[MITgcm-devel] making model_ad_diff.list shorter

[Jean-Michel Campin]

Hi Jean,

Oliver told me that initialise_fixed.F is part of the routines
that OpenAD differentiates. So what I was trying was wrong:
I removed from the diff-list few routines that are called from 
initialise_fixed.F, which was OK with TAF but will not work
with OpenAD.
We can leave it like this, and one day might put something 
little bit more consistent on the TAF side.

Thanks,
Jean-Michel

On Mon, Feb 10, 2014 at 06:56:25AM -0600, Jean Utke wrote:
> On 02/08/2014 10:42 AM, Jean-Michel Campin wrote:
> >I had to revert the changes since it's messing up the OpenAD build
> >(not sure how to go around this).
> Hi Jean-Michel,
> 
> could you describe "messing up"  in some detail?
> I know that at some point I introduced a second pass over the list
> because there were some (perceived?) inconsistencies in that list.
> That second pass may be the source of the problem.
> 
> Jean
> 
> -- 
> Jean Utke
> Argonne National Lab. / MCS (240-2154)
> 9700 S Cass Ave., Argonne, IL, 60439, USA
> phone +1 630 252 4552 / mobile  +1 630 363 5753
> 
> 
> _______________________________________________
> MITgcm-devel mailing list
> MITgcm-devel at mitgcm.org
> http://mitgcm.org/mailman/listinfo/mitgcm-devel