User/jml/operation fms2 the reckoning#1033

Closed

wrongkindofdoctor wants to merge 5474 commits intomom-ocean:dev/gfdlfrom wrongkindofdoctor:user/jml/operation_fms2_the_reckoning

+4,627-1,340

This pull request is big! We're only showing the most recent 250 commits

Commits on Jul 31, 2019

removed a rogue 'vd%' from the 't_grid_str' definition in write_initial_condition, added a default case that throws a fatal error if the 't_grid_str' is empty in write_initial_conditions and save_r…
wrongkindofdoctor
committed

Commits on Aug 12, 2019

Merge pull request #26 from NOAA-GFDL/dev/gfdl
wrongkindofdoctor
authored
Merge branch 'dev/gfdl' into user/jml/operation_fms2_the_reckoning
wrongkindofdoctor
committed

Commits on Aug 19, 2019

added call to mpp_define_io_domain in initialize_ocean_public_type to facilitate use of new open_file, which requires that these values be associated
wrongkindofdoctor
committed

Commits on Aug 21, 2019

Commits on Aug 28, 2019

Commits on Aug 29, 2019

removed 'fms_' from rest of new io routine names to eliminate the pointers, put the new io public routines in abc order
wrongkindofdoctor
committed
removed unused string variable definitions, added deallocate statments for axis_data_CS%axis and axis_data_CS%data
wrongkindofdoctor
committed
added a routine for the dyn_horgrid data type structure G to the MOM_open_file interface
wrongkindofdoctor
committed
fixed the domain argument in the dyn_horgrid open file routine
wrongkindofdoctor
committed
added the fileobjwrite argument to the write_data calls
wrongkindofdoctor
committed
removed unused variable
wrongkindofdoctor
committed
replaced the energy diagnostics create_file and reopen_file calls with the new fms io open, register, write, and close routines
wrongkindofdoctor
committed
add dim_names declaration, redefined the time_vals argument as an array
wrongkindofdoctor
committed
added routine for non-domain-decomposed files to 'MOM_open_file' interface. This accomodates the writing of the vertical coordinate file in MOM_ALE and MOM_coord_initialization.
wrongkindofdoctor
committed
removed the G argument from MOM_initialize_coord and write_vertgrid_file routines, and call to MOM_open_file. This file is not domain-decomposed, and does not need the mpp_domains from G
wrongkindofdoctor
committed
added type FmsNetcdfFile_t for non-domain-decomposed files
wrongkindofdoctor
committed
added type FmsNetcdfFile_t for non-domain-decomposed files
wrongkindofdoctor
committed
removed G argument from MOM_initialize_coord call since the routine now calls write_vertgrid_file without the argument for a non-domain-decomposed write
wrongkindofdoctor
committed
modified write_vertgrid_file to use the fms io interfaces for writing non-domain-decomposed files, added the type FmsNetcdfFile_t
wrongkindofdoctor
committed
working out issues with defining the index values to use for reading the supergrid stuff
wrongkindofdoctor
committed

Commits on Aug 30, 2019

added calls to more new fms-io routines
wrongkindofdoctor
committed
added some temporary routines to query the axis and variable dimesions to help sort out the supergrid dimension index issues
wrongkindofdoctor
committed
re-defined the corner and edge_lengths arguments for read_data, but still having issues
wrongkindofdoctor
committed

Commits on Sep 3, 2019

Commits on Sep 4, 2019

testing hacky crap to deal with symmetric indexing into a temp array for a ferret workaround
wrongkindofdoctor
committed

Commits on Sep 5, 2019

Commits on Sep 6, 2019

removed unused calls, swapped unstructured grid stuff out of MOM_grid_initialize. I think that the temporary buffer size needs to be defined on the data domain rather than the compute domain
wrongkindofdoctor
committed

Commits on Sep 13, 2019

replaced the MOM_read_data calls for x and y into tmpGlbl with read_data calls for non-decomposed file; added the fms io file type FmsNetcdfFile_t; added call to close the file for fileobjread and …
wrongkindofdoctor
committed
minor cleanup
wrongkindofdoctor
committed

