4 |
Notes on tags used in MITgcmUV |
Notes on tags used in MITgcmUV |
5 |
============================== |
============================== |
6 |
|
|
7 |
|
o pkg/mom_common: |
8 |
|
Move computation of length scales to mom_init_fixed.F (avoid recomputation |
9 |
|
at each time-steps*levels). |
10 |
|
o pkg/layers: |
11 |
|
Created a new package for computing volume fluxes in temperature or |
12 |
|
[not yet implemented] density layers, e.g. for calculating residual |
13 |
|
overturning circulations. |
14 |
|
o pkg/diagnostics: |
15 |
|
implement a RS type version for the set of DIAGNOSTICS-FILL subroutines |
16 |
|
o fix few RS/RL type mismatch (check with ifort 11.1 "-warn all" option) |
17 |
|
o pkg/mdsio (& pkg/rw): rework high level S/R interface: |
18 |
|
To avoid mixing type (RS/RL) of input/output array argument, |
19 |
|
replace single mixed-type array with a pair from each type (RS/RL). |
20 |
|
o setting pickupSuff force to start from a pickup (even if nIter0=0): this |
21 |
|
is now implemented for all pkgs (previously only done for the main pickup) |
22 |
|
|
23 |
|
checkpoint61u |
24 |
|
o pkg/obcs: |
25 |
|
- change OB?eta from _RS to _RL and put them into a separate commen block |
26 |
|
- add code to read OB?eta from a file via obcs_prescribe_read: the code |
27 |
|
compiles and does not destroy any other experiments, the data is read |
28 |
|
properly, but the code is not tested in actual applications |
29 |
|
o ini_mnc_vars.F: if available (ALLOW_CAL), use startTime_1 and 2 to |
30 |
|
create a meaningful unit attribute for the time coordinate variable |
31 |
|
o pkg/ebm: add plenty of missing "_d 0" ; |
32 |
|
update results of experiment global_ocean_ebm (fwd + adjoint tests) |
33 |
|
Notes: several pieces of code only works for (very) specific set up |
34 |
|
(domain size, lat range, etc ...); should be documented & stop if one |
35 |
|
of those assumptions is not meet. |
36 |
|
o pkg/seaice: add missing "_d 0" in HSNOW initialisation |
37 |
|
update results of experiments: lab_sea (3 fwd tests + 2 adjoint tests) |
38 |
|
and offline_exf_seaice (seaicetd + adjoint). |
39 |
|
o add missing "_d 0" in ShortWave flux penetration (S/R SWFRAC): |
40 |
|
update results of experiments: lab_sea (4 fwd tests + 3 adjoint), |
41 |
|
seaice_obcs, global_ocean.cs32x15 (icedyn & thsice) and natl_box (x2). |
42 |
|
o few changes to pass when compiling with strick checking of arguments |
43 |
|
across S/R (allow to find few bugs in S/R calls). |
44 |
|
o pkg/seaice: |
45 |
|
- seaice_evp: move if-statement out of loop (TEM) at the cost of 11 new |
46 |
|
2d-fields, for slightly better performance |
47 |
|
|
48 |
|
checkpoint61t |
49 |
|
o pkg/seaice: |
50 |
|
- remove unused variables in seaice_evp |
51 |
|
- reduce number of SQRT in seaice_evp and seaice_calc_viscosities |
52 |
|
o eeset_parms.F, open_copy_data_file.F & ini_parms.F: |
53 |
|
named file replacing SCRATCH files (for TARGET_BGL & TARGET_CRAYXT): |
54 |
|
each processor opens a different file (a fix from Matt). |
55 |
|
o cg2d: |
56 |
|
use simple EXCH (overlap size = 1 and ignore corners), like in cg3d.F : |
57 |
|
this reduces number of EXCH calls by 2 when using exch2. |
58 |
|
For now, keep previous size array for gc2d_nsa.F to avoid Pb with store |
59 |
|
directives and missing simple EXCH (exch_s3d_rx) adjoint code. |
60 |
|
o pkg/thsice: |
61 |
|
add code for Near Infra-Red albedo (from Jeff) |
62 |
|
(turned on with run-time flag: thSIce_calc_albNIR, in data.ice). |
63 |
|
o ini_model_io: |
64 |
|
