Second GPU make Garuda Linux strange

[19topgun93]

Hello

I have now two Geforce GTX 1070 in my Computer. I want one pathtrough to a Win 10 VM. I used GitHub - bryansteiner/gpu-passthrough-tutorial link, found in another thread here on Garuda.

But now I can only post garuda-inxi to this:

System:
  Kernel: 6.9.6-zen1-1-zen arch: x86_64 bits: 64 compiler: gcc v: 14.1.1
    clocksource: tsc avail: hpet,acpi_pm
    parameters: BOOT_IMAGE=/@/boot/vmlinuz-linux-zen
    root=UUID=9bdc1bb8-710e-4a2b-a1fc-d771d79d9fb1 rw rootflags=subvol=@
    quiet resume=UUID=3de67b4e-0081-4321-8b87-d6348302d4da loglevel=3 ibt=off
  Desktop: Cinnamon v: 6.2.2 tk: GTK v: 3.24.42 wm: Muffin v: 6.2.0 tools:
    avail: cinnamon-screensaver,xautolock vt: 7 dm: LightDM v: 1.32.0
    Distro: Garuda base: Arch Linux
Machine:
  Type: Desktop Mobo: ASUSTeK model: CROSSHAIR VI HERO v: Rev 1.xx
    serial: <superuser required> part-nu: SKU UEFI: American Megatrends v: 7704
    date: 12/16/2019
CPU:
  Info: model: AMD Ryzen 9 3900X bits: 64 type: MT MCP arch: Zen 2 gen: 3
    level: v3 note: check built: 2020-22 process: TSMC n7 (7nm)
    family: 0x17 (23) model-id: 0x71 (113) stepping: 0 microcode: 0x8701013
  Topology: cpus: 1x cores: 12 tpc: 2 threads: 24 smt: enabled cache:
    L1: 768 KiB desc: d-12x32 KiB; i-12x32 KiB L2: 6 MiB desc: 12x512 KiB
    L3: 64 MiB desc: 4x16 MiB
  Speed (MHz): avg: 2415 high: 3823 min/max: 2200/4672 boost: enabled
    scaling: driver: acpi-cpufreq governor: schedutil cores: 1: 3699 2: 2200
    3: 2043 4: 2200 5: 2021 6: 3823 7: 2200 8: 2200 9: 2200 10: 2278 11: 2200
    12: 2200 13: 3598 14: 2200 15: 2049 16: 2072 17: 2200 18: 3600 19: 2200
    20: 2200 21: 2200 22: 2200 23: 2056 24: 2137 bogomips: 182391
  Flags: avx avx2 ht lm nx pae sse sse2 sse3 sse4_1 sse4_2 sse4a ssse3 svm
  Vulnerabilities: <filter>

At the Beginnig the PC is a bit laggy, but then it is normal. If I want to create the VM it is hanging in the first creating process of Virt-Manager if I click install. That I cannot post the full garuda-inxi I think it is the second GPU. Are there commands to discover why this is happening?

I also look at the moment in PCI passthrough via OVMF - ArchWiki. So I look to make the changes with garuda-dracut-support I think.

[tbg]

You are not the only one complaining about recent problems using dual GPU’s.

No video after sddm login when 2nd GPU is connected

As title says when i plug in a 2nd gpu (1650) i get no video after sddm login screen, the os is running since i can still hear sounds when plugging and unplugging devices. I already tried every port in each graphics card and still no video. This is on kde plasma 6 wayland and nvidia 555 drivers. I used to run this exact config on fedora 39 (on nvidia 535 drivers) with no troubles, used the second gpu for vfio pass-through for a windows vm.

VFIO/GPU Passthrough explanation pattern

Hello I have two Geforce GTX 1070 in my Computer. I want to passthrough one of them to a VM to run Windows. In VFIO/GPU Passthrough there is a guide GitHub - bryansteiner/gpu-passthrough-tutorial But I don’t get the pattern for Make sure to substitute the correct bus addresses for the devices you’d like to passthrough to your VM (in my case a GPU and SSD). Just in case it’s still unclear, you get the virsh PCI device IDs from the iommu.sh script’s output. Translate the address…

My guess would be that it’s a bug with the nvidia proprietary driver causing this. I have not used any nvidia products in quite some time, so my I’m probably not the best one to offer advice.

Good luck resolving your issue.

---

Edited to add recent bug report:

Bug :nvidia-drivers-555.58 DVI output not working ->black screen

[19topgun93]

I’m rollback to 550 because of the Error with the DVI. So, I will see. Next Setup will be full AMD

[tbg]

See my above post for recent edits that added further info.

[19topgun93]

GPU timeout | lockup - #3 by nPHYN1T3 - Linux - NVIDIA Developer Forums I think I found a thread by NVIDIA about this. So I have to wait now for a fixed. A info for everybody with a second NVIDIA GPU.