[Intel-gfx] [PATCH 0/7] Future preparation patches

[Daniel Vetter]

On Fri, Jul 18, 2014 at 02:04:56PM +0100, Damien Lespiau wrote:
> On Fri, Jul 18, 2014 at 04:53:34PM +0530, Jindal, Sonika wrote:
> > 
> > 
> > On 7/18/2014 4:26 PM, Damien Lespiau wrote:
> > >On Fri, Jul 18, 2014 at 11:04:03AM +0530, sonika.jindal at intel.com wrote:
> > >>From: Sonika Jindal <sonika.jindal at intel.com>
> > >>
> > >>This series prepares future platform enabling by changing HAS_PCH_SPLIT to more
> > >>appropriate check since the code accessed may not have anything to do with
> > >>having PCH or not.
> > >
> > >Hi Sonika,
> > >
> > >HAS_PCH_SPLIT() is true for Ironlake (gen 5) as it's paired with the
> > >Ibex Peak PCH.
> > >
> > >In various patches, the condition needs to be INTEL_INFO(dev)->gen < 5
> > >then.
> > >
> > I am sorry, my understanding was that HAS_PCH_SPLIT is equivalent to
> > (gen > 5 && !(IS_VALLEYVIEW) )
> > So, is it like, HAS_PCH_SPLIT is equivalent to (gen >=5 && !(IS_VALEYVIEW))
> 
> Yes, indeed!

Since the patches need to be respun anyway, I vote for the introduction of
HAS_GMCH_DISPLAY for the gen < 5 || IS_VLV condition. Since vlv (i.e. byt
+ chv) have essentially inherited the gmch display block from gen3/4. I
think that would help the readability of the code quite a bit.

Comments?
-Daniel
-- 
Daniel Vetter
Software Engineer, Intel Corporation
+41 (0) 79 365 57 48 - http://blog.ffwll.ch