Distorted Login Screen - Nvidia Driver - 2x Titan RTX - Failed to initialize the NVIDIA GPU at PCI

First of all: Hello to everyone in this forum, I appreciate any help of you a lot.

Second: I know this seems like a know issue - I googled a lot and I tried to install the nvidia drivers in various ways but I never succeeded so far.

Problem: Once I installed the driver (via Yast, the runfile or the cuda package - I tried many options) the screen looks weird when the login screen is supposed to appear. You usually see either a black screen or a console login with greenish stripes distributed randomly across the screen.

With nomodeset 3 I can login to the console and

hwinfo --gfxcard

tells me that the nvidia drivers seem to be active (as far as I can tell).

07: PCI b300.0: 0300 VGA compatible controller (VGA)
  [Created at pci.386]
  Unique ID: fdCU.P_xIeD+NKiD
  Parent ID: zm+K.ktE97HFsmdF
  SysFS ID: /devices/pci0000:b2/0000:b2:00.0/0000:b3:00.0
  SysFS BusID: 0000:b3:00.0
  Hardware Class: graphics card
  Model: "nVidia TU102"
  Vendor: pci 0x10de "nVidia Corporation"
  Device: pci 0x1e02 "TU102"
  SubVendor: pci 0x10de "nVidia Corporation"
  SubDevice: pci 0x12a3 
  Revision: 0xa1
  Driver: "nvidia"
  Driver Modules: "nvidia"
  Memory Range: 0xfa000000-0xfaffffff (rw,non-prefetchable)
  Memory Range: 0xe0000000-0xefffffff (ro,non-prefetchable)
  Memory Range: 0xf0000000-0xf1ffffff (ro,non-prefetchable)
  I/O Ports: 0xf000-0xffff (rw)
  Memory Range: 0xfb000000-0xfb07ffff (ro,non-prefetchable,disabled)
  IRQ: 45 (no events)
  Module Alias: "pci:v000010DEd00001E02sv000010DEsd000012A3bc03sc00i00"
  Driver Info #0:
    Driver Status: nouveau is not active
    Driver Activation Cmd: "modprobe nouveau"
  Driver Info #1:
    Driver Status: nvidia_drm is active
    Driver Activation Cmd: "modprobe nvidia_drm"
  Driver Info #2:
    Driver Status: nvidia is active
    Driver Activation Cmd: "modprobe nvidia"
  Config Status: cfg=new, avail=yes, need=no, active=unknown
  Attached to: #24 (PCI bridge)

50: PCI 6500.0: 0300 VGA compatible controller (VGA)
  [Created at pci.386]
  Unique ID: ze67.P_xIeD+NKiD
  Parent ID: Iovz.ktE97HFsmdF
  SysFS ID: /devices/pci0000:64/0000:64:00.0/0000:65:00.0
  SysFS BusID: 0000:65:00.0
  Hardware Class: graphics card
  Model: "nVidia TU102"
  Vendor: pci 0x10de "nVidia Corporation"
  Device: pci 0x1e02 "TU102"
  SubVendor: pci 0x10de "nVidia Corporation"
  SubDevice: pci 0x12a3 
  Revision: 0xa1
  Driver: "nvidia"
  Driver Modules: "nvidia"
  Memory Range: 0xd7000000-0xd7ffffff (rw,non-prefetchable)
  Memory Range: 0xc0000000-0xcfffffff (ro,non-prefetchable)
  Memory Range: 0xd0000000-0xd1ffffff (ro,non-prefetchable)
  I/O Ports: 0xb000-0xbfff (rw)
  Memory Range: 0x000c0000-0x000dffff (rw,non-prefetchable,disabled)
  IRQ: 35 (no events)
  Module Alias: "pci:v000010DEd00001E02sv000010DEsd000012A3bc03sc00i00"
  Driver Info #0:
    Driver Status: nouveau is not active
    Driver Activation Cmd: "modprobe nouveau"
  Driver Info #1:
    Driver Status: nvidia_drm is active
    Driver Activation Cmd: "modprobe nvidia_drm"
  Driver Info #2:
    Driver Status: nvidia is active
    Driver Activation Cmd: "modprobe nvidia"
  Config Status: cfg=new, avail=yes, need=no, active=unknown
  Attached to: #72 (PCI bridge)