Commits on Sep 16, 2019

Commits on Sep 17, 2019

Fixed the logic and moved the register and write statements after the computations in write_energy. Need to add corner and edge_lengths arguments to write values for multiple times
wrongkindofdoctor
committed

Commits on Sep 18, 2019

added corner and edge_length indices to the call to write the time data, added use statement for the MOM 'lowercase' function
wrongkindofdoctor
committed

Commits on Sep 19, 2019

Commits on Sep 23, 2019

Commits on Sep 24, 2019

added interface routines to assign 0d,1d, 2d pointers to output arrays, added logic to check dimension names for each variable to determine whether data have time axes and need to be written at eve…
wrongkindofdoctor
committed

Commits on Sep 25, 2019

Commits on Sep 26, 2019

Merge pull request #28 from NOAA-GFDL/dev/gfdl
wrongkindofdoctor
authored
merged in dev/gfdl updates
wrongkindofdoctor
committed
added unlimited and get_variable_unlimited_dimension_index to public MOM_io routines
wrongkindofdoctor
committed
added get_variable_unlimited_dimension_index routine
wrongkindofdoctor
committed
cleaned up unused code, removed the start attribute from data_struct type, added calls to get_unlimited_variable_dimension_index to check for time dimension in output variables, added the optional …
wrongkindofdoctor
committed

Commits on Sep 27, 2019

removed commented-out calls to old write routines, and unused code mods
wrongkindofdoctor
committed

Commits on Oct 2, 2019

Commits on Oct 3, 2019

added optional position argument to diag_axis_init calls for id_xq and id_yq. This didn't fix the array value problems for the diagnostic files
wrongkindofdoctor
committed

Commits on Oct 4, 2019

appended .nc to the hard-coded ocean geometry and vertical coordinate netcdf file names
wrongkindofdoctor
committed
Merge pull request #31 from NOAA-GFDL/dev/gfdl
wrongkindofdoctor
authored
changed the save_restart axis registration loop to use the global io domain indices for the buffer, added the write data calls for the axes to the register loop
wrongkindofdoctor
committed
Merge branch 'dev/gfdl' into user/jml/operation_fms2_the_reckoning
wrongkindofdoctor
committed
messed with the diagnostic writes, and now MOM_shared_initialization is hanging
wrongkindofdoctor
committed
forgot to save
wrongkindofdoctor
committed

Commits on Oct 22, 2019

Merge pull request #32 from NOAA-GFDL/dev/gfdl
wrongkindofdoctor
authored
Merge branch 'dev/gfdl' into user/jml/operation_fms2_the_reckoning

wrongkindofdoctor
authored and
wrongkindofdoctor
committed
fixed typo

wrongkindofdoctor
authored and
wrongkindofdoctor
committed
added new subroutine to return logical vars for use_lath, use_lonh, use_latq, use_lonq based on horizontal grid string classification

wrongkindofdoctor
authored and
wrongkindofdoctor
committed
reworked the the routines to write ocean_geometry.nc,replaced MOM6 register_axis wrapper call with register_axis for the lat and lon coordinates; having trouble getting the correct indices for the …

wrongkindofdoctor
authored and
wrongkindofdoctor
committed
fixed a typo, still messing with domain crap

wrongkindofdoctor
authored and
wrongkindofdoctor
committed

Commits on Oct 23, 2019

fixed the write_data procedure for write_ocean_geometry_file, but still cant get the correct latitude values

wrongkindofdoctor
authored and
wrongkindofdoctor
committed

Commits on Oct 24, 2019

separated a logical statement into 2 parts

wrongkindofdoctor
authored and
wrongkindofdoctor
committed

Commits on Oct 28, 2019

still debugging; looks like the MOM latitudes are off going into the write_geom routine, so will have to see if the MOM6 crew messed with the coordinates recently, or if there's some bizarro magic …

wrongkindofdoctor
authored and
wrongkindofdoctor
committed

Commits on Oct 30, 2019

removed whitespace; identified lat problem with new no-domain read implementation
wrongkindofdoctor
committed

