[MITgcm-devel] removing gT,gS and gPtr

[Jean-Michel Campin]

Hi Samar,

Regarding the current pkg/matrix, I made the changes (last Monday, 08/11)
so that it works without gPtr (I did check, with a diff of the
pkg/matrix output files, between before and after the changes).
I guess If you have other pieces of code (not checked-in) that
use gPtr, it should be possible to modify these in the same way.

Cheers,
Jean-Michel

On Fri, Aug 15, 2014 at 03:36:00AM -0400, Samar Khatiwala wrote:
> Hi Jean-Michel,
> 
> My pkg/matrix uses gPtr from PTRACERS_FIELDS.h but as long as the variable is available in a common 
> block in some other header file I can use that.
> 
> Thanks,
> 
> Samar
> 
> On Aug 14, 2014, at 6:20 PM, Jean-Michel Campin <jmc at ocean.mit.edu> wrote:
> 
> > Hi Oliver, Patrick and others,
> > 
> > I think I am ready to remove gPtr from the common block
> > (PTRACERS_FIELDS.h): it's not longer used but did not check 
> > if it's used within pkg/darwin.
> > 
> > And regarding gT,gS, I can remove them from DYNVARS.h
> > (except if USE_OLD_EXTERNAL_FORCING is defined, but this case
> > is not supported for adjoint built) but that I don't know
> > what to do in autodiff_store.F & autodiff_restore.F:
> > I can comment out the 2 lines there, but should I also shift (by -2) 
> > all the indices in big array "StoreDynVars3D" (probably yes) ?
> > And also, is this part of code tested in our standard
> > verification experiments ?
> > 
> > Cheers,
> > Jean-Michel
> > 
> > _______________________________________________
> > MITgcm-devel mailing list
> > MITgcm-devel at mitgcm.org
> > http://mitgcm.org/mailman/listinfo/mitgcm-devel
>