[Bug 111597] [CI][RESUME] igt@* - fail - Failed assertion: !"GPU hung"

Wed Sep 11 10:31:49 UTC 2019

https://bugs.freedesktop.org/show_bug.cgi?id=111597

--- Comment #5 from Stanislav Lisovskiy <stanislav.lisovskiy at intel.com> ---
Doing now some testing with tgl, when I do submit multiple gpgpu_fill commands
I constantly get this:

(kms_plane_stress:3092) gpu_cmds-CRITICAL: Test assertion failure function
gen7_render_flush, file ../lib/gpu_cmds.c:36:
(kms_plane_stress:3092) gpu_cmds-CRITICAL: Failed assertion: ret == 0
(kms_plane_stress:3092) gpu_cmds-CRITICAL: Last errno: 5, Input/output error
Pausing GPU thread 0 
Stack trace:
  #0 ../lib/igt_core.c:1694 __igt_fail_assert()
  #1 ../lib/gpu_cmds.c:36 gen7_render_flush()
  #2 ../lib/gpgpu_fill.c:356 gen12p1_gpgpu_fillfunc()
  #3 ../tests/kms_plane_stress.c:318 gpu_load()
  #4 /build/glibc-OTsEL5/glibc-2.27/nptl/pthread_create.c:463 start_thread()
  #5 ../sysdeps/unix/sysv/linux/x86_64/clone.S:97 __clone()

Which works quite fine with ICL and other platforms. In dmesg I have this:

[ 3108.643351] hangcheck rcs0
[ 3108.643420] hangcheck        Awake? 2
[ 3108.643428] hangcheck        Hangcheck: 6016 ms ago
[ 3108.643434] hangcheck        Reset count: 0 (global 0)
[ 3108.643440] hangcheck        Requests:
[ 3108.643628] hangcheck                active  1a:4*  prio=2 @ 7900ms:
kms_plane_stres[1347]
[ 3108.643689] hangcheck                ring->start:  0x00008000
[ 3108.643708] hangcheck                ring->head:   0x00000048
[ 3108.643724] hangcheck                ring->tail:   0x00003078
[ 3108.643733] hangcheck                ring->emit:   0x00003080
[ 3108.643738] hangcheck                ring->space:  0x00000f88
[ 3108.643745] hangcheck                ring->hwsp:   0xffff81c0
[ 3108.643753] hangcheck [head 0080, postfix 00c8, tail 0100, batch
0x00000000_007ea000]:
[ 3108.643820] hangcheck [0000] 7a000004 21144c1c fffff080 00000000 00000000
00000000 02800000 00000000
[ 3108.643832] hangcheck [0020] 10400002 ffff81c0 00000000 00000003 04000001
18800101 007ea000 00000000
[ 3108.643841] hangcheck [0040] 04000000 00000000 7a000004 111050a1 ffff81c0
00000000 00000004 00000000
[ 3108.643849] hangcheck [0060] 01000000 04000001 0e40c002 00000000 ffffe0c8
00000000 02800000 00000000
[ 3108.644037] hangcheck        MMIO base:  0x00002000
[ 3108.644085] hangcheck        RING_START: 0x00008000
[ 3108.644098] hangcheck        RING_HEAD:  0x000000c0
[ 3108.644110] hangcheck        RING_TAIL:  0x00003078
[ 3108.644139] hangcheck        RING_CTL:   0x00003001
[ 3108.644158] hangcheck        RING_MODE:  0x00000000
[ 3108.644173] hangcheck        RING_IMR: 00000000
[ 3108.644198] hangcheck        ACTHD:  0x00000000_007ea884
[ 3108.644223] hangcheck        BBADDR: 0x00000000_007ea885
[ 3108.644246] hangcheck        DMA_FADDR: 0x00000000_007eaa80
[ 3108.644257] hangcheck        IPEIR: 0x00000000
[ 3108.644267] hangcheck        IPEHR: 0x25014100
[ 3108.644286] hangcheck        Execlist status: 0x00002098 00000040, entries
12
[ 3108.644295] hangcheck        Execlist CSB read 8, write 8, tasklet queued?
no (enabled)
[ 3108.644318] hangcheck                Active[0: ring:{start:00008000,
hwsp:ffff81c0, seqno:00000003}, rq:  1a:c2  prio=2 @ 7748ms:
kms_plane_stres[1347]
[ 3108.644343] hangcheck                E  1a:4*  prio=2 @ 7901ms:
kms_plane_stres[1347]
[ 3108.644352] hangcheck                E  1a:6  prio=2 @ 7900ms:
kms_plane_stres[1347]
[ 3108.644360] hangcheck                E  1a:8  prio=2 @ 7899ms:
kms_plane_stres[1347]
[ 3108.644368] hangcheck                E  1a:a  prio=2 @ 7898ms:
kms_plane_stres[1347]
[ 3108.644377] hangcheck                E  1a:c  prio=2 @ 7898ms:
kms_plane_stres[1347]
[ 3108.644384] hangcheck                E  1a:e  prio=2 @ 7897ms:
kms_plane_stres[1347]
[ 3108.644392] hangcheck                E  1a:10  prio=2 @ 7896ms:
kms_plane_stres[1347]
[ 3108.644442] hangcheck                ...skipping 88 executing requests...
[ 3108.644450] hangcheck                E  1a:c2  prio=2 @ 7748ms:
kms_plane_stres[1347]
[ 3108.644457] hangcheck HWSP:
[ 3108.644470] hangcheck [0000] 00000000 00000000 00000000 00000000 00000000
00000000 00000000 00000000
[ 3108.644475] hangcheck *
[ 3108.644486] hangcheck [0040] 00010001 00010005 00010001 00010005 00010001
00010005 00010001 00010005
[ 3108.644491] hangcheck *
[ 3108.644499] hangcheck [00a0] 00000000 00000000 00000000 00000000 00000000
00000000 00000000 00000008
[ 3108.644508] hangcheck [00c0] 00000000 00000000 00000000 00000000 00000000
00000000 00000000 00000000
[ 3108.644513] hangcheck *
[ 3108.644563] hangcheck Idle? no
[ 3108.644578] hangcheck Signals:
[ 3108.644676] hangcheck        [1a:44] @ 7846ms
[ 3108.651414] i915 0000:00:02.0: GPU HANG: ecode 12:1:0xdadebeff, in
kms_plane_stres [1347], hang on rcs0
[ 3108.651930] GPU hangs can indicate a bug anywhere in the entire gfx stack,
including userspace.
[ 3108.651945] Please file a _new_ bug report on bugs.freedesktop.org against
DRI -> DRM/Intel
[ 3108.651953] drm/i915 developers can then reassign to the right component if
it's not a kernel issue.
[ 3108.651958] The GPU crash dump is required to analyze GPU hangs, so please
always attach it.
[ 3108.651965] GPU crash dump saved to /sys/class/drm/card0/error

kms_plane_stress is not yet in IGT however, I think there is definitely a bug,
however I don't have any clue what gpu hang might mean.

-- 
You are receiving this mail because:
You are the QA Contact for the bug.
You are on the CC list for the bug.
You are the assignee for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.freedesktop.org/archives/intel-gfx-bugs/attachments/20190911/01dd39c6/attachment-0001.html>