search

bottlesdevs / Bottles

Run Windows software and games on Linux
https://usebottles.com
GNU General Public License v3.0
6.46k stars 276 forks source link

[Bug]: Unable to load libGLX_nvidia.so.0 #3133

Open Jpnyman opened 1 year ago

Jpnyman commented 1 year ago

Describe the bug

On a relatively fresh install of Ubuntu 22.04 with bottles installed from flatpak am running into the same error:

:     __EGL_EXTERNAL_PLATFORM_CONFIG_DIRS
F:     /etc/egl/egl_external_platform.d:/usr/lib/x86_64-linux-gnu/GL/egl/egl_external_platform.d:/usr/share/egl/egl_external_platform.d
F: Running 'bwrap --args 44 bottles'
18:40:47 (ERROR) Unable to load libGLX_nvidia.so.0 
18:40:47 (WARNING) Unable to locate libGLX_nvidia 
18:40:47 (INFO) Bottles Started! 
18:40:48 (INFO) Performing Bottles checks…

I have an unused Nvidia GPU which is bound to vfio and not used by anything except a virtual machine with my AMD GPU being used for everything Linux.

01:00.0 VGA compatible controller [0300]: NVIDIA Corporation GP102 [GeForce GTX 1080 Ti] [10de:1b06] (rev a1)
    Subsystem: NVIDIA Corporation GP102 [GeForce GTX 1080 Ti] [10de:1b06]
    Kernel driver in use: vfio-pci
02:00.0 VGA compatible controller [0300]: Advanced Micro Devices, Inc. [AMD/ATI] Ellesmere [Radeon RX 470/480/570/570X/580/580X/590] [1002:67df] (rev e7)
    Subsystem: Sapphire Technology Limited Nitro+ Radeon RX 570/580/590 [1da2:e366]
    Kernel driver in use: amdgpu

I am unable to use any rendering except for wine3d, with them all erroring out saying I do not have a vulkan capable GPU. For more context see my forum post, follow up and comment on github. Additionally this error gets thrown not only during startup but also seemingly randomly while navigating through the menus of bottles.

18:49:35 (WARNING) Unable to locate libGLX_nvidia 
18:50:31 (INFO) Wine Bridge is available. 
18:50:31 (INFO) Using Wine Bridge -- get_procs 
18:50:31 (ERROR) Unable to load libGLX_nvidia.so.0 
18:50:31 (WARNING) Unable to locate libGLX_nvidia

To Reproduce

Attempt to launch any program utilizing dxvk while having a Nvidia GPU bound to pci-vfio

Package

Flatpak from Flathub

Distribution

Ubuntu 22.04

Debugging Information

Official Package: true
Version: '51.9'
DE/WM: ubuntu
Display:
    X.org: true
    X.org (port): :99.0
    Wayland: true
Graphics:
    vendors:
        nvidia: &id002
            vendor: nvidia
            envs:
                __NV_PRIME_RENDER_OFFLOAD: '1'
                __GLX_VENDOR_LIBRARY_NAME: nvidia
                __VK_LAYER_NV_optimus: NVIDIA_only
            icd: ''
            nvngx_path: null
        amd: &id001
            vendor: amd
            envs:
                DRI_PRIME: '1'
            icd: /usr/lib/x86_64-linux-gnu/GL/vulkan/icd.d/radeon_icd.x86_64.json:/usr/lib/i386-linux-gnu/GL/vulkan/icd.d/radeon_icd.i686.json
    prime:
        integrated: *id001
        discrete: *id002
Kernel:
    Type: Linux
    Version: 6.5.7-zen+
Disk:
    Total: 8356495360
    Free: 8356339712
RAM:
    MemTotal: 15.6GiB
    MemAvailable: 12.0GiB
Bottles_envs: null

Troubleshooting Logs

No response

Additional context

No response

vercingetorx commented 1 year ago

It may be related to the way multiple GPUs are handled:

@staticmethod
    def assume_discrete(vendors: list):
        if "nvidia" in vendors and "amd" in vendors:
            return {"integrated": "amd", "discrete": "nvidia"}
        if "nvidia" in vendors and "intel" in vendors:
            return {"integrated": "intel", "discrete": "nvidia"}
        if "amd" in vendors and "intel" in vendors:
            return {"integrated": "intel", "discrete": "amd"}
        return {}

In your case it looks like the Nvidia GPU is being assumed to be the discrete GPU and the AMD is being assumed as integrated (as in the case of PRIME Render Offload) despite the fact that your AMD GPU is your primary discrete GPU. Hopefully someone with more knowledge can chime in.

In the mean time you could try setting the environment variable __VK_LAYER_NV_optimus=non_NVIDIA_only

Jpnyman commented 1 year ago

I appreciate the feedback.

I have overridden the environment variable in flatpak using sudo flatpak override --env=__VK_LAYER_NV_optimus=non_NVIDIA_only com.usebottles.bottles

Yet bottles gives __VK_LAYER_NV_optimus: NVIDIA_only

I even tried setting it globally in bashrc but no luck, printenv __VK_LAYER_NV_optimus giving non_NVIDIA_only but bottles still being NVIDIA_ONLY

vercingetorx commented 1 year ago

You're right, I tested a couple out and Bottles is not respecting these environment variables. Most likely a separate bug... I don't know there is an easy fix without that being addressed first.

Jpnyman commented 1 year ago

No worries, I appreciate the feedback. Any other pointers?

vercingetorx commented 1 year ago

No not really. The only other thing you might try if you haven't already is setting environment variables within a specific bottle's settings rather than for Bottles itself. Also, I didn't think of it but if you don't have NVIDIA drivers installed then those NVIDIA env variables won't work either way. Setting DRI_PRIME=0 might work as it instructs Linux to use your integrated graphics, which apparently in this case is your AMD GPU. That's all I got.

Jpnyman commented 1 year ago

My AMD GPU is not CPU integrated graphics. I have two dedicated graphics cards. I set the environment both as a flatpak override and in bashrc but no luck. I appreciate the help though.

vercingetorx commented 1 year ago

My AMD GPU is not CPU integrated graphics. I have two dedicated graphics cards.

Correct, but:

prime:
        integrated: *id001
        discrete: *id002

Anyway, good luck.

vercingetorx commented 12 months ago

Hi, I am still looking into this. Would you mind running flatpak run --command=sh com.usebottles.bottles and then ls /usr/lib/x86_64-linux-gnu/GL/vulkan/icd.d/ and post the output. Also want to confirm you are using Mesa drivers?

cbsrobot commented 11 months ago

I have probably the same Problem on Rocky Linux 9 with a NVIDIA card. If it's not related, please let me know.

here is my output:

flatpak run --command=sh com.usebottles.bottles
[📦 com.usebottles.bottles ~]$ ls /usr/lib/x86_64-linux-gnu/GL/vulkan/icd.d/
intel_hasvk_icd.x86_64.json  intel_icd.x86_64.json  lvp_icd.x86_64.json  radeon_icd.x86_64.json  virtio_icd.x86_64.json
[📦 com.usebottles.bottles ~]$

this is the output of running bottles:

➜  ~ flatpak run com.usebottles.bottles 
libEGL warning: DRI3: Screen seems not DRI3 capable
libEGL warning: DRI2: failed to authenticate
libEGL warning: DRI3: Screen seems not DRI3 capable
libEGL warning: DRI2: failed to authenticate
13:19:08 (ERROR) Unable to load libGLX_nvidia.so.0 
13:19:08 (WARNING) Unable to locate libGLX_nvidia 
13:19:08 (INFO) Bottles Started! 
13:19:08 (INFO) Performing Bottles checks… 
13:19:08 (INFO) Dxvks found:
     - dxvk-2.1
     - dxvk-2.3

