[solved] Nvidia driver installed and running, but still software rendering

[baronvonsatchel]

Hey everyone, I've been using LM for almost 2 years, but around a week ago had my first real "Linux moment" that had me reinstall the operating system in order to fix it. Since then, there have been minor glitches here and there but it's been perfectly usable up to this point. When I went to turn my computer on today however, it came up with a message saying that software rendering was enabled and to check the driver manager. I'm also getting all sorts of flickering and black boxes appearing on my monitors. I've spent hours researching the problem and found countless threads with similar problems, but none of the solutions did the trick for me. I've already tried purging and reinstalling the drivers (several times in fact), and adding Nvidia drivers to /etc/modules. Driver manager says I am using Nvidia driver 470 and not Nouveau. Secure boot is disabled. You can see from my inxi -Fxxx below that even though the Nvidia driver is recognized, the renderer is llvmpipe (software rendering) and I think that might be my problem. In either case I don't know how to go about fixing it.

Code: Select all

System:
  Kernel: 5.4.0-86-generic x86_64 bits: 64 compiler: gcc v: 9.3.0 
  Desktop: Cinnamon 5.0.5 wm: muffin 5.0.1 dm: LightDM 1.30.0 
  Distro: Linux Mint 20.2 Uma base: Ubuntu 20.04 focal 
Machine:
  Type: Desktop System: Gigabyte product: Z390 AORUS PRO WIFI v: N/A 
  serial: <filter> 
  Mobo: Gigabyte model: Z390 AORUS PRO WIFI-CF v: x.x serial: <filter> 
  UEFI: American Megatrends v: F10 date: 06/05/2019 
CPU:
  Topology: 8-Core model: Intel Core i7-9700K bits: 64 type: MCP 
  arch: Kaby Lake rev: D L2 cache: 12.0 MiB 
  flags: avx avx2 lm nx pae sse sse2 sse3 sse4_1 sse4_2 ssse3 vmx 
  bogomips: 57600 
  Speed: 800 MHz min/max: 800/4900 MHz Core speeds (MHz): 1: 800 2: 800 
  3: 800 4: 800 5: 800 6: 800 7: 800 8: 800 
Graphics:
  Device-1: NVIDIA TU104 [GeForce RTX 2070 SUPER] vendor: Gigabyte 
  driver: nvidia v: 470.63.01 bus ID: 01:00.0 chip ID: 10de:1e84 
  Display: x11 server: X.Org 1.20.11 driver: nvidia 
  resolution: 2560x1440~60Hz 
  OpenGL: renderer: llvmpipe (LLVM 10.0.0 256 bits) v: 4.5 Mesa 21.1.8 
  compat-v: 3.1 direct render: Yes 

[baronvonsatchel]

After posting this I tried updating the Kernel from the 5.4 you see above to the latest 5.11 in update manager, but that did not fix it either. I remain stumped by this one. Any chance it's the GPU itself that's dying and not a software issue?

[SMG]

Graphics:
Device-1: NVIDIA TU104 [GeForce RTX 2070 SUPER] vendor: Gigabyte
driver: nvidia v: 470.63.01 bus ID: 01:00.0 chip ID: 10de:1e84
Display: x11 server: X.Org 1.20.11 driver: nvidia
resolution: 2560x1440~60Hz
OpenGL: renderer: llvmpipe (LLVM 10.0.0 256 bits) v: 4.5 Mesa 21.1.8
compat-v: 3.1 direct render: Yes

Welcome to the forum, baronvonsatchel.

Where did you obtain the mesa? Stock version in LM20.2 is 21.0.3.

That's the only idea I have at the moment. If you want, I can take a look at your Xorg log file to see if there might be anything there which gives a clue as to what is happening.

Use this command in a terminal to upload the X Server log file to termbin:
cat /var/log/Xorg.0.log | nc termbin.com 9999
It will return with a url address that you should post in your next reply.

[baronvonsatchel]

I think I got it off the Mesa website, I was trying to get OpenGL to work in a Windows VM and updating Mesa was one of the things I tried. (Got it to work in the end, but discovered I needed 4.3 and VMware only marginally supports 4.1 so the whole project was a loss.)

Here is the link to the Xorg log: https://termbin.com/dwng3

[SMG]

I think I got it off the Mesa website, I was trying to get OpenGL to work in a Windows VM and updating Mesa was one of the things I tried.

Thanks for letting me know. I've seen others like that and wondered about the source.

People speak highly of the kisak-ppa for newer mesa versions and I've never had to help someone undo a change because of it, so that is usually what I recommend.

The Xorg log indicates you have a custom config file. Did you set it up to do something special? Normally most installs do not require one.
cat /etc/X11/xorg.conf would display the contents.

