vGPU esxi with Ubuntu 20.04 host, nvidia-smi error

I am having problem installing driver to my Ubuntu 20.04 running inside of esxi 6.7 update 2.
I am using a Tesla P6 GPU.
I used this command to install cuda.

wget https://developer.download.nvidia.com/compute/cuda/repos/ubuntu2004/x86_64/cuda-ubuntu2004.pin
sudo mv cuda-ubuntu2004.pin /etc/apt/preferences.d/cuda-repository-pin-600
wget https://developer.download.nvidia.com/compute/cuda/11.5.0/local_installers/cuda-repo-ubuntu2004-11-5-local_11.5.0-495.29.05-1_amd64.deb
sudo dpkg -i cuda-repo-ubuntu2004-11-5-local_11.5.0-495.29.05-1_amd64.deb
sudo apt-key add /var/cuda-repo-ubuntu2004-11-5-local/7fa2af80.pub
sudo apt-get update
sudo apt-get -y install cuda

I always get an error when I run 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 log

$ dmesg | grep NVRM | head
[  729.533091] NVRM: This PCI I/O region assigned to your NVIDIA device is invalid:
               NVRM: BAR1 is 0M @ 0x0 (PCI:0000:02:02.0)
[  729.533109] NVRM: The NVIDIA probe routine failed for 1 device(s).
[  729.533109] NVRM: None of the NVIDIA devices were initialized.
[  729.689050] NVRM: This PCI I/O region assigned to your NVIDIA device is invalid:
               NVRM: BAR1 is 0M @ 0x0 (PCI:0000:02:02.0)
[  729.689066] NVRM: The NVIDIA probe routine failed for 1 device(s).
[  729.689066] NVRM: None of the NVIDIA devices were initialized.
[  729.822066] NVRM: This PCI I/O region assigned to your NVIDIA device is invalid:
               NVRM: BAR1 is 0M @ 0x0 (PCI:0000:02:02.0)

Any help would be appreciated.