[Nouveau] frequent frozen desktops since upgrade to kernel 5.9

Thu Jan 7 22:15:47 UTC 2021

On Thu, 2021-01-07 at 22:04 +0100, Arsene Wald wrote:
> I've been using my 9 year old machine without issues up to Fedora 32 
> with kernel 5.8. Some weeks ago I upgraded to kernel 5.9. After working 
> for a random duration, my Display becomes mostly frozen, showing 
> noticeable flickering and the mouse reacts extremely slowly if at all.
> Ctrl+Alt+t still opens a terminal but typing is a pain.
> ssh from a different computer is possible with acceptable reaction times.
> Ctrl+Alt+Del logs out session. A reLogin is possible with normal 
> reaction times, but again, only for an unforeseeable time!
> I upgraded to F33, but the issue persisted. Then I did a fresh F33 
> install, from scratch. Unfortunately the unpredictable freezes occur again.
> 
> During an unresponsive period,
> - "top" command shows process "kworker/u32:2-events_unbound" has high 
> CPU usage.
> - "echo l > /proc/sysrq-trigger" followed by "dmesg" shows the 
> following, which hints me to the nouveau module:
> 
> [  890.261108] nouveau 0000:01:00.0: DRM: base-0: timeout
> [  892.261335] ------------[ cut here ]------------
> [  892.261392] WARNING: CPU: 1 PID: 133 at 
> drivers/gpu/drm/nouveau/dispnv50/disp.c:211 nv50_dmac_wait+0x1bd/0x210 
> [nouveau]
> [  892.261393] Modules linked in: xt_CHECKSUM xt_MASQUERADE xt_conntrack 
> ipt_REJECT nf_nat_tftp nf_conntrack_tftp tun bridge stp llc nft_objref 
> nf_conntrack_netbios_ns nf_conntrack_broadcast 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 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 rfkill ip_set nf_tables nfnetlink ip6table_filter 
> ip6_tables iptable_filter vboxnetadp(OE) vboxnetflt(OE) vboxdrv(OE) 
> sunrpc intel_powerclamp iTCO_wdt intel_pmc_bxt coretemp 
> iTCO_vendor_support gpio_ich kvm_intel ppdev pktcdvd kvm irqbypass 
> intel_cstate intel_uncore pcspkr i2c_i801 i2c_smbus snd_hda_codec_via 
> snd_hda_codec_generic snd_hda_codec_hdmi ledtrig_audio snd_hda_intel 
> parport_pc snd_intel_dspcfg parport snd_hda_codec lpc_ich snd_hda_core 
> snd_hwdep snd_seq snd_seq_device snd_pcm asus_atk0110 snd_timer snd
> [  892.261414]  soundcore acpi_cpufreq zram ip_tables nouveau video 
> mxm_wmi wmi i2c_algo_bit drm_kms_helper cec ttm crct10dif_pclmul 
> crc32_pclmul crc32c_intel drm uas ghash_clmulni_intel serio_raw 
> usb_storage r8169 fuse
> [  892.261424] CPU: 1 PID: 133 Comm: kworker/u32:5 Tainted: G        W 
> OE     5.9.16-200.fc33.x86_64 #1
> [  892.261425] Hardware name: System manufacturer System Product 
> Name/P7P55D, BIOS 2101    10/20/2011
> [  892.261458] Workqueue: events_unbound nv50_disp_atomic_commit_work 
> [nouveau]
> [  892.261491] RIP: 0010:nv50_dmac_wait+0x1bd/0x210 [nouveau]
> [  892.261493] Code: 8d 48 04 48 89 4a 68 c7 00 00 00 00 20 49 8b 46 38 
> 41 c7 86 20 01 00 00 00 00 00 00 49 89 46 68 e8 28 fd ff ff e9 8a fe ff 
> ff <0f> 0b 48 83 c4 28 b8 92 ff ff ff 5b 5d 41 5c 41 5d 41 5e c3 49 8b
> [  892.261493] RSP: 0018:ffffb9d7402b7d70 EFLAGS: 00010282
> [  892.261495] RAX: ffffffffffffff92 RBX: 0000000000000003 RCX: 
> 0000000000000000
> [  892.261495] RDX: ffffffffffffff92 RSI: ffffb9d7402b7cb8 RDI: 
> ffffb9d7402b7d58
> [  892.261496] RBP: ffffb9d7402b7d70 R08: 0000000000000000 R09: 
> ffffb9d7402b7c88
> [  892.261496] R10: 0000000000000030 R11: 0000000000000002 R12: 
> 00000000fffffffb
> [  892.261497] R13: ffff9db9cfef4b68 R14: ffff9db9cfef4ba8 R15: 
> 0000000000000000
> [  892.261498] FS:  0000000000000000(0000) GS:ffff9db9d7a40000(0000) 
> knlGS:0000000000000000
> [  892.261499] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
> [  892.261500] CR2: 00007f73aa54b000 CR3: 0000000172a0e004 CR4: 
> 00000000000206e0
> [  892.261501] Call Trace:
> [  892.261539]  base507c_ntfy_set+0x2f/0x90 [nouveau]
> [  892.261571]  nv50_wndw_flush_set+0x68/0x1c0 [nouveau]
> [  892.261604]  nv50_disp_atomic_commit_tail+0x4cd/0x790 [nouveau]
> [  892.261609]  ? __switch_to+0x7f/0x470
> [  892.261612]  process_one_work+0x1b4/0x370
> [  892.261613]  worker_thread+0x53/0x3e0
> [  892.261615]  ? process_one_work+0x370/0x370
> [  892.261616]  kthread+0x11b/0x140
> [  892.261617]  ? __kthread_bind_mask+0x60/0x60
> [  892.261620]  ret_from_fork+0x22/0x30
> [  892.261622] ---[ end trace b96a5d9f7e4edcca ]---
> 
> # lspci | grep VGA
> 01:00.0 VGA compatible controller: NVIDIA Corporation GF108 [GeForce GT 
> 430] (rev a1)
> => connected with dual-link DVI to a 4K Display (3840x2160 resolution)
> 
> Although less frequently, the same problem also appears on my (old) 
> laptop since upgrading to kernel 5.9 (running Manjaro with Kernel 
> 5.9.11-3 and gaphics card: NVIDIA Corporation G96GLM [Quadro FX 770M] at 
> resolution 1920x1200).
> 
> Any ideas how I can (help to) solve this problem?

If you could bisect where the issue started, that'd be a good start.

> Arsène
> _______________________________________________
> Nouveau mailing list
> Nouveau at lists.freedesktop.org
> https://lists.freedesktop.org/mailman/listinfo/nouveau

-- 
Cheers,
 Lyude Paul (she/her)
 Software Engineer at Red Hat