Maximizing Performance with Profiling

Profiling is an analysis activity that tells you where the most time is being spent in executing your source code. The IDE provides the Application Profiler tool so you can sample the execution position of a running process, count how many times functions are called, or measure function runtimes.

From the IDE, you can build an application for profiling and launch it on a QNX Neutrino target. As the application runs, the profiling results are sent to the IDE. Examining these results helps you pinpoint inefficient areas in your code, without having to follow its execution line by line. You can optimize slow functions (e.g., by refining algorithms, reducing heap allocating and freeing operations), then run another profiling session to see the improvement.

The Application Profiler tool can work with the System Profiler, to record function execution times during a kernel event trace. You can then see how other system activity influences your application's performance.