Bug 88119

Summary: [gen4] GPU Hang
Product: DRI Reporter: Kirill Elagin <kirelagin>
Component: DRM/IntelAssignee: Intel GFX Bugs mailing list <intel-gfx-bugs>
Status: CLOSED WORKSFORME QA Contact: Intel GFX Bugs mailing list <intel-gfx-bugs>
Severity: normal    
Priority: medium CC: intel-gfx-bugs
Version: unspecified   
Hardware: x86-64 (AMD64)   
OS: Linux (All)   
Whiteboard:
i915 platform: GM45 i915 features: GPU hang
Attachments:
Description Flags
GPU crash dump
none
Initial X log
none
X log after restart none

Description Kirill Elagin 2015-01-06 17:24:06 UTC 
Created attachment 111861 [details]
GPU crash dump

This is an Acer 4810TG laptop with both intel and radeon videocards, but the discrete one was (hopefully, as I was not using it) DynOff.
I was doing nothing special: Konsole, Firefox, Google Chrome, Zathura running; I was about to launch Okular to print something, but I don’t remember whether I actually launched it when X restarted.

I think I never had such an issue before upgrading the kernel (but also just recently I started using a second screen attached via VGA).


dmesg:

~~~~
Jan 06 19:59:22 kirNote kernel: [drm] stuck on render ring
Jan 06 19:59:22 kirNote kernel: [drm] GPU HANG: ecode 0:0xab0f77f9, in X [2708], reason: Ring hung, action: reset
Jan 06 19:59:22 kirNote kernel: [drm] GPU hangs can indicate a bug anywhere in the entire gfx stack, including userspace.
Jan 06 19:59:22 kirNote kernel: [drm] Please file a _new_ bug report on bugs.freedesktop.org against DRI -> DRM/Intel
Jan 06 19:59:22 kirNote kernel: [drm] drm/i915 developers can then reassign to the right component if it's not a kernel issue.
Jan 06 19:59:22 kirNote kernel: [drm] The gpu crash dump is required to analyze gpu hangs, so please always attach it.
Jan 06 19:59:22 kirNote kernel: [drm] GPU crash dump saved to /sys/class/drm/card1/error
Jan 06 19:59:23 kirNote kernel: [drm] PCIE GART of 1024M enabled (table at 0x000000000025D000).
Jan 06 19:59:23 kirNote kernel: radeon 0000:01:00.0: WB enabled
Jan 06 19:59:23 kirNote kernel: radeon 0000:01:00.0: fence driver on ring 0 use gpu addr 0x0000000020000c00 and cpu addr 0xffff8800b3c10c00
Jan 06 19:59:23 kirNote kernel: radeon 0000:01:00.0: fence driver on ring 3 use gpu addr 0x0000000020000c0c and cpu addr 0xffff8800b3c10c0c
Jan 06 19:59:23 kirNote kernel: radeon 0000:01:00.0: fence driver on ring 5 use gpu addr 0x000000000005c598 and cpu addr 0xffffc9000451c598
Jan 06 19:59:23 kirNote kernel: [drm] ring test on 0 succeeded in 1 usecs
Jan 06 19:59:23 kirNote kernel: [drm] ring test on 3 succeeded in 3 usecs
Jan 06 19:59:23 kirNote kernel: [drm] ring test on 5 succeeded in 1 usecs
Jan 06 19:59:23 kirNote kernel: [drm] UVD initialized successfully.
Jan 06 19:59:23 kirNote kernel: [drm] ib test on ring 0 succeeded in 0 usecs
Jan 06 19:59:23 kirNote kernel: [drm] ib test on ring 3 succeeded in 0 usecs
Jan 06 19:59:23 kirNote kernel: [drm] ib test on ring 5 succeeded
Jan 06 19:59:30 kirNote kernel: [drm] stuck on render ring
Jan 06 19:59:30 kirNote kernel: [drm] GPU HANG: ecode 0:0xfdffffff, in X [2708], reason: Ring hung, action: reset
Jan 06 19:59:30 kirNote kernel: [drm:i915_context_is_banned] *ERROR* gpu hanging too fast, banning!
Jan 06 19:59:31 kirNote kernel: [drm] PCIE GART of 1024M enabled (table at 0x000000000025D000).