Commits on Oct 31, 2019

the lat error occurs during the read into the 2d array. For some reason, the first longitude (nxp=1) has a final latitude value of approx 69. The largest latitude returned by the indexing routine i…

wrongkindofdoctor
authored and
wrongkindofdoctor
committed
Merge branch 'user/jml/operation_fms2_the_reckoning' of github.com:wrongkindofdoctor/MOM6 into user/jml/operation_fms2_the_reckoning

wrongkindofdoctor
authored and
wrongkindofdoctor
committed

Commits on Nov 1, 2019

removed unused variables, added a procedure to loop through the temporary arrays with global y values and search for the longitude with the full ranges of geoLatT and geoLatB values. The indices fo…

wrongkindofdoctor
authored and
wrongkindofdoctor
committed
fixed error check statements

wrongkindofdoctor
authored and
wrongkindofdoctor
committed
deleted old io calls, cleaned up the latitude index checks to throw a fatal if the latitudes are not between 0 and 91 in MOM_grid_initialize

wrongkindofdoctor
authored and
wrongkindofdoctor
committed
whitespace cleanup, removed checks for being on the root pe for r/w/open/close, fixed spacing

wrongkindofdoctor
authored and
wrongkindofdoctor
committed
whitespace cleanup, removed checks for being on the root pe for r/w/open/close, fixed spacing, reduced the axis_data_CS%axis and data sizes to 2 since there are only 2 axes written to the file, add…

wrongkindofdoctor
authored and
wrongkindofdoctor
committed
added use and public declarations for fms-io routines required by MOM_restart

wrongkindofdoctor
authored and
wrongkindofdoctor
committed
cleaned up spacing and unused routines, reduced axis_CS allocatable array sizes to 7, removed fms from new io calls where possible, changed the use statements to reference routines in MOM_io instea…

wrongkindofdoctor
authored and
wrongkindofdoctor
committed

Commits on Nov 4, 2019

removed debugging statements, changed initial values of ymax1 and ymax2 to 999.0, added 0 after FP values ending in decimal pt, made error statments more specific.
wrongkindofdoctor
committed
start updating to new fms
wrongkindofdoctor
committed

Commits on Nov 5, 2019

added new fms io routines get_dimension_size, get_dimension_names, get_num_dimensions
wrongkindofdoctor
committed
added calls to new read, register, dimension size, dimension name routines, added call to scaling routine that I will add to MOM_io to replace the functionality in the MOM_io::MOM_read_data interfaces
wrongkindofdoctor
committed
added calls to new open/close/read and scale_data routines to MOM_state_initialization

wrongkindofdoctor
authored and
wrongkindofdoctor
committed
changed > to greater than in error message b/c fortran can't parse symbols

wrongkindofdoctor
authored and
wrongkindofdoctor
committed
modifying the read_vector routines; need a way to check for lat or lon

wrongkindofdoctor
authored and
wrongkindofdoctor
committed
working on read_vector calls

wrongkindofdoctor
authored and
wrongkindofdoctor
committed

Commits on Nov 6, 2019

modifying write_vector routines, changed the MOM_register_axis routine to MOM_register_diagnostic_axes; need to update calls in the rest of the code

wrongkindofdoctor
authored and
wrongkindofdoctor
committed

Commits on Nov 7, 2019

change MOM_register_aixs to MOM_register_diagnostic_axis, removed interface for MOM_register_axis, replaced old MOM_read_data calls with open_file, register_axis, read_data, close_file procedure

wrongkindofdoctor
authored and
wrongkindofdoctor
committed
change MOM_register_aixs to MOM_register_diagnostic_axis, removed interface for MOM_register_axis, replaced old MOM_read_data calls with open_file, register_axis, read_data, close_file procedure

wrongkindofdoctor
authored and
wrongkindofdoctor
committed
replaced MOM_read_data calls, added checks for axis units of degrees_east and degrees_north to determine whether to pass x or y to register_axis for domain decomposed reads, updated MOM_read_vector…

wrongkindofdoctor
authored and
wrongkindofdoctor
committed

Commits on Nov 8, 2019

changed MOM_register_axis to register_axis, whitespace cleanup, changed the size of allocated data and axis axis_data_cs attributes to 3

