zhijingy / ese519-lab2

ESE519-lab2

The machine I'm using is MacBook Pro, 13-inch, 2018, macOS version is Big Sur 11.7. Processor is 2.3GHz Quad-Core Intel Core i5. To complete the setup, I have cloned the pico-examples and pico-sdk from github.

Step 1: Download required softwares

I have downloaded Visual Studio Code. Visit the website, and downloaded the macOS version.

I used Homebrew to install all the other softwares. If you don't have Homebrew yet, you can folllow the instruction given in the Pico SDK Getting Started Guide. The magic command for installing brew is given as:

$ /bin/bash -c "$(curl -fsSL https://raw.githubusercontent.com/Homebrew/install/master/install.sh)"

After downloading Homebrew, we can follow the Section 9.1.1 in Pico C SDK Getting Started Guide to download CMake and arm-none-eabi-gcc GNU Arm Embedded Toolchain by simply typing in the terminal:

$ brew install cmake
$ brew tap ArmMbed/homebrew-formulae
$ brew install arm-none-eabi-gcc

A serial console is also needed for this lab to view any output from the program running on your board. Screen is a good choice. Simply type "brew install screen" in your terminal. After downloading screen, here is a wonderful guide about how to use it to view the output of your board. Basically to view the output of RP2040 on your terminal, you need to firstly figure out which serial port is assigned to your board. This can be done by typing ls /dev/tty.* without connecting the RP2040 to your computer. You will see something like this:

$ /dev/tty.Bluetooth-Incoming-Port

Then plug RP2040 to your computer, give the ls /dev/tty.* command again, a new serial port will show up. That will be the port for RP2040 on your computer. Here I give the example:

$ /dev/tty.Bluetooth-Incoming-Port	/dev/tty.usbmodem14201

In this case, /dev/tty.usbmodem14201 is the serial port for your RP2040. Typing screen /dev/tty.usbmodem14201 115200 will bring you to the screen mode in which you can monitor the output of your board.

Step 2: Clone required github files

In your terminal, under a directory you consider proper, type in mkdir pico to create a folder called "pico", and cd pico to get into the folder. Then in the "pico" directory, use git to clone the required PICO-SDK and PICO-EXAMPLES files.

If you don't have or never use git, install it on your computer. After installing git onto your computer, clone the files to local "pico" folder by the commands:

$ git clone -b master https://github.com/raspberrypi/pico-sdk.git
$ cd pico-sdk
$ git submodule update --init
$ cd ..
$ git clone -b master https://github.com/raspberrypi/pico-examples.git

If difficulty is met when cloning the pico files, you can also download the files manually by clicking the green "Code" button and choose "Download ZIP". Put the downloaded "pico-sdk" folder and "pico-examples" folder into the "pico" folder.

Step 3: Setup VScode

Open VScode, on the sidebar choose "Extensions", search for "cmake", install the "CMake" and "CMake Tools". The position of "Extensions" and extensions to be downloaded are shown below:

According to the section 9.1.3 in the getting started pdf, open the "pico-examples" folder in VScode, create a folder under the "pico-examples" directory and name it ".vscode". Inside the ".vscode", create a file called "settings.json". Include the following text in the "settings.json" file.

{
  "cmake.environment": {
  "PICO_SDK_PATH":"../../pico-sdk"
  },
}

After editing and saving the json file, go to the "setting" of vscode(the Cog Wheel at the navigation bar on the left-hand side), click "Extensions" and then click "CMake Tools". At "Cmake: Generator", set the Cmake generator as "Unix Makefiles". Then go to the file menu of VScode and click on "Add Folder to Workspace…", then navigate to "pico-examples" folder and click "Okay". The project will load and you’ll (probably) be asked to choose a compiler. If so, choose the "GCC 10.3.1 arm-none-eabi". If not, click the "tool" icon at the bottom of VScode window, and choose the arm-none-eabi compiler as your kit. The icon to choose and compiler to use are indicated as the picture below:

If you can successfully choose the compiler, now you can build all the binaries files by clicking on the "Build" button (with a cog wheel) in the blue bottom bar of the window. The icon is shown below:

This will create a build folder in the "pico-example" directory.

One thing worth notice but not being mentioned in the guide is that even with the "settings.json" defining the value of "PICO_SDK_PATH", it's highly possible that during the build the compiler cannot see the definition of "PICO_SDK_PATH". In this case, go to "setting" again, and click into the "Cmake Tools" window, in both the "Cmake: Configure Environment" and "Cmake: Build Environment" set the variable "PICO_SDK_PATH" as "../../pico-sdk". This can be seen as below:

However, if VScode cannot detect the arm-none-eabi compiler, there's another way to build the binaries files. In your pico-examples directory, cd build, then cd hello_world. Now we are in the hello_world folder, and we are ready to build with make -j8.

Step 4: Run the program on your RP2040

Upon the success of build, we can happily see this kind of result:

Now plug your RP2040 into your computer. Press and hold the RESET button. While holding RESET, press and release the BOOTSEL button. Release the RESET button once you see an RPI-RP2 disk showing up on your desktop as shown below:

Go to Finder and find the "hello_world" folder under the build, visit the usb folder, drag and drop the hello_usb.uf2 file onto the RPI-RP2 icon. As soon as the file is dropped onto it, the RPI-RP2 will disappearr, indicating that it's now running the program. To monitor the output, we can open a terminal window, type screen serial_port_of_board 115200. In my case, it will be screen /dev/tty.usbmodem14201 115200. 115200 is the data transmission speed we set. After entering the screen mode, we can view the output of "hello_world" as shown:

Congrats!!!