the problem, it allows me to use 3d acceleration when occasionally works). 2) I use unity-2d. 3) I activated Metacity compositing (that really helps to feel "like at home"). 4) If I have to use an application that needs 3d acceleration and at the moment it's not working (I check it with glxgears), I open a terminal and run: `export LIBGL_ALWAYS_SOFTWARE=1` and then I run the desired application, but of course, everything is very very slow (keeping the window the smallest possible helps). But at least I can work with Stellarium, and Visual-Python and even Blender (that uses OpenGL). Just Compiz totally fails because it checks if you've software rendering...

I post this because it may be helpful for other users that have the same problem...

De-duped bug #755791, as I duped it to this bug too hastily. I would hate it if *this* bug got duped to another inaccurately, and apologize for any inconvenience.

Though these two bugs (this one and #755791) have similar workarounds (the workarounds Andre posted above are indeed useful), and though they are with the same card, and have at points shared some symptoms, there is not enough shared to say they are the same bug, and there's enough different that it's likely they are not, particularly after re-checking logs between the two. Again, apologies for any inconvenience.

No problem. Since the symptoms are similar, I hope that the solution of any of the two bugs will solve (or at least help) the other.

Hmm, looks like this bug has stalled.

First thing, for those looking for workarounds, you should be able to install the classic (non-gallium) driver and still use the disable-KMS workaround to get back to working 3D. If you haven't located the classic drivers, look in /usr/lib/dri-alternates. You can use the environment variable to set the path, ala:

 LIBGL_DRIVERS_PATH=/usr/lib/dri-alternates glxgears-info

If that works, you can set that option globally. You guys may need to experiment a bit to get an optimal workaround.

"""Sometimes I'm tempted to just shout "Help, help!", but I know it's just more noise. ..if *anyone* knows who to mail that:
1) can move this bug along, and
2) whom it wouldn't be rude to email about this, and
3) preferably actually *wants* to see this fixed,
please let me know!
"""

Here is a course of action I'd suggest. I agree it's unfortunate the upstream bug report got closed without actually being fixed. Part of the trouble is I think this bug report is a bit creakily old and has conglomerated a few different issues which perhaps makes the issue unclear to the developers. So, I would first suggest breaking out a separate bug report from this one, and keep it discrete to a specific, reproducible problem; be very clear in your description, include a reference to this bug report, and identify specific error messages, screenshots, and so on that clearly define the problem. Here's a good way to start filing it:

 $ ubuntu-bug mesa

(In the dialog that pops up, pick the "yes, I was directed here" option.)

Next, it looks to me like upstream wants this tested against current git versions of things; they don't put much priority on issues reported against released versions in distros typically. Fortunately we provide packaged versions of all the upstream stuff you need to test. Here's the links you need:

  * xorg-edgers: https://launchpad.net/~xorg-edgers/+archive/ppa - this will give git versions of mesa, -ati, etc.

  * http://kernel.ubuntu.com/~kernel-ppa/mainline/ - includes drm-next and .39 backports to natty

For this bug I doubt you need to install and test a newer kernel, I think the newer mesa is the more important thing to test. However, many of these kinds of bugs do depend on stuff in the kernel, so if the new mesa doesn't resolve the issue it would be wise to also test either the drm-next kernel or a .39 kernel (like v2.6.39-rc4-natty). For mesa, make sure to only use natty builds. For the kernel, either the oneiric or natty kernels should work on natty, but the natty builds are more likely to work properly. You may need to experiment a bit. Take good notes.

In the case that you *do* find that the newer git builds work properly, upstream will consider the issue solved, so no use bothering them further. However, if you think the issue still should be fixed in natty, there's some additional testing work to do. You can use a git bisection procedure to narrow in on what patch fixed it. See https://wiki.ubuntu.com/Kernel/KernelBisection or https://wiki.ubuntu.com/X/BisectingMesa or https://wiki.ubuntu.com/X/Bisecting as appropriate.

In the case that you *don't* find any success from the newer ...

