-
Notifications
You must be signed in to change notification settings - Fork 185
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
SSAMG interpolation #1083
SSAMG interpolation #1083
Conversation
…le test problem doing up to 4 levels of ssamg. Doing full multilevel all the way down still yields bad convergence... not sure why.
…he correct code path for sstruct matmult
…One is to pass the entire boundary, including the converted struct connections to BoomerAMG interp. The other is to just add a diagonal to A_u. Not sure which will be better, so I am maintaining both for now with a macro option to choose which to use.
…atrix on some procs.
…rent ssamg interp strategies a runtime option.
…ed a place where I was dropping entries in a bad way.
…e a hacky way of doing this implemented, but need to figure out a good way of doing it.
…ata structure where possible.
…struct conversion routines.
@rfalgout @victorapm , there is still a lot of rough/unfinished code in here for the unstructured interpolation pieces, but we had discussed getting this into recmat sooner rather than later, so I think this is an acceptable state to merge. I've done some testing to make sure I'm not changing behavior when unstructured interpolation is turned off, that is, you should get the same results as the base recmat branch if you don't use the new unstructured interpolation stuff. |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Thanks, Wayne!
…nal cleanup and removing some unused code.
…ght, although convergence is not as good as I expect, so I need to check correctness on coarser levels.
…eateFromIndices. It seems this had some unintended consequences that broke some regressions.
Add unstructured interpolation to SSAMG.