Search results for query: vendor-reset

I removed the startup option on TrueNAS, rebooted the host, and ran smartctl to check for things: nvme0n1 is the boot SSD, so ignoring that drive. So the NVMe look good, right? Spinners in the next post due to character limits.

Ich klinke mich mal hier mit meiner Kingston Data Center DC2000B ein: root@pve:~# smartctl -a /dev/nvme0n1 smartctl 7.4 2024-10-15 r5620 [x86_64-linux-6.17.13-2-pve] (local build) Copyright (C) 2002-23, Bruce Allen, Christian Franke, www.smartmontools.org === START OF INFORMATION SECTION ===...

I made all the changes and it start working under ubuntu (even after restart the vm), but no changes under windows, just black screen

This module no longer exists: https://pve.proxmox.com/pve-docs/pve-admin-guide.html#qm_pci_passthrough vendor-reset does not support your GPU. None of these drivers apply to your GPU, therefore this does nothing. intell_iommu=on is spelled wrong and is not needed on Proxmox kernels since...

...proxmox setting: ``` OS: Proxmox VE 9.1.6 x86_64 Kernel: Linux 6.17.13-2-pve /etc/modules: vfio vfio_iommu_type1 vfio_pci vfio_virqfd vendor-reset /etc/modeprobe.d/blacklist.conf and /etc/modeprobe.d/pve-blacklist.conf blacklist nouveau blacklist nvidia blacklist nvidiafb blacklist radeon...

So turns out "stuck in D3" was my PSU kicking the bucket. Haven't dared to update windows drivers past 32.0.101.8331 or upgrade card firmware, guess that's how it's going to be for a while. Dunno if it helps someone: i've changed my hookscript to enable AI stuff in lxc and keep on demand...

Hi, the error PCI_NUM_PINS error could happens because the NVIDIA T4 doesn’t reset properly after a VM shutdown, leaving KVM with invalid PCI interrupt pins. You could try to bind the T4 to VFIO at boot, disable PCIe ASPM and probably vendor-reset script to reset a GPU to a clean state.

...intel_iommu=on iommu=pt. Zusätzlich bitte posten: dmesg | grep -i vfio dmesg | grep -i iommu cat /etc/pve/qemu-server/<VMID>.conf 5. Vendor-Reset-Patch: AMD GPUs brauchen manchmal den vendor-reset Kernel-Patch: apt install pve-headers-$(uname -r) dkms git git clone...

So, what did you do to make progress between the initial post and before vendor-reset, etc?

Thanks to: https://forum.proxmox.com/threads/success-asus-intel-nuc-13-pro-i5-1340h-igpu-passthrough-on-proxmox-ve-9.180742 HARDWARE Model: Mini PC with Intel N100 or Fujitsu Esprimo Q957 with Intel i5 7500 iGPU: Intel UHD Graphics, no additional GPU installed Host OS: Proxmox VE 9.1.5 Guest...

Man I feel you. I have an A750 and got it working just the other day and then I did the driver update and now I have not been able to get it working again. I downgraded the firmware to previous version as well, but no dice. Been banging my head against the wall with this stupid GPU. On VMware...

It seems to be supported (but I don't own this specific GPU). It is in various guides and the vendor-reset github site itself. It only tells me that there was an issue, not what the issue actually was. It does refer to a log file which can help you find more information. If you want to...

...KDIR=/lib/modules/6.17.4-2-pve/build Error! Bad return status for module build on kernel: 6.17.4-2-pve (x86_64) Consult /var/lib/dkms/vendor-reset/0.1.1/build/make.log for more information. So tried earlier proxmox-boot-tool kernel pin 6.17.2-1-pve running dkms install gives this Error...

using this repo: https://github.com/lowell80/vendor-reset.git I can get ``` root@wtr-pro:~# cat /sys/bus/pci/devices/0000\:04\:00.0/reset_method device_specific ``` however it still is not working as expected. I am trying to get the rom file out of the gpu, to see if that is the issue.

cat sriov_totalvfs

Sure, this is the output root@prox:/sys/devices/pci0000:e0/0000:e0:01.1/0000:e1:00.0/0000:e2:01.0/0000:e3:00.0# ls aer config device iommu_group max_link_speed power reset_method resource2_resize sriov_numvfs...

it is in both, shutdown or reboot. I am testing with vendor-reset now to see if that helps

This command is not correct: the period ( .) at the end is missing. Also, the vendor-reset code does not compile with the recent Proxmox kernel version: https://github.com/gnif/vendor-reset/issues/87 . What GPU do you have (exactly)? Is it explicitly supported by vendor-reset? Otherwise helping...

...I get as far as dkms install root@pve-giga:~# cd vendor-reset root@pve-giga:~/vendor-reset# dkms install Error! Arguments <module> and <module-version> are not specified. Usage: add <module>/<module-version> or add -m <module>/<module-version> or add -m <module> -v...

After Intel driver update 32.0.101.8425 which came with a firmware update, and passing the new vbios, the issues returned without using hookscript. First slot takes out the whole host. Second slot keeps host alive but unresponsive and "kernel: vfio-pci 0000:04:00.0: not ready 32767ms after FLR...