[PATCH 06/14] drm/i915: Split current joiner hw state readout

Nautiyal, Ankit K ankit.k.nautiyal at intel.com
Mon Sep 9 07:55:04 UTC 2024


On 9/6/2024 8:59 PM, Ville Syrjälä wrote:
> On Fri, Sep 06, 2024 at 06:27:59PM +0530, Ankit Nautiyal wrote:
>> From: Stanislav Lisovskiy <stanislav.lisovskiy at intel.com>
>>
>> We need to add a new sanity checks and also do
>> some preparations for adding ultrajoiner hw state readout.
>> Lets first split reading of the uncompressed joiner and bigjoiner
>> bit masks into separate functions.
>>
>> v2: Fixed checkpatch warnings (Ankit)
>>
>> Signed-off-by: Stanislav Lisovskiy <stanislav.lisovskiy at intel.com>
>> Signed-off-by: Ankit Nautiyal <ankit.k.nautiyal at intel.com>
>> Reviewed-by: Suraj Kandpal <suraj.kandpal at intel.com>
>> ---
>>   drivers/gpu/drm/i915/display/intel_display.c | 65 +++++++++++++++-----
>>   1 file changed, 48 insertions(+), 17 deletions(-)
>>
>> diff --git a/drivers/gpu/drm/i915/display/intel_display.c b/drivers/gpu/drm/i915/display/intel_display.c
>> index 3278debf47cc..cdc7531311fc 100644
>> --- a/drivers/gpu/drm/i915/display/intel_display.c
>> +++ b/drivers/gpu/drm/i915/display/intel_display.c
>> @@ -3580,51 +3580,82 @@ static bool transcoder_ddi_func_is_enabled(struct drm_i915_private *dev_priv,
>>   	return tmp & TRANS_DDI_FUNC_ENABLE;
>>   }
>>   
>> -static void enabled_joiner_pipes(struct drm_i915_private *dev_priv,
>> -				 u8 *primary_pipes, u8 *secondary_pipes)
>> +static void enabled_uncompressed_joiner_pipes(struct drm_i915_private *dev_priv,
>> +					      u8 *primary_pipes, u8 *secondary_pipes)
>>   {
>>   	struct intel_crtc *crtc;
>>   
>>   	*primary_pipes = 0;
>>   	*secondary_pipes = 0;
>>   
>> +	if (DISPLAY_VER(dev_priv) < 13)
>> +		return;
>> +
>>   	for_each_intel_crtc_in_pipe_mask(&dev_priv->drm, crtc,
>>   					 joiner_pipes(dev_priv)) {
>>   		enum intel_display_power_domain power_domain;
>>   		enum pipe pipe = crtc->pipe;
>>   		intel_wakeref_t wakeref;
>>   
>> -		power_domain = intel_dsc_power_domain(crtc, (enum transcoder) pipe);
>> +		power_domain = POWER_DOMAIN_PIPE(pipe);
>>   		with_intel_display_power_if_enabled(dev_priv, power_domain, wakeref) {
>>   			u32 tmp = intel_de_read(dev_priv, ICL_PIPE_DSS_CTL1(pipe));
>>   
>> -			if (!(tmp & BIG_JOINER_ENABLE))
>> -				continue;
>> -
>> -			if (tmp & PRIMARY_BIG_JOINER_ENABLE)
>> +			if (tmp & UNCOMPRESSED_JOINER_PRIMARY)
>>   				*primary_pipes |= BIT(pipe);
>> -			else
>> +			if (tmp & UNCOMPRESSED_JOINER_SECONDARY)
>>   				*secondary_pipes |= BIT(pipe);
>>   		}
>> +	}
>> +}
>>   
>> -		if (DISPLAY_VER(dev_priv) < 13)
>> -			continue;
>> +static void enabled_bigjoiner_pipes(struct drm_i915_private *dev_priv,
>> +				    u8 *primary_pipes, u8 *secondary_pipes)
>> +{
>> +	struct intel_crtc *crtc;
>>   
>> -		power_domain = POWER_DOMAIN_PIPE(pipe);
>> +	*primary_pipes = 0;
>> +	*secondary_pipes = 0;
> We seem to be missing any kind of check to make sure bigjoiner
> is actually present in the hardware. Or am I just blind?
> If that is the case we need to fix it up before the
> refactoring goes in.

We have check in DP, and in one place while checking for maxdotclock.

But yes its missing here. I am thinking something like: 
https://patchwork.freedesktop.org/patch/612875/?series=133800&rev=7

Will try to introduce a the check here prior to any other patch, that 
can be used in other places.

>
>> +
>> +	for_each_intel_crtc_in_pipe_mask(&dev_priv->drm, crtc,
>> +					 joiner_pipes(dev_priv)) {
>> +		enum intel_display_power_domain power_domain;
>> +		enum pipe pipe = crtc->pipe;
>> +		intel_wakeref_t wakeref;
>> +
>> +		power_domain = intel_dsc_power_domain(crtc, (enum transcoder)pipe);
>>   		with_intel_display_power_if_enabled(dev_priv, power_domain, wakeref) {
>>   			u32 tmp = intel_de_read(dev_priv, ICL_PIPE_DSS_CTL1(pipe));
>>   
>> -			if (tmp & UNCOMPRESSED_JOINER_PRIMARY)
>> +			if (!(tmp & BIG_JOINER_ENABLE))
>> +				continue;
>> +
>> +			if (tmp & PRIMARY_BIG_JOINER_ENABLE)
>>   				*primary_pipes |= BIT(pipe);
>> -			if (tmp & UNCOMPRESSED_JOINER_SECONDARY)
>> +			else
>>   				*secondary_pipes |= BIT(pipe);
>>   		}
>>   	}
>> +}
>> +
>> +static void enabled_joiner_pipes(struct drm_i915_private *dev_priv,
>> +				 u8 *primary_pipes, u8 *secondary_pipes)
>> +{
>> +	u8 primary_uncompressed_joiner_pipes, primary_bigjoiner_pipes;
>> +	u8 secondary_uncompressed_joiner_pipes, secondary_bigjoiner_pipes;
>> +
>> +	enabled_uncompressed_joiner_pipes(dev_priv, &primary_uncompressed_joiner_pipes,
>> +					  &secondary_uncompressed_joiner_pipes);
>> +
>> +	enabled_bigjoiner_pipes(dev_priv, &primary_bigjoiner_pipes,
>> +				&secondary_bigjoiner_pipes);
>> +
>> +	*primary_pipes = 0;
>> +	*secondary_pipes = 0;
> These seem redundant.

Will do away with these.


>> +
>> +	*primary_pipes = primary_uncompressed_joiner_pipes | primary_bigjoiner_pipes;
>>   
>> -	/* Joiner pipes should always be consecutive primary and secondary */
>> -	drm_WARN(&dev_priv->drm, *secondary_pipes != *primary_pipes << 1,
>> -		 "Joiner misconfigured (primary pipes 0x%x, secondary pipes 0x%x)\n",
>> -		 *primary_pipes, *secondary_pipes);
> Where did this go?

I think this was unintended. Will fix this.

Regards,

Ankit

>
>> +	*secondary_pipes = secondary_uncompressed_joiner_pipes | secondary_bigjoiner_pipes;
>>   }
>>   
>>   static enum pipe get_joiner_primary_pipe(enum pipe pipe, u8 primary_pipes, u8 secondary_pipes)
>> -- 
>> 2.45.2


More information about the Intel-gfx mailing list