Error 0x00000116 on nvlddmkm.sys when using vGPU, Passthrough works.

Hello everyone.

I have an issue with a POC for a customner that are trying out vGPU with a K2 Card. I have done everything according to the manual but when I add a vGPU instead of the Passthrough card I get a Error 0x00000116 BSOD on nvlddmkm.sys. Like I said using Passthrough works fine.

I have don tests when using passthrough using GPU-Z, RedWay 3D Turbine, Unique etc etc. and I get the performance I want. But as soon as I switch to vGPU (no matter which one) the server BSOD.

I know the HW is not in the HCL but I can’t see how that has anything to do with it but I might be wrong.


P8Z77 -V PRO
Intel i7 4770 (non k)
Intel Gigabit NIC
XenServer 6.2 SP1 HF4

Windows 2008 R2
4 vCPU
NVIDIA GRID vGPU Software Release 331

Any ideas?


Jörgen Persson

Hello Jörgen,

the reason might be that the K2 is not certified for your Host. You can find all GRID certified servers at

NVIDIA works very close with all vendors on that site to certify the GRID GPUs. GRID GPUs are different to other GPUs and sometimes require BIOS changes from the server vendor for example. The error you are mentioning might be a result of that.

Erik Bohnhorst | Solution Architect – GRID
NVIDIA Corporation

I know that the host itself is not certified but to make the customer pay for the hardware before them knowing that it acctually works is the reason for the POC. I guess then that the only way to show that it works is to leave it at Passthrough then?

I just think it’s wierd since XenServer makes it “hardware unaware” with the XenServer tools and I can’t see how it then “cares” about the hardware XenServer is running on.

If anyone else has any ideas please feel free to respond.

//Jörgen Persson