JetPack 4.4 Developer Preview - L4T R32.4.2 released

Autonomous Machines Jetson & Embedded Systems Announcements
1

We are pleased to announce JetPack 4.4 Developer Preview supporting Jetson AGX Xavier series, Jetson Xavier NX, Jetson TX2 series, Jetson TX1, and Jetson Nano. (JetPack 4.3 remains the latest production release.)

JetPack 4.4 key features include support for Jetson Xavier NX module, new versions of CUDA, TensorRT and cuDNN, and support for the upcoming DeepStream 5.0 Developer Preview.

Please refer to the JetPack Release Notes and L4T Release Notes for additional info.

JetPack 4.4 Highlights:

  • Support for the new Jetson Xavier NX module
  • Support for CUDA 10.2, and developer preview of TensorRT 7.1.0 and cuDNN 8.0.0
  • Support for Vulkan 1.2 and VPI 0.2.0 (Developer Preview)
  • Support for upgrading JetPack and L4T using Debian package management tool
  • Support for Generic Timestamping Engine (GTE) for Jetson AGX Xavier and Jetson Xavier NX
  • Support for Dynamic Frequency Scaling (DFS) for Video Image Compositor (VIC) using actmon
  • SE (Security Engine) samples to demonstrate hardware backed authentication and encryption capabilities of Jetson TX2 series, Jetson AGX Xavier and Jetson Xavier NX modules.
  • Utility to fuse multiple Jetson modules simultaneously
  • Option to specify APP partition size on the microSD card during initial configuration at first boot of Jetson Nano Developer Kit

JetPack 4.4 components:

  • L4T R32.4.2
  • CUDA 10.2
  • cuDNN 8.0.0 (Developer Preview)
  • TensorRT 7.1.0 (Developer Preview)
  • VisionWorks 1.6
  • OpenCV 4.1
  • Vulkan 1.2
  • VPI 0.2.0 (Developer Preview)
  • Nsight Systems 2020.2
  • Nsight Graphics 2020.1
  • Nsight Compute 2019.3

Existing installations of JetPack 4.3 can be upgraded in-place to JetPack 4.4 Developer Preview without re-flashing the device. For more information about upgrading JetPack via the Debian package management tool, please refer to the JetPack documentation here.

Download JetPack…https://developer.nvidia.com/embedded/jetpack

JetPack Release Notes…https://docs.nvidia.com/jetson/jetpack/release-notes/index.html

L4T Release Notes…https://developer.nvidia.com/jetson-linux-driver-package-release-notes-r3242-dp

3

Did you remove the new SDK Manager (1.1)? I downloaded it one our ago from the site, but now it links to 0.9 when clicking “Download sdk manager”. Also, computers running 0.9 says there is an update available, but fails upon update with the following:

SDK Manager version 1.1.0 update failed.
failed to download http://sdkm-a.nvidia.com/builds/SDKManager/1.1.0/6343_20110_28277817/Linux/sdkmanager_1.1.0-6343_amd64.deb

On my SDK Manager 1.1 it also fails now when trying to install JetPack 4.4, with the following error message:

Oops, could not load the NVIDIA SDK details
Failed to get configuration file from the server. Please check your network connectivity.
4

Thanks for reporting this. Looks like the problem is fixed now and correct version will download. Let us know if not.

5

Hi @prlawrence

The L4T R32.4.2 will be the same when you will be release the jetpack 4.4 in production? (Not in Developer Preview)

6

Hi @rbonghi, L4T R32.4.2 BSP will update to R32.4.3 with the production version of JetPack 4.4 to pick up bug fixes.

Thank you,

NV Jetson Team

1 Like
7

When 4.4 official release will be available, would it be possible to just do package upgrade from 4.4 DP to 4.4 Release (without flashing)?

-albertr

8

Hi @albertr, yes you will be able to use the sudo apt-get upgrade mechanism to upgrade from 4.4 DP to the production 4.4 release without needing to re-flash.

9

JetPack 4.4
run nvcc -v
-bash: nvcc: command not found？

10

TensorFlow2.1调用tnensort7出现如下错误：

Traceback (most recent call last):
File “o.py”, line 137, in
converter = trt.TrtGraphConverterV2(input_saved_model_dir=‘hdf52saved_model’,conversion_params=params)
File “/usr/local/lib/python3.6/dist-packages/tensorflow_core/python/compiler/tensorrt/trt_convert.py”, line 899, in init
assert context.executing_eagerly()
AssertionError

11

can I install cuda 10.0 in jetpack 4.4? since some of my model could only run in cuda 10.0.

12

Hi @CoderJustin, please see these posts:

1 Like
13

Thank you!

14

Any idea when we can expect support for Ubuntu 20.04 for JetPack ? ROS 2 requires 20.04. Thanks.

3 Likes
15

I can’t get d32.4 to install properly on my Jetson Xavier NX Developer Kit. SDKManager fails to connect to the target no matter what I do. I can download the image and boot that, but updating fails. After the Ubuntu updates go through, I’m left with:

$ sudo apt-get update
Hit:1 https://repo.download.nvidia.com/jetson/common r32.4 InRelease
Get:2 https://repo.download.nvidia.com/jetson/t194 r32.4 InRelease [2,559 B]
Hit:3 Index of /ubuntu-ports bionic InRelease
Hit:4 Index of /ubuntu-ports bionic-updates InRelease
Hit:5 Index of /ubuntu-ports bionic-backports InRelease
Get:6 https://repo.download.nvidia.com/jetson/t194 r32.4/main arm64 Packages [8,239 B]
Err:6 https://repo.download.nvidia.com/jetson/t194 r32.4/main arm64 Packages
File has unexpected size (8490 != 8239). Mirror sync in progress? [IP: 184.28.219.26 443]
Hashes of expected file:

  • Filesize:8239 [weak]
  • SHA256:bd3c7533684455dc3d1f94a39a1c4afe48d5a020f35b3b784f1ea3787e193cca
  • SHA1:ea31489efd3e7f756fa70d3288d401319b7fce57 [weak]
  • MD5Sum:5c6a54ba6437bb76e7537ff6ba3d1d0c [weak]
    Release file created at: Tue, 28 Apr 2020 06:35:46 +0000
    Hit:7 Index of /ubuntu-ports bionic-security InRelease
    Fetched 2,559 B in 2s (1,296 B/s)
    Reading package lists… Done
    E: Failed to fetch https://repo.download.nvidia.com/jetson/t194/dists/r32.4/main/binary-arm64/Packages.gz File has unexpected size (8490 != 8239). Mirror sync in progress? [IP: 184.28.219.26 443]
    Hashes of expected file:
    - Filesize:8239 [weak]
    - SHA256:bd3c7533684455dc3d1f94a39a1c4afe48d5a020f35b3b784f1ea3787e193cca
    - SHA1:ea31489efd3e7f756fa70d3288d401319b7fce57 [weak]
    - MD5Sum:5c6a54ba6437bb76e7537ff6ba3d1d0c [weak]
    Release file created at: Tue, 28 Apr 2020 06:35:46 +0000
    E: Some index files failed to download. They have been ignored, or old ones used instead.

Now what?

16

The message does indicate that you should try again if there is an update at the site which is in progress. You can try again and see if there is still an error. If not, then there might be some network problems going on, which again, might be temporary.

If instead the problem recurs, and especially if the particular failure differs (such as the unexpected file size being for a different file or the size being off by a different amount), then it probably is a network issue. Some “lossy” networks can be at fault. WiFi tends to do this more than wired if the fault is at the local computer, but it is hard to say. Try again and compare the logs.