[baronvonsatchel]

The Nvidia X Server Settings app saves the configuration to it. Here is the contents:

Code: Select all

# nvidia-settings: X configuration file generated by nvidia-settings
# nvidia-settings:  version 470.57.01


Section "ServerLayout"
    Identifier     "Layout0"
    Screen      0  "Screen0" 0 0
    InputDevice    "Keyboard0" "CoreKeyboard"
    InputDevice    "Mouse0" "CorePointer"
    Option         "Xinerama" "0"
EndSection

Section "Files"
EndSection

Section "Module"
    Load           "dbe"
    Load           "extmod"
    Load           "type1"
    Load           "freetype"
    Load           "glx"
EndSection

Section "InputDevice"

    # generated from default
    Identifier     "Mouse0"
    Driver         "mouse"
    Option         "Protocol" "auto"
    Option         "Device" "/dev/psaux"
    Option         "Emulate3Buttons" "no"
    Option         "ZAxisMapping" "4 5"
EndSection

Section "InputDevice"

    # generated from default
    Identifier     "Keyboard0"
    Driver         "kbd"
EndSection

Section "Monitor"

    # HorizSync source: edid, VertRefresh source: edid
    Identifier     "Monitor0"
    VendorName     "Unknown"
    ModelName      "Dell S2719DGF"
    HorizSync       230.0 - 230.0
    VertRefresh     40.0 - 144.0
    Option         "DPMS"
EndSection

Section "Device"
    Identifier     "Device0"
    Driver         "nvidia"
    VendorName     "NVIDIA Corporation"
    BoardName      "NVIDIA GeForce RTX 2070 SUPER"
EndSection

Section "Screen"

# Removed Option "nvidiaXineramaInfoOrder" "DFP-0"
# Removed Option "metamodes" "DP-2: 2560x1440_144 +0+0, DP-0: nvidia-auto-select +2560+0"
    Identifier     "Screen0"
    Device         "Device0"
    Monitor        "Monitor0"
    DefaultDepth    24
    Option         "Stereo" "0"
    Option         "nvidiaXineramaInfoOrder" "DFP-3"
    Option         "metamodes" "DP-2: 2560x1440_144 +0+0, DP-0: 1920x1080_144 +2560+0"
    Option         "SLI" "Off"
    Option         "MultiGPU" "Off"
    Option         "BaseMosaic" "off"
    SubSection     "Display"
        Depth       24
    EndSubSection
EndSection

[SMG]

Do you happen to have a Timeshift snapshot of when the system was working to which you could revert? Maybe not since you re-installed.

The Nvidia X Server Settings app saves the configuration to it.

Yes, I am aware of that. It appears you are using it to combine your monitors into one. I didn't catch that in the Xorg log, but I can see that in your config file.

There are a set of lines I am used to seeing in the Xorg log with regards to acceleration that I am not seeing. I went back to someone I helped in the past with a similar issue and noticed you both had:

Code: Select all

[     6.912] (II) Initializing extension GLX
[     6.912] (II) Initializing extension GLX
[     6.912] (II) Indirect GLX disabled.
[     6.912] (II) GLX: Another vendor is already registered for screen 0

I asked for the output of glxinfo -B on that thread and we discovered VMWare was running the renderer. In that case, the person reinstalled, and everything worked properly on the next attempt.

VMWare may have nothing to do with your issue, but maybe the result of glxinfo -B will give us a clue as to what vendor has GLX registered for the screen.

[baronvonsatchel]

I noticed the same thing about "another vendor" and searched for it but nothing was super helpful. Here is what glxinfo -B had to say. Looks like the vendor is Mesa.

Code: Select all

name of display: :0
display: :0  screen: 0
direct rendering: Yes
Extended renderer info (GLX_MESA_query_renderer):
    Vendor: Mesa/X.org (0xffffffff)
    Device: llvmpipe (LLVM 10.0.0, 256 bits) (0xffffffff)
    Version: 21.1.8
    Accelerated: no
    Video memory: 32045MB
    Unified memory: no
    Preferred profile: core (0x1)
    Max core profile version: 4.5
    Max compat profile version: 3.1
    Max GLES1 profile version: 1.1
    Max GLES[23] profile version: 3.2
OpenGL vendor string: Mesa/X.org
OpenGL renderer string: llvmpipe (LLVM 10.0.0, 256 bits)
OpenGL core profile version string: 4.5 (Core Profile) Mesa 21.1.8
OpenGL core profile shading language version string: 4.50
OpenGL core profile context flags: (none)
OpenGL core profile profile mask: core profile

OpenGL version string: 3.1 Mesa 21.1.8
OpenGL shading language version string: 1.40
OpenGL context flags: (none)

OpenGL ES profile version string: OpenGL ES 3.2 Mesa 21.1.8
OpenGL ES profile shading language version string: OpenGL ES GLSL ES 3.20

I only have Timeshifts going back about a week. After the first scare of having to reinstall the OS I went out and bought an external drive and started keeping backups. I could try loading one from before I tried to set up the VM in case Mesa is the issue.

UPDATE: I Timeshifted back to before the VM debacle and it is now using the video card again like it is supposed to. So maybe it was installing the never version of Mesa that hijacked the system from Nvidia.

[SMG]

UPDATE: I Timeshifted back to before the VM debacle and it is now using the video card again like it is supposed to. So maybe it was installing the never version of Mesa that hijacked the system from Nvidia.

Thanks for the feedback.

Since your issue appears to be resolved, please go to your first post in this thread, click the pencil icon, and add [SOLVED] to the title so others know you are no longer seeking help on this issue.