I am currently working with a JIT compiler that compiles C++ code. I am trying to modify the compiler so that it is able to handle OpenMP directives. I have passed the -fopenmp flag (among other flags) to the compiler as an argument in:
clang::CompilerInvocation::CreateFromArgs();
When I run a simple test, the compiler tells me that the omp.h header isn’t found when it encounters the #include <omp.h> statement in the code it is trying to compile.
The project is currently setup in a way where it grabs all the headers it needs from a single directory within the project, and omp.h is obviously not there but it is on my system in general and clang with -fopenmp will work outside of this project. Is there a way other than -fopenmp to link the library to the rest of the project so that the compiler can see it? Or am I adding it at the wrong step maybe?
I am a beginner, so I apologize if I am missing something obvious here.
Any advice is appreciated, thanks in advance 