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

Diff of /MITgcm/doc/tag-index

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

-revision 1.1207 by jmc,
Mon Dec 21 02:11:36 2009 UTC
+revision 1.1272 by mlosch,
Tue May 18 12:26:54 2010 UTC
 Line 4 
 $Name$
      Notes on tags used in MITgcmUV
      ==============================
+ o pkg/exf:
+   - rewrite exf_bulk_largeyeager04.F for better vectorization, changes
+   global_ocean.cs32x15/input.icedyn (11 digits of agreement), updated
+   - fix adjoint (used worked for ALLOW_ATM_WIND defined, works for this now)
+   - cosmetic changes to exf_bulkformulae.F to make it look more like
+     exf_bulk_largeyeager04.F
+ o eesupp/src (exch-1):
+   - Fix AD EXCH code for 2-D domain special cases (Nx=1 or Ny=1).
+   - Separate buffer filling and MPI sending: allow EXCH-1 to work for local
+     array (non-shared) when using MPI+MTH. Also reduces number of BARRIER
+     (even without using MPI).
+   - Message mode: move RecvAck setting (indicator of buffer being ready)
+     from send_put to recv_get S/R (was useless before, but not sure if
+     it's much more useful now);
+   - switch the order of sync: MPI-proc 1rst and then threads;
+   - take out spin-waiting code (#undef EXCH_USE_SPINNING), use BARRIER instead.
+ o model/src:
+   initialise dPhiHydX,dPhiHydY inside calc_grad_phi_hyd.F (so that TAF does
+   not drop it in S/R mddynamics, producing NaN in hs94.1x64x5 AD-test exp).
+ o utils/matlab/rdmnc.m: adapt to be able to use the generic matlab netcdf API
+ o pkg/ctrl: require a specific set of control variables and exf-flags for
+   adjoint of seaice (stop in ctrl_check.F if conditions are not satisfied)
+ o model/src:
+  a) 1rst version of a new driver S/R to turn-off output freq and flags after the
+rst forward sweep, to avoid over-writing output files during recomputations
+    recomputations (& grdchk); called from cost_final.F (& ctrl_init_ctrlvar.F ?)
+  b) 2nd version:
+   - add specific pkg S/R for seaice & ptracers (called from S/R TURNOFF_MODEL_IO);
+   - switch-off more output flags (exf_monFreq, chkPtFreq, dumpInitAndLast,
+      writePickupAtEnd) + seaice & ptracers output;
+   - do not switch off MNC flags (otherwise, won't get admonitor netcdf file)
+ o model/src:
+   - new S/R rotate_uv2en_rl (..._rs) that rotates vector fields.
+ o pkg/exch2:
+   - remove "simulationMode" from arg. list of S/R EXCH2_RX1,2_CUBE ;
+   - add argument "signOption" to EXCH2_RX1_CUBE (will be needed for SM exch).
+ o pkg/autodiff & exch2:
+   process exch_uv_agrid_3d_rx.f & exch2_uv_agrid_3d_rx.f through TAF
+   (for now, only if ALLOW_EXCH2 is defined).
+ o pkg/autodiff:
+   - include PACKAGES_CONFIG.h in exch*ad.flow & exch*ad.F
+     (not supposed to be in CPP_OPTIONS.h) to get ALLOW_EXCH2 defined correctly.
+    + clean-up comments (_RS is not always real*4)
+   - fix bug in ADEXCH_UV_3D_RX (without EXCH2, was calling the forward exch S/R)
+   - add a stop in ADEXCH_UV_AGRID_3D_RX if trying to use it with EXCH2:
+     was calling a forward exch2, which is wrong ; used in SEAICE & EXF pkgs.
+ o testreport (AD-tests):
+   no longer link files from dir "../input" (1rst link files from
+   "../input_ad.[xx]" if secondary test "xx" and then from "../input_ad")
+ checkpoint62f (2010/04/29)
+ o model/src/the_main_loop.F:
+   adjust store directives for seaice evp and multicategory according to
+   pkg/ecco/the_main_loop.F
+ o adjoint:
+   global_sum_ad was incomplete (noted many times by many people ;)
+   nice description by Jean U. as follows:
+   * allreduce(var) in global_sum.F can be interpreted as
+     reduce(sum) followed by broadcast of sum.
+   * The adjoint of that is
+     sum(adjoint) followed by broadcast(summed adjoint value)
+     or in other words, allreduce(sum, adjoint vars)
+   We were missing the sum(adjoint)!
+   Benign, except(!) for "global" scalars appearing in tile loops (bi,bj)
