OpenMP lowering from PFT to FIR

[clementval]

As I mentioned in the meeting, I think having more abstraction in the dialect would be nice to have.
There has been some proposal to have a proper representation in the dialect. I could find one of the discussion here but I believe there is an older one that I could not found.

Issues with the current implementation of privatization in OpenMP with Fortran

create a copy of the original variable outside the parallel region and then use this for firstprivatisation in the region? This sounds more reasonable to me. insert a barrier after privatisation (which generally is the first thing to happen in the parallel region) I am concerned if it leads to some overhead if copy a large array. We don’t need to do this for all data types. For fortran, we only need to do this for pointer/(allocatable?), variables in equivalence and the equivalenced va…

We have such abstraction in the OpenACC dialect and you could borrow from it as a starting point.

Regardless of changing the lowering I think that having abstraction drastically simplify the lowering from PFT to MLIR. It is probably also good to carry the semantic to the MLIR layer to open the floor to future optimizations.