Oakley Brunt

Regarding `invoke_rdouble_x_innerproduct_x`, the current implementation in the PSy-KAlite code (given in this issue) calls two loops to achieve the aim. ``` th_idx = omp_get_thread_num()+1 ! calculate inner product of each...

Hey @arporter, I included those lines for context, really. Though I guess it isn't hard for anyone to work out what `th_idx` is. Okay, so will I need to cast...

Okay, I understand more... but also have more questions! Questions: What do you mean when you say the reduction reference "doesn't yet use the PSyIR properly"? Why is REPRODUCIBLE_REDUCTIONS set...

> the code that generates the reduction (using f2pygen) doesn't know about precision Ahhh this is what I was missing, thank you!! > The old way uses functionality from the...

> Thanks for this Oakley. I think you've mostly got the top-level rules sorted but you're missing a section on the rules for the kernel subroutine itself - see inline....

## Proposed changes: ### Allowed metadata types and access 1. field (READ, WRITE, READWRITE) 2. scalar (READ only) ### LFRic Builtins metadata vs User defined kernels metadata LFRic builtins use...

@arporter Can you have a look at the failing build test please? It is repeatedly tripping over on "Run examples" complaining about `ACCEnterData` not being able to find any data...

@arporter Did I hear right this morning that it I should ignore stub generation for now?

@arporter, I can't get Codecov to recognize that I am checking the "uncovered" line (see in `src/psyclone/tests/domain/lfric/dofkern_test.py::test_function_space_bare_undf`). I'm stuck and asking for some help please.

>On master, there is no check that this isn't a Builtin - we always added a field argument. This may be because this method is never called for a Builtin...