[PATCH, fortran] PR20441 -finit-local-zero
Asher Langton
langton2@llnl.gov
Fri Sep 14 12:44:00 GMT 2007
Hi all,
I somehow missed Thomas's original message entirely. I will take a
look at equivalenced variables tonight.
-Asher
On 9/12/07, Tobias Burnus <burnus@net-b.de> wrote:
> Asher Langton wrote:
> > Here's an updated patch.
>
> Thomas König wrote:
> > Equivalenced variables are not initialized by your patch:
> > [...]
> > This behavior is different from g77, so I think it should be either
> > documented, or the behavior changed to match g77.
>
> Could you send an updated patch (e.g. with documentation update only)
> rather soonish. I think we can then still include this patch in stage 3
> (starts today) as the initial patch has been submitted before stage 3
> has started. I really would like to see this patch in 4.3.x (I have some
> programs which don't initialize local variables). I think we can worry
> about finer points (equivalence, derived types) later. Unless you have a
> patch ready which initializes also equivalence.
>
> (I have to admit that on my system Thomas' equivalence example did not
> work: the variables were always zero with any options.)
>
> Tobias
>
More information about the Fortran
mailing list