------- Comment #10 from law at redhat dot com 2006-02-16 03:52 -------
Subject: Re: [4.2 Regression]
25_algorithms/prev_permutation/1.cc on powerpc{64,}-linux and
powerpc-darwin
On Thu, 2006-02-16 at 02:59 +0000, pinskia at gcc dot gnu dot org wrote:
>
> ------- Comment #9 from pinskia at gcc dot gnu dot org 2006-02-16 02:59
> -------
> (In reply to comment #8)
> > > Which looks more like VRP messing up.
> > Can you check something for me. Is sizetype an unsigned type on
> > this platform? And what type is ivtmp?
> sizetype should be "unsigned long int" but I am not 100% sure.
>
> unsigned int ivtmp.60;
Thanks. There's a reasonable possibility this is the same bug
I'm already working in. It's a latent problem in how we set up
TYPE_MAX_VALUE on targets which have an unsigned sizetype.
jeff
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=26304
