Boot from SD Card Rather than eMMC on Jetson Nano Developer Kit?

Robotics & Edge Computing Jetson & Embedded Systems Jetson Nano
1

I have a Nvidia Jetson Nano 4GB with eMMC 16GB devleloper kit for a university project. The current OS is L4T R32.6.1 (Jetpack 4.6) which I believe to be booting from eMMC rather than the micro sd card that I flashed with Jetpack 4.6.1.

Is it possible for the Jetson to boot from the sd card rather than the eMMC? The 14GB size of /dev/mmcblk0p1 is so small and I would like to boot from the sd cardthat has 128GB so I can use it as the main storage and root file system. I don’t have a Linux machine so I cant do any of that SDK stuff. I only have a windows PC. How can this be solved? Please Help a confused college student

2

Just to clarify some points. There are mistakes in your comment that lots of users got confused.

  1. I wrote a guidance here to tell if you are using a NV devkit. From your comment, I think it is not a NV devkit as we don’t sell anything that has both eMMC and sdcard altogether.

  1. If your board has a sdcard, then it is a custom board from specific vendor. We won’t know how to enable that sdcard for you because we don’t have the hardware schematic of that board. You should check the vendor’s website and see if they provided any customized BSP for it.

  2. BSP means the package that is used to flash the board. Vendor will use sample from sdkmanager and create their own one for their own board.

3

Yeah there is no Nvidia logo on the back of the board. Instead the back says:

LEETOP 900-44555-0000 REV:2.0
SUB kit_NANO

It was bought from: Amazon.com: Jetson Nano 4GB RAM 16G eMMC onboard for AI Robotics Machine Learning (Heat Sink Version) : Electronics

I thought this was just a regular Nvidia product at the time because I did not know anything about Jetsons. Would I be able to boot from my micro sd card plugged into the board?

4

The biggest issue I am running into I’d say is that I am trying to install tensorflow 2.5 from Index of /compute/redist/jp/v46/tensorflow
but I keep getting installation errors like “Failed to build wheel” or “Failed to build h5py” even though I am inside a Python3.6 venv and have installed dependencies and prerequisites from here:

Our official TensorFlow release for Jetson Nano!

Python 3.6+JetPack4.6.3

$ sudo apt-get update
$ sudo apt-get install -y python3-pip pkg-config
$ sudo apt-get install -y libhdf5-serial-dev hdf5-tools libhdf5-dev zlib1g-dev zip libjpeg8-dev liblapack-dev libblas-dev gfortran
$ sudo ln -s /usr/include/locale.h /usr/include/xlocale.h
$ sudo pip3 install --verbose 'protobuf<4' 'Cython<3'
$ sudo wget --no-check-certificate https://developer.download.nvidia.com/compute/redist/jp/v461/tensorflow/tensorflow-2.7.0+nv22.1-cp36-cp36m-linux_aarch64.whl
$ sudo pip3 install --verbose tensorflow-2.7.0+nv22.1-cp36-cp36m-linux_aarch64.whl
5

As I already mentioned, you should check the vendor who made that carrier board. We only support NVIDIA devkit case here. Your board is not a NV devkit.

Amazon has mentioned the vendor seems to be “WayPonDEV”. Maybe you shall check if their website provided anything.

Also, I don’t see any reason that you should ask tensorflow question in this post. If it is not related to original topic, please file a new one for it.

6

This topic was automatically closed 14 days after the last reply. New replies are no longer allowed.