Use CPP instead of Switches#476
Conversation
although remove build suppert for NETCDFv3
if (NCTYPE=4) statements on NC4 lines instead
therefore has to be NetCDF4+
Conflicts: model/ftn/ww3_ounf.ftn
…eature/removeNC4switch
…eature/removeNC4switch Conflicts: manual/impl/switch.tex model/bin/make_makefile.sh
…ure/switch2cpp_build_nosrc Conflicts: model/tools/ftn2src.sh
…o feature/switch2cpp_build_src Conflicts: model/src/gx_outp.F90 model/src/w3gridmd.F90 model/src/w3iogrmd.F90 model/src/w3iopomd.F90 model/src/w3odatmd.F90 model/src/w3src4md.F90 model/src/w3srcemd.F90 model/src/w3wavemd.F90 model/src/ww3_ounf.F90 model/src/ww3_ounp.F90 model/src/ww3_outp.F90
|
Okay, the latest develop has been merged in. This branch is ready for further testing. @aliabdolali since the ufs-weather-model results have changed, I cannot make a PR without changing results for this PR until your PR is complete. I will make a subsequent ufs-weahter-model PR with this branch, but as I think this is a fairly large change it should be done separately. I will make a ufs-weather-model issue for this. |
|
All of the code here is up to date and is ready for everyone to do a full set of regression tests. I'll start a new set here at EMC as well. @ukmo-ccbunney @thesser1 (or Mary) @mickaelaccensi can you please all test again? @aliabdolali this is ready for your review as well. I'll post new matrix.comp & ufs-weather-model regtest updates asap. |
|
Full regression test matrix completed ok with only the usual suspects being non-B4B: |
Thanks @ukmo-ccbunney |
|
The regression tests on hera are still in the queue. I'm working on getting the matrix.comp output for hera. The ufs-weather-model tests have run on some but not all platforms I have access too. Will post results when I have them. Thanks @ukmo-ccbunney for running the tests at your end. Do you approve of this PR now? (It still says you request changes). |
|
Issue #494 is stopping me from completing these tests. We will likely need some sort of bug-fix in develop before we can proceed with testing this. |
|
Regression results from my intel compiler ********************* non-identical cases **************************** mww3_test_03/./work_PR2_UQ_MPI_d2 (8 files differ) |
Thanks @thesser1 |
|
I'll post comparisons in the next hour --- so far all tests are looking good for me. Then we just need @mickaelaccensi to confirm things are working |
|
it's running. I'll tell you the results once it's done |
|
Regtests passed on hera, except for known issues: Regression tests also ran on orion, with known issues of netcdf errors matrixDiff.txt ufs-weather-model passed for gaea.intel, hera.intel and Orion.intel for wave tests except for known point output not matching. The branch with tests is here: https://github.com/JessicaMeixner-NOAA/ufs-weather-model/tree/feature/ww3cpp |
|
all regtests passed except usual mww3_test_03 |
5 participants
Pull Request Summary
Use CPP ifdefs instead of switches for pre-processing and convert all code from ftn to F90 files. (ftn->src)
Description
This PR provides a utility in tools/ that converts the WW3 switches into CPP ifdefs. The build is also updated for this change. The utility in model/tools/ftn2src.sh is run to convert the ftn to a src directory.
Co-author: @ukmo-ccbunney
Issue(s) addressed
Check list
Is your feature branch up to date with the authoritative repository (NOAA/develop)? yes (+PR454)
Make sure you have checked the checklist for a developer submitting to develop and updating version number Version number is not currently updated for this.
Reviewers: @aliabdolali @ukmo-ccbunney @thesser1 @mickaelaccensi
Commit Message
Testing
matrixCompFull.txt
matrixCompSummary.txt
matrixDiff.txt
Please indicate the expected changes in the outputs (excluding the known list of non-identical tests).
There should be no output changes