[PATCH] drm/i915/hdcp: Disable HDCP Line Rekeying for HDCP2.2 on HDMI

[Kandpal, Suraj]

> -----Original Message-----
> From: Borah, Chaitanya Kumar <chaitanya.kumar.borah at intel.com>
> Sent: Tuesday, April 30, 2024 12:05 PM
> To: Kandpal, Suraj <suraj.kandpal at intel.com>; intel-gfx at lists.freedesktop.org
> Subject: RE: [PATCH] drm/i915/hdcp: Disable HDCP Line Rekeying for HDCP2.2
> on HDMI
> 
> Hi Suraj,
> 
> > -----Original Message-----
> > From: Kandpal, Suraj <suraj.kandpal at intel.com>
> > Sent: Wednesday, April 24, 2024 10:42 AM
> > To: intel-gfx at lists.freedesktop.org
> > Cc: Borah, Chaitanya Kumar <chaitanya.kumar.borah at intel.com>; Kandpal,
> > Suraj <suraj.kandpal at intel.com>
> > Subject: [PATCH] drm/i915/hdcp: Disable HDCP Line Rekeying for HDCP2.2
> > on HDMI
> >
> > Disable HDCP Line Rekeying when HDCP ver > 1.4 and when we are on
> HDMI
> > TMDS operation for DISPLAY_VER >= 14.
> >
> > --v2
> > -Wa to be mentioned in comments not in commit message [Jani] -Remove
> > blankline [Jani]
> >
> > --v3
> > -No need to write what is being done in comments when code is self
> > explanatory [Jani]
> >
> > --v4
> > -Add comment regarding need of this WA when in TMDS mode [Chaitanya]
> -
> > Write in chicken register for MTL [CHaitanya]
> >
> > Bspec: 49273
> > Bspec: 69964
> > Signed-off-by: Suraj Kandpal <suraj.kandpal at intel.com>
> > ---
> >  drivers/gpu/drm/i915/display/intel_hdcp.c | 26
> +++++++++++++++++++++++
> >  drivers/gpu/drm/i915/i915_reg.h           |  4 ++++
> >  2 files changed, 30 insertions(+)
> >
> > diff --git a/drivers/gpu/drm/i915/display/intel_hdcp.c
> > b/drivers/gpu/drm/i915/display/intel_hdcp.c
> > index d5ed4c7dfbc0..0d8ae6962d34 100644
> > --- a/drivers/gpu/drm/i915/display/intel_hdcp.c
> > +++ b/drivers/gpu/drm/i915/display/intel_hdcp.c
> > @@ -30,6 +30,30 @@
> >  #define KEY_LOAD_TRIES	5
> >  #define HDCP2_LC_RETRY_CNT			3
> >
> > +/*
> > + * WA: 16022217614
> > + * Disable HDCP line rekeying if we are using HDMI and in
> > + * TMDS mode.
> > + */
> 
> Thank you for the change. My idea was more to document the fact that here
> we assume that HDMI is in TDMS mode (which is currently true but might
> change later) and not the WA itself. More on that below.
> 
> > +static void
> > +intel_hdcp_disable_hdcp_line_rekeying(struct intel_encoder *encoder,
> > +				      struct intel_hdcp *hdcp)
> > +{
> > +	struct drm_i915_private *dev_priv = to_i915(encoder->base.dev);
> > +
> 
> Something on the lines of
> 
> /* Here we assume that HDMI is in TMDS mode */
> 
> But you are welcome to come up with something better.
> 
> Also, s/dev_priv/i915/
> 
The rename cannot be done as many of these register write and read function
Macro require the name to be dev_priv

