[Intel-gfx] [RFC 4/7] drm/i915: Program PPS registers

[Kannan, Vandana]

On 20-Oct-14 9:38 PM, Daniel Vetter wrote:
> On Mon, Oct 20, 2014 at 06:20:06PM +0530, Vandana Kannan wrote:
>> Actually set values into PPS related registers. This implementation is
>> equivalent to intel_dp_panel_power_sequencer_registers where the values
>> saved intially are written into registers.
>>
>> Signed-off-by: Vandana Kannan <vandana.kannan at intel.com>
>> ---
>>   drivers/gpu/drm/i915/intel_dp.c    | 80 ++------------------------------------
>>   drivers/gpu/drm/i915/intel_drv.h   |  3 ++
>>   drivers/gpu/drm/i915/intel_panel.c | 70 +++++++++++++++++++++++++++++++++
>>   3 files changed, 76 insertions(+), 77 deletions(-)
>>
>> diff --git a/drivers/gpu/drm/i915/intel_dp.c b/drivers/gpu/drm/i915/intel_dp.c
>> index a433c5f..ca11eb1 100644
>> --- a/drivers/gpu/drm/i915/intel_dp.c
>> +++ b/drivers/gpu/drm/i915/intel_dp.c
>> @@ -248,7 +248,7 @@ unpack_aux(uint32_t src, uint8_t *dst, int dst_bytes)
>>   }
>>
>>   /* hrawclock is 1/4 the FSB frequency */
>> -static int
>> +int
>>   intel_hrawclk(struct drm_device *dev)
>>   {
>>   	struct drm_i915_private *dev_priv = dev->dev_private;
>> @@ -281,11 +281,6 @@ intel_hrawclk(struct drm_device *dev)
>>   	}
>>   }
>>
>> -static void
>> -intel_dp_init_panel_power_sequencer_registers(struct drm_device *dev,
>> -					      struct intel_dp *intel_dp,
>> -					      struct edp_power_seq *out);
>> -
>>   static void pps_lock(struct intel_dp *intel_dp)
>>   {
>>   	struct intel_digital_port *intel_dig_port = dp_to_dig_port(intel_dp);
>> @@ -4716,76 +4711,6 @@ static void intel_dp_init_panel_power_timestamps(struct intel_dp *intel_dp)
>>   	intel_dp->last_backlight_off = jiffies;
>>   }
>>
>> -static void
>> -intel_dp_init_panel_power_sequencer_registers(struct drm_device *dev,
>> -					      struct intel_dp *intel_dp,
>> -					      struct edp_power_seq *seq)
>
> Hm, moving this function looks like it would be clearer as part of patch
> 1?
>
> Otherwise I've done a (very) quick read-through of your series and on a
> high level it looks sane I think. So please sign someone up for the
> detailed review (and make sure that person is aware of that AR) so I can
> merge this.
>
> Thanks, Daniel
>
Thanks Daniel..
I went through the LVDS and DSI parts. If software delays are used for 
both, would it really add more value to have the delays in a function in 
intel_panel.c and call whenever required ?
I'm thinking directly calling a msleep would be simpler but that would 
mean the PPS part wont be in one place in intel_panel.c.
Let me know what you think about this..

- Vandana