This repository has been archived on 2024-06-02. You can view files and clone it, but cannot push or open issues or pull requests.
upsilon-docker/README.md

2.8 KiB

Upsilon docker development environment setup.

Setup steps

Rebinding Caps Lock to Escape

gsettings set org.gnome.desktop.input-sources xkb-options "['caps:swapescape']"

Installing OpenFPGALoader

Then install openFPGALoader. This utility entered the Ubuntu repositories in 23.04. Install and compile it if you do not have it. Install the udev rule so that admin access is not required to load FPGA bitstreams.

Setup Rootless Docker

Docker allows you to run programs in containers, which are isolated environments. Upsilon development (at the Maglab) uses Docker for reproducibility: the environment can be set up automatically, and re-setup whenever needed.

If you have issues with docker, try adding to ~/.config/docker/daemon.json

{
   "storage-driver": "fuse-overlayfs"
}

Download and Install Python3

Install python3-venv (or python3-virtualenv) and python3-pip.

Clone External Repositories

Run make clone. You may need to download the upsilon repositories and put them in the same folder as the Makefile.

Setup Network

Plug in your router/switch to an ethernet port on your computer. If your computer is usually wired to the network, you will need another ethernet port (a PCI card is ideal, but a USB-Ethernet port works).

Set the ethernet port to static ip 192.168.1.100/24, netmask 255.255.255.0, gateway 192.168.1.1. Make sure this is not the default route. Make sure to adjust your firewall to allow traffic on the 192.168.1.0/24 range.

If your local network already uses the 192.168.1.0/24 range, then you must modify upsilon/firmware/soc.py to use different IPs. You must rebuild the SoC after doing this.

Setup Images

Run make images to create all docker images.

Setup and Run Containers

For NAME in hardware, opensbi, buildroot:

  1. Run make $NAME-container to build the container.
  2. Run make $NAME-copy to copy Upsilon's code into the container.
  3. Run make $NAME-execute to build the data.
  4. Run make $NAME-get to retrieve the build artefacts.
  5. (Optionally) run make $NAME-clean to delete the container.

If you do not delete the container you can run

make $NAME-copy $NAME-execute $NAME-get

when you need to rebuild. If you need shell access, run make $NAME-shell.

Flash FPGA

Plug in your FPGA into the USB slot. Then run

openFPGALoader -c digilent upsilon/boot/digilent_arty.bit

Launch TFTP Server

Install py3tftp (pip3 install --user py3tftp). Then run make tftp to launch the TFTP server. Keep this window open.

Launch FPGA Shell

Run litex_term /dev/ttyUSB1. You should get messages in the window with the TFTP server that the FPGA has connected to the server. Eventually you will get a login prompt: you have sucessfully loaded Upsilon onto your FPGA.