+   (noted by J.M. Campin, e.g., in cost_ssh_mean, variable offset).
+   The fix should produce 16-digit accuracy between
+   single-CPU multi-tile (e.g. nSx=2, nPx=1) and
+   multi-CPU single-tile (e.g. nSx=1, nPx-2)
+   We'll probably change this yet again shortly
+ o eesupp/src:
+   - fix global_sum_r4 (wrong from previous modif in Jun 2009; ever used ?)
+ o pkg/icefront
+   - ALLOW_SUBGLACIAL_RUNOFF option adds capability for time-evolving
+     specification of addmass array based on a 2D field, which is added
+     at bottommost wet level at each horizontal location.
+ o model/src:
+  horizontal grid initialisation (other than curvilinear): add code for Exch2 IO
+  layout; (not always compatible with delX,delY setting; commented out for now).
+ o pkg/exch2:
+  change tile to processor setting (case nSy*nPy > 1): allows default model mapping
+  (myX and myYGlobalLow) to be used for simple (1 facet, no blank tile) config.
+ checkpoint62e (2010/04/15)
+ o pkg/obcs:
+   - fix wrong index in OBCS_SPONGE_V for Eastern-OB (reported by David Hebert).
+ o pkg/dic:
+   -fix bug in fe_chem.F for multi-tile (nSx*nSy > 1) set-up;
+     (note: not presently tested in any forward verification exp.)
+    update tutorial_global_oce_biogeo AD output file.
+ o pkg/exf:
+   - account for the variation of albedo as a function of zenith angle
+     by defining ALLOW_ZENITHANGLE in EXF_OPTIONS.h
+ o build_options:
+  - rename "linux_ia64_ifort+mpi_ice_nas" to "linux_amd64_ifort+mpi_ice_nas"
+ o pkg/dic:
+   change DIC atmos-CO2 box (method 3, dic_int1=3) initialisation & restart:
+   - use binary pickup file for restart;
+   - initialise atmos-CO2 box from dic_pCO2 parameter;
+   - fix multi-threaded.
+ o pkg/rbcs:
+  - clean-up RBCS code: add RBCS_OPTIONS.h file ; fix multi-threaded ;
+    change (+ fix IO calls) type of array RBC_mask from _RL to _RS.
+  - add RBCS to exp4 passive tracer 01 (rbcs was previously untested)
+    and update output.txt
+ o pkg/kpp:
+  - add diagnostics for KPP non-local flux of Temp, Salinity and pTracers;
+  - change diagnostic of KPP non-local term: output average of KPPghat*KPPdiffKzS
+    (= ratio of KPP non-local (salt) flux relative to surface-flux) when
+    using pkg/timeave & pkg/diagnostics (diag 'KPPghatK' replaces 'KPPghat ');
+  - change pkg/offline to read-in average KPPghat*KPPdiffKzS from file
+    KPP_ghatKFile (instead of the average of KPPghat from file KPP_ghatFile).
+ o pkg/offline:
+   fix bug in S/R OFFLINE_FIELDS_LOAD regarding surfaceForcing conversion factor
+   (bug was added in version 1.10 on 2007-08-23).
+ o pkg/offline:
+   - no need to specify periodicExternalForcing, etc ... to use offline pack
+   - offline input files no longer just single prec --> parameter offlineLoadPrec
+ o verification/lab_sea:
+   change AD test to multi-tile setup and update all 4 AD output.
+   NOTE: while cost function values remain very similar,
+   gradients are very different, although gradient checks are self-consistent
+   Should try to get to the bottom of this (cost function summation order?)
+ o pkg/dic:
+   in carbon_chem.F, replace log10 (AUTODIFF) with explicit form to avoid
+   a weakness of TAF that drops part of double precision in log10 derivative.
+   <= changes removed since problem has been fixed in current TAF version 1.9.71
+   update AD output of exp. tutorial_dic_adjoffline & tutorial_global_oce_biogeo.
+ o pkg/ecco:
+   add missing "_d 0" in cost_readers.F, cost_readtopex.F & cost_readtopexmean.F
+   update all 4 lab_sea AD output.
+ checkpoint62d (2010/03/22)
+ o build_options:
+   remove '-r8' (and '-i4') from linux_*pgf and linux_*ifort optfiles.
+ o remove unbalanced quote (single or double) in commented line of source code
+   (may cause some warning with some preprocessor/makedepend versions).
+ o pkg/seaice:
+   fix seaice_model.F for adjoint (gael) and update AD results of experiments
+    lab_sea (std + evp + noseaicedyn) and offline_exf_seaice.
+ o testreport:
+   add option '-runonly' (short: '-ro') to just do the run part (without make).
+ o eesupp: remove JAM communication code (obsolete).
+ o add option to genmake2 & testreport to switch "_RS" type to "real*4"
+   (if allowed in CPP_EEOPTIONS.h); update default CPP_EEOPTIONS.h and return
+   to default CPP_EEOPTIONS.h for exp. inverted_barometer & solid-body.cs-32x32x1
+   (update also the output).
+ checkpoint62c (2010/03/02)
+ o tools/build_options :
+    move (some) old optfiles from build_options/ to build_options/unsupported/
+ o model/src and pkg/exf
+    surface relaxation -- separating data and data.exf options
+    The point is to ensure that we specify the surface relaxation
+    either in data or in data.exf (but not half and half) and to
+    avoid an accidental relaxation to 0 (which happened if a time
+    scale was specified but the file name was left empty)
+ o model/src model/inc and pkg/mom_common
+    ALLOW_3D_VISCAH/ALLOW_3D_VISCA4 => 3D specification of Viscosity
+    => addition to the viscosity field as computed by others schemes
+ o tools/build_options:
+   - rename "_generic_mpi" -> "+mpi_generic" (like other "+mpi" optfile)
+     but keep the former optfile for few weeks for smooth transition
+ o pkg/ecco
+   - allow user defined cost function contributions
+         CPP: ALLOW_USERCOST_CONTRIBUTION
+         USAGE: compute contributions in cost_usercost_all.F
+   - note: this is identical to what ALLOW_GENCOST_CONTRIBUTION once was.
+       the two options are complementary.
+ o pkg/smooth
+   - migration from contrib to the main repository
+ checkpoint62b (2010/01/26)
+ o pkg/monitor:
+   - change back mon_printstats_rl.F & mon_stats_rl.F and add a stop
+     (in case someone forgot to update customized pieces of code)
+ o nonHydrostatic:
+   - add Non-hydrostatic free-surface form (selectNHfreeSurf=1)(not fully tested);
+   - add a simple 2-D experiment (short_surf_wave) to test it.
+ o pkg/icefront:
+    adding the shell of, and the hooks to, a new package that
+    will model melting and freezing of vertical glacier ice fronts
+ o momentum:
+   add diagnostics of momentum tendency from Hydrostatic pressure gradient
+    (without surf.pressure term: would need a 2nd filling since implicit term).
+ o add 3 "_d 0"s so that fizhi compiles with xlf on iblade, and replace
+   two " &end" by " &" (in fizhi-cs-*) because S/R NML_SET_TERMINATOR
+   cannot deal with the former
+ o cleaning:
+   - remove all (except OpenAD) old TLM output from results_ad dirs.
+   - remove unused S/R (MDSWRITEMETA, DIAGNOSTICS_ADD2LIST, old DFILE IO S/R)
+   - comment out maskH definition (replaced by maskInC)
+ o pkg/monitor:
+   - change mon_printstats_rl.F & mon_stats_rl.F (were similar to mon_writestats_rl
+     & mon_calc_stats_rl) to be the RL equivalent of mon_printstats_rs & mon_stats_rs.
+ checkpoint62a (2010/01/15)
+ o model/src:
+   - bottom pressure diagnostic: correct for r* (when nonlinFreeSurf >= 4)
+ o pkg/diagnostics:
+   - add optional level number diagnostics (i.e., level number to be define
+     explictly with S/R DIAGNOSTICS_SETKLEV) with parser-code(10)="X".
+   - strictly check for valid parser-code(10) ;
+   - check if adding diag to the list from the right place.
+ o pkg/longstep:
+   - change parameters and fix tracer conservation with rstar/nonlinear free surface
+ o diagnostics:
+  - in all pkgs, replace remaining calls to old S/R DIAGNOSTICS_ADD2LIST with newer
+    S/R DIAGNOSTICS_ADDTOLIST ; fix seaice VICE diagnostics.
+ o pkg/diagnostics:
+  - Add option for writing vertical integral (sum of selected levels) of an output
+     field instead of writing output field levels. Activate this option on a per
+     output-file basis, by setting the 2nd character of fileflag to 'I'.
+  - Add field-specific information relative to level integral in 3rd postion
+     of parser diagnostics code (gdiag): " "=cumulate levels, "R"=integrate
