Access to RT Cores HW from CUDA (or OpenGL)?

Accelerated Computing CUDA CUDA Programming and Performance
1

Which of the following APIs currently (2023) support programming API access direct to the RT Cores (on compatible NVIDIA HW devices):

  • OpenGL
  • Vulkan
  • CUDA

For each of above which DO support direct programmatic RT core access, where can I find the best/current API details?

For instance,
I found an old link to Turing extensions for Vulkan & OpenGL which provide GLSL extensions for RT core access - are these still supported?

I can’t find anything in the CUDA 12x documentation.

I do know Vulkan is supported, just wondering about OpenGL and/or CUDA primarily, as we have not migrated to Vulkan yet.

Thanks,
Professor Reinhardt
University of Washington
Electrical and Computer Engineering Dept

2

You cannot access the RT cores directly from CUDA.

One NVIDIA provided path is via Optix. There is a separate forum for Optix. It is possible to access RT cores from both Vulkan and OpenGL, see here. (Yes, I know that is the link you provided. I’m offering it as evidence to support my claim, not as a complete answer or tutorial for your question.) I won’t be able to give much help with any of that, this particular sub-forum is focused on CUDA. There are separate forums for OpenGL and Vulkan. I see you have already posted in other forums so I believe you will be able to find the various forums for each of these topics.