Yocto Support for NVIDIA Jetson Platforms - Setting up Yocto
Yocto Support for NVIDIA®Jetson™ |
---|
Setting up Yocto |
Flashing Jetson Platform |
Accessing the Board |
Adding NVIDIA Packages |
DeepStream |
Additional Topics |
FAQ |
Contact Us |
Select your Machine
You can apply the instructions in this wiki for multiple devices, so, whenever you see <MACHINE>, replace it (with the <> symbols included) with your target platform. Below you will find the possible <MACHINE> substitutions.
Device | <MACHINE> | |||
---|---|---|---|---|
TX1 | jetson-tx1-devkit | |||
TX2 | jetson-tx2-devkit | jetson-tx2-devkit-tx2i* | jetson-tx2-devkit-4gb* | |
XAVIER | jetson-agx-xavier-devkit* | jetson-agx-xavier-devkit-8gb* | ||
NANO | jetson-nano-devkit | jetson-nano-devkit-emmc* | ||
XAVIER NX | jetson-xavier-nx-devkit* | jetson-xavier-nx-devkit-emmc* | ||
Note: the machines with * have NOT been tested by RidgeRun yet |
Install Dependencies
1. Use the following command to install Yocto dependencies on Ubuntu.
sudo apt-get install gawk wget git-core diffstat unzip texinfo gcc-multilib build-essential chrpath socat libsdl1.2-dev xterm
2. Create a working directory for Yocto
YOCTO_DIR=/home/$USER/yocto-tegra
mkdir $YOCTO_DIR
3. Select the latest branch (as per May 2020)
export BRANCH="dunfell"
4. Download Yocto's Poky base meta-layer
cd $YOCTO_DIR
git clone -b ${BRANCH} git://git.yoctoproject.org/poky.git poky-${BRANCH}
5. Download meta-tegra
git clone -b ${BRANCH}-l4t-r32.5.0 https://github.com/madisongh/meta-tegra.git
Compile Yocto image
1. Initialize the build environment
cd $YOCTO_DIR
source poky-${BRANCH}/oe-init-build-env build
2. Setup the required settings
local.conf
- Open the file $YOCTO_DIR/build/conf/local.conf with your favorite text editor and append the following lines to the end of the file.
- Setup the SSTATE and DOWNLOADS directories with a custom path to avoid re-downloading and re-compiling in the future.
- Add the name of your target platform <MACHINE>.
- Set IMAGE_CLASSES and IMAGE_FSTYPES to easily flash the target board.
- Set 3.6 as the python3 and python3-native preferred version.
The python recipes are NOT included in the main meta-tegra repository yet. |
Note: Replace <MACHINE> with your selected machine (don't delete the quotes "")
MACHINE ?= "<MACHINE>"
DISTRO_FEATURES = "x11 opengl "
IMAGE_CLASSES += "image_types_tegra"
IMAGE_FSTYPES = "tegraflash"
SSTATE_DIR ?= "/home/${USER}/Yocto/sstate_dir"
DL_DIR ?= "/home/${USER}/Yocto/downloads"
PREFERRED_VERSION_python3 = "3.6%"
PREFERRED_VERSION_python3-native = "3.6%"
Optional: By default, Yocto uses all the available cores and threads of the computer. If you want to limit the resources used on the build, you can add the following lines to your $YOCTO_DIR/build/conf/local.conf file. Change the parameters according to your needs and to your host machine specification.
BB_NUMBER_THREADS = '6'
PARALLEL_MAKE = '-j6'
bblayers.conf
Open the file $YOCTO_DIR/build/conf/bblayers.conf with your favorite text editor and add the meta-tegra layer to the build/conf/bblayers.conf file to load the provided recipes:
BBLAYERS ?= " \
/home/${USER}/yocto-tegra/meta-tegra \
/home/${USER}/yocto-tegra/poky-dunfell/meta \
/home/${USER}/yocto-tegra/poky-dunfell/meta-poky \
/home/${USER}/yocto-tegra/poky-dunfell/meta-yocto-bsp \
"
3. Build the image
Use an image with developer tools such as GStreamer, NVIDIA plugins, etc
bitbake core-image-sato-dev
Or you can choose a different image from the following table
<IMAGE> | ||
---|---|---|
core-image-sato-dev | ||
core-image-weston | ||
core-image-base |
Legacy instructions
Please visit the following links for older versions of Jetpack.