13:19:08 (INFO) Vkd3ds found:
     - vkd3d-proton-2.8-1-1e663e6

13:19:08 (INFO) Nvapis found:
     - dxvk-nvapi-v0.6-1-0827efb

13:19:08 (INFO) Latencyflexs found:
     - latencyflex-v0.1.1

13:19:08 (INFO) Runners found:
     - soda-8.0-2
     - soda-7.0-9
     - soda-7.0-8
     - sys-wine-8.0.2

13:19:09 (ERROR) Unable to load libGLX_nvidia.so.0 
13:19:09 (WARNING) Unable to locate libGLX_nvidia 
13:19:09 (ERROR) Nvidia driver libraries haven't been found. DLSS might not work! 
13:19:09 (INFO) Bottles found:
[...]
vercingetorx commented 11 months ago

Thanks for providing that information. Can you please provide the Bottles debug information as well? Are you using a single Nvidia GPU? Which drivers are you using?

cbsrobot commented 11 months ago

I have a Nvidia GPU with the proprietary driver: NVIDIA-SMI 545.23.08 

$ flatpak run --command=bottles-cli com.usebottles.bottles info health-check 
20:03:27 (ERROR) Unable to load libGLX_nvidia.so.0 
20:03:27 (WARNING) Unable to locate libGLX_nvidia 

** (process:349686): WARNING **: 20:03:27.536: Error writing credentials to socket: Error sending message: Broken pipe
20:03:27 (ERROR) Unable to load libGLX_nvidia.so.0 
20:03:27 (WARNING) Unable to locate libGLX_nvidia 
Official Package: true
Version: '51.9'
DE/WM: gnome
Display:
    X.org: true
    X.org (port): :99.0
    Wayland: false
Graphics:
    vendors:
        nvidia:
            vendor: nvidia
            envs:
                __NV_PRIME_RENDER_OFFLOAD: '1'
                __GLX_VENDOR_LIBRARY_NAME: nvidia
                __VK_LAYER_NV_optimus: NVIDIA_only
            icd: ''
            nvngx_path: null
    prime:
        integrated: null
        discrete: null
Kernel:
    Type: Linux
    Version: 5.14.0-362.8.1.el9_3.x86_64
Disk:
    Total: 134871797760
    Free: 134871646208
RAM:
    MemTotal: 251.2GiB
    MemAvailable: 210.7GiB
Bottles_envs: null
cbsrobot commented 11 months ago

After digging a bit deeper it seems Nvidia CUDA team released version 545.23.08, but the official version is 545.23.06 and flatpak uses the official version. After pinning the Nvidia driver to 545.23.06 everything works again. Not sure the original issue is solved though.

Here are the relevant links to read more about it: https://github.com/flathub/org.freedesktop.Platform.GL.nvidia/pull/214 https://forums.developer.nvidia.com/t/545-23-08-is-not-listed-in-the-nvidia-driver-downloads-page/273181/10

vercingetorx commented 11 months ago

Weird. Definitely a different issue than the op; for whatever reason Bottles is/was unable to load your Nvidia library file.

There isn't much to be done on our side as loading this file is handled by a Python Standard Library method.

  try:
      libglx_nvidia = CDLL("libGLX_nvidia.so.0")
  except OSError:
      logging.error("Unable to load libGLX_nvidia.so.0")
      return None

From the Python documentation

When a dependent DLL of the loaded DLL is not found, a OSError error is raised..

My guess is this is the issue on your end.

Vortelf commented 11 months ago

TL;DR: Bottles (and some other apps) are missing the 64bit dependency of the Flatpak Nvidia driver.

To fix the issue, just install it: flatpak install flathub org.freedesktop.Platform.GL.nvidia-545-29-06.

