Arch Linux

emil.s

Member

Registered: 2011-05-22

Posts: 11

Corrupt/artifacting screens after suspend [Wayland | Nvidia | Sddm]

Hello!

I have a very strange issue on my workstation. After starting from suspend, my screen always have some kind of corruption or artifact.

Example:
https://imgur.com/a/64aAs17

Sometimes both screen are artifacting, and sometimes it's just some windows. But there is always something that is broken, meaning that suspend is totally useless for me.
If I switch to the virtual console and restart sddm, everything is back to normal.

The problem is that I don't even know where to start troubleshooting. Is it a driver issue, displaymanager issue or a wayland issue?

Any logs that are of particular interest? (Nothing really obvious in dmesg).

Package versions:
nvidia 555.58.02-17
wayland 1.23.0-1
kwayland 6.1.4-1
sddm 0.21.0-4
linux 6.10.6.arch1-1

Any idea about what could cause this?

Best regards

seth

Member

Registered: 2012-09-03

Posts: 60,787

Re: Corrupt/artifacting screens after suspend [Wayland | Nvidia | Sddm]

https://wiki.archlinux.org/title/NVIDIA … er_suspend

---

How to upload text · How to boot w/o GUI · Disable Windows Fast-Start! · Fix your xinitrc

emil.s

Member

Registered: 2011-05-22

Posts: 11

Re: Corrupt/artifacting screens after suspend [Wayland | Nvidia | Sddm]

https://wiki.archlinux.org/title/NVIDIA … er_suspend

Ah, thanks!

Yeah we are absolutely onto something.

root@ThinkStation: /home/emil #> cat /etc/modprobe.d/nvidia.conf 
options nvidia NVreg_PreserveVideoMemoryAllocations=1

Both services are loaded:

root@ThinkStation: /home/emil #> systemctl status nvidia-suspend.service nvidia-resume.service 
○ nvidia-suspend.service - NVIDIA system suspend actions
     Loaded: loaded (/usr/lib/systemd/system/nvidia-suspend.service; enabled; preset: disabled)
     Active: inactive (dead)

○ nvidia-resume.service - NVIDIA system resume actions
     Loaded: loaded (/usr/lib/systemd/system/nvidia-resume.service; enabled; preset: disabled)
     Active: inactive (dead)

And I'm getting some results. However, not the desired ones...

After suspend, the computer automatically tries to resume after just a few seconds. But this time the screens are just black and it's totally unresponsive. (Can't even SSH login).

And I see the following in the journal:

aug 22 22:28:49 ThinkStation systemd-logind[743]: The system will suspend now!
....
aug 22 22:28:50 ThinkStation systemd[1]: Reached target Sleep.
aug 22 22:28:50 ThinkStation systemd[1]: Starting NVIDIA system suspend actions...
aug 22 22:28:50 ThinkStation suspend[3473]: nvidia-suspend.service
aug 22 22:28:50 ThinkStation logger[3473]: <13>Aug 22 22:28:50 suspend: nvidia-suspend.service
aug 22 22:28:50 ThinkStation kwin_wayland[914]: kwin_wayland_drm: Presentation failed! Permission denied
aug 22 22:28:50 ThinkStation kwin_wayland[914]: kwin_wayland_drm: Presentation failed! Permission denied
aug 22 22:28:50 ThinkStation kernel: snd_hda_codec_hdmi hdaudioC2D0: HDMI: invalid ELD data byte 71
aug 22 22:28:50 ThinkStation kernel: ------------[ cut here ]------------
aug 22 22:28:50 ThinkStation kernel: WARNING: CPU: 9 PID: 3475 at include/linux/rwsem.h:80 follow_pte+0x1de/0x200
aug 22 22:28:50 ThinkStation kernel: Modules linked in: rfkill lm92 intel_rapl_msr intel_rapl_common intel_uncore_frequency intel_uncore_frequency_common sb_edac x86_pkg_temp_thermal intel_powerclamp coretemp kvm_intel kvm snd_hda_codec_realtek snd_hda_codec_generic crct10dif_pclmul snd_hda_scodec_component snd_hda_codec_hdmi crc32_pclmul polyval_clmulni polyval_generic gf128mul uvcvideo snd_hda_intel raid456 ghash_clmulni_intel videobuf2_vmalloc vboxnetflt(OE) sha512_ssse3 nls_iso8859_1 snd_intel_dspcfg nvidia_drm(POE) vboxnetadp(OE) snd_usb_audio vfat async_raid6_recov uvc sha1_ssse3 snd_intel_sdw_acpi nvidia_modeset(POE) fat videobuf2_memops aesni_intel async_memcpy cdc_ether snd_usbmidi_lib vboxdrv(OE) snd_hda_codec iTCO_wdt async_pq videobuf2_v4l2 crypto_simd snd_ump video intel_pmc_bxt usbnet async_xor ee1004 pkcs8_key_parser snd_hda_core snd_rawmidi cryptd mei_wdt iTCO_vendor_support async_tx videodev snd_seq_device rapl ixgbe snd_hwdep r8152 videobuf2_common i2c_i801 think_lmi nvidia_uvm(POE) snd_pcm mii i2c_smbus intel_cstate
aug 22 22:28:50 ThinkStation kernel:  mdio_devres rtsx_usb_ms snd_timer memstick intel_uncore libphy intel_wmi_thunderbolt mc firmware_attributes_class wmi_bmof pcspkr intel_pch_thermal i2c_mux mdio e1000e snd mei_me lpc_ich ptp soundcore mei pps_core dca mousedev joydev md_mod mac_hid nvidia(POE) i2c_dev crypto_user loop dm_mod nfnetlink ip_tables x_tables rtsx_usb_sdmmc hid_generic mmc_core usbhid rtsx_usb btrfs blake2b_generic libcrc32c crc32c_generic xor raid6_pq mxm_wmi nvme crc32c_intel sha256_ssse3 sr_mod ata_generic nvme_core cdrom xhci_pci pata_acpi xhci_pci_renesas nvme_auth wmi
aug 22 22:28:50 ThinkStation kernel: CPU: 9 PID: 3475 Comm: nvidia-sleep.sh Tainted: P           OE      6.10.6-arch1-1 #1 703d152c24f1971e36f16e505405e456fc9e23f8
aug 22 22:28:50 ThinkStation kernel: Hardware name: LENOVO 30B4S01W00/102F, BIOS S00KT73A 05/24/2022
aug 22 22:28:50 ThinkStation kernel: RIP: 0010:follow_pte+0x1de/0x200
aug 22 22:28:50 ThinkStation kernel: Code: cc cc cc 48 81 e2 00 00 00 c0 48 09 c2 48 f7 d2 48 85 fa 75 20 e8 b2 f5 ff ff 48 8b 35 6b f1 5c 01 48 81 e6 00 00 00 c0 eb 8d <0f> 0b 48 3b 1f 0f 83 50 fe ff ff bd ea ff ff ff eb b6 49 8b 3c 24
aug 22 22:28:50 ThinkStation kernel: RSP: 0018:ffffbc3804753a70 EFLAGS: 00010246
aug 22 22:28:50 ThinkStation kernel: RAX: 0000000000000000 RBX: 000076ca89100000 RCX: ffffbc3804753ab0
aug 22 22:28:50 ThinkStation kernel: RDX: ffffbc3804753aa8 RSI: 000076ca89100000 RDI: ffff9bf2b62ae450
aug 22 22:28:50 ThinkStation kernel: RBP: ffffbc3804753af0 R08: ffffbc3804753c48 R09: 0000000000000000
aug 22 22:28:50 ThinkStation kernel: R10: ffff9bf1d934b000 R11: ffffffffc39a0080 R12: ffffbc3804753ab0
aug 22 22:28:50 ThinkStation kernel: R13: ffffbc3804753aa8 R14: ffff9bf08b4e2100 R15: 0000000000000000
aug 22 22:28:50 ThinkStation kernel: FS:  000072adf1e58b80(0000) GS:ffff9bf9d7c80000(0000) knlGS:0000000000000000
aug 22 22:28:50 ThinkStation kernel: CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
aug 22 22:28:50 ThinkStation kernel: CR2: 000020a0128f34a0 CR3: 0000000336766002 CR4: 00000000003706f0
aug 22 22:28:50 ThinkStation kernel: DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
aug 22 22:28:50 ThinkStation kernel: DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
aug 22 22:28:50 ThinkStation kernel: Call Trace:
....

And then about 10k lines of crash dumps. :-|

There is some "kwin_wayland_drm: Presentation failed! Permission denied" message just before it crashes, not sure if it's related?

seth

Member

Registered: 2012-09-03

Posts: 60,787

Re: Corrupt/artifacting screens after suspend [Wayland | Nvidia | Sddm]

aug 22 22:28:50 ThinkStation kernel: Call Trace:
....

You cut off the most interesting part…
The 550xx and 555xx drivers are frequently crahing, S3 cycles seem to be a trigger: https://bbs.archlinux.org/viewtopic.php?id=293400&p=4
There's an inofficial 550xx driver claimed by nvidia to fix this and you can otherwise try the 535xx drivers or nvidia-open (which might still have general issues w/ S3)

If you've nvidia in the initramfs, make sure to have regenerated that after editing modprobe (though the services are supposed to complain if the parameter isn't set)

---

How to upload text · How to boot w/o GUI · Disable Windows Fast-Start! · Fix your xinitrc