Recent content by Eddy Gordo From Tekken

The passthrough works with a work around, but its rather annoying and I'd like to correct it... When I initially start the VM, I have to physically disconnect all devices connected to the PCI card, it will start with no errors. As soon as i connect any usb device, it will freeze and the host...

Thank you for the help! It put me on a path to solving my problem MY SOLUTION... I was at a LAN party and I had a crossover cable with me. so I connected my pve box to a Win10 pc directly with the crossover cable. on the windows pc settings > Ethernet > Network # > IP Settings > edit > select...

thank you, I can not reach the web gui or ssh at the moment to change the network settings. and since the 2 guests are loading immediatly and taking control of the gpu, pve doesn't boot to a console. i'm too far from my home network to simply take it back. is there some way/option during pve...

At my home network I interact with host using the web gui or/and ssh using a local address 192.168.1.___ i took the pc to a new network, 192.168.21.___ and i am unable to connect to the host. the current configuration starts two win10 guests with gpu passthrough automatically when the the host...

I've upgraded to pve 6.3-3 running kernel: 5.4.78-2-pve after upgrade, the amd reset problem returned. this time i used the latest workaround, which is to install a kernel module https://www.nicksherlock.com/2020/11/working-around-the-amd-gpu-reset-bug-on-proxmox/ I have to two win10 vms. vm1...

/etc/modprobe.d/vfio.conf options vfio-pci ids=1002:731f,1002:ab38 disable_vga=1 options vfio-pci ids=10de:1187,10de:0e0a disable_vga=1 options vfio-pci ids=1912:0014 I ran this according to the wiki, then reboot update-initramfs -u -k all

the card worked when tested the card with ubuntu 20.04 live desktop

card is new, and i've only tried it in this rig. i'm currently using the VM without that card passed through. i plugged a mouse into it, then at a node shell put in lsusb. it does not show up there. i will try to boot into a live distro... lspci -n -s 21:00.0 21:00.0 0c03: 1912:0014 (rev 03)

I'm still stuck here...has anyone come across this?

find /sys/kernel/iommu_groups/ -type l /sys/kernel/iommu_groups/55/devices/0000:42:0a.0 /sys/kernel/iommu_groups/83/devices/0000:62:00.0 /sys/kernel/iommu_groups/17/devices/0000:05:00.0 /sys/kernel/iommu_groups/45/devices/0000:40:07.1 /sys/kernel/iommu_groups/73/devices/0000:60:01.0...

dmesg output was too long to post in reply, I've added a .txt file to the original post. Thank You for helping me

I took a suggestion to change the config args line to args: -cpu 'host,+kvm_pv_unhalt,+kvm_pv_eoi,hv_vendor_id=NV43FIX,hidden=1' the VM would not start until i reverted back to args: -cpu 'host,+kvm_pv_unhalt,+kvm_pv_eoi,hv_vendor_id=NV43FIX,kvm=off'

Could someone help me determine what I have done incorrectly. I am attempting to passthrough this card https://www.startech.com/Cards-Adapters/USB-3.0/Cards/3-port-pcie-usb-card-gigabit-ethernet~PEXUSB3S3GE to this VM agent: 1 args: -cpu...

My problem was due to the NAVI reset bug for AMD GPUs. I solved it with the solution here https://forum.level1techs.com/t/navi-reset-kernel-patch/147547/134 patching the kernel worked in my situation with multiple GPU installed I would not imagine this will work for NVIDIA GPUs, and I would not...

hypothetically...If I'm not savvy enough to compile the kernal and patch myself, it appears I have two options. are there any other choices? 1. buy NVIDIA GPU 2. switch to vmware esxi 3. ?