Jump to content
Forums in Read-Only Mode - Please use Reddit ×
  • 0

[SOLVED] New nvidia-drivers, kernel module fails to load, X server fails to start


moldmaker

Question

Hi everybody,

when I installed Funtoo on my rather old computer in April, I was running a Nvidia GeForce GTX 580 VGA adapter. I installed the stage 3 Gnome tarball which gave me the 4.19.98_p1-debian-sources-lts kernel. For the old graphics card I had to mask the newer nvidia-drivers; it is supported by the 390.132. No problem so far, everything running smoothly.
Some days ago, I replaced the old VGA adapter with a newer one, a Nvidia GTX 980 Ti. So it came to my mind to do some software updates too. I decided to install the debian-sources 5.9.6_p1, I also did

emerge -avt linux-firmware

and

ego boot update

The new kernel was starting up properly, but the X server did not start because the old nvidia driver was only working up to kernel 5.4. So I unmasked the newer nvidia.drivers and installed version 450.80. There were no errors during the installation. But still the X server does not start, and now I am stuck. This is the Xorg logfile.

[   324.724] 
X.Org X Server 1.20.8
X Protocol Version 11, Revision 0
[   324.800] Build Operating System: Linux 5.9.6_p1-debian-sources x86_64 Gentoo
[   324.828] Current Operating System: Linux funtoo 5.9.6_p1-debian-sources #1 SMP Sun Dec 20 14:20:26 CET 2020 x86_64
[   324.828] Kernel command line: BOOT_IMAGE=/kernel-debian-sources-x86_64-5.9.6_p1 nomodeset intel_iommu=on snd_hda_intel.power_save=0 real_root=/dev/sda3 rootfstype=ext4 rand_id=R9C5AYUB
[   324.890] Build Date: 20 December 2020  03:45:56PM
[   324.921]  
[   324.953] Current version of pixman: 0.34.0
[   325.018] 	Before reporting problems, check http://wiki.x.org
	to make sure that you have the latest version.
[   325.018] Markers: (--) probed, (**) from config file, (==) default setting,
	(++) from command line, (!!) notice, (II) informational,
	(WW) warning, (EE) error, (NI) not implemented, (??) unknown.
[   325.155] (==) Log file: "/var/log/Xorg.0.log", Time: Sun Dec 20 16:33:37 2020
[   325.191] (==) Using config file: "/etc/X11/xorg.conf"
[   325.227] (==) Using config directory: "/etc/X11/xorg.conf.d"
[   325.264] (==) Using system config directory "/usr/share/X11/xorg.conf.d"
[   325.264] (==) ServerLayout "Layout0"
[   325.264] (**) |-->Screen "Screen0" (0)
[   325.264] (**) |   |-->Monitor "Monitor0"
[   325.265] (**) |   |-->Device "Device0"
[   325.265] (**) |-->Input Device "Keyboard0"
[   325.265] (**) |-->Input Device "Mouse0"
[   325.265] (**) Option "Xinerama" "0"
[   325.265] (==) Automatically adding devices
[   325.265] (==) Automatically enabling devices
[   325.265] (==) Automatically adding GPU devices
[   325.265] (==) Max clients allowed: 256, resource mask: 0x1fffff
[   325.265] (==) FontPath set to:
	/usr/share/fonts/misc/,
	/usr/share/fonts/TTF/,
	/usr/share/fonts/OTF/,
	/usr/share/fonts/Type1/,
	/usr/share/fonts/100dpi/,
	/usr/share/fonts/75dpi/
[   325.265] (==) ModulePath set to "/usr/lib64/xorg/modules"
[   325.265] (WW) Hotplugging is on, devices using drivers 'kbd', 'mouse' or 'vmmouse' will be disabled.
[   325.265] (WW) Disabling Keyboard0
[   325.265] (WW) Disabling Mouse0
[   325.265] (II) Loader magic: 0x55a701434d00
[   325.265] (II) Module ABI versions:
[   325.265] 	X.Org ANSI C Emulation: 0.4
[   325.265] 	X.Org Video Driver: 24.1
[   325.265] 	X.Org XInput driver : 24.1
[   325.265] 	X.Org Server Extension : 10.0
[   325.268] (++) using VT number 2

