[915GM] font corruption on Intel GMA900

[Zhmak]

Sometimes getting incomplete drawing of same glyphs of one font on whole screen.

In the screenshot you can see the "e" is drawn incompletely (both english and cyrillic too).

The bug is happening frequently and spontaneous on my Asus EEEPC 900.

(This issue affects Ubuntu's 10.10 or later. A workaround using the DebugWait parameter is described in comment #40 - https://bugs.launchpad.net/ubuntu/+source/xserver-xorg-video-intel/+bug/745608/comments/40 ).

[lspci]
Nux: lspci: 00:02.0 VGA compatible controller [0300]: Intel Corporation Mobile 915GM/GMS/910GML Express Graphics Controller [8086:2592] (rev 04)

[Bryce Harrington (bryce)]

Hi Zhmak,

Hmm, we've had a number of text corruption bugs however not one like this; and most of those have been resolved now. So this looks like it may be a new bug. Before I send this upstream, I have some questions for you:

1. When did you first start noticing this problem? E.g. did you recently update to 11.04 and start noticing it right away, or has it come on only recently?

2. You say it happens frequently. Do you notice any patterns to what makes it appear? E.g. certain programs you always have running, or certain activities you perform?

3. If you boot into Classic Desktop (no effects), can you reproduce the problem?

4. Could this simply be a bug in the font/language you're using? Can you try experimenting with different theme and language settings and note if you see it with other combinations?

[Zhmak (zhmak)]

Hi Bryce,

1. Can't remember first occurrence. I'm sure that I saw it in Maverick, may be in Lucid. Second screenshot made in Maverick.
2. It happening few times at week. I can surf web with firefox or opening console, nothing unusual.
3. I'll switch to "no effects" mode to find it out.
4. I don't think so. Using only default themes and fonts. You can find that another font affected on second attached screenshot (monospace font on console) .

Also I saw thin grey (transparent?) streak on maximize button yesterday and same streak on close button while writing this reply. Can' take screenshot because changing focus to gnome-screenshot cause to change button color and streak disappear.

Thanks for attention!

[Zhmak (zhmak)]

Bug still appear in "no effects" mode.

I saw it in gnome-screensaver prompt of password to login. I didn't press any key and screen blackened due inactivity. After this I pressed key to bring it back - all glyphs drawn well.

[Bryce Harrington (bryce)]

Alright, thanks, I think all the info needed for sending this upstream has been gathered.

[Zhmak (zhmak)]

I saw this bug again in gnome-screensaver unlock screen. At this time there are three different glyphs corrupted.

[Bryce Harrington (bryce)]

Can you confirm that you are still seeing this problem with the natty release?

If you are, can you test one more thing please - we are now providing builds of upstream's experimental driver and I'd like to confirm it also affects their code as well before we send this bug upstream. Please install the kernel and reboot onto it, reproduce the bug, and then attach a fresh 'dmesg > dmesg.txt' here. Thanks ahead of time!

[Zhmak (zhmak)]

Yes, I still seeing this bug.

Couple hours ago I did upgrade packages to up-to-date versions and rebooted system. Then I saw in unlock screen two different glyphs are stroked (picture attached to post).

How can I install newest kernel or experimental driver?

My current kernel is:
$ uname -a
Linux eee-900 2.6.38-8-generic #42-Ubuntu SMP Mon Apr 11 03:31:50 UTC 2011 i686 i686 i386 GNU/Linux

Version of xserver-xorg-video-intel is 2:2.14.0-4ubuntu7

[Anthony Waitz (anthony-aawaitz)]

Bryce,

I don't know if you received this feedback on the bug from Zhmak (it
was emailed only to me) so I am forwarding it to you.

---------- Forwarded message ----------
From: Zhmak <email address hidden>
Date: Fri, Apr 22, 2011 at 2:50 AM
Subject: [Bug 745608] Re: font corruption on Intel GMA900
To: <email address hidden>

Yes, I still seeing this bug.

Couple hours ago I did upgrade packages to up-to-date versions and
rebooted system. Then I saw in unlock screen two different glyphs are
stroked (picture attached to post).

How can I install newest kernel or experimental driver?

My current kernel is:
$ uname -a
Linux eee-900 2.6.38-8-generic #42-Ubuntu SMP Mon Apr 11 03:31:50 UTC
2011 i686 i686 i386 GNU/Linux

Version of xserver-xorg-video-intel is 2:2.14.0-4ubuntu7

** Attachment added: "stroked letters i unlock screen dialog"
  https://bugs.launchpad.net/ubuntu/+source/xserver-xorg-video-intel/+bug/745608/+attachment/2081933/+files/DSC04531.JPG

--
You received this bug notification because you are a direct subscriber
of a duplicate bug (742260).
https://bugs.launchpad.net/bugs/745608

Title:
 font corruption on Intel GMA900

Status in “xserver-xorg-video-intel” package in Ubuntu:
 Incomplete

Bug description:
 Binary package hint: xserver-xorg-video-intel

 Sometimes getting incomplete drawing of same glyphs of one font on
 whole screen.

 At screenshot you can see "e" drawn incomplete (both english and
 cyrillic too).

 That bug happening frequently and spontaneous on my Asus EEEPC 900.

