This patch adds Altivec support for RAID-6, if appropriately configured
on the ppc or ppc64 architectures. Note that it changes the compile
flags for ppc64 in order to handle -maltivec correctly; this change was
vetted on the ppc64 mailing list and OK'd by paulus.
Signed-off-by: H. Peter Anvin <[email protected]>
"H. Peter Anvin" <[email protected]> wrote:
>
> $(obj)/raid6int1.c: $(src)/raid6int.uc $(src)/unroll.pl FORCE
So we require that raid6int.uc propagate through the system with the x bit
set.
I wonder if it would be safer to stick a $(SHELL) in there?
Andrew Morton <[email protected]> wrote:
>
> "H. Peter Anvin" <[email protected]> wrote:
> >
> > $(obj)/raid6int1.c: $(src)/raid6int.uc $(src)/unroll.pl FORCE
>
> So we require that raid6int.uc propagate through the system with the x bit
> set.
err, I think I misread that. Ignore.
On Wed, 17 Nov 2004 22:52:55 -0800, H. Peter Anvin <[email protected]> wrote:
> This patch adds Altivec support for RAID-6, if appropriately configured
> on the ppc or ppc64 architectures. Note that it changes the compile
> flags for ppc64 in order to handle -maltivec correctly; this change was
> vetted on the ppc64 mailing list and OK'd by paulus.
>
> Signed-off-by: H. Peter Anvin <[email protected]>
>
hpa, are you aware of any other routines which should benefit from altivec?
--
Greetz, Antonio Vargas aka winden of network
Las cosas no son lo que parecen, excepto cuando parecen lo que si son.
Followup to: <[email protected]>
By author: Antonio Vargas <[email protected]>
In newsgroup: linux.dev.kernel
>
> hpa, are you aware of any other routines which should benefit from altivec?
>
Presumably the XOR code used by RAID-5, and quite possibly some of the
cryptography stuff. Unlike most SIMD instruction sets, it should be
possible to write AES using Altivec.
-hpa
On Fri, 19 Nov 2004 13:04:11 +0100, Antonio Vargas <[email protected]> wrote:
> On Fri, 19 Nov 2004 05:05:54 +0000 (UTC), H. Peter Anvin <[email protected]> wrote:
> > Followup to: <[email protected]>
> > By author: Antonio Vargas <[email protected]>
> > In newsgroup: linux.dev.kernel
> > >
> > > hpa, are you aware of any other routines which should benefit from altivec?
> > >
> >
> > Presumably the XOR code used by RAID-5, and quite possibly some of the
> > cryptography stuff. Unlike most SIMD instruction sets, it should be
> > possible to write AES using Altivec.
>
> I'll take a crack at the RAID-5 stuff first then.
>
Here we go... this file is compile-tested on gcc 3.3 from userspace,
it can serve as an starting point for testing.
Signed-off-by: Antonio Vargas <[email protected]>
--
Greetz, Antonio Vargas aka winden of network
Las cosas no son lo que parecen, excepto cuando parecen lo que si son.