Diagnostic updates cause WY failures in branch with parallel fixes

[jeff-cohere]

Some recent updates have cause failures in the WY demos. @bishtgautam submitted PR #221, which applies several fixes to restore the dycore's parallel capabilities. Then PR #220 was merged, adding an interface for diagnostic fields. In a separate branch that contains the changes for #221 rebased to master post #220, the WY demos stopped working. Gautam's assessment:

The transient WY tests are failing. The residual for some cells is garbage at line 1167 of tdywy.c in the rebased branch. But, I can’t figure out where I made the mistake.

We're tracking that issue here so we can go ahead and merge #221.

We should probably add more parallel tests to our regression suite to make sure we don't damage the parallel runs in our future work.

[jeff-cohere]

Hey @bishtgautam , can you show me how to reproduce this issue? I'm wondering if it still occurs, and/or how much we should focus on fixing the WY tests when we know that discretization is going to be reworked.

[nocollier]

I thought we had discussed just removing it as needed? We can always reconstruct it later if it is needed.

…

On Thu, May 5, 2022 at 2:40 PM Jeffrey N. Johnson ***@***.***> wrote: Hey @bishtgautam <https://github.com/bishtgautam> , can you show me how to reproduce this issue? I'm wondering if it still occurs, and/or how much we should focus on fixing the WY tests when we know that discretization is going to be reworked. — Reply to this email directly, view it on GitHub <#224 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AAKFCB5BMUWTVVSCXUO75XLVIQI2XANCNFSM5NSGZFXA> . You are receiving this because you are subscribed to this thread.Message ID: ***@***.***>

[jeff-cohere]

Yeah, maybe we should just "comment out" these regression tests for now. Thanks, @nocollier .