 ProblemType: Bug
 DistroRelease: Ubuntu 11.04
 Package: xserver-xorg-video-intel 2:2.14.0-4ubuntu4
 ProcVersionSignature: Ubuntu 2.6.38-7.39-generic 2.6.38
 Uname: Linux 2.6.38-7-generic i686
 Architecture: i386
 CompizPlugins: No value set for
`/apps/compiz-1/general/screen0/options/active_plugins'
 CompositorRunning: compiz
 DRM.card0.LVDS.1:
  status: connected
  enabled: enabled
  dpms: On
  modes: 1024x600
  edid-base64:
 DRM.card0.VGA.1:
  status: disconnected
  enabled: disabled
  dpms: Off
  modes:
  edid-base64:
 Date: Wed Mar 30 14:44:17 2011
 DistUpgraded: Fresh install
 DistroCodename: natty
 DistroVariant: ubuntu
 GraphicsCard:
  Intel Corporation Mobile 915GM/GMS/910GML Express Graphics
Controller [8086:2592] (rev 04) (prog-if 00 [VGA controller])
    Subsystem: ASUSTeK Computer Inc. Device [1043:82d9]
    Subsystem: ASUSTeK Computer Inc. Device [1043:82d9]
 InstallationMedia: Ubuntu 11.04 "Natty Narwhal" - Alpha i386 (20110321)
 MachineType: ASUSTeK Computer INC. 900
 ProcEnviron:
  LANGUAGE=ru_RU:en
  LANG=ru_RU.UTF-8
  SHELL=/bin/bash
 ProcKernelCmdLine: BOOT_IMAGE=/boot/vmlinuz-2.6.38-7-generic
root=UUID=75374be3-dac3-432a-946e-99728dd11078 ro quiet splash
vt.handoff=7
 Renderer: Unknown
 SourcePackage: xserver-xorg-video-intel
 UpgradeStatus: No upgrade log present (probably fresh install)
 dmi.bios.date: 03/03/2009
 dmi.bios.vendor: American Megatrends Inc.
 dmi.bios.version: 1006
 dmi.board.asset.tag: To Be Filled By O.E.M.
 dmi.board.name: 900
 dmi.board.vendor: ASUSTeK Computer INC.
 dmi.board.version: x.xx
 dmi.chassis.asset.tag: 0x00000000
 dmi.chassis.type: 10
 dmi.chassis.vendor: ASUSTek Computer INC.
 dmi.chassis.version: x.x
 dm...

[Anthony Waitz (anthony-aawaitz)]

I booted a live CD of natty and did not see the problem. Do I need to
install to test?

On Fri, Apr 22, 2011 at 2:40 AM, Bryce Harrington
<email address hidden> wrote:
> Can you confirm that you are still seeing this problem with the natty
> release?
>
> If you are, can you test one more thing please - we are now providing
> builds of upstream's experimental driver and I'd like to confirm it also
> affects their code as well before we send this bug upstream.  Please
> install the kernel and reboot onto it, reproduce the bug, and then
> attach a fresh 'dmesg > dmesg.txt' here.  Thanks ahead of time!
>
> ** Changed in: xserver-xorg-video-intel (Ubuntu)
>       Status: Confirmed => Incomplete
>
> --
> You received this bug notification because you are a direct subscriber
> of a duplicate bug (742260).
> https://bugs.launchpad.net/bugs/745608
>
> Title:
>  font corruption on Intel GMA900
>
> Status in “xserver-xorg-video-intel” package in Ubuntu:
>  Incomplete
>
> Bug description:
>  Binary package hint: xserver-xorg-video-intel
>
>  Sometimes getting incomplete drawing of same glyphs of one font on
>  whole screen.
>
>  At screenshot you can see "e" drawn incomplete (both english and
>  cyrillic too).
>
>  That bug happening frequently and spontaneous on my Asus EEEPC 900.
>
>  ProblemType: Bug
>  DistroRelease: Ubuntu 11.04
>  Package: xserver-xorg-video-intel 2:2.14.0-4ubuntu4
>  ProcVersionSignature: Ubuntu 2.6.38-7.39-generic 2.6.38
>  Uname: Linux 2.6.38-7-generic i686
>  Architecture: i386
>  CompizPlugins: No value set for `/apps/compiz-1/general/screen0/options/active_plugins'
>  CompositorRunning: compiz
>  DRM.card0.LVDS.1:
>   status: connected
>   enabled: enabled
>   dpms: On
>   modes: 1024x600
>   edid-base64:
>  DRM.card0.VGA.1:
>   status: disconnected
>   enabled: disabled
>   dpms: Off
>   modes:
>   edid-base64:
>  Date: Wed Mar 30 14:44:17 2011
>  DistUpgraded: Fresh install
>  DistroCodename: natty
>  DistroVariant: ubuntu
>  GraphicsCard:
>   Intel Corporation Mobile 915GM/GMS/910GML Express Graphics Controller [8086:2592] (rev 04) (prog-if 00 [VGA controller])
>     Subsystem: ASUSTeK Computer Inc. Device [1043:82d9]
>     Subsystem: ASUSTeK Computer Inc. Device [1043:82d9]
>  InstallationMedia: Ubuntu 11.04 "Natty Narwhal" - Alpha i386 (20110321)
>  MachineType: ASUSTeK Computer INC. 900
>  ProcEnviron:
>   LANGUAGE=ru_RU:en
>   LANG=ru_RU.UTF-8
>   SHELL=/bin/bash
>  ProcKernelCmdLine: BOOT_IMAGE=/boot/vmlinuz-2.6.38-7-generic root=UUID=75374be3-dac3-432a-946e-99728dd11078 ro quiet splash vt.handoff=7
>  Renderer: Unknown
>  SourcePackage: xserver-xorg-video-intel
>  UpgradeStatus: No upgrade log present (probably fresh install)
>  dmi.bios.date: 03/03/2009
>  dmi.bios.vendor: American Megatrends Inc.
>  dmi.bios.version: 1006
>  dmi.board.asset.tag: To Be Filled By O.E.M.
>  dmi.board.name: 900
>  dmi.board.vendor: ASUSTeK Computer INC.
>  dmi.board.version: x.xx
>  dmi.chassis.asset.tag: 0x00000000
>  dmi.chassis.type: 10
>  dmi.chassis.vendor: ASUSTek Computer INC.
>  dmi.chassis.version: x.x
>  dmi.modalias: dmi:bvnAmericanMegatrendsInc.:bvr1006:bd03/03/2009:svnASUSTeKComputerINC.:pn900:pvr0704:rvnASUSTeKComp...