Oh, also, once you do file a new bug report, please close this one out - click the downward triangle next to Linux and set the status to Invalid. Include a comment to point people to the fresh new bug report.

Thank you for your expert advice. As soon as I have the availability to follow it, I will. That you took the time to explain the process a little bit better is truly helpful.

FWIW, I'm still using Maverick, and the only way I have been able to get decent video playback and (almost?) no static was to blacklist radeon in /etc/modprobe.d/blacklist.conf. I imagine the following would also work:

# sudo echo "blacklist radeon" > /etc/modprobe.d/radeon-kludge.conf

This carries with it penalties, but at least I can watch non-fullscreen Flash video at something better than 2 fps.

For Bug #755791, in freedesktop.org (bug report 37679 of freedesktop.org), we found a workaround: running `vblank_mode=0 glxgears` gives correct output of glxgears. At the moment, compiz still doesn't work. Try this workaround in an Ubuntu Classic (or Unity 2d) session, with KMS enabled (since disabling it also disables 3d rendering), that is, a session with 3d acceleration enabled.

Does it work also in your case?

It seems to decrease the corruption initally, but doesn't fix the issue. My system was workable for about five minutes or so before corruption became a major issue (but I think it depends more on patterns of usage, though -- if I open more apps, corruption shows up more quickly). Also, running just glxgears, it takes a while (and some dragging of glxgears around the screen) to start in with the corruption.

Apologies I have been too busy at work to test again until now
setting vblank_mode=0 with kms enabled seems to just delay the typical screen corruption that I can usually trigger in about 1min of draggng the glxgears window around the screen.

with vblank_mode=0 it now takes ~5-10min before I can recreate the screen corruption.

Incidentally I just tested with a Mobile X1300 and it does not have the issue... so it seems specific to the X1200 RS690 !!

I did not realise that Ubuntu had shifted to the Gallium Driver... it explains why it suddenly broke in 11.04 ... Thx Andre :-)

Gives me more things test.

Ok, now you have another thing to add to the bug report in freedestop.org...

I have applied the workaround but certain things will still set off the corruption, such as visiting a website with java/flash or maximizing libreoffice or glxgears.

I see the same corruption on a Gateway LT31 with the x1200. The nomodeset workaround does address the corruption though the user is without desktop effects.

Regarding #76, what is the best way to set this for a user's entire desktop session?

Hi all,

I am a completely new Linux user. I moved from a totally broken Win XP environment to Ubuntu 11.04 a couple of weeks ago, expecting a streamlined, fast and functional system, with Unity 3D as a fancy GUI. But from the outset I have had diabolical issues with screen rendering corruption with the default Unity 3D setup and any apps in Unity 2D or even in Classic (no effects) which presumably use OpenGL which I wanted to use, e.g. Scilab (plotting), Oolite “Elite-like” game. The rendering corruption started right from the first: even the “try out” Ubuntu mode running from my USB drive during setup was afflicted! “Installing the whole Natty Ubuntu OS properly from disk will fix the problem,” I thought at first. I was wrong!

Hardware:
- Dell Dimension 5150 tower PC, 1GB physical RAM
- Dual-core Intel Pentium 4 @ 3GHz
- According to Ubuntu lspci -nn I have: ATI Technologies Inc RV370 5B60 [Radeon X300 (PCIE)] [1002:5b60]
01:00.1 Display controller [0380]: ATI Technologies Inc RV370 [Radeon X300SE] [1002:5b70]
-Two monitors: 17” Dell LCD Model E176FPc driven by the 15-pin D-type video output; and an old Apple 19” monitor connected to the DVI output of this video card. Nothing connected to the S-Video socket.
- Have used the monitors “control panel” in Ubuntu to configure an extended desktop across these two monitors, with the Dell as primary to the left, and the Apple to the right.

Have uploaded some screenshots of the rendering issues in Unity3D (a complete wipeout, totally unusable) and within Unity2D and Classic (no effects). They remind me so much of a buffer overrun somewhere...

