protonscr

amdgpu: failed to clear page tables on GEM object close (-16)

dxvkclosed amd radv
doitsujin/dxvk#1912 · opened 2021-01-29 by Webxorcist · updated 2021-01-31 · 4 comments · github
WWebxorcist 2021-01-29 github

I am not sure where this problem originates. I get regular dxvk updates and with every update lately the number of crashes seem to change. Sometimes I can play for hours without a crash. The next day after a dxvk update, some games are barely playable.

I do apologize if I am wrong to make a bug report here and I welcome any tip in the right direction.

It is not happening with al games, but all 3D open world games I play have this issue.

Software information

Cyberpunk 2077, Watch_Dogs 2, World of Warcraft

System information

  • GPU: Radeon RX 5700 XT
  • Kernel: 5.10.9
  • Wine version: Proton Experimental, Proton 5.13.5, Lutris-6.0-rc1-x86_64, Wine Staging 6.0
  • DXVK version: 1.7.1 / 1.7.3-32bit / 1.7.1L/1.7.2L/1.7.3L (Last 3 come with Lutris)
  • libvulkan: 1.2
  • Mesa: 20.3.3

Log files

I can't find the dxvk log files in Steam, but in dmesg:

[ 4139.396954] [drm:amdgpu_dm_atomic_commit_tail [amdgpu]] *ERROR* Waiting for fences timed out!
[ 4144.518133] [drm:amdgpu_job_timedout [amdgpu]] *ERROR* ring gfx timeout, but soft recovered
[ 4151.596625] [drm:amdgpu_gem_va_ioctl [amdgpu]] *ERROR* Couldn't update BO_VA (-16)
[ 4151.596795] [drm:amdgpu_gem_va_ioctl [amdgpu]] *ERROR* Couldn't update BO_VA (-16)
[ 4151.604220] [drm:amdgpu_gem_va_ioctl [amdgpu]] *ERROR* Couldn't update BO_VA (-16)
[ 4151.845264] [drm:amdgpu_gem_va_ioctl [amdgpu]] *ERROR* Couldn't update BO_VA (-16)
[ 4151.845373] [drm:amdgpu_gem_va_ioctl [amdgpu]] *ERROR* Couldn't update BO_VA (-16)
[ 4151.845447] [drm:amdgpu_gem_va_ioctl [amdgpu]] *ERROR* Couldn't update BO_VA (-16)
[ 4151.845537] [drm:amdgpu_gem_va_ioctl [amdgpu]] *ERROR* Couldn't update BO_VA (-16)
[ 4151.845652] [drm:amdgpu_gem_va_ioctl [amdgpu]] *ERROR* Couldn't update BO_VA (-16)
[ 4151.845762] [drm:amdgpu_gem_va_ioctl [amdgpu]] *ERROR* Couldn't update BO_VA (-16)
[ 4151.845854] [drm:amdgpu_gem_va_ioctl [amdgpu]] *ERROR* Couldn't update BO_VA (-16)
[ 4151.883146] [drm:amdgpu_gem_va_ioctl [amdgpu]] *ERROR* Couldn't update BO_VA (-16)
[ 4162.618449] amdgpu 0000:08:00.0: amdgpu: failed to clear page tables on GEM object close (-16)
[ 4162.618499] amdgpu 0000:08:00.0: amdgpu: failed to clear page tables on GEM object close (-16)
  <dozens of repeats removed>
[ 4162.620323] amdgpu 0000:08:00.0: amdgpu: failed to clear page tables on GEM object close (-16)
[ 4162.620326] amdgpu 0000:08:00.0: amdgpu: failed to clear page tables on GEM object close (-16)
[ 4162.620329] amdgpu 0000:08:00.0: amdgpu: failed to clear page tables on GEM object close (-16)
[ 4162.620333] amdgpu 0000:08:00.0: amdgpu: failed to clear page tables on GEM object close (-16)
[ 4162.622056] [drm:amdgpu_gem_va_ioctl [amdgpu]] *ERROR* Couldn't update BO_VA (-16)
[ 4162.735755] [drm:amdgpu_gem_va_ioctl [amdgpu]] *ERROR* Couldn't update BO_VA (-16)
[ 4162.739261] [drm:amdgpu_gem_va_ioctl [amdgpu]] *ERROR* Couldn't update BO_VA (-16)
[ 4162.743243] [drm:amdgpu_gem_va_ioctl [amdgpu]] *ERROR* Couldn't update BO_VA (-16)

