4 |
Notes on tags used in MITgcmUV |
Notes on tags used in MITgcmUV |
5 |
============================== |
============================== |
6 |
|
|
7 |
|
o pkg/autodiff: ALLOW_AUTODIFF_WHTAPEIO |
8 |
|
Option to use mdsio_read_whalos.F/mdsio_write_whalos.F, |
9 |
|
rather than mdsio_readvector.F/mdsio_writevector.F, within |
10 |
|
adread_adwrite.F to read/write adjoint-related tapes to disk. |
11 |
|
o eesup and pkg/mdsio: ALLOW_WHIO |
12 |
|
|
13 |
|
Option to write/read 2D fields to files including tiles halos. |
14 |
|
The main purpose is for adjoint related "tape I/O". |
15 |
|
The secondary purpose is for debugging phases. |
16 |
|
|
17 |
|
A word on those rather specifically intended purposes. |
18 |
|
The code was meant to be stripped down to the bare minimum. |
19 |
|
Typically, there is a unique and basic mapping to the global |
20 |
|
buffer (tile by tile, then proc by proc, then level by level), and |
21 |
|
byteswaps and meta-files are omitted. Indeed the (*_WH*) code is not |
22 |
|
meant to generate user friendly outputs of various shapes and forms. |
23 |
|
Rather it is intended, for the knowledgeabale developers, |
24 |
|
to be largely independent of the many mdsio options, to allow |
25 |
|
outputs of tiles+halos in debugging phases and in adjoint runs, |
26 |
|
and to allow some flexibility in tuning (adoint) I/O depending on each |
27 |
|
file system behavior. With regard to tuning, most of the potential knobs |
28 |
|
(to leave files open, to allocate the I/O indep. of usesinglecpuio, |
29 |
|
to specify directories, to specify the typical size of I/O calls, etc.) |
30 |
|
are not yet included, and will be added if they prove useful. |
31 |
|
o pkg/seaice: |
32 |
|
fix uninitialised arrays in USE_ORIGINAL_SBI version of seaice_solve4temp.F |
33 |
|
o pkg/seaice: |
34 |
|
new S/R seaice_solve4temp.F: this is the merged version of Ian Fenty |
35 |
|
code (seaice_budget_ice_if.F) which enable to recover old standard |
36 |
|
code (seaice_budget_ice.F) when option USE_ORIGINAL_SBI is defined. |
37 |
|
temporary: add "#define USE_ORIGINAL_SBI" at the top of seaice_solve4temp.F. |
38 |
|
o model/src (rStar): |
39 |
|
use simple average (not area weighted) of rStarFac at U,V point when using |
40 |
|
vectorInvariant and selectKEscheme=1,3 (consistent with mom. vertical advection) |
41 |
|
update output of exp. fizhi-cs-32x32x40 and fizhi-cs-aqualev20. |
42 |
|
|
43 |
|
checkpoint62k (2010/09/17) |
44 |
|
o adjoint |
45 |
|
* changed verification/global_ocean.cs32x15 ; adjoint now tests |
46 |
|
Adams-Bashforth 3 + NLFS + RSTAR |
47 |
|
* some more modifs required for ECCO-GODAE v4 |
48 |
|
o pkg/seaice |
49 |
|
add SEAICE_AGE_VOL option to use volume instead of area for |
50 |
|
integrating SEAICE_AGE |
51 |
|
o model/src: |
52 |
|
first check-in of sigma & hybrid-sigma coordinate code ; also includes |
53 |
|
non-linear free-surface parts ; but pressure-gradient not yet coded. |
54 |
|
o pkg/ptracers: |
55 |
|
option for ptracer to convert Salt Relax into additional EmP. |
56 |
|
tested in exp. global_ocean.90x40x15.dwnslp (update output) |
57 |
|
|
58 |
|
checkpoint62j (2010/08/25) |
59 |
|
o pkg/ggl90: |
60 |
|
other minor changes: remove 3-D temp array gTKE (+ add argument myIter) |
61 |
|
o pkg/dic: car_flux.F, phos_flux.F |
62 |
|
- reduce number of exp-calls in order to improve performance |
63 |
|
- clean up a little (replace fields by their macros, etc., fix |
64 |
|
typo in CPP-flag) |
65 |
|
o pkg/ggl90: minor changes |
66 |
|
- GGL90diffKrS was removed --> always use GGL90diffKr |
67 |
|
- GGL90viscAr was removed --> replaced with GGL90viscArU, GGL90viscArV |
68 |
|
- hack of mxlMaxFlag=2 --> ensure mixing between first and second level |
69 |
|
(commented out for now) |
70 |
|
- change in max/min operations to ensure that smoothing is ok |
71 |
|
- smoothing of GGL90viscAr was moved to ggl90_calc.F (as done for GGL90diffKr) |
72 |
|
- always use diffKrNrT as background profile (i.e. never use diffKr field) |
73 |
|
o pkg/cfc: bug fix, AREA(i,j,1,bi,bj) -> AREA(i,j,bi,bj) |
74 |
|
o pkg/dic: |
75 |
|
- model/src/impldiff.F: if TARGET_NEC_SX, extend loop ranges for better |
76 |
|
vectorization |
77 |
|
- small bug fix in insol for the unlikely case of a cartesian grid |
78 |
|
- towards vectorizing the dic package: bio_export.F, car_flux.F, phos_flux.F |
79 |
|
+ now vectorize, which also removes adjoint recomputations; |
80 |
|
+ replace integer maksp1 by _RL maskp1 |
81 |
|
+ replace a**x by exp(x*log(a)) in phos_flux.F for speed |
82 |
|
|
83 |
|
checkpoint62i (2010/07/21) |
84 |
|
o pkg/flt: |
85 |
|
- with OBCS: stop floats which enter the OB region |
86 |
|
o model/src: |
87 |
|
- use tempVertAdvScheme & saltVertAdvScheme (instead of horizontal ones) |
88 |
|
for implicit vertical advection. |
89 |
|
o pkg/seaice: |
90 |
|
- account for melting in seaice age evolution. |
91 |
|
- remove double counting of seaice ridging effect on seaice age. |
92 |
|
|
93 |
|
checkpoint62h (2010/06/18) |
94 |
|
o pkg/exch2: |
95 |
|
- remove exch2_*xy_rx files: call was replaced with exch2_*3d_rx(myNz=1). |
96 |
|
o tools/genmake2 |
97 |
|
- remove "-pdefault" option: redundant with the use of "packages.conf"; |
98 |
|
- remove file pkg/pkg_default : set the default as a pkg group |
99 |
|
("default_pkg_list") defined in pkg/pkg_groups; |
100 |
|
- fix for case pkg/pkg_depend is missing (was hanging before); |
101 |
|
- add option "-pgroups" to specify a different pkg_groups than the default. |
102 |
o verification/adjustment.cs-32x32x1: |
o verification/adjustment.cs-32x32x1: |
103 |
add a "minimal" set-up test (code_min & input_min) which only compiles |
add a "minimal" set-up test (code_min & input_min) which only compiles |
104 |
eesupp/src and the 2 pkgs (exch2 & debug) which don't need model/src & inc. |
eesupp/src and the 2 pkgs (exch2 & debug) which don't need model/src & inc. |