[OE-core] [PATCH] tune-xscale.inc: restore specific tuning

[Khem Raj]

On Tue, Aug 14, 2018 at 7:24 AM Richard Purdie
<richard.purdie at linuxfoundation.org> wrote:
>
> On Tue, 2018-08-14 at 16:10 +0200, Andrea Adami wrote:
> > Since commit
> >
> >  a11bdc36a1be
> >  tune-*: define more generic DEFAULTTUNE to share feed between
> > machines
> >
> > we are using armv5te to share binary feeds.
> >
> > After 008085450417 gcc-8: Fix spurious mcpu/march conflict for xscale
> > we can use again the specific optimizations.
> >
> > As of today, there are no public feeds for the legacy devices so
> > let's prefer
> > the optimal tuning.
> >
> > Signed-off-by: Andrea Adami <andrea.adami at gmail.com>
> > ---
> >  meta/conf/machine/include/tune-xscale.inc | 2 +-
> >  1 file changed, 1 insertion(+), 1 deletion(-)
> >
> > diff --git a/meta/conf/machine/include/tune-xscale.inc
> > b/meta/conf/machine/include/tune-xscale.inc
> > index 0d07333..cd08b93 100644
> > --- a/meta/conf/machine/include/tune-xscale.inc
> > +++ b/meta/conf/machine/include/tune-xscale.inc
> > @@ -1,4 +1,4 @@
> > -DEFAULTTUNE ?= "armv5te"
> > +DEFAULTTUNE ?= "xscale"
>
> Does this make any real world difference except for duplicating feeds?

Usually it should, since compiler should be able to use
xscale specific instructions during codegen  , it will be interesting
to see numbers to measure the gain.

>
> Cheers,
>
> Richard
> --
> _______________________________________________
> Openembedded-core mailing list
> Openembedded-core at lists.openembedded.org
> http://lists.openembedded.org/mailman/listinfo/openembedded-core