Re: [RFC PATCH] drm/i915: push shared dpll enable to encoders on DDI platforms

[Jani Nikula <jani.nikula@intel.com>]

On Fri, 06 Oct 2017, Daniel Vetter <daniel@ffwll.ch> wrote:
> On Thu, Oct 05, 2017 at 02:50:13PM +0300, Jani Nikula wrote:
>> Signed-off-by: Jani Nikula <jani.nikula@intel.com>
>> ---
>>  drivers/gpu/drm/i915/intel_crt.c     |  3 +++
>>  drivers/gpu/drm/i915/intel_ddi.c     | 18 ++++++++++++++++++
>>  drivers/gpu/drm/i915/intel_display.c |  3 ---
>>  3 files changed, 21 insertions(+), 3 deletions(-)
>> 
>> diff --git a/drivers/gpu/drm/i915/intel_crt.c b/drivers/gpu/drm/i915/intel_crt.c
>> index 668e8c3e791d..b3094d606329 100644
>> --- a/drivers/gpu/drm/i915/intel_crt.c
>> +++ b/drivers/gpu/drm/i915/intel_crt.c
>> @@ -255,6 +255,9 @@ static void hsw_pre_pll_enable_crt(struct intel_encoder *encoder,
>>  	WARN_ON(!intel_crtc->config->has_pch_encoder);
>>  
>>  	intel_set_pch_fifo_underrun_reporting(dev_priv, PIPE_A, false);
>> +
>> +	if (intel_crtc->config->shared_dpll)
>> +		intel_enable_shared_dpll(intel_crtc);
>>  }
>
> Looking at haswell_crtc_compute_clock all outputs on hsw+ need a shared
> dpll, except dsi. Here's my proposal:
>
> - Drop the if condition when pushing into encoder callbacks, we statically
>   know which one it is. With that I think this patch here looks good.

Agreed. You think the patch at hand is fine after that as the first
step?

> - Push the clock computation into encoders to get rid of the silly DSI
>   special in haswell_crtc_compute_clock. Which then again will force you
>   to get rid of all the encoder special casing in dpll_mgr->get_dpll, plus
>   passing the encoder argument around.
>
>   I think the prettier way to do this is to pre-fill the clock we want in
>   the encoder compute_config callback, and then also call
>   intel_find_shared_dpll from there.

Are you proposing to do this for all platforms? Or just hsw/ddi+?

Do we still call ->crtc_compute_clock() in intel_crtc_atomic_check(),
and the subsequent call chain will just use the information pre-filled
in compute config?

BR,
Jani.

>
> There's a lot more that should be untangled imo in intel_ddi.c and
> computed in compute_config of the right encoder type instead of if ladders
> all over, but we're slowly getting there I think.
>
> Cheers, Daniel
>
>>  
>>  static void hsw_pre_enable_crt(struct intel_encoder *encoder,
>> diff --git a/drivers/gpu/drm/i915/intel_ddi.c b/drivers/gpu/drm/i915/intel_ddi.c
>> index f1adc2544ab9..c1152c602f08 100644
>> --- a/drivers/gpu/drm/i915/intel_ddi.c
>> +++ b/drivers/gpu/drm/i915/intel_ddi.c
>> @@ -2407,13 +2407,29 @@ static void intel_disable_ddi(struct intel_encoder *intel_encoder,
>>  	}
>>  }
>>  
>> +static void intel_ddi_pre_pll_enable(struct intel_encoder *encoder,
>> +				     const struct intel_crtc_state *pipe_config,
>> +				     const struct drm_connector_state *conn_state)
>> +{
>> +	struct drm_crtc *crtc = pipe_config->base.crtc;
>> +	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
>> +
>> +	if (intel_crtc->config->shared_dpll)
>> +		intel_enable_shared_dpll(intel_crtc);
>> +}
>> +
>>  static void bxt_ddi_pre_pll_enable(struct intel_encoder *encoder,
>>  				   const struct intel_crtc_state *pipe_config,
>>  				   const struct drm_connector_state *conn_state)
>>  {
>> +	struct drm_crtc *crtc = pipe_config->base.crtc;
>> +	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
>>  	uint8_t mask = pipe_config->lane_lat_optim_mask;
>>  
>>  	bxt_ddi_phy_set_lane_optim_mask(encoder, mask);
>> +
>> +	if (intel_crtc->config->shared_dpll)
>> +		intel_enable_shared_dpll(intel_crtc);
>>  }
>>  
>>  void intel_ddi_prepare_link_retrain(struct intel_dp *intel_dp)
>> @@ -2714,6 +2730,8 @@ void intel_ddi_init(struct drm_i915_private *dev_priv, enum port port)
>>  	intel_encoder->enable = intel_enable_ddi;
>>  	if (IS_GEN9_LP(dev_priv))
>>  		intel_encoder->pre_pll_enable = bxt_ddi_pre_pll_enable;
>> +	else
>> +		intel_encoder->pre_pll_enable = intel_ddi_pre_pll_enable;
>>  	intel_encoder->pre_enable = intel_ddi_pre_enable;
>>  	intel_encoder->disable = intel_disable_ddi;
>>  	intel_encoder->post_disable = intel_ddi_post_disable;
>> diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c
>> index 9f2bf3b3f759..6d0573350786 100644
>> --- a/drivers/gpu/drm/i915/intel_display.c
>> +++ b/drivers/gpu/drm/i915/intel_display.c
>> @@ -5490,9 +5490,6 @@ static void haswell_crtc_enable(struct intel_crtc_state *pipe_config,
>>  
>>  	intel_encoders_pre_pll_enable(crtc, pipe_config, old_state);
>>  
>> -	if (intel_crtc->config->shared_dpll)
>> -		intel_enable_shared_dpll(intel_crtc);
>> -
>>  	if (intel_crtc_has_dp_encoder(intel_crtc->config))
>>  		intel_dp_set_m_n(intel_crtc, M1_N1);
>>  
>> -- 
>> 2.11.0
>> 

-- 
Jani Nikula, Intel Open Source Technology Center
_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/intel-gfx