Bug 95068

Summary:

drm/i915: Resetting chip after gpu hang

Product:

DRI

Reporter:

Martial <mttm724>

Component:

DRM/Intel

Assignee:

Intel GFX Bugs mailing list <intel-gfx-bugs>

Status:

CLOSED DUPLICATE

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:

SKL

i915 features:

GPU hang

Attachments:

Description	Flags
GPU crash dump file	none
xrandr information	none

Description Martial 2016-04-22 14:25:27 UTC

Created attachment 123150 [details]
GPU crash dump file

Hi,

My GPU hang and resetting sometime.
Much better since the 4.4.0-18 kernel, but not still super stable.

uname -m =>
  x86_64

uname -r =>
  4.4.0-21-generic

lsb_release -a =>
  No LSB modules are available.
  Distributor ID:	Ubuntu
  Description:	Ubuntu 16.04 LTS
  Release:	16.04
  Codename:	xenial

Machine =>
  Dell XPS 13 9350

BIOS Information =>
  Vendor: Dell Inc.
  Version: 1.3.3
  Release Date: 03/01/2016

Error message into syslog file =>
Apr 22 13:59:55  kernel: [15194.457006] [drm] stuck on render ring
Apr 22 13:59:55  kernel: [15194.457966] [drm] GPU HANG: ecode 9:0:0x87f99ff9, in gnome-shell [2191], reason: Ring hung, action: reset
Apr 22 13:59:55  kernel: [15194.457973] [drm] GPU hangs can indicate a bug anywhere in the entire gfx stack, including userspace.
Apr 22 13:59:55  kernel: [15194.457978] [drm] Please file a _new_ bug report on bugs.freedesktop.org against DRI -> DRM/Intel
Apr 22 13:59:55  kernel: [15194.457982] [drm] drm/i915 developers can then reassign to the right component if it's not a kernel issue.
Apr 22 13:59:55  kernel: [15194.457987] [drm] The gpu crash dump is required to analyze gpu hangs, so please always attach it.
Apr 22 13:59:55  kernel: [15194.457991] [drm] GPU crash dump saved to /sys/class/drm/card0/error
Apr 22 13:59:55  kernel: [15194.460350] drm/i915: Resetting chip after gpu hang
Apr 22 13:59:57  kernel: [15196.457047] [drm] RC6 on
Apr 22 14:00:02  /usr/lib/gdm3/gdm-x-session[2023]: (EE) [mi] EQ overflowing.  Additional events will be discarded until existing events are processed.
Apr 22 14:00:02  /usr/lib/gdm3/gdm-x-session[2023]: (EE)
Apr 22 14:00:02  /usr/lib/gdm3/gdm-x-session[2023]: (EE) Backtrace:
Apr 22 14:00:02  /usr/lib/gdm3/gdm-x-session[2023]: (EE) 0: /usr/lib/xorg/Xorg (xorg_backtrace+0x4e) [0x558d094be50e]
Apr 22 14:00:02  /usr/lib/gdm3/gdm-x-session[2023]: (EE) 1: /usr/lib/xorg/Xorg (mieqEnqueue+0x253) [0x558d0949ffd3]
Apr 22 14:00:02  /usr/lib/gdm3/gdm-x-session[2023]: (EE) 2: /usr/lib/xorg/Xorg (QueuePointerEvents+0x52) [0x558d09378672]
Apr 22 14:00:02  /usr/lib/gdm3/gdm-x-session[2023]: (EE) 3: /usr/lib/xorg/modules/input/evdev_drv.so (0x7fba7d87b000+0x61f3) [0x7fba7d8811f3]
Apr 22 14:00:02  /usr/lib/gdm3/gdm-x-session[2023]: (EE) 4: /usr/lib/xorg/modules/input/evdev_drv.so (0x7fba7d87b000+0x6a5d) [0x7fba7d881a5d]
Apr 22 14:00:02  /usr/lib/gdm3/gdm-x-session[2023]: (EE) 5: /usr/lib/xorg/Xorg (0x558d0930c000+0x94248) [0x558d093a0248]
Apr 22 14:00:02  /usr/lib/gdm3/gdm-x-session[2023]: (EE) 6: /usr/lib/xorg/Xorg (0x558d0930c000+0xb9712) [0x558d093c5712]
Apr 22 14:00:02  /usr/lib/gdm3/gdm-x-session[2023]: (EE) 7: /lib/x86_64-linux-gnu/libc.so.6 (0x7fba84ac3000+0x354a0) [0x7fba84af84a0]
Apr 22 14:00:02  /usr/lib/gdm3/gdm-x-session[2023]: (EE) 8: /lib/x86_64-linux-gnu/libc.so.6 (ioctl+0x5) [0x7fba84bbf685]
Apr 22 14:00:02  /usr/lib/gdm3/gdm-x-session[2023]: (EE) 9: /usr/lib/xorg/modules/drivers/intel_drv.so (0x7fba80c97000+0x2d73e) [0x7fba80cc473e]
Apr 22 14:00:02  /usr/lib/gdm3/gdm-x-session[2023]: (EE) 10: /usr/lib/xorg/modules/drivers/intel_drv.so (0x7fba80c97000+0x63e64) [0x7fba80cfae64]