[Anthony Waitz (anthony-aawaitz)]

While testing from the live CD the screen got totally garbled and the
machine crashed. I guess I should expect this with an alpha version.
I had to power cycle it. This happened after installing Chromium for
further testing (all in the Live CD).

On Fri, Apr 22, 2011 at 9:23 AM, Anthony Waitz <email address hidden> wrote:
> I booted a live CD of natty and did not see the problem.  Do I need to
> install to test?
>
>
> On Fri, Apr 22, 2011 at 2:40 AM, Bryce Harrington
> <email address hidden> wrote:
>> Can you confirm that you are still seeing this problem with the natty
>> release?
>>
>> If you are, can you test one more thing please - we are now providing
>> builds of upstream's experimental driver and I'd like to confirm it also
>> affects their code as well before we send this bug upstream.  Please
>> install the kernel and reboot onto it, reproduce the bug, and then
>> attach a fresh 'dmesg > dmesg.txt' here.  Thanks ahead of time!
>>
>> ** Changed in: xserver-xorg-video-intel (Ubuntu)
>>       Status: Confirmed => Incomplete
>>
>> --
>> You received this bug notification because you are a direct subscriber
>> of a duplicate bug (742260).
>> https://bugs.launchpad.net/bugs/745608
>>
>> Title:
>>  font corruption on Intel GMA900
>>
>> Status in “xserver-xorg-video-intel” package in Ubuntu:
>>  Incomplete
>>
>> Bug description:
>>  Binary package hint: xserver-xorg-video-intel
>>
>>  Sometimes getting incomplete drawing of same glyphs of one font on
>>  whole screen.
>>
>>  At screenshot you can see "e" drawn incomplete (both english and
>>  cyrillic too).
>>
>>  That bug happening frequently and spontaneous on my Asus EEEPC 900.
>>
>>  ProblemType: Bug
>>  DistroRelease: Ubuntu 11.04
>>  Package: xserver-xorg-video-intel 2:2.14.0-4ubuntu4
>>  ProcVersionSignature: Ubuntu 2.6.38-7.39-generic 2.6.38
>>  Uname: Linux 2.6.38-7-generic i686
>>  Architecture: i386
>>  CompizPlugins: No value set for `/apps/compiz-1/general/screen0/options/active_plugins'
>>  CompositorRunning: compiz
>>  DRM.card0.LVDS.1:
>>   status: connected
>>   enabled: enabled
>>   dpms: On
>>   modes: 1024x600
>>   edid-base64:
>>  DRM.card0.VGA.1:
>>   status: disconnected
>>   enabled: disabled
>>   dpms: Off
>>   modes:
>>   edid-base64:
>>  Date: Wed Mar 30 14:44:17 2011
>>  DistUpgraded: Fresh install
>>  DistroCodename: natty
>>  DistroVariant: ubuntu
>>  GraphicsCard:
>>   Intel Corporation Mobile 915GM/GMS/910GML Express Graphics Controller [8086:2592] (rev 04) (prog-if 00 [VGA controller])
>>     Subsystem: ASUSTeK Computer Inc. Device [1043:82d9]
>>     Subsystem: ASUSTeK Computer Inc. Device [1043:82d9]
>>  InstallationMedia: Ubuntu 11.04 "Natty Narwhal" - Alpha i386 (20110321)
>>  MachineType: ASUSTeK Computer INC. 900
>>  ProcEnviron:
>>   LANGUAGE=ru_RU:en
>>   LANG=ru_RU.UTF-8
>>   SHELL=/bin/bash
>>  ProcKernelCmdLine: BOOT_IMAGE=/boot/vmlinuz-2.6.38-7-generic root=UUID=75374be3-dac3-432a-946e-99728dd11078 ro quiet splash vt.handoff=7
>>  Renderer: Unknown
>>  SourcePackage: xserver-xorg-video-intel
>>  UpgradeStatus: No upgrade log present (probably fresh install)
>>  dmi.bios.date: 03/03/2009
>>  dmi.bios.vendor: American Megatrends Inc.
>>  dmi.bi...

[CassieMoondust (cassie-lx)]

I have the same issue with my eeePC 701 4G. After some time fonts begin to get corrupt.
I have a pictures wich i upload as attachment.

Maybe i can help, what can i do to give you more information?

