Question: AGX orin SOM CPU and GPU Memory assignment

Questions

  1. AGX orin CPU Memory size and its memory map
  2. GPU memory size and its memory map
  3. CPU and GPU can share the same physical memory?
    and its share memory size.

image

  1. Linux porting memory location and its max size
  2. Memory assignment can be allocated its memory size by Linux?
    (Memory variable size allocation is possible?)

Where can I find these documents?

Hi,
Please check 1. CUDA for Tegra — CUDA for Tegra 12.3 documentation

A dGPU with separate DRAM memory can be connected to the Tegra device over PCIe or NVLink. It is currently supported only on the NVIDIA DRIVE platform

If you use AGX Orin developer kit, coneecting to dGPU is not supported. Please note this.

For allocating memory on iGPU, if you would like to allocate memory which ca be accessed through GPu an dCPU, you can call cudaMallocManaged():
CUDA Runtime API :: CUDA Toolkit Documentation

Or use NvBufSurface APIs to allocate NvBufSurface:
Jetson Linux API Reference: Main Page | NVIDIA Docs

This topic was automatically closed 14 days after the last reply. New replies are no longer allowed.