Created attachment 130546 [details] dmesg [ 1056.574349] [drm] GPU HANG: ecode 7:0:0xf3cffffe, in opera-developer [2350], reason: Hang on render ring, action: reset [ 1056.574467] [drm] GPU hangs can indicate a bug anywhere in the entire gfx stack, including userspace. [ 1056.574469] [drm] Please file a _new_ bug report on bugs.freedesktop.org against DRI -> DRM/Intel [ 1056.574471] [drm] drm/i915 developers can then reassign to the right component if it's not a kernel issue. [ 1056.574473] [drm] The gpu crash dump is required to analyze gpu hangs, so please always attach it. [ 1056.574475] [drm] GPU crash dump saved to /sys/class/drm/card0/error [ 1056.575266] drm/i915: Resetting chip after gpu hang [ 1064.570246] drm/i915: Resetting chip after gpu hang [ 1072.570387] drm/i915: Resetting chip after gpu hang [ 1082.554632] drm/i915: Resetting chip after gpu hang [ 1090.555692] drm/i915: Resetting chip after gpu hang [ 1109.498631] ================================= [ 1109.498632] [ INFO: inconsistent lock state ] [ 1109.498633] 4.11.0-0.rc3.git0.2.fc26.x86_64+debug #1 Not tainted [ 1109.498634] --------------------------------- [ 1109.498635] inconsistent {RECLAIM_FS-ON-W} -> {IN-RECLAIM_FS-W} usage. [ 1109.498636] kswapd0/97 [HC0[0]:SC0[0]:HE1:SE1] takes: [ 1109.498637] (&sb->s_type->i_mutex_key#21){++++?+}, at: [<ffffffffc0c39acb>] xfs_ilock+0x13b/0x220 [xfs] [ 1109.498675] {RECLAIM_FS-ON-W} state was registered at: [ 1109.498679] mark_held_locks+0x5f/0x90 [ 1109.498680] lockdep_trace_alloc+0x7d/0xe0 [ 1109.498682] kmem_cache_alloc+0x33/0x2e0 [ 1109.498706] kmem_zone_alloc+0x66/0x100 [xfs] [ 1109.498726] xfs_trans_alloc+0x6b/0x130 [xfs] [ 1109.498747] xfs_link+0x1a4/0x3a0 [xfs] [ 1109.498765] xfs_vn_link+0x6a/0xb0 [xfs] [ 1109.498767] vfs_link+0x238/0x320 [ 1109.498769] SyS_link+0x1bf/0x200 [ 1109.498771] entry_SYSCALL_64_fastpath+0x1f/0xc2 [ 1109.498771] irq event stamp: 591437 [ 1109.498774] hardirqs last enabled at (591437): [<ffffffff9813c131>] __call_rcu+0x1f1/0x3d0 [ 1109.498775] hardirqs last disabled at (591436): [<ffffffff9813bfe2>] __call_rcu+0xa2/0x3d0 [ 1109.498777] softirqs last enabled at (589116): [<ffffffff98980c12>] __do_softirq+0x382/0x511 [ 1109.498779] softirqs last disabled at (589109): [<ffffffff980ba95f>] irq_exit+0x10f/0x120 [ 1109.498780] other info that might help us debug this: [ 1109.498781] Possible unsafe locking scenario: [ 1109.498782] CPU0 [ 1109.498782] ---- [ 1109.498783] lock(&sb->s_type->i_mutex_key#21); [ 1109.498785] <Interrupt> [ 1109.498785] lock(&sb->s_type->i_mutex_key#21); [ 1109.498787] *** DEADLOCK *** [ 1109.498788] 2 locks held by kswapd0/97: [ 1109.498789] #0: (shrinker_rwsem){++++..}, at: [<ffffffff9823173b>] shrink_slab.part.43+0x5b/0x590 [ 1109.498793] #1: (&type->s_umount_key#62){++++.+}, at: [<ffffffff982ce68b>] trylock_super+0x1b/0x50 [ 1109.498795] stack backtrace: [ 1109.498797] CPU: 0 PID: 97 Comm: kswapd0 Not tainted 4.11.0-0.rc3.git0.2.fc26.x86_64+debug #1 [ 1109.498798] Hardware name: Gigabyte Technology Co., Ltd. Z87M-D3H/Z87M-D3H, BIOS F11 08/12/2014 [ 1109.498799] Call Trace: [ 1109.498802] dump_stack+0x8e/0xd1 [ 1109.498804] print_usage_bug+0x1dd/0x1f0 [ 1109.498806] mark_lock+0x504/0x5a0 [ 1109.498808] ? print_shortest_lock_dependencies+0x190/0x190 [ 1109.498809] __lock_acquire+0x2b7/0x1320 [ 