[MITgcm-devel] An other compilation problem in seaice_growth.F

Jean-Michel Campin jmc at ocean.mit.edu
Sat Dec 24 08:43:40 EST 2011


Hi, 

I think that even if pkg/seaice is not "usable" with exf & 
#undef ALLOW_ATM_TEMP, it would be good that it compiles.
Seems to be not a very recent problem, but with checkpoint62k
(from Sept 17, 2010), the only 2 S/R that would not compile
when ALLOW_ATM_TEMP is undef were: seaice_growth_if.F
and seaice_budget_ice_if.F

Cheers,
Jean-Michel

On Fri, Dec 23, 2011 at 06:19:36PM -0500, Jean-Michel Campin wrote:
> Hi,
> 
> I have an other problem when compiling seaice_growth.F 
> (this time not related to #undef SEAICE_GROWTH_LEGACY):
> 
> With #undef ALLOW_ATM_TEMP in EXF_OPTIONS.h, 
> EVAP is not available and some of the diags calculation
> at the end of the routine are stopping the compilation:
> for diag 'SIatmFW ':
> 1771      &          - EVAP(I,J,bi,bj)*( ONE - AREApreTH(I,J) )
> and also for diag 'SIhl    ':
> 1800            tmpscal1= EVAP(I,J,bi,bj)*( ONE - AREApreTH(I,J) )
> or:
> 1805            tmpscal1= EVAP(I,J,bi,bj)*( ONE - AREApreTH(I,J) )
> 
> Jean-Michel
> 
> On Fri, Dec 02, 2011 at 03:37:45PM -0500, Jean-Michel Campin wrote:
> > Hi,
> > 
> > I cannot use the non-legacy version of seaice_growth.F
> > with #undef SHORTWAVE_HEATING since it does not compile:
> > 
> > > gfortran -fconvert=big-endian -fimplicit-none -fPIC  -O3 -funroll-loops -c seaice_growth.f
> > > seaice_growth.f:5593.28:
> > >            tmpscal2=SWFRACB * a_QSWbyATM_open(I,J)                     
> > >                            1
> > > Error: Symbol 'swfracb' at (1) has no IMPLICIT type
> > 
> > This is not a problem with #define SEAICE_GROWTH_LEGACY,
> > so for now, I will use the legacy version.
> > 
> > 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



More information about the MITgcm-devel mailing list