Automotive Linux Wiki

The AGL source code and Yocto layers are maintained on the AGL gerrit server. For information on how to create accounts for gerrit see Getting Started with AGL.

Build instructions for supported hardware platforms available at AGL documentation site.

AGL Uses the Repo tool for managing repositories. First you need to setup layers of AGL. You can use these commands.

Prepare Repo:

  $ mkdir ~/bin
  $ export PATH=~/bin:$PATH
  $ curl https://storage.googleapis.com/git-repo-downloads/repo > ~/bin/repo
  $ chmod a+x ~/bin/repo

To download all code from master:

  $ repo init -u https://gerrit.automotivelinux.org/gerrit/AGL/AGL-repo
  $ repo sync
  

To download all layers for the latest stable release and more see the release notes

  $ repo init -b icefish [-m icefish_9.0.0.xml] -u https://gerrit.automotivelinux.org/gerrit/AGL/AGL-repo
  $ repo sync

To download all layers on the current ​release ​branch which may be in the midst of testing or changes prior to the next stable release:

  $ repo init -b icefish -u https://gerrit.automotivelinux.org/gerrit/AGL/AGL-repo
  $ repo sync

Repo set up information, release notes, and more for previous stable versions of AGL

Grumpy Guppy Releases

Funky Flounder Releases

Electric Eel Releases

Daring Dab Releases

Charming Chinook Releases

Brilliant Blowfish Releases

Agile Albacore Releases

Set up the development environment for the desired target hardware. AGL has created a set up script for defining the target build and desired optional features. To get a complete list of the options available run.

  $ source meta-agl/scripts/aglsetup.sh -h

Once you run aglsetup.sh with your desired paramaters, you can build any target desired. Some examples follow.

Here is the list of features for AGL that can be specified in the aglsetup.sh command line:

• in meta-agl
  • agl-devel: activate development options (empty root password, debugger, strace, valgrind …)
  • agl-netboot: enable network boot support through TFTP and NBD (see meta-netboot layer)
  • agl-appfw-smack: enables IoT.bzh Application Framework + SMACK + Cynara
  • agl-demo: enable layer meta-agl-demo and meta-qt5 - required to build agl-demo-platform
  • agl-localdev: add a local layer named “meta-localdev” in meta directory and a local.dev.inc conf file if present

''CHECK OUT: source meta-agl/scripts/aglsetup.sh -h''

For newer features or to get more details on a given feature, take a look at the configuration files stored for each feature and/or each machine in meta-agl/templates and meta-agl-*/templates.

To build the QEMU version of the AGL demo start use machine “qemux86-64” and feature “agl-demo”

  $ source meta-agl/scripts/aglsetup.sh -m qemux86-64 agl-demo agl-netboot   
  
  $ bitbake agl-demo-platform

To get kernel source tree, you have to build kernel once, to do that, you can use these commands,

  $ bitbake virtual/kernel

Then you can find kernel source tree at

  ls build/tmp/work*/*/linux-yocto/*/linux/

or

  ls build/tmp/work*/*/linux-*/*/git/

More detailed instructions are maintained in the README.md file located in meta-agl repository.

Here is a good manual for developing kernel code http://www.yoctoproject.org/docs/current/kernel-dev/kernel-dev.html

You can choose from 2 ways to add code to AGL kernel.

1. Writing a patch and writing recipe to apply it. See 2.2. Modifying an Existing Recipe
2. Writing recipe to build modules out of tree See 2.5. Working with Out-of-Tree Modules

Follow the step by step instructions for building AGL for other supported targets:

• Building AGL for Raspberry Pi
• Building AGL for Renesas Porter
• Building AGL for Renesas Gen3 boards