Bug 92710 - GPU HANG: ecode 7:0:0x85df9ff8, in chrome [9005], reason: Ring hung, action: reset
Summary: GPU HANG: ecode 7:0:0x85df9ff8, in chrome [9005], reason: Ring hung, action: ...
Status: RESOLVED WORKSFORME
Alias: None
Product: Mesa
Classification: Unclassified
Component: Drivers/DRI/i915 (show other bugs)
Version: unspecified
Hardware: Other All
: medium normal
Assignee: Ian Romanick
QA Contact:
URL:
Whiteboard:
Keywords:
: 94742 (view as bug list)
Depends on:
Blocks:
 
Reported: 2015-10-28 15:31 UTC by nicolas.laplante
Modified: 2016-09-26 06:20 UTC (History)
2 users (show)

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


Attachments
GPU crash dump (413.18 KB, application/x-gzip)
2015-10-28 15:32 UTC, nicolas.laplante
Details

Description nicolas.laplante 2015-10-28 15:31:25 UTC
Using chrome, whole desktop froze for a couple seconds and then resumed.

Dmesg: stuck on render ring

GPU crash dump attached.
Comment 1 nicolas.laplante 2015-10-28 15:32:14 UTC
Created attachment 119258 [details]
GPU crash dump
Comment 2 yann 2016-05-20 09:39:41 UTC
*** Bug 94742 has been marked as a duplicate of this bug. ***
Comment 3 yann 2016-09-20 16:30:18 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.

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

Kernel: 4.2.3-300.fc23.x86_64
Platform: Haswell (pci id: 0x0416)
Mesa: [Please confirm your mesa version]

From this error dump, hung is happening in render ring batch with active head at 0x7f903760, with 0x7a000003 (PIPE_CONTROL) as IPEHR.

We note also:
  ERROR: 0x00000105
    TLB page fault error (GTT entry not valid)
    Invalid page directory entry error
    Cacheline containing a PD was marked as invalid

Batch extract (around 0x7f903760):

0x7f903730:      0x7b000005: 3DPRIMITIVE:
0x7f903734:      0x00000104:    tri list random
0x7f903738:      0x00000006:    vertex count
0x7f90373c:      0x00000000:    start vertex
0x7f903740:      0x00000001:    instance count
0x7f903744:      0x00000000:    start instance
0x7f903748:      0x00000000:    index bias
0x7f90374c:      0x7a000003: PIPE_CONTROL
0x7f903750:      0x00101c11:    no write, cs stall, render target cache flush, instruction cache invalidate, texture cache invalidate, vf fetch invalidate, depth cache flush,
0x7f903754:      0x00000000:    destination address
0x7f903758:      0x00000000:    immediate dword low
0x7f90375c:      0x00000000:    immediate dword high
0x7f903760:      0x780e0000: 3DSTATE_CC_STATE_POINTERS
0x7f903764:      0x00007d81:    pointer to COLOR_CALC_STATE at 0x00007d80 (changed)
0x7f903768:      0x7a000003: PIPE_CONTROL
0x7f90376c:      0x00101000:    no write, cs stall, render target cache flush,
0x7f903770:      0x00000000:    destination address
0x7f903774:      0x00000000:    immediate dword low
0x7f903778:      0x00000000:    immediate dword high
0x7f90377c:      0x05000000: MI_BATCH_BUFFER_END
Comment 4 Axel Fischer 2016-09-26 01:36:13 UTC
The problem hasn't reoccurred since I updated kernel and mesa.
Comment 5 Jani Nikula 2016-09-26 06:20:16 UTC
(In reply to Axel Fischer from comment #4)
> The problem hasn't reoccurred since I updated kernel and mesa.

Thanks for the update.


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.