Port ufs-weather-model to acorn (WCOSS2 TDS) #291
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
DusanJovic-NOAA
requested review from
junwang-noaa,
aerorahul,
climbfuji and
MinsukJi-NOAA
DusanJovic-NOAA
changed the title
aerorahul
reviewed
Nov 20, 2020
| setenv CMAKE_CXX_COMPILER CC | ||
| setenv CMAKE_Fortran_COMPILER ftn | ||
| setenv CMAKE_Platform wcoss2 | ||
|
|
There was a problem hiding this comment.
One could just add I_MPI_LINK environment variable in the debug modulefile and eliminate the hard-wired compile time flags.
| @@ -0,0 +1,3 @@ | |||
| set(INLINE_POST ON CACHE BOOL "Enable inline post" FORCE) | |||
| set(PARALLEL_NETCDF ON CACHE BOOL "Enable parallel NetCDF" FORCE) | |||
| set(DEBUG_LINKMPI OFF CACHE BOOL "Enable linkmpi option when DEBUG mode is on" FORCE) | |||
There was a problem hiding this comment.
Are we debugging MPI implementation here?
I see we are hard-wiring flags such as:
if(DEBUG)
if(OPENMP)
set(CMAKE_Fortran_FLAGS "${CMAKE_Fortran_FLAGS} -link_mpi=dbg_mt")
else()
set(CMAKE_Fortran_FLAGS "${CMAKE_Fortran_FLAGS} -link_mpi=dbg")
endif()
endif()
This should be handled by linking to the runtime debug MPI library by setting an environment variable such as I_MPI_LINK
By adding those flags at compile time, we are essentially debugging the MPI implementation.
| @@ -56,6 +56,9 @@ set +x | |||
| if [[ $MACHINE_ID == macosx.* ]] || [[ $MACHINE_ID == linux.* ]]; then | |||
| source $PATHTR/modulefiles/${MACHINE_ID}/fv3 | |||
| else | |||
| if [[ $MACHINE_ID == wcoss2 ]]; then | |||
| source /apps/prod/lmodules/startLmod | |||
There was a problem hiding this comment.
I guess this will eventually be moved to ../NEMS/src/conf/module-setup.*sh.inc?
There was a problem hiding this comment.
Yes. Or if they switch to Lmod version of modules, sourcing this script might not be necessary at all.
| @@ -173,6 +173,33 @@ elif [[ $MACHINE_ID = wcoss_dell_p3 ]]; then | |||
| cp fv3_conf/fv3_bsub.IN_wcoss_dell_p3 fv3_conf/fv3_bsub.IN | |||
| cp fv3_conf/compile_bsub.IN_wcoss_dell_p3 fv3_conf/compile_bsub.IN | |||
|
|
|||
| elif [[ $MACHINE_ID = wcoss2 ]]; then | |||
|
|
|||
| source /apps/prod/lmodules/startLmod | |||
There was a problem hiding this comment.
Once NEMS/src/conf/module-setup.*sh.inc is modified to include wcoss2, this line will not be needed.
|
Merged in #295 |
Closed
pjpegion
pushed a commit
to NOAA-PSL/ufs-weather-model.p7b
that referenced
this pull request
Jul 20, 2021
… UGWP bugfixes, LTP bugfixes, sfcsub update, framework update (ufs-community#291) * ugwpv1_gsldrag bug fixes, and GFS_typdefs updates associated with diag flags for ugwpv1_gsldrag and GFS_DCNV_generic.F90 * Change vertical dimension of arrays atmosphere_heat_diffusivity (aka dkt) and atmosphere_momentum_diffusivity (aka dku), update long names * Updaate .gitmodules and submodule pointers for ccpp-framework and ccpp-physics for code review and testing * Update descriptions of dku and dkt in ccpp/driver/GFS_diagnostics.F90 to match what is in the CCPP metadata * Revert .gitmodules and update submodule pointers for ccpp-framework and ccpp-physics Co-authored-by: Michael Toy <[email protected]>
epic-cicd-jenkins
pushed a commit
that referenced
this pull request
Apr 17, 2023
…sed to avoid bug with do_sppt/skeb/shum namelist entries (#291) * Remove all references to /lfs3 on Jet * Add Ben and Ratko to the CODEOWNERS file * Replace hard-coded make_orog module file with build-level module file in UFS_UTILS * Remove hard-coded make_sfc_climo module file * Fixes after updating fork with authoritative repo * Set ad-hoc stochastic physics scheme magnitudes to -999.0 when not used to avoid bug with do_sppt/skeb/shum namelist entries * Change stanzas in setup.sh for setting the ad-hoc schemes to "TRUE" and "FALSE" from lowercase to uppercase, and move the "if false, then -999.0" block from the generate to the setup script
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.
Description
Porting ufs-weather-model to acorn, WCOSS2 TDS. WCOSS2 is not using Intel CPU, so using
-xCORE-AVX2compile flag creates executable that can not run on non-intel cpus. That flag is replaced by-march=core-avx2Issue(s) addressed
Testing
I ran one test (fv3_ccpp_control) manually on acorn. Input data for full regression tests are not yet available. This system has only 32 compute nodes, I'm not sure if we can run all tests using only 32 nodes.
Full test will be performed on all Tier-1 platforms. No changes in the baselines on Tier-1 platforms.
Dependencies
N/A