I used to run 535.86.05 on Fedora 38 6.5.11 which work flawlessly. From Linux 6.6 on this kernel stopped working.
So I upgraded my system to Fedora 39 running on Linux 6.6.9.
I tried to reinstall the driver, but it fails at building. So instead I downloaded the newest version 35.146.02, which compiles just fine. As before, the installation complains that the DKMS registration failed as it already exists. This occured before and was fine.
At the end of the installation though, it fails due to “the “nvidia-drm” kernel module not being loaded”.
The system at first failed at booting but now seems to boot and run the driver just fine.
The tail of the log file looks like this:
executing: '/usr/bin/chcon -t textrel_shlib_t /usr/lib/libnvidia-fbc.so.535.146.02'...
executing: '/usr/bin/chcon -t textrel_shlib_t /usr/lib/libnvidia-allocator.so.535.146.02'...
executing: '/usr/bin/chcon -t textrel_shlib_t /usr/lib/libnvidia-opticalflow.so.535.146.02'...
executing: '/usr/sbin/ldconfig'...
executing: '/usr/sbin/depmod -a '...
executing: '/usr/bin/systemctl daemon-reload'...
-> done.
-> Driver file installation is complete.
ERROR: Unable to load the 'nvidia-drm' kernel module.
-> Kernel messages:
[ 1328.848569] nvidia 0000:01:00.0: vgaarb: VGA decodes changed: olddecodes=none,decodes=none:owns=io+mem
[ 1329.048450] NVRM: loading NVIDIA UNIX x86_64 Kernel Module 535.146.02 Sun Dec 3 14:06:14 UTC 2023
[ 1329.066733] nvidia_uvm: module uses symbols nvUvmInterfaceDisableAccessCntr from proprietary module nvidia, inheriting taint.
[ 1329.219598] nvidia-uvm: Loaded the UVM driver, major device number 511.
[ 1329.234869] nvidia-modeset: Loading NVIDIA Kernel Mode Setting Driver for UNIX platforms 535.146.02 Sun Dec 3 14:02:44 UTC 2023
[ 1329.237308] [drm] [nvidia-drm] [GPU ID 0x00000100] Loading driver
[ 1329.237310] [drm] Initialized nvidia-drm 0.0.0 20160202 for 0000:01:00.0 on minor 0
[ 1329.242977] [drm] [nvidia-drm] [GPU ID 0x00000100] Unloading driver
[ 1329.254323] nvidia-modeset: Unloading
[ 1329.276914] nvidia-uvm: Unloaded the UVM driver.
[ 1329.303869] nvidia-nvlink: Unregistered Nvlink Core, major device number 235
[ 1760.689234] VFIO - User Level meta-driver version: 0.3
[ 1761.179099] nvidia-nvlink: Nvlink Core is being initialized, major device number 235
[ 1761.180361] nvidia 0000:01:00.0: vgaarb: VGA decodes changed: olddecodes=none,decodes=none:owns=io+mem
[ 1761.381178] NVRM: loading NVIDIA UNIX x86_64 Kernel Module 535.146.02 Sun Dec 3 14:06:14 UTC 2023
[ 1761.407061] nvidia_uvm: module uses symbols nvUvmInterfaceDisableAccessCntr from proprietary module nvidia, inheriting taint.
[ 1761.560293] nvidia-uvm: Loaded the UVM driver, major device number 511.
[ 1761.575518] nvidia-modeset: Loading NVIDIA Kernel Mode Setting Driver for UNIX platforms 535.146.02 Sun Dec 3 14:02:44 UTC 2023
[ 1761.577931] [drm] [nvidia-drm] [GPU ID 0x00000100] Loading driver
[ 1761.577933] [drm] Initialized nvidia-drm 0.0.0 20160202 for 0000:01:00.0 on minor 0
[ 1761.583534] [drm] [nvidia-drm] [GPU ID 0x00000100] Unloading driver
[ 1761.596023] nvidia-modeset: Unloading
[ 1761.616658] nvidia-uvm: Unloaded the UVM driver.
[ 1761.640616] nvidia-nvlink: Unregistered Nvlink Core, major device number 235
ERROR: Installation has failed. Please see the file '/var/log/nvidia-installer.log' for details. You may find suggestions on fixing installation problems in the README available on the Linux driver download page at www.nvidia.com.
It seems to run fine, with current dmesg stating:
[ 6.450921] nvidia: loading out-of-tree module taints kernel.
[ 6.450929] nvidia: module license 'NVIDIA' taints kernel.
[ 6.450930] Disabling lock debugging due to kernel taint
[ 6.450933] nvidia: module verification failed: signature and/or required key missing - tainting kernel
[ 6.450934] nvidia: module license taints kernel.
[ 6.870031] nvidia-nvlink: Nvlink Core is being initialized, major device number 236
[ 6.871126] nvidia 0000:01:00.0: vgaarb: VGA decodes changed: olddecodes=io+mem,decodes=none:owns=io+mem
[ 6.907449] Generic FE-GE Realtek PHY r8169-0-300:00: attached PHY driver (mii_bus:phy_addr=r8169-0-300:00, irq=MAC)
[ 7.034035] r8169 0000:03:00.0 enp3s0: Link is Down
[ 7.079086] NVRM: loading NVIDIA UNIX x86_64 Kernel Module 535.146.02 Sun Dec 3 14:06:14 UTC 2023
[ 7.159341] nvidia-modeset: Loading NVIDIA Kernel Mode Setting Driver for UNIX platforms 535.146.02 Sun Dec 3 14:02:44 UTC 2023
[ 7.164664] [drm] [nvidia-drm] [GPU ID 0x00000100] Loading driver
[ 7.978821] [drm] Initialized nvidia-drm 0.0.0 20160202 for 0000:01:00.0 on minor 0
[ 9.729936] r8169 0000:03:00.0 enp3s0: Link is Up - 1Gbps/Full - flow control off
[ 14.263059] Xorg[1324]: memfd_create() called without MFD_EXEC or MFD_NOEXEC_SEAL set
Does this mean I’m now running a partly installed driver? Are there any installation steps after the check that failed that I am now missing? (It e.g. didn’t ask me about the xorg configuration file)
Am I right in assuming that the only thing that failed is the final check if the driver can actually be loaded? Any idea why that could fail?