LonelyCat124

Ok - I'm going to work through the documentation to work out how things fit together tomorrow and go from there.

Ok - I realised I need another thing that doesn't yet exist - which is OpenMP's single/master directives, as well as `taskwait`. From what I can work out/understand it looks...

Once the single implementation is complete, I'll probably do a `omp master` implementation next (which I think is best to inherit from all the OMP Single code as it is...

Ok - from discussion with Rupert, I will implement `taskwait` next as 2 PRs: 1. The `OMPTaskwaitDirective` - this is a standalone piece of work which can enable manual creation...

Ok, so now that I've implemented the NemoLite2D example using taskwait directives, and the taskloop/single/parallel transformation, I've thought a bit on how I think the taskwait transformation can work. I...

I will probably work on the `OMPTaskwaitTrans` one day this week (probably tomorrow) unless anyone has specific concerns about my implementation plan. I do still have a few questions about...

Hmm ok - the difficulty is then for general PSyclone things that don't map so cleanly onto an array, how to work out the dependencies - I will think.

No - I'll take a look, Rupert had only told me about the `get_forward_dependence` etc. functions available on Nodes.

One thing that doesn't quite seem to fit compared with other transformations is the `_directive` class member. Since this can make many directives am I ok to just not create...

Next on the list is explicit tasks. The basic transformation here is pretty straightforward: `#pragma omp task`, plus additional clauses. The clauses I'm going to implement initially are just the...