This is the mail archive of the
gcc-patches@gcc.gnu.org
mailing list for the GCC project.
Re: {PING] [PATCH] Sign extension elimination
On Thu, Apr 20, 2006 at 07:13:58PM +0100, Thiemo Seufer wrote:
> H. J. Lu wrote:
> > On Thu, Apr 20, 2006 at 10:23:21AM -0700, Mark Mitchell wrote:
> > >
> > > The x86-64 work is probably going to have to wait for 4.3.
> > >
> > > >> What is it about the x86-64 instruction set that is so different from
> > > >> PowerPC in this regard? Are any of our other x86 experts concerned
> > > >> about this pass?
> > > >
> > > > What is so different for x86-64 is all 32bit register moves are
> > > > automatically zero extended to 64bit.
> > >
> > > Whereas on other architectures a 32-bit register move is either (a)
> > > sign-extending, or (b) leaves the high bits undefined?
> > >
> >
> > I would say "leave the high bits unchanged". But Sparc, MIPS, PPC
> > experts have to confirm it.
>
> MIPS requires them to be sign-extended for proper operation.
If MIPS can use the same SEE approach as x86-64, it should be easy to
support MIPS once the x86-64 SEE works.
H.J.