trouble using gfortran

[Sylvain Aguinaga]

Thanks everybody for your help. I'll check if I have the last version
of gfortran (i didn't install it myself). The last remark of Paul
looks very interresting, indeed I don't initialize all the variables,
it's a quite big code (two-phase flow code) but I think it's worth
trying that. I'll check all that stuff.
Thanks again everybody.

Sylvain

2006/4/19, Paul Thomas <paulthomas2@wanadoo.fr>:
> Silvain,
>
> Steve and Tobias have already given you good advice - your English is
> fine but you need to explain more what you are trying to do and what
> goes wrong with it.
>
> As general bit of advice, it is worth while checking if you assume in
> your code that all variables and pointers are initialized to zero/null.
> gfortran does not do this for you but many other compilers do.  In the
> last six months, I have found this to be the biggest hinderance, by far,
> to people getting existing code up and running with gfortran.  HJ Lu,
> myself and now Jerry DeLisle are all trying to get the quantum chemistry
> package, tonto, to run.  HJ has found that it is full of assumed NULL
> initializations.
> I have found the same, recently with a tokamak equilibrium code, proteus.
>
> Who would vote for -fzero-all-variables and -fnull-pointers?
>
> Paul
>
>


--
*************************************************
Sylvain Aguinaga
Ingénieur Aérodynamique
Doctorant PSA Peugeot Citroën/LEA/IMFT/CSTB
06.78.07.32.39
sylvain.aguinaga@gmail.com