NVIDIA vGPU manager in ESXi 6.5 returns an error
Applies to
- NetApp H615C with NVIDIA Tesla T4 GPUs
- ESXi 6.5
Issue
Running command nvidia-smi
in ESXi 6.5 on H615C node returns an error
NVMK:Unknown Error
KVLM failed to initialize
ESXi vmkernel log output:
2020-06-18T08:55:42.016Z cpu2:66269)IOMMU: 2234: Device 0000:18:00.0 placed in new domain 0x43065407f050.
2020-06-18T08:55:42.016Z cpu2:66269)DMA: 963: Protecting DMA engine 'NVIDIADmaEngine'. Putting parent PCI device 0000:18:00.0 in IOMMU domain 0x43065407f050.
2020-06-18T08:55:42.016Z cpu2:66269)DMA: 646: DMA Engine 'NVIDIADmaEngine' created using mapper 'DMAIOMMU'.
2020-06-18T08:55:42.016Z cpu2:66269)NVRM: This is a 64-bit BAR mapped above 16 TB by the system
NVRM: BIOS or the VMware ESXi kernel. This PCI I/O region assigned
NVRM: to your NVIDIA device is not supported by the kernel.
NVRM: BAR1 is 256M @ 0x387fc00$
2020-06-18T08:55:42.064Z cpu2:66269)DMA: 691: DMA Engine 'NVIDIADmaEngine' destroyed.