Description of problem: nVidia proprietary driver 410.78-2.mga7.nonfree fails to build in Plasma 64-bit Mageia-7-beta1 classic iso (24/11 classic iso) install. Version-Release number of selected component (if applicable): How reproducible: Steps to Reproduce: 1. Install (on real UEFI/GPT hardware) Mageia-7-beta1 classic iso (24/11 version) 2. The nvidia proprietary fails to build, so install reverts to the Nouveau driver (which seems quite usable) N.B. Fails if 'proprietary driver' offer accepted during intallation OR if delayed until after reboot. 3. Note that although the 'nvida-settings' desktop icon is selectable, it fails: # /usr/bin/nvidia-settings Xlib: extension "GLX" missing on display ":0". ERROR: NVIDIA driver is not loaded ERROR: Unable to load info from any available system 4. Note: # dkms status nvidia-current, 410.78-2.mga7.nonfree, 4.19.5-desktop-1.mga7, x86_64: installed virtualbox, 5.2.22-2.mga7, 4.19.4-desktop-1.mga7, x86_64: installed-binary from 4.19.4-desktop-1.mga7 virtualbox, 5.2.22-2.mga7, 4.19.5-desktop-1.mga7, x86_64: installed-binary from 4.19.5-desktop-1.mga7 # modprobe nvidia-current modprobe: ERROR: could not insert 'nvidia_current': No such device # dmesg | tail -50 ........ [ 2732.779756] nvidia-nvlink: Nvlink Core is being initialized, major device number 241 [ 2732.779941] NVRM: The NVIDIA probe routine was not called for 1 device(s). [ 2732.779942] NVRM: This can occur when a driver such as: NVRM: nouveau, rivafb, nvidiafb or rivatv NVRM: was loaded and obtained ownership of the NVIDIA device(s). [ 2732.779942] NVRM: Try unloading the conflicting kernel module (and/or NVRM: reconfigure your kernel without the conflicting NVRM: driver(s)), then try loading the NVIDIA kernel module NVRM: again. [ 2732.779942] NVRM: No NVIDIA graphics adapter probed! [ 2732.780023] nvidia-nvlink: Unregistered the Nvlink Core, major device number 241 [ 2958.474910] nvidia-nvlink: Nvlink Core is being initialized, major device number 241 [ 2958.475112] NVRM: The NVIDIA probe routine was not called for 1 device(s). [ 2958.475112] NVRM: This can occur when a driver such as: NVRM: nouveau, rivafb, nvidiafb or rivatv NVRM: was loaded and obtained ownership of the NVIDIA device(s). [ 2958.475113] NVRM: Try unloading the conflicting kernel module (and/or NVRM: reconfigure your kernel without the conflicting NVRM: driver(s)), then try loading the NVIDIA kernel module NVRM: again. [ 2958.475113] NVRM: No NVIDIA graphics adapter probed! [ 2958.475191] nvidia-nvlink: Unregistered the Nvlink Core, major device number 241 5. Note that Sysinfo/Graphical info/OpenGL fails to show the usual info - just fails: "Could not initialise OpenGL/GLX Could not initialise OpenGL (ES)EGL"
CC: (none) => wilcal.int
Assignee: bugsquad => kernelKeywords: (none) => 7beta1CC: (none) => isobuild, marja11
Booting up today after yesterday's 41-package Cauldron s/w update (which installed the 410.78-2 proprietary nvidia driver), I was informed at the start of the boot that I should then reboot to make use of an updated driver. I accepted that reboot offer, and the ensuing reboot came up with that nVidia driver in action! --------------------------------- /usr/bin/nvidia-settings now displays correctly, as does sysinfo/graphics/OpenGL. --------------------------------- The $64 question is: Will that status quo be provided when the next iso install test is done?! Fingers crossed that it will...
P.S. Today's Cauldron update today also installed that nVidia driver, which is still in operation, so the situation seems to have sorted itself out for the time being...
Yep, 410.78-2 contains a fix for what drakx expects to find in an nVidia driver, so it should just keep working
CC: (none) => tmb
The problem appears to have disappeared, so this can be closed in the hope that it does not re-emerge...
Apart from a brief re-appearence of the nouveau on 16/2/2019, this problem seems to be under control, so closing...
Status: NEW => RESOLVEDResolution: (none) => FIXED