JetPack 4.4 Developer Preview - L4T R32.4.2 released

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…

JetPack Release Notes…

L4T Release Notes…

1 Like

Downloaded, installed and running very nicely on Xavier NX (in Jetson Nano Dev Kit rev B01 with modified fan assisted headsink), a very smooth process - thanks to all involved in getting the release out.

1 Like

Small documentation issue… The partition layout page in the Developer Guide has some malformed HTML which makes reading the partition layouts difficult…

Thanks for reporting this. A temporary fix for that page is in the pipe, and you helped the documentation team uncover a bug in one of their tools. Much appreciated.

1 Like

Actually I just realized another issue… In that same section, the layout for the NX P3668-0000 (DevKit module) QSPI-NOR and SDCard are incorrect. They look like old entries or those copied from the Nano. In fact, the P3668-0000 layout is exactly like the P3668-0001 with eMMC. 47 entries in the QSPI-NOR and 11 on the SD Card.

Hello prlawrence,

Do you know, Why /boot/extlinux/extlinux.conf removed from NX sample root file system ?
NX won’t support loading FDT from File system ?

I think if you’re not using the sd card image it’s expected that you’ll create the /boot directory yourself after extracting the sample tarball. The sd card image ( certainly does have extlinux.conf and it does support FDT loading from the /boot partition filesystem.

How can I disable the containerd daemon, it appears to take over a GB of the 8GB (although htop doesn’t show that). I want only the minimum stuff necessary to preserve the memory for GPU doing training in DL. Using systemctl to disable it dosn’t seem to work, I don’t want to purge docker, just disable it.

You should probably create a new topic for this issue but make sure you disable the docker service as well. You could also mask containerd so nothing else can start it… systemctl mask containerd.service

Could you share the details about your modified fan assisted heatsink? (e.g. part numbers and url for purchase, and any modification details) Thank you.

Initially I just used the heatsink (and Noctua NF-A4x20 recommended fan) from the Nano Dev Kit, moved off-centre so that the cut out gave clearance over the high components to one side of the main chip (being off-centre the heatsink isn’t over the high components on the other side). In this arrangement it was only possible to use one of the mounting holes - basically to align the heatsink. So this was good for use on a desk where the setup isn’t disturbed at all. Definitely only considered this a very temporary solution, and certainly not recommended. I’ve since machined a custom heatsink from aluminium, tapped with M2 threads at the correct spacing to fit the mounting holes perfectly. Sorry that isn’t going to be as much help as you were looking for.

Dear helpers,

I have new Jetson xavier nx
SD card image was downloaded from today (July 3 2020)
md5sum verified match 4b24e23867ed3d4b651934d0ac40c92c
sd card was flashed and system up running. But there is an imx219_board_setup error.
Messages generated from using command dmesg shows imx219_board_setup: error during i2c read probe

Here is the related details:
[ 1.995205] tegra_rtc c2a0000.rtc: Tegra internal Real Time Clock
[ 1.995418] i2c /dev entries driver
[ 1.997726] i2c i2c-2: Added multiplexed i2c bus 9
[ 1.998271] i2c i2c-2: Added multiplexed i2c bus 10
[ 1.998277] i2c-mux-gpio cam_i2cmux: 2 port mux on 3180000.i2c adapter
[ 1.998719] imx219 9-0010: tegracam sensor driver:imx219_v2.0.6
[ 2.022179] tegra-i2c 3180000.i2c: no acknowledge from address 0x10
[ 2.022231] imx219 9-0010: imx219_board_setup: error during i2c read probe (-121)
[ 2.022276] imx219 9-0010: board setup failed
[ 2.022381] imx219: probe of 9-0010 failed with error -121
[ 2.022906] imx219 10-0010: tegracam sensor driver:imx219_v2.0.6
[ 2.046138] tegra-i2c 3180000.i2c: no acknowledge from address 0x10
[ 2.046184] imx219 10-0010: imx219_board_setup: error during i2c read probe (-121)
[ 2.046243] imx219 10-0010: board setup failed
[ 2.046364] imx219: probe of 10-0010 failed with error -121
[ 2.047709] max77620-power max20024-power: Event recorder REG_NVERC : 0x0
[ 2.054357] tegra-thermal-throttle bthrot_cdev: clk:cpu max:1907200000, min:115200000 steps:24

I can’t move om with my csi cameras.
What can I do?
Please help,
Thank you