Topo module variables removed from cellgridintegrate call sequence #579
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
They don't need to be in the calling sequence since setaux already assumes they are in topo_module. Some integers were implicitly declared as (kind=4) whereas recent changes to topo_module.f90 in e8afd6b declared them as (kind=8) to deal with larger topo files.
|
@rjleveque sorry I didn't get to this before the merge but will test out on the big files. Sorry for not catching this in #540 . Changes in cellgridintegrate make sense to me. The diffs seem to be messed up for setaux so didn't compare those but regardless, I'll test it out! |
|
Yes, sorry about diffs in setaux, I discovered that Windows CRLFs had been put at the ends of lines sometime years ago, and it differed from the version elsewhere, so I used dos2unix to clean it up. But it was only the calls to cellgridintegrate that were changed. |
|
Ahh makes sense. Good thing to fix |
rjleveque
added a commit
to rjleveque/geoclaw
that referenced
this pull request
Mar 24, 2024
…ellintegrate The same change was made in shallow/setaux.f90 in clawpack#579 but missed here.
bolliger32
added a commit
to ClimateImpactLab/geoclaw
that referenced
this pull request
Mar 26, 2024
* Update src2.f90 coriolis_forcing, wrong q index for multilayer case. * Start merging bouss code.... Copied modified .f* files from $BoussDev repository where they were developed, from $BoussDev subdirectories geoclaw_2d_shallow and amrclaw_2d Added Makefile.bouss to point to these files. * fix Makefile.bouss paths and use orig setprob * fix bouss_module.f90, now compiles * fix names and order of bouss variables and add to data.py * added examples/bouss/radial_shelf_beach_nonlinear, dies with >1 level Copied from https://github.com/rjleveque/ImplicitAMR-paper and modified to use new merged version for testing, still debugging. * turn off 1d comparison in setplot * update Makefile.bouss and add Riemann sovlers, gfixup.f * updated Makefile.bouss * remove R-solvers from example Makefile * missing file and fix setplot in example 1d_radial subdir * Fix restrt by moving bouss tests to bouss_module Also incorporated fix from clawpack#568 to compute next frame number properly depending on whether recreating last frame output. * clean up Makefiles * minor edits, fixed 1d setplot error * simplified petsc flags in bouss Makefile * cleaned up a lot to make nice example * clean up maketopo.py * fix comment in Makefile * examples/bouss/radial_flat added for debugging Dies using bouss_equations=1 with too many dt reductions 1d_radial code dies with bouss_equations=2 when ingoing wave hits origin * Makefile and setrun changes for initial radial_flat tests * import geoplot from clawpack.visclaw for 1d * debugging 1d_radial -- Gaussian at 0 for a few steps blows up with sgn * debugging - runs ok with depth 100m * radial_flat/1d_radial 100m good example * use finer grid in 1d_radial * improve examples/bouss/radial_flat and add README * clean up petsc stuff in radial_flat/Makefile * Removed examples/bouss/radial_shelf_beach_nonlinear available elsewhere Changes to run with geoclaw are being made to the version in the repo https://github.com/rjleveque/ImplicitAMR-paper which also has some other bouss examples. * remove extraneous variables from qinit.f90 and change setrun for 2 level run in radial_flat example * add petscMPIoptions file and README to bouss/examples * use newer shallow/set_eta_init.f90 using int*8 * use shallow/flag2refine2.f90 without mptr arg used only for debugging * hardwire bathy_index and eta_index in shallow/fgout_module.f90 so also works in bouss * fix indent to match shallow in one place * use shallow/getmaxspeed.f90 in bouss * use shallow/tick.f which seems newer * use shallow gfixup.f since prepnewgrids_new not used * removed old huc,hvc stuff from bc2amr.f90 * fix indents in valout.f90 * use shallow/errf1.f and qinit.f90 in bouss * RUNEXE no longer includes * initialize bouss_equations to 2 in data.py * sample setenv.sh bash script for bouss * update petscMPIoptions * changes for compressed row storage, NB: arrays starts at 0 for rhs and soln * bug fix in crs petsc driver, set triplet false in bouss_module * bug in triplet csr testing * found timing bug in advanc * fix timers - remove double counting * fix comment in petscMPIoptions about solver and preconditions used * remove triplet and use logical crs, fix comments * fix lines that were improperly joined * rename buildSparseMatrix files * update buildSparse files with new subroutine names * updated comments in petsc_driver after talking to Barry * remove old unused prepBuild routine * fix test in marching_front.py * set rundata.geo_data.sphere_source = 1 by default in data.py * fix links in examples/tsunami/eta_init_force_dry README and add comment to setrun * revert one accidental file commit * Remove deprecation warning for topotype 1 * Remove deprecation warning for topotype 1 second attempt since wrong file committed previously * another attempt to fix topo_module.f90 * update storm.py to prep for ds.dims change * Better comments and new tests in examples/bouss/radial_flat/Makefile To make sure user has necessary environment variables set to use PETSc, MPI. Otherwise gives an error, e.g. `Makefile:36: *** PETSC_DIR not set. Stop.` Also a new target, so that `make check` prints out the relevant variables. We might want to adopt something similar in `Makefile.common` to be used in all Clawpack examples (showing only CLAW, FFLAGS, and OMP_NUM_THREADS by default). * Update src/2d/shallow/multilayer/setaux.f90 to remove args from gridcellintegrate The same change was made in shallow/setaux.f90 in clawpack#579 but missed here. --------- Co-authored-by: YuchengZhang-hub <[email protected]> Co-authored-by: Randy LeVeque <[email protected]> Co-authored-by: Marsha Berger <[email protected]> Co-authored-by: Kyle Mandli <[email protected]>
