| 4 |
Notes on tags used in MITgcmUV |
Notes on tags used in MITgcmUV |
| 5 |
============================== |
============================== |
| 6 |
|
|
| 7 |
|
checkpoint64 (2012/10/09) |
| 8 |
|
o pkg/shelfice |
| 9 |
|
- in mom_fluxform.F, mom_vecinv.F: |
| 10 |
|
CALL SHELFICE_V_DRAG(bi,bj,k,vFld,KE,KappaRU,vF,myThid) |
| 11 |
|
Bug fix KappaRU -> KappaRV |
| 12 |
|
does not change verification/isomip but may change other setups |
| 13 |
|
(presumably constant KappaRU=KappaRV) |
| 14 |
|
- fix diagnostics SHIU/Vdrag (thanks to Tom Millgate) |
| 15 |
|
o pkg/obcs: |
| 16 |
|
- fixed some indexing bugs in obcs_seaice_sponge.F |
| 17 |
|
- added obcs relaxation subroutines for HSALT and HSNOW |
| 18 |
|
- changes verification/seaice_obcs/results/obcs_seaice_sponge.F |
| 19 |
|
o pkg/seaice: |
| 20 |
|
- add "multicategory" parameterization for snow in analogy to ice, |
| 21 |
|
default is off for now. |
| 22 |
|
o pkg/obcs: |
| 23 |
|
- added obcs_seaice_sponge.F for obcs relaxation of seaice variables; |
| 24 |
|
see verification/seaice_obcs/input.seaiceSponge for example application. |
| 25 |
|
- remove local commom bloc "OBCS_GLOBAL" in obcs_readparms.F and do all the OB |
| 26 |
|
index setting by master-thread. This fix multi-threads Pb in previous modifs. |
| 27 |
|
o pkg/layers : |
| 28 |
|
- add diagnostics to pkg/layers |
| 29 |
|
- move the meat of pkg/layers from layers_calc.F to layers_fluxcalc.F |
| 30 |
|
(from R.Tulloch). Layers_calc.F now calls layers_fluxcalc.F accordingly. |
| 31 |
|
o pkg/generic_advdiff/gad_implicit_r.F : |
| 32 |
|
- fix ADVr diagnostics in case of both implicit vertical advection and diffusion. |
| 33 |
|
o pkg/seaice/seaice_growth.F: |
| 34 |
|
- fix SItflux diagnostics in virtual salt flux + balanceEmpmr case. |
| 35 |
|
o pkg/obcs: |
| 36 |
|
- change default setting of OBCS_UVICE_OLD from defined to undef; |
| 37 |
|
- use new parameter OB_indexNone for null index value (instead of hard-coded 0) |
| 38 |
|
in all pkg/obcs code and in other pkgs; remove k loop in obcs_check_depths.F; |
| 39 |
|
- pass OB_indexNone as (new) argument to S/R CTRL_MASK_SET_[X,Y]Z; |
| 40 |
|
- change OB null index value (OB_indexNone) to -99 ; this fix the problem |
| 41 |
|
with Southern & Western OB at a tile edge (at least in a simple set-up). |
| 42 |
|
o pkg/obcs: |
| 43 |
|
- change settings of OB index in case OB coincides with tile edge (shared by |
| 44 |
|
2 tiles), so that common OB location is registered by both tiles (instead of |
| 45 |
|
only 1). However, for Southern and Western OB, index value = 0 is not |
| 46 |
|
recognized as valid OB index (=default null value) and needs to be fixed. |
| 47 |
|
- modify accordingly OBCS_BALANCE_FLOW and OBCS_MONITOR (avoid double counting) |
| 48 |
|
- new parameter "OB_indexNone" for null index value (for now, still zero) |
| 49 |
|
to replace hard-coded null index value of zero. |
| 50 |
|
o pkg/exch2: |
| 51 |
|
- improve W2_USE_E2_SAFEMODE (faster) and W2_E2_DEBUG_ON: |
| 52 |
|
always check buffer-size (but outside the loop); move checking for valid |
| 53 |
|
index from W2_USE_E2_SAFEMODE to W2_E2_DEBUG_ON; in W2_E2_DEBUG_ON, |
| 54 |
|
print each tile and point connection only if |W2_printMsg|>= 2 and 3. |
| 55 |
|
o pkg/ctrl (pack & unpack): |
| 56 |
|
- in ctrl_pack.F & ctrl_unpack.F, use the same local file-name variable |
| 57 |
|
(locally defined) for different variables (+ remove the variable-specific |
| 58 |
|
file name in CTRL_GENARR.h). For now, only implemented for GENARR2/3D_CONTROL |
| 59 |
