On Fri, Feb 2, 2018 at 12:13 PM, Jordan, Marcel <marcel.jordan@it-novum.com> wrote:

Hi,

i have some NVIDIA Tesla P100 and V100 gpu in our oVirt 4.2 cluster and
searching for a documentation how to use the new vGPU feature. Is there
any documentation out there how i configure it correctly?

--
Marcel Jordan

Possibly check what would become the official documentation for RHEV 4.2, even if it could not map one-to-one with oVirt

Admin guide here:

https://access.redhat.com/documentation/en-us/red_hat_virtualization/4.2-beta/html/administration_guide/sect-host_tasks#Preparing_GPU_Passthrough

Planning and prerequisites guide here:

https://access.redhat.com/documentation/en-us/red_hat_virtualization/4.2-Beta/html/planning_and_prerequisites_guide/requirements#pci_device_requirements

In oVirt 4.2 release notes I see these bugzilla entries that can help too...

https://bugzilla.redhat.com/show_bug.cgi?id=1481007

https://bugzilla.redhat.com/show_bug.cgi?id=1482033

HIH,

Gianluca