1109.498811] ? mark_held_locks+0x5f/0x90 [ 1109.498813] ? trace_hardirqs_on_caller+0xf4/0x1b0 [ 1109.498814] ? trace_hardirqs_on+0xd/0x10 [ 1109.498816] lock_acquire+0xb7/0x210 [ 1109.498837] ? xfs_ilock+0x13b/0x220 [xfs] [ 1109.498858] ? xfs_inactive+0xf5/0x130 [xfs] [ 1109.498860] down_write_nested+0x5e/0xc0 [ 1109.498876] ? xfs_ilock+0x13b/0x220 [xfs] [ 1109.498892] xfs_ilock+0x13b/0x220 [xfs] [ 1109.498908] xfs_inactive+0xf5/0x130 [xfs] [ 1109.498925] xfs_fs_destroy_inode+0xbb/0x2d0 [xfs] [ 1109.498928] destroy_inode+0x3b/0x60 [ 1109.498929] evict+0x13e/0x1a0 [ 1109.498931] dispose_list+0x56/0x80 [ 1109.498933] prune_icache_sb+0x5a/0x80 [ 1109.498934] super_cache_scan+0x137/0x1b0 [ 1109.498936] shrink_slab.part.43+0x1f4/0x590 [ 1109.498938] shrink_slab+0x1b/0x30 [ 1109.498940] shrink_node+0x102/0x300 [ 1109.498942] kswapd+0x334/0x950 [ 1109.498944] ? mem_cgroup_shrink_node+0x330/0x330 [ 1109.498945] kthread+0x12c/0x150 [ 1109.498946] ? kthread_create_on_node+0x60/0x60 [ 1109.498949] ret_from_fork+0x31/0x40 [ 1123.579229] drm/i915: Resetting chip after gpu hang [ 1133.563266] drm/i915: Resetting chip after gpu hang [ 1141.563424] drm/i915: Resetting chip after gpu hang [ 1149.563562] drm/i915: Resetting chip after gpu hang [ 1157.564590] drm/i915: Resetting chip after gpu hang [ 1165.564198] drm/i915: Resetting chip after gpu hang [ 1173.564077] drm/i915: Resetting chip after gpu hang [ 1181.564180] drm/i915: Resetting chip after gpu hang [ 1189.564268] drm/i915: Resetting chip after gpu hang [ 1197.564450] drm/i915: Resetting chip after gpu hang [ 1205.564604] drm/i915: Resetting chip after gpu hang [ 1213.565726] drm/i915: Resetting chip after gpu hang [ 1221.565727] drm/i915: Resetting chip after gpu hang [ 1229.564981] drm/i915: Resetting chip after gpu hang [ 1239.613015] drm/i915: Resetting chip after gpu hang [ 1246.589224] drm/i915: Resetting chip after gpu hang [ 1254.589388] drm/i915: Resetting chip after gpu hang [ 1256.061497] show_signal_msg: 210 callbacks suppressed [ 1256.061504] opera-developer[17061]: segfault at 968 ip 00007fa190c6e2b0 sp 00007ffff97f6090 error 4 in libX11.so.6.3.0[7fa190c41000+138000] [ 1256.972080] LevelDBEnv (2920) used greatest stack depth: 10224 bytes left [ 1257.041641] TaskSchedulerFo (2313) used greatest stack depth: 10144 bytes left [ 1268.029490] asynchronous wait on fence i915:[global]:b8541 timed out [ 1278.269774] [drm:i915_mutex_lock_interruptible [i915]] *ERROR* Timed out waiting for the gpu reset to complete [ 1289.021849] asynchronous wait on fence i915:[global]:b8541 timed out [ 1289.022146] drm/i915: Resetting chip after gpu hang [ 1296.574013] drm/i915: Resetting chip after gpu hang [ 1306.558218] drm/i915: Resetting chip after gpu hang [ 1314.558390] drm/i915: Resetting chip after gpu hang [ 1322.558509] drm/i915: Resetting chip after gpu hang
Created attachment 130547 [details] /sys/class/drm/card0/error
Created attachment 130548 [details] lspci -vvv
Created attachment 130549 [details] lshw
commit 5d4bac5503fcc67dd7999571e243cee49371aef7 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Mar 22 20:59:30 2017 +0000 drm/i915: Restore marking context objects as dirty on pinning
Use of freedesktop.org services, including Bugzilla, is subject to our Code of Conduct. How we collect and use information is described in our Privacy Policy.