This is the mail archive of the fortran@gcc.gnu.org mailing list for the GNU Fortran project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

Re: variable definition context bug?


Andrew,

Thanks for the files (I have been able to use the *.mod
files directly). Indeed your code is very touchy, however
I have found a pattern:

I can get rid of the simple assignments of the kind
abundancesWork=abundancesParent+abundancesWork
but if I touch anything controlling the program flow
(call, if, where, ...) the errors go away:

for instance if I try to remove

    where (moleculesValue < 0.0d0)
!        moleculesValue=0.0d0
    end where

or the 'else' line in

    if (thisNode%componentExists(componentIndex)) then
       thisIndex=Tree_Node_Hot_Halo_Index(thisNode)
       Tree_Node_Hot_Halo_Outflowed_Mass_Standard=thisNode%components(thisIndex)%properties(outflowedMassIndex,propertyValue)
    else
!        Tree_Node_Hot_Halo_Outflowed_Mass_Standard=0.0d0
    end if

This reminds me of a pr which had a similar behavior, but
I cannot ind a way to locate it;-(

What you can do on your side is to try the same approach on
the code generating the modules in order to reduce their sizes
too. Then it would be possible to have a single file test
of managable size.

Cheers,

Dominique


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]