move (from packages_readparms.F & initialise_fixed.F) to ini_model_io.F |
65 |
|
MNC & MONITOR initialisation calls; move MNC calls from {pkg}_readparms.F |
66 |
|
to {pkg}_init_fixed.F for {pkg}= kpp, seaice & shelfice. |
67 |
|
|
68 |
|
checkpoint61s |
69 |
|
o pkg/exch2 (exch2_rxN_cube*): |
70 |
|
remove last BARRIER (no need to synchronise after getting data from shared |
71 |
|
buffer (get) as long as any change to buffer (put,recv) is between BARRIER) |
72 |
|
o change tiling of natl_box (from 2 to 4 tiles) to get it tested with MPI+MTH |
73 |
|
and generate new output (for the 2 tests) using default rotationPeriod. |
74 |
|
o pkg/exch2: |
75 |
|
- always call exch2_*_cube, not exch-1 anymore, if useCubedSphereExchange=F |
76 |
|
(was already the case with AUTODIFF_EXCH2 defined) |
77 |
|
- add bj in exch2 arrays and S/R. |
78 |
|
- can now use pkg/exch2 for regular (non CS) set-up; switch test experiment |
79 |
|
MLAdjust to exch2 (identical results except with MPI -> truncation diff); |
80 |
|
o add pkg/longstep for ptracer timestep a multiple of U/V/T/S timestep |
81 |
|
- adds deltaT argument to the subroutines TIMESTEP_TRACER DWNSLP_APPLY |
82 |
|
GAD_ADVECTION GAD_CALC_RHS GAD_IMPLICIT_R GAD_SOM_ADVECT |
83 |
|
- verification experiment natl_box.longstep |
84 |
|
o build_options: ifort+mpi_aces optfile (following Constantinos advise): |
85 |
|
allow different versions of ifort (8 or 9) and different versions |
86 |
|
of mpi (mpich or mpich2) to be used with the same optfile. |
87 |
|
requires to load a netcdf module and set MPI_INC_DIR (see comments inside |
88 |
|
optfile). Change testing scripts (example_scripts/ACESgrid) accordingly. |
89 |
|
o pkg/seaice: |
90 |
|
fix previous modif (3-level field removed) for old-pickup reading |
91 |
|
|
92 |
|
checkpoint61r |
93 |
|
o pkg/seaice: |
94 |
|
- clean up the 3-time levels of UICE,VICE,HEFF,AREA in three steps: |
95 |
|
1. introduce fields for time step (N-1) |
96 |
|
2. do not use levels 2 and 3 anymore (but keep them) |
97 |
|
3. replace 3D versions of UICE,VICE,HEFF,AREA by 2D versions. This |
98 |
|
includes changes to pkg/thsice/thsice_get_velocity.F, |
99 |
|
model/src/external_forcing_surf.F, pkg/obcs/obcs_apply_seaice.F, and |
100 |
|
pkg/obcs/obcs_apply_uvice.F. |
101 |
|
- completely remove code for leap-frog time stepping for second order |
102 |
|
advection |
103 |
|
- adjusted adjoint accordingly; |
104 |
|
gradient checks have improved by an order of magnitude! |
105 |
|
|
106 |
|
checkpoint61q |
107 |
|
o pkg/seaice: fix bug in free slip boundary conditons for LSOR solver |
108 |
|
- the boundary conditions along open boundaries are unclear, because the |
109 |
|
associated masks are set to zero and are thus treated as closed when |
110 |
|
applying the slip-boundary conditions. |
111 |
|
- move initialisation of grid variables from seaice_init_varia to _fixed |
112 |
|
o adding doc/diags_changes.txt |
113 |
|
o model/src: |
114 |
|
do_the_model_io: rename few pkg S/R which are called from do_the_model_io, |
115 |
|
to avoid "_diags" naming if no relation to pkg/diagnostics. |
116 |
|
ini_theta,ini_salt: fix (at least compile the code) reading from NetCDF |
117 |
|
file and only call PLOT_FIELD_XYZ if debugMode=T. |
118 |
|
remove few PLOT_FIELD_XY calls outside initialisation. |
119 |
|
remove few unnecessary BARRIER. |
120 |
|
o pkg/salt_plume: add SaltPlumeSouthernOcean to allow disabling |
121 |
|
of salt_plume package in Southern Ocean |
122 |
|
o pkg/thsice: |
123 |
|
- fix hOceMxL in overlap (needed for sea-ice advection); |
