July 2026 — EZoni/WarpX: Implemented a critical stability fix for radial geometries in implicit residual evaluations by relocating inverse-volume scaling for charge density (rho) into the particle deposition path. This restores pre-6589 behavior for both explicit and implicit paths, eliminating divergence in theta-implicit hybrid-PIC schemes in RZ coordinates. The change is anchored to commit 478e7d615c4c514f481eead6f1d259d437ca1392 and aligns deposition scaling with the explicit path. Regression validation shows the RZ Ohm solverImplicit PICMI regression test passes with the fix, and there are no checksum changes for existing tests. Co-authored by S. Eric Clark and Claude Fable 5.

June 2026 focused on stabilizing external-field divergence cleaning in WarpX and expanding support for externally-loaded particle B-fields. Delivered a robust projection-based divergence cleaner with opt-in control across electrostatic and magnetostatic solver modes, including per-component cleaning for multi-map particle fields and regression/efficacy tests. Fixed a critical crash when using LoadAppliedField with initial divergence cleaning by introducing guards for external B fields and extending cleaning to particle-field maps; added component-level cleaning to prevent aborts and ensure correct behavior across grid and particle paths. Updated PICMI and documentation to reflect applied-field cleaning and its opt-in usage. Added end-to-end tests (Examples/Tests/projection_div_cleaner) validating efficacy and regression scenarios. Validated on CUDA and OpenMP builds across 2D/3D/RZ, confirming stability and correctness. Business impact includes expanded capability for externally-loaded field configurations without crashes, improving reliability for production simulations and reducing maintenance risk.

February 2026: Focused on stability and correctness of the RZ vector Laplacian in WarpX for cylindrical coordinates. Implemented origin-specific handling to fix division-by-zero and ensure proper application of hyper-resistivity, addressing a critical stability bug in the RZ evaluation. This work improves simulation reliability and accuracy at the coordinate origin, reducing numerical artifacts and increasing confidence in results.