Black screen on Nvidia after updating to 20260428

English Install/Boot/Login
,
81

Lovely. Thanks for the clarifications guys!

82

I have a feeling that itā€™s not 7.0 kernel specifically. It probably highlighted an old problem. https://bugzilla.opensuse.org/show_bug.cgi?id=1212841
I donā€™t know what specifically is causing it, but it seems something is wrong with tools that used to build Nvidia driver. I donā€™t see similar reports on Ubuntu 26.04 or CachyOS.

83

@rektal a mystery for sure. Iā€™m on a Turing GPU same as youā€¦

So from a system environment perspective my settings are;

  • Systemd Boot
  • NO Plymouth
  • Boot options intel_iommu=on fbdev=1 nvidia_drm.modeset=1 rd.driver.blacklist=nouveau loglevel=2
  • kernel-default 7.0.2-1.1
  • nvidia-open-driver-G07-signed-cuda-kmp-default 595.71.05_k7.0.2_1-1.2 and matching nvidia packages
  • Tumbleweed version 20260430

If your willing, as a test if you set booting to just a tty via login as root user (no sudo please) and run systemctl set-default multi-user.target update to the latest Nvidia driver, kernel, snapshot etc and reboot. Do you get back to the tty and can log in as root user and see the output from nvidia-smi etc.

If that does work, run systemctl isolate graphical.target does this work?

Edit: updated environmentā€¦

84

Ok sorry Im new to that and I want to know how does TW works and how to fix it
Thank you for helping me

85

I was asking only and I know many things in tech but I never dealed with any gpu problems and this forums Is to understand and solve problems
Anyway thank you helped me much

86

@Lioli7k The world has moved on since 2023 :wink:, the G06 drivers and hardware are considered legacy. G07+ and the open driver is the main focus.

1 Like
87

7.0.2 and G06 give me the low-resolution graphics. The updated G07 reverts to a blank screen.

88

Same for me. Blank screen but if I try to boot into 7.02 in recovery mode it gives me a pink screen instead of a blue screen of crash.

1 Like
89

@rektal if you also add the boot option enable.plymouth=0 does that make a difference? Itā€™s one thing I removeā€¦

90

I know and I am using open G07 on my system. Legacy or not, Iā€™m triaging a bug, so I will be testing them as well once in a while. So far Iā€™ve tested open G07, proprietary G07 and open G06. All of them are broken for me. Iā€™ve mentioned that issue because Stefan Dirsch thinks it might be related to those black screen shenanigans.

91

RTX4060 and wayland (X immediately crashed) here. 7.0.2 / G-07 open driver installed (my system is likely not on the latest kernel/nvidia) but did not dare to restart yet as there was a rather strange observation:
Had similar issues as others here: black screen (once even pink) after reboot (not even could change to another tty). Booted to previous kernel, but no nvidia driver. After another boot (no clue why it then worked) into 7.0.1 it kind of worked but kde reported repeated restart of desktop effects due restarted graphics. After some AI session and tweaking with nvidia-smi (limiting whatever the AI suggested), kde graphics setting and more, no improvement. Switched off all desktop effects and and set animation speed to immediate ā†’ no improvement. However, computer went into standby and after waking it up, everything seems to run normal (also some java app that was unuasable).
Not sure it runs as smooth as usual due to the desktop effects switched off, but if feels normal.
Reading all the comments here I am reluctant to dare a zypper dup or even a restart.

92

After several days of trying different Nvidia driver versions, the only result Iā€™ve achieved is that when KDE starts up, I get continuous error messages about desktop effects and freezes.

However, if I press the ESC key during system startup to view details, one of two things can happen:

  • The screen turns blue (green or even magenta), so I have to press CTRL+ALT+DEL to restart.

  • KDE starts without problems and everything runs smoothly.

I donā€™t understand why, but for now, it happens every time.

