Bug 28882

Summary: Screen corruption and GPU hangs
Product: Drivers Reporter: Ko Mi (chaostya.test)
Component: Video(DRI - Intel)Assignee: drivers_video-dri-intel (drivers_video-dri-intel)
Severity: normal CC: akpm, amitshah, chris, florian, maciej.rutecki, rjw, tiagomatos
Priority: P1    
Hardware: All   
OS: Linux   
Kernel Version: 2.6.38-rc4 Subsystem:
Regression: Yes Bisected commit-id:
Bug Depends on:    
Bug Blocks: 27352    
Attachments: kern.log

Description Ko Mi 2011-02-11 18:17:20 UTC
This didn't happened in pre 2.6.38 kernels, and happened couple times since I upgraded to 2.6.38-rc*
I'm running Ubuntu with mainline kernel builds from PPA.

All of a sudden screen corrupts showing color stripes.

X.org.0.log contains:

[ 16613.665] (EE) intel(0): Detected a hung GPU, disabling acceleration.
[ 16613.790] (WW) intel(0): i830_uxa_prepare_access: gtt bo map failed: Input/output error
and keeps coming with last message till I restart

But I can switch to another TTY with Ctrl+Alt+F1 and it shows good, I can even login and check logs. But when I get back to X with Ctrl+Alt+F7 I see a normal screen but image is stalled and contains pre corruption picture mouse cursor is shown but system doesn't respond to clicks.

Feb 11 19:40:19 chaostya-laptop kernel: [16613.136032] [drm:i915_hangcheck_elapsed] *ERROR* Hangcheck timer elapsed... GPU hung
Feb 11 19:40:19 chaostya-laptop kernel: [16613.137381] [drm:i915_do_wait_request] *ERROR* i915_do_wait_request returns -11 (awaiting 1740115 at 1740113, next 1740116)
Feb 11 19:40:21 chaostya-laptop kernel: [16614.696036] [drm:i915_hangcheck_elapsed] *ERROR* Hangcheck timer elapsed... GPU hung
Feb 11 19:40:21 chaostya-laptop kernel: [16614.702424] [drm:i915_reset] *ERROR* Failed to reset chip.
Feb 11 19:41:05 chaostya-laptop kernel: [16659.156805] [drm:drm_mode_getfb] *ERROR* invalid framebuffer id
Comment 1 Ko Mi 2011-02-11 18:20:46 UTC
Created attachment 47362 [details]
Comment 2 Ko Mi 2011-02-11 18:23:04 UTC
Created attachment 47372 [details]
Comment 3 Ko Mi 2011-02-14 17:26:02 UTC
This guys' https://bugzilla.kernel.org/show_bug.cgi?id=28642 dmesg and Xorg.log look similar
[ 29935.103] (WW) intel(0): intel_uxa_prepare_access: bo map failed: Input/output error
But he has a newer Dell.
Comment 4 Amit Shah 2011-02-28 11:57:33 UTC
I got this a couple of times.  First time was when mplayer was starting.  Second time was when totem was starting.  Both times, I get the plymouth splash screen shown.  I can then switch to a tty using ctrl+alt+f2, etc.

Happened with vmlinuz-2.6.38-0.rc6.git2.1.fc15.x86_64 recently (which is a Fedora kernel based on the rc6+ snapshot).

dmesg info:

[95566.456090] [drm:i915_hangcheck_elapsed] *ERROR* Hangcheck timer elapsed... GPU hung
[95566.460453] [drm:i915_do_wait_request] *ERROR* i915_do_wait_request returns -11 (awaiting 276360 at 276359, next 276363)
[95568.716017] [drm:i915_hangcheck_elapsed] *ERROR* Hangcheck timer elapsed... GPU hung
[95569.225052] [drm:i915_reset] *ERROR* Failed to reset chip.

00:02.0 VGA compatible controller: Intel Corporation Mobile 4 Series Chipset Integrated Graphics Controller (rev 07) (prog-if 00 [VGA controller])
	Subsystem: Lenovo Device 20e4
	Flags: bus master, fast devsel, latency 0, IRQ 47
	Memory at f2000000 (64-bit, non-prefetchable) [size=4M]
	Memory at d0000000 (64-bit, prefetchable) [size=256M]
	I/O ports at 1800 [size=8]
	Expansion ROM at <unassigned> [disabled]
	Capabilities: [90] MSI: Enable+ Count=1/1 Maskable- 64bit-
	Capabilities: [d0] Power Management version 3
	Kernel driver in use: i915
	Kernel modules: i915
Comment 5 Chris Wilson 2011-03-10 08:34:45 UTC
Amit and Ko, you both have userspace, though different, bugs.