[Bug 110368] [CI][SHARDS] igt at gem_exec_nop@basic-series - fail - Failed assertion: !"GPU hung"

[bugzilla-daemon at freedesktop.org]

https://bugs.freedesktop.org/show_bug.cgi?id=110368

--- Comment #2 from Chris Wilson <chris at chris-wilson.co.uk> ---
Stuck waiting for a CS interrupt. I don't think it's exactly the same as the
garbage we see within the CSB -- as if we hit garbage there I expect the
various asserts to fire. So a CS interrupt not arriving makes more sense?

<7> [1693.074610] hangcheck rcs0
<7> [1693.074625] hangcheck     Hangcheck d2e90f01:d2e90f01 [6022 ms]
<7> [1693.074633] hangcheck     Reset count: 5 (global 0)
<7> [1693.074641] hangcheck     Requests:
<7> [1693.075384] hangcheck     RING_START: 0x007f2000
<7> [1693.075909] hangcheck     RING_HEAD:  0x000035f8
<7> [1693.076492] hangcheck     RING_TAIL:  0x000035f8
<7> [1693.076506] hangcheck     RING_CTL:   0x00003000
<7> [1693.076521] hangcheck     RING_MODE:  0x00000200 [idle]
<7> [1693.076533] hangcheck     RING_IMR: 00000000
<7> [1693.077255] hangcheck     ACTHD:  0x00000000_024035f8
<7> [1693.079415] hangcheck     BBADDR: 0x00000000_00000004
<7> [1693.080162] hangcheck     DMA_FADDR: 0x00000000_00000000
<7> [1693.081461] hangcheck     IPEIR: 0x00000000
<7> [1693.082112] hangcheck     IPEHR: 0x00000000
<7> [1693.084094] hangcheck     Execlist status: 0x00018001 00000000
<7> [1693.085382] hangcheck     Execlist CSB read 0, write 0 [mmio:0], tasklet
queued? no (enabled)
<7> [1693.085504] hangcheck             ELSP[0] count=1, ring:{start:007f2000,
hwsp:fffee200, seqno:000011b8}, rq:  b5f:11b8!  prio=6 @ 8128ms: signaled
<7> [1693.085511] hangcheck             ELSP[1] idle
<7> [1693.085518] hangcheck             HW active? 0x1
<7> [1693.085632] hangcheck             Queue priority hint: 6
<7> [1693.085642] hangcheck             Q  b5f:11ba  prio=6 @ 8128ms:
gem_exec_nop[5054]
<7> [1693.085651] hangcheck             Q  b5f:11bc  prio=4 @ 8127ms:
gem_exec_nop[5054]
<7> [1693.085660] hangcheck             Q  b5f:11be  prio=4 @ 8127ms:
gem_exec_nop[5054]
<7> [1693.085667] hangcheck             Q  b5f:11c0  prio=4 @ 8127ms:
gem_exec_nop[5054]
<7> [1693.085675] hangcheck             Q  b5f:11c2  prio=4 @ 8127ms:
gem_exec_nop[5054]
<7> [1693.085683] hangcheck             Q  b5f:11c4  prio=4 @ 8127ms:
gem_exec_nop[5054]
<7> [1693.085692] hangcheck             Q  b5f:11c6  prio=4 @ 8127ms:
gem_exec_nop[5054]
<7> [1693.085756] hangcheck             ...skipping 82 queued requests...
<7> [1693.085764] hangcheck             Q  b5f:126c  prio=4 @ 8124ms:
gem_exec_nop[5054]
<7> [1693.085997] hangcheck HWSP:
<7> [1693.086004] hangcheck [0000] 00000000 00000000 00000000 00000000 00000000
00000000 00000000 00000000
<7> [1693.086008] hangcheck *
<7> [1693.086016] hangcheck [0040] 10000018 00000040 10008002 00000040 10000018
00000040 10000001 00000000
<7> [1693.086022] hangcheck [0060] 10000018 00000040 10000001 00000000 00000000
00000000 00000000 00000000
<7> [1693.086028] hangcheck [0080] 00000000 00000000 00000000 00000000 00000000
00000000 00000000 00000000
<7> [1693.086033] hangcheck *
<7> [1693.086039] hangcheck [00c0] 00000000 00000000 00000000 00000000 d2e90f01
00000000 00000000 00000000
<7> [1693.086045] hangcheck [00e0] 00000000 00000000 00000000 00000000 00000000
00000000 00000000 00000000
<7> [1693.086050] hangcheck *
<7> [1693.086063] hangcheck Idle? no

However, the CSB pointers are zero... So that implies we did consume whatever
was most recently signaled.

-- 
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/20190409/49d834f5/attachment-0001.html>