I was suffering from the same issue. I was clueless as to why it doesn't load the library when it is present, so I started digging.

flatpak run --command=bottles-cli com.usebottles.bottles info health-check returns everything as it should.

18:43:03 (ERROR) Unable to load libGLX_nvidia.so.0
18:43:03 (WARNING) Unable to locate libGLX_nvidia
18:43:03 (ERROR) Unable to load libGLX_nvidia.so.0
18:43:03 (WARNING) Unable to locate libGLX_nvidia
Official Package: true
Version: '51.9'
DE/WM: plasmawayland
Display:
    X.org: true
    X.org (port): :0
    Wayland: true
Graphics:
    vendors:
        nvidia: &id002
            vendor: nvidia
            envs:
                __NV_PRIME_RENDER_OFFLOAD: '1'
                __GLX_VENDOR_LIBRARY_NAME: nvidia
                __VK_LAYER_NV_optimus: NVIDIA_only
            icd: /usr/lib/i386-linux-gnu/GL/vulkan/icd.d/nvidia_icd.json
            nvngx_path: null
        intel: &id001
            vendor: intel
            envs:
                DRI_PRIME: '1'
            icd: /usr/lib/x86_64-linux-gnu/GL/vulkan/icd.d/intel_hasvk_icd.x86_64.json:/usr/lib/x86_64-linux-gnu/GL/vulkan/icd.d/intel_icd.x86_64.json:/usr/lib/i386-linux-gnu/GL/vulkan/icd.d/intel_hasvk_icd.i686.json:/usr/lib/i386-linux-gnu/GL/vulkan/icd.d/intel_icd.i686.json
    prime:
        integrated: *id001
        discrete: *id002
Kernel:
    Type: Linux
    Version: 6.6.7-arch1-1
Disk:
    Total: 5968003072
    Free: 5967843328
RAM:
    MemTotal: 11.1GiB
    MemAvailable: 4.7GiB
Bottles_envs: null

Here's the output of some debugging I did. Apparently, the Nvidia lib inside the Flatpak sandbox is not a 64-bit one. 

[📦 com.usebottles.bottles ~]$ cat /usr/lib/i386-linux-gnu/GL/vulkan/icd.d/nvidia_icd.json
{
    "file_format_version" : "1.0.0",
    "ICD": {
        "library_path": "libGLX_nvidia.so.0",
        "api_version" : "1.3.260"
    }
}
[📦 com.usebottles.bottles ~]$ find / -name "libGLX_nvidia*"
/app/lib/i386-linux-gnu/GL/nvidia-545-29-06/extra/libGLX_nvidia.so.545.29.06
/app/lib/i386-linux-gnu/GL/nvidia-545-29-06/extra/libGLX_nvidia.so.0
---
[📦 com.usebottles.bottles ~]$ ldconfig -p | grep libGLX
    libGLX_nvidia.so.0 (libc6) => /app/lib/i386-linux-gnu/GL/nvidia-545-29-06/lib/libGLX_nvidia.so.0
    libGLX_mesa.so.0 (libc6,x86-64) => /usr/lib/x86_64-linux-gnu/GL/default/lib/libGLX_mesa.so.0
    libGLX_mesa.so.0 (libc6) => /app/lib/i386-linux-gnu/GL/default/lib/libGLX_mesa.so.0
    libGLX.so.0 (libc6,x86-64) => /usr/lib/x86_64-linux-gnu/libGLX.so.0
    libGLX.so.0 (libc6) => /app/lib/i386-linux-gnu/libGLX.so.0
---
[📦 com.usebottles.bottles ~]$ python
Python 3.10.13 (main, Nov 10 2011, 15:00:00) [GCC 12.2.0] on linux
Type "help", "copyright", "credits" or "license" for more information.
>>> from ctypes import *
>>> libglx_nvidia = CDLL("libGLX_nvidia.so.0")
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
  File "/usr/lib/python3.10/ctypes/__init__.py", line 374, in __init__
    self._handle = _dlopen(self._name, mode)
OSError: libGLX_nvidia.so.0: cannot open shared object file: No such file or directory
>>> libglx_nvidia = CDLL("/app/lib/i386-linux-gnu/GL/nvidia-545-29-06/lib/libGLX_nvidia.so.0")
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
  File "/usr/lib/python3.10/ctypes/__init__.py", line 374, in __init__
    self._handle = _dlopen(self._name, mode)
OSError: /app/lib/i386-linux-gnu/GL/nvidia-545-29-06/lib/libGLX_nvidia.so.0: wrong ELF class: ELFCLASS32
---
[📦 com.usebottles.bottles ~]$ file /app/lib/i386-linux-gnu/GL/nvidia-545-29-06/lib/libGLX_nvidia.so.545.29.06
/app/lib/i386-linux-gnu/GL/nvidia-545-29-06/lib/libGLX_nvidia.so.545.29.06: ELF 32-bit LSB shared object, Intel 80386, version 1 (SYSV), dynamically linked, BuildID[sha1]=9a5a5afeda1131caa1a2e953b7f000414431bc01, stripped