124 |
|
update output of experiment global_ocean.cs32x15.icedyn. |
125 |
|
o pkg/seaice: |
126 |
|
- make the (placeholder-)tracer IceAge a little more meaningful |
127 |
|
by avoiding ridging of age due to convergent ice flow |
128 |
|
- add diagnostics for surface temperature, atmospheric and oceanic |
129 |
|
heat flux, and atmospheric and oceanic ice growth |
130 |
|
|
131 |
|
checkpoint61p |
132 |
|
o eesupp (global_max & global_sum): |
133 |
|
- add one more element to buffer (start at index 0) for GL_MAX/SUM output; |
134 |
|
- remove starting & endding barrier (no longer needed) |
135 |
|
o pkg/mdsio, writelocal: |
136 |
|
works also in multi-threaded when LOCBIN_IO_THREAD_SAFE is undefined |
137 |
|
(remove the STOP): uses shared buffer IO to store data from all threads |
138 |
|
and then let master-thread write nThreads tiles. |
139 |
|
o pkg/mdsio (read/write field): |
140 |
|
- tiled IO done in 1 piece (all levels at a time); |
141 |
|
- new header file "MDSIO_BUFF_3D.h" with 3-D buffers. |
142 |
|
- multi-threaded: allow to read/write local (non-shared) array |
143 |
|
(was already working with singleCpuIO ; now works also without); |
144 |
|
=> no longer needs barrier call after reading a file (added inside |
145 |
|
mds_read/write field). |
146 |
|
- move barrier calls outside gather/scatter_2d to mds_read/write_field |
147 |
|
o pkg/obcs/exf |
148 |
|
- exf_getffieldrec.F: add a few ". _d 0" and |
149 |
|
add code to catch the case fldperiod = 0. |
150 |
|
- obcs_precribe_read.F: change if statements so that they match |
151 |
|
the variable type (_RL) |
152 |
|
- add a separate exf_iprec_obcs to exf_set_obcs (by default equal to |
153 |
|
exf_iprec) as part of namelist EXF_NML_01 |
154 |
|
o pkg/mdsio: |
155 |
|
- read/write tiled (local) files: read/write 1-level tile chunk at a time |
156 |
|
(instead of segment of length sNx); expected to speed up tiled IO. |
157 |
|
o pkg/exch2: |
158 |
|
- take buffer copy from/to array out of S/R exch2_send/recv into new |
159 |
|
S/R exch2_put/get ; Exch of local variable now works with MPI+MTH |
160 |
|
(tested by removing commom block statement in SOLVE_FOR_PRESSURE.h). |
161 |
|
- ad version of send/recv no longer needed (but needs exch2_ad_put/ad_get) |
162 |
|
- simplify argument list of S/R exch2_get_uv_bounds and exch2_get_scal_bounds |
163 |
|
(which replaces exch2_get_recv_bounds & exch2_get_send_bounds) by |
164 |
|
including Topology header file (and common blocks). |
165 |
|
- implement EXCH_IGNORE_CORNERS in scalar exchange (rx1); |
166 |
|
remove 2nd exch call in exch2_s3d_rx (no longer needed). |
167 |
|
- document S/R arg. list (+ update some comments) |
168 |
|
o pkg/seaice: |
169 |
|
- added SEAICE_CLIM_AIR code, which permits modulation of surface air |
170 |
|
temperature and humidity over sea ice based on climatological values. |
171 |
|
- clean up computation of Hibler+Bryan (1987) stress coupling for the case |
172 |
|
of LSR (this change is expected to change the results slightly because now |
173 |
|
it uses slightly difference moduli from the second last LSR solution, but |
174 |
|
that is more consistent with the stress computations; this part of the code |
175 |
|
is not tested in the verification experiments) |
176 |
|
- fix diagnostics for flux of ice in case of B-grid |
177 |
|
- turn strain rates eij and press into global fields within global |
178 |
|
common block in SEAICE.h for more straighforward diagnostics |
179 |
|
- fix diagnostics for principal stress components, press, zeta, eta |
