Bug 92716 - GPU HANG which distorts my screen display and seems to affect sleeping/hibernating after I upgraded from Kubuntu 15.04 to 15.10 on a 64 bits machine
Summary: GPU HANG which distorts my screen display and seems to affect sleeping/hibern...
Status: CLOSED INVALID
Alias: None
Product: DRI
Classification: Unclassified
Component: DRM/Intel (show other bugs)
Version: unspecified
Hardware: x86-64 (AMD64) Linux (All)
: medium major
Assignee: Intel GFX Bugs mailing list
QA Contact: Intel GFX Bugs mailing list
URL:
Whiteboard:
Keywords:
Depends on:
Blocks:
 
Reported: 2015-10-28 20:48 UTC by mishal.strogoff
Modified: 2017-06-02 12:21 UTC (History)
2 users (show)

See Also:
i915 platform: I965G
i915 features: power/suspend-resume


Attachments
Output of log files, some commands and one screenshot (167.98 KB, application/x-zip-compressed)
2015-10-28 20:48 UTC, mishal.strogoff
no flags Details
Output of log files, some commands and one screenshot for the failure at November 24th 2015 (279.33 KB, application/octet-stream)
2015-11-24 16:53 UTC, mishal.strogoff
no flags Details

Description mishal.strogoff 2015-10-28 20:48:11 UTC
Created attachment 119264 [details]
Output of log files, some commands and one screenshot

About one week ago I upgraded to Kubuntu 15.10 and after that I frequently (several times per hour) suffer from distortions in my display screen. I already filed two bugs reports in Ubuntu launchpad site, available at https://bugs.launchpad.net/ubuntu/+source/intel-vaapi-driver/+bug/1509868 and https://bugs.launchpad.net/ubuntu/+source/linux/+bug/1509903 and today, after a failed attempt to hibernate my computer via a GUI menu option in KDE, I reviewed the /var/log/syslog file, where I found some instructions to file a bug in the site bugs.freedesktop.org, so here I am.

Distribution: Kubuntu 15.10
uname -a: Linux pc 4.2.0-16-generic #19-Ubuntu SMP Thu Oct 8 15:35:06 UTC 2015 x86_64 x86_64 x86_64 GNU/Linux
Display connector: VGA (I guess it's DE-15 male plug according to Wikipedia)

I attach some related files and one screenshot (resized to get a smaller size)

Please review this and try to solve it. If I can help with something, let me know.

Note: as I wrote this description, my screen display distorted and my workaround is to press Alt+Tab and minimize and restore the opened applications till I get a more or less decent viewing.
Comment 1 Mika Kuoppala 2015-11-16 11:17:26 UTC
For me it seems that the batchbuffer got overwritten while it was being executed.
Could you upload a few more error states if this is easily reproducible to get more samples. Preferrably all on clean reboots before suspend/resume cycles.
Comment 2 mishal.strogoff 2015-11-24 16:51:02 UTC
(In reply to Mika Kuoppala from comment #1)
> For me it seems that the batchbuffer got overwritten while it was being
> executed.
> Could you upload a few more error states if this is easily reproducible to
> get more samples. Preferrably all on clean reboots before suspend/resume
> cycles.

That's possible, however I don't know the details about batchbuffer or something like that. Today I powered on my computer, a fresh boot, and waited until the failure raised again. I remember what I was doing exactly: I had opened a text editor (Kate), a file browser (Dolphin) and a web browser (Firefox) and Firefox was the active application and in the web page http://www.shodor.org/interactivate/activities/TheMandelbrotSet/ I clicked the "Start Over" button and bump!!! my screen display became a mess!

I will attach some relevant files again.
Comment 3 mishal.strogoff 2015-11-24 16:53:23 UTC
Created attachment 120087 [details]
Output of log files, some commands and one screenshot for the failure at November 24th 2015

The promised new files attached to this bug
Comment 4 mishal.strogoff 2015-12-25 18:53:54 UTC
I've found a way to bypass this annoying bug, thanks to the documentation available at https://wiki.archlinux.org/index.php/Intel_graphics

More than a week has passed without suffering the failure, since I implemented the change suggested there :)

In summary, I created the file /etc/X11/xorg.conf.d/20-intel.conf with this content (I left some commented lines as a reference to other possible workarounds that I didn't test):

# content of file begins here                                                                                                     
Section "Device"                                                                                                                                                        
  Identifier  "Intel Graphics"                                                                                                                                          
  Driver      "intel"                                                                                                                                                   
  Option      "AccelMethod"  "uxa"                                                                                                                                      
  #disabling the GPU usage with the NoAccel option                                                                                                                      
  #Option "NoAccel" "True"                                                                                                                                              
  #Alternatively, try to disable the 3D acceleration only with the DRI option
  #Option "DRI" "False"
EndSection
# content of file ends here                                                                                                     

If you know about another solution or alternate configuration, I would like to know the details.

Thanks!
Comment 5 mishal.strogoff 2015-12-25 18:57:16 UTC
I mean the creation of the file /etc/X11/xorg.conf.d/20-intel.conf solved the failure of the video display. The other problem of the sleeping/hibernating functions continues to be present.
Comment 6 yann 2016-09-20 16:14:00 UTC
(In reply to mishal.strogoff from comment #5)
> I mean the creation of the file /etc/X11/xorg.conf.d/20-intel.conf solved
> the failure of the video display. The other problem of the
> sleeping/hibernating functions continues to be present.

We seem to have neglected the bug a bit, apologies.
There were improvements pushed in kernel that will benefit to your system, so please re-test with latest kernel to see if this issue is still occurring.

If it is still happening, please attached kernel log (consider to add "drm.debug=0x1e log_buf_len=1M" as additional parameter of your boot command line).
Comment 7 Ricardo 2017-02-22 16:12:28 UTC
mishal.strogoff@veryrealemail.com is this GPU hang seen updating to the latest kernel? please review comment before if it still occurs to add necessary information. 

If the hand does not occur please set the bug to resolved

Otherwise if there is no response within 30 days the bug will be closed due to lack of activity.
Comment 8 Ricardo 2017-05-30 17:41:10 UTC
based on the lack of activity and a response from the submitter to update results with latest configuration the bug will be closed


Use of freedesktop.org services, including Bugzilla, is subject to our Code of Conduct. How we collect and use information is described in our Privacy Policy.