Warframe issues with VKD3D-PROTON-BIN 2.7

[Mel34]

Description of the problem

Since update to vkd3d-proton-bin 2.7 I'm having random game related GPU freezes/crashes. System remains responsive to keyboard input and it's possible to restart Xorg session and switch VT's. When the freeze happens, game display gets frozen in one frame. Audio continues to play normally. Swapping VTs and returning to Xorg session leads to garbled display. I've tried disabling the new internal "magic" disk cache to no avail, freezes still occured. Downgrading to a 2.6 stops the problem from reproducing.

DMESG contains the following:

[  619.605846] ------------[ cut here ]------------
[  619.605847] WARNING: CPU: 9 PID: 3254 at drivers/dma-buf/dma-resv.c:289 dma_resv_add_fence+0x120/0x1e0
[  619.605852] Modules linked in: snd_seq_dummy snd_hrtimer snd_seq snd_seq_device ccm algif_aead cbc des_generic libdes ecb algif_skcipher cmac md4 algif_hash af_alg vfat fat intel_rapl_msr intel_rapl_common iwlmvm edac_mce_amd snd_hda_codec_realtek kvm_amd snd_hda_codec_generic gigabyte_wmi wmi_bmof ledtrig_audio snd_hda_codec_hdmi kvm mac80211 irqbypass crct10dif_pclmul snd_hda_intel crc32_pclmul polyval_clmulni snd_intel_dspcfg libarc4 btusb snd_intel_sdw_acpi polyval_generic gf128mul btrtl snd_hda_codec ghash_clmulni_intel iwlwifi btbcm aesni_intel snd_hda_core btintel btmtk crypto_simd snd_hwdep cryptd snd_pcm pcspkr rapl k10temp bluetooth cfg80211 joydev snd_timer mousedev ecdh_generic snd ccp soundcore i2c_piix4 rfkill crc16 wmi tpm_crb tpm_tis tpm_tis_core tpm gpio_amdpt mac_hid rng_core gpio_generic acpi_cpufreq lzo_rle pkcs8_key_parser fuse zram ip_tables x_tables hid_logitech_hidpp hid_logitech_dj btrfs blake2b_generic libcrc32c crc32c_generic usbhid xor raid6_pq amdgpu nvme
[  619.605883]  nvme_core xhci_pci crc32c_intel xhci_pci_renesas nvme_common drm_ttm_helper ttm gpu_sched drm_buddy drm_display_helper cec
[  619.605887] CPU: 9 PID: 3254 Comm: Warframe.x64.ex Not tainted 6.0.6-arch1-1 #1 a46cc4b882cfc11c3bbb09d6a0fab3dcad53b5c2
[  619.605889] Hardware name: Gigabyte Technology Co., Ltd. B550 AORUS MASTER/B550 AORUS MASTER, BIOS F15e 07/20/2022
[  619.605890] RIP: 0010:dma_resv_add_fence+0x120/0x1e0
[  619.605891] Code: ff f0 0f c1 43 38 83 f8 01 0f 84 9e 00 00 00 85 c0 0f 8e a9 00 00 00 48 83 c4 08 5b 5d 41 5c 41 5d 41 5e 41 5f c3 cc cc cc cc <0f> 0b e9 3c ff ff ff 49 8b 76 08 48 8b 7b 28 49 8b 46 28 80 3e 00
[  619.605892] RSP: 0018:ffffaa6184e6fe00 EFLAGS: 00010246
[  619.605893] RAX: ffffffff9d3aa460 RBX: ffff905db88fc158 RCX: 0000000000000002
[  619.605894] RDX: 0000000000000003 RSI: ffff905e6032e840 RDI: ffff905e6032e878
[  619.605895] RBP: ffff905e5e6f4e00 R08: 00007fa2ac2ce620 R09: 0000000000000000
[  619.605895] R10: 0000000000000000 R11: 0000000000000001 R12: ffff905e6032e840
[  619.605896] R13: 0000000000000001 R14: ffff905e6032e840 R15: 0000000000000000
[  619.605896] FS:  000000001819f6c0(0000) GS:ffff90645ea40000(0000) knlGS:0000000011af0000
[  619.605897] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[  619.605898] CR2: 00007fa23c5d4000 CR3: 000000012a82e000 CR4: 0000000000750ee0
[  619.605899] PKRU: 55555554
[  619.605899] Call Trace:
[  619.605900]  <TASK>
[  619.605902]  dma_buf_ioctl+0x31d/0x3a0
[  619.605904]  __x64_sys_ioctl+0x94/0xd0
[  619.605906]  do_syscall_64+0x5f/0x90
[  619.605909]  ? syscall_exit_to_user_mode+0x1b/0x40
[  619.605910]  ? do_syscall_64+0x6b/0x90
[  619.605911]  ? exc_page_fault+0x74/0x170
[  619.605912]  entry_SYSCALL_64_after_hwframe+0x63/0xcd
[  619.605914] RIP: 0033:0x7fa2ac414c0f
[  619.605915] Code: 00 48 89 44 24 18 31 c0 48 8d 44 24 60 c7 04 24 10 00 00 00 48 89 44 24 08 48 8d 44 24 20 48 89 44 24 10 b8 10 00 00 00 0f 05 <89> c2 3d 00 f0 ff ff 77 18 48 8b 44 24 18 64 48 2b 04 25 28 00 00
[  619.605916] RSP: 002b:000000001819e6f0 EFLAGS: 00000246 ORIG_RAX: 0000000000000010
[  619.605917] RAX: ffffffffffffffda RBX: 00007fa210041ac0 RCX: 00007fa2ac414c0f
[  619.605917] RDX: 000000001819e780 RSI: 0000000040086203 RDI: 0000000000000080
[  619.605918] RBP: 000000001819e780 R08: 00007fa2a9e1d258 R09: 00007fa210054170
[  619.605918] R10: 00007fa2ac2ce620 R11: 0000000000000246 R12: 0000000040086203
[  619.605919] R13: 0000000000000080 R14: 0000000000000000 R15: 0000000000000000
[  619.605920]  </TASK>
[  619.605921] ---[ end trace 0000000000000000 ]---