Comparison to running this locally where I have the same driver installed via pacman.

$ file /usr/lib/libGLX_nvidia.so.545.29.06
/usr/lib/libGLX_nvidia.so.545.29.06: ELF 64-bit LSB shared object, x86-64, version 1 (SYSV), dynamically linked, BuildID[sha1]=c8b8a791af8635327f949c0e0d8ebdcc2d1a077b, stripped
---
Python 3.11.6 (main, Nov 14 2023, 09:36:21) [GCC 13.2.1 20230801] on linux
Type "help", "copyright", "credits" or "license" for more information.
>>> from ctypes import *
>>> libglx_nvidia = CDLL("libGLX_nvidia.so.0")
>>> libglx_nvidia
<CDLL 'libGLX_nvidia.so.0', handle 55ba63ba7560 at 0x7f23334b4490>

I decided to check what I have installed in Flatpack and I was surprised that only a 32-bit package was listed as a dependency. 

$ flatpak list | grep "nvidia"
nvidia-545-29-06    org.freedesktop.Platform.GL32.nvidia-545-29-06      1.4 system

A quick Google search led me here, which confirmed that there are both 32 and 64-bit packages and I am missing one. github.com/flathub/org.freedesktop.Platform.GL.nvidia

After I installed it, I ran my bottle with flatpak run --command=bottles-cli com.usebottles.bottles run and the process appeared in nvidia-smi.

XZ02R commented 10 months ago

Hi, I have the exact same problem as OP and our setups are identical.

17:52:01 (INFO) Catalog installers loaded 
17:52:01 (INFO) Catalog dependencies loaded 
17:52:02 (INFO) Catalog components loaded 
17:52:03 (INFO) Setting Key sync=fsync for bottle Gaming… 
17:52:03 (INFO) There is no running wineserver. 
17:52:20 (INFO) Launching an executable… 
17:52:20 (ERROR) Unable to load libGLX_nvidia.so.0 
17:52:20 (WARNING) Unable to locate libGLX_nvidia 
17:52:20 (WARNING) Nouveau driver detected, this may cause issues 
17:52:23 (INFO) Using EasyAntiCheat runtime 
17:52:23 (INFO) Using BattlEye runtime

I also have a multi GPU system where one card has a vfio-pci driver attached and is only used for virtualized workloads.

