Francesc Verdugo

Results 195 comments of Francesc Verdugo
trafficstars

Generalizing the thickness of the ghost layer should be very easy. I think it is as easy as changing the hardcoded `1` by `ghost` in `start-=1` and `stop+=1` in this...

Hi @ZhangFengshun this looks promising, but we do not want to hard code 2 as the thickness of the ghost layer. Instead we want to use the value of variable...

> For parallel stencil computations (e.g. finite-difference methods), it would be nice to have partitioned multidimensional arrays, with support for arbitrary thickness "ghost" overlap regions (so that you could loop...

Perhaps I found the issue. See this comment in the code: https://github.com/fverdugo/GalerkinToolkit.jl/blob/eed905c038c9356653f8c51db40696340d14e500/src/geometry.jl#L3644 We cannot assume a single level of indirection in 3D. We need 2 levels of indirection. In general,...

Good news! ![Screenshot from 2024-04-09 16-37-24](https://github.com/fverdugo/GalerkinToolkit.jl/assets/48508177/7cc21883-9e13-47db-addd-5a0d1888a7c2)

@jfdev001 the correct implementation is in a function I created for debugging here: https://github.com/fverdugo/GalerkinToolkit.jl/blob/d6f91b3374421a56fc4981fca7963e3cfa9b2ced/src/geometry.jl#L3999 can you please move the fix to the `two_level_mesh` function and delete the new function? After...

```julia function label_boundary_faces!(mesh::PMesh;physical_name="boundary") D = num_dims(mesh) face_parts = face_partition(mesh,D-1) v = pfill(1,face_parts) assemble!(v) |> wait map(partition(mesh),partition(v),face_parts) do mymesh, myv, myfaces topo = topology(mymesh) face_to_cells = face_incidence(topo,d,D) local_to_owner_face = local_to_owner(myfaces) part...

Hi @jfdev001 Try to use the implementation in my last comment. (I introduced some changes just after our meeting). https://github.com/fverdugo/GalerkinToolkit.jl/issues/65#issuecomment-2058598819

Now I remember that we just defined a "dummy" face partition in the parallel two_level_mesh, right? I think that there is nothing wrong with my label_boundary_faces! funciton. The problem is...

In this file there are different ways of computing the "strength" graph of a sparse matrix, also for vector-valued problems: https://github.com/pyamg/pyamg/blob/e1fe54c93be1029c02ddcf84c2338a607b088703/pyamg/strength.py#L275