Update externals based on ESCOMP/CESM tag 'cesm3_0_beta03' #76
Conversation
There should be minimal differences between the EarthWorksOrg and ESCOMP versions; EarthWorksOrg/CTSM just has fsurdat files for mpasa30 and mpasa7p5 that upstream doesn't have.
This time it also includes submodules that are commented out in .gitmodules
NOTE: EWOrg/CAM is updated to cam6_4_033 instead of cam6_4_032 since EarthWorks already is using the files added in _033.
Sync select files with ESCOMP/CESM
02e8599 Bump to 0.9.2 8be4e76 Merge pull request EarthWorksOrg#64 from ESMCI/fix/newremote e850ebe fix one get another ee5deca avoid spurious warnings d485f0a add tmate for debugging 024d2e4 try updating poetry 59015d4 fix issues with sparse config b85ce7b Bump to 0.9.0 2e8d985 Merge pull request EarthWorksOrg#62 from ESMCI/fix/nonexistanttag dada084 more changes to add status d826532 better error checks 7825f29 update to provide warning 532c244 Bump to 0.8.5 5d2093a Merge pull request EarthWorksOrg#60 from ESMCI/fix/worktree 4749ece remove debug print statement 56b2024 fix for git worktree issue git-subtree-dir: .lib/git-fleximod git-subtree-split: 02e859950fd79064ad5b64565ce1073bea5d3ffa
Merge commit '0be1d1f4c84b06ce541d4399fd3e9d436ac63dec' into update/cesm3_0_beta03 Command: git subtree pull --squash -P .lib/git-fleximod https://github.com/ESMCI/git-fleximod v0.9.2
gdicker1
added
external
|
Thanks @gdicker1 ! Have you tested this with the nvhpc compiler (on CPU or GPU)? I'm having difficulties building the cesm3_0_beta03 tag with nvphc. I'm seeing a "rc=2" error while building the atm library. I see this with version 24.3 and 24.9. I'm just wondering if you've seen something similar with the EarthWorks code? |
|
@sherimickelson I haven't, but I was just about to kick off the testing for this PR. I'll watch out for this and reply with what I see. I don't recall seeing this for the cesm3_0_beta02 update. |
|
@sherimickelson, I didn't have any problems building EarthWorks tests (using nvhpc/24.3 stack on Derecho). I could try to take a peak if you send me a path to your testcase and/or sandbox. Also I'm looking into namelist differences and differences in the baseline vs test results before I'm sure of this PR. |
|
Thanks for double-checking, @gdicker1. I wonder if something in my environment is causing this. |
@sherimickelson yes, I'll get right on modifying EarthWorksOrg/ccs_configs_cesm. I'm thinking it will be a separate PR so I can also add some restart tests. |
Bring in changes to develop that were part of EarthWorksOrg/EarthWorks PR#75. Ensure externals are at the right commits.
Use tags from merged Sub-PRs and update URLs in .gitmodules
|
Results from testing with results from #75 as baseline:
As I said, the changes seem reasonable for how different the namelists are and changes around CTSM files. I plan to push the merged branches soon. |
This PR catches up to the current development in ESCOMP/CESM. This includes updating some externals (a.k.a. submodules) through PRs, updating git-fleximod, syncing the cime_config folder, and other miscellaneous changes to match CESM.