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

Assorted fixes while supporting PETSc #700

Merged
merged 10 commits into from
Jan 27, 2020

Conversation

guyer
Copy link
Member

@guyer guyer commented Jan 27, 2020

No description provided.

- Terminate sweeps if they go too long

  this example could run forever if the solver doesn't converge

- clean up messy doctest structure

- Remove litter from tests

  Tests should not pollute user directories with temp files

- Import math from numerix

  numerix routines are no longer imported globally. Horrible construction of
  chemotaxis tests meant we didn't catch these two.

- Change chemotaxis examples into proper doctests

- Disable archaic tests

  - disable examples/cahnHilliard/mesh2DCoupled tests
  - disable `levelSet/electroChem` tests

  These tests have [all kinds of problems](usnistgov#682)

- Fix formatting of results
Fix timetests formatting
- Some combination of linux and Python 3.8 fails to close /dev/null
  on other processors.
- Writing on only one file deadlocks in
  `fipy.meshes.representations.gridRepresentation._GridRepresentation._test`:
  https://app.circleci.com/jobs/github/usnistgov/fipy/1226
  https://app.circleci.com/jobs/github/usnistgov/fipy/1225
  https://app.circleci.com/jobs/github/usnistgov/fipy/1217
  https://travis-ci.org/usnistgov/fipy/jobs/627999300
- Stipulate binary

  Avoid doubling memory with `dumps()`,
  but prevent `/dev/null` closure problem
- Limit Gmsh to < 4.0
  Gmsh 4.0 doesn't partition correctly
  Addresses usnistgov#697
- Export ghosts with Gmsh 4.x
  [Ghosts are broken](https://gitlab.onelab.info/gmsh/gmsh/issues/733),
  but export them anyway.
- Change refinement metric
- Limit refinements
  Gmsh refinement seems to stall out
- Don't partition tiny meshes
- Tolerate mangled Gmsh version string

  Assume Gmsh 3.0 if we can't parse the useless nonsense
  that Gmsh stuffs into stderr
- Ensure GmshMesh uses correct communicator

  More than one overlap will "downgrade" to a serial communicator
  because Gmsh doesn't give us more than one layer of ghosts
@guyer guyer force-pushed the Assorted_fixes_while_supporting_PETSc branch from 5946f8c to f71f15b Compare January 27, 2020 18:20
@guyer guyer requested a review from wd15 January 27, 2020 19:43
@guyer guyer merged commit 115e729 into usnistgov:master Jan 27, 2020
@guyer guyer deleted the Assorted_fixes_while_supporting_PETSc branch January 27, 2020 20:47
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Projects
None yet
Development

Successfully merging this pull request may close these issues.

2 participants