| 4 |
Notes on tags used in MITgcmUV |
Notes on tags used in MITgcmUV |
| 5 |
============================== |
============================== |
| 6 |
|
|
| 7 |
|
o pkg/bbl: |
| 8 |
|
- fix call to FIND_RHO_2D in bbl_calc_rho.F (case multi-tiles per proc); |
| 9 |
|
update global_with_exf output.yearly.txt file. |
| 10 |
|
o pkg/cal: |
| 11 |
|
- start to clean-up cal_set.F ; fix for the case where baseTime is non-zero. |
| 12 |
|
- remove 2 unsused and very short (1 line only) functions ; |
| 13 |
|
- remove unsused subroutines from AD-Diff list. |
| 14 |
|
o model/src: |
| 15 |
|
- start to implement clean wait to stop in initialization S/R (config_check.F, |
| 16 |
|
packages_check.F and pkg/obcs/obcs_check.F + pkg/mypackage) |
| 17 |
|
o folowing PRINT_MESSAGE changes: |
| 18 |
|
- remove reminder multiple print to the same file (STDOUT) when multi-threads. |
| 19 |
|
o pkg/bbl: |
| 20 |
|
- modification so it can work with z* |
| 21 |
|
- replace deltaT with dTtracerLev(k) for asynchronous time stepping |
| 22 |
|
- global_with_exf/input.yearly verification; changes output.yearly.txt |
| 23 |
|
o eesupp/src: |
| 24 |
|
- PRINT_MESSAGE: use OpenMP critical instruction (similar to PRINT_ERROR); |
| 25 |
|
writing to unit 0 (if unit=errorMessageUnit) extended to the case myThid=1 |
| 26 |
|
- CHECK_THREADS: specific code if using OpenMP threading: improve checking of |
| 27 |
|
# of threads across processors (e.g., different env OMP_NUM_THREADS setting) |
| 28 |
|
- improve special preliminary call to EESET_PARMS (if eedata is read 2 times). |
| 29 |
|
- when using MPI, always set usingMPI=T ; set default accordingly and stop |
| 30 |
|
if not consistent. |
| 31 |
|
- ALL_PROC_DIE: add a BARRIER after MPI_FINALIZE call ; skip this when thread |
| 32 |
|
argument is zero (in case called within single-thread section). |
| 33 |
|
o pkg/fizhi: |
| 34 |
|
clean-up turbulence cold-start switch: decided in fizhi_init_vars.F, stored |
| 35 |
|
in common bloc (fizhi_coms.h) and then passed as argument up to S/R TURBIO. |
| 36 |
|
o pkg/exch2: |
| 37 |
|
- add argument "updateCorners" to S/R exch2_get_uv_bounds (enable to |
| 38 |
|
activate EXCH_IGNORE_CORNERS in vector EXCH S/R). |
| 39 |
|
- switch 1rst EXCH call to IGNORE_CORNERS (instead of UPDATE_CORNERS) |
| 40 |
|
but keep the 2nd with UPDATE_CORNERS: this prevents overwriting good data |
| 41 |
|
with bad ones coming from a not-yet-updated halo. |
| 42 |
|
This should fix the problem when sNx <> sNy and GCD(sNx,sNy) < 2*Olx. |
| 43 |
|
o pkg/frazil: |
| 44 |
|
move FrzForcT diag from frazil_diagnostics_state.F to frazil_calc_rhs.F |
| 45 |
|
http://mitgcm.org/pipermail/mitgcm-devel/2012-March/005384.html |
| 46 |
|
o model/src/config_check.F: add a check for OLx>sNx/OLy>sNy |
| 47 |
|
o verification/lab_sea |
| 48 |
|
* #undef SEAICE_EXCLUDE_FOR_EXACT_AD_TESTING |
| 49 |
|
* #define ALLOW_SST_CONTROL & ALLOW_SST_CONTROL |
| 50 |
|
* nTimeSteps 3 -> 4 |
| 51 |
|
* changed reference output accordingly |
| 52 |
|
o pkg/ctrl: |
| 53 |
|
* fix ctrl_get_gen for xx_sst, xx_sss |
| 54 |
|
o pkg/fizhi: |
| 55 |
|
- fix 2 calls to S/R GRD2MSC (argument igrd) in update_earth_exports.F |
| 56 |
|
update results of exp. fizhi-cs-32x32x40 (fail with only 6 digits for cg2d) |
| 57 |
|
- by default, use new version of S/R GETPWHERE (#define TRY_NEW_GETPWHERE); |
| 58 |
|
fix few out-of-bounds indices in referencing vars (initialise igrd to 1); |
| 59 |
|
clean-up fizhi_turb.F (remove unused & uninitialised qdiaglsm array). |
| 60 |
|
o pkg/gmredi: |
| 61 |
|
- add diagnostic of Redi (main diagnonal) vertical diffusive flux of Temp |
| 62 |
|
o pkg/fizhi: |
| 63 |
|
