Fix bugs in inactive top cells feature #5
Closed
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
cbegeman
force-pushed
the
ocn/bugfix-inactive-top-cells
branch
from
4d42267 to
64e846f
Compare
cbegeman
force-pushed
the
ocn/bugfix-inactive-top-cells
branch
2 times, most recently
from
de0f78a to
2236e64
Compare
cbegeman
force-pushed
the
ocn/bugfix-inactive-top-cells
branch
from
2236e64 to
7c0f132
Compare
added 24 commits
September 8, 2022 10:17
- Comment out Langmuir enhancement (lamult) and mixed layer depth (HML) variables, since they are not in the new wavewatch source code. - Modify the field output specification. The outputs are now grouped into 9 different groups. The flag array to turn them on/off is now 2D - Add field output flag variables for coupling variables (not used) - Update call to w3init - Increase size of odat variable to account for coupling variables
- The NOTYPE variable wasn't being assigned and defaulted to 0. The loop that goes over output types to call the write routines was being skipped over. - Setting the NOTYPE variable to 7 in wav_comp_mct.F90 fixes the issue.
- Add w3_clean command - Get rid of redundant environment variable specification - Switch NETCDF_CONFIG variable to intel
- This is the cleanest way to include this file in the build process
This reverts commit 3a4455f.
- Use WW3's w3_source script to pre-process the source code for the ww3_shel program only, instead of using w3_make to pre-process and compile all programs - w3_source creates a .tar.gz file that must be untared before calling the e3sm makefile
- Only include ww3_shel files since those are now the only ones generated by buildlib - ww3_grid files are also included, since that program was built using w3_make to create the mod_def.ww3 file
- Add argument to call to w3init in wav_comp_mct.F90 - Update paths to ww3 repo in buildlib. Also handle setting up the wwatch3.env file (if it doesn't already exist) by calling w3_setup - Point to correct mod_def.ww3 and restart.ww3 files in buildnml and namelist_definition_ww3.xml
- This file will get copied to WW3 repo and when w3_setup is run, it will cause this to be used as the switch file for pre-processing the .ftn source code
- This makes sure that the WW3 git working directory is clean (no untracked files) after building
- Add fields in seq_flds_mod.F90 - Set w2x indicies in ww3_cpl_indicies.F90 - Set x2o indicies in mpaso)cpl_indices.F90 - Pack variables in wav_comp_mod.F90 - No modifications to MPAS-O source code yet
Fixes: 1) get_case_env: needs to load Config 2) bless_test_results: Never bless a failed test 3) PEM: fix potential invalid rootpe 4) fix for multi driver restarts Changes: 1) cprnc: add a data type check 2) Updates config_batch.xml to facilitate queue specific flags e.g. -w docker would only apply when the queue name is long. 3) Updates workflow to build and push containers for PRs 4) Shorten test-job-id with --jenkins-id option 5) improve the pfs test by copying software_environment.txt [BFB]
Double PEs to prevent walltime timeouts on Ascent for
- ERS.hcru_hcru.I20TRGSWCNPRDCTCBC.elm-erosion
- ERS.ne30pg2_r05_EC30to60E2r2.GPMPAS-JRA.mosart-rof_ocn_2way
(on Crusher too)
- ERS.r05_r05.IELM.elm-V2_ELM_MOSART_features
[BFB]
Update to github.io in mpas license website Until 2021 github supported the alternate web address github.com. In 2021 it changed exclusively to github.io. This PR updates the link to the MPAS license on all MPAS headers. It only changes code comments, not working code. [BFB]
…SM-Project#5201) Remove unneeded mesh declaration Mesh variables are declared in mpas_ocn_mesh.F rather than individual routines. Fixes E3SM-Project#5200 [BFB]
add elm input files to the land-atm 1850 and 20tr compsets and make sure the simulations use correct namelist files. [NBFB]
…oject#5231) Update CIME submodule ... to 977c1f6cd9283d03e356c8370a32302be4991cf4 Fixes: 1) get_case_env: needs to load Config 2) bless_test_results: Never bless a failed test 3) PEM: fix potential invalid rootpe 4) fix for multi driver restarts Changes: 1) cprnc: add a data type check 2) Updates config_batch.xml to facilitate queue specific flags e.g. -w docker would only apply when the queue name is long. 3) Updates workflow to build and push containers for PRs 4) Shorten test-job-id with --jenkins-id option 5) improve the pfs test by copying software_environment.txt [BFB]
