Ali Ramadhan

Rename `@wrapper` to `@impero`?

3

Wrapper seems like a pretty generic name for a macro so I was wondering if renaming it to something like `@impero` would make sense?

Allow `FieldTimeSeries` to pass keyword arguments to `jldopen`

11

This PR takes some code I had to pass kwargs to `jldopen` through `FieldTimeSeries`. But from the dicussion in issue #3680 it sounds like maybe a more future-proof way to...

Resolves #3735

Is `HydrostaticFreeSurfaceModel` supposed to support a rigid lid?

6

I noticed a file called `rigid_lid.jl` exists (https://github.com/CliMA/Oceananigans.jl/blob/b3be950ef6a1f2139c2f65e083f20f1f7958ea55/src/Models/HydrostaticFreeSurfaceModels/rigid_lid.jl) so I was curious whether `HydrostaticFreeSurfaceModel` can run with a free surface, presumably by passing `free_surface = nothing`. It does error (see...

Support passing keyword arguments to `jldopen` when using `FieldTimeSeries`

3

I am interested in reading JLD2 files in parallel using multiple threads. This is actually now supported by JLD2.jl after https://github.com/JuliaIO/JLD2.jl/pull/477 provided you pass the `parallel_read = true` kwarg to...

`PartialCellBottom` needs to be adapted correctly for the GPU when using a `HydrostaticFreeSurfaceModel`

4

I know `PartialCellBottom` is unexported and therefore experimental but just opening an issue that it doesn't currently work on GPUs with `HydrostaticFreeSurfaceModel`. I believe it's an Adapt issue. It works...

Update Project.toml to use Oceananigans.jl v0.92

5

Hope this PR is okay. I'm hoping to keep OceanBioME.jl compatible with the latest version of Oceananigans.jl. Just opening this PR to see if tests pass with Oceananigans.jl v0.92. Don't...

Illegal memory access on CPU and GPU with `LagrangianParticles` + `HydrostaticFreeSurfaceModel` + `LatitudeLongitudeGrid`

1

I'm trying to add some particles to a hydrostatic model on a lat-lon grid, but ran into some CUDA memory issues. After reducing down to a MWE I noticed that...

Probably as a result of PR #3847, on the `main` branch precompiation fails due to some method overwriting. This may be harmless as all tests passed anyways, e.g. see the...

Error using `ScalarDiffusivity` where the viscosity or diffusivity are functions with parameters

6

I believe the below MWE should work according to the `ScalarDiffusivity` docstring, but from the error it seems to be expecting `funky_diffusion(x, y, z, t)` instead of `funky_diffusion(x, y, z,...