How to reproduce the issue:

Launch warframe, enter Elite sanctuary onslaught game mode in a public match, proceed clearing zones and entering portal to the next zone until you get a frozen game. For me it's usually taking the portal to the 3rd or the 4th zone.

System information:

• Distro name and Version: Archlinux (testing repos enabled)
• Kernel version: Linux bababool 6.0.6-arch1-1 # 1 SMP PREEMPT_DYNAMIC Sat, 29 Oct 2022 14:08:39 +0000 x86_64 GNU/Linux
• Custom kernel: N/A
• AMD official driver version: N/A

Hardware description:

• CPU: AMD 5600X
• GPU: 07:00.0 VGA compatible controller [0300]: Advanced Micro Devices, Inc. [AMD/ATI] Navi 22 [Radeon RX 6700/6700 XT/6750 XT / 6800M] [1002:73df] (rev c5)
• System Memory: 32 GB ddr4 @ 3200 MHz
• Display(s): LG 27" 144Hz
• Type of Display Connection: DP

Log files

• dmesg.log
• journal.log
• Xorg.0.log

Visuals

• Game freeze

• Glitchy Xorg after changing VTs

[ms178]

I also see something similar in dmesg with Sniper Elite 5 (Vulkan) and get frequent crashes with the same screen output as seen in the screenshot with Battlefield 1 (DX12). This is on a Vega 56 with a GCC 12.2.1 compiled Mesa-git.

