Getting Started with the EMAC OE SDK UPDATE 2019

Redirect page
Jump to: navigation, search
TODO: {{#todo: Complete (03.31.2015-13:42->BS+);(04.08.2015-15:00->BS+);(04.09.15-14:00->MD+);(04.21.2015-10:15->BS+);(04.24.2015-20:00->MD-);(04.28.2015-12:20->MD+);(04.29.2015-16:00->BS+);(04.29.2015-17:30->MD+);(04.30.2015-15:30->KY+); (07.22.2015-11:00->BS+)(10.07.2015-16:32->KY+)|Brian Serrano|OE 5.0,BS,MD,Complete,KY}}

The EMAC OE SDK is distributed with a set of example projects intended to demonstrate how to use the EMAC OE toolchain and libraries. This page demonstrates the process of compiling an example project and running it on the target machine. Also, it provides a straightforward guide to the essential steps you need to follow to get started with cross compiling a simple program with gcc and running the program on your embedded machine.

Getting Started with the EMAC OE SDK

Installing EMAC SDK

A Linux developer system is required to install the EMAC software development kit.

For those without a Linux developer system we recommend downloading Oracle VM VirtualBox and installing EMAC's pre-configured Linux Ubuntu virtual machine. All steps involved can be found here:
Set up virtual machine with EMAC_OE_SDK pre-intalled
Those with a Linux developer system already configured can follow the below steps:
SDK Install

The most current EMAC_OE_SDK release is version 5.1, all higher versions should be disregarded.

Connecting to a Target Board

Primarily a board is connected to through it's debug serial port. If it also connected to a network and the IP address is known, ssh can be used to connect as well. More information on establishing a physical connection to a board is available on the Serial Connections and Network Connections page.

The next step after establishing a physical connection to the board is logging in. For more information visit the System Log In page.

Setting the Filesystem to Read-Write

The root filesystem on a machine running EMAC OE Linux is mounted read-only by default. In order to put files on the board, you may need to make the filesystem read-write. The Linux Filesystem Organization page has more information regarding which parts of the filesystem are mounted read-only by default versus which parts are always writeable. Note that EMAC recommends installing your program into a read-only portion of the filesystem (according to the guide) to safeguard it from filesystem corruption (such as may be caused by removing power from the board without shutting down the operating system first, which is normal in embedded systems).

To remount the root filesytem in read-write mode, enter the following into the terminal:

root@ipac9x25:~# oemntrw

This will only change the root filesystem to read-write for the current boot. When the system is rebooted, the root filesystem will once again be mounted read-only.

Transferring Files

The command line syntax for transferring a file using the SSH protocol is scp file user@host:/directory. SCP, or Secure Copy, is a way of securely transferring files between a local and remote host. For example, to send the file example.text to the /usr/bin directory of a system with the IP address, enter the following command:

developer@ldc:~# scp example.text root@

Alternatively, you can leave off the path after the colon if you want the file to go directly into the root user's home directory.

developer@ldc:~# scp example.text root@

Remote Execution

SSH can also be used to execute programs on remote systems without logging in. The syntax for SSH remote execution is ssh user@host "my_command -args file". The following is an example of a command to run a program on a board with the IP address

developer@ldc:~# ssh root@ "/path/to/executable -args"

Basic Compiling

The two subsections below show the two common options for how to compile source code. The first subsection demonstrates how to use the EMAC CMake tool, while the second subsection demonstrates how to compile c code manually .

CMake Compiling

Host Machine Compiling

This section demonstrates how to use the EMAC CMake tool to generate CMake files automatically for a project. When using the EMAC SDK there are currently two options for cross compiling:

  • arm
  • x86

For the purposes of this guide, the arm option will be used for the listed examples.

Navigate to the directory where the project will be located. Then run the CMake new project script.

developer@ldc:~# cd projects
developer@ldc:~/projects# export PATH=/opt/emac/5.1/bin:$PATH
developer@ldc:~/projects# oe_init_project -n hello.c

If desired, please enter a name for this project, otherwise press Enter to use the default: hello_emac

-- Creating new project directory...
-- Creating new source file...
-- Building custom CMakeLists.txt file...
-- Done.

