NVIDIA geforce Titan X not detected on Ubuntu 18.04 kernel 4.18.0-22 : Failed to initialize the NVIDIA GPU at PCI:2:0:0

I Installed drivers and Cuda-10.0 by adding cuda-repos from Nvidia Site. But the OS refuses to detect the GPU card

Error I see in the log: Failed to initialize the NVIDIA GPU at PCI:2:0:0

$ inxi -SGx
System: Host: graviton Kernel: 4.18.0-22-generic x86_64 bits: 64 gcc: 7.3.0 Console: tty 0
Distro: Ubuntu 18.04.2 LTS
Graphics: Card: NVIDIA GM200 [GeForce GTX TITAN X] bus-ID: 02:00.0
Display Server: N/A driver: (unloaded: nvidia) tty size: 145x45 Advanced Data: N/A out of X

$ cat /var/log/Xorg.0.log

[ 313.228] () NVIDIA(0): Depth 24, (–) framebuffer bpp 32
[ 313.228] (==) NVIDIA(0): RGB weight 888
[ 313.228] (==) NVIDIA(0): Default visual is TrueColor
[ 313.228] (==) NVIDIA(0): Using gamma correction (1.0, 1.0, 1.0)
[ 313.228] (II) Applying OutputClass “nvidia” options to /dev/dri/card0
[ 313.228] (
) NVIDIA(0): Option “AllowEmptyInitialConfiguration”
[ 313.228] (**) NVIDIA(0): Enabling 2D acceleration
[ 313.228] (II) Loading sub module “glxserver_nvidia”
[ 313.228] (II) LoadModule: “glxserver_nvidia”
[ 313.229] (II) Loading /usr/lib/xorg/modules/extensions/libglxserver_nvidia.so
[ 313.232] (II) Module glxserver_nvidia: vendor=“NVIDIA Corporation”
[ 313.232] compiled for 4.0.2, module version = 1.0.0
[ 313.232] Module class: X.Org Server Extension
[ 313.232] (II) NVIDIA GLX Module 410.48 Thu Sep 6 06:24:45 CDT 2018
[ 313.520] (EE) NVIDIA(GPU-0): Failed to initialize the NVIDIA GPU at PCI:2:0:0. Please
[ 313.520] (EE) NVIDIA(GPU-0): check your system’s kernel log for additional error
[ 313.520] (EE) NVIDIA(GPU-0): messages and refer to Chapter 8: Common Problems in the
[ 313.520] (EE) NVIDIA(GPU-0): README for additional information.
[ 313.520] (EE) NVIDIA(GPU-0): Failed to initialize the NVIDIA graphics device!
[ 313.520] (EE) NVIDIA(0): Failing initialization of X screen 0
[ 313.520] (II) UnloadModule: “nvidia”
[ 313.520] (II) UnloadSubModule: “glxserver_nvidia”
[ 313.520] (II) Unloading glxserver_nvidia
[ 313.520] (II) UnloadSubModule: “wfb”
[ 313.520] (II) UnloadSubModule: “fb”

nvidia-bug-report.log.gz (573 KB)

Please check if secure boot is enabled in bios and disable it.
Please run nvidia-bug-report.sh as root and attach the resulting .gz file to your post. Hovering the mouse over an existing post of yours will reveal a paperclip icon.


Secure boot in UEFI is disable. I checked before installing the drivers.

Installed nvidia-driver-430 and cuda-10.0

$ nvidia-smi
No devices were found

$ modinfo /var/lib/dkms/nvidia/430.26/4.18.0-22-generic/x86_64/module/nvidia.ko
filename: /var/lib/dkms/nvidia/430.26/4.18.0-22-generic/x86_64/module/nvidia.ko
alias: char-major-195-*
version: 430.26
supported: external
license: NVIDIA
srcversion: 89BDA0F56877588EC9454C6
alias: pci:v000010DEd00000E00svsdbc04sc80i00*
alias: pci:v000010DEdsvsdbc03sc02i00
alias: pci:v000010DEdsvsdbc03sc00i00
depends: ipmi_msghandler
retpoline: Y
name: nvidia
vermagic: 4.18.0-22-generic SMP mod_unload

nvidia-bug-report.log.gz (573 KB)

You’re running into

[  100.589306] NVRM: RmInitAdapter failed! (0x25:0x40:1103)
[  100.589323] NVRM: rm_init_adapter failed for device bearing minor number 0

might be a hardware failure. Check for a system bios update, try reseating card, check if it works in another system.

We checked that by going back to Ubuntu 14.04 from 18.04. GPU works fine in 14.04. problem is specific to 18.04.
Any clues why it is failing in 18.04

Did you upgrade the bios yet?
Please try a current 418/430 driver.

I installed 430 driver only. Not sure about bios update. Will try that