On Mon, 2015-11-23 at 09:18 +1100, Cyril Bur wrote:
> On Fri, 20 Nov 2015 22:01:04 +1100
> Michael Ellerman wrote:
> > On Wed, 2015-11-18 at 14:26 +1100, Cyril Bur wrote:
> > > diff --git a/arch/powerpc/kernel/entry_64.S
> > > b/arch/powerpc/kernel/entry_64.S
> > > index
On Wed, 2015-11-18 at 14:26 +1100, Cyril Bur wrote:
> Currently the FPU, VEC and VSX facilities are lazily loaded. This is
> not a
> problem unless a process is using these facilities.
I would prefer to say facilities are "enabled" and registers are
"loaded". You're mixing the two here.
>
On Fri, 20 Nov 2015 22:01:04 +1100
Michael Ellerman wrote:
> On Wed, 2015-11-18 at 14:26 +1100, Cyril Bur wrote:
> > diff --git a/arch/powerpc/kernel/entry_64.S b/arch/powerpc/kernel/entry_64.S
> > index c8b4225..46e9869 100644
> > --- a/arch/powerpc/kernel/entry_64.S
> >
On Wed, 2015-11-18 at 14:26 +1100, Cyril Bur wrote:
> diff --git a/arch/powerpc/kernel/entry_64.S b/arch/powerpc/kernel/entry_64.S
> index c8b4225..46e9869 100644
> --- a/arch/powerpc/kernel/entry_64.S
> +++ b/arch/powerpc/kernel/entry_64.S
> @@ -210,7 +210,54 @@ system_call:
Currently the FPU, VEC and VSX facilities are lazily loaded. This is not a
problem unless a process is using these facilities.
Modern versions of GCC are very good at automatically vectorising code, new
and modernised workloads make use of floating point and vector facilities,
even the kernel