Repository navigation
Bring CMEPS share code into this repository, and organize share code into subdirectories - #88
Conversation
…tory and finish refactoring dead nuopc components
…l of shr_sys_abort with ESMF_FAILURE tracing Conflicts: src/share/nuopc/glc_elevclass_mod.F90
Update CIME to ESMCI cime5.8.2-2 Squash merge of jgfouca/branch-for-to-acme-2019-05-06 Features * Nuopc update * nldas_forcing_v2: This is a regional, 0.125 degree data set over the U.S., for use in forcing CTSM. * Change DATM_CLMNCEP_YR_ALIGN for present-day compsets: Make DATM_CLMNCEP_YR_ALIGN match DATM_CLMNCEP_YR_START. This will make the model year align with the forcing year. For example, if RUN_STARTDATE is set to 2000-01-01, then the model will use year-2000 atmospheric forcing data in the first year, etc. non-BFB for any present-day compsets using datm (mostly I-cases and G-cases). Bug fix: * Fixes to NAG flags in scripts/Tools/Makefile [non-BFB] (datm)
* esmci_remote_for_split/master: (1657 commits) Fix typo Small upgrade to e3sm_cime_mgmt send run two TestStatus.log to case1 log fix mpi-serial dont require optional arguments Use RawConfigParser to load saved submit options Reset Anvil changes Fix incorrect var refference Upate env_mach_specific schema and CIME docs Add more information in comment Also allow machine-mpirun level customization of the run_misc_suffix Remove lines referring to rcp compsets with nldas2 forcing Move custom run_exe into mpirun spec Add NLDAS2 datm forcing option Allow machine level customization of the run_exe increase wallclock time for Vnuopc_Ln9.f19_g17.X test check existence of scalar fields before querying their value replace tabs with spaces in buildnml minor fix in mediator code to merge it with FV3 branch Generalize DATM_CLMNCEP_YR_ALIGN for present-day compsets ...
…#3547) Update CIME to ESMCI cime5.8.20 Squash merge of jgfouca/branch-for-to-acme-2020-04-14 This is the first cime update in half a year, lots of changes. Features: * allow for specification of vertical grid in ATM_GRID * if no env vars, print comment to assist user in env_mach_specific.xml * case.setup: add --keep option for clean/reset actions to keep local changes to generated files * case.setup: refresh build Macros when doing a reset/clean * cprnc: cime will build a copy in sharelibs if no global version available * check_input_data: add ftp timeout, make sure failover to svn is working * added support for new OMIP compsets in POP as well as better handling of the different versions of JRA forcings * Pio2 updated to latest master * reorganize mct and nuopc data, xcpl and stub components based on coupler * new auto-triaging feature for tests, will report potentially broken commits to TestStatus.log * coupler: Allow writing T-forcing cplhist files even when running with SGLC * Add support for UFS model! Not really relevant to E3SM, but is a major development in CIME * archiver: Extend archiving support to accept a regex pattern * coupler: Update for compatibility with latest CMEPS Bug fixes: * check_input_data: fix issue with pop input files that should not be found * fix default for PIO_REARRANGER * docn: Small fix to allow scam to run with aquaplanet * fix issue with resubmit on systems that use ssh * fix misleading output in CaseStatus * Update description for FORCE_BUILD_SMP * archiver: pop dd history files were not identified by get_extension * NOTE: MPAS history files are now being processed as they should have been all along * Improve scripts_regression_tests test cleanup * dlnd: Some gnu tests were giving an error of different character lengths in dlnd. * coupler: seq flux should be initialized even if for some reason the namelist isn't read * env_batch.xml: we can now handle the case where multiple queues have the same name * ERI: fix tests that were less than a day, or with START_TOD nonzero * test hist: Exclude .ref1 and .ref2 from TestStatus files * coupler: fix issue with e3sm field Fixes: #2324 [BFB]
* esmci_remote_for_split/master: (2458 commits) fix cime_model to stdout issue extend allowed string length Add a comment in the Makefile improve format fix formatting revert path change add make check, fix tests fix format issues change type to double for derived vars Updated filter on calculating normalized rms. Now, a variable must be >0, not nearly 0, to be normalized. Also, variables that cannot be normalized generate a huge rms. Added check that the normalizing denominator is >0 when calculating a relative rms Fix typo Update for cime5.8.23 Missing credit in shr_flux_atmOcn fix more pylint issues, unrelated to pr fix pylint issue fix atm level spec so that it does not match 1x1_brazil Add missing ROF2OCN grid for r05_TO_gx3v7 fix for issue 3452 erroneous log error due to FV3 additions to Makefile, and bug fix for FV3 compset definition. Update NCPL values ...
