[MITgcm-devel] gad_calc_rhs

[Jean-Michel Campin]

Hi Jean,

Thanks a lot for fixing this (I saw this morning tests went well).

And sorry to have disturbed your holiday. I was ready to tolerate 
a broken OpenAD code for 1 day (there was many little changes
and wanted to have 1 full good test with TAF first).

I did the minimum (added data.autodiff in verification/OpenAD/input_oad)
but was not sure how to proceed with the other pieces.

Also, I realised that I made some changes during the last 2.5 years 
in some source files but forgot to update the local version in
OpenAD/code_oad_all/. I will try to import these changes made in
 gad_implicit_r.F
 ini_linear_phisurf.F
 initialise_fixed.F
(not much new in write_state.F)

All the Best,
Jean-Michel

On Wed, Jul 04, 2012 at 10:17:13PM -0500, Jean Utke wrote:
> Hi Jean-Michel,
> 
> Souviens-toi du jour du repos pour le sanctifier !
> 
> Today is July 4th  and you make changes to the code anyway  :-)
> 
> Just FYI,  I added the new AUTODIFF_PARAMS  file included by
> gad_calc_rhs  to
> the list of files to be converted to modules or else
> we get an out-of-order  USE statement.
> Actually I think I could do away with the whole common block
> to module converions because some years ago I added logic
> to OpenAD to deal with common blocks.   But I never had the
> time to try it out and streamline the mitgcm build part.
> 
>  Best wishes,
> 
> Jean
> 
> -- 
> Jean Utke
> Argonne National Lab. / MCS (240-1156)
> 9700 S Cass Ave., Argonne, IL, 60439, USA
> phone +1 630 252 4552 / mobile  +1 630 363 5753
>