Edoardo Zoni

Results 49 issues of Edoardo Zoni

There are situations where it is necessary to fill the guard cells when inverse FFTs of fields and/or currents are computed. For example, we want to fill the guard cells...

component: spectral
component: parallelization

To-do: - [x] Improve post-processing analysis - [x] Test and improve based on feedback (Cc @oshapoval) To be decided whether a Python script or a Jupyter notebook is preferred. This...

component: post-processing

Fix #3232. Filtering will also need to be moved to the right locations, though (depends on another open PR). To-do: merge #3302 first and rebase afterwards. Currently not supporting (to...

bug
component: spectral
bug: affects latest release

To-do: - [ ] Update documentation - [ ] Add CI tests (Langmuir, consistency checks) - [ ] Add rho quadratic case?

enhancement
component: spectral

Could something like this be useful? A small WarpX function, `WarpX::AllocInitMultiFab`, that allows us to allocate a MultiFab and initialize at the same time its values using `setVal`. For development...

To-do: - [x] Split `ApplyFilterandSumBoundaryJ` into new functions `ApplyFilterJ` and `SumBoundaryJ` - [x] Use new function(s) `ApplyFilterJ` in MR part of `AddCurrentFromFineLevelandSumBoundary`: - [x] fix the periodicity used to call...

component: parallelization

Follow up on #2679. Needs #2748, #2839, #2937, #2958. To-do: - [x] Deposition of rho at half time step - [x] Equations w/o div cleaning and w/o time averaging -...

component: spectral

With the Vay deposition, apply filtering on D, after it is deposited, and sum the guard cells of J, after it is computed from D. Before this PR, both filtering...

bug
component: spectral
bug: affects latest release

Having calls such as `rho[lev]->setVal(0.0, 0, WarpX::ncomps, rho[lev]->nGrowVect());` is potentially buggy if `rho[lev]->nComp()` is different than `WarpX::ncomps`, which is possible (depending on the algorithms used): https://github.com/ECP-WarpX/WarpX/blob/af5b5764f8a1657a3da901447a3ddf0c358c0b7c/Source/WarpX.cpp#L1872-L1877 It seems to me...

bug
cleaning

Here's a few points that need to be addressed in order to optimize the synchronization of nodal points in WarpX: - [x] Merge synchronization of nodal points into `FillBoundary`: -...

Performance optimization