Primary display adapter: #7

Here is the Xorg.0.log

    86.864] 
X.Org X Server 1.19.6
Release Date: 2017-12-20
    86.865] X Protocol Version 11, Revision 0
    86.866] Build Operating System: openSUSE SUSE LINUX
    86.866] Current Operating System: Linux linux-zmit 4.12.14-lp150.12.48-default #1 SMP Tue Feb 12 14:01:48 UTC 2019 (268f014) x86_64
    86.866] Kernel command line: BOOT_IMAGE=/boot/vmlinuz-4.12.14-lp150.12.48-default root=UUID=73c4710a-d791-4a53-9d43-7263b71ef5be resume=/dev/disk/by-id/ata-Samsung_SSD_860_EVO_250GB_S3YGNC0K827753L-part4 splash=silent quiet showopts nomodeset 3
    86.868] Build Date: 14 March 2019  12:00:00PM
    86.868]  
    86.869] Current version of pixman: 0.34.0
    86.870]     Before reporting problems, check http://wiki.x.org
    to make sure that you have the latest version.
    86.870] Markers: (--) probed, (**) from config file, (==) default setting,
    (++) from command line, (!!) notice, (II) informational,
    (WW) warning, (EE) error, (NI) not implemented, (??) unknown.
    86.872] (==) Log file: "/var/log/Xorg.0.log", Time: Tue Apr  9 18:00:13 2019
    86.874] (==) Using config directory: "/etc/X11/xorg.conf.d"
    86.874] (==) Using system config directory "/usr/share/X11/xorg.conf.d"
    86.875] (==) No Layout section.  Using the first Screen section.
    86.875] (==) No screen section available. Using defaults.
    86.875] (**) |-->Screen "Default Screen Section" (0)
    86.875] (**) |   |-->Monitor "<default monitor>"
    86.875] (==) No monitor specified for screen "Default Screen Section".
    Using a default monitor configuration.
    86.875] (==) Automatically adding devices
    86.875] (==) Automatically enabling devices
    86.875] (==) Automatically adding GPU devices
    86.875] (==) Max clients allowed: 256, resource mask: 0x1fffff
    86.877] (WW) The directory "/usr/share/fonts/misc/sgi" does not exist.
    86.877]     Entry deleted from font path.
    86.877] (==) FontPath set to:
    /usr/share/fonts/misc:unscaled,
    /usr/share/fonts/Type1/,
    /usr/share/fonts/100dpi:unscaled,
    /usr/share/fonts/75dpi:unscaled,
    /usr/share/fonts/ghostscript/,
    /usr/share/fonts/cyrillic:unscaled,
    /usr/share/fonts/truetype/,
    built-ins
    86.877] (==) ModulePath set to "/usr/lib64/xorg/modules"
    86.877] (**) Extension "XFree86-DGA" is disabled
    86.877] (II) The server relies on udev to provide the list of input devices.
    If no devices become available, reconfigure udev or disable AutoAddDevices.
    86.877] (II) Loader magic: 0x55577d201d00
    86.877] (II) Module ABI versions:
    86.877]     X.Org ANSI C Emulation: 0.4
    86.877]     X.Org Video Driver: 23.0
    86.877]     X.Org XInput driver : 24.1
    86.877]     X.Org Server Extension : 10.0
    86.879] (++) using VT number 1