My system has a Ryzen 9800X3D and an RTX 5090, the Nvidia driver (nvidia-open-driver-G07-signed-cuda-kmp-default) is version 595.71.05 and the kernel is 7.0.2-1-default.

93

@Lioli7k So what parameters are set for the nvidia drivers in the permutations you have tried?

For example here I have;

cat /proc/driver/nvidia/params | sort

CoherentGPUMemoryMode: ""
CreateImexChannel0: 0
DeviceFileGID: 0
DeviceFileMode: 438
DeviceFileUID: 0
DmaRemapPeerMmio: 1
DynamicPowerManagement: 3
DynamicPowerManagementVideoMemoryThreshold: 200
EnableDbgBreakpoint: 0
EnableGpuFirmware: 18
EnableGpuFirmwareLogs: 2
EnableMSI: 1
EnablePCIeGen3: 0
EnablePCIERelaxedOrderingMode: 0
EnableResizableBar: 0
EnableS0ixPowerManagement: 1
EnableStreamMemOPs: 0
EnableSystemMemoryPools: 529
EnableUserNUMAManagement: 1
ExcludedGpus: ""
GpuBlacklist: ""
GrdmaPciTopoCheckOverride: 0
IgnoreMMIOCheck: 0
ImexChannelCount: 2048
InitializeSystemMemoryAllocations: 1
KMallocHeapMaxSize: 0
MemoryPoolSize: 0
ModifyDeviceFiles: 1
NvLinkDisable: 0
OpenRmEnableUnsupportedGpus: 1
PreserveVideoMemoryAllocations: 1
RegisterPCIDriver: 1
RegistryDwords: ""
RegistryDwordsPerDevice: ""
ResmanDebugLevel: 4294967295
RmLogonRC: 1
RmMsg: ""
RmNvlinkBandwidthLinkCount: 0
RmProfilingAdminOnly: 1
S0ixPowerManagementVideoMemoryThreshold: 256
TegraGpuPgMask: 4294967295
TemporaryFilePath: "/var/tmp"
UseKernelSuspendNotifiers: 0
UsePageAttributeTable: 4294967295
VMallocHeapMaxSize: 0
94

I have that option as a boot param, looking through the logs I did notice thisā€¦

