X displays various artifacts, white lines all over the screen

[Lev Kandel]

Binary package hint: xserver-xorg-video-ati

See attached screenshot. Dots and lines start appearing right after X starts. After a while (usually a day or two), entire areas of the screen start filling up with flickering lines.
Note that my xorg.conf has XAA enabled because EXA causes X to lock up (bug #345224), but I've tried switching back to EXA and saw the same artifacts in the short time before X froze.

ProblemType: Bug
Architecture: amd64
DistroRelease: Ubuntu 9.04
Package: xserver-xorg-video-ati 1:6.12.1-0ubuntu2
ProcEnviron:
 LANG=en_US.UTF-8
 SHELL=/bin/bash
ProcVersion: Linux version 2.6.28-13-generic (buildd@yellow) (gcc version 4.3.3 (Ubuntu 4.3.3-5ubuntu4) ) #44-Ubuntu SMP Tue Jun 2 07:55:09 UTC 2009
SourcePackage: xserver-xorg-video-ati
Uname: Linux 2.6.28-13-generic x86_64

[lspci]
00:00.0 Memory controller [0580]: nVidia Corporation CK804 Memory Controller [10de:005e] (rev a3)
     Subsystem: ABIT Computer Corp. Device [147b:1c0c]
01:00.0 VGA compatible controller [0300]: ATI Technologies Inc RV380 [Radeon X600 (PCIE)] [1002:5b62]
     Subsystem: ATI Technologies Inc Device [1002:0422]

[Bryce Harrington (bryce)]

[This is an automatic notification.]

A new version of the -ati driver is now available in Karmic.

This is a significant update to -ati which brings in kernel mode-setting
(currently disabled) and scores of fixes for DRI2, EXA, etc.

I've posted the new version of this driver to the following PPA,
would you mind testing it and seeing if it resolves the bug you
reported?

  https://edge.launchpad.net/~bryceharrington/+archive/ppa/+sourcepub/709908/+listing-archive-extra

If you're not running this release of Ubuntu, you can try booting the Karmic
LiveCD and loading the PPA onto it, and then log out/in to restart X.
ISOs are available at http://cdimages.ubuntu.com/releases/

After testing Karmic, report back here whether it's still an issue or not,
and if it is please post a fresh Xorg.0.log and 'dmesg' output.

Note there could be new bugs... please file these as new reports using
the command 'ubuntu-bug xorg'.

[Lev Kandel (y5b2rqi02)]

Worse than before. Screen corruption starts happening immediately on login (in fact, even before -- on the login screen). What's more: in Jaunty, corruption was extremely rare with DRI turned off (I've seen it once in 1-2 months of use). Now even with DRI off, I see screen corruption after several minutes of use. Attaching screenshot, Xorg.0.log and dmesg output.

Bryce, any idea what it would take for this to get fixed? Having to turn DRI off is bad enough (desktop becomes infuriatingly slow), and now even that doesn't help.

[Bryce Harrington (bryce)]

> Bryce, any idea what it would take for this to get fixed?

Thanks for testing; yes, this should go upstream as its next step now that it's confirmed against this version. Setting to Confirmed for now.

[Bryce Harrington (bryce)]

Forwarding this bug from Ubuntu reporter Lev Makhlis:
http://bugs.launchpad.net/ubuntu/+source/xserver-xorg-video-ati/+bug/391439

[Problem]
X displays various artifacts, dots and white lines all over the screen, same effect whether XAA or EXA is enabled. (EXA also causes a freeze).

Reported against Jaunty originally but confirmed to still happen on Karmic with a recent git snapshot of -ati.

[Original Description]
See attached screenshot. Dots and lines start appearing right after X starts. After a while (usually a day or two), entire areas of the screen start filling up with flickering lines.
Note that my xorg.conf has XAA enabled because EXA causes X to lock up (bug #345224), but I've tried switching back to EXA and saw the same artifacts in the short time before X froze.

