libtool piecewise linking

Alexandre Oliva aoliva@redhat.com
Fri Aug 22 20:29:00 GMT 2003


On Aug 22, 2003, Marc Espie <espie@nerim.net> wrote:

> Yeah, but if that's the only argument against linker scripts, which DO
> fix an actual, genuine problem, then it's a pretty thin one.

It would introduce a hidden regression in the libtool testsuite, so it
wouldn't be acceptable for libtool CVS, and therefore it is not
acceptable for us here.

There is a better way to do it, but it takes a bit more effort.  My
argument is not against using linker scripts.  It's against using them
in a fragile way.  Fragile software is bad.  If we were to refuse to
accept such files, it wouldn't be as much of a problem.  But accepting
them and then failing in incomprehensible ways is exactly the reason
why this patch is not acceptable.  However, I wouldn't like to see it
gratuitously reject files that used to be acceptable, or that would be
acceptable if the command line didn't get too big.

-- 
Alexandre Oliva   Enjoy Guarana', see http://www.ic.unicamp.br/~oliva/
Red Hat GCC Developer                 aoliva@{redhat.com, gcc.gnu.org}
CS PhD student at IC-Unicamp        oliva@{lsd.ic.unicamp.br, gnu.org}
Free Software Evangelist                Professional serial bug killer



More information about the Gcc-patches mailing list