Apr 22 14:00:02  /usr/lib/gdm3/gdm-x-session[2023]: (EE) 11: /usr/lib/xorg/modules/drivers/intel_drv.so (0x7fba80c97000+0x580ac) [0x7fba80cef0ac]
Apr 22 14:00:02  /usr/lib/gdm3/gdm-x-session[2023]: (EE) 12: /usr/lib/xorg/modules/drivers/intel_drv.so (0x7fba80c97000+0x39def) [0x7fba80cd0def]
Apr 22 14:00:02  /usr/lib/gdm3/gdm-x-session[2023]: (EE) 13: /usr/lib/xorg/modules/drivers/intel_drv.so (0x7fba80c97000+0x39ff2) [0x7fba80cd0ff2]
Apr 22 14:00:02  /usr/lib/gdm3/gdm-x-session[2023]: (EE) 14: /usr/lib/xorg/Xorg (0x558d0930c000+0x13d101) [0x558d09449101]
Apr 22 14:00:02  /usr/lib/gdm3/gdm-x-session[2023]: (EE) 15: /usr/lib/xorg/Xorg (0x558d0930c000+0x4fb87) [0x558d0935bb87]
Apr 22 14:00:02  /usr/lib/gdm3/gdm-x-session[2023]: (EE) 16: /usr/lib/xorg/Xorg (0x558d0930c000+0x53bbf) [0x558d0935fbbf]
Apr 22 14:00:02  /usr/lib/gdm3/gdm-x-session[2023]: (EE) 17: /usr/lib/xorg/Xorg (0x558d0930c000+0x57c43) [0x558d09363c43]
Apr 22 14:00:02  /usr/lib/gdm3/gdm-x-session[2023]: (EE) 18: /lib/x86_64-linux-gnu/libc.so.6 (__libc_start_main+0xf0) [0x7fba84ae3830]
Apr 22 14:00:02  /usr/lib/gdm3/gdm-x-session[2023]: (EE) 19: /usr/lib/xorg/Xorg (_start+0x29) [0x558d0934df59]
Apr 22 14:00:02  /usr/lib/gdm3/gdm-x-session[2023]: (EE)
Apr 22 14:00:02  /usr/lib/gdm3/gdm-x-session[2023]: (EE) [mi] These backtraces from mieqEnqueue may point to a culprit higher up the stack.
Apr 22 14:00:02  /usr/lib/gdm3/gdm-x-session[2023]: (EE) [mi] mieq is *NOT* the cause.  It is a victim.
Apr 22 14:00:03  kernel: [15202.457207] [drm] stuck on render ring
Apr 22 14:00:03  kernel: [15202.458218] [drm] GPU HANG: ecode 9:0:0x87f99ff9, in gnome-shell [2191], reason: Ring hung, action: reset
Apr 22 14:00:03  kernel: [15202.461209] drm/i915: Resetting chip after gpu hang
Apr 22 14:00:03  /usr/lib/gdm3/gdm-x-session[2023]: [mi] Increasing EQ size to 1024 to prevent dropped events.
Apr 22 14:00:03  /usr/lib/gdm3/gdm-x-session[2023]: [mi] EQ processing has resumed after 85 dropped events.
Apr 22 14:00:03  /usr/lib/gdm3/gdm-x-session[2023]: [mi] This may be caused by a misbehaving driver monopolizing the server's resources.
Apr 22 14:00:05  kernel: [15204.457253] [drm] RC6 on
Apr 22 14:00:11  kernel: [15210.481450] [drm] stuck on render ring
Apr 22 14:00:11  kernel: [15210.481922] [drm] GPU HANG: ecode 9:0:0x85dfbfff, in chrome [5738], reason: Ring hung, action: reset
Apr 22 14:00:11  kernel: [15210.484181] drm/i915: Resetting chip after gpu hang
Apr 22 14:00:12  /usr/lib/gdm3/gdm-x-session[2023]: The XKEYBOARD keymap compiler (xkbcomp) reports:
Apr 22 14:00:12  /usr/lib/gdm3/gdm-x-session[2023]: > Warning:          Type "ONE_LEVEL" has 1 levels, but <RALT> has 2 symbols
Apr 22 14:00:12  /usr/lib/gdm3/gdm-x-session[2023]: >                   Ignoring extra symbols
Apr 22 14:00:12  /usr/lib/gdm3/gdm-x-session[2023]: Errors from xkbcomp are not fatal to the X server
Apr 22 14:00:12  /usr/lib/gdm3/gdm-x-session[2023]: The XKEYBOARD keymap compiler (xkbcomp) reports:
Apr 22 14:00:12  /usr/lib/gdm3/gdm-x-session[2023]: > Warning:          Type "ONE_LEVEL" has 1 levels, but <RALT> has 2 symbols
Apr 22 14:00:12  /usr/lib/gdm3/gdm-x-session[2023]: >                   Ignoring extra symbols
Apr 22 14:00:12  /usr/lib/gdm3/gdm-x-session[2023]: Errors from xkbcomp are not fatal to the X server

Comment 1 Martial 2016-04-22 14:28:05 UTC

Created attachment 123151 [details]
xrandr information

Comment 2 yann 2016-05-18 15:30:02 UTC

For both bugs:
- Hung happening on "render" ring, 
- Application is "gnome-shell",
- HW/PCI_ID is "0x1926", 
- Instruction/IPETH is "0x78260000"

*** This bug has been marked as a duplicate of bug 94101 ***

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.