Jan 06 19:59:31 kirNote kernel: radeon 0000:01:00.0: WB enabled
Jan 06 19:59:31 kirNote kernel: radeon 0000:01:00.0: fence driver on ring 0 use gpu addr 0x0000000020000c00 and cpu addr 0xffff8800b3c10c00
Jan 06 19:59:31 kirNote kernel: radeon 0000:01:00.0: fence driver on ring 3 use gpu addr 0x0000000020000c0c and cpu addr 0xffff8800b3c10c0c
Jan 06 19:59:31 kirNote kernel: radeon 0000:01:00.0: fence driver on ring 5 use gpu addr 0x000000000005c598 and cpu addr 0xffffc9000451c598
Jan 06 19:59:31 kirNote kernel: [drm] ring test on 0 succeeded in 1 usecs
Jan 06 19:59:31 kirNote kernel: [drm] ring test on 3 succeeded in 3 usecs
Jan 06 19:59:32 kirNote kernel: [drm] ring test on 5 succeeded in 1 usecs
Jan 06 19:59:32 kirNote kernel: [drm] UVD initialized successfully.
Jan 06 19:59:32 kirNote kernel: [drm] ib test on ring 0 succeeded in 0 usecs
Jan 06 19:59:32 kirNote kernel: [drm] ib test on ring 3 succeeded in 0 usecs
Jan 06 19:59:32 kirNote kernel: [drm] ib test on ring 5 succeeded
Jan 06 19:59:34 kirNote kernel: audit: type=1701 audit(1420563574.952:405): auid=4294967295 uid=0 gid=0 ses=4294967295 pid=2708 comm="X" exe="/usr/bin/Xorg" sig=6
Jan 06 19:59:35 kirNote kernel: plugin-containe[18920]: segfault at 7fa2ff8450f8 ip 00007fa3009e79ff sp 00007fa2f8857bb0 error 4 in libflashplayer.so[7fa30039d000+1188000]
Jan 06 19:59:35 kirNote kernel: audit: type=1701 audit(1420563575.239:406): auid=1000 uid=1000 gid=1000 ses=1 pid=18920 comm="plugin-containe" exe="/usr/lib64/firefox/plugin-container" sig=11
Jan 06 19:59:49 kirNote kernel: audit: type=1006 audit(1420563589.477:407): pid=18996 uid=0 old-auid=4294967295 auid=1000 old-ses=4294967295 ses=2 res=1
Jan 06 20:04:36 kirNote kernel: [drm:i9xx_check_fifo_underruns] *ERROR* pipe A underrun
Jan 06 20:04:36 kirNote kernel: [drm:i965_irq_handler] *ERROR* pipe A underrun
~~~~


~~~~
# X -version

X.Org X Server 1.15.0
Release Date: 2013-12-27
X Protocol Version 11, Revision 0
Build Operating System: Linux 3.12.21-gentoo-r1kirNote x86_64 Gentoo
Current Operating System: Linux kirNote 3.17.7-gentookirNote #1 SMP Sat Jan 3 02:06:17 MSK 2015 x86_64
Kernel command line: BOOT_IMAGE=/kernel-3.17.7-gentoo root=UUID=b075b01b-2b39-45eb-8208-dcbbdcb1c0e3 ro
Build Date: 14 December 2014  09:47:53AM
 
Current version of pixman: 0.32.4
        Before reporting problems, check http://wiki.x.org
        to make sure that you have the latest version.
~~~~
Comment 1 Chris Wilson 2015-01-06 20:58:30 UTC 
It all looked fairly benign, then I realised it was gen4, where even a simple multi-vertex PRIMITIVE could crash. It looks like you have an old xf86-video-intel, could you please attach the Xorg.0.log? (Especially the Xorg.0.log.old after a hang + restart)
Comment 2 Kirill Elagin 2015-01-06 22:33:08 UTC 
Created attachment 111870 [details]
Initial X log
Comment 3 Kirill Elagin 2015-01-06 22:35:43 UTC 
Created attachment 111871 [details]
X log after restart

Yes, xf86-video-intel is somewhat old, as is xorg-server itself: both are the latest stable on Gentoo.
Comment 4 Jani Saarinen 2016-12-09 11:22:56 UTC 
Really old issue.
Is this issue still seen with latest kernel?
Comment 5 Jani Saarinen 2017-03-08 08:20:39 UTC 
No feedback in many months, closing. If issue still seen, please re-open.

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.