[Bug 101857] [IGT] [BYT] Fail some sub test on gem_evict_everything

Thu Jul 20 22:44:35 UTC 2017

https://bugs.freedesktop.org/show_bug.cgi?id=101857

--- Comment #3 from Chris Wilson <chris at chris-wilson.co.uk> ---
Lots of hilarity, lets start with

[80030.768913] ================================
[80030.768921] WARNING: inconsistent lock state
[80030.768930] 4.13.0-rc1-drm-tip-ww29-commit-f1c32d6+ #1 Tainted: G     U      
[80030.768942] --------------------------------
[80030.768949] inconsistent {RECLAIM_FS-ON-W} -> {IN-RECLAIM_FS-W} usage.
[80030.768960] kswapd0/37 [HC0[0]:SC0[0]:HE1:SE1] takes:
[80030.768968]  (&dev_priv->rps.hw_lock){+.+.?.}, at: [<ffffffffa008508f>]
intel_disable_gt_powersave.part.19+0x1f/0x260 [i915]
[80030.769050] {RECLAIM_FS-ON-W} state was registered at:
[80030.769063]   mark_held_locks+0x6f/0xa0
[80030.769073]   lockdep_trace_alloc+0x96/0xe0
[80030.769082]   kmem_cache_alloc_trace+0x31/0x2e0
[80030.769139]   i915_gem_object_create_stolen_for_preallocated+0x90/0x470
[i915]
[80030.769196]   intel_init_gt_powersave+0x7e5/0x8a0 [i915]
[80030.769256]   intel_modeset_gem_init+0xd/0x20 [i915]
[80030.769307]   i915_driver_load+0xa1a/0x16c0 [i915]
[80030.769360]   i915_pci_probe+0x34/0xa0 [i915]
[80030.769370]   pci_device_probe+0xa3/0x130
[80030.769381]   driver_probe_device+0x299/0x440
[80030.769390]   __driver_attach+0xde/0xe0
[80030.769399]   bus_for_each_dev+0x61/0xa0
[80030.769407]   driver_attach+0x19/0x20
[80030.769416]   bus_add_driver+0x1fb/0x270
[80030.769425]   driver_register+0x5b/0xd0
[80030.769434]   __pci_register_driver+0x5b/0x60
[80030.769447]   snd_pcm_new_stream+0xb/0x1e0 [snd_pcm]
[80030.769458]   do_one_initcall+0x3f/0x160
[80030.769468]   do_init_module+0x5a/0x1fa
[80030.769477]   load_module+0x21e0/0x2590
[80030.769485]   SyS_finit_module+0xbc/0xf0
[80030.769495]   entry_SYSCALL_64_fastpath+0x1c/0xb1
[80030.769503] irq event stamp: 1494801
[80030.769512] hardirqs last  enabled at (1494801): [<ffffffff81884f82>]
_raw_spin_unlock_irqrestore+0x52/0x60
[80030.769527] hardirqs last disabled at (1494800): [<ffffffff81884db3>]
_raw_spin_lock_irqsave+0x13/0x50
[80030.769542] softirqs last  enabled at (1494626): [<ffffffff81086349>]
__do_softirq+0x1d9/0x4a0
[80030.769556] softirqs last disabled at (1494603): [<ffffffff81086789>]
irq_exit+0xa9/0xc0
[80030.769569] 
               other info that might help us debug this:
[80030.769580]  Possible unsafe locking scenario:

[80030.769591]        CPU0
[80030.769597]        ----
[80030.769603]   lock(&dev_priv->rps.hw_lock);
[80030.769613]   <Interrupt>
[80030.769619]     lock(&dev_priv->rps.hw_lock);
[80030.769629] 
                *** DEADLOCK ***

