[MITgcm-devel] [MITgcm-cvs] MITgcm/pkg/seaice CVS Commit

[Martin Losch]

Hi there,

the recomputations that I was alluding to look like this:
TAF RECOMPUTATION LOOP WARNING DOLOOP_STMT ad_input_code.f:384544 in seaice_lsr
 extensive recomputations are required.
 The responsible variables are : av,bi,bj,bv,cv,i,rhsv,seaicemaskv,sny,vice,vrt1,vrt2,wfav,zero

And they go away with staf -version 2.1.38. I didn't check when they appear (currently we are at 2.3.4 but 2.2.38 didn't work either), but it's pretty obvious that the current staf gets confused because "zero" is a constant fortran parameter, so is "sny", i,bi,bj are loop indices, etc.
Maybe something to report to fastOpt?

Martin


On Dec 21, 2012, at 2:21 PM, Martin Losch <mlosch at forge.csail.mit.edu> wrote:

> Update of /u/gcmpack/MITgcm/pkg/seaice
> In directory forge:/tmp/cvs-serv23962/pkg/seaice
> 
> Modified Files:
> 	seaice_lsr.F 
> Log Message:
> - move calculation of matrix coefficients into separate routine; for now
>  this routine remains in the same file; the main purpose of this change
>  is to be able to reuse the coe in s/r seaice_preconditioner
> - reminder to myself I noticed a lot of recomputations for seaice_lsr,
>  but the do not have anything to do with this change, needs to be looked at
> 
> 
> _______________________________________________
> MITgcm-cvs mailing list
> MITgcm-cvs at mitgcm.org
> http://mitgcm.org/mailman/listinfo/mitgcm-cvs