Bug 95267

Summary:	[SKL] GPU HANG: ecode 9:0:0x86dfbff9 in chromium [1493], reason: Ring hung, action: reset
Product:	Mesa	Reporter:	Daniel Franke <dfoxfranke>
Component:	Drivers/DRI/i965	Assignee:	Ian Romanick <idr>
Status:	RESOLVED INVALID	QA Contact:
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:	/sys/class/drm/card0/error dmesg

Description Daniel Franke 2016-05-04 19:51:05 UTC

Created attachment 123466 [details]
/sys/class/drm/card0/error

On my new Dell XPS 13 Developer Edition, i7-6560U (stepping 3) chipset, I get frequent hangs (usually within the first minute) in almost any application that utilizes OpenGL, notably including chromium until I disable hardware acceleration. I'm running the latest NixOS, with the kernel switched to 4.6.0-rc4 in order to get the fix for bug #94102. Error and dmesg dumps attached.

Comment 1 Daniel Franke 2016-05-04 19:51:26 UTC

Created attachment 123467 [details]
dmesg

Comment 2 Daniel Franke 2016-05-04 20:04:40 UTC

Setting enable_rc6=0 appears to work around the problem.

Comment 3 Daniel Franke 2016-05-16 17:28:39 UTC

I just updated from kernel 4.6.0-rc4 to 4.6.0-rc6. This appears to have solved the crashes (or at least made them far less frequent), even with rc6 enabled. This leads me to strongly suspect that this report is a duplicate of #94161, but I'll leave it open and let the devs mark it as such if they see fit.

Comment 4 yann 2016-09-02 16:04:03 UTC

Assigning to Mesa product.

From this error dump, hung is happening in render ring batch with active head
at 0xfe8c37bc, with 0x79000002 (3DSTATE_DRAWING_RECTANGLE) as IPEHR.

Batch extract (around 0xfe8c37bc):

0xfe8c3794:      0x000000d2: MI_NOOP
Bad length 5 in 3DSTATE_STENCIL_BUFFER, expected 3-3
0xfe8c3798:      0x78060003: 3DSTATE_STENCIL_BUFFER
0xfe8c379c:      0x8100087f:    dword 1
0xfe8c37a0:      0xff9e7000:    dword 2
0xfe8c37a4:      0x00000000:    dword 3
0xfe8c37a8:      0x00000000:    dword 4
0xfe8c37ac:      0x78040001: 3DSTATE_CLEAR_PARAMS
0xfe8c37b0:      0x3f800000:    dword 1
0xfe8c37b4:      0x00000001:    dword 2
0xfe8c37b8:      0x79000002: 3DSTATE_DRAWING_RECTANGLE
0xfe8c37bc:      0x00000000:    top left: 0,0
0xfe8c37c0:      0x02280415:    bottom right: 1045,552
0xfe8c37c4:      0x00000000:    origin: 0,0
0xfe8c37c8:      0x784a0000: 3D UNKNOWN: 3d_965 opcode = 0x784a
0xfe8c37cc:      0x00000000: MI_NOOP

Comment 5 yann 2016-11-04 15:18:00 UTC

Please test a new version of Mesa (12 or 13) and mark as REOPENED
if you can reproduce and RESOLVED/* if you cannot reproduce.

Comment 6 Annie 2017-02-10 22:38:46 UTC

Dear Reporter,

This Mesa bug has been in the "NEEDINFO" status for over 60 days. I am closing this bug based on lack of response but feel free to reopen if resolution is still needed. Please ensure you're supplying the correct information as requested.

Thank you.

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.