This is the mail archive of the
gcc@gcc.gnu.org
mailing list for the GCC project.
Re: Local optimization on i386 ?
- To: law at cygnus dot com
- Subject: Re: Local optimization on i386 ?
- From: Marc Espie <espie at quatramaran dot ens dot fr>
- Date: Tue, 30 Nov 1999 14:38:09 +0100
- Cc: egcs at egcs dot cygnus dot com
- Organization: Ecole Normale Superieure de Paris
In article <26996.943919594@upchuck> you write:
>
> In message <19991130004402.A1867@liafa1.liafa.jussieu.fr>you write:
> > Is this normal behavior on a 386 ?
>Unfortunately, yes.
Oh, ok... I meant, is there a good reason why the code is like this.
> > I don't know enough about 386, but should't this be
> > subl $20, %esp ?
> >
> > I assume I'm missing some obvious reason why there are two separate
> > subs...
>Because they're emitted by two completely different parts of the compiler.
So this looks like a misoptimization bug...
Hum... I believe I could coerce gas into peep-holing that :)
Isn't there a peep-hole pass or equivalent in gcc that could handle this ?