[Sitsofe Wheeler (sitsofe)]

I see this too. I've filed it upstream as https://bugs.freedesktop.org/show_bug.cgi?id=36326 .

[Sitsofe Wheeler (sitsofe)]

I forgot to say that I am using an EeePC 900 too and its chipset is an 915GM.

[Ferry Toth (ftoth)]

Yes I have it too on Eepc 900 and also on desktop with 'Intel Corporation 82945G/GZ Integrated Graphics Controller (rev 02)". All with Kubuntu.

See attachment following.

Had the same in Debian Squeeze. (same driver I guess, and older KDE).

Do not see the problem on Natty Kubuntu with Nvidia card.

Ferry

[Ferry Toth (ftoth)]

Check the 'v' in ex-yogi van de wijs

Some sizes look good, some not good.

[Ferry Toth (ftoth)]

And it's not just fonts, check these icons.

[Robert Hooker (sarvatt)]

Sorry for asking again, but do you have natty-updates enabled and 2:2.14.0-4ubuntu7.1 installed that you can test again with? It contains a fix that is relevant to this bug

[Ferry Toth (ftoth)]

Yes, I have natty-updates enabled and 2:2.14.0-4ubuntu7.1 installed.

The above screenshots were with the latest packages (natty-proposed not enabled).

I looks like the longer I have an application running the more distored the fonts. I can hardly read what I'm typing right now.

Could it be related to some font cache?

Ferry

[Toucan (karmiktoucan)]

I have the same bug on EEEPC900:
OS : Debian GNU/Linux
Kernel versions : 2.6.38-2-686 and 2.6.39-rc5
Drivers: xserver-xorg-intel 2.14.0 and 2.15.0-1

With Kernel 2.6.32-5-686 this bug doesn't appear.

[S. Christian Collins (s-chriscollins)]

On my system, this bug only seems to occur after resuming from suspend. Has anybody else here noticed a similar trend?

** My System **
PC: HP Pavilion dv1550se laptop
CPU: Intel(R) Pentium(R) M processor 1.60GHz
RAM: 1GB DDR400
Video: Mobile 915GM/GMS/910GML Express Graphics Controller
Sound: 82801FB/FBM/FR/FW/FRW (ICH6 Family) AC'97 Audio Controller
OS: Xubuntu 11.04 i386

[Ferry Toth (ftoth)]

No, on one system I never suspend. It seems more related to the time an application or the system is running.

BTW it also occured on Debian Squeeze with kernels earlier then 2.6.38. I am not sure but I think it got introduced with the intel driver 2.14.

Ferry

[Toucan (karmiktoucan)]

On my netbook (eeePC 900) with kernel 2.6.32-5-686 and intel drivers 2.14.0 or 2.15.0 this bug does not appear, but with newer kernels ( 2.6.38 or 2.6.39-rc5) there is corruptions.

[Ferry Toth (ftoth)]

Hmm, This problem seems to depend on the actual Intel chip:

eeepc 900: sometimes the broken fonts and other anomalies occur, restarting X seems to make them go away.
Chip set: VGA compatible controller: Intel Corporation Mobile 915GM/GMS/910GML Express Graphics Controller (rev 04)

asus mother board 1: always broken fonts, switching to different fonts or sizes can make the screen readable again.
Chip set: VGA compatible controller: Intel Corporation 82945G/GZ Integrated Graphics Controller (rev 02)

asus mother board2: no problems do far.
Chip set: VGA compatible controller: Intel Corporation 82G33/G31 Express Integrated Graphics Controller (rev 10)

All 3 systems running same Natty Kubuntu, 915GM and 82G33 in i386, 82945G in amd64.

Ferry

[Ferry Toth (ftoth)]

Given how badly readable the display is on 82945G I would say the importance should be higher then 'low'.

[Sitsofe Wheeler (sitsofe)]

Ferry:
Is the corruption that you see on the 945G the same as what you see on the EeePC/in the screenshot in comment #1 or is it slightly different?

[Ferry Toth (ftoth)]

Yes, and no. I'm not using a dutch localization, so you I don't normally have Cyrillic fonts :-)

Also I'm using Kubuntu (though with evolution).

But both on Eeepc and 945G the problem is that fonts are broken. Each broken letter shows the same distortion as shown in screenshot $1 (where the 'e' is broken).

But the broken letters are not the same with each boot. Also in firefox when you increase the font size it changes and at a certain font size will go away.

With Eeepc the system is usable mostly, with 845G text can become totally unreadable.

Screen shots in #16 - #19 were taken on the 945G, but after I had been fiddling with changing the default kwin fonts, so it didn't even look so bad as normally. But after reboot the at first 'readable' font are again broken.

But as shown in screen shot #19 also icons are sometimes distorted (as you can see in the navigation iicons).

Ferry

[dmiranda (dmiranda)]

I have the same problem using Ubuntu natty completely updated.

[Ferry Toth (ftoth)]

This is related to bugzilla.kernel.org #26002.

I think.

Ferry

[Sitsofe Wheeler (sitsofe)]

Ferry:
It's really hard to tell and there's a risk that a lot of issues that might be need to be separate are lumped together.

I've seen different types of corruption on my EeePC 900 with an 915GM since 10.10:
Horizontal line(s) through a glyph. Affects the same character of that font/size repeatedly, problem persists past a repaint. Screenshot: https://bugs.freedesktop.org/attachment.cgi?id=45730
A rectangular/square "block" of the glyph is missing. Affects the same character of that font/size repeatedly, problem persists past a repaint. Probably the same as the above but might be different. Tends to be rarer than the first type of corruption.