Do you want to create a build directory for this project? (y/n) y

-- Creating build directory...

Do you want to run cmake for this project? (y/n) y

-- Using system compiler
-- The C compiler identification is GNU 4.8.2
-- The CXX compiler identification is GNU 4.8.2
-- Check for working C compiler: /usr/bin/cc
-- Check for working C compiler: /usr/bin/cc -- works
-- Detecting C compiler ABI info
-- Detecting C compiler ABI info - done
-- Check for working CXX compiler: /usr/bin/c++
-- Check for working CXX compiler: /usr/bin/c++ -- works
-- Detecting CXX compiler ABI info
-- Detecting CXX compiler ABI info - done
-- Configuring done
-- Generating done
-- Build files have been written to: /home/developer/projects/hello_emac/hello_emac-build

Do you want to compile this project? (y/n) y

Scanning dependencies of target hello_emac
[100%] Building C object CMakeFiles/hello_emac.dir/hello.c.o
Linking C executable hello_emac
[100%] Built target hello_emac

The executable, in this case, is now inside the hello_emac/hello_emac-build directory.

Target Machine Compiling

The CMake project script has now made a project directory that contains the following:

  • CMakeLists.txt
  • Source code file (hello.c in this case)
  • README file
  • Desktop Build Directory (hello_emac-build in this case)

The CMakeLists.txt contains the required information to automatically create a Makefile for a given architecture. This was created by the EMAC oe_init_project script, and will need to be modified over time as a project grows. The comments in the file generated by the EMAC tool will provide a good starting point for how to add additional source files and how to perform other common tasks related to maintaining your CMake build system as your project grows. The CMake project provides extensive documentation on how to work with these files.

The source code file generated by the script (hello.c) contains a basic Hello World style program.

The README file contains more information on using CMake with the EMAC 5.X SDK.

The Desktop Build Directory (hello_emac-build) contains the executable hello_emac, the Makefile, and various cache files. These were automatically created by CMake and by the build system, and can be recreated at any time.

It is useful to have a Desktop Build Directory because it is easier (in the beginning) to use the desktop to verify all code changes before cross-compiling for a target board. This will be useful until the application under development depends upon resources which are only available on the target hardware, such as certain devices drivers or the touchscreen (if so equipped).

Use the following steps to cross-compile the project and send it to a target board.

  1. In a terminal, navigate to the base directory of the project.

    If the target board being used is x86, then change all occurrences of arm in the following sections below to x86.
  2. Create a build directory for cross compiling.

    developer@ldc:~/projects/hello_emac# mkdir hello_emac-build-arm
  3. Change directories into the newly created directory.

    developer@ldc:~/projects/hello_emac# cd hello_emac-build-arm
  4. Run cmake using the target architecture.

    developer@ldc:~/projects/hello_emac/hello_emac-build-arm# cmake .. -DARCH:STRING=arm
  5. Compile the code using make.

    developer@ldc:~/projects/hello_emac/hello_emac-build-arm# make

    The make command creates the target executable in the hello_emac-build-arm directory.

  6. Now copy the executable to the target board. Use any of the options listed under the Transferring Files section.

    The executable is now located on the target device at the designated transfer directory. It now can be run remotely (covered here) OR from the target device by navigating to the transfer directory and issuing the run command (./executable_name).

Manual Compiling

Host Machine Compiling

In these examples we are using the Ipac-9x25 board. Your board's processor's architecture will determine which file needs to be sourced. These files come from the EMAC SDK toolchain.

Create a file called hello.c using a text editor such as vi, nano, or gedit. Then copy and paste the source code from the Hello World System Log Example section in the text editor of your choice.

Once you've created and saved the file with the source code, use the following syntax to compile the program called hello.c:

developer@ldc:~# gcc -o hello hello.c

If there is no error in your code then the compiler will successfully create an executable file called hello in the current directory.

