[MITgcm-devel] [MITgcm-cvs] MITgcm/doc CVS Commit

[Jean-Michel Campin]

Hi Gael,

I think I know where the Pb is:
your new S/R ctrl_cost_driver.F has a wrong include (cal.h) which
is causing Pb in experiments that does not compile pkg/cal
(should produce some errors at makedepend stage ? and when using CPP
at make step).

In fact, the TAF ad test on acesgrid using pgi is also showing a lot of 
fails (experiments that don't compile pkg/cal).
Regarding the OpenAD test, it's likely to be the same issue
(since none of the tested OpenAD exp is using pkg/cal).

And since we are here (in ctrl_cost_driver.F), could we follow the
same usual model convention regarding the order of there 3 arguments:
 myTime, myIter, myThid 
and finally, local var "iarr" is not going to be used
unless ALLOW_GENTIM2D_CONTROL is defined (independently of the 2 other options:
 ALLOW_GENARR3D_CONTROL & ALLOW_GENARR2D_CONTROL )

Cheers,
Jean-Michel

On Thu, Sep 18, 2014 at 09:58:50AM -0400, Jean-Michel Campin wrote:
> Hi Gael, Patrick and Oliver,
> 
> Seems that some OpenAD tests did fail last night (but all OK before).
> The error msg (at compile time) is:
> > ERROR: AssemblerError: parser failed :unable to assemble contents of a unit named ctrl_cost_driver
> 
> This occur on:
> 1) acresgrid (mpi & non mpi), all experiments
> 2) stomp , only 1 experiment (OpenAD) is tested there, and fails
>    with same error.
> 
> Now what is rather strange is that the OpenAD test on baudelaire
> (Using Patrick OpenAD built) did not show any problem.
> Is this a matter of different version of OpneAD ?
> 
> Cheers,
> Jean-Michel
> 
> On Wed, Sep 17, 2014 at 04:58:54PM -0400, Gael Forget wrote:
> > Update of /u/gcmpack/MITgcm/doc
> > In directory baudelaire:/srv/scratch/gforget/MITgcm/doc
> > 
> > Modified Files:
> > 	tag-index 
> > Log Message:
> > 
> > document latest cahnges
> > 
> > 
> > 
> > _______________________________________________
> > MITgcm-cvs mailing list
> > MITgcm-cvs at mitgcm.org
> > http://mitgcm.org/mailman/listinfo/mitgcm-cvs
> 
> _______________________________________________
> MITgcm-devel mailing list
> MITgcm-devel at mitgcm.org
> http://mitgcm.org/mailman/listinfo/mitgcm-devel