Based on discussion in ESCOMP/CAM#404
Add module to read ozone_coupling_nl This will work together with ESCOMP/CMEPS#212. However, this can safely be brought in prior to that PR, because this code is not yet invoked from anywhere.
* moved files into ufs * implemented different interface for ufs flux_atmocn_mod * add compile fixes for ufs * removed reference to util and replaced with ufs * clean up of Makefile * removed references to use of med_kind_mod and introduced ufs_kind_mod * removed unneeded files needed by ufs * Optional tiled history files for ATM (#257)
Fixes an issue with how coefficients are applied to individual fire components.
…oupling through CMEPS (ESCOMP#61) * Add Wave slow loop coupling and Field Exchanges for waves in nems * Updates CMEPS for dynamic allocation in med_internalstate
### Description of changes Update shr_pio_mod, removing mct centric initialization. Improve logging and make sure logs go to the correct component. Depends on cime branch async_io_in_esmf ### Specific notes Contributors other than yourself, if any: CMEPS Issues Fixed (include github issue #): Are changes expected to change answers? bfb Any User Interface Changes (namelist or namelist defaults changes)? The modelio namelist has been eliminated and the parameters from that file were moved to nuopc.runconfig, The xml variable PIO_ASYNC_INTERFACE was changed from a global variable to one that could be set individually for each component. ### Testing performed Testing performed if application target is CESM: - [X] (recommended) CIME_DRIVER=nuopc scripts_regression_tests.py - machines: cheyenne intel - details (e.g. failed tests): IRT_N2_Vmct_Ln9.f19_g16_rx1.A.cheyenne_intel fails, apparently expected. - [ ] (recommended) CESM testlist_drv.xml - machines and compilers: - details (e.g. failed tests): - [x] (optional) CESM prealpha test - machines and compilers - details (e.g. failed tests): - [ ] (other) please described in detail - machines and compilers - details (e.g. failed tests): Testing performed if application target is UFS-coupled: - [ ] (recommended) UFS-coupled testing - description: - details (e.g. failed tests): Testing performed if application target is UFS-HAFS: - [ ] (recommended) UFS-HAFS testing - description: - details (e.g. failed tests): ### Hashes used for testing: - [ ] CESM: - repository to check out: https://github.com/ESCOMP/CESM.git - branch/hash: - [ ] UFS-coupled, then umbrella repostiory to check out and associated hash: - repository to check out: - branch/hash: - [ ] UFS-HAFS, then umbrella repostiory to check out and associated hash: - repository to check out: - branch/hash:
…MP#65) * Refactor source and destination masking * Refactor esmFieldsExchange with the advertise and initialize phases * Implement a check that each component pair for a field advertise is present. * Permit coupling of ATM-WAV with the CMEPS mediator.
modified: cesm/nuopc_cap_share/seq_drydep_mod.F90
new file: cesm/nuopc_cap_share/shr_drydep_mod.F90
modified: cesm/nuopc_cap_share/shr_drydep_mod.F90
Revert "first step - reorder pio_init and move to ensemble_driver"
merge to ESCOMP cmeps1.1.41
Sync w/ latest UFS changes ### Description of changes Adds recent changes for CMEPS in UFS ### Specific notes Adds ufs_trace wrapper to enable profiling for use in UFS. Adds capability to use mapfiles in UFS global applications. Removes HYCOM field exchanges in UFS/HAFS mode (HYCOM has been removed from our component list). Removes deprecated flux scheme for UFS (ocean_flux_scheme == -1) Updates ufs/ccpp as required for ccpp changes Changes should not touch any CESM code and should be B4B.
…m_v0 merge noresm cmeps1.1.57_noresm_v0 ### Description of changes Merge changes from noresm cmeps1.1.57_noresm_v0 back to ESCOMP ### Specific notes - In particular, this has enthalpy computation changes in the mediator that would occur for NorESM if CAM computed the enthalpy. However, although these hooks are in CMEPS - the computation of enthalpy in CAM has not been brought over to the main development branch yet. - In addition if the nuopc.config variable `histaux_l2x1yrg = .true. ` then if there is ocn2glc_coupling then the following mediator auxiliary file will be created: ` trim(case_name),'.cpl',trim(inst_tag),'.hx.lnd2glc.',trim(nexttime_str),'.nc'` if there is lnd2glc_coupling then the following mediator auxiliary file will be created: ` trim(case_name),'.cpl',trim(inst_tag),'.hx.lnd2glc.',trim(nexttime_str),'.nc'` Also - this PR might potentially interfere with PR ESCOMP/CMEPS#683 Contributors other than yourself, if any: None CMEPS Issues Fixed: None Are changes expected to change answers? bfb Any User Interface Changes (namelist or namelist defaults changes)? ### Testing performed Ran CESM prealpha testing on derecho with intel and gnu compilers, with baseline comparisons. Tests passed except expected failures.
