Adds a broadcast of the namelist value 'urbantvmapalgo' #2534
Conversation
Fix unit tests for nstep starting at 1 instead of 0
Add sowing window input files Previously, one could run crops with either (a) sowing windows defined by the hemisphere-specific start and end dates on the paramfile or (b) prescribed sowing dates specified by input file stream_fldFileName_sdate. This PR replaces the latter with two new input files, stream_fldFileName_swindow_start and stream_fldFileName_swindow_end.
…RITY Fix RXCROPMATURITY test in zerothtstep branch
...and other mods that I'm preserving from closed PR ESCOMP#2019, such as - changes to long_names and - treating avgflag as a tape (not field) trait for 'I' and 'L' tapes
… it in so it is consistent across all ranks when not using the default value ('nn').
|
@briandobbins I'm rebasing this to come in on b4b-dev, so it can come in quicker. |
ekluzek
added
enhancement
|
Is this ready to merge? |
Merge tmp-241219 branch to master Includes three tmp-241219 tags: tmp-241219.n01.ctsm5.3.016 Merge b4b-dev: nfix_method options Houlton (default), Bytnerowicz (option) tmp-241219.n02.ctsm5.3.016 FATES hydro test update tmp-241219.n03.ctsm5.3.016 Bug fix for izumi nag tests to pass (b4b unless using Bytnerowicz) Fixes ESCOMP#2924 Fix problem with izumi nag tests Fixes ESCOMP#2878 Remove fates_allom_smode shell_command update in FatesColdHydro testmod Fixes ESCOMP#2869 Update temperature cost function for symbiotic nfix in FUN Changes answers as documented in the ChangeLog. slevis resolved conflicts: doc/ChangeLog doc/ChangeSum
…_bounds Change history time to be equal to the middle of the time bounds
slevis resolved conflicts: .gitmodules
ctsm5.3.019: Stop running 0th time step
Includes merge conflicts in src/cpl/lilac/lnd_comp_esmf.F90 due to PR ESCOMP#2912.
These were resolved with PR ESCOMP#2912.
Merge b4b-dev
Merge tag ctsm5.3.020 into b4b-dev
…rrently used at ultra-high res (largely to get around an issue we had with online regridding here, which can be another approach taken later)
|
Well, I'm embarrassed to say I've only now gotten back to this (from last May!), but added one more change - enabling the 'redist' method for the 'urbantvmapalgo' variable in bld/namelist_files/namelist_definition_ctsm.xml. Happy to hop on a quick call if anyone has questions or needs additional changes, but it looks like since the rebase wasn't needed, this is good to go? Apologies for the ridiculously long delay! |
|
Thanks for the update @briandobbins. The problem you saw with redist is a more general problem that's important to correct throughout the code (#1912). So I'll do that on top of what you've done here and bring it in. That fix is something we really need to have corrected by the release, since it means you can't use many of these options for a list of other fields either. We are all busy. We have lots of PR's that we can't bring in for many months as well... |
Adds a broadcast of the namelist value (urbantvmapalgo) after reading it in so it is consistent across all ranks when not using the default value ('nn').
Description of changes
Simple one-line change; a namelist value wasn't being shared with ranks other than the main one reading the namelist. This lead to hangs when trying a different algorithm.
Specific notes
Contributors other than yourself, if any:
CTSM Issues Fixed (include github issue #):
Fixes #2533 (Done)
Fix #1912 (TODO)
Are answers expected to change (and if so in what way)?
No
Any User Interface Changes (namelist or namelist defaults changes)?
No
Does this create a need to change or add documentation? Did you do so?
No
Testing performed, if any:
Limited testing, but it's a one-line logical change. If more is needed, let me know. I just did a simple test with and without the change, looking at the values on all ranks.