Skip to content

Conversation

cianwilson
Copy link
Contributor

Just a draft at the moment to keep track/refresh my memory of the changes.

cianwilson and others added 15 commits July 7, 2021 01:00
Fails because the preconditioner matrix is assumed to exist, when it doesn't.

Also depends on an options path in Full_Projection that isn't compatible with being called from solenoidal projection instead of
momentum equation.
…e right.

Also some fixes to two tests that failed the updated logic because they were themselves inconsistent.
…xistence) preconditioner_matrix.

Also allow correct velocity to take in a option path as it also has assumed locations for velocity solver options.

Add a test that appears to pass that uses Schur complement solenoidal projection.  Need to look at this more.
…e necessary but unrelated to other changes in this branch.
…ions.

diag and lump appear to be best with tolerances increased for no and mass due to poor convergence
…c numbering so that we get a second order halo when it's required.
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.

1 participant