[Mesa-dev] [PATCH 1/2] intel/common: Add an is_kabylake field to gen_device_info

[Jason Ekstrand]

On Tue, Nov 8, 2016 at 1:53 PM, Matt Turner <mattst88 at gmail.com> wrote:

> On Tue, Nov 8, 2016 at 1:21 PM, Jason Ekstrand <jason at jlekstrand.net>
> wrote:
> > Most of the 3-D engine Kaby Lake is identical to Sky Lake.  However,
> there
>
> While hyphenating 3D looks a little odd to me, Skylake is definitely
> just a single word. (Strangely, Kaby Lake is indeed two words)
>
> > are a few small differences that we need to be able to detect.
> >
> > Signed-off-by: Jason Ekstrand <jason at jlekstrand.net>
> > ---
> >  src/intel/common/gen_device_info.c | 14 +++++++++-----
> >  src/intel/common/gen_device_info.h |  1 +
> >  2 files changed, 10 insertions(+), 5 deletions(-)
> >
> > diff --git a/src/intel/common/gen_device_info.c
> b/src/intel/common/gen_device_info.c
> > index 30df0b2..3ff98f0 100644
> > --- a/src/intel/common/gen_device_info.c
> > +++ b/src/intel/common/gen_device_info.c
> > @@ -427,6 +427,10 @@ static const struct gen_device_info
> gen_device_info_bxt_2x6 = {
> >   * There's no KBL entry. Using the default SKL (GEN9) GS entries value.
> >   */
> >
> > +#define KBL_FEATURES \
>
> We don't have subgen FEATURES #defines for anything else. bxt, for
> instance just sets .is_broxton in its couple of fields. Not wrong, but
> doesn't seem particularly necessary for a single field.
>

We do for Haswell which is why I did it this way.


> I'd prefer to just put .is_kabylake in the KBL structs, unless you've
> got further plans.
>

Sure.  I don't care much either way.


> With that fixed, both patches are
>
> Reviewed-by: Matt Turner <mattst88 at gmail.com>
>

Thanks
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.freedesktop.org/archives/mesa-dev/attachments/20161108/1eab7e82/attachment.html>