To verify this, enter the following command:

developer@ldc:~# ls -l hello*

-rwxrwxr-x 1 bserrano bserrano 9583 Apr 6 12:45 hello
-rwxr-xr-x 1 bserrano bserrano 129 Apr 6 12:28 hello.c

To run the program, enter the following command:

developer@ldc:~# ./hello

Hello EMAC OE!


developer@ldc:~# /path/to/hello

Hello EMAC OE!

Target Board Compiling

To compile for the target board, you must source the environment-setup-armv5e-emac-linux-gnueabi file.

developer@ldc:~# source /opt/emac/5.1/environment-setup-armv5e-emac-linux-gnueabi

Once you are in the directory with the source file, enter the following command:

developer@ldc:~# $CC -o hello hello.c

Once you source the file in your current terminal, you can only use it to cross-compile your program for the target board. You can no longer compile it for your host machine. To compile it to your host machine, simply open a new terminal.

After sourcing the file, copy the program over to the target board using scp. Enter the following command:

developer@ldc:~# scp hello root@

root@'s password:

hello 100% 9583 9.4KB/s 00:00

After copying the program, you can now execute the program on the target board. Enter the following command:

developer@ldc:~# ssh root@ ./hello

root@'s password:
Hello EMAC OE!

Remote Debugging

When working with embedded systems the binary is usually compiled on a development machine with a different CPU architecture than what is on the target machine. This can be a problem when, as is typically the case, the target machine lacks the system resources to run a debugger. In these cases, it is possible to use the GNU debugger (gdb) on the development machine to remotely debug the target machine provided it has a flavor-matched version of a program named gdbserver. All EMAC OE builds are packaged with a flavor-matched gdbserver to simplify the setup process for developers.

A flavor-matched gdbserver is one which is built against the same source code revision as the gdb client for your desktop, because gdb does not have a stable interface for attaching to the gdbserver. In fact, the interface changes slightly with almost every release, so if you find yourself having difficulty getting gdb and gdbserver to communicate with each other, make sure they're the exact same version before trying any other debugging steps.

For more information visit the Remote Debugging EMAC OE SDK Projects with gdbserver page.


Hello World System Log Example

This example will print Hello EMAC OE! to the syslog facility as well as the console. This will allow you to log, debug, and send status messages to the system logger.

To compile and run this program, see the sections above.

#include <stdio.h>
#include <stdlib.h>
#include <unistd.h>
#include <syslog.h>

int main(int argc, char** argv)
    char message[] = "Hello EMAC OE!";

    openlog("slog", LOG_PID|LOG_CONS, LOG_USER);
    syslog(LOG_INFO, "%s", message);

    printf("%s\n", message);

    return 0;

This is extremely useful because it allows you to save a record of the output that you might not see first hand.

To verify the output of the program went into syslog, enter the following command:

developer@ldc:~# tail /var/log/syslog

Apr 7 14:10:06 ENG-26-LX dhclient: DHCPACK of from
Apr 7 14:10:06 ENG-26-LX dhclient: bound to -- renewal in 3306 seconds.
Apr 7 14:17:01 ENG-26-LX CRON[21193]: (root) CMD ( cd / && run-parts --report /etc/cron.hourly)
Apr 7 14:54:41 ENG-26-LX hpcups[21266]: prnt/hpcups/HPCupsFilter.cpp 689: First raster data plane..
Apr 7 14:55:31 hpcups[21266]: last message repeated 3 times
Apr 7 15:05:12 ENG-26-LX dhclient: DHCPREQUEST of on eth0 to port 67
Apr 7 15:05:12 ENG-26-LX dhclient: DHCPACK of from
Apr 7 15:05:12 ENG-26-LX dhclient: bound to -- renewal in 3559 seconds.
Apr 7 15:17:01 ENG-26-LX CRON[21302]: (root) CMD ( cd / && run-parts --report /etc/cron.hourly)
Apr 7 15:27:08 ENG-26-LX slog[21375]: Hello EMAC OE!

As you can see on the bottom line, your program output has been recorded and date stamped in syslog.

For more information on system logging visit the System Logging page.

Further Information

Where to Go Next
Pages with Related Content