Repository navigation

Eulerian SUPG advection-diffusion with symbolic BDF and Adams-Moulton orders (#657 follow-on) - #673

Closed

lmoresi wants to merge 14 commits into

developmentfrom

feature/eulerian-supg-transport

Closed

Eulerian SUPG advection-diffusion with symbolic BDF and Adams-Moulton orders (#657 follow-on)#673
lmoresi wants to merge 14 commits into
developmentfrom
feature/eulerian-supg-transport

Commits on Sep 3, 2026
- Expose the BDF and Adams-Moulton coefficient symbols on the DDt managers
  
  lmoresi
  and
  claude
  committed
  f7d1a21
- Pack and index auxiliary fields by DM field, not by position in mesh.vars
  
  lmoresi
  and
  claude
  committed
  e84dea9
- Add the RotatingGaussian transport oracle; fix the integral-norm error for scalar variables
  
  lmoresi
  and
  claude
  committed
  ccfc23b
- Extract the per-element timestep estimate shared by the advection-diffusion solvers
  
  lmoresi
  and
  claude
  committed
  6d8e2b7
- Skip the mesh-owned multigrid pickup for a solver that owns its preconditioner
  
  lmoresi
  and
  claude
  committed
  9c8a125
- Eulerian advection-diffusion with SUPG: BDF and Adams-Moulton orders from the symbolic history
  
  lmoresi
  and
  claude
  committed
  b686530
- Design note for the Eulerian SUPG solver; BDF2 becomes the default from the integrator study
  
  lmoresi
  and
  claude
  committed
  451efe3
- Integrator study, res 64: BDF3 grows slowly on pure advection at any Courant number
  
  lmoresi
  and
  claude
  committed
  f076781
- AdvDiffusionSUPG takes the semi-Lagrangian solver's interface: a drop-in replacement
  
  lmoresi
  and
  claude
  committed
  2b9941c
- Drop the integrator argument: order and theta already reach every safe scheme
  
  lmoresi
  and
  claude
  committed
  252165e
- An accuracy-based timestep for the Eulerian solver; credit NengLu in the module and note
  
  lmoresi
  and
  claude
  committed
  417f7b8
Commits on Sep 4, 2026

Navigation Menu

Repository navigation

Eulerian SUPG advection-diffusion with symbolic BDF and Adams-Moulton orders (#657 follow-on) - #673

Eulerian SUPG advection-diffusion with symbolic BDF and Adams-Moulton orders (#657 follow-on)#673
lmoresi wants to merge 14 commits into
developmentfrom
feature/eulerian-supg-transport

Commits on Sep 3, 2026

Expose the BDF and Adams-Moulton coefficient symbols on the DDt managers

Pack and index auxiliary fields by DM field, not by position in mesh.vars

Add the RotatingGaussian transport oracle; fix the integral-norm error for scalar variables

Extract the per-element timestep estimate shared by the advection-diffusion solvers

Skip the mesh-owned multigrid pickup for a solver that owns its preconditioner

Eulerian advection-diffusion with SUPG: BDF and Adams-Moulton orders from the symbolic history

Design note for the Eulerian SUPG solver; BDF2 becomes the default from the integrator study

Integrator study, res 64: BDF3 grows slowly on pure advection at any Courant number

AdvDiffusionSUPG takes the semi-Lagrangian solver's interface: a drop-in replacement

Drop the integrator argument: order and theta already reach every safe scheme

An accuracy-based timestep for the Eulerian solver; credit NengLu in the module and note

Commits on Sep 4, 2026

Match the Krylov tolerance to the SNES tolerance, and make preconditioner="fmg" a real switch on the SUPG solver

Design note: the 512^2 rows at matched tolerance

Let theta be set after construction, as the semi-Lagrangian solver allows