180 |
|
to give something meaning full |
181 |
|
o change tiling of global_ocean.cs32x15 (12 tiles, 32x16) and generate |
182 |
|
new output (all 4 tests) using default rotationPeriod. |
183 |
|
o pkg KPP: |
184 |
|
- add flag (LimitHblStable) to avoid limiting BL depth under stable conditions |
185 |
|
- fix vertical indices in RI_IWMIX (match between KPP grid from 0 to Nr+1 |
186 |
|
and normal grid from 1 to Nr) |
187 |
|
o Fix computations of L4rdt in mom_calc_visc.F |
188 |
|
o eesupp: |
189 |
|
new S/R ALL_PROC_DIE (like EEDIE, without irrelevant threads err msg): |
190 |
|
should be called before a "stop" if we know that all process are going |
191 |
|
to stop; if one Proc does call this S/R, will hang in MPI_FINALISE. |
192 |
|
|
193 |
|
checkpoint61o |
194 |
|
o pkg/exch2: change check_depths to issue just a warning (instead of error+stop) |
195 |
|
o model/src/ini_curvilinear_grid.F: |
196 |
|
compute angleSin/CosC when they are not read from a file |
197 |
|
o pkg/seaice: |
198 |
|
add pseudo time stepping for LSOR |
199 |
|
o pkg/exch2: |
200 |
|
temporary fix for exch2_s3d_rx; update results of exp. hs94.cs-32x32x5.impIGW |
201 |
|
o pkg/seaice: |
202 |
|
Adding separate sea ice albedo, drag coefficients, and HO in Southern |
203 |
|
Ocean from those in Arctic Ocean. See *_south in SEAICE_PARAMS.h |
204 |
|
o SingleCpuIO: |
205 |
|
fix bug put before chkp61n (May 11), when more than 2 MPI cpu are used. |
206 |
|
o pkg/seaice: |
207 |
|
put exchanged variable in (local) common block for MPI+Multi-Threads |
208 |
|
to work (necessary with exch1, and for now with exch2 too). |
209 |
|
o verification: to get more experiments tested with mpi, mth and mpi+mth, |
210 |
|
change tiling of: |
211 |
|
aim.5l_Equatorial_Channel, aim.5l_LatLon, dome, exp4, global_ocean.90x40x15, |
212 |
|
hs94.cs-32x32x5, MLAdjust, offline_exf_seaice & tutorial_plume_on_slope |
213 |
|
update output of exp: |
214 |
|
exp4, global_ocean.90x40x15(+.dwslp), MLAdjust(all 5), |
215 |
|
tutorial_plume_on_slope and hs94.cs-32x32x5 (use default rotationPeriod) |
216 |
|
+ ideal_2D_ocean: use default rotation period + test Global_Sum_SINGLECPU |
217 |
|
(note: output.txt generated on faulks using only 1 tile) |
218 |
|
o eesupp & mdsio (SingleCpuIO): |
219 |
|
- remove MISSING_TILE_IO (from MDSIO_OPTIONS.h), replaced by run-time |
220 |
|
parameter "W2_useE2ioLayOut" (read from file "data.exch2"). |
221 |
|
Note: was not effective for globalFile & useSingleCpuIO=F; now fixed. |
222 |
|
- move MDSIO_SCPU.h (pkg/mdsio) to EEBUFF_SCPU.h (eesupp/inc). |
223 |
|
- fix GLOBAL_SUM_SINGLECPU when using Exch2; re-use same buffers and same |
224 |
|
gather/scatter S/R as with SingleCpuIO (=> 1 less 2D global RL array). |
225 |
|
o pkg/exch2: |
226 |
|
- new code to set-up W2-Exch2 topology (replace matlab-topology-generator): |
227 |
|
read parameter file "data.exch2" if it exists; otherwise try default regular |
228 |
|
cube without blank-tile; new header file "W2_EXCH2_SIZE.h"; |
229 |
|
- update matlab-topology-generator (exch2_setup_cs6_print.m) to be compatible |
230 |
|
with new code (not to be used, just as a backup solution). |
231 |
|
- add examples (e.g., for lat-lon-cap_120) of parameter file "data.exch2" |
232 |
|
in utils/exch2/input. |
233 |
|
|
234 |
checkpoint61n |
checkpoint61n |
235 |
o eesupp & mdsio (SingleCpuIO): |
o eesupp & mdsio (SingleCpuIO): |
236 |
- move mapping to global io-buffer inside gather_2d/scater_2d: this save |
- move mapping to global io-buffer inside gather_2d/scater_2d: this save |