/dev/nvhost* documentation

Hi,

This might be a dumb question. I have trouble adding local resources of TX2 to AWS Greengrass (basically a container). If I do “ls /dev/nvhost*”, the terminal prints out a lot of GPU resources. I wonder if there is a documentation explaining what they are and what they do so that I can choose the right combination of them.
Thanks!

Hi,

Could you check this comment first?
https://devtalk.nvidia.com/default/topic/1052324/jetson-nano/jetson-nano-aws-greengrass-/post/5341970/#5341970

Thanks.

Hi,

I have used Greengrass before. I am trying to activate Kinesis Video Stream with Greengrass. This requires GPU resources from TX2. So I need some documentation about nvhost*.
Thanks!

Hi,

To enable GPU, you will need to add the access of these device nodes:

/dev/nvhost-ctrl
/dev/nvhost-ctrl-gpu
/dev/nvhost-prof-gpu
/dev/nvmap
/dev/nvhost-gpu
/dev/nvhost-as-gpu

Check this wiki to get more information: https://elinux.org/Jetson_Zoo#Docker

Thanks.

Hi,

Thank you for replying!
These do not seem to be enough. Maybe the issue is non-GPU-related. T_T