This is with Sniper Elite 5:

[ 5. Nov 09:55] ------------[ cut here ]------------
[  +0,000003] WARNING: CPU: 2 PID: 2854 at drivers/dma-buf/dma-resv.c:289 0xffffffff9dc4b51f
[  +0,000004] Modules linked in: intel_rapl_msr mei_wdt intel_rapl_common sb_edac x86_pkg_temp_thermal intel_powerclamp coretemp crct10dif_pclmul crc32_pclmul ghash_clmulni_in>
[  +0,000032] CPU: 2 PID: 2854 Comm: sniper5_vulkan. Tainted: G           O       6.0.7-1-cachyos-bore #1 f7adbff77bb9b8276d2fceb4a67f84cf3b674ef3
[  +0,000003] Hardware name: LENOVO GAMING TF/X99-TF Gaming, BIOS CX99DE26 10/10/2020
[  +0,000001] RIP: 0010:0xffffffff9dc4b51f
[  +0,000002] Code: 83 fe 01 0f 84 d4 00 00 00 45 85 f6 0f 8f 03 ff ff ff 48 83 c4 08 be 03 00 00 00 5b 5d 41 5c 41 5d 41 5e 41 5f e9 81 b6 c5 ff <0f> 0b e9 35 fe ff ff 4d 8b >
[  +0,000001] RSP: 0018:ffff9c7c4c98fe98 EFLAGS: 00010246
[  +0,000002] RAX: 0000000000000002 RBX: ffff8b3628f95958 RCX: ffffffff9e728580
[  +0,000001] RDX: 0000000000000003 RSI: ffff8b37b061bbc0 RDI: ffff8b37b061bbf8
[  +0,000001] RBP: ffff8b35a2df4c00 R08: ffff8b3787e59618 R09: 0000000000000002
[  +0,000001] R10: ffff8b361366f6c0 R11: ffff8b3495241400 R12: ffff8b37b061bbc0
[  +0,000000] R13: 0000000000000001 R14: ffffffff9dc47a20 R15: ffff8b37b061bbc0
[  +0,000001] FS:  00007f96940a5f00(0000) GS:ffff8b42bf680000(0000) knlGS:0000000067fe0000
[  +0,000002] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[  +0,000000] CR2: 00007f968fc06000 CR3: 0000000b362d6003 CR4: 00000000001706e0
[  +0,000002] Call Trace:
[  +0,000002]  <TASK>
[  +0,000000]  ? 0xffffffff9dc47a20
[  +0,000001]  ? 0xffffffff9dc47d51
[  +0,000001]  ? 0xffffffff9d72036f
[  +0,000000]  ? 0xffffffff9e0320eb
[  +0,000001]  ? 0xffffffff9e20006a
[  +0,000001]  </TASK>
[  +0,000001] ---[ end trace 0000000000000000 ]---

These are seen after Battlefield 1 crashed:

[EDIT:] I've exchanged the second screenshot with a more recent crash with a DEBUG-Kernel with more meaningful output.

[Astrobald]

Same issue on Watch Dogs Legion:

lutris-GE-Proton7-33 + vkd3d-proton-git + mesa-tkg-git Ryzen 5 5600X + RX 5500XT + B550 AORUS ELITE V2

