Raspberry Pi: Cross compile HelloWorld

In this tutorial you will learn the basics how to cross compile a simple program for the Raspberry Pi 2 Model B.

Why would you like to cross compile?

The two major reasons why you would like to cross compile are:

1. Increasing speed of compilation
2. Decreasing wear out of SD card

Besides of that the lack of (disk) space can be an argument to decide to cross compile.

In this tutorial, cross compilation is performed at a desktop-computer with an Intel i5 processor, running Ubuntu 15.10 at 64 bits.

1. Create your working environment

   First create a new working directory and step into this directory.

   mkdir ~/raspberry
   cd ~/raspberry
   

2. Retrieve the Raspberry Tools

   Clone the Raspberry Pi Tools GIT repository. Among the tools is a complete cross compilation environment.

   git clone https://github.com/raspberrypi/tools

3. Create your program

   Compose a simple program which outputs a text on the console. Create a file named
   ~/raspberry/helloworld.c with your favourite text editor like ‘vi‘ or ‘gedit’ containing these contents:

   #include 
   
   int main(void){
           printf("Hello world!\n\n");
           return 0;
   }
   

4. Create a Makefile

   Create a new file ~/raspberry/Makefile with these contents. Be sure to use TABs instead of SPACEs as indentation characters.

   helloworld: helloworld.o
   	$(CC) $(LDFLAGS) helloworld.o -o helloworld
   helloworld.o: helloworld.c
   	$(CC) $(CFLAGS) -c helloworld.c
   clean:
   	rm *.o helloworld
   

5. Compile your program

   First we will export the environment variable CC containing as value the location of the cross compiler. At the same line we will start the make process.

   export CC=~/raspberry/tools/arm-bcm2708/arm-bcm2708hardfp-linux-gnueabi/bin/arm-bcm2708hardfp-linux-gnueabi-gcc && make

6. Check output

   When everything went well, in the current directory two new files are created: helloworld.o and helloworld. The later is of interest. This is our compiled program for the target environment. Check the target architecture by issuing the command:

   file helloworld
   

   … the output should read something like this:

   helloworld: ELF 32-bit LSB executable, ARM, EABI5 version 1 (SYSV), dynamically linked, interpreter /lib/ld-linux.so.3, for GNU/Linux 3.1.9, not stripped
   

   Here we see that it is compiled for a 32 bit ARM architecture. This is at least completely different as the architecture of the build computer: 64 bit, Intel i5.

7. Transfer to Raspberry Pi

   Now we must transfer this file to the Raspberry Pi. For simplicity is assumed that the Raspberry Pi is up-and-running already, including a network connection and SSH-daemon, so that we don’t have to leave our chair. In other cases you can use USB-storage or the SD-card to transfer the file. We can reach the Raspberry Pi via SSH using the host name ‘raspberrypi’ and can login as user ‘pi’ and password ‘raspberry’ (these are the defaults for Raspbian).

   scp helloworld pi@raspberrypi:/home/pi/
   

8. Try your program

   Now login into the Raspberry Pi and try your program

   ssh pi@raspberrypi
   ./helloworld
   

   The output should be: “Hello world!”

   In case you see the error “Permission denied“, check the execution bit of the program file and try again.

   chmod +x helloworld
   ./helloworld