--- MITgcm/doc/tag-index 2011/08/29 19:46:22 1.1498 +++ MITgcm/doc/tag-index 2011/10/27 22:40:16 1.1523 @@ -1,9 +1,144 @@ -$Header: /home/ubuntu/mnt/e9_copy/MITgcm/doc/tag-index,v 1.1498 2011/08/29 19:46:22 jmc Exp $ +$Header: /home/ubuntu/mnt/e9_copy/MITgcm/doc/tag-index,v 1.1523 2011/10/27 22:40:16 jmc Exp $ $Name: $ Notes on tags used in MITgcmUV ============================== +o pkg/seaice (lsr solver): + - put coriolis term into FORCEX,Y (as in seaice_evp.F; previously in rhsU,V only); + update output (change truncation) of FWD test exp lab_sea & global_ocean.cs32x15 + - fix SEAICEuseFREEDRIFT for new OBCS version (ndef OBCS_UVICE_OLD) + - new parameter "LSR_mixIniGuess" allows (if=2,4) to mix free-drift solution + into initial guess for linearised solver; use weight ~ (1/err)^2,4. + - compute (with new S/R) and print initial (+ free-driff sol. if available) + and final residual of linearised system. +o pkg/pp81: add diagnostics for vertical eddy viscosity and diffusivity + (requires new subroutines pp81_init_fixed.F/pp81_diagnostics_init.F) +o pkg/obcs: Stevens BCs (still only for east and west) + - fix Stevens BCs by using intermediate velocities of previous + time step as "baroclinic" contribution rather velocity of one grid cell + inside the boundaries; more consistent but makes the entire treatment + more complicated, requires new pickup files and two new S/Rs: + obcs_copy_uv_n and obcs_save_uv_n + keep old version with flag OBCS_STEVENS_USE_INTERIOR_VELOCITY + (undefined by default) + - silently move computation of some grid factors into separate loops to + save some recomputations + - add a hack for non-linear free surface + - add a verification experiment to exp4 that tests this code +o make gluemnc/xplodemnc work with (moron) non-GNU shell tools + +checkpoint63d (2011/10/25) +o pkg/obcs + seaice-dynamics: + - new implementation calls OBCS_APPLY_UVICE (in seaice_dynsolver.F) before + solving for uIce,vIce and prevent SEAICE_LSR or SEAICE_EVP to modify OB + values (consistent with an implicit method). + - add option OBCS_UVICE_OLD (in OBCS_OPTIONS.h, defined by default) to + return to older implementation (same results as before); note that none of + the "experimental" OBCS_SEAICE options are implemented in the new version. + - switch test exp. seaice_obcs to use new version (update output). +o pkg/seaice: + - add few stop (B-grid & CS-exch, B-grid & obcs, AUTODIFF & SEAICE_maskRHS) + - add parenthesis in Water-Ice drag calculations (evp & lsr); + - in seaice_lsr.F, simplify and fix AA3 setting in case sNx or sNy=1; + and fix case SEAICE_waterDrag=0 (but unstable if no or little seaice) + - update output of test exp lab_sea that uses evp or lsr (fwd+ad) + and global_ocean.cs32x15 (icedyn & seaice fwd+ad). +o pkg/layers: + - Include potential density as new coordinate (Thanks to David Munday) +o pkg/dic: + - Remove subroutine CALC_PCO2_APPROX_CO3 from carbon_chem.F and add + carbonate computation/output to CALC_PCO2_APPROX +o pkg/generic_advdiff + - make TAF generate code that vectorizes: + gad_dst3fl_adv_*.F, gad_u3c4_impl_r.F +o model/src: + - improve vectorization also for adjoint code of + solve_tridiagonal and solve_pentadiagonal. As this requires up to + 5 new 3D-fields, this code is only turned on for the combination of + ALLOW_AUTODIFF_TAMC and TARGET_NEC_SX. Also remove error message + for singular matrix check (only for TARGET_NEC_SX). + - remove hard-wiring of CPP-flag ALLOW_SOLVERS_KLOOPINSIDE to + ALLOW_AUTODIFF. This flag now needs to be set outside of + solve_tri/pentadiagonal.F +o pkg/exf: + - fix previous modif in exf_readparms.F (revision 1.66: treatment of + optional namelist EXF_NML_OBCS). + - add many debug-print (to trace down where pkg/cal error msg comes from) +o pkg/mom_vecinv/mom_vi_hdissip.F + - rearrange loops so that vectorization is improved and in particular + the adjoint routine vectorizes. The code is a little longer now, but I + think easier to read. + Note: CPP-flag