https://bugs.freedesktop.org/show_bug.cgi?id=109692
Bug ID: 109692 Summary: deadlock occurs during GPU reset Product: DRI Version: XOrg git Hardware: Other OS: All Status: NEW Severity: normal Priority: medium Component: DRM/AMDgpu Assignee: dri-devel@lists.freedesktop.org Reporter: mikhail.v.gavrilov@gmail.com
Created attachment 143419 --> https://bugs.freedesktop.org/attachment.cgi?id=143419&action=edit dmesg
Steps for reproduce: 1. $ git clone git://people.freedesktop.org/~agd5f/linux -b amd-staging-drm-next 2. $ make bzImage && make module 3. # make modules_install && make install 4. Launch "Shadow of the Tomb Raider" --- Here GPU hung occurs --- and after few time --- Here start GPU reset --- --- Here Deadlock occurs ---
[ 291.746741] amdgpu 0000:0b:00.0: [gfxhub] no-retry page fault (src_id:0 ring:158 vmid:7 pasid:32774, for process SOTTR.exe pid 5250 thread SOTTR.exe pid 5250) [ 291.746750] amdgpu 0000:0b:00.0: in page starting at address 0x0000000000002000 from 27 [ 291.746754] amdgpu 0000:0b:00.0: VM_L2_PROTECTION_FAULT_STATUS:0x0070113C [ 297.135183] [drm:amdgpu_dm_atomic_commit_tail [amdgpu]] *ERROR* Waiting for fences timed out. [ 302.255032] [drm:amdgpu_dm_atomic_commit_tail [amdgpu]] *ERROR* Waiting for fences timed out. [ 302.265813] [drm:amdgpu_job_timedout [amdgpu]] *ERROR* ring gfx timeout, signaled seq=13292, emitted seq=13293 [ 302.265950] [drm:amdgpu_job_timedout [amdgpu]] *ERROR* Process information: process SOTTR.exe pid 5250 thread SOTTR.exe pid 5250 [ 302.265974] amdgpu 0000:0b:00.0: GPU reset begin!
[ 302.266337] ====================================================== [ 302.266338] WARNING: possible circular locking dependency detected [ 302.266340] 5.0.0-rc1-drm-next-kernel+ #1 Tainted: G C [ 302.266341] ------------------------------------------------------ [ 302.266343] kworker/5:2/871 is trying to acquire lock: [ 302.266345] 000000000abbb16a (&(&ring->fence_drv.lock)->rlock){-.-.}, at: dma_fence_remove_callback+0x1a/0x60 [ 302.266352] but task is already holding lock: [ 302.266353] 000000006e32ba38 (&(&sched->job_list_lock)->rlock){-.-.}, at: drm_sched_stop+0x34/0x140 [gpu_sched] [ 302.266358] which lock already depends on the new lock.
[ 302.266360] the existing dependency chain (in reverse order) is: [ 302.266361] -> #1 (&(&sched->job_list_lock)->rlock){-.-.}: [ 302.266366] drm_sched_process_job+0x4d/0x180 [gpu_sched] [ 302.266368] dma_fence_signal+0x111/0x1a0 [ 302.266414] amdgpu_fence_process+0xa3/0x100 [amdgpu] [ 302.266470] sdma_v4_0_process_trap_irq+0x6e/0xa0 [amdgpu] [ 302.266523] amdgpu_irq_dispatch+0xc0/0x250 [amdgpu] [ 302.266576] amdgpu_ih_process+0x84/0xf0 [amdgpu] [ 302.266628] amdgpu_irq_handler+0x1b/0x50 [amdgpu] [ 302.266632] __handle_irq_event_percpu+0x3f/0x290 [ 302.266635] handle_irq_event_percpu+0x31/0x80 [ 302.266637] handle_irq_event+0x34/0x51 [ 302.266639] handle_edge_irq+0x7c/0x1a0 [ 302.266643] handle_irq+0xbf/0x100 [ 302.266646] do_IRQ+0x61/0x120 [ 302.266648] ret_from_intr+0x0/0x22 [ 302.266651] cpuidle_enter_state+0xbf/0x470 [ 302.266654] do_idle+0x1ec/0x280 [ 302.266657] cpu_startup_entry+0x19/0x20 [ 302.266660] start_secondary+0x1b3/0x200 [ 302.266663] secondary_startup_64+0xa4/0xb0 [ 302.266664] -> #0 (&(&ring->fence_drv.lock)->rlock){-.-.}: [ 302.266668] _raw_spin_lock_irqsave+0x49/0x83 [ 302.266670] dma_fence_remove_callback+0x1a/0x60 [ 302.266673] drm_sched_stop+0x59/0x140 [gpu_sched] [ 302.266717] amdgpu_device_pre_asic_reset+0x4f/0x240 [amdgpu] [ 302.266761] amdgpu_device_gpu_recover+0x88/0x7d0 [amdgpu] [ 302.266822] amdgpu_job_timedout+0x109/0x130 [amdgpu] [ 302.266827] drm_sched_job_timedout+0x40/0x70 [gpu_sched] [ 302.266831] process_one_work+0x272/0x5d0 [ 302.266834] worker_thread+0x50/0x3b0 [ 302.266836] kthread+0x108/0x140 [ 302.266839] ret_from_fork+0x27/0x50 [ 302.266840] other info that might help us debug this:
[ 302.266841] Possible unsafe locking scenario:
[ 302.266842] CPU0 CPU1 [ 302.266843] ---- ---- [ 302.266844] lock(&(&sched->job_list_lock)->rlock); [ 302.266846] lock(&(&ring->fence_drv.lock)->rlock); [ 302.266847] lock(&(&sched->job_list_lock)->rlock); [ 302.266849] lock(&(&ring->fence_drv.lock)->rlock); [ 302.266850] *** DEADLOCK ***
[ 302.266852] 5 locks held by kworker/5:2/871: [ 302.266853] #0: 00000000d133fb6e ((wq_completion)"events"){+.+.}, at: process_one_work+0x1e9/0x5d0 [ 302.266857] #1: 000000008a5c3f7e ((work_completion)(&(&sched->work_tdr)->work)){+.+.}, at: process_one_work+0x1e9/0x5d0 [ 302.266862] #2: 00000000b9b2c76f (&adev->lock_reset){+.+.}, at: amdgpu_device_lock_adev+0x17/0x40 [amdgpu] [ 302.266908] #3: 00000000ac637728 (&dqm->lock_hidden){+.+.}, at: kgd2kfd_pre_reset+0x30/0x60 [amdgpu] [ 302.266965] #4: 000000006e32ba38 (&(&sched->job_list_lock)->rlock){-.-.}, at: drm_sched_stop+0x34/0x140 [gpu_sched] [ 302.266971] stack backtrace: [ 302.266975] CPU: 5 PID: 871 Comm: kworker/5:2 Tainted: G C 5.0.0-rc1-drm-next-kernel+ #1 [ 302.266976] Hardware name: System manufacturer System Product Name/ROG STRIX X470-I GAMING, BIOS 1103 11/16/2018 [ 302.266980] Workqueue: events drm_sched_job_timedout [gpu_sched] [ 302.266982] Call Trace: [ 302.266987] dump_stack+0x85/0xc0 [ 302.266991] print_circular_bug.isra.0.cold+0x15c/0x195 [ 302.266994] __lock_acquire+0x134c/0x1660 [ 302.266998] ? add_lock_to_list.isra.0+0x67/0xb0 [ 302.267003] lock_acquire+0xa2/0x1b0 [ 302.267006] ? dma_fence_remove_callback+0x1a/0x60 [ 302.267011] _raw_spin_lock_irqsave+0x49/0x83 [ 302.267013] ? dma_fence_remove_callback+0x1a/0x60 [ 302.267016] dma_fence_remove_callback+0x1a/0x60 [ 302.267020] drm_sched_stop+0x59/0x140 [gpu_sched] [ 302.267065] amdgpu_device_pre_asic_reset+0x4f/0x240 [amdgpu] [ 302.267110] amdgpu_device_gpu_recover+0x88/0x7d0 [amdgpu] [ 302.267173] amdgpu_job_timedout+0x109/0x130 [amdgpu] [ 302.267178] drm_sched_job_timedout+0x40/0x70 [gpu_sched] [ 302.267183] process_one_work+0x272/0x5d0 [ 302.267188] worker_thread+0x50/0x3b0 [ 302.267191] kthread+0x108/0x140 [ 302.267194] ? process_one_work+0x5d0/0x5d0 [ 302.267196] ? kthread_park+0x90/0x90 [ 302.267199] ret_from_fork+0x27/0x50 [ 302.692194] amdgpu 0000:0b:00.0: [drm:amdgpu_ring_test_helper [amdgpu]] *ERROR* ring kiq_2.1.0 test failed (-110) [ 302.692234] [drm:gfx_v9_0_hw_fini [amdgpu]] *ERROR* KCQ disable failed [ 302.768931] amdgpu 0000:0b:00.0: GPU BACO reset [ 303.278874] amdgpu 0000:0b:00.0: GPU reset succeeded, trying to resume [ 303.279006] [drm] PCIE GART of 512M enabled (table at 0x000000F400900000). [ 303.279072] [drm:amdgpu_device_gpu_recover [amdgpu]] *ERROR* VRAM is lost! [ 303.279234] [drm] PSP is resuming... [ 303.426601] [drm] reserve 0x400000 from 0xf400d00000 for PSP TMR SIZE [ 303.572227] [drm] UVD and UVD ENC initialized successfully. [ 303.687727] [drm] VCE initialized successfully. [ 303.689585] [drm] recover vram bo from shadow start [ 303.722757] [drm] recover vram bo from shadow done [ 303.722761] [drm] Skip scheduling IBs! [ 303.722791] amdgpu 0000:0b:00.0: GPU reset(2) succeeded! [ 303.722811] [drm] Skip scheduling IBs! [ 303.722838] [drm] Skip scheduling IBs! [ 303.722846] [drm] Skip scheduling IBs! [ 303.722854] [drm] Skip scheduling IBs! [ 303.722863] [drm] Skip scheduling IBs! [ 303.722871] [drm] Skip scheduling IBs!