    86.882] (II) systemd-logind: took control of session /org/freedesktop/login1/session/_31
    86.882] (II) xfree86: Adding drm device (/dev/dri/card0)
    86.883] (II) systemd-logind: got fd for /dev/dri/card0 226:0 fd 11 paused 0
    86.883] (II) xfree86: Adding drm device (/dev/dri/card1)
    86.884] (II) systemd-logind: got fd for /dev/dri/card1 226:1 fd 12 paused 0
    86.895] (--) PCI:*(0:101:0:0) 10de:1e02:10de:12a3 rev 161, Mem @ 0xd7000000/16777216, 0xc0000000/268435456, 0xd0000000/33554432, I/O @ 0x0000b000/128, BIOS @ 0x????????/131072
    86.895] (--) PCI: (0:179:0:0) 10de:1e02:10de:12a3 rev 161, Mem @ 0xfa000000/16777216, 0xe0000000/268435456, 0xf0000000/33554432, I/O @ 0x0000f000/128, BIOS @ 0x????????/524288
    86.895] (II) LoadModule: "glx"
    86.898] (II) Loading /usr/lib64/xorg/modules/extensions/libglx.so
    86.969] (II) Module glx: vendor="NVIDIA Corporation"
    86.969]     compiled for 4.0.2, module version = 1.0.0
    86.969]     Module class: X.Org Server Extension
    86.969] (II) NVIDIA GLX Module  418.40.04  Fri Mar 15 00:52:32 CDT 2019
    86.973] (II) Scanning /etc/X11/xorg_pci_ids directory for additional PCI ID's supported by the drivers
    86.973] (II) Scanning /etc/X11/xorg_pci_ids directory for additional PCI ID's supported by the drivers
    86.973] (II) Scanning /etc/X11/xorg_pci_ids directory for additional PCI ID's supported by the drivers
    86.973] (==) Matched nvidia as autoconfigured driver 0
    86.973] (==) Matched nouveau as autoconfigured driver 1
    86.973] (==) Matched nv as autoconfigured driver 2
    86.973] (==) Matched nvidia as autoconfigured driver 3
    86.973] (==) Matched nouveau as autoconfigured driver 4
    86.973] (==) Matched nv as autoconfigured driver 5
    86.973] (==) Matched nvidia as autoconfigured driver 6
    86.973] (==) Matched nouveau as autoconfigured driver 7
    86.973] (==) Matched nv as autoconfigured driver 8
    86.973] (==) Matched modesetting as autoconfigured driver 9
    86.973] (==) Matched fbdev as autoconfigured driver 10
    86.973] (==) Matched vesa as autoconfigured driver 11
    86.973] (==) Assigned the driver to the xf86ConfigLayout
    86.973] (II) LoadModule: "nvidia"
    86.973] (II) Loading /usr/lib64/xorg/modules/drivers/nvidia_drv.so
    86.986] (II) Module nvidia: vendor="NVIDIA Corporation"
    86.986]     compiled for 4.0.2, module version = 1.0.0
    86.986]     Module class: X.Org Video Driver
    86.986] (II) LoadModule: "nouveau"
    86.986] (II) Loading /usr/lib64/xorg/modules/drivers/nouveau_drv.so
    86.992] (II) Module nouveau: vendor="X.Org Foundation"
    86.992]     compiled for 1.19.6, module version = 1.0.15
    86.992]     Module class: X.Org Video Driver
    86.992]     ABI class: X.Org Video Driver, version 23.0
    86.992] (II) LoadModule: "nv"
    86.993] (WW) Warning, couldn't open module nv
    86.993] (II) UnloadModule: "nv"
    86.993] (II) Unloading nv
    86.993] (EE) Failed to load module "nv" (module does not exist, 0)
    86.993] (II) LoadModule: "modesetting"
    86.993] (II) Loading /usr/lib64/xorg/modules/drivers/modesetting_drv.so
    86.994] (II) Module modesetting: vendor="X.Org Foundation"
    86.994]     compiled for 1.19.6, module version = 1.19.6
    86.994]     Module class: X.Org Video Driver
    86.994]     ABI class: X.Org Video Driver, version 23.0
    86.994] (II) LoadModule: "fbdev"
    86.994] (II) Loading /usr/lib64/xorg/modules/drivers/fbdev_drv.so
    86.995] (II) Module fbdev: vendor="X.Org Foundation"
    86.995]     compiled for 1.19.6, module version = 0.4.4
    86.995]     Module class: X.Org Video Driver
    86.995]     ABI class: X.Org Video Driver, version 23.0
    86.995] (II) LoadModule: "vesa"
    86.995] (II) Loading /usr/lib64/xorg/modules/drivers/vesa_drv.so
    86.995] (II) Module vesa: vendor="X.Org Foundation"
    86.995]     compiled for 1.19.6, module version = 2.4.0
