[Bug 112225] [CI][BAT] igt at i915_selftest@live_blt - dmesg-fail - igt_copy_blt failed with error 9

[bugzilla-daemon at freedesktop.org]

https://bugs.freedesktop.org/show_bug.cgi?id=112225

--- Comment #2 from Chris Wilson <chris at chris-wilson.co.uk> ---
All we have of interest is

<4>[  502.237878] RIP: 0010:__free_pages+0x0/0x20
<4>[  502.237884] Code: e8 a5 f0 ff ff f7 c5 00 02 00 00 74 0a e8 58 be f7 ff
55 9d 5b 5d c3 55 9d 5b 5d e9 4a c0 f7 ff 66 2e 0f 1f 84 00 00 00 00 00 <f0> ff
4f 34 75 0b 85 f6 74 05 e9 c1 dd ff ff eb 8f f3 c3 0f 1f 00
<4>[  502.237888] RSP: 0018:ffffc9000034be00 EFLAGS: 00010246
<4>[  502.237892] RAX: 0000000000000002 RBX: ffff888057ceccc0 RCX:
0000000000000001
<4>[  502.237895] RDX: ffff888057ceccc0 RSI: 0000000000000000 RDI:
0000000000000000
<4>[  502.237898] RBP: 0000000000000dfd R08: ffffea00000880c8 R09:
0000000000000000
<4>[  502.237901] R10: 0000000000000000 R11: 0000000000000000 R12:
ffff88805bb83cd0
<4>[  502.237904] R13: ffff888057ceccc0 R14: ffff88805c18d180 R15:
0000100000000000
<4>[  502.237908] FS:  0000000000000000(0000) GS:ffff888100100000(0000)
knlGS:0000000000000000
<4>[  502.237911] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
<4>[  502.237914] CR2: 0000000000000034 CR3: 0000000066af4001 CR4:
00000000000606e0
<3>[  502.237919] BUG: sleeping function called from invalid context at
./include/linux/percpu-rwsem.h:38
<3>[  502.237922] in_atomic(): 0, irqs_disabled(): 1, non_block: 0, pid: 4028,
name: igt/blt-1
<4>[  502.237925] INFO: lockdep is turned off.
<4>[  502.237928] irq event stamp: 296790
<4>[  502.237932] hardirqs last  enabled at (296789): [<ffffffff8121fab8>]
free_unref_page+0x48/0x60
<4>[  502.237937] hardirqs last disabled at (296790): [<ffffffff81001c0a>]
trace_hardirqs_off_thunk+0x1a/0x20
<4>[  502.237943] softirqs last  enabled at (294316): [<ffffffff81c00385>]
__do_softirq+0x385/0x47f
<4>[  502.237948] softirqs last disabled at (294301): [<ffffffff810b7faa>]
irq_exit+0xba/0xc0
<4>[  502.237953] CPU: 1 PID: 4028 Comm: igt/blt-1 Tainted: G     UD          
5.4.0-rc6-CI-CI_DRM_7279+ #1
<4>[  502.237956] Hardware name: GOOGLE Peppy/Peppy, BIOS MrChromebox
02/04/2018
<4>[  502.237959] Call Trace:
<4>[  502.237967]  dump_stack+0x67/0x9b
<4>[  502.237973]  ___might_sleep+0x178/0x260
<4>[  502.237979]  exit_signals+0x2b/0x350
<4>[  502.237983]  do_exit+0xa3/0xd70
<4>[  502.238116]  ? i915_gem_object_copy_blt+0x2a0/0x2a0 [i915]
<4>[  502.238124]  ? kthread+0x119/0x130
<4>[  502.238131]  rewind_stack_do_exit+0x17/0x20
<3>[  502.270759] vaddr[0]=3e8c1322, expected=ee6fa1a1

which smells of a locking bug. But we lack the start of the oops.

If it is the locking bug, we expect it to strike any machine in live_blt, so
optimistically we might get a better trace.

-- 
You are receiving this mail because:
You are the QA Contact for the bug.
You are the assignee for the bug.
You are on the CC list for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.freedesktop.org/archives/intel-gfx-bugs/attachments/20191107/669d83db/attachment.html>