swap() on x86 platforms.

[amylaar@cygnus.co.uk]

> Here's GCC-2.7.2.  (I formatted the table differently.)  You can see
> that the older GCC could actually do induction variable elimination in
> C at -O2.  The newer egcs-121697 cannot do it -- perhaps someone will
> dig around to find out why.

No need to dig around - it was disabled on purpose, to allegedly fix the
bugs associated with biv elimination.  I think the overall result is
to disable about 95% of the biv elimination and about 98% of the biv
elimination bugs.

It is possible to solve this properly, but it needs some register
lifetime information at loop optimization time.