1. Proposed title of this feature request

Enable Nvidia CX-8 SuperNIC for AI-RAN use cases

2. What is the nature and description of the request?

Enable the usage of Nvidia ConnectX-8 Super NIC for AI-RAN use cases of Nokia. 

3. Why does the customer need this? (List the business requirements here)

1. Configurable by NNO (Nvidia Network Operator) for 5G workloads connectivity
2. CX-8 to support cryptographic acceleration, so CPU-intensive cryptographic operations can be offloaded to accelerators.
3. CX-8 to support Secure boot
4. It needs to support the line rates of 200 gbps, 400 gbps & 800 gbps. 
5. It needs to host the PTP/sync operator from Nokia, therefore it should have been tested with its own Nvidia operator for PTP functionality.
   1. "NIC synchronization feature support - I want that the NIC used in the solution has a high precision physical clock and supports HW time stamping, so that I can synchronize the host to high precision time sources."  DNG ID - 13247401
6. Enable configuration of (DOCA) RDMA between NICs and GPUs, so that high-performance data paths can be made available to applications
   1. "Configuration of RDMA - I want the Cloud Infrastructure management system to enable the configuration of RDMA between NICs and GPUs, so that high-performance data paths can be made available to applications" DNG ID - 13325989

4. List any affected packages or components.

Nvidia drivers, DoCA, etc 

5. What has been supported by certification and what is covered in RHELBU epics

As per this doc --> https://docs.google.com/document/d/10eoSJL6iJmpTfhTI_UV8Rr9GzdNPGl3sHdRS_93Vkd8/edit?usp=sharing