This is the mail archive of the
gcc@gcc.gnu.org
mailing list for the GCC project.
Re: More altivec array initialisation troubles
- From: Daniel Egger <degger at fhm dot edu>
- To: Aldy Hernandez <aldyh at redhat dot com>
- Cc: Graham Stott <grahams at redhat dot com>,GCC Developer Mailinglist <gcc at gcc dot gnu dot org>
- Date: 23 Feb 2002 12:13:09 +0100
- Subject: Re: More altivec array initialisation troubles
- References: <FA24B416-27E4-11D6-BCEB-000393750C1E@redhat.com>
Am Fre, 2002-02-22 um 23.39 schrieb Aldy Hernandez:
> > static const vector signed short test4[4] =
> > {
> > (vector signed short){1, 2, 3, 4, 5, 6, 7, 8},
> > (vector signed short){11, 12, 13, 14, 15, 16, 17, 18},
> > (vector signed short){21, 22, 23, 24, 25, 26, 27, 28},
> > (vector signed short){31, 32, 33, 34, 35, 36, 37, 38}
> > };
> >
> > does. This should be fixed anyhow as it's fairly simple to get wrong
> > and the warning is sort of confusing, not to mention that the
> > behaviour is not really sensible (at least to me).
> perhaps errors should be more sane though.
Have to be. The warnings don't really make sense and the generated code
is simply wrong and doesn't make any sense at all, no matter how it
really works.
--
Servus,
Daniel