Arch Linux

I created a new profile to see what happens on default settings. Here is some recordings for LTS and stable kernels.

https://drive.google.com/drive/folders/19kBULZ3OKNISdoFBkmCF66uqq4vLwVI5

nouveau crash

Oct 19 09:14:24 arch kernel: ------------[ cut here ]------------
Oct 19 09:14:24 arch kernel: WARNING: CPU: 4 PID: 99 at drivers/gpu/drm/nouveau/nvif/vmm.c:68 nvif_vmm_put+0x73/0x80 [nouveau]
Oct 19 09:14:24 arch kernel: 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 intel_rapl_msr intel_rapl_common x86_pkg_temp_thermal intel_powerclamp coretemp kvm_intel snd_hda_codec_conexant snd_hda_codec_generic snd_hda_codec_hdmi kvm joydev ledtrig_audio snd_hda_intel snd_intel_dspcfg soundwire_intel soundwire_generic_allocation ath9k irqbypass soundwire_cadence ath9k_common nouveau crct10dif_pclmul snd_hda_codec ath9k_hw mxm_wmi crc32_pclmul ghash_clmulni_intel snd_hda_core i2c_algo_bit ath3k aesni_intel ath btusb uvcvideo btrtl snd_hwdep ttm soundwire_bus btbcm crypto_simd toshiba_acpi videobuf2_vmalloc btintel vfat iTCO_wdt cryptd sparse_keymap glue_helper intel_pmc_bxt videobuf2_memops mei_hdcp at24 fat iTCO_vendor_support wmi_bmof industrialio videobuf2_v4l2 lzo_rle mac80211 bluetooth drm_kms_helper snd_soc_core videobuf2_common rapl snd_compress intel_cstate intel_uncore cec videodev ac97_bus
Oct 19 09:14:24 arch kernel: snd_pcm_dmaengine ecdh_generic syscopyarea snd_pcm sysfillrect ecc mc mousedev sysimgblt cfg80211 psmouse crc16 fb_sys_fops snd_timer snd soundcore i2c_i801 mei_me libarc4 atl1c mei lpc_ich i2c_smbus mac_hid pcspkr toshiba_bluetooth video rfkill wmi pkcs8_key_parser crypto_user acpi_call(OE) drm fuse agpgart zram bpf_preload ip_tables x_tables xfs libcrc32c crc32c_generic hid_logitech_hidpp hid_logitech_dj usbhid serio_raw atkbd libps2 crc32c_intel ums_realtek sr_mod uas cdrom xhci_pci usb_storage xhci_pci_renesas i8042 serio
Oct 19 09:14:24 arch kernel: CPU: 4 PID: 99 Comm: kworker/4:1 Tainted: G OE 5.10.74-1-lts #1
Oct 19 09:14:24 arch kernel: Hardware name: TOSHIBA SATELLITE L750/Base Board Product Name, BIOS 3.60 06/26/2012
Oct 19 09:14:24 arch kernel: Workqueue: events nouveau_cli_work [nouveau]
Oct 19 09:14:24 arch kernel: RIP: 0010:nvif_vmm_put+0x73/0x80 [nouveau]
Oct 19 09:14:24 arch kernel: Code: 8b 06 48 89 f3 b9 10 00 00 00 48 89 e2 be 02 00 00 00 48 89 44 24 08 e8 8b e2 ff ff 85 c0 75 0a 48 c7 43 08 00 00 00 00 eb bf <0f> 0b eb f2 e8 54 f7 2c de 0f 1f 40 00 66 66 66 66 90 53 48 83 ec
Oct 19 09:14:24 arch kernel: RSP: 0018:ffffa77c002dfde0 EFLAGS: 00010282
Oct 19 09:14:24 arch kernel: RAX: 00000000fffffffe RBX: ffffa77c002dfe08 RCX: 0000000000000000
Oct 19 09:14:24 arch kernel: RDX: 0000000000000000 RSI: ffffa77c002dfd50 RDI: ffffa77c002dfdf0
Oct 19 09:14:24 arch kernel: RBP: ffff9c4469d9c900 R08: 0000000000000000 R09: 0000000000000000
Oct 19 09:14:24 arch kernel: R10: 0000000000000030 R11: 0000000000000018 R12: ffff9c4469da4b40
Oct 19 09:14:24 arch kernel: R13: dead000000000100 R14: ffff9c4440c8bb08 R15: ffff9c4469da4070
Oct 19 09:14:24 arch kernel: FS: 0000000000000000(0000) GS:ffff9c4587500000(0000) knlGS:0000000000000000
Oct 19 09:14:24 arch kernel: CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
Oct 19 09:14:24 arch kernel: CR2: 0000561926781eb4 CR3: 00000001216a4005 CR4: 00000000000606e0
Oct 19 09:14:24 arch kernel: Call Trace:
Oct 19 09:14:24 arch kernel: nouveau_vma_del+0x89/0xd0 [nouveau]
Oct 19 09:14:24 arch kernel: nouveau_gem_object_delete_work+0x36/0x60 [nouveau]
Oct 19 09:14:24 arch kernel: nouveau_cli_work+0xcc/0x120 [nouveau]
Oct 19 09:14:24 arch kernel: process_one_work+0x1b6/0x350
Oct 19 09:14:24 arch kernel: worker_thread+0x53/0x3e0
Oct 19 09:14:24 arch kernel: ? process_one_work+0x350/0x350
Oct 19 09:14:24 arch kernel: kthread+0x11b/0x140
Oct 19 09:14:24 arch kernel: ? kthread_associate_blkcg+0xa0/0xa0
Oct 19 09:14:24 arch kernel: ret_from_fork+0x22/0x30
Oct 19 09:14:24 arch kernel: ---[ end trace a9579ccf5de5fdb2 ]---

Please check with a different/older (-lts) kernel to make sure it's not drm related.

Still reproducible on current LTS, stable, zen, liquorix.
I have never managed to success for at least one year.

I will check older ones(starting from two years probably).

Anything i can do ?

kwin 5.23.2-1
xf86-video-nouveau 1.0.17-1

ask at #nouveau (OFTC)

Yeah i learned that it's a known bug/regression and someone is working on it, also some fixes have been pushed. Gonna wait for next kernel(s) to try.