|
(to be extended to other local file names in ctrl.h). |
| 60 |
|
o pkg/flt: |
| 61 |
|
- add a FLT_Iter0 option (similar to PTRACERS_Iter0). |
| 62 |
|
o pkg/mdsio: |
| 63 |
|
- fix missing STOP in mdsio_readvec_loc.F (case ioUnit=0) when file is missing |
| 64 |
|
o tools/genmake2: |
| 65 |
|
- add similar dependences for target EXE_AD & EXE_FTL as we have for EXECUTABLE |
| 66 |
|
(added by Alistair in version 1.51 for building without make depend); |
| 67 |
|
- remove Header files list (H_SRC_FILES) from the dependence of targets |
| 68 |
|
ad_input_code.f & ftl_input_code.f (not clear why it was added in the first |
| 69 |
|
place in version 1.23): this save going through staf when unrelated header |
| 70 |
|
file has been modified. |
| 71 |
|
Note: for an intermediate target (e.g., ad_input_code.f or adtaf) without |
| 72 |
|
"make depend" step, still need to do "make links" before "make adtaf". |
| 73 |
|
o eesupp/src (+ few pkg src files) & genmake2: |
| 74 |
|
- finish to remove ALWAYS_USE_MPI in source code that TAF does not see; |
| 75 |
|
- no longer set ALWAYS_USE_MPI in Makefile (was done by genmake2). |
| 76 |
|
- improve genmake2 checking for ALLOW_${PKG} in CPP_OPTIONS.h files (distinguish |
| 77 |
|
upper/lower case ; account for space between "#" and undef/define) |
| 78 |
|
o pkg/flt (with exch2): |
| 79 |
|
- cleaning: fix so that it compiles; move remaining MPI calls out of flt_exch2.F; |
| 80 |
|
- simplify threads synchronisation; fix (many issues) multi-threaded with MPI. |
| 81 |
|
|
| 82 |
|
checkpoint63s (2012/09/05) |
| 83 |
|
o pkg/ecco: |
| 84 |
|
- remove #ifdef ALLOW_SMOOTH_CORREL3D brackets. |
| 85 |
|
- add more relevant #ifdef ALLOW_SMOOTH ones. |
| 86 |
|
- sort out useAtmWind, useSMOOTH, ctrlSmoothCorrel2D. |
| 87 |
|
o pkg/ctrl: |
| 88 |
|
- CTRL_OPTIONS.h : document CPPs related to smoothing; ALLOW_SMOOTH_CORREL2D/3D, |
| 89 |
|
ALLOW_CTRL_SMOOTH, ALLOW_SMOOTH_CTRL2D, ALLOW_SMOOTH_CTRL3D. |
| 90 |
|
- ctrl.h : document run time params doInitXX, doMainPack, doMainUnpack, doPackDiag, |
| 91 |
|
doSinglePrecTapelev (audodiff param!), ctrlSmoothCorrel2D (new) and |
| 92 |
|
ctrlSmoothCorrel3D (new). |
| 93 |
|
The added ctrlSmoothCorrel2D/3D replace the old ALLOW_SMOOTH_CORREL2D/3D CPP. |
| 94 |
|
- ctrl_readparms.F : initialise ctrlSmoothCorrel2D/3D and read from namelist. |
| 95 |
|
- ctrl_init.F ctrl_pack.F ctrl_unpack.F : switch off ustress/vstress |
| 96 |
|
controls when useAtmWind, and vice versa. |
| 97 |
|
- ctrl_map_ini_ecco.F : UVEL0, VVEL0, edtaux, edtauy cannot be smoothed as tracers. |
| 98 |
|
- ctrl_get_gen.F : replace #ifdef ALLOW_SMOOTH_CORREL2D with |
| 99 |
|
#ifdef ALLOW_SMOOTH, IF ctrlSmoothCorrel2D, IF useSMOOTH. |
| 100 |
|
- ctrl_map_ini_gen.F : replace #ifdef ALLOW_SMOOTH_CORREL3D with |
| 101 |
|
#ifdef ALLOW_SMOOTH, IF ctrlSmoothCorrel3D, IF useSMOOTH. |
| 102 |
|
- ctrl_set_pack_xy[z].F, ctrl_set_unpack_xy[z].F: |
| 103 |
|
replace ALLOW_SMOOTH_CORREL2D/3D brackets with ctrlSmoothCorrel2D/3D. |
| 104 |
|
o pkg/smooth: |
| 105 |
|
- remove un-necessary includes. |
| 106 |
|
- remove ALLOW_SMOOTH*D and ALLOW_SMOOTH_CORREL*D brakets. |
| 107 |
|
Those CPP options were never defined, and not necessary. |
| 108 |
|
o model/src: |
| 109 |
|
- fix CPP brackets in ini_mixing.F (GM_background_K belongs to gmredi). |
| 110 |
|
- standard treatment of useSMOOTH. |
| 111 |
|
o pkg/autodiff: |
| 112 |
|
- add LOGICAL useGGL90inAdMode, useGGL90inFwdMode, useSALT_PLUMEinAdMode, |
| 113 |
|
useSALT_PLUMEInFwdMode allowing to switch off ggl90 and salt_plume adjoint. |