I have no additional proprietary drivers installed, e.g. ATI/Radeon. I am loathe to install any because, having surfed widely to try and find a sure-fire resolution to this issue, I have seen so many cases where advised changes have messed up the user's system in their particular case that I have stopped short of making changes: I am too naïve a Linux user to start messing about with it to that level.

Is this thread still running? In the spirit of offering up more evidence for the nature and scope of the rendering corruption issues out here in the community, would it be helpful if I described my symptoms in more detail...? I can describe various reproducible (on my system) instances if needed...

Here's an example of how (presumably) OpenGL content gets stuffed even with correctly-rendered Ubuntu Classic (no effects) window surroundings... Just to confirm, this is ATI x300 (sorry if this is the wrong place for the report).

I don't know if this works for you, but:

Ubuntu has two types of release -- one of them is "LTS", or "Long-term
support." I'm running Ubuntu 10.04 Lucid Lynx, because I can have nifty
desktop effects and accelerated graphics, which I find both pretty and
extremely useful. To do so, you would need to:

1) Install Lucid Lynx
2) Once you're at the login screen, log into a 'failsafe gnome' session
   (down at the bottom of the login screen, after you click your
   username, there's a 'session' option. Change it to 'failsafe gnome'
3) run the following command:
gksudo gedit /etc/modprobe.d/radeon-kms-workaround.conf
4) type the following line:
options radeon modeset=0
5) Save, and reboot.

What it does:
forces usage of an older driver
What it provides an x1200 user:
Accelerated desktop with compiz
What it provides an x300 user:
no idea, but it should be better than the default.

I simply haven't had the time to push this issue through to xorg and get
it changed in the new driver -- I've just been too busy. But there was
a nice writeup earlier about what needs to be done, if anyone can take
up that challenge.

-Brian

I use Ubuntu since 2005 and recently also see rendering problems with my ATI X600 and others like X1200. But I never saw problems like your screenshots. I recommend you to create a new Ubuntu bug report (by executing "ubuntu-bug" in a terminal) or directly upstream on https://bugs.freedesktop.org where the graphics system is hosted.

Meanwhile you could try to:
- boot with the kernel option "nomodeset" (press F6 on boot menu screen)
- login with "Ubuntu classic (without effects)", or uninstall Compiz effects completely by executing "sudo apt-get remove compiz-core" in a terminal
- downgrade to the latest Long Term Support release "Ubuntu 10.04 LTS"

OK, eode and oliver-joos, I'll give your suggestions a shot. Many thanks for your rapid response!

UPDATE: The mods suggested above sadly did not help with Natty. I've done a total, clean install of 10.04 LTS UbuntuStudio (as I wanted the low-latency kernel for music apps).

With UbuntuStudio 10.04:
* Video so far works now without a hitch, straight out of the box with my ATI RV370 [Radeon X300] PCI card! *
What I think is OpenGL content now works just as well as anything, and I haven't seen any rendering issues yet. Overall, video seems much more solid and stable. No further command-line mods or workarounds needed at this point in time.

Not only that, but with the addition of just the Medibuntu package and alsa-firmware on top, I can even drive my EMU (Creative Labs) 0404 PCI soundcard for audio and MIDI as well!

Next step: create the new bug report re: apocalyptic Natty rendering, as per Oliver's suggestion #89.

Thanks again Brian and Oliver for your excellent advice!

I have been doing testing of 11.10 Ubuntu beta 2 on a Gateway Netbook LT31 with the ATI Radeon X1270 (RS690M) graphics card. The system is unusable due to video corruption. As a side note, this problem was fixed in earlier releases of Ubuntu but its back now and results in a totally unusable system.

*nod* See the above comments on starting a new bug and pushing this
issue through to xorg. I'm not able to right now -- I just dont' have
the time -- I've defaulted to using Lucid, with the radeon-kms.conf fix.

The issue is that there's an entirely new driver framework, and it
doesn't work for these cards (all the people here and more -- it's a
common card in low-end laptops). This means that wayland doesn't work
for us, either. I'm hoping I or someone else will have the time to push
this all the way up to xorg, and get resolution there before the next
LTS release.

