Proxmox - PCIe passthrough - Nvidia - Tesla - P40 as Host device
₹600-1500 INR
Closed
Posted 8 months ago
₹600-1500 INR
Paid on delivery
I am looking for a freelancer who has experience with Proxmox and PCIe passthrough for Nvidia Tesla P40. The primary purpose of the passthrough is for "Other" (client to specify).
Ideal Skills and Experience:
- Proxmox expertise, specifically with version 8.0.4
- Experience with Nvidia Tesla P40 GPU installation on Proxmox server
- Proficient in configuring and implementing PCIe passthrough
- Understanding of virtual machine GPU acceleration and related technologies
We have already did the pcie as RAW device / Host :
lspci -nnk | grep -i nvidia -A2
37:00.0 3D controller [0302]: NVIDIA Corporation GP102GL [Tesla P40] [10de:1b38] (rev a1)
Subsystem: NVIDIA Corporation GP102GL [Tesla P40] [10de:11d9]
Kernel driver in use: vfio-pci
Kernel modules: nvidiafb, nouveau
5b:00.0 PCI bridge [0604]: Intel Corporation Sky Lake-E PCI Express Root Port A [8086:2030] (rev 04)
Subsystem: Hewlett Packard Enterprise Sky Lake-E PCI Express Root Port A [1590:00ea]
--
af:00.0 3D controller [0302]: NVIDIA Corporation GP102GL [Tesla P40] [10de:1b38] (rev a1)
Subsystem: NVIDIA Corporation GP102GL [Tesla P40] [10de:11d9]
Kernel driver in use: vfio-pci
Kernel modules: nvidiafb, nouveau
d7:00.0 PCI bridge [0604]: Intel Corporation Sky Lake-E PCI Express Root Port A [8086:2030] (rev 04)
Subsystem: Hewlett Packard Enterprise Sky Lake-E PCI Express Root Port A [1590:00ea]
But when we use NVIDIA-SMI : NVIDIA-SMI has failed because it couldn't communicate with the NVIDIA driver. Make sure that the latest NVIDIA driver is installed and running.
DMESG : NVRM: This PCI I/O region assigned to your NVIDIA device is invalid:
NVRM: BAR0 is 0M @ 0x0 (PCI:0000:01:00.0)
nvidia: probe of 0000:01:00.0 failed with error -1
NVRM: The NVIDIA probe routine failed for 1 device(s).
NVRM: None of the NVIDIA devices were initialized.
nvidia-nvlink: Unregistered Nvlink Core, major device number 238
If you have the necessary skills and experience with Proxmox, Nvidia Tesla P40, and PCIe passthrough, please reach out to discuss the project further.
Hello,
I have been working with Proxmox PCI passthrough and vGPU where few VMs can share same GPU card.
Some NVIDIA models i have worked with: T4, V100, and P40, etc. For your references:
https://www.freelancer.com/projects/linux/Proxmox-Expert-needed-for-cloning
https://www.freelancer.com/projects/linux/help-with-proxmox-36932788
https://www.freelancer.com/projects/c-programming/additional-help-proxmox
https://www.freelancer.com/projects/linux/Acceso-usando-proxmox-linux
https://www.freelancer.com/projects/c-programming/Project-for-Hai-Proxmox-Networking
From what you posted, the host configuration is correct since I can see vfio-pci is in place.
We can discuss over chat.
Hello,
I am an expert in system administration and full stack development. I currently am the SysAdmin for more than 15 AWS accounts of various businesses along with several other cloud and on-prem servers. I have deployed several Proxmox servers before and I have also passed through various PCIe devices and I can get you stored out in under 1 hour.
Please contact me on chat to discuss further.
Thanking you,
Navaneeth Krishna K.
Hhi I am experienced in this and I can start right now but i have few doubts and questions lets have a quick chat and get it started waiting for your replyyy
Greetings,
I have gained experience in utilizing Proxmox PCI passthrough and vGPU technology to enable multiple virtual machines to share a single GPU card. Specifically, I have worked with various NVIDIA models such as T4, P4, and P40, M10 and others.
I'm ready to start right now.