I was trying to run a docker on orin developer kit, and got the following err message:
Error response from daemon: failed to create task for container: failed to create shim task: OCI runtime create failed: nvidia-container-runtime did not terminate successfully: exit status 1: time="2023-09-14T09:59:51-07:00" level=error msg="failed to create NVIDIA Container Runtime: failed to construct OCI spec modifier: requirements not met: cuda>=11.8||brand=tesla&&driver>=450&&driver<451||brand=tesla&&driver>=470&&driver<471||brand=unknown&&driver>=470&&driver<471||brand=nvidia&&driver>=470&&driver<471||brand=nvidiartx&&driver>=470&&driver<471||brand=geforce&&driver>=470&&driver<471||brand=geforcertx&&driver>=470&&driver<471||brand=quadro&&driver>=470&&driver<471||brand=quadrortx&&driver>=470&&driver<471||brand=titan&&driver>=470&&driver<471||brand=titanrtx&&driver>=470&&driver<471||brand=tesla&&driver>=510&&driver<511||brand=unknown&&driver>=510&&driver<511||brand=nvidia&&driver>=510&&driver<511||brand=nvidiartx&&driver>=510&&driver<511||brand=geforce&&driver>=510&&driver<511||brand=geforcertx&&driver>=510&&driver<511||brand=quadro&&driver>=510&&driver<511||brand=quadrortx&&driver>=510&&driver<511||brand=titan&&driver>=510&&driver<511||brand=titanrtx&&driver>=510&&driver<511||brand=tesla&&driver>=515&&driver<516||brand=unknown&&driver>=515&&driver<516||brand=nvidia&&driver>=515&&driver<516||brand=nvidiartx&&driver>=515&&driver<516||brand=geforce&&driver>=515&&driver<516||brand=geforcertx&&driver>=515&&driver<516||brand=quadro&&driver>=515&&driver<516||brand=quadrortx&&driver>=515&&driver<516||brand=titan&&driver>=515&&driver<516||brand=titanrtx&&driver>=515&&driver<516 not met"
: unknown
Error: failed to start containers: trt2023
Does that means my driver on orin is too old for this docker? and if it is, would you please tell me where I could find the available version for orin kit and how to update the drivers? my current driver is:
$ cat /proc/driver/nvidia/version
NVRM version: NVIDIA UNIX Open Kernel Module for aarch64 35.2.1 Release Build (buildbrain@mobile-u64-6348-d7000) Tue Jan 24 15:26:51 PST 2023
GCC version: gcc version 9.3.0 (Buildroot 2020.08)
$ nvcc -V
nvcc: NVIDIA (R) Cuda compiler driver
Copyright (c) 2005-2022 NVIDIA Corporation
Built on Sun_Oct_23_22:16:07_PDT_2022
Cuda compilation tools, release 11.4, V11.4.315
Build cuda_11.4.r11.4/compiler.31964100_0
$ cat /sys/module/nvidia/version
35.2.1
Hi @czxu what container are you trying to run? It appears to be built for x86, not Jetson. You should use one of the l4t or jetpack containers instead (or a container derived from these):
@czxu I’m unfamiliar with that registry and unable to access it. I would ask the organizers of the hackathon if they have a version for Jetson. Otherwise, l4t-jetpack already includes the TensorRT libraries and development headers.
Thank you so much dusty_nv! That’s super helpful! It would be super helpful if we could have a image for jetson, cause I also tried build the env on orin env and it’s painful… (a lot of conflicts especially torch version)
Sorry @czxu, I wasn’t aware of this TRT Hackathon event, however the organizers may know more about what is in the container. If it needs PyTorch, then I would recommend using l4t-pytorch as base container (which also already includes TensorRT)
May I know if we have any images for orin developer kit (I am using jetpack 5) to deploy TRT for some network examples (e.x. bevformer)? Recently I am trying to play with the orin to convert and deploy bevformer, I came across some related posts and find you are an expert in jetson dockers, so I would like to know if you have any recommend images for my case lol
Hi Chengzhe, the l4t-jetpack container includes TensorRT and it’s development headers/libraries, in addition to the tensorrt python package, so it already has it. And l4t-pytorch is based on l4t-jetpack, so it has TensorRT too.
Sounds great! I will using the l4t images you mentioned, thank you!
BTW, do we have any images with mm packages (mmcv-full, mmdet, mmsegmentation, etc.)? I found it super slow when building wheel for mmcv-full on orin jetson lol