Implementing DMA with CUDA?

Hello good people!

Is it at all possible to implement Direct Memory Access on GeForce RTX 3080 (using CUDA) since it’s not listed under supported GPUs for GPUDirect. Implementing DMA is crucial since I’m trying to minimize latency for data stream: frame grabber => GPU memory.

I’m using modern OpenGL to render provided frames!

OS: Ubuntu 20.04
CPU: AMD Ryzen Threadripper 3960X
GPU: Nvidia GeForce RTX 3080
FrameGrabber: Kaya Komodo
Motherboard: ASUS TRX40

Thank You and have a great day!
Best regards,
Filip