/[MITgcm]/MITgcm/doc/tag-index
ViewVC logotype

Diff of /MITgcm/doc/tag-index

Parent Directory Parent Directory | Revision Log Revision Log | View Revision Graph Revision Graph | View Patch Patch

revision 1.1950 by dimitri, Thu Dec 19 01:06:09 2013 UTC revision 1.1960 by jmc, Sat Jan 25 18:52:38 2014 UTC
# Line 4  $Name$ Line 4  $Name$
4      Notes on tags used in MITgcmUV      Notes on tags used in MITgcmUV
5      ==============================      ==============================
6    
7  o useSingleCpuInput separates single-CPU input from single-CPU output  o genmake2 & testreport:
8  o pkg/seaice/seaice_lsr.F: fix an initialisation bug for the    - add genmake2 option "-oad" to generate a Makefile for OpenAD built and
9    SEAICE_VECTORIZE_LSR_ZEBRA option that lead to NaNs on a NEC-SX8      in this case, use tools/adjoint_options/adjoint_oad as default AD optfile.
10        Update testreport accordingly. No need to set "export OPENAD=true"
11        in adjoint_options/adjoint_oad anymore (will be removed later).
12    o pkg/shelfice:
13      - put T & S shelfice forcing into model surfaceForcing_T,_S (for now,
14        only if SHELFICEboundaryLayer=F); Change model/src/external_forcing.F
15        so that surface forcing applies to k=kSurf if using pkg/shelfice.
16    o pkg/ctrl, pkg/grdchk, pkg/ecco:
17      - Changes to make GENARR more modular (i.e. invisible if not used)
18    o model/src/temp_integrate.F and salt_integrate.F
19      - re-include store directives that were removed on Dec 27th.
20    o model/src/external_forcing_surf.F
21      - add storage directives for EmPmR and PmEpR to avoid recomputation of
22        seaice_model in the adjoint that is problematic when seaice_model is
23        turned off in the adjoint. The resolved issue dated back to July when
24        EXTERNAL_FORCING_SURF call was moved.
25        Update global_ocean.cs32x15/*/output_adm.seaice_dynmix.txt accordingly.
26    o pkg/exf/exf_check.F
27      - fix a small bug again (originally found by Wentao Liu, now found again)
28      - streamline error output
29    o pkg/streamice and aim_v23:
30      - replace a few "_EXCH_XY_RL (" with "_EXCH_XY_RL(", so that
31        Apple's /usr/bin/cpp knows what to do:
32        Apple LLVM version 5.0 (clang-500.2.79) (based on LLVM 3.3svn)
33        Target: x86_64-apple-darwin13.0.0
34    
35    checkpoint64s (2014/01/05)
36    o tools/adjoint_options:
37      - switch to fastopt.net TAF server (instead of default fastopt.de) for
38        AD optfiles adjoint_diva & adjoint_f95 (similar to adjoint_default).
39    o model/src:
40      - remove some unnecessary TAF storage directives in forward_step.F and
41        in dynamics.F (note: not always removed by TAF, e.g., double storage
42        of kappaRU,kappaRV in former version of dynamics.F).
43    o model/src & pkg/ptracers:
44      - call directly GAD_IMPLICIT_R & IMPLDIFF + DWNSLP_APPLY from temp_integrate.F
45        salt_integrate.F and ptracers_integrate.F (instead of from thermodynamics.F)
46        and remove ptracers_dwnslp_apply.F & ptracers_implicit.F
47      - call MATRIX_STORE_TENDENCY_EXP for each tracer (in ptracers_integrate.F)
48      - add fZon & fMer as output argument of S/R GAD_CALC_RHS.
49      - update TAF AD-output of experiments global_ocean.90x40x15 (13 for AD-grad).
50    o useSingleCpuInput separates single-CPU input from single-CPU output;
51        for now only implemented for in exf_interp_read.F
52    o verification/advect_xy:
53      - fix missing EXCH call in local version of ini_salt.F and update output.
54    o momentum diagnostics:
55      - fix which deltaT is used in Total-Tendency diagnostics.
56      - fix typo in diagnostics for momentum tendency from AB.
57    o pkg/seaice/seaice_lsr.F: fix an initialisation bug for the
58        SEAICE_VECTORIZE_LSR_ZEBRA option that lead to NaNs on a NEC-SX8
59    
60  checkpoint64r (2013/12/08)  checkpoint64r (2013/12/08)
61  o verification/*/results:  o verification/*/results:
# Line 112  o verification: Line 161  o verification:
161  o pkg/cost:  o pkg/cost:
162    - define a "final cost function" for each tile, and use GLOBAL_SUM_TILE_RL    - define a "final cost function" for each tile, and use GLOBAL_SUM_TILE_RL
163      (instead of _GLOBAL_SUM_RL) to get the final cost from all procs & threads:      (instead of _GLOBAL_SUM_RL) to get the final cost from all procs & threads:
164      o this fixes global-sum value if multi-threads (since GLOBAL_SUM_TILE_RL      o this fixes global-sum value if multi-threads (since GLOBAL_SUM_RL
165        cannot be applied to var in common block).        cannot be applied to var in common block).
166      o allows to get results independent of tile-per-proc setting (if using      o allows to get results independent of tile-per-proc setting (if using
167        slower GLOBAL_SUM_SEND_RECV option).        slower GLOBAL_SUM_SEND_RECV option).
Legend:
Removed from v.1.1950  
changed lines
  Added in v.1.1960
  ViewVC Help
Powered by ViewVC 1.1.22