The 945GM I have never shows the above problems (although it shows another issue - https://bugzilla.redhat.com/attachment.cgi?id=493733 ). I've also yet to see any font corruption on a machine with a 965G graphics chipset...

All the corruption above happens without needing to suspend/hibernate and resume.

We need to be clear on what types of corruption happen, where they happen, what triggers the corruption and which graphics card chipset is being used. Another thing to check is whether turning off tiling by putting

Section "Device"
    Identifier "Intel"
    Driver "intel"
    Option "Tiling" "false"
EndSection

in /etc/X11/xorg.conf and then rebooting changes whether the corruption appears (the above is purely for testing purposes though - don't run daily with such a configuration as it will be very slow!).

[Ferry Toth (ftoth)]

Sitofe,

Currently on my eeepc I see these both problems sporadically, but it is very workable, so I don't dare to complain.
I have had these same corruptions on the eeepc before. It seems to be coming and going with linux releases.

On the desktop with 945G at times ff becomes unreadable, until I press increase font size 2 times (ctrl-+).

Both problems you mention can be observed (exactly as you say, it affects a certain character/font/size combination, and eventually more and more become broken, switching to another size/font temporarily solves the problem), but the font problem is the most common and disturbing.

Note that while some complain about their terminal program, I don't see it in konsole (KDE terminal). This may be due to the font used?

So yes, I see the things you see, right now the 945G is affected most.

I can try tiling=false, and will report back.

But I don't know what else I can do. Any other suggestions?

Ferry

[Sitsofe Wheeler (sitsofe)]

Ferry:
I have good suggestions I'm afraid. Just out of interest, what CPU and which screen resolution is the 945G running at? How much RAM? I'm wondering if there's any sort of pattern to this problem...

[Ferry Toth (ftoth)]

Is there anyway I can automatically collect all relevant info as in the bug description?

Now I created a new bug report #785388. I can hardly believe that is the easy way.

Ferry

[Sitsofe Wheeler (sitsofe)]

Ferry:
Thanks for posting that. A quick glance shows the 945G machine is very different to the 915GM EeePC - it is using a 64 distro, has 4GBytes of memory the screen resolution is 1920x1200 connected by VGA etc. The only commonality I could see is that they are both ASUStek machines. Sigh.

[Ferry Toth (ftoth)]

The are both Kubuntu Natty using the Intel driver.

I could boot the 64-bit 945G with a 32-bit USB stick and see if that helps.

And I could switch temporarily to a lower resolution, but not 1024x600 I guess.

Actually the 64-bit machine has only 2GB of RAM not 4.

I'll report back with the results.

Ferry

[Sitsofe Wheeler (sitsofe)]

Ferry:
I would be surprised if 64 or 32 bit made a difference. Basically your results have ruled out that it's related to whether a 64 bit OS is used. Screen resolution difference also seem unlikely (I've reproduced the problem at 640x480 on my EeePC). I would be surprised if it is RAM related.

[Ferry Toth (ftoth)]

I think so too.

It will probably turn out to be related to the 945G chip, or if I am unlucky, to the revision of the chip.

But I have had these problems in the passed, and as the kernel got updated they went away again . Soi I am sure it can be done.

Ferry

[Sitsofe Wheeler (sitsofe)]

Ferry:
Can you test if having

Section "Device"
    Identifier "Intel"
    Driver "intel"
    Option "DebugWait" "true"
EndSection

in xorg.conf helps?

[Chris Wilson (ickle)]

To be honest, I had hoped that this would have proven to be the unfenced-BLT bug. The known w/a for UXA is to enable DebugWait, which implies that UXA is not handling its domains correctly.

The most likely candidate is:

diff --git a/src/intel_uxa.c b/src/intel_uxa.c
index 76a3146..40e3b67 100644
--- a/src/intel_uxa.c
+++ b/src/intel_uxa.c
@@ -1379,8 +1379,8 @@ Bool intel_uxa_init(ScreenPtr screen)
        }

        /* PutImage */
- intel->uxa_driver->put_image = intel_uxa_put_image;
- intel->uxa_driver->get_image = intel_uxa_get_image;
+ //intel->uxa_driver->put_image = intel_uxa_put_image;
+ //intel->uxa_driver->get_image = intel_uxa_get_image;

        intel->uxa_driver->prepare_access = intel_uxa_prepare_access;
        intel->uxa_driver->finish_access = intel_uxa_finish_access;

[Chris Wilson (ickle)]

I've pushed a tree to remove some of the optimisations to see if they are the cause of the corruption:

http://cgit.freedesktop.org/~ickle/xf86-video-intel #glyph-cache-bug

Can you please try this branch to see if does the trick?

Rough guide to testing:

$ git clone git://people.freedesktop.org/~ickle/xf86-video-intel
$ cd xf86-video-intel
$ git checkout origin/glyph-cache-bug
$ ./autogen.sh --prefix=/usr
$ make
$ sudo make install
# restart X

[Sitsofe Wheeler (sitsofe)]

Chris:
I'm going to struggle to even test your simple patch. How easy is it to cross compile this on a 64 bit Fedora for a 32 bit Ubuntu with a different xorg?

