This is the mail archive of the
gcc-help@gcc.gnu.org
mailing list for the GCC project.
Re: Problem when using optimization on aix 5.2 and gcc 4.1.1
- From: Andrew Haley <aph-gcc at littlepinkcloud dot COM>
- To: "Thomas Mittelstaedt" <T dot Mittelstaedt at cadenas dot de>
- Cc: gcc-help at gcc dot gnu dot org, "j dot starek at cadenas dot de" <j dot starek at cadenas dot de>
- Date: Tue, 22 May 2007 10:11:49 +0100
- Subject: Re: Problem when using optimization on aix 5.2 and gcc 4.1.1
- References: <op.tsoshl0c3gkkc3@cprog027> <18001.54238.150546.172512@zebedee.pink> <op.tspzr6j03gkkc3@cprog027>
Thomas Mittelstaedt writes:
> On Mon, 21 May 2007 19:16:14 +0200, Andrew Haley
> <aph-gcc@littlepinkcloud.COM> wrote:
>
> ...
> > I suggest, to begin with, that you try running your program under
> > Valgrind. Also, be very sure to compile your program with -Wall and
> > fix whatever warnings you come across.
>
> valgrind on aix? Don't think so.
Sorry, didn't notice that.
> Compiling with -Wall is not easy for us right now, too. There simply are
> too many!! :-(
Well, that's your clue, isn't it?
> Could there be a problem with our build of libstdc++?
Oh, of course that's possible. But here we have to deal in
probabilities. While it is possible that libstdc++ is at fault, the
fact that the senistivity to optimization is when building your
program, not when building libstdc++, suggests otherwise.
Andrew.