[Intel-gfx] [PATCH V2] drm/i915: (VLV2) Fix the hotplug detection bits

[Ville Syrjälä]

On Tue, Jan 21, 2014 at 06:45:54PM +0100, Daniel Vetter wrote:
> On Tue, Jan 21, 2014 at 10:22:31AM -0700, Todd Previte wrote:
> > These bits are in reverse order in the header from those defined in
> > the specification. Change the bit positions for ports B and D to
> > correctly match the spec.
> > 
> > - Added sign-off
> > 
> > Signed-off-by: Todd Previte <tprevite at gmail.com>
> > 
> > ---
> >  drivers/gpu/drm/i915/i915_reg.h | 4 ++--
> >  1 file changed, 2 insertions(+), 2 deletions(-)
> > 
> > diff --git a/drivers/gpu/drm/i915/i915_reg.h b/drivers/gpu/drm/i915/i915_reg.h
> > index 10ecf90..2d77b51 100644
> > --- a/drivers/gpu/drm/i915/i915_reg.h
> > +++ b/drivers/gpu/drm/i915/i915_reg.h
> > @@ -2083,9 +2083,9 @@
> >   * Please check the detailed lore in the commit message for for experimental
> >   * evidence.
> >   */
> > -#define   PORTD_HOTPLUG_LIVE_STATUS               (1 << 29)
> > +#define   PORTD_HOTPLUG_LIVE_STATUS               (1 << 27)
> >  #define   PORTC_HOTPLUG_LIVE_STATUS               (1 << 28)
> > -#define   PORTB_HOTPLUG_LIVE_STATUS               (1 << 27)
> > +#define   PORTB_HOTPLUG_LIVE_STATUS               (1 << 29)
> 
> As Jani explained, this will break g4x. And it might very well be that
> this particular lie has been carried forward until baytrail, so I think we
> must cross check Bspec first with the reality of shipping platforms.

FYI I just checked this on Jani's ELK, and it agrees with the spec.
Port B is bit 29 and port C is bit 28. Both are HDMI/DVI ports on the
board. No DP connectors present unfortunately. Which explains why
things work since we don't check the live status on HDMI ports.

-- 
Ville Syrjälä
Intel OTC