Jago Strong-Wright

@tomchor, the "fix" I made was just a hack to keep the previous value in the next unused halo point: https://github.com/jagoosw/OpenBoundaries.jl/blob/e2d041b0418dc015ea6a85a6d408d0c47d483b8e/src/perturbation_advection.jl#L94-L95 But @glwagner has made changes to the kernel launching...

Ah so I've realised this isn't the fix we needed, and I was just hiding it from myself in the profile because I replaced the function by writing it in...

I'll stream line those two line before I merge

> I think this has been superseded by #3834. @jagoosw let us know if it is the case or not. Yeah definitely superseded!

That all makes sense, I hadn't considered how big a change that was. This is just a convenience constructor for that alternative so probably isn't the best solution (and I...

Since you're using the non-hydrostatic model its here: https://github.com/CliMA/Oceananigans.jl/blob/main/src/Models/NonhydrostaticModels/compute_nonhydrostatic_tendencies.jl and the functions called from compute_G... are here: https://github.com/CliMA/Oceananigans.jl/blob/main/src/Models/NonhydrostaticModels/nonhydrostatic_tendency_kernel_functions.jl

The light integration doesn't occur in the compute_Gc kernel but during the update_state step

Perhaps you could write up how todo it and convert this roll a discussion?

This preprint might be of general interest https://arxiv.org/pdf/2502.14148 And this discusses the performance of the physics: https://doi.org/10.1029/2024MS004465 - as it eludes to there hasn't been much focus on optimising Oceanaingans...

Hopefully I've written this somewhere but I (and Oceananigans) try to follow [YASGuide](https://github.com/jrevels/YASGuide)