There's a message above on the necessary process in the fairly-recent
history.

@Brian: you already pushed this bug upstream, didn't you? See https://bugs.freedesktop.org/show_bug.cgi?id=35457

I have a Gateway LT3103u.

I've finally found a way to reproduce the corruption I experience reliably- I point Firefox 7 at http://google.com/nexus . Not sure what it is on that site that triggers it, but

I've had corruption on Natty 64 bit and Maverick 64 bit too and am currently running lucid 31 bit:

tony@blackgat32:~$ uname -a
Linux blackgat32 2.6.32-33-generic #72-Ubuntu SMP Fri Jul 29 21:08:37 UTC 2011 i686 GNU/Linux
tony@blackgat32:~$ cat /etc/issue
Ubuntu 10.04.3 LTS \n \l

Sorry about the truncated above. I just testted this site, and get the same result as wess (as above)

Not sure what it is on that site that triggers it, but

SOLVED

Hi again. Sorry about the last two comments. I was in a carpool on CA-17 when I wrote them. I think. Anyway, I tried many of the above on my LT3103u and the only one that worked for me was passing nomodeset (not any of the modeset=0 variants) on the kernel commandline (first tested interactively by catching grub in the act of booting, and then by putting it in the /boot/grub/grub.cfg via /etc/default/grub and update-grub).

For completeness, and to help out others who come across this via web search, here's the chip in question:

$ lspci|grep VGA
01:05.0 VGA compatible controller: ATI Technologies Inc RS690M [Radeon X1200 Series]

And I am in fact running 32 bit Lucid on this 64 bit machine, not 31 bit ;-)

Now that this issue is solved, I'm ready to bump the distro back up a notch (I'd downgraded to try to get this resolved)

Oh, and for the record, it made no difference on this machine whether I was an upgraded user or a newly created user.

Tony

@ agodshall

You posted - "I tried many of the above on my LT3103u and the only one that worked for me was passing nomodeset (not any of the modeset=0 variants) on the kernel commandline (first tested interactively by catching grub in the act of booting, and then by putting it in the /boot/grub/grub.cfg via /etc/default/grub and update-grub)."

I'm very new to Ubuntu/Linux OS environment, so please, if you don't mind, walk me through your suggestion. How do i get to the kernel commandline?

For your info, i'm using Gateway LT3105g & i experience the same graphical corruption others have mentioned in this thread.

Thanks.

Brendan.

Sorry, i forgot to add that i'm running Ubuntu 11.10. I've tried 11.04, 10.10 & even 10.04 but the same thing happened - graphical corruption, even when booting from Live CD. I've checked the ISO's & they're all good. I've used them to install Ubuntu in Intel-based laptops & experienced no problem at all. It's just sad that i can't install Ubuntu on my own system.

Processor 1.2GHz AMD Athlon 64 L110
Memory 2GB, 667MHz DDR2
Hard drive 250GB 5,400rpm
Chipset AMD RS690E
Graphics ATI Radeon X1270

The bug is not a driver issue but a hardware issue. The OEMs of x1200 series locked down the bios to the integrated graphic card. To fix this issue you need to flash a custom bios to your netbook/laptop and and change the bios settings. I did this two weeks ago. my lt3103u with the x1270 chip is running great. No need to turn off KMS and it works with the gallium driver with are the standard on the newer releases of ubuntu. I posted instructions to ubuntu forum. Here is the link:

http://ubuntuforums.org/showthread.php?p=11682072#post11682072

Ooh, this means this is *VERY* likely a sideport ram issue, and should
narrow down the bugfixing process significantly.

> http://ubuntuforums.org/showthread.php?p=11682072#post11682072

Thanks. 100% working on packard bell dot ma. It also fixed various windows
problems.