wrongkindofdoctor
authored and
wrongkindofdoctor
committed
changed MOM_register_axis to register_axis, whitespace cleanup, changed the size of allocated data and axis axis_data_cs attributes to 3

wrongkindofdoctor
authored and
wrongkindofdoctor
committed
removed unused parameter

wrongkindofdoctor
authored and
wrongkindofdoctor
committed
cleanup, removed the cartesian axis attribute from the axis_data_type

wrongkindofdoctor
authored and
wrongkindofdoctor
committed
changed MOM_get_axis_data to MOM_get_diagnostic_axis_data

wrongkindofdoctor
authored and
wrongkindofdoctor
committed
whitespace cleanup, edited MOM_get_diagnostic_axis_data description

wrongkindofdoctor
authored and
wrongkindofdoctor
committed
added new function MOM_register_variable_axes to query axes associated with variables and register them, changed the variables in MOM_register_diagnostic_axes to camelCase

wrongkindofdoctor
authored and
wrongkindofdoctor
committed
added new open, MOM_register_variable_axes, read_data, close routine

wrongkindofdoctor
authored and
wrongkindofdoctor
committed

Commits on Nov 12, 2019

Merge pull request #33 from NOAA-GFDL/dev/gfdl
wrongkindofdoctor
authored
merged in latest dev/gfdl updates

wrongkindofdoctor
authored and
wrongkindofdoctor
committed
added new interface routines MOM_get_nc_corner_edgelengths that set variable corner and edge_length arrays based on get_var_size and/or user-specified indices

wrongkindofdoctor
authored and
wrongkindofdoctor
committed
FINALLY replaced all MOM_read_data_calls with new fms-io. General structure is open, register_variable_axes, read, close. Modules reading data slices include calls to MOM_get_nc_corner_edgelengths,…

wrongkindofdoctor
authored and
wrongkindofdoctor
committed

Commits on Nov 13, 2019

started debugging new IO, managed to fine more effing MOM_read_data calls, so I'm replacing those with the new io, too.
wrongkindofdoctor
committed

Commits on Nov 14, 2019

fixed the calls to MOM_open_file, added checks for the unlimited dimension where code assumes presence of time dimension, but does not specify it in the data buffer (e.g., there was no loop that wa…
wrongkindofdoctor
committed
FINALLY FINISHED REPLACING THE MOM_READ_DATA CALLS WITH READ_DATA,etc...
wrongkindofdoctor
committed
removed unused routines and calls including create_file, reopen_file, mom_read_data
wrongkindofdoctor
committed
cleaned more unused routines, but found more MOM_read_calls in the config_src drivers. Lord.
wrongkindofdoctor
committed
more cleanup
wrongkindofdoctor
committed
added write_version_number back to MOM_io b/c still needed by drivers
wrongkindofdoctor
committed
updating drivers; need to figure out best way to specify 10+ variables w/potentially different file names
wrongkindofdoctor
committed

Commits on Nov 15, 2019

Commits on Nov 18, 2019

removed calls to fms io_infra_end and get_filename_appendix
wrongkindofdoctor
committed
cleaned up typos, added function num_timelevels back to MOM_io with new IO calls since some of the driver routines use the function
wrongkindofdoctor
committed
fixed some more compile-time errors
wrongkindofdoctor
committed
more compile-time debugging
wrongkindofdoctor
committed
Lots of compile-time debugging. The code now compiles with gnu7 in debug, and is ready for testing
wrongkindofdoctor
committed

Commits on Nov 19, 2019

Commits on Nov 20, 2019

Commits on Nov 21, 2019

put create_file back in MOM_io and modified the routine to use the new fms-io interfaces

wrongkindofdoctor
authored and
wrongkindofdoctor
committed

Commits on Nov 22, 2019

added new write_field interfaces to MOM_io.

