> What's that?
> 
> cc   -o indent indent.o io.o lexi.o parse.o pr_comment.o args.o
> ld: indent.o: RRS text relocation at 0x38f6 for "__toupper_tab_"
> ld: indent.o: RRS text relocation at 0x38ea for "__ctype_"
> ld: io.o: RRS text relocation at 0x5304 for "__ctype_"

I'm also seeing this on a current (as of today) x86 system.  There are
two parts to this horror.  1) The above gripe at link time 2) after
updating the running system to the latest current, every program that
uses the dynamic libc comes up with this gripe:

/usr/libexec/ld.so: symbol ___sF at 0x4008b2d0 in /usr/lib/libc.so.12.33 changed size: expected 264, actual 0
/usr/libexec/ld.so: symbol __ctype_ at 0x4008b424 in /usr/lib/libc.so.12.33 changed size: expected 4, actual 0
/usr/libexec/ld.so: symbol _optind at 0x4008a030 in /usr/lib/libc.so.12.33 changed size: expected 4, actual 0

Some programs will refuse to run, others will just have a very noisy
startup.  Luckily enough of the build system worked to allow a second
build.  Things got *much* saner as soon as the new lib's were dropped
into place.  I'm now at the point were all the new programs from the
second pass have been installed, and it appears things are back to
normal.

Now dare I run rdist to push the new system to all the other machines?
I'm somewhat concerned that things will die as soon as /usr/lib/ gets
pushed across the net.

-wolfgang
-- 
Wolfgang Rupprecht    <wolfgang@wsrcc.com>     http://www.wsrcc.com/wolfgang/
Cut down on spam.   Install ORBS support:    http://www.dorkslayers.com/orbs/