[Intel-gfx] [PATCH v3 5/8] drm/i915/dp: Use sink_format in dp_is_ycbcr420

[Nautiyal, Ankit K]

On 10/20/2022 10:24 PM, Ville Syrjälä wrote:
> On Tue, Oct 11, 2022 at 12:04:44PM +0530, Ankit Nautiyal wrote:
>> Use the new crtc_state member sink_format to check if DP DFP will use
>> YCBCR420.
>> The earlier intel_dp->dfp members are not required as decision to use
>> color format conversion by the PCON is computed and stored in
>> crtc_state. Also drop the intel_dp argument in the helper function.
>>
>> Signed-off-by: Ankit Nautiyal <ankit.k.nautiyal at intel.com>
>> ---
>>   drivers/gpu/drm/i915/display/intel_dp.c | 13 +++++--------
>>   1 file changed, 5 insertions(+), 8 deletions(-)
>>
>> diff --git a/drivers/gpu/drm/i915/display/intel_dp.c b/drivers/gpu/drm/i915/display/intel_dp.c
>> index 99d72b345907..d8ef4dccf0c1 100644
>> --- a/drivers/gpu/drm/i915/display/intel_dp.c
>> +++ b/drivers/gpu/drm/i915/display/intel_dp.c
>> @@ -1177,19 +1177,16 @@ static bool intel_dp_supports_dsc(struct intel_dp *intel_dp,
>>   		drm_dp_sink_supports_dsc(intel_dp->dsc_dpcd);
>>   }
>>   
>> -static bool intel_dp_is_ycbcr420(struct intel_dp *intel_dp,
>> -				 const struct intel_crtc_state *crtc_state)
>> +static bool intel_dp_is_ycbcr420(const struct intel_crtc_state *crtc_state)
>>   {
>> -	return crtc_state->output_format == INTEL_OUTPUT_FORMAT_YCBCR420 ||
>> -		(crtc_state->output_format == INTEL_OUTPUT_FORMAT_YCBCR444 &&
>> -		 intel_dp->dfp.ycbcr_444_to_420);
>> +	return crtc_state->sink_format == INTEL_OUTPUT_FORMAT_YCBCR420;
>>   }
>>   
>>   static int intel_dp_hdmi_compute_bpc(struct intel_dp *intel_dp,
>>   				     const struct intel_crtc_state *crtc_state,
>>   				     int bpc, bool respect_downstream_limits)
>>   {
>> -	bool ycbcr420_output = intel_dp_is_ycbcr420(intel_dp, crtc_state);
>> +	bool ycbcr420_output = intel_dp_is_ycbcr420(crtc_state);
>>   	int clock = crtc_state->hw.adjusted_mode.crtc_clock;
> I think I'd take this a notch further and start passing
> the sink_format to most (maybe even all?) places that
> currently take a 'bool ycbcr420_output'. That could also
> serve as a slight step towards 4:2:2 support.

Makes sense, will modify the functions as suggested.

Regards,

Ankit

>
>>   
>>   	/*
>> @@ -1986,7 +1983,7 @@ intel_dp_compute_output_format(struct intel_encoder *encoder,
>>   
>>   	crtc_state->output_format = intel_dp_output_format(connector, ycbcr_420_only);
>>   
>> -	if (ycbcr_420_only && !intel_dp_is_ycbcr420(intel_dp, crtc_state)) {
>> +	if (ycbcr_420_only && !intel_dp_is_ycbcr420(crtc_state)) {
>>   		drm_dbg_kms(&i915->drm,
>>   			    "YCbCr 4:2:0 mode but YCbCr 4:2:0 output not possible. Falling back to RGB.\n");
>>   		crtc_state->output_format = INTEL_OUTPUT_FORMAT_RGB;
>> @@ -2001,7 +1998,7 @@ intel_dp_compute_output_format(struct intel_encoder *encoder,
>>   	ret = intel_dp_compute_link_config(encoder, crtc_state, conn_state,
>>   					   respect_downstream_limits);
>>   	if (ret) {
>> -		if (intel_dp_is_ycbcr420(intel_dp, crtc_state) ||
>> +		if (intel_dp_is_ycbcr420(crtc_state) ||
>>   		    !connector->base.ycbcr_420_allowed ||
>>   		    !drm_mode_is_420_also(info, adjusted_mode))
>>   			return ret;
>> -- 
>> 2.25.1