wrongkindofdoctor
authored and
wrongkindofdoctor
committed

Commits on Nov 28, 2019

merge in latest dev/gfdl updates (#36 )
wrongkindofdoctor
authored
Merge branch 'dev/gfdl' into user/jml/operation_fms2_the_reckoning

wrongkindofdoctor
authored and
wrongkindofdoctor
committed
added new wrapper routines for field_size and field_exists to MOM-io

wrongkindofdoctor
authored and
wrongkindofdoctor
committed

Commits on Dec 3, 2019

Merge branch 'dev/gfdl' of git://github.com/NOAA-GFDL/MOM6 into NOAA-GFDL-dev/gfdl

wrongkindofdoctor
authored and
wrongkindofdoctor
committed
Merge branch 'NOAA-GFDL-dev/gfdl' into dev/gfdl

wrongkindofdoctor
authored and
wrongkindofdoctor
committed
Merge branch 'dev/gfdl' of github.com:wrongkindofdoctor/MOM6 into dev/gfdl

wrongkindofdoctor
authored and
wrongkindofdoctor
committed
removed unused calls to open_file and file_exists from driver files
wrongkindofdoctor
committed
fixed compile-time errors
wrongkindofdoctor
committed

Commits on Dec 4, 2019

more compile-time debugging
wrongkindofdoctor
committed

Commits on Dec 5, 2019

added scale option to write field
wrongkindofdoctor
committed

Commits on Dec 6, 2019

Merge in latest dev/gfdl updates (#40 )
wrongkindofdoctor
authored
fixed compile-time errors
wrongkindofdoctor
committed
Merge branch 'user/jml/operation_fms2_the_reckoning' of github.com:wrongkindofdoctor/MOM6 into user/jml/operation_fms2_the_reckoning
wrongkindofdoctor
committed
Merge branch 'dev/gfdl' into user/jml/operation_fms2_the_reckoning
wrongkindofdoctor
committed
whitespace cleanup
wrongkindofdoctor
committed

Commits on Dec 9, 2019

Merge branch 'dev/gfdl' into user/jml/operation_fms2_the_reckoning
wrongkindofdoctor
authored
fixed doxygen errors in interface definitions

wrongkindofdoctor
authored and
wrongkindofdoctor
committed
Merge pull request #41 from NOAA-GFDL/dev/gfdl
wrongkindofdoctor
authored
added checks to create_file that do a non-domain-decomposed read if the layout and io_layout are only on 1 pe
wrongkindofdoctor
committed

Commits on Dec 10, 2019

whitespace cleanup
wrongkindofdoctor
committed

Commits on Dec 11, 2019

added checks for G, dG, GV to create_file before setting axis data and metadata
wrongkindofdoctor
committed

Commits on Dec 12, 2019

removed unused 'fields' argument from create_file calls
wrongkindofdoctor
committed

Commits on Dec 13, 2019

whitespace cleanup
wrongkindofdoctor
committed

Commits on Dec 16, 2019

create_file updates
wrongkindofdoctor
committed

Commits on Dec 17, 2019

Merge pull request #42 from NOAA-GFDL/dev/gfdl
wrongkindofdoctor
authored
Merge branch 'dev/gfdl' into user/jml/operation_fms2_the_reckoning

wrongkindofdoctor
authored and
wrongkindofdoctor
committed
removed calls to write_field for non-dd interfaces in write_field_DD routines

wrongkindofdoctor
authored and
wrongkindofdoctor
committed
moved the calls to get_var_dimension_features and get_global_io_domain_indices out of conditional statement since the indices require the variable axis names
wrongkindofdoctor
committed

Commits on Jan 6, 2020

Merge pull request #43 from NOAA-GFDL/dev/gfdl
wrongkindofdoctor
authored
Merge branch 'dev/gfdl' into user/jml/operation_fms2_the_reckoning
wrongkindofdoctor
committed
removed the G argument from write_field calls for variables with vertical dimensions only
wrongkindofdoctor
committed

Commits on Jan 7, 2020

replaced var_exists with direct call to variable_exists in write_field_2d_dd still having problems with axes not being associated with the mpp domain
wrongkindofdoctor
committed

Commits on Jan 10, 2020

added read calls to write_field interfaces to get the domain-decomposed axis indices to pass to write_data, reorded the write calls and get_current_time calls
wrongkindofdoctor
committed

