[ghemical-devel] Re: redhat 9 g++ has no dynamic_cast problems

Michael Banck mbanck@gmx.net
Tue, 15 Apr 2003 11:54:33 +0200


On Tue, Apr 15, 2003 at 12:20:21PM +0300, Tommi Hassinen wrote:
> Yesterday and today I installed our RedHat 9 machine, and after testing
> both v1.00 and v1.50 versions I can tell that the RedHat 9 compiler

gah :(

> [thassine@shrike thassine]$ g++ --version
> g++ (GCC) 3.2.2 20030222 (Red Hat Linux 3.2.2-5)

Debian ships a 3.2.3-prerelease in unstable right now. Hmm.

> has no problems dealing with dynamic_cast<>() statements. The both
> versions work fine without any patching (like they are in CVS). I haven't
> tried with --enable-mpqc yet but my guess is that it works as well.

I've just stumbled over http://bugs.debian.org/188943 ("g++-3.2: code
using dynamic_cast causes segfaults when -fno-rtti is used"), but I guess that
does not apply to use, right?

>        Mandrake-9.0            ok
>        RedHat 8.0              ok
>        RedHat 9.0              ok
>        Debian "testing"        problems
>        Debian "unstable"       problems

Does RedHat-8 use g++-3.2 already? I'll try to find the RH-9 Changelog
and see if they perhaps applied some patches to fix this problem.

> > > OK, I've tried to automakefy libghemical. The appended patch
> 
> Wow!  :)  Thanks! I haven't had time to look it yet, but I will this week
> or the next one, and I'll try to apply automake in the new mopac7 library
> as well.

I did something like that for mopac7 a while (maybe a year) ago. I can
try to dig that up if you want.
 
> This might be a dumb question, but you mentioned that aclocal.m4 is
> now obsolete as well. So, how the use of automake will affect the
> "custom" settings of the configure-script (for example the detection
> of sc-config that we have and which works fine); can we still keep
> them as they are now?

Custom m4-macros go into acinclude.m4, I put the sc-config stuff in
there. Running 'aclocal' then generates aclocal.m4 out of configure.in
and acinclude.m4, AFAICT. As always, it's debatable whether the
generated aclocal.m4 should be put into CVS or not :)

Michael