[MITgcm-devel] changes in the_main_loop.F

[Jean-Michel Campin]

Hi,

I think that the reason why my testreport fail to compile for
lab_sea & obcs_ctrl (both using pkg/ecco) is related to 
a weakness of the Makefile dependencies regarding AD_CONFIG.h
(because I run 2 times testreport one after the other, with no clean).

Cheers,
Jean-Michel

On Thu, Mar 01, 2012 at 12:07:46PM -0800, Matthew Mazloff wrote:
> Hi Jean-Michel
> 
> Perhaps just add
> ALLOW_ADJOINT_RUN
> 
> to
> MITgcm/verification/obcs_ctrl/code_ad/ECCO_CPPOPTIONS.h
> 
> Would that fix it?
> 
> -Matt
> 
> 
> On Mar 1, 2012, at 12:00 PM, Jean-Michel Campin wrote:
> 
> >Hi Gael,
> >
> >I am going to revert to previous version (=1.120) of "model/src/
> >the_main_loop.F",
> >since
> >1) the Makefile is now missing some dependencies (we talk about this
> >yearterday), in particular the_main_loop.f is missing dependencies to
> >checkpoint_lev3_directives.h and checkpoint_lev2_directives.h
> >(checked with hs94.1x64x5/build/Makefile)
> >
> >2) few experiments are failing (see:
> >http://mitgcm.org/testing/results/2012_03/tr_danton_20120301_0/summary.txt
> >), hs94.1x64x5
> >  isomip & isomip.htd
> >  tutorial_global_oce_biogeo
> >  tutorial_global_oce_optim
> >  tutorial_tracer_adjsens
> >) because of this error:
> >>forrtl: severe (193): Run-Time Check Failure. The variable
> >>'adthe_main_loop_$ILEV_1' is being used without being defined
> >
> >3) still need to check what happens to lab_sea & obcs_ctrl (fail
> >to compile).
> >
> >Cheers,
> >Jean-Michel
> >
> >
> >_______________________________________________
> >MITgcm-devel mailing list
> >MITgcm-devel at mitgcm.org
> >http://mitgcm.org/mailman/listinfo/mitgcm-devel
> 
> 
> _______________________________________________
> MITgcm-devel mailing list
> MITgcm-devel at mitgcm.org
> http://mitgcm.org/mailman/listinfo/mitgcm-devel