HP ProLiant DL385p Gen8 and nVidia Tesla M40

[Alex-ZiX]

I've bought the nVidia Tesla M40 GPU and tries to install it into my HP ProLiant DL385p Gen8 server. I use Hyper-V Server 2019 operating system. And it boots up normally. But I can't setup the original nVidia drivers - it says that no compatible hardware found. I want to pass this GPU to the one of virtual machines using DDA (Discrete Device Assignment) but script from MS which checks passibility of using DDA tells me that this GPU supports line-based interrupts only. DDA needs message-based interrupts for working. So I think that may be my server just doesn't support this GPU. I have nVidia GRID K2 GPU installed at the same server and it works fine.

After some googling I've found that i should enable "4G decoding" or increase "MMIO memory" but there is no such options in my RBSU.

[ManBha]

Hello,

As per the driver the card is not supported by the server.

https://support.hpe.com/hpesc/public/swd/detail?swItemId=MTX-5c87402ea5454b37b57fe67ef6#tab4

Supported Servers:

• HPE ProLiant DL380 Gen9
• HPE ProLiant XL190r Gen9
• HPR ProLiant XL270d Gen9

Please do refer the below document for instructions to enable IOMMU. {page 38]

https://support.hpe.com/hpesc/public/docDisplay?docId=c00191707

Thanks.

[Kaushik_ray_1]

You can go into the bios/RBSU

Once in press CTRL+A and you will see a hidden option called SERVICE OPTION

In the hidden option scroll down to PCI Express 64-Bit BAR Support and enable it. 

Save the bios and exit

Your M40 should work now. 

Good luck