SUNStepper basics based on MRIStepInnerStepper#463
Conversation
|
@drreynolds @gardner48 I have made the changes we discussed on Tuesday. Let me know what you think. |
drreynolds
left a comment
There was a problem hiding this comment.
I like how clean this is now. I noticed a few items for comment (see below).
a030d57 to
c04f24a
Compare
Steven-Roberts
left a comment
There was a problem hiding this comment.
Initial pass at latest changes
|
After some more thought and discussion with David on IDA and passing around I'd like to propose the following to make SunStepper self starting (all you need is
where there's a function which can be specified by users (we could have a default that uses |
|
I actually need the OneStep function in the adjoint work, so I have added it back into here. |
|
It looks like the only remaining issue with the CI is that the documentation updates have some label errors: |
|
Yeah, wasn't sure how to best handle that. Those are references to parts of the docs that will be added in the operator splitting PR (which is merged after this). I can temporarily comment out or remove the references out to get the CI passing. |
I forget, but is the "merge" plan to approve/merge this PR before #572? If that's the case, then perhaps we should move these bits of documentation into #572 so that both "squashed" and merged PRs can pass all CI tests? |
gardner48
left a comment
There was a problem hiding this comment.
One minor change (missing process error call) that I'll apply and approve in a sec.
Add SUNStepper class for evolving IVPs --------- Co-authored-by: Steven Roberts <roberts115@llnl.gov> Co-authored-by: David Gardner <gardner48@llnl.gov> Co-authored-by: Daniel R. Reynolds <reynolds@smu.edu>
I am opening this PR as a draft to gather feedback. This renames/moves
MRIStepInnerSteppertoSUNStepperso that @Steven-Roberts and I can extend it for other purposes.