| 4 |
Notes on tags used in MITgcmUV |
Notes on tags used in MITgcmUV |
| 5 |
============================== |
============================== |
| 6 |
|
|
| 7 |
|
o make gluemnc/xplodemnc work with (moron) non-GNU shell tools |
| 8 |
|
|
| 9 |
|
checkpoint63d (2011/10/25) |
| 10 |
|
o pkg/obcs + seaice-dynamics: |
| 11 |
|
- new implementation calls OBCS_APPLY_UVICE (in seaice_dynsolver.F) before |
| 12 |
|
solving for uIce,vIce and prevent SEAICE_LSR or SEAICE_EVP to modify OB |
| 13 |
|
values (consistent with an implicit method). |
| 14 |
|
- add option OBCS_UVICE_OLD (in OBCS_OPTIONS.h, defined by default) to |
| 15 |
|
return to older implementation (same results as before); note that none of |
| 16 |
|
the "experimental" OBCS_SEAICE options are implemented in the new version. |
| 17 |
|
- switch test exp. seaice_obcs to use new version (update output). |
| 18 |
|
o pkg/seaice: |
| 19 |
|
- add few stop (B-grid & CS-exch, B-grid & obcs, AUTODIFF & SEAICE_maskRHS) |
| 20 |
|
- add parenthesis in Water-Ice drag calculations (evp & lsr); |
| 21 |
|
- in seaice_lsr.F, simplify and fix AA3 setting in case sNx or sNy=1; |
| 22 |
|
and fix case SEAICE_waterDrag=0 (but unstable if no or little seaice) |
| 23 |
|
- update output of test exp lab_sea that uses evp or lsr (fwd+ad) |
| 24 |
|
and global_ocean.cs32x15 (icedyn & seaice fwd+ad). |
| 25 |
|
o pkg/layers: |
| 26 |
|
- Include potential density as new coordinate (Thanks to David Munday) |
| 27 |
|
o pkg/dic: |
| 28 |
|
- Remove subroutine CALC_PCO2_APPROX_CO3 from carbon_chem.F and add |
| 29 |
|
carbonate computation/output to CALC_PCO2_APPROX |
| 30 |
|
o pkg/generic_advdiff |
| 31 |
|
- make TAF generate code that vectorizes: |
| 32 |
|
gad_dst3fl_adv_*.F, gad_u3c4_impl_r.F |
| 33 |
|
o model/src: |
| 34 |
|
- improve vectorization also for adjoint code of |
| 35 |
|
solve_tridiagonal and solve_pentadiagonal. As this requires up to |
| 36 |
|
5 new 3D-fields, this code is only turned on for the combination of |
| 37 |
|
ALLOW_AUTODIFF_TAMC and TARGET_NEC_SX. Also remove error message |
| 38 |
|
for singular matrix check (only for TARGET_NEC_SX). |
| 39 |
|
- remove hard-wiring of CPP-flag ALLOW_SOLVERS_KLOOPINSIDE to |
| 40 |
|
ALLOW_AUTODIFF. This flag now needs to be set outside of |
| 41 |
|
solve_tri/pentadiagonal.F |
| 42 |
|
o pkg/exf: |
| 43 |
|
- fix previous modif in exf_readparms.F (revision 1.66: treatment of |
| 44 |
|
optional namelist EXF_NML_OBCS). |
| 45 |
|
- add many debug-print (to trace down where pkg/cal error msg comes from) |
| 46 |
|
o pkg/mom_vecinv/mom_vi_hdissip.F |
| 47 |
|
- rearrange loops so that vectorization is improved and in particular |
| 48 |
|
the adjoint routine vectorizes. The code is a little longer now, but I |
| 49 |
|
think easier to read. |
| 50 |
|
Note: CPP-flag MOM_VI_ORIGINAL_VISCA4 is not tested. |
| 51 |
|
o pkg/seaice: |
| 52 |
|
- allow to specify initial seaice velocity from files (uIceFile,vIceFile) |
| 53 |
|
o pkg/exch2: |
| 54 |
|
- alternative code (without tile x tile matrix) for cumulated (global) sum; |
| 55 |
|
used as default; keep using matrix version in adjustment.cs-32x32x1. |
| 56 |
|
o tools/build_options: |
| 57 |
|
- after merging some MPI and/or multi-threads specific optfiles so that |
| 58 |
|
the modified default optfile support both built (using genmake2 -mpi |
| 59 |
|
for the MPI built, using genmake2 -omp for OpenMP built), |
| 60 |
|
move now obsolete optfiles to tools/build_options/unsupported/ |
| 61 |
|
(add copy of email sent to support list in doc/optfiles_changes.txt) |
| 62 |
|
|
| 63 |
|
checkpoint63c (2011/09/28) |
| 64 |
|
o pkg/ebm: |
| 65 |
|
- fix ebm_zonalmean.F for several procs in Ydir (nPy > 1); update output of |
| 66 |
|
test exp. global_ocean_ebm (affects machine truncation, 12 digits for cg2d) |
| 67 |
|
o pkg/seaice (SItracer pickup) |
| 68 |
|
* Simple code for pickup of SItracer fields (#define ALLOW_SITRACER) |
| 69 |
|
To be done: |
| 70 |
|
- thorough testing of missing fields in S/R seaice_read_pickup |
| 71 |
|
- thorough matching between generic tracers and specific fields used |
| 72 |
|
(e.g. volume-weighted age, area-weighted age, salinity, ...) |
| 73 |
|
o pkg/grdchk: |
| 74 |
|
- all procs calculate and print finite-difference gradient (for testreport) |
| 75 |
|
- change description output used by testreport; |
| 76 |
|
- AD testing: also compare and report Finite-difference gradient; |
| 77 |
|
=> update all output_adm*.txt |
| 78 |
|
o test exp. global_ocean_ebm: |
| 79 |
|
- switch back useRealFreshWaterFlux (no problem anymore with run-off); |
| 80 |
|
and turn on exactConserve; update (fwd) output. |
| 81 |
|
o pkg/seaice (LSR): |
| 82 |
|
- print to STDOUT number of solver iterations at SEAICE_monFreq interval. |
| 83 |
|
- modify solver to deal cleanly with u/v switch at face edges (CS-grid): |
| 84 |
|
* identical results with non-CS topology (but save up to half of EXCH calls) |
| 85 |
|
* no much changes in cs-32 verification tests since no ice at face edges |
| 86 |
|
(differences due to convergence criteria, since nb of iterations changed) |
| 87 |
|
* expect some differences on lat-lon-cap grid. |
| 88 |
|
update FWD results of global_ocean.cs32x15 test exp. (icedyn & seaice) |
| 89 |
o pkg/exch2: |
o pkg/exch2: |
| 90 |
- add new CPP option to disable Cumul-Sum matrix (to save storage when using |
- add new CPP option to disable Cumul-Sum matrix (to save storage when using |
| 91 |
many tiles); until alternative coding is added, just add a stop. |
many tiles); until alternative coding is added, just add a stop. |
Parent Directory
| 
Revision Log
| 
Revision Graph
| 
Patch