[Intel-gfx] [PATCH 3/6] drm/i915/icl: Set TC type to unknown when a sudden disconnection happen

[Souza, Jose]

On Tue, 2018-10-30 at 19:51 +0200, Ville Syrjälä wrote:
> On Wed, Oct 10, 2018 at 02:35:05PM -0700, José Roberto de Souza
> wrote:
> > Otherwise it would be in a inconsistent state as port is
> > disconnected
> > but with a valid tc type.
> > 
> > Cc: Paulo Zanoni <paulo.r.zanoni at intel.com>
> > Signed-off-by: José Roberto de Souza <jose.souza at intel.com>
> > ---
> >  drivers/gpu/drm/i915/intel_dp.c | 7 ++++---
> >  1 file changed, 4 insertions(+), 3 deletions(-)
> > 
> > diff --git a/drivers/gpu/drm/i915/intel_dp.c
> > b/drivers/gpu/drm/i915/intel_dp.c
> > index d3f31103b8ec..d3af6aa1959a 100644
> > --- a/drivers/gpu/drm/i915/intel_dp.c
> > +++ b/drivers/gpu/drm/i915/intel_dp.c
> > @@ -4863,6 +4863,9 @@ static void icl_update_tc_port_type(struct
> > drm_i915_private *dev_priv,
> >  			      type_str);
> >  }
> >  
> > +static void icl_tc_phy_disconnect(struct drm_i915_private
> > *dev_priv,
> > +				  struct intel_digital_port *dig_port);
> > +
> >  /*
> >   * This function implements the first part of the Connect Flow
> > described by our
> >   * specification, Gen11 TypeC Programming chapter. The rest of the
> > flow (reading
> > @@ -4917,9 +4920,7 @@ static bool icl_tc_phy_connect(struct
> > drm_i915_private *dev_priv,
> >  	if (dig_port->tc_type == TC_PORT_TYPEC &&
> >  	    !(I915_READ(PORT_TX_DFLEXDPSP) &
> > TC_LIVE_STATE_TC(tc_port))) {
> >  		DRM_DEBUG_KMS("TC PHY %d sudden disconnect.\n",
> > tc_port);
> > -		val = I915_READ(PORT_TX_DFLEXDPCSSS);
> > -		val &= ~DP_PHY_MODE_STATUS_NOT_SAFE(tc_port);
> > -		I915_WRITE(PORT_TX_DFLEXDPCSSS, val);
> > +		icl_tc_phy_disconnect(dev_priv, dig_port);
> >  		return false;
> >  	}
> 
> Patches 1-3
> Reviewed-by: Ville Syrjälä <ville.syrjala at linux.intel.com>
> 

Thanks, I will rebase and reoder the patches so the one without
reviewed-by goes last and someone can merge the first 5.

> But icl_tc_port_connected() still looks somewhat busted. If the
> port changes from legacy|typec to tbt we won't configure the
> safe mode knob correctly.

For what I understood, firmware will hang in the disconnected state
until driver change to safe mode when comming form legacy|typec.
But I will give a try into spliting into a connector for each type like
you suggested.

> 
> I think we want something like this:
> 
> + if (!is_legacy && !is_typec)
> + 	icl_tc_phy_disconnect(dev_priv, intel_dig_port);
> + 
>   if (!is_legacy && !is_typec && !is_tbt) {
> - 	icl_tc_phy_disconnect(dev_priv, intel_dig_port);
>   	return false;
>   }
> 
> >  
> > -- 
> > 2.19.1
> > 
> > _______________________________________________
> > Intel-gfx mailing list
> > Intel-gfx at lists.freedesktop.org
> > https://lists.freedesktop.org/mailman/listinfo/intel-gfx