Il giorno 19 febbraio 2012 01:25, Brian Visel
<email address hidden>ha scritto:

> Ooh, this means this is *VERY* likely a sideport ram issue, and should
> narrow down the bugfixing process significantly.
>
> > http://ubuntuforums.org/showthread.php?p=11682072#post11682072
>
> --
> You received this bug notification because you are subscribed to a
> duplicate bug report (569709).
> https://bugs.launchpad.net/bugs/556782
>
> Title:
> [rs690m] Graphics corruption with ati x1200
>
> Status in The Linux Kernel:
> Confirmed
> Status in X.org XServer - ATI gfx chipset driver:
> Fix Released
> Status in “linux” package in Ubuntu:
> Triaged
>
> Bug description:
> [Edit]
> This bug has been around for a bit over a year now, and has gotten rather
> long, so this is a brief summary.
> There is a memory corruption issue that affects users of the rs690m, to
> varying degrees.
> For many most people, it makes the desktop unusable.
>
> Workaround: *Note: This is a workaround, not a fix. It will give you a
> usable system.
> [Natty]: Regression: This workaround now only provides 2d/software
> rendering, and one must either:
> * Choose choose the "Ubuntu Classic" session from the GDM Login screen
> or
> * Install the "unity-2d" package.
>
> [code]
> sudo su
> echo options radeon modeset=0 > /etc/modprobe.d/radeon-kms.conf
> exit
> [/code]
> Note that this doesn't totally fix the issue, but brings your desktop to
> a workable state.
>
> Freedesktop.org has dealt with one bug having to do with graphics
> corruption on the rs690m. That bug has been fixed, but the issue of
> graphics corruption in general has not been resolved. A new bug has been
> opened with freedesktop.org to continue pushing through the resolution of
> this issue.
> [/edit]
>
> ProblemType: Bug
> DistroRelease: Ubuntu 10.04
> Package: xorg 1:7.5+3ubuntu1
> ProcVersionSignature: Ubuntu 2.6.32-19.28-generic 2.6.32.10+drm33.1
> Uname: Linux 2.6.32-19-generic i686
> Architecture: i386
> Date: Tue Apr 6 12:58:29 2010
> DkmsStatus: Error: [Errno 2] No such file or directory
> InstallationMedia: Ubuntu 9.10 "Karmic Koala" - Release i386 (20091028.5)
> MachineType: Gateway LT31
> ProcCmdLine: BOOT_IMAGE=/vmlinuz-2.6.32-19-generic
> root=UUID=617a7a50-d35f-4b03-8bf1-f91ec024381b ro quiet splash
> ProcEnviron:
> LANG=en_US.utf8
> SHELL=/bin/bash
> SourcePackage: xorg
> dmi.bios.date: 06/18/2009
> dmi.bios.vendor: Phoenix Technologies LTD
> dmi.bios.version: v1.3201
> dmi.board.name: SJM11-YK
> dmi.board.vendor: Gateway
> dmi.board.version: Not Applicable
> dmi.chassis.type: 10
> dmi.chassis.vendor: Gateway
> dmi.chassis.version: N/A
> dmi.modalias:
> dmi:bvnPhoenixTechnologiesLTD:bvrv1.3201:bd06/18/2009:svnGateway:pnLT31:pvrNotApplicable:rvnGateway:rnSJM11-YK:rvrNotApplicable:cvnGateway:ct10:cvrN/A:
> dmi.product.name: LT31
> dmi.product.version: Not Applicable
> dmi.sys.vendor: Gateway
> system:
> distro: Ubuntu
> codename: lucid
> architecture: i686
> kernel: 2.6.32-19-generic
>
> [lspci]
> 01:05...

I know it's not much, but I'm placing a $50 bounty on this if someone takes control of this bug and gets it fixed in Pangolin. I'll pay via PayPal.

I have 2 systems with screen corruption and square mouse pointer:

*desktop precise 64bit 12.04.1 clean installed on ati x800
*laptop precise 32bit 12.04.1 on lubuntu clean installed on ati mobile

This is what i tryed:

*adding 'nomodeset' to /etc/default/grub = compiz got disabled, screen lags
*sudo service lightdm restart = did not fix the screen corruption
*generated a xorg.log and adding the line swcursor = did not fix the screen corruption
*in ccsm category opengl, disabled sync to vblank and texture filter to fast = did not fix the screen corruption
*tryed another mouse icon theme = did not fix the screen corruption

I had this bug on previous versions of ubuntu aswell it did not matter if i upgraded or clean installed

every cold boot in the morning this mouse pointer and screen got corrupted, after a reboot its gone
few hours wait or next morning: the screen corruption is back again...