    86.995]     Module class: X.Org Video Driver
    86.995]     ABI class: X.Org Video Driver, version 23.0
    86.996] (II) NVIDIA dlloader X Driver  418.40.04  Fri Mar 15 00:54:41 CDT 2019
    86.996] (II) NVIDIA Unified Driver for all Supported NVIDIA GPUs
    86.996] (II) NOUVEAU driver 
    86.996] (II) NOUVEAU driver for NVIDIA chipset families :
    86.996]     RIVA TNT        (NV04)
    86.996]     RIVA TNT2       (NV05)
    86.996]     GeForce 256     (NV10)
    86.996]     GeForce 2       (NV11, NV15)
    86.996]     GeForce 4MX     (NV17, NV18)
    86.996]     GeForce 3       (NV20)
    86.996]     GeForce 4Ti     (NV25, NV28)
    86.996]     GeForce FX      (NV3x)
    86.996]     GeForce 6       (NV4x)
    86.996]     GeForce 7       (G7x)
    86.996]     GeForce 8       (G8x)
    86.996]     GeForce GTX 200 (NVA0)
    86.996]     GeForce GTX 400 (NVC0)
    86.996] (II) modesetting: Driver for Modesetting Kernel Drivers: kms
    86.997] (II) FBDEV: driver for framebuffer: fbdev
    86.997] (II) VESA: driver for VESA chipsets: vesa
    86.997] (II) systemd-logind: releasing fd for 226:0
    86.998] (II) Loading sub module "fb"
    86.998] (II) LoadModule: "fb"
    86.999] (II) Loading /usr/lib64/xorg/modules/libfb.so
    87.000] (II) Module fb: vendor="X.Org Foundation"
    87.000]     compiled for 1.19.6, module version = 1.0.0
    87.000]     ABI class: X.Org ANSI C Emulation, version 0.4
    87.000] (II) Loading sub module "wfb"
    87.000] (II) LoadModule: "wfb"
    87.000] (II) Loading /usr/lib64/xorg/modules/libwfb.so
    87.001] (II) Module wfb: vendor="X.Org Foundation"
    87.001]     compiled for 1.19.6, module version = 1.0.0
    87.001]     ABI class: X.Org ANSI C Emulation, version 0.4
    87.002] (II) Loading sub module "ramdac"
    87.002] (II) LoadModule: "ramdac"
    87.002] (II) Module "ramdac" already built-in
    87.003] (II) systemd-logind: releasing fd for 226:1
    87.004] (EE) [drm] Failed to open DRM device for (null): -2
    87.004] (WW) Falling back to old probe method for modesetting
    87.004] (WW) Falling back to old probe method for fbdev
    87.004] (II) Loading sub module "fbdevhw"
    87.004] (II) LoadModule: "fbdevhw"
    87.005] (II) Loading /usr/lib64/xorg/modules/libfbdevhw.so
    87.005] (II) Module fbdevhw: vendor="X.Org Foundation"
    87.005]     compiled for 1.19.6, module version = 0.0.2
    87.005]     ABI class: X.Org Video Driver, version 23.0
    87.005] (II) NVIDIA(0): Creating default Display subsection in Screen section
    "Default Screen Section" for depth/fbbpp 24/32
    87.005] (==) NVIDIA(0): Depth 24, (==) framebuffer bpp 32
    87.005] (==) NVIDIA(0): RGB weight 888
    87.005] (==) NVIDIA(0): Default visual is TrueColor
    87.005] (==) NVIDIA(0): Using gamma correction (1.0, 1.0, 1.0)
    87.006] (**) NVIDIA(0): Enabling 2D acceleration
    87.819] (EE) NVIDIA(GPU-0): Failed to initialize the NVIDIA GPU at PCI:101:0:0.  Please
    87.819] (EE) NVIDIA(GPU-0):     check your system's kernel log for additional error
    87.819] (EE) NVIDIA(GPU-0):     messages and refer to Chapter 8: Common Problems in the
    87.819] (EE) NVIDIA(GPU-0):     README for additional information.
    87.819] (EE) NVIDIA(GPU-0): Failed to initialize the NVIDIA graphics device!
    87.819] (EE) NVIDIA(0): Failing initialization of X screen 0
    87.819] (II) UnloadModule: "nvidia"
    87.819] (II) UnloadSubModule: "wfb"
    87.819] (II) UnloadSubModule: "fb"
    87.819] (EE) Screen(s) found, but none have a usable configuration.
    87.819] (EE) 
Fatal server error:
    87.819] (EE) no screens found(EE) 
    87.819] (EE) 
Please consult the The X.Org Foundation support 
     at http://wiki.x.org
 for help. 
    87.819] (EE) Please also check the log file at "/var/log/Xorg.0.log" for additional information.
    87.819] (EE) 
    87.822] (EE) Server terminated with error (1). Closing log file.