and

[37912.592312] [drm:amdgpu_dm_atomic_commit_tail [amdgpu]] *ERROR* Waiting for fences timed out!
[37912.592412] [drm:amdgpu_dm_atomic_commit_tail [amdgpu]] *ERROR* Waiting for fences timed out!
[37917.466218] [drm:amdgpu_job_timedout [amdgpu]] *ERROR* ring gfx_0.0.0 timeout, signaled seq=8797063, emitted seq=8797065
[37917.466285] [drm:amdgpu_job_timedout [amdgpu]] *ERROR* Process information: process Cyberpunk2077.e pid 20444 thread Cyberpunk2:cs0 pid 20482
[37917.466289] amdgpu 0000:0b:00.0: amdgpu: GPU reset begin!
[37917.466588] ------------[ cut here ]------------
[37917.466671] WARNING: CPU: 27 PID: 19645 at drivers/gpu/drm/amd/amdgpu/../display/dc/dcn20/dcn20_resource.c:3240 dcn20_validate_bandwidth_fp+0x8d/0xd0 [amdgpu]
[37917.466672] Modules linked in: uinput snd_seq_dummy snd_hrtimer snd_seq xt_CHECKSUM xt_MASQUERADE xt_conntrack ipt_REJECT xt_tcpudp nf_nat_tftp nft_objref nf_conntrack_tftp bridge stp llc af_packet nft_fib_inet nft_fib_ipv4 nft_fib_ipv6 nft_fib nft_reject_inet nf_reject_ipv4 nf_reject_ipv6 nft_reject nft_ct nft_chain_nat nf_tables ebtable_nat ebtable_broute ip6table_nat ip6table_mangle ip6table_raw ip6table_security iptable_nat nf_nat nf_conntrack nf_defrag_ipv6 nf_defrag_ipv4 iptable_mangle iptable_raw iptable_security ip_set nfnetlink ebtable_filter ebtables ip6table_filter ip6_tables iptable_filter ip_tables x_tables bpfilter dmi_sysfs xfs dm_crypt hid_steam snd_hda_codec_realtek snd_hda_codec_generic ledtrig_audio snd_hda_codec_hdmi snd_hda_intel snd_intel_dspcfg soundwire_intel soundwire_generic_allocation soundwire_cadence snd_hda_codec snd_hda_core soundwire_bus eeepc_wmi snd_usb_audio snd_soc_core edac_mce_amd asus_wmi snd_usbmidi_lib kvm_amd snd_hwdep battery snd_rawmidi
[37917.466701]  sparse_keymap snd_compress snd_seq_device mc snd_pcm_dmaengine rfkill kvm tiny_power_button irqbypass joydev video snd_pcm efi_pstore pcspkr wmi_bmof mxm_wmi sp5100_tco k10temp snd_timer igb i2c_piix4 snd soundcore dca button acpi_cpufreq nls_iso8859_1 nls_cp437 vfat fat fuse configfs hid_generic usbhid amdgpu iommu_v2 gpu_sched ttm i2c_algo_bit drm_kms_helper syscopyarea sysfillrect sysimgblt fb_sys_fops cec xhci_pci xhci_pci_renesas rc_core xhci_hcd drm crct10dif_pclmul crc32_pclmul ghash_clmulni_intel usbcore aesni_intel nvme glue_helper ccp crypto_simd nvme_core cryptd wmi btrfs blake2b_generic libcrc32c crc32c_intel xor raid6_pq sg dm_multipath dm_mod scsi_dh_rdac scsi_dh_emc scsi_dh_alua i2c_dev msr efivarfs
[37917.466733] CPU: 27 PID: 19645 Comm: kworker/27:1 Not tainted 5.10.7-1-default #1 openSUSE Tumbleweed
[37917.466734] Hardware name: System manufacturer System Product Name/ROG STRIX X570-F GAMING, BIOS 3001 12/04/2020
[37917.466737] Workqueue: events drm_sched_job_timedout [gpu_sched]
[37917.466817] RIP: 0010:dcn20_validate_bandwidth_fp+0x8d/0xd0 [amdgpu]
