The game The Evil Within Demo lead to GPU hang when played first cut scene.
Submitted by Mikhail Gavrilov @Mikhail
Assigned to Default DRI bug account
Link to original bug (#107867)
Description
Created attachment 141490
dmesg
- Install steam
- Enable steam play
- Launch game under Fedora 30 (Rawhide)
- Not skip first cut scene
[ 246.521518] [drm:amdgpu_job_timedout [amdgpu]] *ERROR* ring gfx timeout, signaled seq=27174, emitted seq=27176
[ 246.521534] [drm] GPU recovery disabled.
[ 332.537451] kworker/dying (23) used greatest stack depth: 11392 bytes left
[ 369.399369] INFO: task kworker/u32:8:204 blocked for more than 120 seconds.
[ 369.399384] Tainted: G C 4.19.0-0.rc2.git2.1.fc30.x86_64 #1 (closed)
[ 369.399386] "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message.
[ 369.399389] kworker/u32:8 D10560 204 2 0x80000000
[ 369.399406] Workqueue: events_unbound commit_work [drm_kms_helper]
[ 369.399410] Call Trace:
[ 369.399420] ? __schedule+0x2e2/0xa70
[ 369.399424] ? native_sched_clock+0x3e/0xa0
[ 369.399432] ? dma_fence_default_wait+0x242/0x330
[ 369.399437] schedule+0x2f/0x90
[ 369.399441] schedule_timeout+0x31c/0x4f0
[ 369.399516] ? mark_held_locks+0x57/0x80
[ 369.399520] ? _raw_spin_unlock_irqrestore+0x4b/0x60
[ 369.399527] ? dma_fence_default_wait+0x242/0x330
[ 369.399530] dma_fence_default_wait+0x26e/0x330
[ 369.399535] ? dma_fence_release+0x120/0x120
[ 369.399541] dma_fence_wait_timeout+0x182/0x200
[ 369.399547] reservation_object_wait_timeout_rcu+0x236/0x4e0
[ 369.399619] amdgpu_dm_do_flip+0x112/0x340 [amdgpu]
[ 369.399722] amdgpu_dm_atomic_commit_tail+0x72b/0xd90 [amdgpu]
[ 369.399730] ? wait_for_completion_timeout+0x73/0x1a0
[ 369.399748] commit_tail+0x3d/0x70 [drm_kms_helper]
[ 369.399753] process_one_work+0x27d/0x5c0
[ 369.399764] worker_thread+0x3c/0x390
[ 369.399770] ? process_one_work+0x5c0/0x5c0
[ 369.399774] kthread+0x120/0x140
[ 369.399778] ? kthread_create_worker_on_cpu+0x70/0x70
[ 369.399783] ret_from_fork+0x27/0x50
[ 369.399927]
Showing all locks held in the system:
[ 369.399936] 1 lock held by khungtaskd/94:
[ 369.399937] #0: 000000007b8e38cf (rcu_read_lock){....}, at: debug_show_all_locks+0x15/0x183
[ 369.399954] 3 locks held by kworker/u32:8/204:
[ 369.399956] #0: 00000000265972fd ((wq_completion)"events_unbound"){+.+.}, at: process_one_work+0x1f3/0x5c0
[ 369.399967] #1 (closed): 0000000005b55269 ((work_completion)(&state->commit_work)){+.+.}, at: process_one_work+0x1f3/0x5c0
[ 369.399977] #2: 00000000857176ee (reservation_ww_class_mutex){+.+.}, at: amdgpu_dm_do_flip+0xd6/0x340 [amdgpu]
[ 369.400157] 1 lock held by EvilWithinDemo./3714:
[ 369.400171] =============================================
**Attachment 141490**, "dmesg":
dmesg3.txt