search

NVIDIA-AI-IOT / deepstream_360_d_smart_parking_application

Describes the full end to end smart parking application that is available with DeepStream 5.0
MIT License
340 stars 114 forks source link

Running application on Nvidia Jetson TX2 / Xavier #1

Open Shrunoti opened 5 years ago

Shrunoti commented 5 years ago

I've tried running this application on Nvidia Jetson TX2 and Nvidia Xavier. I've faced difficulty running the run.sh file in perception server as 'nvidia-docker' is not supported on arm64 architecture. Is there any work around that?

Also will the deepstream 3.0 be supported on these two devices?

mengzhangjian commented 5 years ago

@Shrunoti ，you should install jetson4.2.1 on your tx2, that includes nvidia-docker

Shaked commented 5 years ago

@mengzhangjian is it out already? I can't seem to find it

fork88888 commented 5 years ago

I set tegra-docker with GPU support on Jetson Xavier. Next I have run container with deviceQuery application. I see that GPU is supported:

deviceQuery, CUDA Driver = CUDART, CUDA Driver Version = 10.0, CUDA Runtime Version = 10.0, NumDevs = 1 Result = PASS

But when I try to run deepstream_360d:3.0-18.11 container I see the following:

standard_init_linux.go:211: exec user process caused "exec format error" failed to resize tty, using default size

I think that means that the conteiner is for x86 architecture.

Is there deepstream_360d container for arm64v8 architecture?