4 |
Notes on tags used in MITgcmUV |
Notes on tags used in MITgcmUV |
5 |
============================== |
============================== |
6 |
|
|
7 |
|
o verification/matrix_example: |
8 |
|
- split domain in 8 tiles (2x4); add SIZE.h_mpi and eedata.mth for |
9 |
|
MPI and mutlti-threaded testing. |
10 |
|
o pkg/matrix: |
11 |
|
- change name of S/R MATRIX_INIT to S/R MATRIX_INIT_VARIA ; |
12 |
|
- fix for multi-tiles (add bi,bj indices to exp/impMatrixCounter); |
13 |
|
- fix for multi-threaded (remove DATA statement); |
14 |
|
- use standard DIFFERENT_MULTIPLE function; |
15 |
|
- add MATRIX_OPTIONS.h |
16 |
|
o verification/matrix_example: |
17 |
|
To allow to split the domain in several tiles, change the domain size |
18 |
|
from 31x31 to 32x32 by adding 1 row (on Southern side) + 1 column (on |
19 |
|
Western side) of land points. |
20 |
|
o (*_)write_pickup: |
21 |
|
- fix few bugs in mypackage_read/write_pickup.F |
22 |
|
- in MDS_WR_METAFILES calls, replace "myTime" by a local copy into |
23 |
|
array of length 1 (was caught by compiler syntax-check). |
24 |
|
o pkg/diagnostics (velocity potential solver): |
25 |
|
- fix initialisation of main diagonal matrix (for overlap values) |
26 |
|
o tools/build_options: |
27 |
|
- towards single optfile for MPI and non-MPI built: merge few +mpi optfile |
28 |
|
into previously non MPI one (linux_am64 pgf77,open64,pathf90,ifort11) |
29 |
|
- linux_amd64_gfortran: merge +mpi_generic version into standard optfile: |
30 |
|
* move IEEE flags to DEVEL section; take IEEE flags from +mpi_generic ; |
31 |
|
* FAST case: for now, keep -O3 with ini_masks_etc.F in NOOPTFILES list; |
32 |
|
and use "-devel" testreport option for standard testing on baudelaire. |
33 |
|
o pkg/grdchk: |
34 |
|
- Assume that all gradient-check position are within the same tile |
35 |
|
corresponding to global indices (iGloTile,jGloTile) and that |
36 |
|
positions (i,j indices or nbeg) are local to this tile; |
37 |
|
- set tile local-indices and proc. number according to tile global-indices; |
38 |
|
- remove grdchkwhichproc from namelist (retired param); |
39 |
|
- send adjoint gradient to all procs so that all procs print it to STDOUT; |
40 |
|
- switch on MPI test for global_ocean.90x40x15 AD test experiment. |
41 |
|
o adding package bbl (Bottom Boundary Layer) |
42 |
|
description in MITgcm/pkg/bbl/bbl_description.tex |
43 |
|
example/test experiment in MITgcm_contrib/bbl |
44 |
|
o adjoint verifications |
45 |
|
- changed 5 experiments to multi-tiles |
46 |
|
* global_ocean.90x40x15 |
47 |
|
* obcs_ctrl |
48 |
|
* offline_exf_seaice |
49 |
|
* tutorial_dic_adjoffline |
50 |
|
* tutorial_tracer_adjsens |
51 |
|
changed reference output for those that crossed tile boundaries |
52 |
|
(changed choices of gradient check positions) |
53 |
|
- with same tiling, switch on MPI test for the last 4 (add SIZE.h_mpi) |
54 |
|
- temporary bug fix in grdchk_loc.F for obcs |
55 |
|
o genmake2 & testreport |
56 |
|
- add target "obj" (just for OBJFILES files) to Makefile; |
57 |
|
- add option to specify FEXTRAFLAGS on command line. |
58 |
|
o pkg/mypackage: |
59 |
|
- remove bi,bj arguments from argument list of S/R MYPACKAGE_CALC_RHS |
60 |
|
|
61 |
|
checkpoint63a (2011/08/02) |
62 |
|
o pkg/diagnostics (velocity potential): |
63 |
|
- fix velocity-potential calculation when using OBCS: balance net inflow |
