Roadmap: open-issue priority

Abierto
  • ¿Algo no está bien?
#109 0 comentarios 0 reacciones 0 asignados Ver en GitHub

Nadie ha tomado este issue todavía.

Evaluación

Dificultad
4/5
Tiempo estimado
3-5 días
Aptitud para principiantes
45/100
Tipo de issue
Documentación
Claridad
Bastante claro
Estado de actividad
Activo
Stack tecnológico
julia
Área
documentation

Línea de trabajo

Start with issue #109 and compare its priority table against the currently open issues listed there. Update the order, metadata, and rationale when issues are filed or closed; done means the body remains the single source of truth for what should be worked on next.

Escrito por el modelo de indexación a partir del texto del issue.

Descripción

The priority order of every open issue. Order: correctness first, then high-impact / low-effort, then high-impact / higher-effort, then low-impact, then investigations. The meta issues (#49, #51, #79, #80, #81, #82) aren't listed on their own: each one ranks with its highest child. When an issue is closed or filed, edit this body so it stays the single source of truth for "what next". Priority lives here rather than in labels.

Last triaged 2026-09-30 against v0.2.0. At that point there were no open bugs: #40 was closed as fixed by 9fc1b02.

Rank Issue Meta Impact / effort Why here
1 #108 re-enable Enzyme tests skipped on Julia 1.13 #82 high / S–M The dκ parameter-gradient invariant is currently checked on only half the CI matrix
2 #62 Robin BC #80 high / M Box-face scalar Robin is self-contained and unlocks wall-coupled systems
3 #61 active-cell mask #80 high / M Downstream correctness fix (spurious O₂ sink). Needed under every #69 outcome
4 #65 SciMLOperators jac_prototype adapter #81 high / M Concrete consumer, and the trait prerequisites landed in #89
5 #51 single-grid apply! throughput — high / L Largest CPU gap (≥2× unused bandwidth). The executor choice is a DESIGN decision
6 #70 anisotropic diffusion ∇·(D∇u) #80 high / L Baseline cardiac-EP physics. Stage 1 is S–M once the ScalingOp adjoint is fixed
7 #30 fold test/multigpu into the MDLA suite #79 med / S Cheap, and the upstream blockers are cleared
8 #43 GPU/packed-field AD parity #82 med / M Step 1 (CPU packed gradient + rule-hit assertion) can land now
9 #47 explicit distributed-backend selection #79 med / M Core-only, and a hard prerequisite for #46
10 #72 AMR regrid cadence + buffer zone #49 med / M The only AMR item that affects correctness (front on coarse cells)
11 #52 fold α·I + β·L into one stencil pass #51 med / M ~1.7× on explicit Euler, independent of threading
12 #83 thread the BlockForest path #51 med / M Leaves are disjoint. Reuses #51's executor, so it follows #51
13 #73 (Split)ODEFunction + integrator search #81 med / L Gated on #65
14 #63 stacked multi-field operator — med / L Generalizes a hand-fused downstream kernel. #107 is the scalar-diagonal precursor
15 #54 FluxDivergence primitive — med / L Most of the FV rollout has shipped. Four items remain
16 #46 ImplicitGlobalGrid halo backend #79 med / XL Blocked by #47. No current multi-node user
17 #78 batch the coarse–fine flux rewrite #49 low / M Only matters for GPU packed Diffusion with many coarse–fine faces
18 #41 pullback rules for parameter-free linear leaves #82 low / L Performance only: gradients are already exact via the tape
19 #94 source-centric transposed CSR for the fill adjoint #49 low / L Off the mul! hot path, and design questions are still open
20 #69 pick the non-box geometry seam #80 med / L An investigation. It gates only #70 stages 2–3, not #61

—
🤖 Ranked by Claude, who promises the order wasn't picked by how fun each one sounds.
claude-opus-5-5[1m] (high) · claude-code 2.1.285 · unreviewed

Lenguaje dominante
Julia
Estrellas
3
Forks
1
Merge medio
4 d 19 h
PR fusionados (30 d)
5

Preparar el entorno

Este proyecto no incluye contenedor de desarrollo, Dockerfile ni guía de contribución, así que la configuración corre por tu cuenta: empieza por su README y consulta nuestra guía para la primera contribución para los pasos generales.

Primeros pasos

  1. Lee el issue completo y luego la guía de contribución del proyecto.
  2. Comenta en el issue que vas a ocuparte — evita que dos personas hagan lo mismo.
  3. Haz un fork del repositorio y trabaja en una rama.
  4. Abre un pull request que haga referencia al número del issue.

Más de RallypointOne/MatrixFreeOperators.jl

Todos los issues de RallypointOne/MatrixFreeOperators.jl

Issues similares

Más issues de Julia

Recibe los nuevos issues en tu correo

Un resumen breve de issues de GitHub para principiantes.