Commits on Jan 13, 2020

Commits on Jan 15, 2020

moved the energypath_nc definition out of the logic blockfor CS%previous calls because the write_field interface requires the filepath
wrongkindofdoctor
committed

Commits on Jan 16, 2020

Commits on Jan 17, 2020

Merge pull request #46 from NOAA-GFDL/dev/gfdl
wrongkindofdoctor
authored
(*)Set answers_2018 in calls to solve_linear_system

Hallberg-NOAA
authored and
wrongkindofdoctor
committed
+Added subroutine linear_solver to regrid_solvers

Hallberg-NOAA
authored and
wrongkindofdoctor
committed
(*)Simplified edge-value linear systems

Hallberg-NOAA
authored and
wrongkindofdoctor
committed
(*)Added explicit boundary edge value estimates

Hallberg-NOAA
authored and
wrongkindofdoctor
committed
(*)Got a test working with edge_slopes_implicit_h5

Hallberg-NOAA
authored and
wrongkindofdoctor
committed
(*)Improved algorithms in edge_slopes_implicit_h5

Hallberg-NOAA
authored and
wrongkindofdoctor
committed
Use linear_solver in edge_slopes_implicit_h5

Hallberg-NOAA
authored and
wrongkindofdoctor
committed
Made regrid_edge_slopes more concise

Hallberg-NOAA
authored and
wrongkindofdoctor
committed
+Merged regrid_edge_slopes into regrid_edge_values

Hallberg-NOAA
authored and
wrongkindofdoctor
committed
(*)Corrected a sign error in edge_slopes_implicit_h3

Hallberg-NOAA
authored and
wrongkindofdoctor
committed
(*)Use end_value_h4 in edge_slopes_implicit_h3

Hallberg-NOAA
authored and
wrongkindofdoctor
committed
(*)Limit fractional thicknesses in end_value_h4

Hallberg-NOAA
authored and
wrongkindofdoctor
committed
+Add optional argument tol to test_answers

Hallberg-NOAA
authored and
wrongkindofdoctor
committed
(*)Minor refactoring of edge_values_explicit_h4

Hallberg-NOAA
authored and
wrongkindofdoctor
committed
Added dOxygen comments to test_line

Hallberg-NOAA
authored and
wrongkindofdoctor
committed

Commits on Jan 22, 2020

still having nothing but problems with the tc3 layout domains. I think getting the compute indices should work, but there are size mismatches for the 1-pe IO case, and it requires getting rid of an…
wrongkindofdoctor
committed

Commits on Jan 24, 2020

defined the pelist for non-domain-decomposed writes in create_file to pass to open_file to fix issue with multiple pe's being identified as the root and attempting to open the file
wrongkindofdoctor
committed

Commits on Jan 27, 2020

changed TINY back to EPSILON to account for possible roundoff error in either direction
wrongkindofdoctor
committed

Commits on Jan 28, 2020

removed index routine calls--not needed w/new IO check. Need to add register_axis calls to remaining DD write_field calls, and add pe list calls to non-DD write_field routines
wrongkindofdoctor
committed

Commits on Jan 29, 2020

removed the unused THREADING option from create_file
wrongkindofdoctor
committed

Commits on Jan 30, 2020

still having nothing but problems with the tc3 layout domains. I think getting the compute indices should work, but there are size mismatches for the 1-pe IO case, and it requires getting rid of an…

wrongkindofdoctor
authored and
wrongkindofdoctor
committed
Merge branch 'dev/gfdl' into user/jml/operation_fms2_the_reckoning
wrongkindofdoctor
authored
removed call to flush_file from MOM_sum_output b/c no longer needed

wrongkindofdoctor
authored and
wrongkindofdoctor
committed
Merge branch 'user/jml/operation_fms2_the_reckoning' of github.com:wrongkindofdoctor/MOM6 into user/jml/operation_fms2_the_reckoning
wrongkindofdoctor
committed
white space cleanup
wrongkindofdoctor
committed