bolliger32
added a commit
to ClimateImpactLab/geoclaw
that referenced
this pull request
Mar 29, 2024
* Update src2.f90 coriolis_forcing, wrong q index for multilayer case. * Start merging bouss code.... Copied modified .f* files from $BoussDev repository where they were developed, from $BoussDev subdirectories geoclaw_2d_shallow and amrclaw_2d Added Makefile.bouss to point to these files. * fix Makefile.bouss paths and use orig setprob * fix bouss_module.f90, now compiles * fix names and order of bouss variables and add to data.py * added examples/bouss/radial_shelf_beach_nonlinear, dies with >1 level Copied from https://github.com/rjleveque/ImplicitAMR-paper and modified to use new merged version for testing, still debugging. * turn off 1d comparison in setplot * update Makefile.bouss and add Riemann sovlers, gfixup.f * updated Makefile.bouss * remove R-solvers from example Makefile * missing file and fix setplot in example 1d_radial subdir * Fix restrt by moving bouss tests to bouss_module Also incorporated fix from clawpack#568 to compute next frame number properly depending on whether recreating last frame output. * clean up Makefiles * minor edits, fixed 1d setplot error * simplified petsc flags in bouss Makefile * cleaned up a lot to make nice example * clean up maketopo.py * fix comment in Makefile * examples/bouss/radial_flat added for debugging Dies using bouss_equations=1 with too many dt reductions 1d_radial code dies with bouss_equations=2 when ingoing wave hits origin * Makefile and setrun changes for initial radial_flat tests * import geoplot from clawpack.visclaw for 1d * debugging 1d_radial -- Gaussian at 0 for a few steps blows up with sgn * debugging - runs ok with depth 100m * radial_flat/1d_radial 100m good example * use finer grid in 1d_radial * improve examples/bouss/radial_flat and add README * clean up petsc stuff in radial_flat/Makefile * Removed examples/bouss/radial_shelf_beach_nonlinear available elsewhere Changes to run with geoclaw are being made to the version in the repo https://github.com/rjleveque/ImplicitAMR-paper which also has some other bouss examples. * remove extraneous variables from qinit.f90 and change setrun for 2 level run in radial_flat example * add petscMPIoptions file and README to bouss/examples * use newer shallow/set_eta_init.f90 using int*8 * use shallow/flag2refine2.f90 without mptr arg used only for debugging * hardwire bathy_index and eta_index in shallow/fgout_module.f90 so also works in bouss * fix indent to match shallow in one place * use shallow/getmaxspeed.f90 in bouss * use shallow/tick.f which seems newer * use shallow gfixup.f since prepnewgrids_new not used * removed old huc,hvc stuff from bc2amr.f90 * fix indents in valout.f90 * use shallow/errf1.f and qinit.f90 in bouss * RUNEXE no longer includes * initialize bouss_equations to 2 in data.py * sample setenv.sh bash script for bouss * update petscMPIoptions * changes for compressed row storage, NB: arrays starts at 0 for rhs and soln * bug fix in crs petsc driver, set triplet false in bouss_module * bug in triplet csr testing * found timing bug in advanc * fix timers - remove double counting * fix comment in petscMPIoptions about solver and preconditions used * remove triplet and use logical crs, fix comments * fix lines that were improperly joined * rename buildSparseMatrix files * update buildSparse files with new subroutine names * updated comments in petsc_driver after talking to Barry * remove old unused prepBuild routine * fix test in marching_front.py * set rundata.geo_data.sphere_source = 1 by default in data.py * fix links in examples/tsunami/eta_init_force_dry README and add comment to setrun * revert one accidental file commit * Remove deprecation warning for topotype 1 * Remove deprecation warning for topotype 1 second attempt since wrong file committed previously * another attempt to fix topo_module.f90 * Better comments and new tests in examples/bouss/radial_flat/Makefile To make sure user has necessary environment variables set to use PETSc, MPI. Otherwise gives an error, e.g. `Makefile:36: *** PETSC_DIR not set. Stop.` Also a new target, so that `make check` prints out the relevant variables. We might want to adopt something similar in `Makefile.common` to be used in all Clawpack examples (showing only CLAW, FFLAGS, and OMP_NUM_THREADS by default). * Update src/2d/shallow/multilayer/setaux.f90 to remove args from gridcellintegrate The same change was made in shallow/setaux.f90 in clawpack#579 but missed here. * Fix empty time offset and pandas reading deprecation warning * remove examples/run_tests.py and point to new common script in clawutil * update examples/README.txt regarding env variables * download gulf_carribean.tt3 from geoclaw.org --------- Co-authored-by: YuchengZhang-hub <[email protected]> Co-authored-by: Randy LeVeque <[email protected]> Co-authored-by: Marsha Berger <[email protected]> Co-authored-by: Kyle Mandli <[email protected]>
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.
WIP: Still need to test some more before merging.
Remove
topo_modulevariables from calling sequence ofcellgridintegrateThey don't need to be in the calling sequence since setaux already assumes they are in topo_module.
Some integers were implicitly declared as (kind=4) whereas recent changes to topo_module.f90 in #540 declared them as (kind=8) to deal with larger topo files. As a result the code dies when reading some large topo files when i0topo overflows 4 bytes and was set to 0 in
gridcellintegrate.@bolliger32: You might want to take a look and see if it still works for your big files.