4 |
Notes on tags used in MITgcmUV |
Notes on tags used in MITgcmUV |
5 |
============================== |
============================== |
6 |
|
|
7 |
o pkg/exf: |
checkpoint62g (2010/05/19) |
8 |
|
o eesupp/src (exch-1): |
9 |
|
- split exch_rx.F into Forward S/R (exch1_rx.F) and Adjoint S/R (exch1_rx_ad.F) |
10 |
|
and remove argument "simulationMode"; same thing for all exch_*rx_cube.F |
11 |
|
(renamed to exch1_*rx_cube.F). |
12 |
|
- change accordingly exch_*_rx.template and AD version in pkg/autodiff |
13 |
|
to call new exch1 S/R ; update Makefile and .cvsignore. |
14 |
|
- replace call to exch2_xy_rx S/R with exch2_*3d_rx(myNz=1) (will remove |
15 |
|
exch2_xy_rx later) |
16 |
|
Note: after updating your code, need to run "make clean" in eesupp/src |
17 |
|
o test experiment bottom_ctrl_5x5: |
18 |
|
new output_adm.txt with fewer grdchk & less monitor output. |
19 |
|
o model/src: |
20 |
|
remove the call to ini_cori in initialise_varia.F (just keep the usual one |
21 |
|
in initialise_fixed.F). More consistent with coriolis & horizontal grid |
22 |
|
beeing treated in the same way (ini_grid is not called a 2nd time). |
23 |
|
o pkg/exf: |
24 |
- rewrite exf_bulk_largeyeager04.F for better vectorization, changes |
- rewrite exf_bulk_largeyeager04.F for better vectorization, changes |
25 |
global_ocean.cs32x15/input.icedyn (11 digits of agreement), updated |
global_ocean.cs32x15/input.icedyn (11 digits of agreement), updated |
26 |
- fix adjoint (used worked for ALLOW_ATM_WIND defined, works for this now) |
- fix adjoint (used worked for ALLOW_ATM_WIND defined, works for this now) |
34 |
- Message mode: move RecvAck setting (indicator of buffer being ready) |
- Message mode: move RecvAck setting (indicator of buffer being ready) |
35 |
from send_put to recv_get S/R (was useless before, but not sure if |
from send_put to recv_get S/R (was useless before, but not sure if |
36 |
it's much more useful now); |
it's much more useful now); |
37 |
- switch the order of sync: MPI-proc 1rst and then threads; |
- Switch the order of sync: MPI-proc 1rst and then threads; |
38 |
- take out spin-waiting code (#undef EXCH_USE_SPINNING), use BARRIER instead. |
- take out spin-waiting code (#undef EXCH_USE_SPINNING), use BARRIER instead. |
39 |
|
- fix AD code for communication mode ='get' (but not used). |
40 |
o model/src: |
o model/src: |
41 |
initialise dPhiHydX,dPhiHydY inside calc_grad_phi_hyd.F (so that TAF does |
initialise dPhiHydX,dPhiHydY inside calc_grad_phi_hyd.F (so that TAF does |
42 |
not drop it in S/R mddynamics, producing NaN in hs94.1x64x5 AD-test exp). |
not drop it in S/R mddynamics, producing NaN in hs94.1x64x5 AD-test exp). |