+     vertically (x level-thickness x hFac), "r"=same but without hFac.
+ o do_the_model_io:
+   - pass end-of-model-run flag as argument, from forward_step to diagnostics_write:
+     in case useSIGREG, will write (averaged) diagnostics (if dumpAtLast) before
+     writing pickups and stopping.
+ o pkg/diagnostics (for now, only with mdsio):
+   - write time information to meta file (and modify accordingly rdmds.m) :
+     time-averaged diagnostics output: write starting and ending time of the
+     time-averaging interval ; snap-shot diagnostics output remains unchanged;
+ o pkg/timeave:
+   - S/R TIMEAVE_NORMALIZ: change size of cumulated-time array (was previously
+     always Nr), now assumed to be equal to number of levels of averaged field;
+     only used for DIC_COST monthly mean arrays: this fixes dic_aver_final.F
+ o pkg/sbo:
+   - fix ocean bottom pressure global-average & 2-D time-ave output (was
+     neither initialised nor filled).
+   - add starting iteration in suffix of global-average output file
+     (to avoid over-writing the file after a restart)
+   - use standard MDSIO pkg S/R instead of SBO_WRITEVECTOR (fix byte-swap Pb).
+ o time-average:
+   - use simple cumulative-time array (no k index) in all pkgs: {PKG}_timeAve(bi,bj)
+     (pkgs: aim_v23, bulk_force, dic, gmredi, kpp, land, ocn_compon_interf,
+      ggl90, my82, opps, pp81, ptracers, sbo, seaice, shelfice, thsice)
+     this fix DIC_COST cumulative-time array (previously: bi,bj,k)
+   - fix 1rst call (nIter0) snap-shot output and accumulation (pkg/shelfice)
+   - fix missing accumulation for pkg/opps.
+   - cumulate fields only if doing time-ave output ({PKG}_taveFreq > 0)
+      (pkgs: ptracers, kpp, ocn_compon_interf, ggl90, my82, pp81);
+   - fix initialisation of time-ave vars and do only full time-step increment
+     (i.e.: no first & last "half time-step") (pkgs: ggl90, my82, opps, pp81);
+ o pkg/layers:
+   - use standard RW pkg S/R to write the output
+     and remove local write_fld version (which had a problem).
+   - use simple cumulative-time array (no level index) to fix time-ave S/R calls.
+   - fix LAYERS_CALC argument list.
+ o pkg/timeave:
+   - store cumulative-time in a simple tiled array (drop the level index)
+   - add new S/R (TIMEAVE_NORMALIZE) which uses simple cumulative-time array.
+ o pkg/ocn,aim_compon_interf
+   - changes for multi-threaded coupled run. (tested with ifort-11 on danton)
+ checkpoint62  (2009/12/23)
+ o switch on useOBCSbalance in tutorial_plume_on_slope exp. and update output.
+     (was not previously tested in any other exp.)
  o pkg/monitor:
    - clean-up _RL/_RS version: add specific S/R for statistics with mask & volume:
      mon_writestats_rl/s,F mon_calc_stats_rl/s.F ; will keep the old one
      (mon_printstats_rl/s,F mon_stats_rl/s.F) just for bare statistics.
    - use hFac & 2-D mask (instead of 3-D mask + hFac in previous mon_stats_rl.F).
    - change del2 calculation.
-Line 14 
 o pkg/monitor:
+Line 272 
 o pkg/monitor:
    - shorter monitor suffix for forcing stats ("forcing" instead of "extforcing").
  o mask & obcs:
    - define 2-D "interior" masks (i.e., zero beyond OB) at 3 locations (C, W & S).
    - move global area computation from ini_masks_etc.F to ini_linear_phisurf.F
      (called after packages_init_fixed)
  o pkg/obcs:
    - move bi,bj loops into obcs_calc, so obcs_prescribe_read is called only once.
-Line 28 
 o nonHydrostatic:
+Line 286 
 o nonHydrostatic:
      (in prep. for new NH free surface form).
  o pkg/obcs:
    add files & OB-variables for wVel (Non-Hydrostatic) with useOBCSprescribe
  o model/src:
    fix 2-D solver initialisation for deep-atmosphere (p-coord) with topography
  checkpoint61z (2009/12/01)

 Legend:



Removed from v.1.1207
 


changed lines


 
Added in v.1.1272
 Legend:



Removed from v.1.1207
 


changed lines


 
Added in v.1.1272
-Removed from v.1.1207
+Added in v.1.1272

	ViewVC Help
Powered by ViewVC 1.1.22