[Chris Wilson (ickle)]

Nigh-on impossible unless you have the entire build environment for your target as well (-intel will depend upon the X server ABI as defined in the headers, which of course vary over time).

Don't worry the fix is just to switch to SNA, which will happen soon enough.

[Andris Berzins (pkix)]

I have the same issue running DELL Latitude D430 with Intel Mobile 945GM/GMS/GME, 943/940GML Express Integrated Graphics Controller.

I can confirm, that SNA workaround fixes the issue.

$ cat /etc/X11/xorg.conf
Section "Device"
 Identifier "Card0"
 Driver "intel"
 Option "AccelMethod" "sna"
EndSection

[Sitsofe Wheeler (sitsofe)]

Ok I've built the intel driver on the git branch (the log mentioned it was version 2.20.16) mentioned and the problem is still there.

[Chris Wilson (ickle)]

Created attachment 71918
Flush around glyph updates

[Chris Wilson (ickle)]

Created attachment 71919
Flush around glyph updates

Rushed.

[Sitsofe Wheeler (sitsofe)]

Created attachment 71944
Disable glyph cache

[Sitsofe Wheeler (sitsofe)]

Created attachment 71949
Screenshot of patched driver with filled in glyphs

If you look at this screenshot which is using the patch on http://paste.debian.net/218151/ you will see that some blocks have lines through them

[Sitsofe Wheeler (sitsofe)]

Created attachment 71953
Screenshot 2 of patched driver with filled in glyphs

This was taken using the patch on http://paste.debian.net/218164/ but no errors were logged...

[Chris Wilson (ickle)]

Sitosfe, if you get a chance, please could you try -intel.git (or 2.20.18)? I've implemented stricter throttling, so I am wondering if that worksaround the issue like DebugWait.

[Chris Wilson (ickle)]

Sitsofe reports that commit 441ef916ae6569c88b3d6abaf7fea4d69be49d76
Author: Chris Wilson <email address hidden>
Date: Thu Jan 10 19:14:21 2013 +0000

    intel: Throttle harder

does improve matters, but I'm still at a loss to explain why.

[Chris Wilson (ickle)]

*** Bug 37782 has been marked as a duplicate of this bug. ***

[Chris Wilson (ickle)]

*** Bug 49396 has been marked as a duplicate of this bug. ***

[Chris Wilson (ickle)]

*** Bug 39851 has been marked as a duplicate of this bug. ***

[Chris Wilson (ickle)]

*** Bug 55881 has been marked as a duplicate of this bug. ***

[Paulepanter (paulepanter)]

(In reply to comment #76)
> Sitsofe reports that commit 441ef916ae6569c88b3d6abaf7fea4d69be49d76
> Author: Chris Wilson <email address hidden>
> Date: Thu Jan 10 19:14:21 2013 +0000
>
> intel: Throttle harder
>
> does improve matters, but I'm still at a loss to explain why.

Finally I got some time to test that patch too. I checked out branch `debian-unstable` from the Debian packaging Git repository [1] (based on release 2.19.0)

    commit 8ab872522ae0891b8fb52fbb31db52d581c3efa3
    Author: Julien Cristau <email address hidden>
    Date: Sun Sep 16 20:47:05 2012 +0200

        Upload to unstable

and applied the following patch on top of it.

    commit f49c3cacd25ffb78ccfff49b001ba89375044145
    Author: Chris Wilson <email address hidden>
    Date: Thu Jan 10 19:14:21 2013 +0000

        intel: Throttle harder

Limited testing shows that the glyph corruption is gone. Unfortunately performance seems to be slower too. That means for example maximizing a window it does not happen instantly and I can see how it is redrawn. So basically the same behavior with `"DebugWait" "true"`, if I remember correctly.

[1] http://anonscm.debian.org/gitweb/?p=pkg-xorg/driver/xserver-xorg-video-intel.git;a=summary

[Chris Wilson (ickle)]

(In reply to comment #81)
> and applied the following patch on top of it.
>
> commit f49c3cacd25ffb78ccfff49b001ba89375044145
> Author: Chris Wilson <email address hidden>
> Date: Thu Jan 10 19:14:21 2013 +0000
>
> intel: Throttle harder
>
> Limited testing shows that the glyph corruption is gone. Unfortunately
> performance seems to be slower too. That means for example maximizing a
> window it does not happen instantly and I can see how it is redrawn. So
> basically the same behavior with `"DebugWait" "true"`, if I remember
> correctly.

What I realised later was that Sitosfe was hitting the bug in libdrm-intel that was causing it to fallback to swrast. After applying that patch, you also need libdrm-intel-2.4.41 or just http://cgit.freedesktop.org/mesa/drm/commit/?id=fdda97007b1dbf95beb16a0e3510fd36c89e8c33

Paul, can you update libdrm and see if the bug returns?

[Paulepanter (paulepanter)]

(In reply to comment #82)
> (In reply to comment #81)
> > and applied the following patch on top of it.
> >
> > commit f49c3cacd25ffb78ccfff49b001ba89375044145
> > Author: Chris Wilson <email address hidden>
> > Date: Thu Jan 10 19:14:21 2013 +0000
> >
> > intel: Throttle harder
> >
> > Limited testing shows that the glyph corruption is gone. Unfortunately
> > performance seems to be slower too. That means for example maximizing a
> > window it does not happen instantly and I can see how it is redrawn. So
> > basically the same behavior with `"DebugWait" "true"`, if I remember
> > correctly.
>
> What I realised later was that Sitosfe was hitting the bug in libdrm-intel
> that was causing it to fallback to swrast. After applying that patch, you
> also need libdrm-intel-2.4.41 or just
> http://cgit.freedesktop.org/mesa/drm/commit/
> ?id=fdda97007b1dbf95beb16a0e3510fd36c89e8c33
>
> Paul, can you update libdrm and see if the bug returns?

