On Thu, 2018-11-29 at 15:03 -0800, Rodrigo Vivi wrote:
> On Mon, Nov 26, 2018 at 04:37:07PM -0800, José Roberto de Souza
> wrote:
> > According to eDP spec, sink can required specific selective update
> > granularity that source must comply.
> > Here caching the value if required and checking source supports it.
> > 
> > Cc: Rodrigo Vivi <rodrigo.vivi@intel.com>
> > Cc: Dhinakaran Pandiyan <dhinakaran.pandiyan@intel.com>
> > Signed-off-by: José Roberto de Souza <jose.souza@intel.com>
> > ---
> >  drivers/gpu/drm/i915/i915_drv.h  |  1 +
> >  drivers/gpu/drm/i915/intel_psr.c | 32
> > ++++++++++++++++++++++++++++++++
> >  2 files changed, 33 insertions(+)
> > 
> > diff --git a/drivers/gpu/drm/i915/i915_drv.h
> > b/drivers/gpu/drm/i915/i915_drv.h
> > index f763b30f98d9..cbcd85af95bf 100644
> > --- a/drivers/gpu/drm/i915/i915_drv.h
> > +++ b/drivers/gpu/drm/i915/i915_drv.h
> > @@ -506,6 +506,7 @@ struct i915_psr {
> >  	ktime_t last_exit;
> >  	bool sink_not_reliable;
> >  	bool irq_aux_error;
> > +	u16 su_x_granularity;
> >  };
> >  
> >  enum intel_pch {
> > diff --git a/drivers/gpu/drm/i915/intel_psr.c
> > b/drivers/gpu/drm/i915/intel_psr.c
> > index 7607a58a6ec0..9215c9052381 100644
> > --- a/drivers/gpu/drm/i915/intel_psr.c
> > +++ b/drivers/gpu/drm/i915/intel_psr.c
> > @@ -257,6 +257,21 @@ static u8
> > intel_dp_get_sink_sync_latency(struct intel_dp *intel_dp)
> >  	return val;
> >  }
> >  
> > +static u16 intel_dp_get_su_x_granulartiy(struct intel_dp
> > *intel_dp)
> > +{
> > +	u16 val = 0;
> > +	ssize_t r;
> > +
> > +	if (!(intel_dp->psr_dpcd[1] & DP_PSR2_SU_GRANULARITY_REQUIRED))
> > +		return val;
> > +
> > +	r = drm_dp_dpcd_read(&intel_dp->aux, DP_PSR2_SU_X_GRANULARITY,
> > &val, 2);
> > +	if (r != 2)
> > +		DRM_WARN("Unable to read DP_PSR2_SU_X_GRANULARITY\n");
> > +
> > +	return val;
> > +}
> > +
> >  void intel_psr_init_dpcd(struct intel_dp *intel_dp)
> >  {
> >  	struct drm_i915_private *dev_priv =
> > @@ -311,6 +326,8 @@ void intel_psr_init_dpcd(struct intel_dp
> > *intel_dp)
> >  		if (dev_priv->psr.sink_psr2_support) {
> >  			dev_priv->psr.colorimetry_support =
> >  				intel_dp_get_colorimetry_status(intel_d
> > p);
> > +			dev_priv->psr.su_x_granularity =
> > +				intel_dp_get_su_x_granulartiy(intel_dp)
> > ;
> >  		}
> >  	}
> >  }
> > @@ -525,6 +542,21 @@ static bool intel_psr2_config_valid(struct
> > intel_dp *intel_dp,
> >  		return false;
> >  	}
> >  
> > +	if (dev_priv->psr.su_x_granularity) {
> > +		/*
> > +		 * HW will always send full lines in SU blocks, so X
> > will
> > +		 * always be 0 and we only need to check the width to
> > validate
> > +		 * horizontal granularity.
> > +		 * About vertical granularity HW works by SU blocks
> > starting
> > +		 * at each 4 lines with height of 4 lines, what eDP
> > states
> > +		 * that sink should support.
> > +		 */
> > +		if (crtc_hdisplay % dev_priv->psr.su_x_granularity) {
> > +			DRM_DEBUG_KMS("PSR2 not enabled, HW can not
> > match sink SU granularity requirement\n");
> > +			return false;
> 
> I wonder if regardless this bit we still need to do a sort of
> check anyway because spec states:
> 
> "
> Sets the grid pattern granularity in the X direction.
> A value of 0 indicates that no X-coordinate granularity requirement
> exists other than
> the standard restrictions, wherein the:
> •
>  Starting X-coordinate must be evenly divisible by 16
>  •
>  Rectangle width must be evenly divisible by 4
> "

Hum, the x will always be 0 so it will always be divisible by 16 but
the width could not be divisible by 4 in the odd resolutions. I will
add this check, thanks.


> 
> Also, why we are just checking X granularity and not Y? (0074h)
> 
> (maybe it would be useful to introduce along with previous patch
> that I had just reviewed)


It is in the comment above but coping from eDP spec:

A value of 00h, 01h, 02h, or 04h should be supported by the Sink device
to ensure
interoperability with various Source devices. A value of 08h or 10h may
be considered
for system-specific implementations.

Our height is multiple of 4 so we are safe even if sink have a y
granularity set.

> 
> > +		}
> > +	}
> > +
> >  	return true;
> >  }
> >  
> > -- 
> > 2.19.2
> >