May 03 18:40:55 hp-suse kernel: ------------[ cut here ]------------
May 03 18:40:55 hp-suse kernel: Unpatched return thunk in use. This should not happen!
May 03 18:40:55 hp-suse kernel: WARNING: arch/x86/kernel/cpu/bugs.c:3736 at __warn_thunk+0x10/0x20, CPU#2: (udev-worker)/998
May 03 18:40:55 hp-suse kernel: Modules linked in: snd_ump(+) rc_dtt200u videobuf2_vmalloc snd_soc_core(+) nvidia(O+) snd_hda_core realtek btintel pcspkr uvc hp_wmi acpi_cpufreq(-) snd_rawmidi i2c_i801 dvb_usb_dtt200u(+) efi_pst>
May 03 18:40:55 hp-suse kernel:  hid_generic scsi_dh_emc typec_ucsi scsi_dh_rdac roles sdhci scsi_dh_alua nvme_keyring typec sg cqhci nvme_auth ghash_clmulni_intel mxm_wmi aesni_intel hkdf scsi_mod mmc_core intel_lpss_pci thunde>
May 03 18:40:55 hp-suse kernel: CPU: 2 UID: 0 PID: 998 Comm: (udev-worker) Tainted: G           O        7.0.2-1-default #1 PREEMPT(full) openSUSE Tumbleweed  ab90b4c9940707f9cafa19bdad80b2cec52dbe51
May 03 18:40:55 hp-suse kernel: Tainted: [O]=OOT_MODULE
May 03 18:40:55 hp-suse kernel: Hardware name: HP OMEN by HP Laptop 17-cb1xxx/8748, BIOS F.10 06/04/2020
May 03 18:40:55 hp-suse kernel: RIP: 0010:__warn_thunk+0x10/0x20
May 03 18:40:55 hp-suse kernel: Code: 00 48 98 e9 92 69 e4 00 66 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 0f 1f 40 d6 0f 1f 44 00 00 48 8d 3d 80 4b 26 02 <67> 48 0f b9 3a c3 cc cc cc cc 66 0f 1f 44 00 00 90 90 90 90 90>
May 03 18:40:55 hp-suse kernel: RSP: 0018:ffffcf1dc1d47ab0 EFLAGS: 00010282
May 03 18:40:55 hp-suse kernel: RAX: 0000000000000000 RBX: 0000000000000000 RCX: 0000000000000000
May 03 18:40:55 hp-suse kernel: RDX: 0000000000000000 RSI: 0000000000000286 RDI: ffffffffa7944a70
May 03 18:40:55 hp-suse kernel: RBP: ffffcf1dc1d47b00 R08: 0000000000000286 R09: 0000000000000020
May 03 18:40:55 hp-suse kernel: R10: 0000000000000000 R11: 0000000000000000 R12: ffffffffc1c0b010
May 03 18:40:55 hp-suse kernel: R13: 0000000000000000 R14: 000055aa047cc510 R15: ffff8cf8f06fff00
May 03 18:40:55 hp-suse kernel: FS:  00007f17c0fca240(0000) GS:ffff8d0015658000(0000) knlGS:0000000000000000
May 03 18:40:55 hp-suse kernel: CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
May 03 18:40:55 hp-suse kernel: CR2: 00007feed2a90024 CR3: 0000000124996004 CR4: 00000000007706f0
May 03 18:40:55 hp-suse systemd-fsck[1101]: fsck.fat 4.2 (2021-01-31)
May 03 18:40:55 hp-suse systemd-fsck[1101]: /dev/nvme1n1p1: 16 files, 1121/63965 clusters
May 03 18:40:55 hp-suse kernel: PKRU: 55555554
May 03 18:40:55 hp-suse kernel: Call Trace:
May 03 18:40:55 hp-suse kernel:  <TASK>
May 03 18:40:55 hp-suse kernel:  warn_thunk_thunk+0x16/0x30
May 03 18:40:55 hp-suse kernel:  nvidia_init_module+0x2b/0x730 [nvidia 576e50faea88f7ca6014e024e2ab0ec9e0fdc9ab]
May 03 18:40:55 hp-suse kernel:  ? __SCT__tp_func_nvswitch_dev_sxid+0xa2fc/0xa2fc [nvidia 576e50faea88f7ca6014e024e2ab0ec9e0fdc9ab]
May 03 18:40:55 hp-suse kernel:  ? do_one_initcall+0x46/0x320
May 03 18:40:55 hp-suse kernel:  ? ima_post_read_file+0xe0/0x100
May 03 18:40:55 hp-suse kernel:  ? do_init_module+0x46/0x280
May 03 18:40:55 hp-suse kernel:  ? do_init_module+0x84/0x280
May 03 18:40:55 hp-suse kernel:  ? init_module_from_file+0xd8/0x140
May 03 18:40:55 hp-suse kernel:  ? init_module_from_file+0xd8/0x140
May 03 18:40:55 hp-suse kernel:  ? idempotent_init_module+0x100/0x2f0
May 03 18:40:55 hp-suse kernel:  ? __x64_sys_finit_module+0x71/0xe0
May 03 18:40:55 hp-suse kernel:  ? do_syscall_64+0x11c/0x600
May 03 18:40:55 hp-suse kernel:  ? switch_fpu_return+0x56/0xf0
May 03 18:40:55 hp-suse kernel:  ? do_syscall_64+0x2d6/0x600
May 03 18:40:55 hp-suse kernel:  ? __x64_sys_openat+0x61/0xa0
May 03 18:40:55 hp-suse kernel:  ? do_syscall_64+0x11c/0x600
May 03 18:40:55 hp-suse kernel:  ? switch_fpu_return+0x56/0xf0
May 03 18:40:55 hp-suse kernel:  ? do_syscall_64+0x2d6/0x600
May 03 18:40:55 hp-suse kernel:  ? do_syscall_64+0x11c/0x600
May 03 18:40:55 hp-suse kernel:  ? do_syscall_64+0x11c/0x600
May 03 18:40:55 hp-suse kernel:  ? clear_bhb_loop+0x50/0xa0
May 03 18:40:55 hp-suse kernel:  ? clear_bhb_loop+0x50/0xa0
May 03 18:40:55 hp-suse kernel:  ? entry_SYSCALL_64_after_hwframe+0x76/0x7e
May 03 18:40:55 hp-suse systemd[1]: Finished File System Check on /dev/disk/by-uuid/05A2-357A.
May 03 18:40:55 hp-suse kernel:  </TASK>
May 03 18:40:55 hp-suse kernel: ---[ end trace 0000000000000000 ]---
May 03 18:40:55 hp-suse kernel: nvidia-nvlink: Nvlink Core is being initialized, major device number 235
95