Please let me know if there's a solution yet after all these years...

I confirm that adding GRUB_CMDLINE_LINUX_DEFAULT="quiet splash nomodeset" to /etc/default/grub
removes the screen corruption and square mouse pointer at pc boot, but compiz is unworkable, wobbly windows
are disabled, unity icon laucher size default as in 2D, screen lags etc...

If anyone know a fix to have compiz enabled without nomodeset please let me know

A good rule of thumb to follow is separate hardware=separate bugs.

The X1200 (mobile/integrated chip) problem seems rather intractable, although some people have had success disabling/reducing the amount of sideport RAM - see comments. #138 and #149.

The X800 problem should be easier to fix, althought the fact that it only happens when the hardware is cold could mean a hardware bug - does that card work OK on Windows? Can you please attach the output of lspci -vvnn for the X800 card?

@madbiologist: the card worked well on windows yes, here's the output of lspci:

http://paste.ubuntu.com/1183794/

I realize that the last comment was almost a year ago, but apparently this bug is still around. I have it with my Gateway LT31xx laptop, with the same graphics card as the bug title. I was able to find a solution to this problem. It requires a BIOS flash with a modified bios file that enables certain options (North Bridge, IIRC).

The solution was based on this post --> http://ubuntuforums.org/showthread.p...2#post11682072 .

I went to the thread they referenced (http://forum.notebookreview.com/acer...ml#post6368072), and found the modified bios file for my laptop, a Gateway LT31xx. The modified bios file gives you an option to select only 'UMA' rather than 'UMA+Sideport'. I flashed this new bios file, after using the correct phlash.exe command (found in the first post of the notebookreview thread), and when I rebooted, there was the option. I changed that and everything looks great --- no tearing at all. I don't know if I have 3D compositing, but I do have xcompmgr running, so I think compositing works fine.

I think this is the solution that everyone is looking for in regard to the screen tearing problem. If you have a different laptop, I would suggest searching the notebookreview thread to see if a modified bios file was created for your specific model.

Sorry about the links in the previous comment. They should be, respectively,

http://ubuntuforums.org/showthread.php?t=1866046&p=11682072#post11682072

http://forum.notebookreview.com/acer/480992-acer-laptop-phoenix-bios-bios-mod-request-19.html#post6368072

Brian Visel, this bug was reported a while ago and there hasn't been any activity in it recently. We were wondering if this is still an issue? If so, could you please test for this with the latest development release of Ubuntu? ISO images are available from http://cdimage.ubuntu.com/daily-live/current/ .

If it remains an issue, could you please run the following command in the development release from a Terminal (Applications->Accessories->Terminal), as it will automatically gather and attach updated debug information to this report:

apport-collect -p linux <replace-with-bug-number>

Also, could you please test the latest upstream kernel available following https://wiki.ubuntu.com/KernelMainlineBuilds ? It will allow additional upstream developers to examine the issue. Please do not test the daily kernel folder, but the one all the way at the bottom. Once you've tested the upstream kernel, please comment on which kernel version specifically you tested. If this bug is fixed in the mainline kernel, please add the following tags:
kernel-fixed-upstream
kernel-fixed-upstream-VERSION-NUMBER

where VERSION-NUMBER is the version number of the kernel you tested. For example:
kernel-fixed-upstream-v3.11.1

This can be done by clicking on the yellow circle with a black pencil icon next to the word Tags located at the bottom of the bug description. As well, please remove the tag:
needs-upstream-testing

If the mainline kernel does not fix this bug, please add the following tags:
kernel-bug-exists-upstream
kernel-bug-exists-upstream-VERSION-NUMBER

As well, please remove the tag:
needs-upstream-testing

Once testing of the upstream kernel is complete, please mark this bug's Status as Confirmed. Please let us know your results. Thank you for your understanding.

