Bug 110469

Created attachment 144037 [details]
dmesg log

When using discrete radeon gpu on my laptop in most application(via PRIME) I constantly get gpu hung.

01:00.0 Display controller: Advanced Micro Devices, Inc. [AMD/ATI] Sun XT [Radeon HD 8670A/8670M/8690M / R5 M330 / M430 / R7 M520] (rev 83)
	Subsystem: Hewlett-Packard Company Sun XT [Radeon HD 8670A/8670M/8690M / R5 M330 / M430 / Radeon 520 Mobile]
	Physical Slot: 0
	Control: I/O+ Mem+ BusMaster+ SpecCycle- MemWINV- VGASnoop- ParErr- Stepping- SERR- FastB2B- DisINTx+
	Status: Cap+ 66MHz- UDF- FastB2B- ParErr- DEVSEL=fast >TAbort- <TAbort- <MAbort- >SERR- <PERR- INTx-
	Latency: 0, Cache Line Size: 64 bytes
	Interrupt: pin A routed to IRQ 32
	Region 0: Memory at d0000000 (64-bit, prefetchable) [size=256M]
	Region 2: Memory at f0400000 (64-bit, non-prefetchable) [size=256K]
	Region 4: I/O ports at 4000 [size=256]
	Expansion ROM at f0440000 [disabled] [size=128K]
	Capabilities: <access denied>
	Kernel driver in use: radeon
	Kernel modules: radeon, amdgpu

I can reproduce this bug in glmark2. bug ALWAYS happens on [desktop] stage of benchmark. also this is not hardware problem(gpu works fine in windows/ no overheating)
Please help me, I cant use discrete card for a year now because this bug.

There is call trace in dmesg somehere in the middle(full logs attached):

Apr 18 23:45:40 HP kernel: radeon 0000:01:00.0: ring 3 stalled for more than 238640msec
Apr 18 23:45:40 HP kernel: radeon 0000:01:00.0: GPU lockup (current fence id 0x00000000000257d3 last fence id 0x00000000000257d8 on ring 3)
Apr 18 23:45:40 HP kernel: radeon 0000:01:00.0: ring 0 stalled for more than 239107msec
Apr 18 23:45:40 HP kernel: radeon 0000:01:00.0: GPU lockup (current fence id 0x0000000000012b81 last fence id 0x0000000000012b84 on ring 0)
Apr 18 23:45:40 HP kernel: INFO: task kworker/u8:0:3018 blocked for more than 120 seconds.
Apr 18 23:45:40 HP kernel:       Tainted: G           OE     5.0.3-pf #1
Apr 18 23:45:40 HP kernel: "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message.
Apr 18 23:45:40 HP kernel: kworker/u8:0    D    0  3018      2 0x80000080
Apr 18 23:45:40 HP kernel: Workqueue: events_unbound commit_work [drm_kms_helper]
Apr 18 23:45:40 HP kernel: Call Trace:
Apr 18 23:45:40 HP kernel:  ? __schedule+0x505/0x14d0
Apr 18 23:45:40 HP kernel:  ? ieee802_11_parse_elems_crc+0x16a/0x650 [mac80211]
Apr 18 23:45:40 HP kernel:  schedule+0x28/0x90
Apr 18 23:45:40 HP kernel:  schedule_timeout+0x23d/0x2e0
Apr 18 23:45:40 HP kernel:  ? dce110_timing_generator_get_crtc_scanoutpos+0x88/0x110 [amdgpu]
Apr 18 23:45:40 HP kernel:  dma_fence_default_wait+0x204/0x270
Apr 18 23:45:40 HP kernel:  ? dma_fence_wait_timeout+0x100/0x100
Apr 18 23:45:40 HP kernel:  dma_fence_wait_timeout+0xd9/0x100
Apr 18 23:45:40 HP kernel:  reservation_object_wait_timeout_rcu+0x1f2/0x370
Apr 18 23:45:40 HP kernel:  amdgpu_dm_do_flip+0x14a/0x4a0 [amdgpu]
Apr 18 23:45:40 HP kernel:  ? amdgpu_dm_atomic_commit_tail+0x5f9/0xbc0 [amdgpu]
Apr 18 23:45:40 HP kernel:  amdgpu_dm_atomic_commit_tail+0x5f9/0xbc0 [amdgpu]
Apr 18 23:45:40 HP kernel:  commit_tail+0x3d/0x70 [drm_kms_helper]
Apr 18 23:45:40 HP kernel:  process_one_work+0x1f4/0x3f0
Apr 18 23:45:40 HP kernel:  worker_thread+0x2d/0x3e0
Apr 18 23:45:40 HP kernel:  ? process_one_work+0x3f0/0x3f0
Apr 18 23:45:40 HP kernel:  kthread+0x112/0x130
Apr 18 23:45:40 HP kernel:  ? kthread_park+0x80/0x80
Apr 18 23:45:40 HP kernel:  ret_from_fork+0x1f/0x40