I am trying to run the template OpenMP "Hello World" project, according to the directions here:
http://processors.wiki.ti.com/index.php/OpenMP_on_C6000
Per the directions, the only modification I made to the template project is the following line added to the .cfg file:
OpenMP.autoDnldCore = false;
The project builds without issues. When I click the debug button, it proceeds to load the .out file onto the individual cores. Core 1 loads first (and without issue), but when it gets to core 2 it gets stuck. I eventually have to kill Code Composer through the task manager. Below is a screenshot of where it gets stuck.
I can get the project to work by loading the cores manually. I disable "Run to symbol main on program load or restart" for each core. I then manually load each core with the .out file. Each core is suspended at .cinit00. I then run cores 1,2,3, then core 0. And I get the expected output.
What is causing this?