I’m quite new to OpenSuse and I’m running out of ideas here :slight_smile: Any help is greatly appreaceated.

DO you know which NVIDIA driver you installed thee ae several version for different families. This is a very new card so I’d be sure to use the 05 version. If this is the one installed check with NVIDIA to see what they recommend.

Does Plymouth remain installed? https://bugzilla.opensuse.org/show_bug.cgi?id=1090451 suggests it should not be.

Thank you for the suggestions - I will try it with the 05 again and without Plymouth. I can check this tomorrow.
I think I tried the 04 as well as 05 driver so far. Maybe it is Plymouth then.

Would you recommend the Yast Nvidia repo or the Nvidia rpm file: https://www.nvidia.com/download/driverResults.aspx/141506/en-us ?

The one on the web seems older then the one I see here 415.25 there 418.56 from the repo (note I still run 42,3 so not sure what is up on 15.0)

Either should work. down side of installing the hard way (ie from web version) is that you must reinstall after any kernel update. From the repo it is automatic but you do need to reboot

Hi OP how does it go? I’m planning to put an RTX on my machine but I was afraid there will be a problem since it’s a new hardware and not sure if it will be easily supported, planning to play some games on leap it would be could if you can give us some status on it. Thank you.

What I did:

  • Fresh installation of Leap 15.0
  • zypper update
  • reboot (with nomodeset)
  • added nvidia repo (via yast)
  • uninstalled plymouth and plymouth-dracut (via yast)
  • installed x11-video-nvidiaG05, nvidia-glG05, nvidia-gfxG05-kmp-default, nvidia-computeG05 (+accepted automatic changes) (via yast)
  • reboot (with nomodeset)

Screen still looks like this:
https://www.dropbox.com/s/6zag9sg11lhf8m1/20190411_100629.jpg?dl=0

Did you try booting **without **nomodeset?

As the last step: yes - I tried it with and without nomodeset.

Maybe some additional info can help to figure out the error source:

  • There is no xorg.conf file but I think the configuration of X now takes place ithe corresponding folder.
  • After I did as described above hwinfo --gfxcard still shows that nouveau is not active, but nvidia is
  • The Xorg.0.log still states “Failed to initialize the NVIDIA GPU at PCI:101:0:0” and “Screen(s) found, but none have a usable configuration.”

Can I provide any further log files or information that might help?

