Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

[WIP] Refactor PDE operator for multigrid #401

Draft
wants to merge 23 commits into
base: devel
Choose a base branch
from

Conversation

landinjm
Copy link
Contributor

@landinjm landinjm commented Jan 14, 2025

Do not review. Posted to collaborate with @fractalsbyx.

Refactoring matrix free operator (in computeRHS and etc.) to support multigrid and GPU.

Summary of issues that are closed:
#394,

Issues in progress
#393
#398
#386
#376
#351
#346
#323
#313
#311
#301
#294
#291
#290
#285
#276
#273
#253
#218
#210
#174
#90
#86
#65

@landinjm
Copy link
Contributor Author

We can just integrate the work in #379 as we rebuild computeRHS, computeLHS & postprocess.

Also #398 can be resolved.

@landinjm landinjm force-pushed the refactor_PDE_operator branch from 90eaee1 to f0f68f6 Compare January 15, 2025 22:23
…ect should be its own class like we do for conditional ostreams.

There are still somes cases where std::cout should be replaced with the pout_base of an assertion
@landinjm
Copy link
Contributor Author

VariableValueContainer can be deleted too

@landinjm
Copy link
Contributor Author

landinjm commented Jan 16, 2025

Checklist of features left to integrate

  • AMR
  • Newton update & nonlinear solves
  • Multiple equations
  • User-facing implementations
  • Initial conditions
  • Boundary conditions
  • Grain growth features
  • User-controlled inputs
  • File read-in
  • Checkpoints
  • Transient problems
  • Nucleation

landinjm and others added 9 commits January 16, 2025 12:05
Broke solveIncrement into four smaller sections:
1. Aux vars that are not nonlinear (sequential)
2. Explicit vars (concurrent)
3. Linear solves (sequential)
4. Nonlinear solve (concurrent)

TODO/NOTES:
I broke the skip_time_dependent thing.
Changes not yet propogated.
Need some documentation.
Prefer to pass var_index into solver rather than use member variable currentFieldIndex.
Maybe removed need for member variable currentIncrement.
computeExplicitRHS-like functions are commented with "HERE" to be replaced
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

2 participants