[dataflow] remove try_auto_increment from regmove

Jeffrey Law law@redhat.com
Mon Nov 6 17:31:00 GMT 2006

On Mon, 2006-11-06 at 06:27 +0100, Paolo Bonzini wrote:
> > I have no problem with this patch.  I would like to see it go in. 
> > However, the ppc only has pre* auto insns and the x86 has none. 
> >   
> Yes, you're right about the ppc (the x86 bootstrap was only to be sure 
> the patch triggered no warning, which would break the bootstrap on all 
> machines).  I'll hold on.
If you wanted to do a before/after assembly comparison, you might
try a generic m68k target.  Or an SH target, which IIRC was the
original processor regmove was designed to help.


More information about the Gcc-patches mailing list