[   50.737589] WARNING: CPU: 3 PID: 2130 at drivers/dma-buf/dma-resv.c:289 dma_resv_add_fence+0x11d/0x1e0
[   50.737594] Modules linked in: snd_seq_dummy snd_seq_midi snd_hrtimer snd_seq_midi_event snd_seq joydev mousedev rfkill qrtr zram vmnet(OE) snd_usb_audio snd_usbmidi_lib snd_rawmidi snd_seq_device mc usbhid vfat fat intel_rapl_msr intel_rapl_common amdgpu snd_hda_codec_realtek kvm_amd snd_hda_codec_generic wmi_bmof gigabyte_wmi ledtrig_audio snd_hda_codec_hdmi kvm snd_hda_intel ipt_REJECT snd_intel_dspcfg irqbypass nf_reject_ipv4 crct10dif_pclmul snd_hda_codec xt_limit crc32_pclmul ghash_clmulni_intel snd_hwdep gpu_sched xt_addrtype aesni_intel snd_hda_core drm_buddy r8169 crypto_simd snd_pcm drm_ttm_helper cryptd xt_tcpudp ttm realtek snd_timer ccp rapl mdio_devres drm_display_helper snd xt_conntrack k10temp libphy rng_core cec i2c_piix4 soundcore nf_conntrack nf_defrag_ipv6 gpio_amdpt nf_defrag_ipv4 gpio_generic libcrc32c wmi mac_hid ip6table_filter ip6_tables iptable_filter sch_cake vmmon(OE) vmw_vmci uinput crypto_user fuse bpf_preload ip_tables x_tables ext4 crc32c_generic crc16
[   50.737631]  mbcache jbd2 nvme nvme_core crc32c_intel xhci_pci nvme_common xhci_pci_renesas
[   50.737635] CPU: 3 PID: 2130 Comm: WatchDogsLegion Tainted: G        W  OE      6.0.7-271-tkg-tt #1 a0af0d186dd6c5424c9ead5dd5be361e36bb80d5
[   50.737637] Hardware name: Gigabyte Technology Co., Ltd. B550 AORUS ELITE V2/B550 AORUS ELITE V2, BIOS F15d 07/20/2022
[   50.737638] RIP: 0010:dma_resv_add_fence+0x11d/0x1e0
[   50.737640] Code: ff f0 0f c1 43 38 83 f8 01 0f 84 a5 00 00 00 85 c0 0f 8e b0 00 00 00 48 83 c4 08 5b 5d 41 5c 41 5d 41 5e 41 5f c3 cc cc cc cc <0f> 0b e9 3f ff ff ff 49 8b 75 08 48 8b 7b 28 49 8b 45 28 80 3e 00
[   50.737641] RSP: 0018:ffffb0a5c5707e70 EFLAGS: 00010246
[   50.737643] RAX: ffffffffa538a420 RBX: ffff8f80fbaf2158 RCX: 0000000000000002
[   50.737643] RDX: 0000000000000003 RSI: ffff8f80a84256c0 RDI: ffff8f80a84256f8
[   50.737644] RBP: ffff8f80a84a6c00 R08: 000000002d06dd00 R09: 0000000000000000
[   50.737645] R10: 0000000000000000 R11: 0000000000000001 R12: ffff8f80a84256c0
[   50.737646] R13: ffff8f80a84256c0 R14: ffff8f7fc362ce01 R15: 0000000000000000
[   50.737647] FS:  000000002d06f6c0(0000) GS:ffff8f86deec0000(0000) knlGS:0000000009070000
[   50.737648] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[   50.737649] CR2: 00007f4c609fb000 CR3: 000000019604c000 CR4: 0000000000750ee0
[   50.737650] DR0: 000000007ffe0288 DR1: 000000007ffe02e8 DR2: 0000000000000000
[   50.737650] DR3: 0000000000000000 DR6: 00000000ffff0ff0 DR7: 00000000ffff06aa
[   50.737651] PKRU: 55555554
[   50.737652] Call Trace:
[   50.737653]  <TASK>
[   50.737654]  dma_buf_ioctl+0x319/0x3a0
[   50.737656]  __x64_sys_ioctl+0x91/0xd0
[   50.737659]  do_syscall_64+0x5c/0x90
[   50.737661]  entry_SYSCALL_64_after_hwframe+0x63/0xcd
[   50.737663] RIP: 0033:0x7f4c64264c0f
[   50.737665] Code: 00 48 89 44 24 18 31 c0 48 8d 44 24 60 c7 04 24 10 00 00 00 48 89 44 24 08 48 8d 44 24 20 48 89 44 24 10 b8 10 00 00 00 0f 05 <89> c2 3d 00 f0 ff ff 77 18 48 8b 44 24 18 64 48 2b 04 25 28 00 00
[   50.737666] RSP: 002b:000000002cf6f080 EFLAGS: 00000246 ORIG_RAX: 0000000000000010
[   50.737667] RAX: ffffffffffffffda RBX: 00007f4a78096df0 RCX: 00007f4c64264c0f
[   50.737668] RDX: 000000002cf6f110 RSI: 0000000040086203 RDI: 0000000000000021
[   50.737669] RBP: 000000002cf6f110 R08: 00007f4aaba8e350 R09: 00007f4a780a9ea0
[   50.737669] R10: 000000002d06dd00 R11: 0000000000000246 R12: 0000000040086203
[   50.737670] R13: 0000000000000021 R14: 0000000000000000 R15: 0000000000000000
[   50.737672]  </TASK>
[   50.737672] ---[ end trace 0000000000000000 ]---
[   63.305998] ------------[ cut here ]------------

