Summary: | Random loss of video output, laptop connected via docking station to two external (DVI, VGA) monitors rotated in portrait mode | ||||||||
---|---|---|---|---|---|---|---|---|---|
Product: | DRI | Reporter: | ilakast | ||||||
Component: | DRM/Intel | Assignee: | Ville Syrjala <ville.syrjala> | ||||||
Status: | CLOSED WORKSFORME | QA Contact: | Intel GFX Bugs mailing list <intel-gfx-bugs> | ||||||
Severity: | major | ||||||||
Priority: | medium | CC: | ddstreet, ilakast, intel-gfx-bugs | ||||||
Version: | XOrg git | ||||||||
Hardware: | x86 (IA32) | ||||||||
OS: | Linux (All) | ||||||||
Whiteboard: | |||||||||
i915 platform: | I965GM | i915 features: | GPU hang | ||||||
Attachments: |
|
Description
ilakast
2014-06-13 23:41:49 UTC
xrandr -q Screen 0: minimum 320 x 200, current 2048 x 1280, maximum 32767 x 32767 LVDS1 connected (normal left inverted right x axis y axis) 1400x1050 60.0 + 60.0 50.0 1280x1024 60.0 1280x960 60.0 1360x768 59.8 60.0 1152x864 60.0 1024x768 60.0 800x600 60.3 56.2 640x480 59.9 VGA1 connected 1024x1280+1024+0 left (normal left inverted right x axis y axis) 376mm x 301mm 1280x1024 60.0*+ 75.0 1280x960 75.0 60.0 1152x864 75.0 1024x768 75.1 70.1 60.0 832x624 74.6 800x600 72.2 75.0 60.3 56.2 640x480 75.0 72.8 66.7 60.0 720x400 70.1 DVI1 connected 1024x1280+0+0 left (normal left inverted right x axis y axis) 376mm x 301mm 1280x1024 60.0*+ 1024x768 60.0 800x600 60.3 640x480 60.0 720x400 70.1 VIRTUAL1 disconnected (normal left inverted right x axis y axis) not a lot of useful info from Xorg.0.log.old [ 29970.500] (EE) intel(0): Detected a hung GPU, disabling acceleration. [ 29970.500] (EE) intel(0): When reporting this, please include /sys/class/drm/card0/error and the full dmesg. [ 30000.860] (II) intel(0): resizing framebuffer to 1400x1050 The loss of display is a characteristic of a GPU hang on that machine. Ville has been looking at improving the GPU reset which may help, but really that is a secondary problem. The primary issue is the cause of the GPU hang, for which we need the /sys/class/drm/card0/error captured before you reboot. Thank you for your input Chris. I will try to think of a way to get the content of /sys/class/drm/card0/error with no display, before rebooting. Any ideas on how to practically achieve that, most welcome. Created attachment 101195 [details]
Output of /sys/class/drm/card0/error
Managed to ssh into the crashed laptop, so please find attached. Any feedback would be GREATLY appreciated.
To help with debugging, it seems that crashes are not that random after all. The majority of times are connected to the Minitube application (version 2.1.6, latest as of today). Not necessarily with playback but also when stopping a video or when switching from fullscreen to window mode and vice versa. Created attachment 101979 [details]
latest crash when opening a pdf file (evince)
Output from /sys/class/drm/card0/error
(In reply to comment #7) > Created attachment 101979 [details] > latest crash when opening a pdf file (evince) > > Output from /sys/class/drm/card0/error This crash actually happened when opening a pdf file with evince, so my previous statement is now not true. Crashes are not software-related. The batch buffer gets overwritten. This seems to happen as it is being read, but I couldn't spot the culprit within the batch, so I presume it is the render cache that gets flushed after userspace has written the new batch. You definitely want to test with the latest kernel and xf86-video-intel. (In reply to comment #9) > You definitely want to test with the latest kernel and xf86-video-intel. Reporter, please try them. I have very very similar problem with i915 on an old dell optiplex 745 with ubuntu 14.04. I found this bug by googling "/build/buildd/linux-3.13.0/drivers/gpu/drm/i915/intel_display.c:922" I can consequently reproduce it by opening https://chrome.google.com/webstore/detail/user-agent-switcher-for-c/djflhoibgkdhkhhcedjiklpkjnoahfmg in google chrome. I have tested using UXA instead of SNA without that making any difference. So i googled ubuntu .deb package for newer linux kernels. 3.14.x - no change. same thing happens. 3.16.x - GPU crashes but kernel does not completely die. The window manager died though. 3.17.x - same as 3.16. I also found some deb package for xf86-video-intel-2.99.216 + git20141016 but issue is still there. I could try build xf86-video-intel from git and do git bisect, but I'd need some hint where to start from, eg, what would likely be a starting point for 'good'. I will not bisect kernel though, not on this old box. one more thing, its 64bit ubuntu here (aka amd64). Have you tried an updated kernel yet? Getting a current dmesg and crash dump might help, assuming this bug still exists... Timeout, closing. Please reopen if the problem persists with latest kernels. |
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.