Hi,
I saw the announcement for the Jetson Orin Nano Super Developer Kit but there’s no means presently for current installations to update to support the new frequencies. Could you post also an overlay on what to update for existing Jetpack 6.1 installations (i.e. bpmp dtb, nvpmodel.conf etc.)?
Hi,
Please check the release page:
Jetson Linux | NVIDIA Developer
NEW: Jetson Linux 36.4 now supports the new boosted performance mode [MAXN mode] for Jetson Orin Nano Developer Kit, that improves the AI compute performance by 1.7x. See ‘Installing JetPack’ section for a new SD Card Image that supports MAXN mode as well as instructions on installing the new image through NVIDIA SDK Manager on JetPack 6.1
You can re-flash Orin Nano developer kit through SDKManager.
I understand, but it doesn’t say what to update in Jetson Linux | NVIDIA Developer Driver Package (BSP) to unlock the MAXN mode when flashing through the BSP along with other customizations.
Hi,
Jetpack 6.1 is refreshed to enable the mode. Please re-download it and install through SDKManager. Or run the manual command:
Quick Start — NVIDIA Jetson Linux Developer Guide 1 documentation
Yes but the link https://developer.nvidia.com/downloads/embedded/l4t/r36_release_v4.0/release/Jetson_Linux_R36.4.0_aarch64.tbz2 in Jetson Linux | NVIDIA Developer is not updated with the jetson-orin-nano-devkit-super configuration. Right now all you can do is to install the official SD image, but can’t reflash if you have customized UEFI etc.
Hi,
For information, is your Orin Nano 8GB or 4GB? Now we support it on Orin Nano 8GB.
Please run SDKManager and select Orin Nano 8GB sku5. And select download only. And you shall see the packages downloaded to
Downloads/nvidia/sdkm_downloads$ ls
cudnn-local-tegra-repo-ubuntu2204-9.3.0_1.0-1_arm64.deb.mtd
hwdata
Jetson_Linux_R36.4.2_aarch64.tbz2
sdkml3_jetpack_61_rev1.json
Tegra_Linux_Sample-Root-Filesystem_R36.4.2_aarch64.tbz2
I am using Jetson Orin Nano 8GB dev kit -(P3767-0003) not official. I already installed jetpack6.1. But the power mode do not have MAXN mode. Is the MAXN mode only available on Jetson Orin Nano 8GB dev kit -(P3767-0005) official ?
I am using 8GB orin nano and having the same exact problem
I have the same problem on jetson orin nano 8Gb running jetpack 6.1
Thanks, worked. Jetson_Linux_R36.4.2_aarch64.tbz2 and Tegra_Linux_Sample-Root-Filesystem_R36.4.2_aarch64.tbz2 is what I was looking for.
will this update can be applied to Orin Nano 4GB. If so, when will you be able to get the update?
Why are the updated 36.4.2 packages required for the new power modes only available in new images and not on the apt repo? Until these packages are available in the apt repo please post a direct link to the tarfile that typically contains them Jetson_Linux_R36.4.2_aarch64.tbz2
nvidia-container-toolkit-base/now 1.14.2-1 arm64 [installed,local]
nvidia-container-toolkit/now 1.14.2-1 arm64 [installed,local]
nvidia-l4t-3d-core/now 36.4.2-20241212160716 arm64 [installed,local]
nvidia-l4t-apt-source/now 36.4.2-20241212160716 arm64 [installed,local]
nvidia-l4t-bootloader/now 36.4.2-20241212160716 arm64 [installed,local]
nvidia-l4t-camera/now 36.4.2-20241212160716 arm64 [installed,local]
nvidia-l4t-configs/now 36.4.2-20241212160716 arm64 [installed,local]
nvidia-l4t-core/now 36.4.2-20241212160716 arm64 [installed,local]
nvidia-l4t-cuda-utils/now 36.4.2-20241212160716 arm64 [installed,local]
nvidia-l4t-cuda/now 36.4.2-20241212160716 arm64 [installed,local]
nvidia-l4t-cudadebuggingsupport/now 12.6-34622040.0 arm64 [installed,local]
nvidia-l4t-display-kernel/now 5.15.148-tegra-36.4.2-20241212160716 arm64 [installed,local]
nvidia-l4t-dla-compiler/now 36.4.2-20241212160716 arm64 [installed,local]
nvidia-l4t-firmware/now 36.4.2-20241212160716 arm64 [installed,local]
nvidia-l4t-gbm/now 36.4.2-20241212160716 arm64 [installed,local]
nvidia-l4t-graphics-demos/now 36.4.2-20241212160716 arm64 [installed,local]
nvidia-l4t-gstreamer/now 36.4.2-20241212160716 arm64 [installed,local]
nvidia-l4t-init/now 36.4.2-20241212160716 arm64 [installed,local]