[37917.466818] Code: 00 7b 35 22 85 14 1f 00 00 75 2f 31 d2 f2 0f 11 85 58 26 00 00 48 89 ee 4c 89 e7 e8 6d f6 ff ff 89 c2 22 95 14 1f 00 00 75 30 <0f> 0b 48 89 9d 58 26 00 00 5b 5d 41 5c c3 75 c9 48 89 9d 58 26 00
[37917.466819] RSP: 0018:ffffb86c8380fc08 EFLAGS: 00210246
[37917.466821] RAX: 0000000000000001 RBX: 4079400000000000 RCX: 000000000004feb9
[37917.466821] RDX: 0000000000000000 RSI: 9dcc1f596280c38f RDI: 00000000000331a0
[37917.466822] RBP: ffff9b6bc51a0000 R08: ffff9b706018c000 R09: ffff9b6d01290000
[37917.466823] R10: ffff9b706018c000 R11: 0000000100000001 R12: ffff9b6d01290000
[37917.466823] R13: ffff9b6d01290000 R14: ffff9b6cc004f000 R15: ffff9b6bc51a0000
[37917.466825] FS:  0000000000000000(0000) GS:ffff9b73cf0c0000(0000) knlGS:0000000000000000
[37917.466826] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[37917.466826] CR2: 00007f5081147000 CR3: 00000003b3432000 CR4: 0000000000750ee0
[37917.466827] PKRU: 55555554
[37917.466828] Call Trace:
[37917.466901]  dcn20_validate_bandwidth+0x24/0x40 [amdgpu]
[37917.466972]  dc_validate_global_state+0x2f2/0x390 [amdgpu]
[37917.467040]  ? dc_rem_all_planes_for_stream+0xcb/0x110 [amdgpu]
[37917.467112]  dm_suspend+0x174/0x1b0 [amdgpu]
[37917.467159]  amdgpu_device_ip_suspend_phase1+0x75/0xd0 [amdgpu]
[37917.467207]  ? amdgpu_fence_process+0x4d/0x130 [amdgpu]
[37917.467253]  amdgpu_device_ip_suspend+0x1c/0x60 [amdgpu]
[37917.467335]  amdgpu_device_pre_asic_reset+0x185/0x19c [amdgpu]
[37917.467410]  amdgpu_device_gpu_recover.cold+0x5cf/0x95d [amdgpu]
[37917.467477]  amdgpu_job_timedout+0x11c/0x140 [amdgpu]
[37917.467481]  drm_sched_job_timedout+0x66/0xf0 [gpu_sched]
[37917.467484]  process_one_work+0x1df/0x370
[37917.467486]  worker_thread+0x50/0x400
[37917.467488]  ? process_one_work+0x370/0x370
[37917.467489]  kthread+0x11b/0x140
[37917.467491]  ? __kthread_bind_mask+0x60/0x60
[37917.467493]  ret_from_fork+0x22/0x30
[37917.467495] ---[ end trace 44dabd0fe84ee29d ]---
[37918.029218] amdgpu 0000:0b:00.0: [drm:amdgpu_ring_test_helper [amdgpu]] *ERROR* ring kiq_2.1.0 test failed (-110)
[37918.029248] [drm:gfx_v10_0_hw_fini [amdgpu]] *ERROR* KGQ disable failed
[37918.284934] amdgpu 0000:0b:00.0: [drm:amdgpu_ring_test_helper [amdgpu]] *ERROR* ring kiq_2.1.0 test failed (-110)
[37918.284965] [drm:gfx_v10_0_hw_fini [amdgpu]] *ERROR* KCQ disable failed
[37918.540910] [drm:gfx_v10_0_hw_fini [amdgpu]] *ERROR* failed to halt cp gfx
[37918.576127] [drm] free PSP TMR buffer
[37918.621072] amdgpu 0000:0b:00.0: amdgpu: BACO reset
[37920.715594] amdgpu 0000:0b:00.0: amdgpu: GPU reset succeeded, trying to resume
[37920.715811] [drm] PCIE GART of 512M enabled (table at 0x0000008000300000).
[37920.715832] [drm] VRAM is lost due to GPU reset!
[37920.731948] [drm] PSP is resuming...
[37920.908107] [drm] reserve 0x900000 from 0x81fe400000 for PSP TMR
[37921.100105] amdgpu 0000:0b:00.0: amdgpu: RAS: optional ras ta ucode is not available