I have problems with a clean install of 13.04 x64 and an ATI RS690M graphics card.
Everything is slow, menu shows artifacts and random patterns.
Proprietary drivers do not seem to support this version of ubuntu any more.

HP Compaq 6715b
lspci: VGA compatible controller: Advanced Micro Devices [AMD] nee ATI RS690M [Radeon Xpress 1200/1250/1270]

kuntergunt, if you have a bug in Ubuntu, the Ubuntu Kernel team, Ubuntu Bug Control team, and Ubuntu Bug Squad would like you to please file a new report by executing the following in a terminal while booted into a Ubuntu repository kernel (not a mainline one) via:
ubuntu-bug linux

For more on this, please read the official Ubuntu documentation:
Ubuntu Bug Control and Ubuntu Bug Squad: https://wiki.ubuntu.com/Bugs/BestPractices#X.2BAC8-Reporting.Focus_on_One_Issue
Ubuntu Kernel Team: https://wiki.ubuntu.com/KernelTeam/KernelTeamBugPolicies#Filing_Kernel_Bug_reports
Ubuntu Community: https://help.ubuntu.com/community/ReportingBugs#Bug_reporting_etiquette

When opening up the new report, please feel free to subscribe me to it.

Please note, not filing a new report would delay your problem being addressed as quickly as possible.

No need exists to comment here at this time. After reading the above documentation in it's entirety, if you have further questions, you are welcome to redirect them to the appropriate mailing list or forum via http://www.ubuntu.com/support/community/mailinglists , or you may contact me directly.

Thank you for your understanding.

Brian Visel, this bug report is being closed due to your last comment https://bugs.launchpad.net/ubuntu/+source/linux/+bug/556782/comments/184 regarding you no longer have the hardware. For future reference you can manage the status of your own bugs by clicking on the current status in the yellow line and then choosing a new status in the revealed drop down box. You can learn more about bug statuses at https://wiki.ubuntu.com/Bugs/Status. Thank you again for taking the time to report this bug and helping to make Ubuntu better. Please submit any future bugs you may find.

According to https://git.kernel.org/cgit/linux/kernel/git/torvalds/linux.git/commit/?id=35eecf052250f663f07a4cded7d3503fd1b50729 and https://bugs.freedesktop.org/show_bug.cgi?id=35457 this bug is finally fixed in the upstream 3.13-rc5 kernel. A PPA of this kernel is available at http://kernel.ubuntu.com/~kernel-ppa/mainline/ and instructions on how to install and uninstall it are available at https://wiki.ubuntu.com/Kernel/MainlineBuilds

This bug was NOT ressolved even in Kernel 3.17!

Even with latest Enlighment (E19), I experience text corruption when enabling hardware acceleration in Compositing window.
When switching to Software rendering, there is no issue.

KDE freezes simply after certain period of work if hardware acceleration is used.

ShinobiTeno, thank you for your comment. Unfortunately, this bug report is not scoped to you, your problem, or your hardware. So your problem and hardware may be tracked, could you please file a new report with Ubuntu by executing the following in a terminal while booted into the default Ubuntu kernel (not a mainline one) via:
ubuntu-bug linux

For more on this, please read the official Ubuntu documentation:
Ubuntu Bug Control and Ubuntu Bug Squad: https://wiki.ubuntu.com/Bugs/BestPractices#X.2BAC8-Reporting.Focus_on_One_Issue
Ubuntu Kernel Team: https://wiki.ubuntu.com/KernelTeam/KernelTeamBugPolicies#Filing_Kernel_Bug_reports
https://wiki.ubuntu.com/Kernel/Policies/DuplicateBugs
Ubuntu Community: https://help.ubuntu.com/community/ReportingBugs#Bug_reporting_etiquette

When opening up the new report, please feel free to subscribe me to it.

As well, please do not announce in this report you created a new bug report.

Thank you for your understanding.

Helpful bug reporting tips:
https://wiki.ubuntu.com/ReportingBugs