nvidia-l4t-initrd/now 36.4.2-20241212160716 arm64 [installed,local]
nvidia-l4t-jetson-io/now 36.4.2-20241212160716 arm64 [installed,local]
nvidia-l4t-jetson-multimedia-api/now 36.4.2-20241212160716 arm64 [installed,local]
nvidia-l4t-jetsonpower-gui-tools/now 36.4.2-20241212160716 arm64 [installed,local]
nvidia-l4t-kernel-dtbs/now 5.15.148-tegra-36.4.2-20241212160716 arm64 [installed,local]
nvidia-l4t-kernel-headers/now 5.15.148-tegra-36.4.2-20241212160716 arm64 [installed,local]
nvidia-l4t-kernel-oot-headers/now 5.15.148-tegra-36.4.2-20241212160716 arm64 [installed,local]
nvidia-l4t-kernel-oot-modules/now 5.15.148-tegra-36.4.2-20241212160716 arm64 [installed,local]
nvidia-l4t-kernel/now 5.15.148-tegra-36.4.2-20241212160716 arm64 [installed,local]
nvidia-l4t-libwayland-client0/now 36.4.2-20241212160716 arm64 [installed,local]
nvidia-l4t-libwayland-cursor0/now 36.4.2-20241212160716 arm64 [installed,local]
nvidia-l4t-libwayland-egl1/now 36.4.2-20241212160716 arm64 [installed,local]
nvidia-l4t-libwayland-server0/now 36.4.2-20241212160716 arm64 [installed,local]
nvidia-l4t-multimedia-utils/now 36.4.2-20241212160716 arm64 [installed,local]
nvidia-l4t-multimedia/now 36.4.2-20241212160716 arm64 [installed,local]
nvidia-l4t-nvfancontrol/now 36.4.2-20241212160716 arm64 [installed,local]
nvidia-l4t-nvml/now 36.4.2-20241212160716 arm64 [installed,local]
nvidia-l4t-nvpmodel-gui-tools/now 36.4.2-20241212160716 arm64 [installed,local]
nvidia-l4t-nvpmodel/now 36.4.2-20241212160716 arm64 [installed,local]
nvidia-l4t-nvsci/now 36.4.2-20241212160716 arm64 [installed,local]
nvidia-l4t-oem-config/now 36.4.2-20241212160716 arm64 [installed,local]
nvidia-l4t-openwfd/now 36.4.2-20241212160716 arm64 [installed,local]
nvidia-l4t-optee/now 36.4.2-20241212160716 arm64 [installed,local]
nvidia-l4t-pva/now 36.4.2-20241212160716 arm64 [installed,local]
nvidia-l4t-tools/now 36.4.2-20241212160716 arm64 [installed,local]
nvidia-l4t-vulkan-sc-dev/now 36.4.2-20241212160716 arm64 [installed,local]
nvidia-l4t-vulkan-sc-samples/now 36.4.2-20241212160716 arm64 [installed,local]
nvidia-l4t-vulkan-sc-sdk/now 36.4.2-20241212160716 arm64 [installed,local]
nvidia-l4t-vulkan-sc/now 36.4.2-20241212160716 arm64 [installed,local]
nvidia-l4t-wayland/now 36.4.2-20241212160716 arm64 [installed,local]
nvidia-l4t-weston/now 36.4.2-20241212160716 arm64 [installed,local]
nvidia-l4t-x11/now 36.4.2-20241212160716 arm64 [installed,local]
nvidia-l4t-xusb-firmware/now 36.4.2-20241212160716 arm64 [installed,local]
Was able to get it by reverse engineering the sdkmanager online json starting with https://developer.download.nvidia.com/sdkmanager/sdkm-config/main/sdkml1_repo.json and eventually finding the path of the tarfile I was looking for.
https://developer.nvidia.com/assets/embedded/secure/tools/files/jetpack-sdks/jetpack-6.1/Jetson_61_b218/Jetson_Linux_R36.4.2_aarch64.tbz2
so download and extract that tarfile and then update to the latest debs using similar commands
sudo apt install --only-upgrade ~/Downloads/Linux_for_Tegra/nv_tegra/l4t_deb_packages/*.deb ~/Downloads/Linux_for_Tegra/tools/*.deb ~/Downloads/Linux_for_Tegra/bootloader/*.deb ~/Downloads/Linux_for_Tegra/kernel/*.deb
Installing the updated debs is not enough to get the “super” clocks as the original device trees for the jetson orin devkits were not updated but instead new “super” variant device trees were made. switching to the “super” variant can be done by manually updating /etc/nv_boot_control.conf (obtained from the rootfs of the nvidia rev 1 image, TNSPEC 3767-300-0005--1-0-jetson-orin-nano-devkit-super-) to the super variant and then re-configuring the nvidia-l4t-bootloader package so that the new “super” dtb is flashed to qspi on reboot. once that is done and /etc/nvpmodel.conf is removed and allowed to re-generate on boot the devkit uses the “super” variant without having to flash a new system
Hi,
The next Jetpack 6.2 will enable the mode for Orin Nano 4GB. Please wait for the next release.
This topic was automatically closed 14 days after the last reply. New replies are no longer allowed.