…t#5233) Adding HOMMEXX_ENABLE_GPU to Hommexx_config in order for scream code to pick it up. Scream builds of homme do not use config file in theta-l_kokkos. [bfb]
Add WAVEWATCH III as new coupled wave component This is a stealth feature PR that adds the WAVEWATCH III submodule, cime configuration, namelist generation, build infrastracture, and coupling interface. Wave only, G case + waves, and B case + waves compsets are included along with model resolutions that include the wave mesh intended for v1 and v2 LR configurations. The initial coupled, active wave feedback is a parameterization for Langmuir mixing. [NML] for configurations with active ocean [BFB]
…ct#5187) Fix out-of-bounds index for vertical mixing with one active layer The current implicit vertical mixing routine fails when a column has one active layer, because of references to k+1 and k−1. This PR adds a check on each edge, and if there is only one active layer, it applies the implicit drag term but not the vertical mixing terms. When only a single layer exists, implicit vertical mixing is not needed (or even valid), but implicit drag is still desirable. There are two cases where the model may have a single layer: 1. when the whole domain is single layer; and 2. when individual columns have only one active vertical cell, like near coastlines with wetting-and-drying. Fixes E3SM-Project#4849 [BFB]
…#5196) Use reduced g-point data when using RRTMGP in EAM and E3SM-MMF. This should half the cost of RRTMGP in E3SM and bring performance in line with RRTMG, which was already using a similar number of quadrature points. This will be non-BFB only for the tests that use RRTMGP (eam-rrtmgp/eam-rrtmgpxx testmods, and all of the MMF tests), but BFB for all other tests. [non-BFB] for RRTMGP tests
…ject#5232) Add zeroing of velocity to ocean init mode Explicitly sets normal velocity to zero at the beginning of init mode. Each test case can then overwrite the zeros with initial velocities, but most do not. In the past the unassigned arrays wrote out zeros, but this is not true now, as described in E3SM-Project#5230. This code is used to create MPAS-Ocean initial conditions. It is not compiled in the E3SM code base for simulations. Fixes E3SM-Project#5230 [BFB]
cbegeman
force-pushed
the
ocn/bugfix-inactive-top-cells
branch
2 times, most recently
from
38431f9 to
719a98f
Compare
|
This PR has been migrated and split: |
mark-petersen
pushed a commit
that referenced
this pull request
May 31, 2023
cee/15.0.0 with GPU MPI buffers can crash in a system lib like this: #4 0x00007fffe159e35b in (anonymous namespace)::do_free_with_callback(void*, void (*)(void*)) [clone .constprop.0] () from /opt/cray/pe/cce/15.0.0/cce/x86_64/lib/libtcmalloc_minimal.so.1 #5 0x00007fffe15a8f16 in tc_free () from /opt/cray/pe/cce/15.0.0/cce/x86_64/lib/libtcmalloc_minimal.so.1 #6 0x00007fffe99c2bcd in _dlerror_run () from /lib64/libdl.so.2 #7 0x00007fffe99c2481 in dlopen@@GLIBC_2.2.5 () from /lib64/libdl.so.2 #8 0x00007fffea7bce42 in _ad_cray_lock_init () from /opt/cray/pe/lib64/libmpi_cray.so.12 #9 0x00007fffed7eb37a in call_init.part () from /lib64/ld-linux-x86-64.so.2 #10 0x00007fffed7eb496 in _dl_init () from /lib64/ld-linux-x86-64.so.2 #11 0x00007fffed7dc58a in _dl_start_user () from /lib64/ld-linux-x86-64.so.2 #12 0x0000000000000001 in ?? () #13 0x00007fffffff42e7 in ?? () #14 0x0000000000000000 in ?? () Work around this by using cee/14.0.3.
xylar
pushed a commit
that referenced
this pull request
Nov 8, 2023
Update development branch to E3SM master
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
This PR fixes 2 bugs that affect the inactive top cells feature (i.e., when
minLevelCell > 1). One inocn_diagnostic_solve_sshand the other affects the sea ice salinity flux inocn_surface_bulk_forcing.No changes in the solution should occur when there are no inactive top cells.
This PR also includes a cleanup of Redi. The original clunky workaround was designed to avoid non-bfb changes in BGC tracers as discussed here, and this revision introduces those non-bfb changes as @maltrud determined that they are insignificant.
One commit f59b505 increases the generalizability of vertical coordinates by relaxing the assumption that
restingThicknessspansssh=0tobottomDepth. As such, it is tangentially related to inactive top cells.