Did it work passably with a FOSS DDX before you rebooted using nomodeset? Do you know it’s not too new to be supported using the much older 15.0 kernel? 15.1 is due for release before June, so you might want to give its beta or TW a try.

Before I installed the Nvidia Packages I could load Gnome and use the Desktop (with a low resolution) As far as I can remember nouveau was active back then.

Did you do as suggested?

    87.819] (EE) NVIDIA(GPU-0): Failed to initialize the NVIDIA GPU at PCI:101:0:0.  Please
    87.819] (EE) NVIDIA(GPU-0):     check your system's kernel log for additional error
    87.819] (EE) NVIDIA(GPU-0):     messages and refer to Chapter 8: Common Problems in the
    87.819] (EE) NVIDIA(GPU-0):     README for additional information.

Nomodeset](SDB:nomodeset: Work Around Graphic Upgrade & Installation Obstacles - openSUSE Wiki) is a troubleshooting parameter.

After I did as described above hwinfo --gfxcard still shows that nouveau is not active, but nvidia is
Usually ‘inxi -Gxx’ is more useful than hwinfo, but not so much if you can’t run it from within an X session. Within an X session inxi reports both kernel drivers and DDX running and optional, and makes device ID easy to copy & paste for lookup.

Here it is (var/log/messages): http://termbin.com/sim5

I took a look before, but actually I don’t really know what I have to look for.

Here is inxi -Gxx: https://termbin.com/ko2a

(unloaded: Nvidia does’n sound too good :-/) I executed the command after I logged in via the console (no GUI) if it is of any help, i can go back to the system state before I installed the drivers and call it from an x session.

Here is apparently a problem:


2019-04-11T13:13:45.559350+02:00 linux-zmit kernel:     7.723693] resource sanity check: requesting [mem 0x000c0000-0x000fffff], which spans more than PCI Bus 0000:00 [mem 0x000c4000-0x000c7fff window]
2019-04-11T13:13:45.559360+02:00 linux-zmit kernel:     7.723834] caller _nv000934rm+0x1bf/0x1f0 [nvidia] mapping multiple BARs
2019-04-11T13:13:45.867364+02:00 linux-zmit kernel:     8.031619] NVRM: RmInitAdapter failed! (0x26:0xffff:1106)
2019-04-11T13:13:45.867381+02:00 linux-zmit kernel:     8.031646] NVRM: rm_init_adapter failed for device bearing minor number 0

The first two messages should be warnings; but the next two mean nvidia kernel driver failed for whatever reasons. You may have more luck asking on nvidia support forums.

Oh, thank you - now I have a good starting point for further investigations :slight_smile:

In some Nvidia Forums I read that this might be related to the “secure-boot” option.
Do you have any recommendations on this topic?

IMHO Secure boot is security theater It protects the initial boot stack from changes but if someone can get to it then they already own your machine…:’(

A. Revert your “nomodeset” or “modesetting” settings or try next steps after a clean install.

B. Try to update X11 (X.org, Display Server):

  1. Add repository: https://download.opensuse.org/repositories/X11:/XOrg/openSUSE_Leap_15.0/
  2. Make “Full repository Vendor change”: SDB:Vendor change update - openSUSE Wiki
  3. Update your system

When I tried to update only Mesa packets without other components my system had hanged (ATI - AMD videochip).

C. Then renew your Nvidia drivers:

  1. Uninstall Nvidia drivers (if installed)
  2. Uninstall Nouveau drivers
  3. Install Nvidia G05 (the newest) drivers (SDB:NVIDIA drivers - openSUSE Wiki)

[HR][/HR]

nvidia-nvlink: Nvlink Core is being initialized, major device number 244

What about NVLink bridge?
Try to start working with a single videocard, then add second.

In some Nvidia Forums I read that this might be related to the “secure-boot” option.
Do you have any recommendations on this topic?

Try to set it “Off”.
If you succeed with your goals, then you can set it “On”.