This is the mail archive of the
fortran@gcc.gnu.org
mailing list for the GNU Fortran project.
Re: diagnostics
Il giorno mer, 27/08/2008 alle 20.25 +0200, Thomas Koenig ha scritto:
> On Wed, 2008-08-27 at 15:20 +0200, Salvatore Filippone wrote:
> > Hmm, mistery deepens: compiling with 4.3.0 instead of 4.3.1 does NOT
> > give an error.....
>
> Hi Salvatore,
>
> this makes this a (potential) regression.
>
> Could you file a PR, with a test case that can reproduce the problem
> (even if it's large)?
I'll see what can be done tomorrow; if you are interested, I can give
you all the source codes offline (it's really impractical to upload them
to bugzilla), I would prefer cutting down to a self contained test, but
I don't know how long it will take me to do it.
Would the output of -fdump-{whatever} be helpful (since it can be
generated quickly)?
>
>
> > > --------------------------------------------
> > > Given that the type in question is defined only once in a well-defined
> > > place, I am wondering why the compiler is throwing this error at me. Any
> > > help ?
>
> Just a stab in the dark, because we really don't have a lot of
> information... Did you try throwing away all your *.mod files and
> recompiling?
Yes, I did, multiple times.
Thanks
Salvatore