Regards,
Suraj Kandpal

> > +	if (encoder->type != INTEL_OUTPUT_HDMI)
> > +		return;
> > +
> > +	if (DISPLAY_VER(dev_priv) >= 14) {
> > +		if (IS_METEORLAKE(dev_priv))
> > +			intel_de_rmw(dev_priv, MTL_CHICKEN_TRANS(hdcp-
> > >cpu_transcoder),
> > +				     HDCP_LINE_REKEY_DISABLE, 1);
> 
> Since we are only setting the HDCP_LINE_REKEY_DISABLE bit (and clearing
> nothing) the correct use of the function would be
> 
> 			intel_de_rmw(dev_priv, MTL_CHICKEN_TRANS(hdcp-
> >cpu_transcoder), 0, HDCP_LINE_REKEY_DISABLE);
> 
> > +		else
> > +			intel_de_rmw(dev_priv, TRANS_DDI_FUNC_CTL(hdcp-
> > >cpu_transcoder),
> > +				     TRANS_DDI_HDCP_LINE_REKEY_DISABLE,
> > 1);
> 
> Same here.
> 
> > +	}
> > +}
> > +
> >  static int intel_conn_to_vcpi(struct intel_atomic_state *state,
> >  			      struct intel_connector *connector)  { @@ -2005,6
> > +2029,8 @@ static int _intel_hdcp2_enable(struct intel_atomic_state
> > +*state,
> >  		    connector->base.base.id, connector->base.name,
> >  		    hdcp->content_type);
> >
> > +	intel_hdcp_disable_hdcp_line_rekeying(connector->encoder, hdcp);
> > +
> >  	ret = hdcp2_authenticate_and_encrypt(state, connector);
> >  	if (ret) {
> >  		drm_dbg_kms(&i915->drm, "HDCP2 Type%d  Enabling Failed.
> > (%d)\n", diff --git a/drivers/gpu/drm/i915/i915_reg.h
> > b/drivers/gpu/drm/i915/i915_reg.h index 3f34efcd7d6c..39b1a2d516fe
> > 100644
> > --- a/drivers/gpu/drm/i915/i915_reg.h
> > +++ b/drivers/gpu/drm/i915/i915_reg.h
> > @@ -4613,6 +4613,8 @@
> >  					    [TRANSCODER_D] =
> > _CHICKEN_TRANS_D))
> >  #define _MTL_CHICKEN_TRANS_A	0x604e0
> >  #define _MTL_CHICKEN_TRANS_B	0x614e0
> > +#define _MTL_CHICKEN_TRANS_C	0x624e0
> > +#define _MTL_CHICKEN_TRANS_D	0x634e0
> 
> These look good.
> 
> Regards
> 
> Chaitanya
> 
> >  #define MTL_CHICKEN_TRANS(trans)	_MMIO_TRANS((trans), \
> >  						    _MTL_CHICKEN_TRANS_A,
> > \
> >  						    _MTL_CHICKEN_TRANS_B)
> > @@ -4631,6 +4633,7 @@
> >  #define   DP_FEC_BS_JITTER_WA		REG_BIT(15)
> >  #define   PSR2_VSC_ENABLE_PROG_HEADER	REG_BIT(12)
> >  #define   DP_DSC_INSERT_SF_AT_EOL_WA	REG_BIT(4)
> > +#define   HDCP_LINE_REKEY_DISABLE	REG_BIT(0)
> >
> >  #define DISP_ARB_CTL	_MMIO(0x45000)
> >  #define   DISP_FBC_MEMORY_WAKE		REG_BIT(31)
> > @@ -5630,6 +5633,7 @@ enum skl_power_gate {
> >  #define  TRANS_DDI_EDP_INPUT_B_ONOFF	(5 << 12)
> >  #define  TRANS_DDI_EDP_INPUT_C_ONOFF	(6 << 12)
> >  #define  TRANS_DDI_EDP_INPUT_D_ONOFF	(7 << 12)
> > +#define  TRANS_DDI_HDCP_LINE_REKEY_DISABLE	REG_BIT(12)
> >  #define  TRANS_DDI_MST_TRANSPORT_SELECT_MASK
> 	REG_GENMASK(11,
> > 10)
> >  #define  TRANS_DDI_MST_TRANSPORT_SELECT(trans)	\
> >  	REG_FIELD_PREP(TRANS_DDI_MST_TRANSPORT_SELECT_MASK,
> > trans)
> > --
> > 2.43.2