GPU with VMware, nvidia-smi: No devices were found

I have a GTX 1070 and I’m trying to use it with VMware vsphere.

My ubuntu version is 16.04, and related outputs are as following:

lspci output:
$ lspci | grep NVIDIA
13:00.0 VGA compatible controller: NVIDIA Corporation Device 1b81 (rev a1)
13:00.1 Audio device: NVIDIA Corporation Device 10f0 (rev a1)

device files:
crw-rw-rw- 1 root root 195, 0 Aug 29 19:46 /dev/nvidia0
crw-rw-rw- 1 root root 195, 255 Aug 29 19:46 /dev/nvidiactl
crw-rw-rw- 1 root root 246, 0 Aug 29 19:46 /dev/nvidia-uvm

driver version:
NVRM version: NVIDIA UNIX x86_64 Kernel Module 396.54 Tue Aug 14 19:02:34 PDT 2018
GCC version: gcc version 5.4.0 20160609 (Ubuntu 5.4.0-6ubuntu1~16.04.10)

driver gpu info:
Model: GeForce GTX 1070
IRQ: 65
GPU UUID: GPU-???-???-???-???-???
Video BIOS: ??.??.??.??.??
Bus Type: PCIe
DMA Size: 47 bits
DMA Mask: 0x7fffffffffff
Bus Location: 0000:13:00.0
Device Minor: 0

I installed the CUDA 9.2 following the instructions here:
https://developer.nvidia.com/cuda-downloads?target_os=Linux&target_arch=x86_64&target_distro=Ubuntu&target_version=1604&target_type=deblocal

The commands I ran are:
sudo dpkg -i cuda-repo-ubuntu1604-9-2-local_9.2.148-1_amd64.deb
sudo apt-key add /var/cuda-repo-<version>/7fa2af80.pub
sudo apt-get update
sudo apt-get install cuda

But when I nvidia-smi, it says “No devices found”. I’m wondering if I did any of the steps incorrectly?

Thanks in advance!

This is expected behavior. Not all NVIDIA GPUs are supported or intended to be used within a virtualized environment. The driver attempts to enforce this.

Thank you for the fast reply. But I have seen a post about using GTX 1070 passthrough with ESXi. Setting up GTX-1070 Passthrough with ESXi | by David Ramsay | Medium

I followed the same steps but nvidia-smi still can’t see the device. Thanks.