[80030.769643] 2 locks held by kswapd0/37:
[80030.769651]  #0:  (shrinker_rwsem){++++..}, at: [<ffffffff8119a12a>]
shrink_slab.part.17+0x4a/0x5f0
[80030.769671]  #1:  (&dev->struct_mutex){+.+.+.}, at: [<ffffffffa00c454e>]
shrinker_lock+0x2e/0x80 [i915]
[80030.769737] 
               stack backtrace:
[80030.769749] CPU: 1 PID: 37 Comm: kswapd0 Tainted: G     U         
4.13.0-rc1-drm-tip-ww29-commit-f1c32d6+ #1
[80030.769763] Hardware name:
\xffffffff\xffffffff\xffffffff\xffffffff\xffffffff\xffffffff\xffffffff\xffffffff\xffffffff\xffffffff\xffffffff\xffffffff\xffffffff\xffffffff\xffffffff\xffffffff\xffffffff\xffffffff\xffffffff\xffffffff\xffffffff\xffffffff\xffffffff\xffffffff\xffffffff\xffffffff\xffffffff\xffffffff\xffffffff\xffffffff\xffffffff\xffffffff\xffffffff
\xffffffff\xffffffff\xffffffff\xffffffff\xffffffff\xffffffff\xffffffff\xffffffff\xffffffff\xffffffff\xffffffff\xffffffff\xffffffff\xffffffff\xffffffff\xffffffff\xffffffff\xffffffff\xffffffff\xffffffff\xffffffff\xffffffff\xffffffff\xffffffff\xffffffff\xffffffff\xffffffff\xffffffff\xffffffff\xffffffff\xffffffff\xffffffff\xffffffff/DN2820FYK,
BIOS FYBYT10H.86A.0057.2017.0303.1852 03/03/2017
[80030.769779] Call Trace:
[80030.769789]  dump_stack+0x67/0x97
[80030.769799]  print_usage_bug.part.26+0x245/0x25e
[80030.769809]  mark_lock+0x121/0x630
[80030.769818]  ? print_shortest_lock_dependencies+0x1a0/0x1a0
[80030.769828]  __lock_acquire+0x36d/0x1790
[80030.769839]  lock_acquire+0x9e/0x1e0
[80030.769847]  ? lock_acquire+0x9e/0x1e0
[80030.769900]  ? intel_disable_gt_powersave.part.19+0x1f/0x260 [i915]
[80030.769955]  ? intel_disable_gt_powersave.part.19+0x1f/0x260 [i915]
[80030.769966]  __mutex_lock+0x83/0x980
[80030.770019]  ? intel_disable_gt_powersave.part.19+0x1f/0x260 [i915]
[80030.770030]  ? trace_hardirqs_on_caller+0x118/0x180
[80030.770083]  ? intel_disable_gt_powersave.part.19+0x1f/0x260 [i915]
[80030.770094]  ? trace_hardirqs_on+0xd/0x10
[80030.770151]  ? fwtable_read32+0x88/0x2c0 [i915]
[80030.770162]  mutex_lock_nested+0x16/0x20
[80030.770171]  ? mutex_lock_nested+0x16/0x20
[80030.770224]  intel_disable_gt_powersave.part.19+0x1f/0x260 [i915]
[80030.770278]  intel_sanitize_gt_powersave+0x1f/0x30 [i915]
[80030.770335]  i915_gem_reset+0x96/0xf0 [i915]
[80030.770387]  i915_reset+0x132/0x1c0 [i915]
[80030.770443]  __i915_wait_request_check_and_reset.isra.12.part.13+0x28/0x30
[i915]
[80030.770504]  i915_wait_request+0x804/0x8a0 [i915]
[80030.770514]  ? krealloc+0x74/0xc0
[80030.770524]  ? wake_up_q+0x70/0x70
[80030.770532]  ? wake_up_q+0x70/0x70
[80030.770588]  i915_gem_object_wait_fence+0x69/0x100 [i915]
[80030.770644]  i915_gem_object_wait+0x2c7/0x3f0 [i915]
[80030.770653]  ? _raw_spin_unlock_irqrestore+0x52/0x60
[80030.770710]  i915_gem_object_unbind+0x47/0x150 [i915]
[80030.770767]  i915_gem_shrink+0x278/0x580 [i915]
[80030.770777]  ? mark_held_locks+0x6f/0xa0
[80030.770834]  i915_gem_shrinker_scan+0xae/0xc0 [i915]
[80030.770891]  ? i915_gem_shrinker_scan+0xae/0xc0 [i915]
[80030.770902]  shrink_slab.part.17+0x1f3/0x5f0
[80030.770912]  shrink_node+0x2e8/0x2f0
[80030.770921]  kswapd+0x2e9/0x940
[80030.770931]  kthread+0x10f/0x150
[80030.770939]  ? mem_cgroup_shrink_node+0x330/0x330
[80030.770949]  ? kthread_create_on_node+0x40/0x40
[80030.770958]  ? kthread_create_on_node+0x40/0x40
[80030.770967]  ret_from_fork+0x27/0x40

-- 
You are receiving this mail because:
You are the assignee for the bug.
You are on the CC list 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/20170720/18f283b0/attachment.html>