MOM_VI_ORIGINAL_VISCA4 is not tested. +o pkg/seaice: + - allow to specify initial seaice velocity from files (uIceFile,vIceFile) +o pkg/exch2: + - alternative code (without tile x tile matrix) for cumulated (global) sum; + used as default; keep using matrix version in adjustment.cs-32x32x1. +o tools/build_options: + - after merging some MPI and/or multi-threads specific optfiles so that + the modified default optfile support both built (using genmake2 -mpi + for the MPI built, using genmake2 -omp for OpenMP built), + move now obsolete optfiles to tools/build_options/unsupported/ + (add copy of email sent to support list in doc/optfiles_changes.txt) + +checkpoint63c (2011/09/28) +o pkg/ebm: + - fix ebm_zonalmean.F for several procs in Ydir (nPy > 1); update output of + test exp. global_ocean_ebm (affects machine truncation, 12 digits for cg2d) +o pkg/seaice (SItracer pickup) +* Simple code for pickup of SItracer fields (#define ALLOW_SITRACER) + To be done: + - thorough testing of missing fields in S/R seaice_read_pickup + - thorough matching between generic tracers and specific fields used + (e.g. volume-weighted age, area-weighted age, salinity, ...) +o pkg/grdchk: + - all procs calculate and print finite-difference gradient (for testreport) + - change description output used by testreport; + - AD testing: also compare and report Finite-difference gradient; + => update all output_adm*.txt +o test exp. global_ocean_ebm: + - switch back useRealFreshWaterFlux (no problem anymore with run-off); + and turn on exactConserve; update (fwd) output. +o pkg/seaice (LSR): + - print to STDOUT number of solver iterations at SEAICE_monFreq interval. + - modify solver to deal cleanly with u/v switch at face edges (CS-grid): + * identical results with non-CS topology (but save up to half of EXCH calls) + * no much changes in cs-32 verification tests since no ice at face edges + (differences due to convergence criteria, since nb of iterations changed) + * expect some differences on lat-lon-cap grid. + update FWD results of global_ocean.cs32x15 test exp. (icedyn & seaice) +o pkg/exch2: + - add new CPP option to disable Cumul-Sum matrix (to save storage when using + many tiles); until alternative coding is added, just add a stop. +o pkg/kpp: + - move ptracer/salt correction to KPP diffusivity from calc_3d_diffusivity.F + to S/R KPP_CALC_DIFF_PTR (+ add tracer-index to argument list). +o optfiles: + - merged few MPI specific (*+mpi*) and multi-threads specific (*+mth) optfiles + into corresponding standard one, which now support different build: + (when using "genmake2 -mpi" for MPI built, "genmake2 -omp" for OpenMP built) + - add "darwin_amd64_gfortran" optfile (initially from Constantinos, tested by Gael) +o pkg/obcs (seaice): + - add missing EXCH for uIce,vIce after applying OBCS; + - remove S/R EXF_SWAPFFIELDS_X/Y , disable S/R EXF_SET_OBCS_X/Y by adding + argument = number of levels to S/R EXF_SWAPFFIELDS_XZ/YZ and S/R + EXF_SET_OBCS_XZ/YZ ; update obcs_exf_load.F; + - fix EXF_SET_OBCS_XZ/YZ for multi-threaded (assuming loaded arrays are shared) + - verification exp. seaice_obcs: split domain in 2 tiles ; add SIZE.h_mpi + and eedata.mth for MPI and mutlti-threaded testing. +o pkg/flt: + - make FLT EXCH working with multi-threads (and MPI + multi-threads); + - use buffer for floats IO (trajectories & profiles); size set in FLT_SIZE.h; + - fix pickup and trajectories + profiles output with multi-threads; + - add parameter to select group of fields to write along trajectory and + profile (default unchanged); + - only call PORT_RAND* function when flt_noise <> 0; + - add multi-threaded test for verification exp. flt_example. +o eesupp/src: + - fix S/R ALL_PROC_DIE for MPI+multi-threads. + - USE_OMP_THREADING: check that number of threads (from eedata) matches + actual number (from env.var. setting). o pkg/ebm: - fix multi-threading in ebm_area_t.F ; clean-up EBM.h header file. - add multi-threaded test for verification exp. global_ocean_ebm