I even compared the setting files between my working systems and the not working one to find differences or any stale settings. I couldnā€™t find any difference.This issue is a quite tricky one.

@malcolmlewis To answer some questions/recommendations you had towards rektal:

  • the affected system does not have plymouth installed
  • nvidia_drm.modeset=1 is set by default via /usr/lib/modprobe.d/50-nvidia.conf which is part of nvidia-common-G0x
96

@FrankyU2 I see the same thunk warningā€¦ system still working fine.

1 Like
97

Itā€™s a known issue and itā€™s safe to ignore.

1 Like
98

@hui yes, but from memory there was a moment in time I added it for gdmā€¦ Iā€™m just showing what I have :wink:

99

Hmā€¦ Havenā€™t looked at them. I leave all on default. At the moment Iā€™m running on 6.19.12 and have them as following:

CoherentGPUMemoryMode: ""
CreateImexChannel0: 0
DeviceFileGID: 0
DeviceFileMode: 438
DeviceFileUID: 0
DmaRemapPeerMmio: 1
DynamicPowerManagement: 3
DynamicPowerManagementVideoMemoryThreshold: 200
EnableDbgBreakpoint: 0
EnableGpuFirmware: 18
EnableGpuFirmwareLogs: 2
EnableMSI: 1
EnablePCIeGen3: 0
EnablePCIERelaxedOrderingMode: 0
EnableResizableBar: 0
EnableS0ixPowerManagement: 1
EnableStreamMemOPs: 0
EnableSystemMemoryPools: 529
EnableUserNUMAManagement: 1
ExcludedGpus: ""
GpuBlacklist: ""
GrdmaPciTopoCheckOverride: 0
IgnoreMMIOCheck: 0
ImexChannelCount: 2048
InitializeSystemMemoryAllocations: 1
KMallocHeapMaxSize: 0
MemoryPoolSize: 0
ModifyDeviceFiles: 1
NvLinkDisable: 0
OpenRmEnableUnsupportedGpus: 1
PreserveVideoMemoryAllocations: 1
RegisterPCIDriver: 1
RegistryDwords: ""
RegistryDwordsPerDevice: ""
ResmanDebugLevel: 4294967295
RmLogonRC: 1
RmMsg: ""
RmNvlinkBandwidthLinkCount: 0
RmProfilingAdminOnly: 1
S0ixPowerManagementVideoMemoryThreshold: 256
TegraGpuPgMask: 4294967295
TemporaryFilePath: "/var/tmp"
UseKernelSuspendNotifiers: 0
UsePageAttributeTable: 4294967295
VMallocHeapMaxSize: 0
100

@Lioli7k so itā€™s the same :wink: