[Intel-gfx] [PATCH v4 08/23] drm/i915/adl_p: Handle TC cold

[Souza, Jose]

On Mon, 2021-05-17 at 16:18 -0700, José Roberto de Souza wrote:
> On Mon, 2021-05-17 at 17:53 +0300, Imre Deak wrote:
> > On Fri, May 14, 2021 at 08:10:20PM -0700, Matt Roper wrote:
> > > From: José Roberto de Souza <jose.souza at intel.com>
> > > 
> > > On ADL-P TC cold is exited and blocked when legacy aux is powered,
> > > that is exacly the same of what ICL need for static TC ports.
> > > 
> > > TODO: When a TBT hub or monitor is connected it will cause TBT and
> > > legacy aux to be powered at the same time, hopefully this will not
> > > cause any issues but if it do, some rework will be needed.
> > > 
> > > v2:
> > >  - skip icl_tc_port_assert_ref_held() warn on, adl-p uses aux to
> > >    block TC cold
> > > 
> > > BSpec: 55480
> > > Cc: Imre Deak <imre.deak at intel.com>
> > > Signed-off-by: José Roberto de Souza <jose.souza at intel.com>
> > > Signed-off-by: Clinton Taylor <Clinton.A.Taylor at intel.com>
> > > Signed-off-by: Matt Roper <matthew.d.roper at intel.com>
> > > ---
> > >  drivers/gpu/drm/i915/display/intel_display_power.c | 3 ++-
> > >  drivers/gpu/drm/i915/display/intel_tc.c            | 2 +-
> > >  2 files changed, 3 insertions(+), 2 deletions(-)
> > > 
> > > diff --git a/drivers/gpu/drm/i915/display/intel_display_power.c b/drivers/gpu/drm/i915/display/intel_display_power.c
> > > index 54c6d65011ee..29d2f1d0cffd 100644
> > > --- a/drivers/gpu/drm/i915/display/intel_display_power.c
> > > +++ b/drivers/gpu/drm/i915/display/intel_display_power.c
> > > @@ -551,7 +551,8 @@ static void icl_tc_port_assert_ref_held(struct drm_i915_private *dev_priv,
> > >  	if (drm_WARN_ON(&dev_priv->drm, !dig_port))
> > >  		return;
> > >  
> > > -	if (DISPLAY_VER(dev_priv) == 11 && dig_port->tc_legacy_port)
> > > +	if (IS_ALDERLAKE_P(dev_priv) ||
> > 
> > A TC port reference is held whenever enabling the port's AUX power
> > domain (so whenever blocking TC-cold for instance), so this shouldn't be
> > needed.
> 
> I'm kinda of rusty in this area so I could be making a mistake but I don't see how tc_lock is held or tc_link_refcount is != 0 when enabling aux to do
> a DPCD transaction when the port is no driving image to the sink.
> Will do some experiments and update here.

Ops please ignore the above.
Yep we don't need it, when coming from tc_cold_block() the mutex is held and when doing a DPCD transaction intel_tc_port_lock() is called before get
the aux power well.

So will drop it, thanks for pointing it out.

> 
> > 
> > OTOH, the !aux_powered check in intel_tc_port_reset_mode() needs this
> > exception, since there TC-cold must be blocked and so AUX will be
> > enabled as opposed to other platforms.
> > 
> > Also, in icl_tc_phy_aux_power_well_enable() we need to avoid the power
> > well enabling timeout error message, since it won't get enabled unless
> > something is actually plugged to the TC connector.
> 
> Agree with both above, planning to squash patch.txt to fix it and make sure we don't have similar issues with future platforms.
> 
> > 
> > > +	    (DISPLAY_VER(dev_priv) == 11 && dig_port->tc_legacy_port))
> > >  		return;
> > >  
> > >  	drm_WARN_ON(&dev_priv->drm, !intel_tc_port_ref_held(dig_port));
> > > diff --git a/drivers/gpu/drm/i915/display/intel_tc.c b/drivers/gpu/drm/i915/display/intel_tc.c
> > > index e325463acddd..1b108dea5fed 100644
> > > --- a/drivers/gpu/drm/i915/display/intel_tc.c
> > > +++ b/drivers/gpu/drm/i915/display/intel_tc.c
> > > @@ -28,7 +28,7 @@ tc_cold_get_power_domain(struct intel_digital_port *dig_port)
> > >  {
> > >  	struct drm_i915_private *i915 = to_i915(dig_port->base.base.dev);
> > >  
> > > -	if (DISPLAY_VER(i915) == 11)
> > > +	if (IS_ALDERLAKE_P(i915) || DISPLAY_VER(i915) == 11)
> > >  		return intel_legacy_aux_to_power_domain(dig_port->aux_ch);
> > >  	else
> > >  		return POWER_DOMAIN_TC_COLD_OFF;
> > > -- 
> > > 2.25.4
> > > 
>