[Bug 111895] [CI][BAT] igt at i915_selftest@live_execlists/live_unlite_switch - dmesg-fail: timed out, cancelling all further testing

Thu Oct 3 15:59:43 UTC 2019

https://bugs.freedesktop.org/show_bug.cgi?id=111895

--- Comment #5 from Chris Wilson <chris at chris-wilson.co.uk> ---
In another example,

<0> [629.616137] i915_sel-4301    0.... 618083258us : __engine_unpark: vcs0
<0> [629.616327] i915_sel-4301    0.... 618083271us : __gt_unpark: 
<0> [629.616530] i915_sel-4301    0.... 618084676us : __i915_request_commit:
vcs0 fence e80d:2
<0> [629.616734] i915_sel-4301    0d..1 618084704us : __i915_request_submit:
vcs0 fence e80d:2, current 0
<0> [629.616928] i915_sel-4301    0d..1 618084711us :
__execlists_submission_tasklet: vcs0: queue_priority_hint:-2147483648,
submit:yes
<0> [629.617121] i915_sel-4301    0d..1 618084716us : trace_ports: vcs0: submit
{ e80d:2, 0:0 }
<0> [629.617325] i915_sel-4301    0.... 618084786us : __i915_request_commit:
vcs0 fence e80e:2
<0> [629.617518]    dmesg-922     1..s. 618084787us : process_csb: vcs0 cs-irq
head=3, tail=4
<0> [629.617711]    dmesg-922     1..s. 618084792us : process_csb: vcs0 csb[4]:
status=0x00000001:0x00000000
<0> [629.617905]    dmesg-922     1..s. 618084798us : trace_ports: vcs0:
promote { e80d:2*, 0:0 }
<0> [629.618109]    dmesg-922     1d.s1 618084814us : __i915_request_submit:
vcs0 fence e80e:2, current 0
<0> [629.618302]    dmesg-922     1d.s1 618084823us :
__execlists_submission_tasklet: vcs0: queue_priority_hint:-2147483648,
submit:yes
<0> [629.618496]    dmesg-922     1d.s1 618084829us : trace_ports: vcs0: submit
{ e80d:2*, e80e:2 }
<0> [629.618700] i915_sel-4301    0.... 618084861us : __i915_request_commit:
vcs0 fence e80d:4
<0> [629.618892]    dmesg-922     1..s. 618084907us : process_csb: vcs0 cs-irq
head=4, tail=5
<0> [629.619085]    dmesg-922     1..s. 618084908us : process_csb: vcs0 csb[5]:
status=0x00008002:0x00000001
<0> [629.619279]    dmesg-922     1..s. 618084912us : trace_ports: vcs0:
preempted { e80d:2*, 0:0 }
<0> [629.619472]    dmesg-922     1..s. 618084916us : trace_ports: vcs0:
promote { e80d:2*, e80e:2 }
<0> [629.619666]    dmesg-922     1d.s1 618084926us :
__execlists_submission_tasklet: vcs0: expired last=e80d:2, prio=3, hint=3
<0> [629.619869]    dmesg-922     1d.s1 618084930us : __i915_request_unsubmit:
vcs0 fence e80e:2, current 0
<0> [629.620073]    dmesg-922     1d.s1 618084936us : __i915_request_unsubmit:
vcs0 fence e80d:2, current 1
<0> [629.620277]    dmesg-922     1d.s1 618084950us : __i915_request_submit:
vcs0 fence e80d:2, current 1
<0> [629.620481]    dmesg-922     1d.s1 618084955us : __i915_request_submit:
vcs0 fence e80e:2, current 0
<0> [629.620674]    dmesg-922     1d.s1 618084958us :
__execlists_submission_tasklet: vcs0: queue_priority_

<7> [629.633154] __intel_gt_set_wedged vcs0
<7> [629.633159] __intel_gt_set_wedged  Awake? 2
<7> [629.633165] __intel_gt_set_wedged  Hangcheck: 0 ms ago
<7> [629.633170] __intel_gt_set_wedged  Reset count: 0 (global 0)
<7> [629.633175] __intel_gt_set_wedged  Requests:
<7> [629.633183] __intel_gt_set_wedged  MMIO base:  0x00012000
<7> [629.633201] __intel_gt_set_wedged  RING_START: 0x00000000
<7> [629.633211] __intel_gt_set_wedged  RING_HEAD:  0x00000000
<7> [629.633220] __intel_gt_set_wedged  RING_TAIL:  0x00000000
<7> [629.633232] __intel_gt_set_wedged  RING_CTL:   0x00000000
<7> [629.633245] __intel_gt_set_wedged  RING_MODE:  0x00000200 [idle]
<7> [629.633254] __intel_gt_set_wedged  RING_IMR: fffffeff
<7> [629.633270] __intel_gt_set_wedged  ACTHD:  0x00000000_00000000
<7> [629.633286] __intel_gt_set_wedged  BBADDR: 0x00000000_00000000
<7> [629.633303] __intel_gt_set_wedged  DMA_FADDR: 0x00000000_00000000
<7> [629.633312] __intel_gt_set_wedged  IPEIR: 0x00000000
<7> [629.633321] __intel_gt_set_wedged  IPEHR: 0x00000000
<7> [629.633334] __intel_gt_set_wedged  Execlist status: 0x00000301 00000000,
entries 6
<7> [629.633339] __intel_gt_set_wedged  Execlist CSB read 4, write 4, tasklet
queued? no (enabled)
<7> [629.633421] __intel_gt_set_wedged          E  e80d:2!  prio=3 @ 11565ms:
[i915]
<7> [629.633469] __intel_gt_set_wedged          E  e80e:2!+  prio=3 @ 11565ms:
signaled
<7> [629.633477] __intel_gt_set_wedged          E  e80d:4!  prio=2 @ 11565ms:
[i915]
<7> [629.633482] __intel_gt_set_wedged HWSP:
<7> [629.633490] __intel_gt_set_wedged [0000] 00000000 00000000 00000000
00000000 00000000 00000000 00000000 00000000
<7> [629.633494] __intel_gt_set_wedged *
<7> [629.633502] __intel_gt_set_wedged [0040] 00008002 00000001 00000814
00000001 00000818 00000001 00000001 00000000
<7> [629.633508] __intel_gt_set_wedged [0060] 00000818 00000001 00008002
00000001 00000000 00000000 00000000 00000004
<7> [629.633515] __intel_gt_set_wedged [0080] 00000000 00000000 00000000
00000000 00000000 00000000 00000000 00000000
<7> [629.633520] __intel_gt_set_wedged *

where it is much clearer that unlite-restore hit it's mark and tricked the HW
into doing something naughty.

-- 
You are receiving this mail because:
You are on the CC list for the bug.
You are the assignee for the bug.
You are the QA Contact for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.freedesktop.org/archives/intel-gfx-bugs/attachments/20191003/52f77955/attachment.html>