I applied the commit you mentioned on top of

    libdrm (2.4.40-1~deb7u2) sid; urgency=low

and rebuilt the package. Unfortunately I think there is no difference. The glyph corruption is still gone but it is also still slow, so swrast still might be used.

How can I check, what is used, without `glxinfo` installed? I could not spot anything useful in `/var/log/Xorg.0.log`.

[Chris Wilson (ickle)]

Basically inspect 'sudo perf top' for pixman functions. There is a slim chance you trigger bug 59711 (as in I have no idea what that is about...) and end up with something in the Xorg.log.

[Paulepanter (paulepanter)]

(In reply to comment #83)
> (In reply to comment #82)
> > (In reply to comment #81)

[…]

> > Paul, can you update libdrm and see if the bug returns?
>
> I applied the commit you mentioned on top of
>
> libdrm (2.4.40-1~deb7u2) sid; urgency=low
>
> and rebuilt the package. Unfortunately I think there is no difference. The
> glyph corruption is still gone but it is also still slow, so swrast still
> might be used.
>
> How can I check, what is used, without `glxinfo` installed? I could not spot
> anything useful in `/var/log/Xorg.0.log`.

It looks like swrast is not used.

    OpenGL renderer string: Mesa DRI Intel(R) 915GM x86/MMX/SSE2
    OpenGL version string: 1.4 Mesa 8.0.5

Here is the full output.

$ glxinfo
name of display: :0
display: :0 screen: 0
direct rendering: Yes
server glx vendor string: SGI
server glx version string: 1.4
server glx extensions:
    GLX_ARB_multisample, GLX_EXT_import_context, GLX_EXT_texture_from_pixmap,
    GLX_EXT_visual_info, GLX_EXT_visual_rating, GLX_MESA_copy_sub_buffer,
    GLX_OML_swap_method, GLX_SGI_swap_control, GLX_SGIS_multisample,
    GLX_SGIX_fbconfig, GLX_SGIX_pbuffer, GLX_SGIX_visual_select_group,
    GLX_INTEL_swap_event
client glx vendor string: Mesa Project and SGI
client glx version string: 1.4
client glx extensions:
    GLX_ARB_create_context, GLX_ARB_create_context_profile,
    GLX_ARB_get_proc_address, GLX_ARB_multisample, GLX_EXT_import_context,
    GLX_EXT_visual_info, GLX_EXT_visual_rating, GLX_EXT_framebuffer_sRGB,
    GLX_EXT_create_context_es2_profile, GLX_MESA_copy_sub_buffer,
    GLX_MESA_multithread_makecurrent, GLX_MESA_swap_control,
    GLX_OML_swap_method, GLX_OML_sync_control, GLX_SGI_make_current_read,
    GLX_SGI_swap_control, GLX_SGI_video_sync, GLX_SGIS_multisample,
    GLX_SGIX_fbconfig, GLX_SGIX_pbuffer, GLX_SGIX_visual_select_group,
    GLX_EXT_texture_from_pixmap, GLX_INTEL_swap_event
GLX version: 1.4
GLX extensions:
    GLX_ARB_get_proc_address, GLX_ARB_multisample, GLX_EXT_import_context,
    GLX_EXT_visual_info, GLX_EXT_visual_rating, GLX_MESA_copy_sub_buffer,
    GLX_MESA_multithread_makecurrent, GLX_MESA_swap_control,
    GLX_OML_swap_method, GLX_OML_sync_control, GLX_SGI_make_current_read,
    GLX_SGI_swap_control, GLX_SGI_video_sync, GLX_SGIS_multisample,
    GLX_SGIX_fbconfig, GLX_SGIX_pbuffer, GLX_SGIX_visual_select_group,
    GLX_EXT_texture_from_pixmap
OpenGL vendor string: Tungsten Graphics, Inc
OpenGL renderer string: Mesa DRI Intel(R) 915GM x86/MMX/SSE2
OpenGL version string: 1.4 Mesa 8.0.5
OpenGL extensions:
    GL_ARB_multisample, GL_EXT_abgr, GL_EXT_bgra, GL_EXT_blend_color,
    GL_EXT_blend_minmax, GL_EXT_blend_subtract, GL_EXT_copy_texture,
    GL_EXT_polygon_offset, GL_EXT_subtexture, GL_EXT_texture_object,
    GL_EXT_vertex_array, GL_EXT_compiled_vertex_array, GL_EXT_texture,
    GL_EXT_texture3D, GL_IBM_rasterpos_clip, GL_ARB_point_parameters,
    GL_EXT_draw_range_elements, GL_EXT_packed_pixels, GL_EXT_point_parameters,
    GL_EXT_rescale_normal, GL_EXT_separate_specular_color,
    GL_EXT_texture_edge_clamp, GL_SGIS_generate_mipmap,
    GL_SGIS_texture_border_clamp, GL_SGIS_texture_edge_clamp,
    GL_SGIS_textu...