| 114 |
|
o pkg/exch2: |
| 115 |
|
- add "if usingMPI" where it was missing (in exch2_rx?_cube*.template) |
| 116 |
|
+ start to remove ALWAYS_USE_MPI in source code that TAF does not see. |
| 117 |
|
- make new S/R W2_MAP_PROCS out of w2_eeboot.F to setup exch2 CommonFlag |
| 118 |
|
and tiles to process mapping. |
| 119 |
|
- re-arrange 3 header files (W2_EXCH2_PARAMS.h, _BUFFER.h & _TOPOLOGY.h) |
| 120 |
|
moving /EXCH2_FILLVAL_RX/ & /W2_EXCH2_COMMFLAG/ to W2_EXCH2_TOPOLOGY.h; |
| 121 |
|
rename "exch2_tProc" to "W2_tileProc" and move seting of e2FillValue_RX |
| 122 |
|
from w2_eeboot.F to w2_readparms.F. |
| 123 |
|
o eesupp/src: |
| 124 |
|
- set exchNeedsMemsync & exchUsesBarrier according to "usingMPI" (instead of |
| 125 |
|
ALWAYS_USE_MPI) + moved from ini_communication_patterns.F to eeset_parms.F; |
| 126 |
|
also move setting of exchCollectStatistics from exch_init.F to eeset_parms.F |
| 127 |
|
- in eedie.F and all_proc_die.F, always call MPI_FINALIZE when ALLOW_USE_MPI |
| 128 |
|
is defined (remove condition on usingMPI) to match cases where MPI_INIT |
| 129 |
|
has been called. |
| 130 |
|
- fix bug I put in previous modif of exch_sm_3d_rx.template (not much used); |
| 131 |
|
- change default (non-MPI) value of myPid (from 1 to 0) to match MPI settings |
| 132 |
|
(so that on 1 proc, get same myPid independently of usingMPI) |
| 133 |
|
- add a stop if multi-procs but usingMPI=F ; stop also if usingMPI=T and |
| 134 |
|
useCubedSphereExchange with EXCH-1 (safer). |
| 135 |
|
- add "if usingMPI" where it was missing (+ start to remove ALWAYS_USE_MPI). |
| 136 |
|
o pkg/ggl90: |
| 137 |
|
- fix bug in filling corner mask array "mskCor" (FILL_CS_CORNER_TR_RL call) |
| 138 |
|
used in ALLOW_GGL90_SMOOTH code. |
| 139 |
|
o pkg/flt: |
| 140 |
|
- fix typo in FLT_WITHOUT_Y_PERIODICITY (the effect was to ignore stopping |
| 141 |
|
floats at the Southern end of the domain). |
| 142 |
|
o eesupp/inc: |
| 143 |
|
- add a list of CPP-options which can be set by Makefile (DEF_IN_MAKEFILE.h); |
| 144 |
|
but not used (and not to be used) anywhere, just for documentation. |
| 145 |
|
o pkg/autodiff, exf, ptracers & seaice (AD & TL output): |
| 146 |
|
- comment out the very nasty "writeBinaryPrec = writeStatePrec" line, since: |
| 147 |
|
a) can damage previously written records in any other multi-record files. |
| 148 |
|
b) if needed, can easily do it cleanly, by calling S/R WRITE_REC_3D_RL/S. |
| 149 |
|
o tools/genmake2: |
| 150 |
|
- define variable MAKE within Makefile and use it in place of 'make'; |
| 151 |
|
- update AD_CONFIG.h (if needed) before building any of the executable |
| 152 |
|
(forward, adjoint or tangent-lin); this involves a 2nd invocation of make; |
| 153 |
|
and add the usual #ifndef/#define AD_CONFIG_H in AD_CONFIG.h; |
| 154 |
|
- always print usual warning comment 'this file is automatically generated' |
| 155 |
|
from script convert_cpp_cmd2defines (+ remove this warning from genmake2). |
| 156 |
|
o pkg/cheapaml: |
| 157 |
|
- updated to correspond to the code used in accepted paper (Nico Wienders). |
| 158 |
|
o pkg/exf: |
| 159 |
|
added run time switch useAtmWind to replace ALLOW_ATM_WIND cpp switch. |
| 160 |
|
ALLOW_ATM_WIND now just sets the useAtmWind default (see exf_readparms.F) |
| 161 |
|
and force defines ALLOW_BULKFORMULAE (EXF_OPTIONS.h). |
| 162 |
|
o pkg/exf, autodiff, ctrl, ecco and seaice: |
| 163 |
|
remove ALLOW_ATM_WIND brackets, or replace them with useAtmWind ones. |
| 164 |
|
o pkg/ctrl, ecco: |