01:00.0 VGA compatible controller [0300]: NVIDIA Corporation GA106 [GeForce RTX 3060 Lite Hash Rate] [10de:2504] (rev a1) (prog-if 00 [VGA controller])
        Subsystem: ASUSTeK Computer Inc. GA106 [GeForce RTX 3060 Lite Hash Rate] [1043:8810]
        Flags: fast devsel, IRQ 16, IOMMU group 2
        Memory at f6000000 (32-bit, non-prefetchable) [size=16M]
        Memory at c0000000 (64-bit, prefetchable) [size=256M]
        Memory at d0000000 (64-bit, prefetchable) [size=32M]
        I/O ports at e000 [size=128]
        Expansion ROM at 000c0000 [disabled] [size=128K]
        Capabilities: <access denied>
        Kernel driver in use: vfio-pci
        Kernel modules: nouveau

01:00.1 Audio device [0403]: NVIDIA Corporation GA106 High Definition Audio Controller [10de:228e] (rev a1)
        Subsystem: ASUSTeK Computer Inc. GA106 High Definition Audio Controller [1043:8810]
        Flags: fast devsel, IRQ 17, IOMMU group 2
        Memory at f7080000 (32-bit, non-prefetchable) [size=16K]
        Capabilities: <access denied>
        Kernel driver in use: vfio-pci
        Kernel modules: snd_hda_intel

07:00.0 VGA compatible controller [0300]: Advanced Micro Devices, Inc. [AMD/ATI] Ellesmere [Radeon RX 470/480/570/570X/580/580X/590] [1002:67df] (rev c7) (prog-if 00 [VGA controller])
        Subsystem: XFX Pine Group Inc. Radeon RX 480 [1682:9480]
        Flags: bus master, fast devsel, latency 0, IRQ 132, IOMMU group 15
        Memory at e0000000 (64-bit, prefetchable) [size=256M]
        Memory at f0000000 (64-bit, prefetchable) [size=2M]
        I/O ports at c000 [size=256]
        Memory at f7100000 (32-bit, non-prefetchable) [size=256K]
        Expansion ROM at f7140000 [disabled] [size=128K]
        Capabilities: <access denied>
        Kernel driver in use: amdgpu
        Kernel modules: amdgpu

07:00.1 Audio device [0403]: Advanced Micro Devices, Inc. [AMD/ATI] Ellesmere HDMI Audio [Radeon RX 470/480 / 570/580/590] [1002:aaf0]
        Subsystem: XFX Pine Group Inc. Ellesmere HDMI Audio [Radeon RX 470/480 / 570/580/590] [1682:aaf0]
        Flags: bus master, fast devsel, latency 0, IRQ 135, IOMMU group 15
        Memory at f7160000 (64-bit, non-prefetchable) [size=16K]
        Capabilities: <access denied>
        Kernel driver in use: snd_hda_intel
        Kernel modules: snd_hda_intel

Additional info on this setup is that my motherboard doesn't support choosing a primary display gpu, so it treats the Nvidia card on the top slot as the primary card until the part in the boot process where grub loads the OS, There the nvidia driver will get blacklisted so that the DE (Plasma) is forced to use the second card as the display.

Package

Flathub

Distribution

Debian 12 Bookworm

Steps to reproduce

  1. Create a new bottle with the gaming preset
  2. Run executable with the bottle

I have also tried the environment variables listed earlier to make the bottle use my "integrated" gpu (the second discrete gpu) with no success.

Applications used for testing

Debugging Information

Official Package: true
Version: '51.10'
DE/WM: plasmawayland
Display:
    X.org: true
    X.org (port): :1
    Wayland: true
Graphics:
    vendors:
        nvidia: &amp;id002
            vendor: nvidia
            envs:
                DRI_PRIME: '1'
            icd: ''
            nvngx_path: null
        amd: &amp;id001
            vendor: amd
            envs:
                DRI_PRIME: '1'
            icd: /usr/lib/x86_64-linux-gnu/GL/vulkan/icd.d/radeon_icd.x86_64.json
    prime:
        integrated: *id001
        discrete: *id002
Kernel:
    Type: Linux
    Version: 6.1.0-16-amd64
Disk:
    Total: 33646112768
    Free: 33645953024
RAM:
    MemTotal: 62.7GiB
    MemAvailable: 31.7GiB
Bottles_envs: null
Hyphaed commented 4 months ago

same, after installing nvidia 560 drivers on Ubuntu

I built flatpak from source but didnt solve the problem

any workaround?

