| 3 |
Notes on tags used in MITgcmUV |
Notes on tags used in MITgcmUV |
| 4 |
============================== |
============================== |
| 5 |
|
|
| 6 |
|
o Fixing OPTIONS files for use of exf: |
| 7 |
|
- ALLOW_ECCO not needed |
| 8 |
|
- ECCO_CPPOPTIONS.h should not be included from |
| 9 |
|
EXF_OPTIONS.h/CAL_OPTIONS.h, but from CPP_OPTIONS.h |
| 10 |
|
as has been the case in the past |
| 11 |
|
- All cal/exf related options are currently set in |
| 12 |
|
single header file. Should remain so until we |
| 13 |
|
agree on a consistent new policy. |
| 14 |
|
o Removed all CPP options for cal package. |
| 15 |
|
cal has no more CPP options. |
| 16 |
|
|
| 17 |
|
checkpoint52b_post |
| 18 |
|
o new version: "thsice" of thermodynamic sea-ice pkg: |
| 19 |
|
- only 1 call from "forward_step"; conserve energy, fresh-water & salt. |
| 20 |
|
- modify fluxes but do not change theta directly; |
| 21 |
|
- light cleaning of bulk_force pkg (still needed with thsice). |
| 22 |
|
- add forcing fields + data files in global_ocean.cs32x15/inp_thsice |
| 23 |
|
+ output_thsice.txt to test thsice & bulk_force pkgs. |
| 24 |
|
o put back grid & coriolis monitor output. |
| 25 |
|
checkpoint52b_pre |
| 26 |
|
|
| 27 |
|
o call "do_the_model_io" at the beginning (=> dump initial state) |
| 28 |
|
o use rhoConstFresh to define convertEmP2rUnit (was done only in ocean-P) |
| 29 |
|
o sort out interplay between tamc.h and PTRACERS_OPTIONS.h |
| 30 |
|
(need info on NUMBER_OF_PTRACERS) |
| 31 |
|
o missing PTRACERS headers in some routines |
| 32 |
|
o default tamc.h in pkg/autodiff supposed to crash |
| 33 |
|
to ensure proper customization |
| 34 |
|
|
| 35 |
|
checkpoint52a_post |
| 36 |
|
o modifications to make FREEZE (_SURFACE) flux visible to pkg/kpp |
| 37 |
|
- moved surfaceTendencyTice from pkg/seaice to main code |
| 38 |
|
- subroutine FREEZE_SURFACE now limits only surface temperature |
| 39 |
|
and is called in FORWARD_STEP (if allowFreezing=.TRUE.) |
| 40 |
|
=> new output.txt for global_ocean.cs32x15 & global_with_exf |
| 41 |
|
- keep the old version of FREEZE for backward compatibility; |
| 42 |
|
global_ocean.90x40x15 left unchanged using useOldFreezing=.TRUE. |
| 43 |
|
- corresponding modifs for TAF directives |
| 44 |
|
o added surface flux output variables to TIMEAVE_STATVARS |
| 45 |
|
o time-averaged output for pkg/ptracers |
| 46 |
|
o fix problem with ECCO-related divided adjoint |
| 47 |
|
snapshot file was missing some ad-variables, |
| 48 |
|
but no warning issued by TAF |
| 49 |
|
|
| 50 |
|
checkpoint52a_pre |
| 51 |
|
o some bug fixes for #undef REAL4_IS_SLOW |
| 52 |
|
|
| 53 |
|
checkpoint52 = ecco_c52_e35 |
| 54 |
|
o after testing adjoints of checkpoint51u_post |
| 55 |
|
|
| 56 |
|
checkpoint51u_post |
| 57 |
|
o this is the post-merge tag of the ecco-branch |
| 58 |
|
(needs another clean checkout and re-testing) |
| 59 |
|
o various natl_box_adjoint configs and results |
| 60 |
|
but no input fields |
| 61 |
|
o obcs adjoint not yet merged! |
| 62 |
|
o adding ECCO global 1x1 and 2x2 degree configs |
| 63 |
|
o updated/extended some store directives for seaice |
| 64 |
|
o pkg/ecco now containes ecco-specific part of cost function |
| 65 |
|
o top level routines the_main_loop, forward_step |
| 66 |
|
supersede those in model/src/ |
| 67 |
|
previous input data.cost now in data.ecco |
| 68 |
|
(new namelist ecco_cost_nml) |
| 69 |
|
o in ctrl/ cleaned some cross-dependencies and updated CPP options |
| 70 |
|
|
| 71 |
|
checkpoint51t_post |
| 72 |
|
o this is the pre-merge tag of the ecco-branch |
| 73 |
|
|
| 74 |
|
checkpoint51s_post |
| 75 |
|
o cleanup: removed various obsolete #define-s |
| 76 |