Commits on Jan 31, 2020

removed the start argument from the calls to read_data the map to the read_data_2d_noDD_diag_axes routine since entire array is read in before indexing
wrongkindofdoctor
committed

Commits on Feb 3, 2020

whitespace cleanup
wrongkindofdoctor
committed
Merge pull request #47 from NOAA-GFDL/dev/gfdl
wrongkindofdoctor
authored

Commits on Feb 6, 2020

reverted whitespace changes
wrongkindofdoctor
committed
changed FMS_COMMIT to master in .testing Makefile
wrongkindofdoctor
committed
changed time_level to timelevel in MOM_solo_driver MOM_read_data, MOM_read_vector calls
wrongkindofdoctor
committed

Commits on Feb 7, 2020

more whitespace cleanup
wrongkindofdoctor
committed
whitespace cleanup, fixed syntax error in MOM_grid_initialize
wrongkindofdoctor
committed
removed unsupported doxygen tag
wrongkindofdoctor
committed
switched time_level argument to timelevel in oda_driver
wrongkindofdoctor
committed
fixed write_field time_level argument in MOM_sum_output.F90
wrongkindofdoctor
committed
replaced get_cpu_affinity with fms_cpu_affinity_get and set_cpu_affinity with fms_cpu_affinity_set
wrongkindofdoctor
committed
added module use fms_affinity fms_affinity_set, fms_affinity_get to MOM_io. Added use statements for fms_affinity_set,get to MOM_driver and mom_domains
wrongkindofdoctor
committed

Commits on Feb 10, 2020

Merge pull request #48 from NOAA-GFDL/dev/gfdl
wrongkindofdoctor
authored
Merge branch 'dev/gfdl' into user/jml/operation_fms2_the_reckoning
wrongkindofdoctor
committed
added procedure to set the io_layout for 1-pe jobs to restore_state

wrongkindofdoctor
authored and
wrongkindofdoctor
committed
redirected get_affinity and set_affinity calls to routines in fms_affinity.F90, and fixed arguments in set_affinity

wrongkindofdoctor
authored and
wrongkindofdoctor
committed

Commits on Feb 11, 2020

Merge branch 'user/jml/operation_fms2_the_reckoning' of github.com:wrongkindofdoctor/MOM6 into user/jml/operation_fms2_the_reckoning
wrongkindofdoctor
committed
removed extraneous calls to open file from write_field_noDD routines, added pelist definition and argument to open_file call in write_scalar
wrongkindofdoctor
committed
declared variable i in write_scalar
wrongkindofdoctor
committed
removed extraneous open_file and pelist definition from write_field_1d_nodd
wrongkindofdoctor
committed
replaced get_cpu_affinity with fms_cpu_affinity_get and set_cpu_affinity with fms_cpu_affinity_set

wrongkindofdoctor
authored and
wrongkindofdoctor
committed
Added option to trailer.py for line length excluding comments

adcroft
authored and
wrongkindofdoctor
committed
Correct spelling in HI doxumentation

adcroft
authored and
wrongkindofdoctor
committed
Adds global extents to hor_index type

adcroft
authored and
wrongkindofdoctor
committed
Adds MOM_random module

adcroft
authored and
wrongkindofdoctor
committed
Segment index added for several OBC segment flags

marshallward
authored and
wrongkindofdoctor
committed
Replaced intent(out) with intent(inout) in EOS

adcroft
authored and
wrongkindofdoctor
committed
added procedure to set the io_layout for 1-pe jobs to restore_state

wrongkindofdoctor
authored and
wrongkindofdoctor
committed
Merge branch 'user/jml/operation_fms2_the_reckoning' of github.com:wrongkindofdoctor/MOM6 into user/jml/operation_fms2_the_reckoning
wrongkindofdoctor
committed
added check that sum of base_cpu and adder is at least 1 that passes the sum to set_affinity if true, and 1 if false to prevent the new fms_affinity_set routine from evaluating cpuset_sz as 0 for 1…
wrongkindofdoctor
committed
changed get_affinity back to get_cpu_affinity, set_affinity back to set_cpu_affinity

wrongkindofdoctor
authored and
wrongkindofdoctor
committed