[Intel-gfx] [PATCH] drm/i915/hdmi: Prune modes that require HDMI2.1 FRL

[Nautiyal, Ankit K]

Hi Arun,

Thanks for the comments.

Please find my response inline.

On 7/8/2022 9:30 AM, Murthy, Arun R wrote:
>
>> -----Original Message-----
>> From: Intel-gfx <intel-gfx-bounces at lists.freedesktop.org> On Behalf Of Ankit
>> Nautiyal
>> Sent: Thursday, July 7, 2022 10:57 AM
>> To: intel-gfx at lists.freedesktop.org
>> Subject: [Intel-gfx] [PATCH] drm/i915/hdmi: Prune modes that require
>> HDMI2.1 FRL
>>
>> HDMI2.1 requires some higher resolution video modes to be enumerated
>> only if HDMI2.1 Fixed Rate Link (FRL) is supported.
>> Current platforms do not support FRL transmission so prune modes that
>> require HDMI2.1 FRL.
>>
> If the hardware doesn't support FRL then it basically blocks HDMI2.1 feature.
> Then it wont be possible to use any resolution above 4k60 is it?


Yes right. As I understand, the HDMI2.1a supersedes HDMI2.0b, and so the

platforms  supporting HDMI2.0 must now pass the HDMI2.1 CTS.
The HDMI2.1a spec introduces Marketing Feature names for 4K100, 4K120,
8k at 50, 8k at 60 with suffix A, and B.
Suffix A meaning mode supported without compression, and B meaning, mode
supported with compression.

There are CTS tests that expect these modes not to be enumerated, if the
source does support the given requirements.


>
>> Signed-off-by: Ankit Nautiyal <ankit.k.nautiyal at intel.com>
>> ---
>>   drivers/gpu/drm/i915/display/intel_hdmi.c | 21 +++++++++++++++++++++
>>   1 file changed, 21 insertions(+)
>>
>> diff --git a/drivers/gpu/drm/i915/display/intel_hdmi.c
>> b/drivers/gpu/drm/i915/display/intel_hdmi.c
>> index ebd91aa69dd2..93c00b61795f 100644
>> --- a/drivers/gpu/drm/i915/display/intel_hdmi.c
>> +++ b/drivers/gpu/drm/i915/display/intel_hdmi.c
>> @@ -1974,6 +1974,20 @@ intel_hdmi_mode_clock_valid(struct
>> drm_connector *connector, int clock,
>>   	return status;
>>   }
>>
>> +/*
>> + * HDMI2.1 requires higher resolution modes like 8k60, 4K120 to be
>> + * enumerated only if FRL is supported. Platforms not supporting FRL
>> + * must prune these modes.
>> + */
>> +static bool
>> +hdmi21_frl_quirk(int dotclock, bool frl_supported) {
>> +	if (dotclock >= 600000 && !frl_supported)
>> +		return true;
>> +
>> +	return false;
>> +}
>> +
>>   static enum drm_mode_status
>>   intel_hdmi_mode_valid(struct drm_connector *connector,
>>   		      struct drm_display_mode *mode)
>> @@ -2001,6 +2015,13 @@ intel_hdmi_mode_valid(struct drm_connector
>> *connector,
>>   		clock *= 2;
>>   	}
>>
>> +	/*
>> +	 * Current Platforms do not support HDMI2.1 FRL mode of
>> transmission,
>> +	 * so prune the modes that require FRL.
>> +	 */
>> +	if (hdmi21_frl_quirk(clock, false))
>> +		return MODE_BAD;
>> +
> Instead of setting this frl_supported as false, can we get this info from hardware, so that when
> our hardware supports it later it would be easy to enable this.

We can have something like:

src_supports_frl()

{

/* FRL not supported in

return false;

}

Later we can add check for Platform when HDMI2.1 FRL is supported and 
rate parsed from VBT.

Regards,

Ankit


>
> Thanks and Regards,
> Arun R Murthy
> -------------------