[   325.268] (--) controlling tty is VT number 2, auto-enabling KeepTty
[   325.271] (II) systemd-logind: took control of session /org/freedesktop/login1/session/_32
[   325.276] (--) PCI: (2@0:0:0) 14f1:8852:d470:9022 rev 2, Mem @ 0xf9a00000/2097152
[   325.276] (--) PCI:*(6@0:0:0) 10de:17c8:1043:8548 rev 161, Mem @ 0xfa000000/16777216, 0xd0000000/268435456, 0xce000000/33554432, I/O @ 0x0000ec00/128, BIOS @ 0x????????/131072
[   325.276] (II) LoadModule: "glx"
[   325.276] (II) Loading /usr/lib64/xorg/modules/extensions/libglx.so
[   325.277] (II) Module glx: vendor="X.Org Foundation"
[   325.277] 	compiled for 1.20.8, module version = 1.0.0
[   325.277] 	ABI class: X.Org Server Extension, version 10.0
[   325.277] (II) LoadModule: "nvidia"
[   325.277] (II) Loading /usr/lib64/xorg/modules/drivers/nvidia_drv.so
[   325.277] (II) Module nvidia: vendor="NVIDIA Corporation"
[   325.277] 	compiled for 1.6.99.901, module version = 1.0.0
[   325.277] 	Module class: X.Org Video Driver
[   325.277] (II) NVIDIA dlloader X Driver  450.80.02  Wed Sep 23 00:53:01 UTC 2020
[   325.277] (II) NVIDIA Unified Driver for all Supported NVIDIA GPUs
[   325.277] (II) Loading sub module "fb"
[   325.277] (II) LoadModule: "fb"
[   325.277] (II) Loading /usr/lib64/xorg/modules/libfb.so
[   325.277] (II) Module fb: vendor="X.Org Foundation"
[   325.277] 	compiled for 1.20.8, module version = 1.0.0
[   325.277] 	ABI class: X.Org ANSI C Emulation, version 0.4
[   325.277] (II) Loading sub module "wfb"
[   325.277] (II) LoadModule: "wfb"
[   325.277] (II) Loading /usr/lib64/xorg/modules/libwfb.so
[   325.277] (II) Module wfb: vendor="X.Org Foundation"
[   325.278] 	compiled for 1.20.8, module version = 1.0.0
[   325.278] 	ABI class: X.Org ANSI C Emulation, version 0.4
[   325.278] (II) Loading sub module "ramdac"
[   325.278] (II) LoadModule: "ramdac"
[   325.278] (II) Module "ramdac" already built-in
[   325.281] (EE) NVIDIA: Failed to initialize the NVIDIA kernel module. Please see the
[   325.281] (EE) NVIDIA:     system's kernel log for additional error messages and
[   325.281] (EE) NVIDIA:     consult the NVIDIA README for details.
[   325.281] (EE) No devices detected.
[   325.281] (EE) 
Fatal server error:
[   325.282] (EE) no screens found(EE) 
[   325.282] (EE) 
Please consult the The X.Org Foundation support 
	 at http://wiki.x.org
 for help. 
[   325.282] (EE) Please also check the log file at "/var/log/Xorg.0.log" for additional information.
[   325.282] (EE) 
[   325.505] (EE) Server terminated with error (1). Closing log file.

"Failed to initialize the NVIDIA kernel module" seems to be the problem. When I looked in dmesg i find nothing that gives me a hint, except this:

[    1.984255] crc16: module verification failed: signature and/or required key missing - tainting kernel

 

But I do not know if this is related to the nvidia kernel module or something else. I have attached the dmesg file to this post. Does anyone have an idea what went wrong? Am I missing something?

Mm.

 

dmesg

Edited by moldmaker
Link to comment
Share on other sites

3 answers to this question

Recommended Posts

  • 0

I can answer this myself, sort of. I downloaded the official linux drivers file from the NVidia website, the exact same version as I installed with portage, 450.80.02.

I ran the script with the same result (kernel module fails to load, X does not start), but the nvidia installer gave me a hint to use the option -no-unified-memory. Since I do not need Cuda, I gave it a try and it worked. Now the kernel module can be loaded. After that, I re-installed the ebuild with the new Use-Flag -cuda, and it works as well.

Mm.

Link to comment
Share on other sites

  • 0
On 12/21/2020 at 5:45 PM, moldmaker said:

I can answer this myself, sort of. I downloaded the official linux drivers file from the NVidia website, the exact same version as I installed with portage, 450.80.02.

I ran the script with the same result (kernel module fails to load, X does not start), but the nvidia installer gave me a hint to use the option -no-unified-memory. Since I do not need Cuda, I gave it a try and it worked. Now the kernel module can be loaded. After that, I re-installed the ebuild with the new Use-Flag -cuda, and it works as well.

Mm.

 

Hi there. Use `-no-unified-memory` where?

Thank you.

Link to comment
Share on other sites

  • 0
6 hours ago, NikosAlexandris said:

 

Hi there. Use `-no-unified-memory` where?

Thank you.

Use it as an argument when executing the install script from the NVidia website.  The driver's install script is named NVIDIA-Linux-x86_64-450.80.02.run, so if you are as root in the same directory where the driver script is located, it is

./NVIDIA-Linux-x86_64-450.80.02.run --no-unified-memory

Mm.

Link to comment
Share on other sites

×
×
  • Create New...