Zawaken commented 2 months ago

Running into the same issue here, not sure if this will fix the issue, but I haven't gotten around to setting vfio-pci for the Nvidia GPU after installing NixOS

0a:00.0 VGA compatible controller: NVIDIA Corporation GP104 [GeForce GTX 1080] (rev a1)
    Subsystem: ASUSTeK Computer Inc. Device 85e8
    Kernel driver in use: nouveau
    Kernel modules: nvidiafb, nouveau
--
0d:00.0 VGA compatible controller: Advanced Micro Devices, Inc. [AMD/ATI] Navi 21 [Radeon RX 6800/6800 XT / 6900 XT] (rev c0)
    Subsystem: XFX Limited Speedster MERC 319 AMD Radeon RX 6900 XT Black
    Kernel driver in use: amdgpu
    Kernel modules: amdgpu

On my Gigabyte Aorus X570 motherboard I have set the second slot, where the AMD card is as the initial display output, I will try seeing if I can manage to blacklist the nvidia card and set it to use the vfio-pci driver.

Package

flathub

Distribution

NixOS 24.11

Steps to reproduce

  1. Create a new bottle with the gaming preset
  2. Try installing battle.net in the bottle

Application

I tried installing battle.net

Debugging information

Official Package: true
Version: '51.13'
DE/WM: null
Display:
    X.org: true
    X.org (port): :0
    Wayland: true
Graphics:
    vendors:
        nvidia: &amp;id002
            vendor: nvidia
            envs:
                __NV_PRIME_RENDER_OFFLOAD: '1'
                __GLX_VENDOR_LIBRARY_NAME: nvidia
                __VK_LAYER_NV_optimus: NVIDIA_only
            icd: ''
            nvngx_path: null
        amd: &amp;id001
            vendor: amd
            envs:
                DRI_PRIME: '1'
            icd: /usr/lib/x86_64-linux-gnu/GL/vulkan/icd.d/radeon_icd.x86_64.json:/usr/lib/i386-linux-gnu/GL/vulkan/icd.d/radeon_icd.i686.json
    prime:
        integrated: *id001
        discrete: *id002
Kernel:
    Type: Linux
    Version: 6.6.50
Disk:
    Total: 16795332608
    Free: 16795185152
RAM:
    MemTotal: 31.3GiB
    MemAvailable: 26.6GiB
Bottles_envs: null
xblackvenomx commented 1 week ago

Hi

I was having the same issue a moment ago, running void linux with xfce4 and the flatpak installation of bottles, with a iGPU and an nvidia dGPU (4060 ti 16gb, to be precise, paired with a 4750G cpu). It turned out, as I did system updates and flatpak updates in parallel, I needed to fire flatpak update again to catch up on the correlating GL package vor the updated nvidia drivers. That did the trick, so my assumption is that there is some sort of version conflict.

Maybe that'll help some of you.

Cheers.

BradHeff commented 1 week ago

I was having the same issue, Drivers from Nvidia site installed. Single GPU GeForce RTX 4080. Just run flatpak update it should search for the missing drivers on flatpak side of things.


:: flatpak update
Looking for updates…

Info: runtime org.kde.Platform branch 6.6 is end-of-life, with reason:
   We strongly recommend moving to the latest stable version of the Platform and SDK
Info: applications using this runtime:
   com.obsproject.Studio

        ID                                              Branch Op Remote  Download
 1. [✓] org.freedesktop.Platform.GL.nvidia-550-127-05   1.4    i  flathub 308.0 MB / 308.1 MB
 2. [\] org.freedesktop.Platform.GL32.nvidia-550-127-05 1.4    i  flathub 173.0 MB / 308.0 MB```