Organize source code into four subdirectories - const, coupling, science, and util - and get the system and unit test builds working with this new organization. Note that I separated out the water species constants from shr_wtracers_mod to avoid having anything in the science directory depend on anything in the coupling directory; this particularly helped in keeping the shr_wiso unit tests clear of other dependencies (which was a pre-existing issue, but it showed up in my test run after this directory separation). Claude (Opus 5.5) assisted with analyzing the new directory layout, giving suggestions, writing the documentation, and (importantly) updating the CMakeLists files. (I reviewed all changes from Claude.)
These are needed by the CTSM unit testing. Previously they were in CMEPS and the CTSM unit test build included them via a different mechanism. Now that they're in share, it makes sense to include them via the same mechanism as other share sources.
| @@ -0,0 +1,24 @@ | |||
| module shr_water_species_mod | |||
There was a problem hiding this comment.
Note that I separated out the water species constants from
shr_wtracers_mod to avoid having anything in the science directory
depend on anything in the coupling directory; this particularly helped
in keeping the shr_wiso unit tests clear of other dependencies (which
was a pre-existing issue, but it showed up in my test run after this
directory separation).
There was a problem hiding this comment.
I wasn't sure whether to put shr_scam_mod in science or util. Based on what happens there, util seems most appropriate.
|
I have run aux_cime_baselines in the context of cesm3_0_alpha11d, with this branch along with the accompanying CMEPS branch (which I'll open a PR for soon). Tests pass and are bit-for-bit except expected failures. |
|
@mvertens - this does not include the change to the handling of |
|
See also the companion PR ESCOMP/CMEPS#721; the CMEPS PR depends on this CESM_share PR. |
|
I realize that this is a hard PR to review. Here is a summary that will hopefully make it easier: For the most part, the added F90 files here are moved over exactly from CMEPS. Exceptions are:
Various CMakeLists.txt files, as well as buildlib.csm_share, were changed to accommodate the reorganization of files. The top-level README.md and src/README.md were also changed to describe this reorganization. |
mvertens
left a comment
There was a problem hiding this comment.
This looks really great. It would be good if possible if we could change the filename of buildlib.csm_share to buildlib.share. If that just adds a new cime dependency it can be defferred.
Have backwards compatibility via a sym link. This is one step towards removing csm/cesm from this repository. Note that the library is still named csm_share because changing that would require numerous changes in multiple repositories.
Move share code out of CMEPS ### Description of changes This PR reorganizes the code that was in nuopc_cap_share. Previously this code was built as part of the share library - not CMEPS. This is now moved into the CESM_share repository in ESCOMP/EPS_Share#88, so that library boundaries reflect the repository boundaries. Specifically: (1) driver_pio_mod and esm_utils_mod are now in the cesm driver directory, since they are only used from there (2) all other files from nuopc_cap_share have been moved into CESM_share (3) shr_flux_mod has also been moved into CESM_share (this module was also built as part of the share library) **For CESM and NorESM, this PR depends on ESCOMP/EPS_Share#88; this CMEPS PR must be brought in at the same time or after that CESM_share PR.** ### Specific notes Contributors other than yourself, if any: Mariana Vertenstein helped figure out the organization we want CMEPS Issues Fixed (include github issue #): - Resolves #605 - Resolves #400 - Helps with #402 - Helps with #458 - Helps with #465 Are changes expected to change answers? (specify if bfb, different at roundoff, more substantial) - bfb Any User Interface Changes (namelist or namelist defaults changes)? no ### Testing performed aux_cime_baselines in the context of cesm3_0_alpha11d, with this branch along with the accompanying CESM_share branch. Tests pass and are bit-for-bit except expected failures.
Rename buildlib.csm_share to buildlib.share This updates config_files to agree with the change in file name made in ESCOMP/EPS_Share#88. That PR was backwards compatible, so it's safe for the CESM_share change to come in before this CIME change, but this CIME change cannot come in before the CESM_share change.
The main goal of this PR is to bring the CMEPS nuopc_cap_share code, as well as flux_atmocn/shr_flux_mod.F90, into this repository. Previously, the share library was built with a mix of code in this repository and code in nuopc_cap_share plus shr_flux_mod. This PR makes the library boundaries reflect the repository boundaries, which is more intuitive and will help address various needs (for unit testing, etc.). I brought over most git history of these modules from the CMEPS repository.
This led the src directory to grow larger, so I felt it was time to organize this directory into subdirectories. This PR does that, too, organizing code into four subdirectories: const, coupling, science, and util.
This PR also renames buildlib.csm_share to buildlib.share, but maintains a sym link to the old one for backwards compatibility.