This article is part of a ‘mini series’ about hidden gems, tips and tricks around Eclipse.
The topic of this one is how to organize projects in the workspace with ‘Working Sets’.
Working Set Selection
This article is part of a ‘mini series’ about hidden gems, tips and tricks around Eclipse.
The topic of this one is how to comment and un-comment a block of source code.
Block Comment
This article is part of a ‘mini series’ about hidden gems, tips and tricks around Eclipse.
The topic of this one is how use Code Completion in the editor.
Library Code Completion
This article is part of a ‘mini series’ about hidden gems, tips and tricks around Eclipse.
The topic of this one is how to show where a project is located on the host system.
Show in System Explorer
This article is part of a ‘mini series’ about hidden gems, tips and tricks around Eclipse.
The topic of this one is how to accelerate the start of the debugger.
Nothing to be done
Creating a new project with Eclipse for a microcontroller these days is fairly easy, and I have the choice if I want to start the project with C or C++:
Choice of C and C++ for a new project
Still the embedded microcontroller world is dominated by C and not C++. So while it is easy to start with a C++ project, most vendor provided example or tutorial project are C projects. So how can I transform such project to C++?
When using a logging framework it is useful to use the current file name or line number. The ANSI C/C++ standard defines the __LINE__ and __FILE_ preprocessor macros for this.
But what about the project name, if it is a release or debug build, the microcontroller used or other things like the operating system which was used to build the binary?
Target Chip Name
This (and even more) can be easily provided by Eclipse to the C/C++ application being built with CDT.
This is a follow-up article of my earlier project presented in “FatFS, MinIni, Shell and FreeRTOS for the NXP K22FN512“. I wanted to extend it with a USB MSD (memory stick) device: The USB storage device gets automatically mounted, and depending on a configuration (.ini) file on the memory device I can perform various actions, for example automatically copy data from the SD card to the USB device. For example the system logs data, and to get the data I insert the memory stick, it copies the data on it and automatically unmounts it, and I can remove the memory stick.
FRDM-K22F USB Host Mode with Memory Sticks
The NMI is a special interrupt on ARM Cortex-M architecture: as the name indicates, it cannot be ‘masked’ by the usual ‘disable interrupts’ flags (PRIMASK, BASEPRI), similar to the Reset signal.
cortex-m-vector-table (Source: adapted from arm.com)
Dealing with the reset signal is kind of obvious, and most designs and boards have it routed to a reset button or similar. The NMI is less obvious if you don’t pay attention to it: most ARM-Cortex implementations and boards have the NMI signal routed to a pin and are ‘hiding’ it in the schematics behind a normal GPIO pin or port: if you don’t pay attention to the NMI functionality, the board might not work as intended.
An essential tool especially developing larger applications or distributed firmware is to use logging. This article presents an open source logging framework I’m using. It is small and easy to use and can log to a console, to a file on the host or even to a file on an embedded file system as FatFS.
Log Output
MCU on Eclipse 