Guillaume Cottenceau wrote:
>
> Stefan van der Eijk <[EMAIL PROTECTED]> writes:
>
> [...]
>
> > > * Sun Apr 01 2001 Guillaume Cottenceau <[EMAIL PROTECTED]> 1.1.8-4mdk
> > >
> > > - re-integrate 1.2.0 backported patches, in a non-broken way this time
> > > (hopefully)
>
> [...]
>
> > I'm still experiencing some problems building this package on my
> > alpha... Please take a look at this:
>
> [...]
>
> > gcc -shared SDL.lo SDL_error.lo SDL_fatal.lo -Wl,--whole-archive
> > main/.libs/libarch.al audio/.libs/libaudio.al video/.libs/libvideo.al
> > events/.libs/libevents.al cdrom/.libs/libcdrom.al
> > thread/.libs/libthread.al timer/.libs/libtimer.al
> > endian/.libs/libendian.al file/.libs/libfile.al
> > hermes/.libs/libhermes.al -Wl,--no-whole-archive main/.libs/libarch.al
> > audio/.libs/libaudio.al video/.libs/libvideo.al
> > events/.libs/libevents.al cdrom/.libs/libcdrom.al
> > thread/.libs/libthread.al timer/.libs/libtimer.al
> > endian/.libs/libendian.al file/.libs/libfile.al
> > hermes/.libs/libhermes.al -L/usr/lib /usr/lib/libesd.so
> > /usr/lib/libaudiofile.so -lm -L/usr/X11R6/lib -lX11 -lXext -lXxf86vm
> > -lXxf86dga -lc -Wl,-soname -Wl,libSDL-1.0.so.0 -o
> > .libs/libSDL-1.0.so.0.6.2
> > collect2: ld terminated with signal 11 [Segmentation fault], core dumped
> > /usr/bin/ld: mmxp_32.lo: invalid string offset 1836278062 >= 257 for
> > section `.strtab'
>
> This I doubt the patch on 1.1.8 could be the origin...
It's all quite strange... I have been able to compile 1.1.8-1mdk before
(the result is:
ftp://ftp.free.fr/pub/Distributions_Linux/Mandrake-devel/cooker/alpha/Mandrake/RPMS/libSDL1.1-1.1.8-1mdk.alpha.rpm
and
ftp://ftp.free.fr/pub/Distributions_Linux/Mandrake-devel/cooker/alpha/Mandrake/RPMS/libSDL1.1-devel-1.1.8-1mdk.alpha.rpm
) so the thing has compiled before... Now, I've got 1mdk through 4mdk on
my system, the 4mdk was rebuilt by the robot, and you saw the result.
I'm rebuilding the other ones now so i can compare the buildoutputs to
see whats the difference. I now think that it's not the SDL package, but
something else in the system that might have changed (new gcc
perhaps?)...
Stefan
>
> --
> Guillaume Cottenceau - http://us.mandrakesoft.com/~gc/
