Skip to content

Conversation

@c-randall
Copy link
Contributor

@c-randall c-randall commented Mar 18, 2025

There were a couple typos in the preconditioner wrappers for IDA:

  1. aux_data.r_vec -> aux_data.rvec_tmp to be consistent with IDA_data
  2. The prec_solvefn.evaluate() call was missing yp_tmp
  3. Documentation referred to spbcg as a linear solver, but code expected spbcgs

I also added test_iterative.py which uses the Robertson problem to test all three iterative solvers. I had to skip these tests for extended-precision builds because I used np.linalg.solve which cannot handle longdouble.

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