Re: [wiget@pld.org.pl] libc/3237: symbol __udivdi3, version GLIBC_2.0 not defined in file libc.so.6 with link time reference

[Franz Sirl <Franz dot Sirl-kernel at lauterbach dot com>]

On Friday 12 April 2002 20:09, Geoff Keating wrote:
> > From: Franz Sirl <Franz.Sirl-kernel@lauterbach.com>
> > Date: Fri, 12 Apr 2002 18:41:15 +0200
> >
> > The same just happened to me, I guess this means I have to add divdi3.c
> > to sysdeps/powerpc/Makefile?
>
> Yes, if it was once exported it has to be included forever.

Hmm, Jakub may have a point though, I wonder why this happened at all:

This is the gcc-2.95.4 compiled glibc sendmail was compiled against:

[fsirl@entropy:~/rh70/glibc/RPMS/ppc]$ readelf -a ./lib/libc-2.2.5.so |egrep 
'(moddi3|divdi3)'
  3444: 001123a0  1176 FUNC    LOCAL  DEFAULT   10 __divdi3
  3769: 00113094   912 FUNC    LOCAL  DEFAULT   10 __umoddi3
  3775: 00112c9c  1016 FUNC    LOCAL  DEFAULT   10 __udivdi3
  4414: 00112878  1060 FUNC    LOCAL  DEFAULT   10 __moddi3

And is the now-installed gcc-3.1 compiled glibc:

[fsirl@entropy:~/rh70/glibc/RPMS/ppc]$ readelf -a /lib/libc-2.2.5.so |egrep 
'(moddi3|divdi3)'
  2367: 0010310c   220 FUNC    LOCAL  HIDDEN   10 __divdi3
  2692: 00104140    44 FUNC    LOCAL  HIDDEN   10 __umoddi3
  2698: 00103c58    36 FUNC    LOCAL  HIDDEN   10 __udivdi3
  3337: 001036ac   232 FUNC    LOCAL  HIDDEN   10 __moddi3

Shouldn't have ld used the symbols from libgcc.a instead of the LOCAL symbol 
from glibc? Or does LOCAL/HIDDEN make the difference here? But then why 
should it take __divdi3/__moddi3 from libgcc.a, but __udivdi3/__umoddi3 from 
libc.so?

Puzzled,
Franz.