Bug 92697 - GPU HANG: ecode 9:0:0x85dffffb while playing XCOM: enemy unknown on Skylake-S/HD Graphics 530
Summary: GPU HANG: ecode 9:0:0x85dffffb while playing XCOM: enemy unknown on Skylake-S...
Status: CLOSED WORKSFORME
Alias: None
Product: Mesa
Classification: Unclassified
Component: Drivers/DRI/i915 (show other bugs)
Version: 11.0
Hardware: x86-64 (AMD64) Linux (All)
: medium normal
Assignee: Ian Romanick
QA Contact:
URL:
Whiteboard:
Keywords:
Depends on:
Blocks:
 
Reported: 2015-10-27 20:23 UTC by dvlohp
Modified: 2016-10-05 10:33 UTC (History)
2 users (show)

See Also:
i915 platform: SKL
i915 features: GPU hang

Attachments
dmesg output (60.56 KB, text/plain)
2015-10-27 20:23 UTC, dvlohp 		Details
crash dump (17.86 KB, text/plain)
2015-10-27 20:24 UTC, dvlohp 		Details
View All

Description dvlohp 2015-10-27 20:23:56 UTC 
Created attachment 119227 [details]
dmesg output

When playing "XCOM: enemy unknown, the game sometime freezes for some seconds or even crashes.

dmesg shows [drm] GPU HANG: ecode 9:0:0x85dffffb, in RenderingThread [1709], reason: Ring hung, action: reset.

I tried with latest drm-intel-nightly from 27.10.2015 (commit 34d1da7d864295c6411788d84b44567f029defd6)

Hardware is an i5-6500 and an Asus H170 Pro Gaming board.
Resolution 1920x1080, the screen is connected via HDMI.
I run debian/unstable with mesa 11.0.4 (released this weekend).

I'll attach a complete dmesg output and GPU crash dump.
Comment 1 dvlohp 2015-10-27 20:24:35 UTC 
Created attachment 119228 [details]
crash dump
Comment 2 Dirk Podolak 2015-11-23 06:24:07 UTC 
I got a very similar issue, sometimes every minute, sometimes every hour. Seems to be worse when doing things that "heavily" involve GPU like dragging a window around on the screen. Seems that using chrome triggers it more often.

I also got the Asus MoBo and an Intel i5-6600 using internal graphics (hd530) with default i915 driver on Suse Leap 42.1, OpenGL is 3.0 Mesa 11.0.2. I use a 4k monitor connected via HDMI.

dmesg:
[ 1991.626832] [drm] stuck on render ring
[ 1991.627178] [drm] GPU HANG: ecode 9:0:0x85dffffb, in Xorg [19240], reason: Ring hung, action: reset
[ 1991.629384] drm/i915: Resetting chip after gpu hang
[ 1993.613889] [drm] RC6 on
Comment 3 yann 2016-09-20 17:04:48 UTC 
There were workaround for SKL and improvements pushed in kernel and Mesa that will benefit to your system, so please re-test with latest kernel & Mesa to see if this issue is still occurring.

In parallel, assigning to Mesa product (please let me know if I am mistaken with this GPU Hang).

Kernel: 4.3.0-rc7+
Platform: Skylake (pci id: 0x1912)
Mesa: 11.0.4

From this error dump, hung is happening in render ring batch with active head at 0xe375ac34, with 0x7a000004 (PIPE_CONTROL) as IPEHR.

Batch extract (around 0xe375ac34):

0xe375abe8:      0x7b000005: 3DPRIMITIVE: fail sequential
0xe375abec:      0x00000104:    vertex count
0xe375abf0:      0x000003c0:    start vertex
0xe375abf4:      0x00000000:    instance count
0xe375abf8:      0x00000001:    start instance
0xe375abfc:      0x00000000:    index bias
0xe375ac00:      0x00000000: MI_NOOP
Bad count in PIPE_CONTROL
0xe375ac04:      0x7a000004: PIPE_CONTROL: no write, no depth stall, no RC write flush, no inst flush
0xe375ac08:      0x00000000:    destination address
0xe375ac0c:      0x00000000:    immediate dword low
0xe375ac10:      0x00000000:    immediate dword high
Bad count in PIPE_CONTROL
0xe375ac1c:      0x7a000004: PIPE_CONTROL: no write, no depth stall, no RC write flush, no inst flush
0xe375ac20:      0x00101c11:    destination address
0xe375ac24:      0x00000000:    immediate dword low
0xe375ac28:      0x00000000:    immediate dword high
0xe375ac34:      0x78210000: 3D UNKNOWN: 3d_965 opcode = 0x7821
0xe375ac38:      0x00003980: MI_NOOP
Comment 4 dvlohp 2016-10-04 19:31:23 UTC 
Unfortunately I mislaid the save game exhibiting this problem.
I tried to reproduce it with other save games but failed. So you can close the issue if you like.
Comment 5 yann 2016-10-05 10:33:32 UTC 
(In reply to dvlohp from comment #4)
> Unfortunately I mislaid the save game exhibiting this problem.
> I tried to reproduce it with other save games but failed. So you can close
> the issue if you like.

Closing then. Please reopen the bug if it is occurring again with latest kernel & Mesa version.

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.