64 |
|
at each level (new S/R OBCS_DIAG_BALANCE). |
65 |
|
- fix some of the problems with OBCS in stream-function calculation. |
66 |
|
o pkg/exf/exf_readparms.F |
67 |
|
- make treatment of optional namelists (EXF_NML_04, EXF_NML_SGRUNOFF, |
68 |
|
EXF_NML_OBCS) more consistent, so that you need not have these lists |
69 |
|
in data.exf when they are not used. |
70 |
|
o optim/optim_readdata.F: |
71 |
|
- add some output about obcs variables that might help debugging |
72 |
|
- add a check about nvarlength |
73 |
|
o pkg/ctrl: once again, fix computation of endrec in s/r ctrl_init_rec |
74 |
|
o pkg/diagnostics (velocity potential): |
75 |
|
- fix mask for OBCS (still problems in stream-function with OBCS); |
76 |
|
- add specific parameter (default = main code CG2D params) for solver; |
77 |
|
- in case of poor convergence, use solution corresponding to lowest residual. |
78 |
|
o pkg/seaice/seaice_growth.F: small changes that permit vectorizable |
79 |
|
adjoint code. In general statements like tmpscal=fld(i,j) are problematic |
80 |
|
and should be avoided where possible. |
81 |
|
o model/src (rStar): |
82 |
|
- include r* effect on reference-state (PhiRef) in totPhiHyd |
83 |
|
(so that it's really the potential anomaly at the cell center); |
84 |
|
affect solution only if z* with full pressure in EOS. |
85 |
|
- test-exp global_ocean.90x40x15: reduce viscA4 (was not stable) and |
86 |
|
update output (because of z* in totPhiHyd and viscA4 changes); |
87 |
|
- keep the previous diagnostics as "PHIHYDcR" (closer to potential anomaly |
88 |
|
@ constant r) until interpolation at constant r is implemented. |
89 |
|
o pkg/diagnostics |
90 |
|
- fix RCENTER diagnostics in case cell center is not at interface middle |
91 |
|
o add code for TEOS-10 (www.teos-10.org, McDougall et al. 2011). Use |
92 |
|
this eos with eosType = 'TEOS10', in data (PARM01). This eos implies |
93 |
|
that THETA and SALT are "conservative temperature" and "absolute |
94 |
|
salinity" |
95 |
|
o pkg/ctrl |
96 |
|
- fix computation of endrec in s/r ctrl_init_rec |
97 |
|
- ctrl_pack/unpack_*: remove unused variables and some obsolete code |
98 |
|
o pkg/seaice/seaice_growth.F |
99 |
|
- fix recomputation with FENTY_AREA_EXPANSION_CONTRACTION code, |
100 |
|
remove then obsolete STORE directives |
101 |
|
- fix STORE directives for tice and tices, so that only the |
102 |
|
appropriate part of the fields are stored (and not the entire |
103 |
|
field); update global_ocean.cs32x15.seaice AD output. |
104 |
|
- cosmetic changes: adjust CPP flags and indentation for better |
105 |
|
legibilty |
106 |
|
o pkg/autodiff: add mnc-output of adBottomDragFld (code contributed by |
107 |
|
Chris Wilson) |
108 |
|
o pkg/gmredi: |
109 |
|
Add Sub-Meso Eddies parameterisation from Baylor (for now, disabled with TAF), |
110 |
|
and test it in verification exp. "front_relax.bvp" (update output.bvp.txt) |
111 |
|
|
112 |
|
checkpoint63 (2011/07/11) |
113 |
o pkg/exch2: |
o pkg/exch2: |
114 |
- rename+move: nTiles in W2_EXCH2_PARAMS.h --> exch2_nTiles in W2_EXCH2_TOPOLOGY.h |
- rename+move: nTiles in W2_EXCH2_PARAMS.h --> exch2_nTiles in W2_EXCH2_TOPOLOGY.h |
115 |
- exch2 version of Cumul-Sum routine: S/R W2_SET_MAP_CUMSUM set up the |
- exch2 version of Cumul-Sum routine: S/R W2_SET_MAP_CUMSUM set up the |