system-info.txt

[K0bin]

Please open separate issues. While at first glance it seems like it's related, different games do different things.

[Astrobald]

Just a note to say that the kernel error occurs when I deactivate the V-Sync graphics option in the game or when I launch the game without the V-Sync graphics option activated. This also happens when gamemode and gamescope are disabled. Thank you!

https://youtu.be/vMNfXdLPdSk

WatchDogsLegion-kernel-error.log

Edit: The bug also appears when I switch from windowed mode to full screen mode with the V-Sync graphics option activated. kernel v6.0.7 --> kernel bug in any case kernel v5.19.17 --> no bug at all

[Blisto91]

@Mel34 Hi there. Are you still having these issues?

[Mel34]

I've gone back to using DX11 (dxvk) so can't say if it's still reproducible on my machine. Sorry for the late comment.

[AyAyEm]

Not sure if it's related but i had a similar issue in the new update, whispers in the wall

[90311.659652] amdgpu 0000:09:00.0: amdgpu: [gfxhub] page fault (src_id:0 ring:24 vmid:6 pasid:32792, for process Warframe.x64.ex pid 574727 thread vkd3d_queue pid 574826)
[90311.659661] amdgpu 0000:09:00.0: amdgpu:   in page starting at address 0x0000800001568000 from client 0x1b (UTCL2)
[90311.659665] amdgpu 0000:09:00.0: amdgpu: GCVM_L2_PROTECTION_FAULT_STATUS:0x00601430
[90311.659667] amdgpu 0000:09:00.0: amdgpu:      Faulty UTCL2 client ID: SQC (data) (0xa)
[90311.659670] amdgpu 0000:09:00.0: amdgpu:      MORE_FAULTS: 0x0
[90311.659672] amdgpu 0000:09:00.0: amdgpu:      WALKER_ERROR: 0x0
[90311.659674] amdgpu 0000:09:00.0: amdgpu:      PERMISSION_FAULTS: 0x3
[90311.659675] amdgpu 0000:09:00.0: amdgpu:      MAPPING_ERROR: 0x0
[90311.659677] amdgpu 0000:09:00.0: amdgpu:      RW: 0x0
[90322.098558] [drm:amdgpu_job_timedout [amdgpu]] *ERROR* ring gfx_0.0.0 timeout, signaled seq=33605949, emitted seq=33605951
[90322.098901] [drm:amdgpu_job_timedout [amdgpu]] *ERROR* Process information: process Warframe.x64.ex pid 574727 thread vkd3d_queue pid 574826

it froze my screen and then it just crashed my xorg session (nothing unusual in the log)

• cpu: Ryzen 5 1600
• gpu: 09:00.0 VGA compatible controller: Advanced Micro Devices, Inc. [AMD/ATI] Navi 23 [Radeon RX 6600/6600 XT/6600M] (rev c7)
• kernel: 6.6.6-zen1-1-zen x86_64 GNU/Linux
• os: Garuda KDE (Arch Linux)
• display server: xorg
• proton: GE-Proton 8-25
• mesa: 1:23.2.1-2
• vkd3d: 7531dd5