Bug 93049

Summary:

Skylake machines randomly hang

Product:

DRI

Reporter:

mike

Component:

DRM/Intel

Assignee:

Olivier Berthier <olivierx.berthier>

Status:

CLOSED WORKSFORME

QA Contact:

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

Severity:

major

Priority:

medium

CC:

daniel, intel-gfx-bugs, mike

Version:

XOrg git

Hardware:

x86-64 (AMD64)

OS:

Linux (All)

Whiteboard:

i915 platform:

SKL

i915 features:

GPU hang

Attachments:

Description	Flags
dmesg trace	none

Description mike 2015-11-20 19:57:46 UTC

Intel Skylake graphics hangs randomly but seems to happen faster while running minecraft, i presume because the graphics is being heavily used. The whole system becomes is unresponsive. This happens on all 3 of my skylake systems (2 laptops and 1 desktop).

Nov 09 03:02:05 laptop kernel: [drm] stuck on render ring
Nov 09 03:02:05 laptop kernel: [drm] GPU HANG: ecode 9:0:0x84df9ffc, in java [2403], reason: Ring hung, action: reset
Nov 09 03:02:05 laptop kernel: [drm] GPU hangs can indicate a bug anywhere in the entire gfx stack, including userspace.
Nov 09 03:02:05 laptop kernel: [drm] Please file a _new_ bug report on bugs.freedesktop.org against DRI -> DRM/Intel
Nov 09 03:02:05 laptop kernel: [drm] drm/i915 developers can then reassign to the right component if it's not a kernel issue.
Nov 09 03:02:05 laptop kernel: [drm] The gpu crash dump is required to analyze gpu hangs, so please always attach it.
Nov 09 03:02:05 laptop kernel: [drm] GPU crash dump saved to /sys/class/drm/card0/error
Nov 09 03:02:05 laptop kernel: drm/i915: Resetting chip after gpu hang
Nov 09 03:02:07 laptop kernel: [drm] RC6 on
Nov 09 03:02:23 laptop kernel: [drm] stuck on render ring
Nov 09 03:02:23 laptop kernel: [drm] GPU HANG: ecode 9:0:0x84df7cfc, in java [2403], reason: Ring hung, action: reset
Nov 09 03:02:23 laptop kernel: drm/i915: Resetting chip after gpu hang
Nov 09 03:02:25 laptop kernel: [drm] RC6 on


This happens on both ubuntu and arch linux (xf86-video-intel 1:2.99.917+478+gdf72bc5-2)

Comment 1 Jani Nikula 2015-11-23 08:49:22 UTC

(In reply to mike from comment #0)
> Nov 09 03:02:05 laptop kernel: [drm] The gpu crash dump is required to
> analyze gpu hangs, so please always attach it.
> Nov 09 03:02:05 laptop kernel: [drm] GPU crash dump saved to
> /sys/class/drm/card0/error

Comment 2 mike 2015-11-24 02:55:29 UTC

(In reply to Jani Nikula from comment #1)
> (In reply to mike from comment #0)
> > Nov 09 03:02:05 laptop kernel: [drm] The gpu crash dump is required to
> > analyze gpu hangs, so please always attach it.
> > Nov 09 03:02:05 laptop kernel: [drm] GPU crash dump saved to
> > /sys/class/drm/card0/error

That file is not created so I cannot attach it. Is there a reason why it would not be created or could it be somewhere else?

Comment 3 cprigent 2015-11-24 17:57:57 UTC

Bug scrub:
Hi Olivier,
Could you try to reproduce it.
Thanks

Comment 4 mike 2015-11-26 11:01:08 UTC

Created attachment 120138 [details]
dmesg trace

Comment 5 iDanoo 2015-12-08 05:46:00 UTC

I'm also having this issue, can help provide any other information required.

It happens completely intermittantly with no trigger that I can pick up. Usually within 15-60min of boot. 

I'm currently passing these kernel params on boot: i915.preliminary_hw_support=1 drm.debug=0 drm.vblankoffdelay=1 i915.semaphores=0 i915.modeset=1 i915.use_mmio_flip=1 i915.powersave=1 i915.enable_ips=1 i915.disable_power_well=1 i915.enable_hangcheck=1 i915.enable_cmd_parser=1 i915.fastboot=0 i915.enable_ppgtt=1 i915.reset=0 i915.lvds_use_ssc=0 i915.enable_psr=0 

Am now testing UXA accel method with Xorg to see if there's any improvement as someone stated on another site that it may help.

Comment 6 iDanoo 2015-12-08 06:08:55 UTC

Confirmed setting UXA accel method works - however performance drops as expected to an almost unusable state. 

------------------------------------------------------------
~ » sudo cat /usr/share/X11/xorg.conf.d/20-intel.conf                                                                                                                                                
Section "Device"
   Identifier  "Intel Graphics"
   Driver      "intel"
   Option      "AccelMethod"  "uxa"
EndSection
------------------------------------------------------------

Comment 7 iDanoo 2015-12-11 06:16:59 UTC

Appears fixed running 4.4.0-rc4 kernel. Currently up to 4 hours without a lock-up.

Comment 8 iDanoo 2015-12-13 01:13:23 UTC

Issue resolved running Kernel 4.4.0-rc4

Comment 9 cprigent 2016-02-22 16:56:15 UTC

So 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.