[Mitgcm-support] Re: not clear for me

[mitgcm-support at dev.mitgcm.org]

The problem is (as usual) related to the adjoint.
There are some partial recomputation issues in this routines.
It seems that these issues are benign, so don't cause real harm for now.
Will tell you more about it when I am back.

Regarding the duplicate routines,
I was going to suggest to have a thrd directory, e.g. mom_shared/
which cotains routines that are shared between both packages.

Greetings from rainy Nice
-Patrick



Jean-Michel Campin wrote:
> Hi,
> 
> Since we have a clean initialization of kappaRV & kappaRU 
> in dynamics.F, at the beginning of the bi,bj loop from which
> mom_fluxform or mom_vecinv & mom_u,v_bottomdrag.F are called,
> I don't understand exactly thoses comments, at the 
> beginning of pkg/mom_fluxform/mom_v_bottomdrag.F :
> 
> 
>>C $Header: /u/gcmpack/MITgcm/pkg/mom_fluxform/mom_v_bottomdrag.F,v 1.7 2003/03/07 04:40:44 heimbach Exp $
>>C $Name:  $
>>
>>#include "CPP_OPTIONS.h"
>>
>>crg butterflies
>>crg using undefined values of kapparv in mom_v_bottomdrag
>>crg mom_v_bottomdrag must have iMin,iMax,jMin,jMax as arguments
>>crg and only use values being calculated in calc_viscosity
>>crg There are probably many more routines in mom_fluxform and mom_vecin
>>crg using undefined values.
>>
>>CBOP
>>C !ROUTINE: MOM_V_BOTTOMDRAG
> 
> 
> By the way, the 2 versions of the S/R in mom_fluxform and mom_vecinv
> are becoming very different, and its sometime tricky to know which
> one will be used. And since it seems difficult to really "disable" one
> of those 2 packages, why don't we decide to always compile one 
> (with all the common S/R in it) and make the 2nd one really optional ?
> 
> Jean-Michel


-- 
_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/
Patrick Heimbach ... Massachusetts Institute of Technology
FON: +1/617/253-5259 .................. EAPS, Room 54-1518
FAX: +1/617/253-4464 ............. 77 Massachusetts Avenue
mailto:heimbach at mit.edu ............... Cambridge MA 02139
http://www.mit.edu/~heimbach/ ..................... U.S.A.