Architecture: amd64
DistroRelease: Ubuntu 9.04
Package: xserver-xorg-video-ati 1:6.12.1-0ubuntu2
ProcEnviron:
 LANG=en_US.UTF-8ProcVersion: Linux version 2.6.28-13-generic (buildd@yellow) (gcc version 4.3.3 (Ubuntu 4.3.3-5ubuntu4) ) #44-Ubuntu SMP Tue Jun 2 07:55:09 UTC 2009
SourcePackage: xserver-xorg-video-ati
Uname: Linux 2.6.28-13-generic x86_64

[Bryce Harrington (bryce)]

Created an attachment (id=29134)
Screenshot.png

[Bryce Harrington (bryce)]

Created an attachment (id=29135)
Screenshot.png (against jaunty)

[Bryce Harrington (bryce)]

Created an attachment (id=29136)
Xorg.0.log

[Bryce Harrington (bryce)]

Created an attachment (id=29137)
dmesg

[agd5f (agd5f)]

please try with the latest git from master. make sure you have:
794ae743c305331741e7cf45dd71c755fe325ed5

[Bryce Harrington (bryce)]

Lev Makhlis - I've forwarded this bug upstream to https://bugs.freedesktop.org/show_bug.cgi?id=23662 - please subscribe yourself to this bug, in case they need further information or wish you to test something. Thanks ahead of time!

[Bryce Harrington (bryce)]

A Ubuntu package of 794ae743 can be found in this PPA for testing:

https://edge.launchpad.net/~xorg-edgers/+archive/ppa
 + xserver-xorg-video-ati - 1:6.12.99+git20090902.794ae743-0ubuntu0tormod

[93fy5qj02 (93fy5qj02)]

Tested the package above; I still get the same screen corruption.

[93fy5qj02 (93fy5qj02)]

Some additional info:
I see two qualitatively different types of corruption: (A) dots, lines and patches in random places; and (B) flickering lines across the screen. I don't know if they have the same root cause or not. The screenshot from Jaunty has both, and I still see both with 794ae743. That said, "B" manifests less frequently (typically within a day or two, whereas "A" starts happening right away), and when "B" does happen, a logout attempt causes X to freeze with 100% CPU (logout works fine before then). Today when the B-type corruption manifested, I captured a syslog with DRM debugging before and after a logout attempt. The X freeze in response to logout seems to be the same problem as bug 23661. Attaching that syslog.

[93fy5qj02 (93fy5qj02)]

Created an attachment (id=29244)
syslog with DRM debugging

"Corrupted" X freezes in response to logout attempt

[agd5f (agd5f)]

Does:
Option "DisplayPriority" "HIGH"
in the device section of your config help?

[93fy5qj02 (93fy5qj02)]

No; at least, I still see type "A" corruption right away. I can wait and see if type "B" still manifests eventually.
By the way, I've already tried AGPMode -- that didn't help either (blank screen with AGPMode=2, same problems as before with 1,4,8).

[agd5f (agd5f)]

AGPmode won't do anything in your case as your card is not AGP. Please try the following options and see if any of them help:
Option "DRI" "FALSE"
Option "EXANoComposite"
Option "EXANoUploadToScreen"
Option "EXANoDownloadFromScreen"

[93fy5qj02 (93fy5qj02)]

I've tried turning DRI off and that decreases, but does not eliminate corruption. In Jaunty, I saw corruption after about a month with DRI off, but with the latest snapshot I start seeing it after several minutes.

Just to clarify, the corruption happens with both EXA and XAA. I normally use XAA because with EXA, X freezes very soon (bug 23661). I can try and see if the EXA options make a difference in EXA mode if that's what you want.

[93fy5qj02 (93fy5qj02)]

With AccelMethod=exa, EXANoComposite, EXANoUploadToScreen and EXANoDownloadFromScreen, screen corruption remains the same. But I haven't seen X freeze so far.

[Michel Dänzer (michel-daenzer)]

Did it ever work properly with older versions of X? Can you rule out a hardware problem, e.g. does it work in another operating system?

[93fy5qj02 (93fy5qj02)]

There wasn't any corruption in Intrepid and before. It started when I upgraded to Jaunty. I'll see if I can confirm that with a live CD.

[93fy5qj02 (93fy5qj02)]

It appears to be a hardware problem that developed over time. I rebooted into an old Fedora Core 9 installation with fglrx driver and now it also showed corruption, although less and different. A new graphics card (of a different model) works fine. I don't know for sure if bug 23661 was also caused by hardware (Fedora never froze), but it might as well be, and I won't be able to debug it anyway, so feel free to close it, too. Thanks for your patience with this.

[Bryce Harrington (bryce)]

According to the upstream bug report, this is confirmed to have been due to hardware degradation.