| 165 |
|
allow to compile both ALLOW_U/VSTRESS_CONTROL and ALLOW_U/VWIND_CONTROL. |
| 166 |
|
Depending on useAtmWind, one is inactive, and the other is active (see |
| 167 |
|
exf_getffields.F/exf_getsurfacefluxes.F). |
| 168 |
|
o tools/genmake2: |
| 169 |
|
- improve error msg when failing in 2nd test of check_for_broken_Ff |
| 170 |
|
- fix option '-makefile' short-name (-mf according to genmake2 -help) |
| 171 |
|
- source local config file "genmake_local" just after processing the genmake2 |
| 172 |
|
command arguments (previously done before processing arg). |
| 173 |
|
o pkg/grdchk: |
| 174 |
|
- add a call to TURNOFF_MODEL_IO for Tangent-Linear run (in this case, call |
| 175 |
|
to this routine from cost_final has been dropped in g_cost_final) |
| 176 |
|
- improve printed information (more explicit msg, no longer using "print *," |
| 177 |
|
fix fcpertminus printed value). |
| 178 |
|
o model/src: |
| 179 |
|
- move call to COST_INIT_VARIA out of packages_init_variables.F directly |
| 180 |
|
into initialise_varia.F (just after calling PACKAGES_INIT_VARIABLES): |
| 181 |
|
this prevents getting cost-funct reset in adinitialise_varia recomputations. |
| 182 |
|
o pkg/dic: |
| 183 |
|
- move initialisation of "freefe" array from from dic_biotic_forcing.F to |
| 184 |
|
fe_chem.F: this prevents TAF from dropping this initialisation in TLM code. |
| 185 |
|
o pkg/grdchk: |
| 186 |
|
- grdchk_main.F : added CALL GLOBAL_SUM_INT( ierr , myThid ) |
| 187 |
|
- grdchk_getadxx.F etc. : omit I/O if ierr.EQ.0 |
| 188 |
|
o pkg/autodiff/ctrl/grdchk : |
| 189 |
|
- attempt to sort out what works and what does not, |
| 190 |
|
with regard to I/O, and to warn/stop accordingly. |
| 191 |
|
o pkg/grdchk: |
| 192 |
|
- attempt to fix grdchk in the case of useSingleCpuIO. |
| 193 |
|
- "if ( myProcId .EQ. grdchkwhichproc .AND. ierr .EQ. 0 )" |
| 194 |
|
statements were moved from grdchk_main.F and into |
| 195 |
|
grdchk_getadxx.F grdchk_getxx.F grdchk_setxx.F. |
| 196 |
|
- a global sum is applied to record numbers icvrec/layer |
| 197 |
|
out of grdchk_loc so that they be passed to I/O routines. |
| 198 |
|
o pkg/smooth: |
| 199 |
|
- fix recomputation that appeared with nlfs : use smooth_recip_hFacC, |
| 200 |
|
smooth_hFacW, smooth_hFacS (copied at init_fixed) in 3d case. |
| 201 |
o verification/testreport: |
o verification/testreport: |
| 202 |
- add option '-tlm' for Tangent-Linear test (-adm,-oad,-tlm are all exclusive); |
- add option '-tlm' for Tangent-Linear test (-adm,-oad,-tlm are all exclusive); |
| 203 |
- can also use compressed (gzip) reference output file; |
- can also use compressed (gzip) reference output file; |
| 204 |
- change format (in pkg/grdchk) of TLM output (+ add forward gradient) |
- change format (in pkg/grdchk) of TLM output (+ add forward gradient) |
| 205 |
to be easier to collect by testreport. |
to be easier to collect by testreport. |
| 206 |
- update existing TLM output (half out-of-date by more than 3 yrs). |
- update existing TLM output (half out-of-date by more than 3 yrs). |
| 207 |
|
- report NETCDF & LAPACK settings (from genmake2) in output dir. |
| 208 |
|
- report run-time in each individual summary (previously only for AD test); |
| 209 |
|
for TLM test, add also TAF version. |
| 210 |
|
- add daily TLM test on (old) aces cluster + weekly test on baudelaire. |
| 211 |
|
|
| 212 |
checkpoint63r (2012/08/15) |
checkpoint63r (2012/08/15) |
| 213 |
o tools/genmake2: |
o tools/genmake2: |
Parent Directory
| 
Revision Log
| 
Revision Graph
| 
Patch