How to check the JetPack Version

Autonomous Machines Jetson & Embedded Systems Jetson TX2
#1

Is there any way to check the JetPack Version instead of just checking the L4T version?

I found a way to check it out is to use JetsonInfo.py.

It will get the information like :

NVIDIA Jetson TX2
L4T 28.2.1 [ JetPack 3.3 or 3.2.1 ]
Board :t186ref
Ubuntu 16.04 LTS
Kernel Vision : 4.4.38-tegra
CUDA 9.0.252

But it seems that if it has the same L4T version, it can’t identify which jetpack version it is.

2 Likes
Can I determine DS and JP version at python level?
OpenCV Building with CUDA CUDNN - No CUDNN
#2

Given a TX2 board, there is no way to check which JetPack version it is flashed with. You can only check its L4T info. For example, cat /etc/nv_tegra_release on board.

4 Likes
#3

Hi Sir,

OK. Got it. Thank you for your information.

best regards,
Lilian.lin

Compile opencv with cuda
#4

FYI, JetPack never installs to the Jetson. Only L4T or packages, so there isn’t any JetPack version associated with the Jetson other than possibly the install GUI front end was JetPack of a certain version. On the other hand, so far as I know, there is only one L4T version associated with each JetPack release.

#5

I hope this can help you:

JetPack 3.2.1 L4T R28.2.1 cuDNN v7.0.5 TensorRT 3.0 GA
JetPack 3.2 L4T R28.2
JetPack 3.3 L4T R28.2.1 cuDNN v7.1.5 TensorRT 4.0 GA YES
##############################################
how to find the version of cudnn

cat /usr/include/cudnn.h | grep CUDNN_MAJOR -A 2


#define CUDNN_MAJOR 7
#define CUDNN_MINOR 1
#define CUDNN_PATCHLEVEL 5
--
#define CUDNN_VERSION    (CUDNN_MAJOR * 1000 + CUDNN_MINOR * 100 + CUDNN_PATCHLEVEL)

#include "driver_types.h"

which means version 7.15

##############################################

#6

This command is not working. Can you please suggest me how to get the Nvidia version for Jetson TX2.

#7

If you want the L4T release for the Jetson itself, then “head -n 1 /etc/nv_tegra_release”. This is what flash installs.

1 Like
#8

I am looking for the same Information on the Jetson Nano Board. I think I use Jetpack 4.2 but I want to be sure. Unfortunatly the path /etc/nv_tegra_release is not existing in my nano build. Our image is from this Link https://developer.nvidia.com/jetson-nano-sd-card-image-r322 and I am to 99% sure it is the Jetpack 4.2 but still want to test it.

1 Like
#9

Did you slove the problem?

#10

There is no /etc/nv_tegra_release in JetPack 4.2.x images.

May I suggest that future JetPack images come with a simple file somewhere that contains the JetPack version. Seems like a pretty straightforward thing to add to your (NVIDIA’s) build process.

#11

Hi all, please see this post to read the version from the nvidia-l4t-core package:

https://devtalk.nvidia.com/default/topic/1057675/jetson-nano/read-l4t-version-from-jetpack-4-2-1/post/5363182/#5363182

1 Like
#12

sudo apt-cache show nvidia-jetpack
It will show something like:

Package: nvidia-jetpack
Version: 4.4-b144
Architecture: arm64
Maintainer: NVIDIA Corporation

31 Likes
#14

Please go through the link https://github.com/rbonghi/jetson_stats.

2 Likes
#16

Perfect answer!! :-)

#17

Great answer!

2 Likes
#18

@hector.lxm

sudo apt-cache show nvidia-jetpack
It will show something like:

Package: nvidia-jetpack
Version: 4.4-b144
Architecture: arm64
Maintainer: NVIDIA Corporation

5 Likes
#19

Mark this as solution if possible.

#21

All I get is:

$ sudo apt-cache show nvidia-jetpack
N: Unable to locate package nvidia-jetpack
E: No packages found

#22

What about “dpkg -l | grep -i 'jetpack”?