- fix few initialisation and multi-tile Pb: aqualev & gridalt now pass the |
| 64 |
|
(old)aces open64 tuv test. |
| 65 |
|
o pkg/icefront: fix bug and reorder loop in icefront_init_varia.F |
| 66 |
|
o pkg/mom_vecinv: |
| 67 |
|
- fix 1 loop range in mom_vecinv.F to be like all the other loops which |
| 68 |
|
set/update gu,gv/Diss (fix Pb of uninitialised fVerUkp in admom_vecinv |
| 69 |
|
with non-lin free-surf). |
| 70 |
|
o model/src: move k-loop outside of i/j-loops when (re-)computing R_low |
| 71 |
|
in ini/update_masks_etc.F in order to avoid -O3 optimization problems with |
| 72 |
|
some compilers (gfortran 4.6.0 on MacOS) |
| 73 |
|
o momentum: |
| 74 |
|
- separate fVer?(:,:,kUp) & fVer?(:,:,kDown) in argument list of MOM_FLUXFORM |
| 75 |
|
& MOM_VECINV subroutines (to help TAF). This fixes Pb of uninitialised |
| 76 |
|
fVer?(kUp) in admom_fluxform with non-lin free-surf or bottom control. |
| 77 |
|
|
| 78 |
|
checkpoint63k (2012/03/17) |
| 79 |
|
o model/src: |
| 80 |
|
- selectAddFluid (ALLOW_ADDFLUID): add missing term in horizontal momentum |
| 81 |
|
flux-form (mom_fluxform.F) and vertical momentum (calc_gw.F) equations. |
| 82 |
|
- SOLVE_DIAGONAL: switch default to use adjoinable & vectorized version |
| 83 |
|
(#undef SOLVE_DIAGONAL_LOWMEMORY); update output of exp. front_relax.bvp. |
| 84 |
|
o model/src: solve_*diagonal.F |
| 85 |
|
- replace division with multiplication by inverse (similar to original version) |
| 86 |
|
update output of AD exp. global_ocean.90x40x15. |
| 87 |
|
- fix pkg/gmredi/gmredi_calc_psi_bvp.F & pkg/ggl90/ggl90_calc.F |
| 88 |
|
to enable the use of the vectorized & differentiable solve_tridiagonal.F |
| 89 |
|
- test exp. global_ocean.90x40x15/input_ad: switch to advection scheme 20 for |
| 90 |
|
vertical advection (to get solve_tridiagonal tested in AD mode). |
| 91 |
|
o pkg/seaice: ECCO-CLIVAR merge: |
| 92 |
|
1. seaice AREA relaxation |
| 93 |
|
* CPP option: SEAICE_ALLOW_AREA_RELAXATION |
| 94 |
|
* runtime flag: SEAICE_tauAreaObsRelax |
| 95 |
|
2. modified global mean cost imbalance code: CPP option |
| 96 |
|
ALLOW_SSH_GLOBMEAN_COST_CONTRIBUTION |
| 97 |
|
o another iteration on solve_*diagonal.F |
| 98 |
|
Make choices independent of AUTODIFF |
| 99 |
|
Default will be the adjoinable vectorized version. |
| 100 |
|
Alternative choices are: |
| 101 |
|
* non-adjoinable original version with low memory footprint: |
| 102 |
|
#define SOLVE_DIAGONAL_LOWMEMORY |
| 103 |
|
* still adjoinable but does not vectorize (k as inner-most loop): |
| 104 |
|
#define SOLVE_DIAGONAL_KINNER (keeping SOLVE_DIAGONAL_LOWMEMORY undef) |
| 105 |
|
o testreport |
| 106 |
|
- add option to use a home made wrapper around "make" |
| 107 |
|
- remove MITGCM_IEEE env. variable (no longer active) |
| 108 |
|
o pkg/shelfice |
| 109 |
|
- improve computation of friction velocity for velocity dependent |
| 110 |
|
exchange coefficients: include velocities into SHELFICEBoundaryLayer |
| 111 |
|
block to average velocities vertically |
| 112 |
|
- add TAF store directives for SHI_ALLOW_GAMMAFRICT |
| 113 |
|
o Modify diagonal solvers, now three versions (upon request): |
| 114 |
|
#ifndef SOLVE_DIAGONAL_AUTODIFF |
| 115 |
|
1. default version (dating back a while ago) |
| 116 |
|
#else /* def SOLVE_DIAGONAL_AUTODIFF */ |
| 117 |
|
# ifdef SOLVE_DIAGONAL_KINNER |
| 118 |
|
2. adjoinable version with k-innermost loop (G. Forget) |
| 119 |
|
# else |
| 120 |
|
3. adjoinable vectorized version of 2. |
| 121 |
|
# endif |
| 122 |
|
#endif /* SOLVE_DIAGONAL_AUTODIFF */ |
| 123 |
|
All flags are currently set within routines, but could/should be moved to CPP-options. |
| 124 |
|
