As a beginner with Virtual Machines and Linux I have my trouble understanding how to properly upgrade Linux on the Tegra - Board. I found some good explanations but they all were too advanced for a beginner - a student - and not a professional in the field.
Therefore, I would like to know how to properly upgrade the Linux Version on the Tegra X1 Board with a Windows machine
Step1:
Make sure you have the following items:
These are all the things needed before.
Step 2
Download Linux 14.04 - it is the only distribution compatible with the Tegra Board at the moment.
http://releases.ubuntu.com/14.04/ubuntu-14.04.4-desktop-amd64.iso
Step 3
Download and install Oracle Virtual Machine. You will need admin privilages for install.
http://download.virtualbox.org/virtualbox/5.1.8/VirtualBox-5.1.8-111374-Win.exe
Step 4
Set up an NVIDIA Developer Account
https://developer.nvidia.com/group/node/873376/subscribe/og_user_node?downloadable_file=874988
Step 5
Set up the virtual machine. To do so start the installed Oracle VM Virtual Box (see Step 3). On the upper left you find the button "new". After clicking on it a window will pop up. At the bottom you can change to "Expert-Mode". In the field "Name" you can give it a fitting phrase like "Ubuntu for Tegra" etc.. The next field Typ should be obiously Linux, and the Version 64 bit.
Set the memory size to a good fitting size, depending on how much RAM you got on your machine. It works fine with 6 GB, anything smaller could lead to some lagging, but will still run.
Put the radio button in the middle so it will create a hard drive.
Go to the next step by hitting "Create"
Step 6
In this window two inputs are interesting. Firstly, it needs a path where to create the virtual environment. So choose a path to a disk that has enough space. Secondly, how many space you give to the environment - 50 gb will work fine.
Step 7
Launch it by selecting the newly created environemt and then hitting "Start". You will be asked to choose a medium to be booted. Here select the downloaded .iso file from step 2.
Step 8
You will be greeted with the Linux-Install environemnt. You can choose between "Try" and "Install". You must choose "Install".
Step 9
After you instaleld Linux you must restart the Virtual Machine. To do so, you can either do it by "normaly" shutting down Linux via the GUI or the command Line tool or from the VM-Software directly by right-clicking on the running virtual machine - close - power down.
Step 10
You may encounter the problem that you do not see the full screen of the Linux environment. To fix this you need to restart the virtual machine. On the virtual machine display at the top bar you can see the entry "devices". If you click on it a drop down menu will open, the last point is "guest additions", click on it and install them. After that reboot the virtual machine.
Step 11
On the virtual machine log in on your NVIDIA Account and download the latest Jetpack Version.
https://developer.nvidia.com/embedded/downloads?#?dn=jetpack-for-l4t-2-3 https://developer.nvidia.com/embedded/jetpack
Step 12
After downloading a file with the extension .run should be in your Downloads-Folder. This is the installation file needed, but it is not executable yet. To make it executable open a shell (right click on the upper left ubuntu symbol, search for terminal and open it). Go to the Downloads folder with:
cd ~/Downloads
and make the run file executable:
sudo chmod u+x *.run
Step 12
Run the .run file with
sudo *./run
Step 12
In the installer choose the board and the software you want to be installed, also agree the software license agreements. After some downloading time the installer will open a terminal.
If the prompt asks you about Network Layout. If it does, choose eth0 if you have you board connected via ethernet cable, if it is connected via Wi-fi choose wlan0.
Step 13
You need to put the Tegra Board in recovery mode. Make sure that all your data is saved, since it will wipe everything clean. Follow the instructions on the terminal to put the Tegra Board in recovery mode. If directions unclear follow this youtube video (which also includes some followign steps):
Step 14
When you think the Tegra is in recovery mode check by doing the following: At the virtual machine, on the top bar go the "devices" and then to USB. Select the NVIDIA entry. If it is not there, the board is not in recovery mode. Make sure that this was really selected. (It is highlighted blue when selected)
Step 15
Back in the Linux virtual machine enter lsusb on a second terminal. If there is an entry with NVIDIA Corp the tegra board was successfully put into recovery mode. Press enter, now the flashing starts - this will take some time
Step 16
After flashing finishes, the jetson board will auomatically boot. Connect it to an HDMI Cable and Monitor. If a login is asked, the username and password are ubuntu.
Step 17 Connect to the Internet Either connect it to the ethernet or a Wifi - depending on what you have chosen at step 12. You may need to disable Wifi to connect via ethernet cable.
Step 18 If you use static IPs you can skip this part since you already now the IP-Adress you gave to the tegra board. If not you have to run
ifconfig
in a terminal. It will show you your adapters and what IP-Adresses they have, note the one that you chose.
Step 19
The Post Installation in the virtual machine either has given up and was unable to determine the IP adress of the Jetson Developer Kit or has found it. If it has not found it give it manually the ip adress you found out by entering "2". After this a GUI is shown where you can enter the Ip-Adress and the username password combo, which is ubuntu.
Step 20
After hitting Next the installation will continue.