Bug 93470 - [HSW] GPU HANG: ecode 7:0:0x85dfbff8, in chrome [2700], reason: Ring hung, action: reset
Summary: [HSW] GPU HANG: ecode 7:0:0x85dfbff8, in chrome [2700], reason: Ring hung, ac...
Status: RESOLVED INVALID
Alias: None
Product: Mesa
Classification: Unclassified
Component: Drivers/DRI/i965 (show other bugs)
Version: unspecified
Hardware: x86-64 (AMD64) Linux (All)
: medium normal
Assignee: Ian Romanick
QA Contact:
URL:
Whiteboard:
Keywords:
Depends on:
Blocks:
 
Reported: 2015-12-21 18:35 UTC by Henning Knut Skoglund
Modified: 2017-02-10 22:39 UTC (History)
1 user (show)

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


Attachments
cat /sys/class/drm/card0/error +dmesg + journal + xrandr (448.36 KB, text/plain)
2015-12-21 18:35 UTC, Henning Knut Skoglund
Details

Description Henning Knut Skoglund 2015-12-21 18:35:30 UTC
Created attachment 120629 [details]
cat /sys/class/drm/card0/error +dmesg + journal + xrandr

After resume from sleep I got GPU Hang on fedora rawhide in chrome. I don't know how to reproduce the issue, but it seems like its related to suspend (it has occured several times before after resume from suspend).

des. 21 19:12:02 satellite kernel: [drm] GPU HANG: ecode 7:0:0x85dfbff8, in chrome [2700], reason: Ring hung, action: reset
des. 21 19:12:02 satellite kernel: [drm] GPU hangs can indicate a bug anywhere in the entire gfx stack, including userspace.
des. 21 19:12:02 satellite kernel: [drm] Please file a _new_ bug report on bugs.freedesktop.org against DRI -> DRM/Intel
des. 21 19:12:02 satellite kernel: [drm] drm/i915 developers can then reassign to the right component if it's not a kernel issue.
des. 21 19:12:02 satellite kernel: [drm] The gpu crash dump is required to analyze gpu hangs, so please always attach it.
des. 21 19:12:02 satellite kernel: [drm] GPU crash dump saved to /sys/class/drm/card0/error
des. 21 19:12:02 satellite kernel: drm/i915: Resetting chip after gpu hang
des. 21 19:12:02 satellite google-chrome.desktop[2599]: [2700:2700:1221/191202:ERROR:gles2_cmd_decoder.cc(12198)] Onscreen context lost via ARB/EXT_robustness. Reset status = GL_GUILTY_CONTEXT_RESET_KHR
des. 21 19:12:02 satellite google-chrome.desktop[2599]: [2700:2700:1221/191202:ERROR:gles2_cmd_decoder.cc(3581)]   GLES2DecoderImpl: Context reset detected after MakeCurrent.
des. 21 19:12:02 satellite google-chrome.desktop[2599]: [2700:2700:1221/191202:ERROR:gpu_command_buffer_stub.cc(1186)] Exiting GPU process because some drivers cannot recover from problems.
des. 21 19:12:02 satellite google-chrome.desktop[2599]: [2599:2599:1221/191202:ERROR:gpu_process_transport_factory.cc(621)] Lost UI shared context.
Comment 1 cprigent 2016-03-04 15:24:42 UTC
Henning,
I see only garbage in the attachment. If possible please update it.
Comment 2 cprigent 2016-03-04 16:01:08 UTC
Sorry, I just understood attachment is a tar.gz.
Reassigned to the mailing list.
Comment 3 yann 2016-09-20 09:32:16 UTC
There were 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.

Kernel: 4.4.0-0.rc5.git3.1.fc24.x86_64
Platform: HSW-ULT (pci id: 0x0a16)
Mesa: [Please confirm your mesa version]

In the meantime, assigning to Mesa product.

From this error dump, hung is happening in render ring batch with active head at 0x79968574, with 0x7a000003 (PIPE_CONTROL) as IPEHR.

Batch extract (around 0x79968574):

0x79968544:      0x7b000005: 3DPRIMITIVE:
0x79968548:      0x00000104:    tri list random
0x7996854c:      0x00000006:    vertex count
0x79968550:      0x00000000:    start vertex
0x79968554:      0x00000001:    instance count
0x79968558:      0x00000000:    start instance
0x7996855c:      0x00000000:    index bias
0x79968560:      0x7a000003: PIPE_CONTROL
0x79968564:      0x00101c11:    no write, cs stall, render target cache flush, instruction cache invalidate, texture cache invalidate, vf fetch invalidate, depth cache flush,
0x79968568:      0x00000000:    destination address
0x7996856c:      0x00000000:    immediate dword low
0x79968570:      0x00000000:    immediate dword high
0x79968574:      0x780e0000: 3DSTATE_CC_STATE_POINTERS
0x79968578:      0x00007b01:    pointer to COLOR_CALC_STATE at 0x00007b00 (changed)
0x7996857c:      0x7a000003: PIPE_CONTROL
0x79968580:      0x00101000:    no write, cs stall, render target cache flush,
0x79968584:      0x00000000:    destination address
0x79968588:      0x00000000:    immediate dword low
0x7996858c:      0x00000000:    immediate dword high
0x79968590:      0x05000000: MI_BATCH_BUFFER_END
Comment 4 yann 2016-11-04 15:42:04 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 5 Annie 2017-02-10 22:39:06 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.