Array bounds for 1. are as before, for 2.,3. cover the entire domain |
| 125 |
|
(could be changed for 1. too). |
| 126 |
|
o verification/lab_sea (fwd): |
| 127 |
|
- update the 4 fwd lab_sea test experiment: no longer SEAICE_GROWTH_LEGACY, |
| 128 |
|
use 7 categories (except hb87), new solve4temp params (except salt_plume), |
| 129 |
|
constant salinity (=4) or variable salinity (salt_plume) done through |
| 130 |
|
SItracer. |
| 131 |
|
o pkg/seaice: |
| 132 |
|
- rename parameters: SIsalFrac to SEAICE_saltFrac & SIsal0 to SEAICE_salt0; |
| 133 |
|
- remove old parameters ( SEAICE_freeze, SEAICEturbFluxFormula, SEAICE_gamma_t, |
| 134 |
|
SEAICE_gamma_t_frz, SEAICE_availHeatTaper & SEAICE_availHeatFracFrz) |
| 135 |
|
from SEAICE_PARMS.h ; still need to fix seaice_cost_sst.F (broken when |
| 136 |
|
SEAICE_freeze got retired) but already better than previously with |
| 137 |
|
SEAICE_freeze=123456.7 deg.C; |
| 138 |
|
- move setting of facOpenGrow/Melt from SEAICE_INIT_FIXED to SEAICE_READPARMS |
| 139 |
|
(safer multi-threaded setting); always set SEAICEuseEVP; |
| 140 |
|
- setting of ocean-ice turb. flux coeff: moved from SEAICE_CHECK to |
| 141 |
|
SEAICE_READPARMS & SEAICE_INIT_FIXED (this fixes wrong summary report); |
| 142 |
|
stop if multiple specifications for the same coeff; make sure default |
| 143 |
|
SEAICE_mcPheePiston is compatible with drF(1) & deltaT. |
| 144 |
|
o pkg/obcs: |
| 145 |
|
- allow to switch to upwind 1rst order advection scheme for the advective |
| 146 |
|
flux computation at the open-boundary; disabled for adjoint built. |
| 147 |
|
o pkg/ptracers: |
| 148 |
|
- rename PTRACERS_RESTART.h -> PTRACERS_START.h |
| 149 |
|
- rename ptracers_fields_blocking_exchanges.F -> ptracers_fields_blocking_exch.F |
| 150 |
|
- add parameter to decide when to start a tracer; use internal flag to |
| 151 |
|
know which tracer to step-forward; switch this internal flag at the top |
| 152 |
|
of forward_step.F (new S/R PTRACERS_SWITCH_ONOFF); for now, not enabled |
| 153 |
|
for adjoint built. |
| 154 |
|
o pkg/seaice: |
| 155 |
|
- add SEAICE_SIZE.h anywhere in the code where SEAICE.h is included ; |
| 156 |
|
- move MULTDIM declaration and setting from SEAICE.h to SEAICE_SIZE.h ; |
| 157 |
|
- keep experiment offline_exf_seaice truly single category (set MULTDIM=1). |
| 158 |
|
o pkg/ecco: |
| 159 |
|
- fix writing of output file DiagnosCost_ClimSalt/Theta |
| 160 |
|
and DiagnosCost_DriftSalt/Theta (if multi-tiles per proc). |
| 161 |
|
o pkg/seaice: |
| 162 |
|
- add missing initialisation of COR_ICE array in dynsolver.F |
| 163 |
|
- merging SEAICE_MULTICATEGORY define and undef cases. Now the same code is |
| 164 |
|
compiled either way, and is tested in all exps (multidim was only tested |
| 165 |
|
in lab_sea ad before). |
| 166 |
|
- cosmetic change to prepare for future addition of ITD. no change to results. |
| 167 |
|
no defaults change. |
| 168 |
|
- added run time param SEAICE_multDim (7 for SEAICE_MULTICATEGORY, 1 otherwise). |
| 169 |
|
- by lack of adequate verification experiments, I further used custom versions |
| 170 |
|
of global_ocean.cs32x15 to test multicat or not, with pickup or not. All |
| 171 |
|
should be correct. |
| 172 |
|
- avoid recomputations in ad. |
| 173 |
o array initialisation in adjoint mode: |
o array initialisation in adjoint mode: |
| 174 |
1) pkg/mom_common/mom_calc_ke.F: |
1) pkg/mom_common/mom_calc_ke.F: |
| 175 |
add full array initialisation of KE inside mom_calc_ke.F (ALLOW_AUTODIFF_TAMC): |
add full array initialisation of KE inside mom_calc_ke.F (ALLOW_AUTODIFF_TAMC): |
Parent Directory
| 
Revision Log
| 
Revision Graph
| 
Patch