[MITgcm-devel] broken adjoint?

[Jean-Michel Campin]

Hi Patrick,

Sorry for this.
I remember running the ad test on faulks after changing
the 2 monitors (exf & seaice), and it passed.
(and the automatic test did also).
Then I thought it would be OK ...

Jean-Michel

On Fri, Jun 30, 2006 at 07:11:03PM -0400, Patrick Heimbach wrote:
> 
> Dimitris,
> 
> someone moved exf_monitor from a place 
> invisible to taf to a place visible to taf.
> flow directives needed to be updated.
> 
> Should be ok now (update exf_ad.flow).
> 
> -p.
> 
> 
> On Fri, 2006-06-30 at 09:53, Dimitris Menemenlis wrote:
> > Patrick, something seems wrong with adjointability of current version of code.
> > Taf errors seem to come from subroutine thermodynamics and exf_monitor but could 
> > also be what I just checked in?  Any suggestions what to look for?
> > Gotta run to catch the game.  Talk later, Dimitris
> > 
> > 
> > > 1197344     CADJ STORE uFld(:,:), vFld(:,:), wFld(:,:) = comlev1_bibj_k, key=kkey, byte=isbyte
> > > TAF ERROR                                                                      12^
> > > keyword REC, KIND, or SHAPE expected
> > > 1197345     CADJ STORE uTrans(:,:), vTrans(:,:)        = comlev1_bibj_k, key=kkey, byte=isbyte
> > > TAF ERROR                                                                      12^
> > > keyword REC, KIND, or SHAPE expected
> > > 1197346     CADJ STORE xA(:,:), yA(:,:)                = comlev1_bibj_k, key=kkey, byte=isbyte
> > > TAF ERROR                                                                      12^
> > > keyword REC, KIND, or SHAPE expected
> > > TAF ERROR    Code of subroutine  exf_monitor not seen and no flow information set,
> > > TAF ERROR    assuming all arguments are input and output
> > > Segmentation fault
> > > 144.977u 0.293s 2:25.25 100.0%  0+0k 0+0io 0pf+0w
> > > 
> > > -------------------------------------------
> > >  something went wrong
> > >  Please send email to support at fastopt.de
> > > -------------------------------------------
> > _______________________________________________
> > MITgcm-devel mailing list
> > MITgcm-devel at mitgcm.org
> > http://mitgcm.org/mailman/listinfo/mitgcm-devel
> -- 
> --------------------------------------------------------
> Patrick Heimbach   Massachusetts Institute of Technology
> FON: +1/617/253-5259                  EAPS, Room 54-1518
> FAX: +1/617/253-4464             77 Massachusetts Avenue
> mailto:heimbach at mit.edu               Cambridge MA 02139
> http://www.mit.edu/~heimbach/                        USA
> 
> _______________________________________________
> MITgcm-devel mailing list
> MITgcm-devel at mitgcm.org
> http://mitgcm.org/mailman/listinfo/mitgcm-devel