[37921.124100] amdgpu 0000:0b:00.0: amdgpu: RAP: optional rap ta ucode is not available
[37921.124102] amdgpu 0000:0b:00.0: amdgpu: SMU is resuming...
[37921.124108] amdgpu 0000:0b:00.0: amdgpu: smu driver if version = 0x00000036, smu fw if version = 0x00000037, smu fw version = 0x002a3d00 (42.61.0)
[37921.124109] amdgpu 0000:0b:00.0: amdgpu: SMU driver if version not matched
[37921.126617] amdgpu 0000:0b:00.0: amdgpu: SMU is resumed successfully!
[37921.278267] [drm] kiq ring mec 2 pipe 1 q 0
[37921.288760] [drm] VCN decode and encode initialized successfully(under DPG Mode).
[37921.288890] [drm] JPEG decode initialized successfully.
[37921.288909] amdgpu 0000:0b:00.0: amdgpu: ring gfx_0.0.0 uses VM inv eng 0 on hub 0
[37921.288910] amdgpu 0000:0b:00.0: amdgpu: ring comp_1.0.0 uses VM inv eng 1 on hub 0
[37921.288910] amdgpu 0000:0b:00.0: amdgpu: ring comp_1.1.0 uses VM inv eng 4 on hub 0
[37921.288911] amdgpu 0000:0b:00.0: amdgpu: ring comp_1.2.0 uses VM inv eng 5 on hub 0
[37921.288912] amdgpu 0000:0b:00.0: amdgpu: ring comp_1.3.0 uses VM inv eng 6 on hub 0
[37921.288913] amdgpu 0000:0b:00.0: amdgpu: ring comp_1.0.1 uses VM inv eng 7 on hub 0
[37921.288913] amdgpu 0000:0b:00.0: amdgpu: ring comp_1.1.1 uses VM inv eng 8 on hub 0
[37921.288914] amdgpu 0000:0b:00.0: amdgpu: ring comp_1.2.1 uses VM inv eng 9 on hub 0
[37921.288915] amdgpu 0000:0b:00.0: amdgpu: ring comp_1.3.1 uses VM inv eng 10 on hub 0
[37921.288916] amdgpu 0000:0b:00.0: amdgpu: ring kiq_2.1.0 uses VM inv eng 11 on hub 0
[37921.288916] amdgpu 0000:0b:00.0: amdgpu: ring sdma0 uses VM inv eng 12 on hub 0
[37921.288917] amdgpu 0000:0b:00.0: amdgpu: ring sdma1 uses VM inv eng 13 on hub 0
[37921.288918] amdgpu 0000:0b:00.0: amdgpu: ring vcn_dec uses VM inv eng 0 on hub 1
[37921.288918] amdgpu 0000:0b:00.0: amdgpu: ring vcn_enc0 uses VM inv eng 1 on hub 1
[37921.288919] amdgpu 0000:0b:00.0: amdgpu: ring vcn_enc1 uses VM inv eng 4 on hub 1
[37921.288920] amdgpu 0000:0b:00.0: amdgpu: ring jpeg_dec uses VM inv eng 5 on hub 1
[37921.291741] amdgpu 0000:0b:00.0: amdgpu: recover vram bo from shadow start
[37921.316953] amdgpu 0000:0b:00.0: amdgpu: recover vram bo from shadow done
[37921.316954] [drm] Skip scheduling IBs!
[37921.316955] [drm] Skip scheduling IBs!
[37921.316977] [drm] Skip scheduling IBs!
[37921.316978] [drm] Skip scheduling IBs!
[37921.316986] amdgpu 0000:0b:00.0: amdgpu: GPU reset(2) succeeded!
[37921.317035] [drm:amdgpu_cs_ioctl [amdgpu]] *ERROR* Failed to initialize parser -125!
Ddoitsujin maintainer 2021-01-30 github

This sounds like a driver bug or unstable GPU more than anything else. Note that Cyberpunk is a D3D12 game and does not use DXVK.

WWebxorcist 2021-01-30 github

Ok thanks. I wasn't sure about it using DXVK.

For the driver, that is freedesktop.org I think?

Ddoitsujin maintainer 2021-01-30 github

yes.

WWebxorcist 2021-01-31 github

Ok thank you. Please close this ticket.

Proton versions