[Paulepanter (paulepanter)]

(In reply to comment #84)
> Basically inspect 'sudo perf top' for pixman functions.

Alright.

> There is a slim chance you trigger bug 59711 (as in I have no idea what that
> is about...) and end up with something in the Xorg.log.

Unfortunately with two patches applied, using `xrandr` to enable the external screen seems to lock up the system. I have to find more time to look into this.

[Chris Wilson (ickle)]

Not an issue for raring (due to it using SNA).

[Chris Wilson (ickle)]

*** Bug 65433 has been marked as a duplicate of this bug. ***

[João M. S. Silva (joao-m-santos-silva)]

I have removed xorg-x11-drv-intel thinking it would fall back to xorg-x11-drv-vesa but it seems to have fallen back to i915 from the kernel. Is this correct?

Anyway, the font corruption issue disappeared.

What is the difference between using the driver from xorg-x11-drv-intel and the one from the kernel?

[Chris Wilson (ickle)]

(In reply to comment #88)
> I have removed xorg-x11-drv-intel thinking it would fall back to
> xorg-x11-drv-vesa but it seems to have fallen back to i915 from the kernel.
> Is this correct?

No. I guess you are using xorg-x11-drv-modesetting as that is the modern fallback.

> Anyway, the font corruption issue disappeared.

The right fix is to enable Option "AccelMethod" "SNA".

[thinkpad (fellowsgarden)]

"Years ago" I applied the solution described here http://askubuntu.com/a/49941/16023

I'm still on "precise" (not "raring").

Lately, the issue's been cropping up more often again.

What's the latest consensus on how to resolve this issue in "precise" ?

Anything else worth trying?

[Diego (gran-diego)]

Had this same issue. Workaround in #167 works for me.

Ubuntu Raring on Dell Latitude E4300. Intel GM45 video card (i915 module)

[Linux-testing-marantz (linux-testing-marantz)]

Created attachment 93730
Screenshot 1 of 5, terminal filled with X's, corruption changes after scrot

[Linux-testing-marantz (linux-testing-marantz)]

Created attachment 93731
Screenshot 1 of 5, terminal filled with X's, corruption changes after scrot

[Linux-testing-marantz (linux-testing-marantz)]

Created attachment 93732
Screenshot 2 of 5, terminal filled with X's, corruption changes after scrot

[Linux-testing-marantz (linux-testing-marantz)]

Created attachment 93733
Screenshot 3 of 5, terminal filled with X's, corruption changes after scrot

[Linux-testing-marantz (linux-testing-marantz)]

Created attachment 93734
Screenshot 4 of 5, terminal filled with X's, corruption changes after scrot

[Linux-testing-marantz (linux-testing-marantz)]

Created attachment 93735
Screenshot 5 of 5, terminal filled with X's, corruption changes after scrot

[Linux-testing-marantz (linux-testing-marantz)]

Created attachment 93736
Log files dmesg + Xorg,0,log showing machine info for screenshots above

[Sitsofe Wheeler (sitsofe)]

Screenshots and logs for <email address hidden> are on a system running
Kernel 3.11.0-12-generic #19-Ubuntu
xorg-intel 2:2.99.904-0ubuntu2
SNA is on: intel(0): SNA initialized with Eaglelake (gen4.5) backend
Chipset is a Q45/Q43.

linux.testing.marantz:
I've add you to the CC list of this bug report so you will see a reply. It is a good practice to do this otherwise you will miss followups...

It is unlikely that the bug you are seeing is the one filed here as your chipset is not an i915 series and you are using SNA on a recent kernel+xorg. I would strongly urge you to open new bug, attach one of your screenshots, your xorg and dmesg logs and then report the versions of your kernel, xorg and xorg intel along with steps on how to reproduce the issue in a comment.

[Sitsofe Wheeler (sitsofe)]

I don't see this in Ubuntu 14.04 as it uses SNA and UXA is going to die so I think we can finally close this.

[Peter Nowee (peter-nowee)]

I ran into this bug on a fresh install of Debian 8 (Jessie) RC2 on an EeePC 701 4G with Intel i915GM graphics. Linux kernel 3.16, X.Org 1.16.4, X.Org intel module 2.21.15.

Using SNA instead of UXA, as suggested above, seems to still solve it:

# cat /etc/X11/xorg.conf
Section "Device"
        Identifier "Intel Graphics"
        Driver "intel"
        Option "AccelMethod" "sna"
EndSection

I see that Debian is preparing to move to SNA [1]:

> xserver-xorg-video-intel (2:2.99.911+git20140529-1~exp2) experimental; urgency=medium
> ..
> * Stop overriding default upstream acceleration method (i.e. default to SNA
> instead of UXA).
>
> -- Vincent Cheng <email address hidden> Mon, 02 Jun 2014 23:47:23 -0700

However, that change is still in Debian's "experimental" area, not in the upcoming stable (Jessie) release.

(Links to corresponding Debian bug reports in Paul's comment 44 above.)

[1] http://metadata.ftp-master.debian.org/changelogs//main/x/xserver-xorg-video-intel/experimental_changelog

[Ubuntu QA Website (ubuntuqa)]

This bug has been reported on the Ubuntu ISO testing tracker.

A list of all reports related to this bug can be found here:
http://iso.qa.ubuntu.com/qatracker/reports/bugs/745608