|
o pkg/grdchk |
| 77 |
|
- renamed 'ph-grd 3' to 'grad-res' |
| 78 |
|
- merged from ecco-branch |
| 79 |
|
(remaining bug fixes for obcs gradient checks) |
| 80 |
|
- additional high-precision output for testreport |
| 81 |
|
(grep for 'precision_grdchk_result') |
| 82 |
|
|
| 83 |
|
checkpoint51r_post |
| 84 |
|
o cleaned up or removed CPP_OPTIONS.h files |
| 85 |
|
o updated results for some of the tests to reflect the new |
| 86 |
|
optfiles (-ieee: -O0 instead of -O3) |
| 87 |
|
|
| 88 |
|
checkpoint51q_post |
| 89 |
|
o cd_code: |
| 90 |
|
- moved cd_scheme.F -> cd_code_scheme.F |
| 91 |
|
- separate read_checkpoint from cd_code_ini_vars.F |
| 92 |
|
- separated cd_code part from write_checkpoint |
| 93 |
|
- updated AD_SOURCE, generated .flow |
| 94 |
|
- added CD_CODE_VARS.h to the_main_loop |
| 95 |
|
- Moving cd_code specific store dir.s to pkg/cd_code |
| 96 |
|
o ctrl package totally restructured |
| 97 |
|
- pack/unpack now optional and decoupled from |
| 98 |
|
xx_/adxx_ I/O |
| 99 |
|
- ctrl_pack/unpack cleaned |
| 100 |
|
(new routines ctrl_init_ctrlvar.F, pkg/ctrl/ctrl_init_wet.F) |
| 101 |
|
- confined inclusion of AD_CONFIG.h to where necessary. |
| 102 |
|
o genmake2: changed AD dependencies from SRCFILES to AD_FILES |
| 103 |
|
|
| 104 |
|
checkpoint51p_post |
| 105 |
|
o create initial cd_code package |
| 106 |
|
- WARNING: the initialization of fields within packages |
| 107 |
|
is broken because INI_FIELDS (where READ_CHECKPOINT is |
| 108 |
|
called) is called *before* PACKAGES_INIT_VARIABLES. |
| 109 |
|
This seems to be backwards since |
| 110 |
|
1) its important to initially zero the entire field including |
| 111 |
|
the "halo" region |
| 112 |
|
2) zeroing needs to be done *before* reading pickup info |
| 113 |
|
and the pickup call might or might not happen in any |
| 114 |
|
particular run |
| 115 |
|
- this needs an overhaul for R2 |
| 116 |
|
o cleanup verification/*/code/GMREDI_OPTIONS.h |
| 117 |
|
o cleanup testreport |
| 118 |
|
|
| 119 |
|
checkpoint51o_post |
| 120 |
|
o bringing AD-related verif's up to date with genmake2 |
| 121 |
|
- carbon/ |
| 122 |
|
- global_ocean.90x40x15/ |
| 123 |
|
- hs.94.1x64x5/ |
| 124 |
|
- (front_relax/ ) |
| 125 |
|
o (more) consistent directory structure for AD-related verif.'s |
| 126 |
|
o cleaning ALLOW_GRADIENT_CHECK -> ALLOW_GRDCHK |
| 127 |
|
o cleaning some ALLOW_TANGENTLINEAR_RUN -> ALLOW_AUTODIFF |
| 128 |
|
o bug fix in find_alpha.F for MDJWF: |
| 129 |
|
- modif. to alpha = 1/D*( dN/dT - rho*dD/Dt) to account for |
| 130 |
|
change rho -> rho-rhoConst |
| 131 |
|
- replace call find_rho to find_rhonum |
| 132 |
|
o initialisation of rFlx extended to full array (required by TAF) |
| 133 |
|
and shifted to thermodynamics |
| 134 |
|
o removed PTRACERS.h in ptracers routine |
| 135 |
|
o added surfacetendencyPtr to S/R parameter list pracers_forcing |
| 136 |
|
o Correcting initialisations for fVerU, fVerV. |
| 137 |
|
o helping TAF to recognize dependency on kUp for |
| 138 |
|
fVerU, fVerV |
| 139 |
|
o Replacing delZ by delR in pkg/ctrl/ |
| 140 |
|
checkpoint51o_pre |
| 141 |
|
|
| 142 |
|
o undid all of cp51 pending some cleanups and discussion |
| 143 |
|
o added the [#include "AD_CONFIG.h"] statement to all files that need |
| 144 |
|
it for adjoint/tl #defines |
| 145 |
|
o re-worked the build logic in genmake2 to support AD_CONFIG.h |
| 146 |
|
o removed tools/genmake since it no longer works |
| 147 |
|
|
| 148 |
checkpoint51n_post |
checkpoint51n_post |
| 149 |
o modifications to make FREEZE flux visible to pkg/kpp |
o modifications to make FREEZE flux visible to pkg/kpp |
Parent Directory
| 
Revision Log
| 
Revision Graph
| 
Patch